Original Research

Asthma results in approximately 125,000 hospitalizations for children annually in the United States.^1,2 The National Heart, Lung, and Blood Institute guidelines recommend that children with persistent asthma be treated with a daily controller medication, ie, an inhaled corticosteroid (ICS).³ Hospitalization for an asthma exacerbation provides an opportunity to optimize daily controller medications and improve disease self-management by providing access to medications and teaching appropriate use of complicated inhalation devices.

To reduce readmission⁴ by mitigating low rates of postdischarge filling of ICS prescriptions,^5,6 a strategy of “meds-in-hand” was implemented at discharge. “Meds-in-hand” mitigates medication access as a barrier to adherence by ensuring that patients are discharged from the hospital with all required medications in hand, removing any barriers to filling their initial prescriptions.⁷ The Asthma Improvement Collaborative at Cincinnati Children’s Hospital Medical Center (CCHMC) previously applied quality improvement methodology to implement “meds-in-hand” as a key intervention in a broad strategy that successfully reduced asthma-specific utilization for the 30-day period following an asthma-related hospitalization of publicly insured children from 12% to 7%.^8,9

At the onset of the work described in this manuscript, children hospitalized with an acute exacerbation of persistent asthma were most often treated with an ICS while inpatients in addition to a standard short course of oral systemic corticosteroids. Conceptually, inpatient administration of ICS provided the opportunity to teach effective device usage with each inpatient administration and to reinforce daily use of the ICS as part of the patient’s daily home medication regimen. However, a proportion of patients admitted for an asthma exacerbation were noted to receive more than one ICS inhaler during their admission, most commonly due to a change in dose or type of ICS. When this occurred, the initially dispensed inhaler was discarded despite weeks of potential doses remaining. While some hospitals preferentially dispense ICS devices marketed to institutions with fewer doses per device, our pharmacy primarily dispensed ICS devices identical to retail locations containing at least a one-month supply of medication. In addition to the wasted medication, this practice resulted in additional work by healthcare staff, unnecessary patient charges, and potentially contributed to confusion about the discharge medication regimen.

Our specific aim for this quality improvement study was to reduce the monthly percentage of admissions for an acute asthma exacerbation treated with >1 ICS from 7% to 4% over a six-month period.

Context

CCHMC is a quaternary care pediatric health system with more than 600 inpatient beds and 800-900 inpatient admissions per year for acute asthma exacerbation. The Hospital Medicine service cares for patients with asthma on five clinical teams across two different campuses. Care teams are supervised by an attending physician and may include residents, fellows, or nurse practitioners. Patients hospitalized for an acute asthma exacerbation may receive a consult from the Asthma Center consult team, staffed by faculty from either the Pediatric Pulmonology or Allergy/Immunology divisions. Respiratory therapists (RTs) administer inhaled medications and provide asthma education.

Planning the Intervention

Our improvement team included physicians from Hospital Medicine and Pulmonary Medicine, an Asthma Education Coordinator, a Clinical Pharmacist, a Pediatric Chief Resident, and a clinical research coordinator. Initial interventions targeted a single resident team at the main campus before spreading improvement activities to all resident teams at the main campus and then the satellite campus by February 2017.

Development of our process map (Figure 1) revealed that the decision for ordering inpatient ICS treatment frequently occurred at admission. Subsequently, the care team or consulting team might make a change in the ICS to fine-tune the outpatient medication regimen given that admission for asthma often results from suboptimal chronic symptom control. Baseline analysis of changes in ICS orders revealed that 81% of ICS changes were associated with a step-up in therapy, defined as an increase in the daily dose of the ICS or the addition of a long-acting beta-agonist. The other common ICS adjustment, accounting for 17%, was a change in corticosteroid without a step-up in therapy, (ie, beclomethasone to fluticasone) that typically occurred near the end of the hospitalization to accommodate outpatient insurance formularies, independent of patient factors related to illness severity.

Improvement Activities

We conceived key drivers or domains that would be necessary to address to effect change. Key drivers included a standardized process for delayed initiation of ICS and confirmation of outpatient insurance prescription drug coverage, prescriber education, and real-time failure notification.

PDSA Interventions

PDSA 1 & 2: Standardized Process for Initiation of ICS

Our initial tests of change targeted the timing of when an ICS was ordered during hospitalization for an asthma exacerbation. Providers were instructed to delay ordering an ICS until the patient’s albuterol treatments were spaced to every three hours and to include a standardized communication prompt within the albuterol order. The prompt instructed the RT to contact the provider once the patient’s albuterol treatments were spaced to every three hours and ask for an ICS order, if appropriate. This intervention was abandoned because it did not reliably occur.

The subsequent intervention delayed the start of ICS treatment by using a PRN indication advising that the ICS was to be administered once the patient’s albuterol treatments were spaced to every three hours. However, after an error resulted in the PRN indication being included on a discharge prescription for an ICS, the PRN indication was abandoned. Subsequent work to develop a standardized process for delayed initiation of ICS occurred as part of the workflow to address the confirmation of outpatient formulary coverage as described next.

PDSA 3: Prioritize the Use of the Institution’s Outpatient Pharmacy

Medication changes that occurred because of outpatient insurance formulary denials were a unique challenge; they required a medication change after the discharge treatment plan had been finalized, and a prescription was already submitted to the outpatient pharmacy. In addition, neither our inpatient electronic medical record nor our inpatient hospital pharmacy has access to decision support tools that incorporate outpatient prescription formulary coverage. Alternatively, outpatient pharmacies have a standard workflow that routinely confirms insurance coverage before dispensing medication. The institutional policy was modified to allow for the inpatient administration of patient-supplied medications, pursuant to an inpatient order. Patient-supplied medications include those brought from home or those supplied by the outpatient pharmacy.

Subsequently, we developed a standardized process to confirm outpatient prescription drug coverage by using our hospital-based outpatient pharmacy to dispense ICS for inpatient treatment and asthma education. This new workflow included placing an order for an ICS at admission as a patient-supplied medication with an administration comment to “please administer once available from the outpatient pharmacy” (Figure 1). Then, once the discharge medication plan is finalized, the prescription is submitted to the outpatient pharmacy. Following verification of insurance coverage, the outpatient pharmacy dispenses the ICS, allowing it to be used for patient education and inpatient administration. If the patient is ineligible to have their prescription filled by the outpatient pharmacy for reasons other than formulary coverage, the ICS is dispensed from the hospital inpatient pharmacy as per the previous standard workflow. Inpatient ICS inhalers are then relabeled for home use per the existing practice to support medications-in-hand.

Further workflow improvements occurred following the development of an algorithm to help the outpatient pharmacy contact the correct inpatient team, and augmentation of the medication delivery process included notification of the RT when the ICS was delivered to the inpatient unit.

PDSA 4: Prescriber Education

Prescribers received education regarding PDSA interventions before testing and throughout the improvement cycle. Education sessions included informal coaching by the Asthma Education Coordinator, e-mail reminders containing screenshots of the ordering process, and formal didactic sessions for ordering providers. The Asthma Education Coordinator also provided education to the nursing and respiratory therapy staff regarding the implemented process and workflow changes.

PDSA 5: Real-Time Failure Notification

To supplement education for the complicated process change, the improvement team utilized a decision support tool (Vigilanz Corp., Chicago, IL) linked to EMR data to provide notification of real-time process failures. When a patient with an admission diagnosis of asthma had a prescription for an ICS verified and dispensed by the inpatient pharmacy, an automated message with relevant patient information would be sent to a member of the improvement team. Following a brief chart review, directed feedback could be offered to the ordering provider, allowing the prescription to be redirected to the outpatient pharmacy.

Study of the Improvement

Patients of all ages, with the International Classification of Diseases, Ninth Revision, and Tenth Revision codes for asthma (493.xx or J45.xx) were included in data collection and analysis if they were treated by the Hospital Medicine service, as the first inpatient service or after transfer from the ICU, and prescribed an ICS with or without a long-acting beta-agonist. Data were collected retrospectively and aggregated monthly. The baseline period was from January 2015 through October 2016. The intervention period was from November 2016 through March 2018. The prolonged baseline and study periods were utilized to understand the seasonal nature of asthma exacerbations.

Measures

Our primary outcome measure was defined as the monthly number of patients admitted to Hospital Medicine for an acute asthma exacerbation administered more than one ICS divided by the total number of asthma patients administered at least one dose of an ICS (patient-supplied or dispensed from the inpatient pharmacy). A full list of ICS is included in the appendix Table.

A secondary process measure approximated our adherence to obtaining ICS from the outpatient pharmacy for inpatient use. All medications administered during hospitalization are documented in the medication administration report. However, only medications dispensed from the inpatient pharmacy are associated with a patient charge. Patient-supplied medications, including those dispensed from the hospital outpatient pharmacy, are not associated with an inpatient charge. Therefore, the secondary process measure was defined as the monthly number of asthma patients administered an ICS not associated with an inpatient charge divided by the total number of asthma patients administered an ICS.

A cost outcome measure was developed to track changes in the average cost of an ICS included on inpatient bills during hospitalization for an asthma exacerbation. This outcome measure was defined as the total monthly cost, using the average wholesale price, of the ICS included on the inpatient bill for an asthma exacerbation, divided by the total number of asthma patients administered at least one dose of an ICS (patient supplied or dispensed from the inpatient pharmacy). The costs of patient-supplied medications (including those dispensed from the outpatient pharmacy) are not included on inpatient hospital bills or this secondary outcome measure.

Our a priori intent was to reduce ICS medication waste while maintaining a highly reliable system that included inpatient administration and education with ICS devices and maintain our medications-in-hand practice. A balancing measure was developed to monitor the reliability of inpatient administration of ICS. It was defined as the monthly number of patients who received a discharge prescription for an ICS and were administered an ICS while an inpatient divided by the total number of asthma patients with a discharge prescription for an ICS.

Analysis

Measures were evaluated using statistical process control charts and special cause variation was determined by previously established rules. Our primary, secondary, and balancing measures were all evaluated using a p-chart with variable subgroup size. The cost outcome measure was evaluated using an X-bar S control chart.^11-13

Primary Outcome Measure

During the baseline period, 7.4% of patients admitted to Hospital Medicine for an acute asthma exacerbation were administered more than one ICS, ranging from 0%-20% of patients per month (Figure 2). Following the start of our interventions, we met criteria for special cause allowing adjustment of the centerline.¹³ The mean percentage of patients receiving more than one ICS decreased from 7.4% to 0.7%. Figure 2 includes the n-value displayed each month and represents all patients admitted to the Hospital Medicine service with an asthma exacerbation who were administered at least one ICS.

Secondary Process Measure

During the baseline period, there were only rare occurrences (less than 1%) of a patient-supplied ICS being administered during an asthma admission. Following the start of our intervention period, the frequency of inpatient administration of patient-supplied ICS showed a rapid increase and met rules for special cause with an increase in the mean percent from 0.7% to 50% (Figure 3). The n-value displayed each month represents all patients admitted to the Hospital Medicine service for an asthma exacerbation administered at least one ICS.

Cost Outcome Measure

The average cost of an ICS billed during hospitalization for an acute asthma exacerbation was $236.57 per ICS during the baseline period. After the intervention period, the average inpatient cost for ICS decreased by 62% to $90.25 per ICS (Figure 4).

Balancing Measure

Our team reduced the monthly percent of children hospitalized with an acute asthma exacerbation administered more than one ICS from 7.4% to 0.7% after implementation of a new workflow process for ordering ICS utilizing the hospital-based outpatient pharmacy. The new workflow delayed ordering and administration of the initial inpatient ICS treatment, allowing time to consider a step-up in therapy. The brief delay in initiating ICS is not expected to have clinical consequence given the concomitant treatment with systemic corticosteroids. In addition, the outpatient pharmacy was utilized to verify insurance coverage reliably prior to dispensing ICS, reducing medication waste, and discharge delays due to outpatient medication formulary conflicts.

Our hospital’s previous approach to inpatient asthma care resulted in a highly reliable process to ensure patients were discharged with medications-in-hand as part of a broader system that effectively decreased reutilization. However, the previous process inadvertently resulted in medication waste. This waste included nearly full inhalers being discarded, additional work by the healthcare team (ordering providers, pharmacists, and RTs), and unnecessary patient charges.

While the primary driver of our decision to use the outpatient pharmacy was to adjudicate insurance prescription coverage reliably to prevent waste, this change likely resulted in a financial benefit to patients. The average cost per asthma admission of an inpatient billed for ICS using the average wholesale price, decreased by 62% following our interventions. The decrease in cost was primarily driven by using patient-supplied medications, including prescriptions newly filled by the on-site outpatient pharmacy, whose costs were not captured in this measure. While our secondary measure may underestimate the total expense incurred by families for an ICS, families likely receive their medications at a lower cost from the outpatient pharmacy than if the ICS was provided by an inpatient pharmacy. The average wholesale price is not what families are charged or pay for medications, partly due to differences in overhead costs that result in inpatient pharmacies having significantly higher charges than outpatient pharmacies. In addition, the 6.7% absolute reduction of our primary measure resulted in direct savings by reducing inpatient medication waste. Our process results in 67 fewer wasted ICS devices ($15,960) per 1,000 admissions for asthma exacerbation, extrapolated using the average cost ($238.20, average wholesale price) of each ICS during the baseline period.

Our quality improvement study had several limitations. (1) The interventions occurred at a single center with an established culture that embraces quality improvement, which may limit the generalizability of the work. (2) Our process verified insurance coverage with a hospital-based outpatient pharmacy. Some ICS prescriptions continued to be dispensed from the inpatient pharmacy, limiting our ability to verify insurance coverage. Local factors, including regulatory restrictions and delivery requirements, may limit the generalizability of using an outpatient pharmacy in this manner. (3) We achieved our goal of decreasing medication waste, but our a priori goal was to maintain our commitment to our established practice of interactive patient education with an ICS device as well as medications-in-hand at time of discharge. Our balancing measure showed a decrease in the percent of patients with a discharge prescription for an ICS who also received an inpatient dose of that ICS. This implies a decreased fidelity in our previously established education protocols. We had postulated that this occurred when the patient-supplied medication arrived on the day of discharge, but not close to when the medication was scheduled on the medication administration report, preventing administration. However, this is not a direct measure of patients receiving medications-in-hand or interactive medication education. Both may have occurred without administration of the ICS. (4) Despite a hospital culture that embraces quality improvement, this project required a significant change in the workflow that required considerable education at the time of implementation to integrate the new process reliably. However, once the process was in place, we have been able to sustain our improvement with limited educational investment.

Implementation of a new process for ordering ICS that emphasized delaying treatment until all necessary information was available and using an outpatient pharmacy to confirm insurance formulary coverage reduced the waste associated with more than one ICS being prescribed during a single admission.

Acknowledgments

The authors thank Sally Pope, MPH and Dr. Michael Carlisle, MD for their contribution to the quality improvement project. Thank you to Drs. Karen McDowell, MD and Carolyn Kercsmar, MD for advisement of our quality improvement project.

The authors appreciate the following individuals for their invaluable contributions. Dr. Hoefgen conceptualized and designed the study, was a member of the primary improvement team, carried out initial analysis, drafted the initial manuscript, and reviewed and revised the manuscript. Drs. Jones and Torres Garcia, and Mr. Hare were members of the primary improvement team who contributed to the design of the quality improvement study and interventions, ongoing data interpretation, and critically reviewed the manuscript. Dr. Courter contributed to the conceptualization and designed the study, was a member of the primary improvement team, designed data collection instruments, and critically reviewed and revised the manuscript. Dr. Simmons conceptualized and designed the study, critically reviewed the manuscript for important intellectual content, and reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Disclaimer

The information or content and conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsements be inferred by the BHPR, HRSA, DHHS, or the U.S. Government.

Asthma results in approximately 125,000 hospitalizations for children annually in the United States.^1,2 The National Heart, Lung, and Blood Institute guidelines recommend that children with persistent asthma be treated with a daily controller medication, ie, an inhaled corticosteroid (ICS).³ Hospitalization for an asthma exacerbation provides an opportunity to optimize daily controller medications and improve disease self-management by providing access to medications and teaching appropriate use of complicated inhalation devices.

To reduce readmission⁴ by mitigating low rates of postdischarge filling of ICS prescriptions,^5,6 a strategy of “meds-in-hand” was implemented at discharge. “Meds-in-hand” mitigates medication access as a barrier to adherence by ensuring that patients are discharged from the hospital with all required medications in hand, removing any barriers to filling their initial prescriptions.⁷ The Asthma Improvement Collaborative at Cincinnati Children’s Hospital Medical Center (CCHMC) previously applied quality improvement methodology to implement “meds-in-hand” as a key intervention in a broad strategy that successfully reduced asthma-specific utilization for the 30-day period following an asthma-related hospitalization of publicly insured children from 12% to 7%.^8,9

At the onset of the work described in this manuscript, children hospitalized with an acute exacerbation of persistent asthma were most often treated with an ICS while inpatients in addition to a standard short course of oral systemic corticosteroids. Conceptually, inpatient administration of ICS provided the opportunity to teach effective device usage with each inpatient administration and to reinforce daily use of the ICS as part of the patient’s daily home medication regimen. However, a proportion of patients admitted for an asthma exacerbation were noted to receive more than one ICS inhaler during their admission, most commonly due to a change in dose or type of ICS. When this occurred, the initially dispensed inhaler was discarded despite weeks of potential doses remaining. While some hospitals preferentially dispense ICS devices marketed to institutions with fewer doses per device, our pharmacy primarily dispensed ICS devices identical to retail locations containing at least a one-month supply of medication. In addition to the wasted medication, this practice resulted in additional work by healthcare staff, unnecessary patient charges, and potentially contributed to confusion about the discharge medication regimen.

Our specific aim for this quality improvement study was to reduce the monthly percentage of admissions for an acute asthma exacerbation treated with >1 ICS from 7% to 4% over a six-month period.

Context

CCHMC is a quaternary care pediatric health system with more than 600 inpatient beds and 800-900 inpatient admissions per year for acute asthma exacerbation. The Hospital Medicine service cares for patients with asthma on five clinical teams across two different campuses. Care teams are supervised by an attending physician and may include residents, fellows, or nurse practitioners. Patients hospitalized for an acute asthma exacerbation may receive a consult from the Asthma Center consult team, staffed by faculty from either the Pediatric Pulmonology or Allergy/Immunology divisions. Respiratory therapists (RTs) administer inhaled medications and provide asthma education.

Planning the Intervention

Our improvement team included physicians from Hospital Medicine and Pulmonary Medicine, an Asthma Education Coordinator, a Clinical Pharmacist, a Pediatric Chief Resident, and a clinical research coordinator. Initial interventions targeted a single resident team at the main campus before spreading improvement activities to all resident teams at the main campus and then the satellite campus by February 2017.

Development of our process map (Figure 1) revealed that the decision for ordering inpatient ICS treatment frequently occurred at admission. Subsequently, the care team or consulting team might make a change in the ICS to fine-tune the outpatient medication regimen given that admission for asthma often results from suboptimal chronic symptom control. Baseline analysis of changes in ICS orders revealed that 81% of ICS changes were associated with a step-up in therapy, defined as an increase in the daily dose of the ICS or the addition of a long-acting beta-agonist. The other common ICS adjustment, accounting for 17%, was a change in corticosteroid without a step-up in therapy, (ie, beclomethasone to fluticasone) that typically occurred near the end of the hospitalization to accommodate outpatient insurance formularies, independent of patient factors related to illness severity.

Improvement Activities

We conceived key drivers or domains that would be necessary to address to effect change. Key drivers included a standardized process for delayed initiation of ICS and confirmation of outpatient insurance prescription drug coverage, prescriber education, and real-time failure notification.

PDSA Interventions

PDSA 1 & 2: Standardized Process for Initiation of ICS

Our initial tests of change targeted the timing of when an ICS was ordered during hospitalization for an asthma exacerbation. Providers were instructed to delay ordering an ICS until the patient’s albuterol treatments were spaced to every three hours and to include a standardized communication prompt within the albuterol order. The prompt instructed the RT to contact the provider once the patient’s albuterol treatments were spaced to every three hours and ask for an ICS order, if appropriate. This intervention was abandoned because it did not reliably occur.

The subsequent intervention delayed the start of ICS treatment by using a PRN indication advising that the ICS was to be administered once the patient’s albuterol treatments were spaced to every three hours. However, after an error resulted in the PRN indication being included on a discharge prescription for an ICS, the PRN indication was abandoned. Subsequent work to develop a standardized process for delayed initiation of ICS occurred as part of the workflow to address the confirmation of outpatient formulary coverage as described next.

PDSA 3: Prioritize the Use of the Institution’s Outpatient Pharmacy

Medication changes that occurred because of outpatient insurance formulary denials were a unique challenge; they required a medication change after the discharge treatment plan had been finalized, and a prescription was already submitted to the outpatient pharmacy. In addition, neither our inpatient electronic medical record nor our inpatient hospital pharmacy has access to decision support tools that incorporate outpatient prescription formulary coverage. Alternatively, outpatient pharmacies have a standard workflow that routinely confirms insurance coverage before dispensing medication. The institutional policy was modified to allow for the inpatient administration of patient-supplied medications, pursuant to an inpatient order. Patient-supplied medications include those brought from home or those supplied by the outpatient pharmacy.

Subsequently, we developed a standardized process to confirm outpatient prescription drug coverage by using our hospital-based outpatient pharmacy to dispense ICS for inpatient treatment and asthma education. This new workflow included placing an order for an ICS at admission as a patient-supplied medication with an administration comment to “please administer once available from the outpatient pharmacy” (Figure 1). Then, once the discharge medication plan is finalized, the prescription is submitted to the outpatient pharmacy. Following verification of insurance coverage, the outpatient pharmacy dispenses the ICS, allowing it to be used for patient education and inpatient administration. If the patient is ineligible to have their prescription filled by the outpatient pharmacy for reasons other than formulary coverage, the ICS is dispensed from the hospital inpatient pharmacy as per the previous standard workflow. Inpatient ICS inhalers are then relabeled for home use per the existing practice to support medications-in-hand.

Further workflow improvements occurred following the development of an algorithm to help the outpatient pharmacy contact the correct inpatient team, and augmentation of the medication delivery process included notification of the RT when the ICS was delivered to the inpatient unit.

PDSA 4: Prescriber Education

Prescribers received education regarding PDSA interventions before testing and throughout the improvement cycle. Education sessions included informal coaching by the Asthma Education Coordinator, e-mail reminders containing screenshots of the ordering process, and formal didactic sessions for ordering providers. The Asthma Education Coordinator also provided education to the nursing and respiratory therapy staff regarding the implemented process and workflow changes.

PDSA 5: Real-Time Failure Notification

To supplement education for the complicated process change, the improvement team utilized a decision support tool (Vigilanz Corp., Chicago, IL) linked to EMR data to provide notification of real-time process failures. When a patient with an admission diagnosis of asthma had a prescription for an ICS verified and dispensed by the inpatient pharmacy, an automated message with relevant patient information would be sent to a member of the improvement team. Following a brief chart review, directed feedback could be offered to the ordering provider, allowing the prescription to be redirected to the outpatient pharmacy.

Study of the Improvement

Patients of all ages, with the International Classification of Diseases, Ninth Revision, and Tenth Revision codes for asthma (493.xx or J45.xx) were included in data collection and analysis if they were treated by the Hospital Medicine service, as the first inpatient service or after transfer from the ICU, and prescribed an ICS with or without a long-acting beta-agonist. Data were collected retrospectively and aggregated monthly. The baseline period was from January 2015 through October 2016. The intervention period was from November 2016 through March 2018. The prolonged baseline and study periods were utilized to understand the seasonal nature of asthma exacerbations.

Measures

Our primary outcome measure was defined as the monthly number of patients admitted to Hospital Medicine for an acute asthma exacerbation administered more than one ICS divided by the total number of asthma patients administered at least one dose of an ICS (patient-supplied or dispensed from the inpatient pharmacy). A full list of ICS is included in the appendix Table.

A secondary process measure approximated our adherence to obtaining ICS from the outpatient pharmacy for inpatient use. All medications administered during hospitalization are documented in the medication administration report. However, only medications dispensed from the inpatient pharmacy are associated with a patient charge. Patient-supplied medications, including those dispensed from the hospital outpatient pharmacy, are not associated with an inpatient charge. Therefore, the secondary process measure was defined as the monthly number of asthma patients administered an ICS not associated with an inpatient charge divided by the total number of asthma patients administered an ICS.

A cost outcome measure was developed to track changes in the average cost of an ICS included on inpatient bills during hospitalization for an asthma exacerbation. This outcome measure was defined as the total monthly cost, using the average wholesale price, of the ICS included on the inpatient bill for an asthma exacerbation, divided by the total number of asthma patients administered at least one dose of an ICS (patient supplied or dispensed from the inpatient pharmacy). The costs of patient-supplied medications (including those dispensed from the outpatient pharmacy) are not included on inpatient hospital bills or this secondary outcome measure.

Our a priori intent was to reduce ICS medication waste while maintaining a highly reliable system that included inpatient administration and education with ICS devices and maintain our medications-in-hand practice. A balancing measure was developed to monitor the reliability of inpatient administration of ICS. It was defined as the monthly number of patients who received a discharge prescription for an ICS and were administered an ICS while an inpatient divided by the total number of asthma patients with a discharge prescription for an ICS.

Analysis

Measures were evaluated using statistical process control charts and special cause variation was determined by previously established rules. Our primary, secondary, and balancing measures were all evaluated using a p-chart with variable subgroup size. The cost outcome measure was evaluated using an X-bar S control chart.^11-13

Primary Outcome Measure

During the baseline period, 7.4% of patients admitted to Hospital Medicine for an acute asthma exacerbation were administered more than one ICS, ranging from 0%-20% of patients per month (Figure 2). Following the start of our interventions, we met criteria for special cause allowing adjustment of the centerline.¹³ The mean percentage of patients receiving more than one ICS decreased from 7.4% to 0.7%. Figure 2 includes the n-value displayed each month and represents all patients admitted to the Hospital Medicine service with an asthma exacerbation who were administered at least one ICS.

Secondary Process Measure

During the baseline period, there were only rare occurrences (less than 1%) of a patient-supplied ICS being administered during an asthma admission. Following the start of our intervention period, the frequency of inpatient administration of patient-supplied ICS showed a rapid increase and met rules for special cause with an increase in the mean percent from 0.7% to 50% (Figure 3). The n-value displayed each month represents all patients admitted to the Hospital Medicine service for an asthma exacerbation administered at least one ICS.

Cost Outcome Measure

The average cost of an ICS billed during hospitalization for an acute asthma exacerbation was $236.57 per ICS during the baseline period. After the intervention period, the average inpatient cost for ICS decreased by 62% to $90.25 per ICS (Figure 4).

Balancing Measure

Our team reduced the monthly percent of children hospitalized with an acute asthma exacerbation administered more than one ICS from 7.4% to 0.7% after implementation of a new workflow process for ordering ICS utilizing the hospital-based outpatient pharmacy. The new workflow delayed ordering and administration of the initial inpatient ICS treatment, allowing time to consider a step-up in therapy. The brief delay in initiating ICS is not expected to have clinical consequence given the concomitant treatment with systemic corticosteroids. In addition, the outpatient pharmacy was utilized to verify insurance coverage reliably prior to dispensing ICS, reducing medication waste, and discharge delays due to outpatient medication formulary conflicts.

Our hospital’s previous approach to inpatient asthma care resulted in a highly reliable process to ensure patients were discharged with medications-in-hand as part of a broader system that effectively decreased reutilization. However, the previous process inadvertently resulted in medication waste. This waste included nearly full inhalers being discarded, additional work by the healthcare team (ordering providers, pharmacists, and RTs), and unnecessary patient charges.

While the primary driver of our decision to use the outpatient pharmacy was to adjudicate insurance prescription coverage reliably to prevent waste, this change likely resulted in a financial benefit to patients. The average cost per asthma admission of an inpatient billed for ICS using the average wholesale price, decreased by 62% following our interventions. The decrease in cost was primarily driven by using patient-supplied medications, including prescriptions newly filled by the on-site outpatient pharmacy, whose costs were not captured in this measure. While our secondary measure may underestimate the total expense incurred by families for an ICS, families likely receive their medications at a lower cost from the outpatient pharmacy than if the ICS was provided by an inpatient pharmacy. The average wholesale price is not what families are charged or pay for medications, partly due to differences in overhead costs that result in inpatient pharmacies having significantly higher charges than outpatient pharmacies. In addition, the 6.7% absolute reduction of our primary measure resulted in direct savings by reducing inpatient medication waste. Our process results in 67 fewer wasted ICS devices ($15,960) per 1,000 admissions for asthma exacerbation, extrapolated using the average cost ($238.20, average wholesale price) of each ICS during the baseline period.

Our quality improvement study had several limitations. (1) The interventions occurred at a single center with an established culture that embraces quality improvement, which may limit the generalizability of the work. (2) Our process verified insurance coverage with a hospital-based outpatient pharmacy. Some ICS prescriptions continued to be dispensed from the inpatient pharmacy, limiting our ability to verify insurance coverage. Local factors, including regulatory restrictions and delivery requirements, may limit the generalizability of using an outpatient pharmacy in this manner. (3) We achieved our goal of decreasing medication waste, but our a priori goal was to maintain our commitment to our established practice of interactive patient education with an ICS device as well as medications-in-hand at time of discharge. Our balancing measure showed a decrease in the percent of patients with a discharge prescription for an ICS who also received an inpatient dose of that ICS. This implies a decreased fidelity in our previously established education protocols. We had postulated that this occurred when the patient-supplied medication arrived on the day of discharge, but not close to when the medication was scheduled on the medication administration report, preventing administration. However, this is not a direct measure of patients receiving medications-in-hand or interactive medication education. Both may have occurred without administration of the ICS. (4) Despite a hospital culture that embraces quality improvement, this project required a significant change in the workflow that required considerable education at the time of implementation to integrate the new process reliably. However, once the process was in place, we have been able to sustain our improvement with limited educational investment.

Implementation of a new process for ordering ICS that emphasized delaying treatment until all necessary information was available and using an outpatient pharmacy to confirm insurance formulary coverage reduced the waste associated with more than one ICS being prescribed during a single admission.

Acknowledgments

The authors thank Sally Pope, MPH and Dr. Michael Carlisle, MD for their contribution to the quality improvement project. Thank you to Drs. Karen McDowell, MD and Carolyn Kercsmar, MD for advisement of our quality improvement project.

The authors appreciate the following individuals for their invaluable contributions. Dr. Hoefgen conceptualized and designed the study, was a member of the primary improvement team, carried out initial analysis, drafted the initial manuscript, and reviewed and revised the manuscript. Drs. Jones and Torres Garcia, and Mr. Hare were members of the primary improvement team who contributed to the design of the quality improvement study and interventions, ongoing data interpretation, and critically reviewed the manuscript. Dr. Courter contributed to the conceptualization and designed the study, was a member of the primary improvement team, designed data collection instruments, and critically reviewed and revised the manuscript. Dr. Simmons conceptualized and designed the study, critically reviewed the manuscript for important intellectual content, and reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Disclaimer

The information or content and conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsements be inferred by the BHPR, HRSA, DHHS, or the U.S. Government.

Asthma results in approximately 125,000 hospitalizations for children annually in the United States.^1,2 The National Heart, Lung, and Blood Institute guidelines recommend that children with persistent asthma be treated with a daily controller medication, ie, an inhaled corticosteroid (ICS).³ Hospitalization for an asthma exacerbation provides an opportunity to optimize daily controller medications and improve disease self-management by providing access to medications and teaching appropriate use of complicated inhalation devices.

To reduce readmission⁴ by mitigating low rates of postdischarge filling of ICS prescriptions,^5,6 a strategy of “meds-in-hand” was implemented at discharge. “Meds-in-hand” mitigates medication access as a barrier to adherence by ensuring that patients are discharged from the hospital with all required medications in hand, removing any barriers to filling their initial prescriptions.⁷ The Asthma Improvement Collaborative at Cincinnati Children’s Hospital Medical Center (CCHMC) previously applied quality improvement methodology to implement “meds-in-hand” as a key intervention in a broad strategy that successfully reduced asthma-specific utilization for the 30-day period following an asthma-related hospitalization of publicly insured children from 12% to 7%.^8,9

At the onset of the work described in this manuscript, children hospitalized with an acute exacerbation of persistent asthma were most often treated with an ICS while inpatients in addition to a standard short course of oral systemic corticosteroids. Conceptually, inpatient administration of ICS provided the opportunity to teach effective device usage with each inpatient administration and to reinforce daily use of the ICS as part of the patient’s daily home medication regimen. However, a proportion of patients admitted for an asthma exacerbation were noted to receive more than one ICS inhaler during their admission, most commonly due to a change in dose or type of ICS. When this occurred, the initially dispensed inhaler was discarded despite weeks of potential doses remaining. While some hospitals preferentially dispense ICS devices marketed to institutions with fewer doses per device, our pharmacy primarily dispensed ICS devices identical to retail locations containing at least a one-month supply of medication. In addition to the wasted medication, this practice resulted in additional work by healthcare staff, unnecessary patient charges, and potentially contributed to confusion about the discharge medication regimen.

Our specific aim for this quality improvement study was to reduce the monthly percentage of admissions for an acute asthma exacerbation treated with >1 ICS from 7% to 4% over a six-month period.

Context

CCHMC is a quaternary care pediatric health system with more than 600 inpatient beds and 800-900 inpatient admissions per year for acute asthma exacerbation. The Hospital Medicine service cares for patients with asthma on five clinical teams across two different campuses. Care teams are supervised by an attending physician and may include residents, fellows, or nurse practitioners. Patients hospitalized for an acute asthma exacerbation may receive a consult from the Asthma Center consult team, staffed by faculty from either the Pediatric Pulmonology or Allergy/Immunology divisions. Respiratory therapists (RTs) administer inhaled medications and provide asthma education.

Planning the Intervention

Our improvement team included physicians from Hospital Medicine and Pulmonary Medicine, an Asthma Education Coordinator, a Clinical Pharmacist, a Pediatric Chief Resident, and a clinical research coordinator. Initial interventions targeted a single resident team at the main campus before spreading improvement activities to all resident teams at the main campus and then the satellite campus by February 2017.

Development of our process map (Figure 1) revealed that the decision for ordering inpatient ICS treatment frequently occurred at admission. Subsequently, the care team or consulting team might make a change in the ICS to fine-tune the outpatient medication regimen given that admission for asthma often results from suboptimal chronic symptom control. Baseline analysis of changes in ICS orders revealed that 81% of ICS changes were associated with a step-up in therapy, defined as an increase in the daily dose of the ICS or the addition of a long-acting beta-agonist. The other common ICS adjustment, accounting for 17%, was a change in corticosteroid without a step-up in therapy, (ie, beclomethasone to fluticasone) that typically occurred near the end of the hospitalization to accommodate outpatient insurance formularies, independent of patient factors related to illness severity.

Improvement Activities

We conceived key drivers or domains that would be necessary to address to effect change. Key drivers included a standardized process for delayed initiation of ICS and confirmation of outpatient insurance prescription drug coverage, prescriber education, and real-time failure notification.

PDSA Interventions

PDSA 1 & 2: Standardized Process for Initiation of ICS

Our initial tests of change targeted the timing of when an ICS was ordered during hospitalization for an asthma exacerbation. Providers were instructed to delay ordering an ICS until the patient’s albuterol treatments were spaced to every three hours and to include a standardized communication prompt within the albuterol order. The prompt instructed the RT to contact the provider once the patient’s albuterol treatments were spaced to every three hours and ask for an ICS order, if appropriate. This intervention was abandoned because it did not reliably occur.

The subsequent intervention delayed the start of ICS treatment by using a PRN indication advising that the ICS was to be administered once the patient’s albuterol treatments were spaced to every three hours. However, after an error resulted in the PRN indication being included on a discharge prescription for an ICS, the PRN indication was abandoned. Subsequent work to develop a standardized process for delayed initiation of ICS occurred as part of the workflow to address the confirmation of outpatient formulary coverage as described next.

PDSA 3: Prioritize the Use of the Institution’s Outpatient Pharmacy

Medication changes that occurred because of outpatient insurance formulary denials were a unique challenge; they required a medication change after the discharge treatment plan had been finalized, and a prescription was already submitted to the outpatient pharmacy. In addition, neither our inpatient electronic medical record nor our inpatient hospital pharmacy has access to decision support tools that incorporate outpatient prescription formulary coverage. Alternatively, outpatient pharmacies have a standard workflow that routinely confirms insurance coverage before dispensing medication. The institutional policy was modified to allow for the inpatient administration of patient-supplied medications, pursuant to an inpatient order. Patient-supplied medications include those brought from home or those supplied by the outpatient pharmacy.

Subsequently, we developed a standardized process to confirm outpatient prescription drug coverage by using our hospital-based outpatient pharmacy to dispense ICS for inpatient treatment and asthma education. This new workflow included placing an order for an ICS at admission as a patient-supplied medication with an administration comment to “please administer once available from the outpatient pharmacy” (Figure 1). Then, once the discharge medication plan is finalized, the prescription is submitted to the outpatient pharmacy. Following verification of insurance coverage, the outpatient pharmacy dispenses the ICS, allowing it to be used for patient education and inpatient administration. If the patient is ineligible to have their prescription filled by the outpatient pharmacy for reasons other than formulary coverage, the ICS is dispensed from the hospital inpatient pharmacy as per the previous standard workflow. Inpatient ICS inhalers are then relabeled for home use per the existing practice to support medications-in-hand.

Further workflow improvements occurred following the development of an algorithm to help the outpatient pharmacy contact the correct inpatient team, and augmentation of the medication delivery process included notification of the RT when the ICS was delivered to the inpatient unit.

PDSA 4: Prescriber Education

Prescribers received education regarding PDSA interventions before testing and throughout the improvement cycle. Education sessions included informal coaching by the Asthma Education Coordinator, e-mail reminders containing screenshots of the ordering process, and formal didactic sessions for ordering providers. The Asthma Education Coordinator also provided education to the nursing and respiratory therapy staff regarding the implemented process and workflow changes.

PDSA 5: Real-Time Failure Notification

To supplement education for the complicated process change, the improvement team utilized a decision support tool (Vigilanz Corp., Chicago, IL) linked to EMR data to provide notification of real-time process failures. When a patient with an admission diagnosis of asthma had a prescription for an ICS verified and dispensed by the inpatient pharmacy, an automated message with relevant patient information would be sent to a member of the improvement team. Following a brief chart review, directed feedback could be offered to the ordering provider, allowing the prescription to be redirected to the outpatient pharmacy.

Study of the Improvement

Patients of all ages, with the International Classification of Diseases, Ninth Revision, and Tenth Revision codes for asthma (493.xx or J45.xx) were included in data collection and analysis if they were treated by the Hospital Medicine service, as the first inpatient service or after transfer from the ICU, and prescribed an ICS with or without a long-acting beta-agonist. Data were collected retrospectively and aggregated monthly. The baseline period was from January 2015 through October 2016. The intervention period was from November 2016 through March 2018. The prolonged baseline and study periods were utilized to understand the seasonal nature of asthma exacerbations.

Measures

Our primary outcome measure was defined as the monthly number of patients admitted to Hospital Medicine for an acute asthma exacerbation administered more than one ICS divided by the total number of asthma patients administered at least one dose of an ICS (patient-supplied or dispensed from the inpatient pharmacy). A full list of ICS is included in the appendix Table.

A secondary process measure approximated our adherence to obtaining ICS from the outpatient pharmacy for inpatient use. All medications administered during hospitalization are documented in the medication administration report. However, only medications dispensed from the inpatient pharmacy are associated with a patient charge. Patient-supplied medications, including those dispensed from the hospital outpatient pharmacy, are not associated with an inpatient charge. Therefore, the secondary process measure was defined as the monthly number of asthma patients administered an ICS not associated with an inpatient charge divided by the total number of asthma patients administered an ICS.

A cost outcome measure was developed to track changes in the average cost of an ICS included on inpatient bills during hospitalization for an asthma exacerbation. This outcome measure was defined as the total monthly cost, using the average wholesale price, of the ICS included on the inpatient bill for an asthma exacerbation, divided by the total number of asthma patients administered at least one dose of an ICS (patient supplied or dispensed from the inpatient pharmacy). The costs of patient-supplied medications (including those dispensed from the outpatient pharmacy) are not included on inpatient hospital bills or this secondary outcome measure.

Our a priori intent was to reduce ICS medication waste while maintaining a highly reliable system that included inpatient administration and education with ICS devices and maintain our medications-in-hand practice. A balancing measure was developed to monitor the reliability of inpatient administration of ICS. It was defined as the monthly number of patients who received a discharge prescription for an ICS and were administered an ICS while an inpatient divided by the total number of asthma patients with a discharge prescription for an ICS.

Analysis

Measures were evaluated using statistical process control charts and special cause variation was determined by previously established rules. Our primary, secondary, and balancing measures were all evaluated using a p-chart with variable subgroup size. The cost outcome measure was evaluated using an X-bar S control chart.^11-13

Primary Outcome Measure

During the baseline period, 7.4% of patients admitted to Hospital Medicine for an acute asthma exacerbation were administered more than one ICS, ranging from 0%-20% of patients per month (Figure 2). Following the start of our interventions, we met criteria for special cause allowing adjustment of the centerline.¹³ The mean percentage of patients receiving more than one ICS decreased from 7.4% to 0.7%. Figure 2 includes the n-value displayed each month and represents all patients admitted to the Hospital Medicine service with an asthma exacerbation who were administered at least one ICS.

Secondary Process Measure

During the baseline period, there were only rare occurrences (less than 1%) of a patient-supplied ICS being administered during an asthma admission. Following the start of our intervention period, the frequency of inpatient administration of patient-supplied ICS showed a rapid increase and met rules for special cause with an increase in the mean percent from 0.7% to 50% (Figure 3). The n-value displayed each month represents all patients admitted to the Hospital Medicine service for an asthma exacerbation administered at least one ICS.

Cost Outcome Measure

The average cost of an ICS billed during hospitalization for an acute asthma exacerbation was $236.57 per ICS during the baseline period. After the intervention period, the average inpatient cost for ICS decreased by 62% to $90.25 per ICS (Figure 4).

Balancing Measure

Our team reduced the monthly percent of children hospitalized with an acute asthma exacerbation administered more than one ICS from 7.4% to 0.7% after implementation of a new workflow process for ordering ICS utilizing the hospital-based outpatient pharmacy. The new workflow delayed ordering and administration of the initial inpatient ICS treatment, allowing time to consider a step-up in therapy. The brief delay in initiating ICS is not expected to have clinical consequence given the concomitant treatment with systemic corticosteroids. In addition, the outpatient pharmacy was utilized to verify insurance coverage reliably prior to dispensing ICS, reducing medication waste, and discharge delays due to outpatient medication formulary conflicts.

Our hospital’s previous approach to inpatient asthma care resulted in a highly reliable process to ensure patients were discharged with medications-in-hand as part of a broader system that effectively decreased reutilization. However, the previous process inadvertently resulted in medication waste. This waste included nearly full inhalers being discarded, additional work by the healthcare team (ordering providers, pharmacists, and RTs), and unnecessary patient charges.

While the primary driver of our decision to use the outpatient pharmacy was to adjudicate insurance prescription coverage reliably to prevent waste, this change likely resulted in a financial benefit to patients. The average cost per asthma admission of an inpatient billed for ICS using the average wholesale price, decreased by 62% following our interventions. The decrease in cost was primarily driven by using patient-supplied medications, including prescriptions newly filled by the on-site outpatient pharmacy, whose costs were not captured in this measure. While our secondary measure may underestimate the total expense incurred by families for an ICS, families likely receive their medications at a lower cost from the outpatient pharmacy than if the ICS was provided by an inpatient pharmacy. The average wholesale price is not what families are charged or pay for medications, partly due to differences in overhead costs that result in inpatient pharmacies having significantly higher charges than outpatient pharmacies. In addition, the 6.7% absolute reduction of our primary measure resulted in direct savings by reducing inpatient medication waste. Our process results in 67 fewer wasted ICS devices ($15,960) per 1,000 admissions for asthma exacerbation, extrapolated using the average cost ($238.20, average wholesale price) of each ICS during the baseline period.

Our quality improvement study had several limitations. (1) The interventions occurred at a single center with an established culture that embraces quality improvement, which may limit the generalizability of the work. (2) Our process verified insurance coverage with a hospital-based outpatient pharmacy. Some ICS prescriptions continued to be dispensed from the inpatient pharmacy, limiting our ability to verify insurance coverage. Local factors, including regulatory restrictions and delivery requirements, may limit the generalizability of using an outpatient pharmacy in this manner. (3) We achieved our goal of decreasing medication waste, but our a priori goal was to maintain our commitment to our established practice of interactive patient education with an ICS device as well as medications-in-hand at time of discharge. Our balancing measure showed a decrease in the percent of patients with a discharge prescription for an ICS who also received an inpatient dose of that ICS. This implies a decreased fidelity in our previously established education protocols. We had postulated that this occurred when the patient-supplied medication arrived on the day of discharge, but not close to when the medication was scheduled on the medication administration report, preventing administration. However, this is not a direct measure of patients receiving medications-in-hand or interactive medication education. Both may have occurred without administration of the ICS. (4) Despite a hospital culture that embraces quality improvement, this project required a significant change in the workflow that required considerable education at the time of implementation to integrate the new process reliably. However, once the process was in place, we have been able to sustain our improvement with limited educational investment.

Implementation of a new process for ordering ICS that emphasized delaying treatment until all necessary information was available and using an outpatient pharmacy to confirm insurance formulary coverage reduced the waste associated with more than one ICS being prescribed during a single admission.

Acknowledgments

The authors thank Sally Pope, MPH and Dr. Michael Carlisle, MD for their contribution to the quality improvement project. Thank you to Drs. Karen McDowell, MD and Carolyn Kercsmar, MD for advisement of our quality improvement project.

The authors appreciate the following individuals for their invaluable contributions. Dr. Hoefgen conceptualized and designed the study, was a member of the primary improvement team, carried out initial analysis, drafted the initial manuscript, and reviewed and revised the manuscript. Drs. Jones and Torres Garcia, and Mr. Hare were members of the primary improvement team who contributed to the design of the quality improvement study and interventions, ongoing data interpretation, and critically reviewed the manuscript. Dr. Courter contributed to the conceptualization and designed the study, was a member of the primary improvement team, designed data collection instruments, and critically reviewed and revised the manuscript. Dr. Simmons conceptualized and designed the study, critically reviewed the manuscript for important intellectual content, and reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Disclaimer

The information or content and conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsements be inferred by the BHPR, HRSA, DHHS, or the U.S. Government.

Hip fractures in the elderly are associated with significant morbidity and mortality.¹ These are typically fragility fractures since they are caused by mechanical forces that would ordinarily not result in a serious injury, such as a fall from or below standing level. The incidence of hip fractures in the United States is expected to increase as the population ages; estimates project 512,000 hip fractures with an associated cost of $16 billion annually by the year 2040.² Timely surgery is recommended for hip fracture patients as delayed surgery beyond 24 to 48 hours of presentation is associated with increased morbidity and mortality.^3-6 Time to surgery (TTS) has been shown to be the major potentially modifiable risk factor in the management of a hip fracture.⁷

Factors that have been noted to influence TTS include the American Society of Anesthesiologists’ (ASA) score, the day of the week of hospital admission, and preoperative testing.^8,9 Preoperative cardiology consultation and subsequent cardiac testing, in particular, can increase the TTS and length of stay (LOS) without changing perioperative management.^9,10 In our review of literature, we could not identify any studies specifically looking at the impact of preoperative specialty consults on short-term mortality or comparison of care provided by hospitalists alone versus additionally involving subspecialists such as cardiologists. To our knowledge, there are no studies that have categorized recommendations from a preoperative specialty consult as minor, moderate, or major.

Our study evaluated whether preoperative specialty consults meaningfully change management and influence outcomes for hip fracture patients. At our institution, all hip fracture patients are admitted to the hospitalist service and comanaged with the orthopedic team. The hospitalist physician performs the preoperative evaluation as part of the admission history and physical exam. Preoperative specialty consult(s), if needed, are requested only by the hospitalist team. A consultant such as a cardiologist provides input; however, final management decisions are coordinated by the hospitalist physician.

Study Design

We performed a retrospective cohort study of patients aged 50 years and older who underwent surgery for an isolated fragility fracture of the hip at Hartford Hospital, a level one trauma and tertiary care medical center, within the 24-month period from April 2015 to March 2017. Fragility hip fracture is defined as one occurring from a fall of a height of standing or less. A consult referred to a specialty or subspecialty consultation, other than hospital medicine, obtained prior to surgery. Patients with additional skeletal trauma and periprosthetic fractures were excluded. A total of 491 unique patients met the inclusion criteria, and data were obtained from chart review and an orthopedic surgery registry. The Hartford Hospital Institutional Review Board approved this study.

Our primary predictor was the presence or absence of a preoperative specialty consultation requested by the hospitalist. We also analyzed the following: covariates of demographics (age, sex, race), the ASA score, and severity of comorbidities using the Charlson comorbidity index (CCI) with a Quan modification;¹¹ “R program package, International Classification of Disease (ICD)”¹² was used to calculate the CCI using ICD-9 and ICD-10 diagnostic codes.

The primary outcome measures were TTS (measured in hours), LOS (measured in days), complications, and preoperative specialty consult resulting in a change in perioperative management. TTS was defined as the time elapsed from the presentation at the emergency department (ED) to surgery start. For transfer or direct admission patients, the time of admission was used in place of time of presentation. The measured complications included postoperative venous thromboembolic events, surgical site infection, myocardial infarction, stroke, and sepsis. Secondary outcome measures included 30-day mortality, readmission rate, and rate of return to OR. There were no elective or planned readmissions postoperatively on review of our institution’s orthopedic surgery registry.

Our team performed an extensive chart review including reviewing the admission note, consulting physician notes, and relevant test results. Our senior investigator (MK) then rated each preoperative specialty consult on appropriateness, the relative strength of the consultant’s recommendation, and resulting change in perioperative management. Cardiology consultations were deemed reasonable if a patient’s cardiac risk was considered elevated by the admitting physician or an active cardiac condition was present (suggestion of or clear evidence for acute coronary syndrome, acute congestive heart failure, uncontrolled arrhythmia, or symptomatic valvular disease). The determination of “elevated cardiac risk” was made, if admit note contained verbiage expressing concern for further evaluation for cardiac issues or words such as “high risk” or “elevated risk”. A specific guideline-based score such as the revised cardiac risk index was not consistently available in this retrospective chart review. A noncardiology consult was deemed reasonable only if it would have been warranted for the specific clinical situation­—for example, a neurology consult for an acute stroke or a pulmonary consult for acute respiratory failure. Consult recommendations or outcomes were rated as minor, moderate, or major (see Table 1 for detailed criteria). Some consults may generate more than one recommendation, in these cases, we determined that a major recommendation supersedes a moderate or minor recommendation and only one was counted in the final analysis. Next we determined if a consult recommendation led to a change in perioperative or therapeutic management, defined as a medication or dosage change, need to delay surgery to stabilize an unstable medical condition, invasive procedures (such as thoracentesis or cardiac catheterization) or change in postoperative monitoring. As a way of clarification, a consult may have a minor recommendation such as an EKG but if no other recommendations were given and there was no change in therapeutic management such as a medication change, this would be considered as a “no change”.

Statistical Analysis

The main analyses compared the two patient subgroups (with or without preoperative specialty consults) around outcome measures. Primary outcome measures were TTS, LOS, complications, and consult resulting in a change in perioperative management. Secondary outcome measures were 30-day readmit, return to OR, and mortality. A preliminary analysis was conducted to explore distributions for TTS and LOS. As expected, none met the assumptions of normality and were thus analyzed with Wilcoxon ranked-sum tests. The other outcomes were dichotomous and analyzed with chi-square tests of proportion or Fisher’s exact test when the expected cell frequencies were too low. Dichotomized variables for TTS (within 24 hours and 48 hours) and LOS (within five days, the median LOS for this cohort) were calculated and subsequently analyzed with additional chi-square tests of proportion or Fisher’s exact test¹³. To explore the effect of preoperative specialty consults independent of potential confounders, logistic regression analyses predicting each of the dichotomous outcomes were conducted with age and CCI used as predictors in addition to the main variable of whether or not there was a preoperative specialty consult. Since the CCI and ASA scores were highly intercorrelated, only the former was chosen for the multivariate analyses based on the consistent algorithm used to calculate CCI.

Additional analyses with the subgroup of patients with a preoperative specialty consult explored whether the consult was reasonable, the relative strength of resulting recommendation and whether it resulted in a change in management. The statistical approach used was the same as for the other dichotomous outcomes. All analyses used 0.05 as the level of statistical significance; SPSSv21 (IBM, Armonk, New York) was the statistical software used.

The sample size for this retrospective analysis was determined by the available number of patients meeting the inclusion criteria. An a priori power calculation was done to determine if the expected volume would be sufficient for the multivariate analysis; the presence of a complication was selected for calculation. Based on an expected volume of approximately 500 and an estimate of a 10% serious complication rate, it was determined that the sample could support the analysis of up to five predictor variables, sufficient for the main variable and four potential confounders; this was considered adequate.¹⁴ Propensity scoring was considered but did not offer any advantages to logistic regression because we only had two observed covariates: CCI and age.

A total of 491 unique patients met our inclusion criteria, 177 patients had a preoperative specialty consult. Of these 177 patients, 24 patients had more than one consult; hence, the total number of consults was 201. Most of the consults were cardiology (159). Others were Infectious disease (11), Pulmonology (10), Neurology (7), and Miscellaneous (14, which included Nephrology, Gastroenterology, Hematology, and Oncology).

No significant differences were found between the consult and no-consult groups with respect to gender, race, body mass index, type of anesthesia, and day of the week of surgery. We did note that patients with a consult were older and had a significantly higher CCI and ASA score (Table 2).

Patients with preoperative specialty consults were older and had more comorbidities than patients without consults. Our findings suggest that consults contribute to delays to surgery and may lead to higher LOS and higher risk of 30-day readmission after controlling for age and comorbidities in a multivariate analysis. This observation is significant considering that consults were requested more frequently on patients with a higher comorbidity burden and included patients who did not get additional preoperative testing, suggesting that a delay from waiting for a consult alone may be deleterious. This was a unique observation in our study; prior studies examining this subject have attributed delays to additional testing and not consults alone. Even though most consult requests appear to be reasonable according to our criteria, the majority of recommendations were minor (72.9%), and 62.7% of consults resulted in no change in perioperative management. Major changes in perioperative management were noted in only 4.5% of patients.

Our finding that a majority of patients in the consult group had no significant change in perioperative management raises an important area of potential improvement in the care of hip fracture patients. We believe that narrowing indications for preoperative specialty consults may result in shorter TTS and LOS for this group of frail elderly patients without sacrificing the quality of care. Since all patients in our study were comanaged by hospitalists and patients without additional consults had similar or better outcomes, we believe that hospitalist physicians are well positioned to provide standardized comanagement to this patient group without additional consultation unless absolutely necessary.

The primary limitation of our study was that this was a retrospective case analysis. The designation of minor, moderate, or major recommendation was done after the consults were already completed, and it may not be possible to predict that a consult results in no change without it being actually performed. Additionally, our classification of recommendations is somewhat arbitrary and subjective; for example, some readers might argue that a medication change counts as a moderate recommendation. We rated a medication change to be minor as we believe that an experienced hospitalist may likely make such management decisions on their own, and if this is the only recommendation from a consult, it is not additional information critical to patient care. There may also be an “unmeasured complexity” noted by the admitting physician, which was not necessarily accounted for by multivariate analysis of age and CCI but one that led to higher mortality and readmissions. However, we feel that this “unmeasured complexity” is likely inconsequential as the vast majority of consults did not result in any change in management. We did adjust for covariates as noted, but some confounding by indication is likely to remain. Additionally, categorization of consult recommendations and consequent changes by one physician could be considered subjective. We did control for this by having another physician review the entire dataset and rate it independently for interrater reliability with excellent correlation and kappa, although these may be inflated to some degree because our chart review did not account for variability among chart extractors.

A prospective evaluation of a clinical protocol that delineates reasonable indications for a preoperative consult would be helpful to validate our findings. In our study, we noted moderate or major recommendations from a cardiologist only in cases where an active cardiac condition was suspected by the hospitalist requesting the consult; hence, limiting preoperative cardiology consults to active cardiac conditions may be a reasonable approach to evaluate in a prospective study.

In conclusion, a majority of preoperative specialty consults do not appear to meaningfully influence management and may indirectly increase morbidity by delaying surgery and extending hospital stays. Our data suggest that unless the patient is clinically unstable and likely to require active management by a consultant prior to hip fracture repair, consults may offer limited benefit. Appropriately standardized perioperative management of this patient group by hospitalist physicians appears to manage most hip fracture patients as effectively with faster TTS and shorter hospital LOS.

Acknowledgments

The authors would like to thank John Corradi, PhD (Research Department at Hartford Hospital) for his input in calculating the Charlson comorbidity index.^12,13

Hip fractures in the elderly are associated with significant morbidity and mortality.¹ These are typically fragility fractures since they are caused by mechanical forces that would ordinarily not result in a serious injury, such as a fall from or below standing level. The incidence of hip fractures in the United States is expected to increase as the population ages; estimates project 512,000 hip fractures with an associated cost of $16 billion annually by the year 2040.² Timely surgery is recommended for hip fracture patients as delayed surgery beyond 24 to 48 hours of presentation is associated with increased morbidity and mortality.^3-6 Time to surgery (TTS) has been shown to be the major potentially modifiable risk factor in the management of a hip fracture.⁷

Factors that have been noted to influence TTS include the American Society of Anesthesiologists’ (ASA) score, the day of the week of hospital admission, and preoperative testing.^8,9 Preoperative cardiology consultation and subsequent cardiac testing, in particular, can increase the TTS and length of stay (LOS) without changing perioperative management.^9,10 In our review of literature, we could not identify any studies specifically looking at the impact of preoperative specialty consults on short-term mortality or comparison of care provided by hospitalists alone versus additionally involving subspecialists such as cardiologists. To our knowledge, there are no studies that have categorized recommendations from a preoperative specialty consult as minor, moderate, or major.

Our study evaluated whether preoperative specialty consults meaningfully change management and influence outcomes for hip fracture patients. At our institution, all hip fracture patients are admitted to the hospitalist service and comanaged with the orthopedic team. The hospitalist physician performs the preoperative evaluation as part of the admission history and physical exam. Preoperative specialty consult(s), if needed, are requested only by the hospitalist team. A consultant such as a cardiologist provides input; however, final management decisions are coordinated by the hospitalist physician.

Study Design

We performed a retrospective cohort study of patients aged 50 years and older who underwent surgery for an isolated fragility fracture of the hip at Hartford Hospital, a level one trauma and tertiary care medical center, within the 24-month period from April 2015 to March 2017. Fragility hip fracture is defined as one occurring from a fall of a height of standing or less. A consult referred to a specialty or subspecialty consultation, other than hospital medicine, obtained prior to surgery. Patients with additional skeletal trauma and periprosthetic fractures were excluded. A total of 491 unique patients met the inclusion criteria, and data were obtained from chart review and an orthopedic surgery registry. The Hartford Hospital Institutional Review Board approved this study.

Our primary predictor was the presence or absence of a preoperative specialty consultation requested by the hospitalist. We also analyzed the following: covariates of demographics (age, sex, race), the ASA score, and severity of comorbidities using the Charlson comorbidity index (CCI) with a Quan modification;¹¹ “R program package, International Classification of Disease (ICD)”¹² was used to calculate the CCI using ICD-9 and ICD-10 diagnostic codes.

The primary outcome measures were TTS (measured in hours), LOS (measured in days), complications, and preoperative specialty consult resulting in a change in perioperative management. TTS was defined as the time elapsed from the presentation at the emergency department (ED) to surgery start. For transfer or direct admission patients, the time of admission was used in place of time of presentation. The measured complications included postoperative venous thromboembolic events, surgical site infection, myocardial infarction, stroke, and sepsis. Secondary outcome measures included 30-day mortality, readmission rate, and rate of return to OR. There were no elective or planned readmissions postoperatively on review of our institution’s orthopedic surgery registry.

Our team performed an extensive chart review including reviewing the admission note, consulting physician notes, and relevant test results. Our senior investigator (MK) then rated each preoperative specialty consult on appropriateness, the relative strength of the consultant’s recommendation, and resulting change in perioperative management. Cardiology consultations were deemed reasonable if a patient’s cardiac risk was considered elevated by the admitting physician or an active cardiac condition was present (suggestion of or clear evidence for acute coronary syndrome, acute congestive heart failure, uncontrolled arrhythmia, or symptomatic valvular disease). The determination of “elevated cardiac risk” was made, if admit note contained verbiage expressing concern for further evaluation for cardiac issues or words such as “high risk” or “elevated risk”. A specific guideline-based score such as the revised cardiac risk index was not consistently available in this retrospective chart review. A noncardiology consult was deemed reasonable only if it would have been warranted for the specific clinical situation­—for example, a neurology consult for an acute stroke or a pulmonary consult for acute respiratory failure. Consult recommendations or outcomes were rated as minor, moderate, or major (see Table 1 for detailed criteria). Some consults may generate more than one recommendation, in these cases, we determined that a major recommendation supersedes a moderate or minor recommendation and only one was counted in the final analysis. Next we determined if a consult recommendation led to a change in perioperative or therapeutic management, defined as a medication or dosage change, need to delay surgery to stabilize an unstable medical condition, invasive procedures (such as thoracentesis or cardiac catheterization) or change in postoperative monitoring. As a way of clarification, a consult may have a minor recommendation such as an EKG but if no other recommendations were given and there was no change in therapeutic management such as a medication change, this would be considered as a “no change”.

Statistical Analysis

The main analyses compared the two patient subgroups (with or without preoperative specialty consults) around outcome measures. Primary outcome measures were TTS, LOS, complications, and consult resulting in a change in perioperative management. Secondary outcome measures were 30-day readmit, return to OR, and mortality. A preliminary analysis was conducted to explore distributions for TTS and LOS. As expected, none met the assumptions of normality and were thus analyzed with Wilcoxon ranked-sum tests. The other outcomes were dichotomous and analyzed with chi-square tests of proportion or Fisher’s exact test when the expected cell frequencies were too low. Dichotomized variables for TTS (within 24 hours and 48 hours) and LOS (within five days, the median LOS for this cohort) were calculated and subsequently analyzed with additional chi-square tests of proportion or Fisher’s exact test¹³. To explore the effect of preoperative specialty consults independent of potential confounders, logistic regression analyses predicting each of the dichotomous outcomes were conducted with age and CCI used as predictors in addition to the main variable of whether or not there was a preoperative specialty consult. Since the CCI and ASA scores were highly intercorrelated, only the former was chosen for the multivariate analyses based on the consistent algorithm used to calculate CCI.

Additional analyses with the subgroup of patients with a preoperative specialty consult explored whether the consult was reasonable, the relative strength of resulting recommendation and whether it resulted in a change in management. The statistical approach used was the same as for the other dichotomous outcomes. All analyses used 0.05 as the level of statistical significance; SPSSv21 (IBM, Armonk, New York) was the statistical software used.

The sample size for this retrospective analysis was determined by the available number of patients meeting the inclusion criteria. An a priori power calculation was done to determine if the expected volume would be sufficient for the multivariate analysis; the presence of a complication was selected for calculation. Based on an expected volume of approximately 500 and an estimate of a 10% serious complication rate, it was determined that the sample could support the analysis of up to five predictor variables, sufficient for the main variable and four potential confounders; this was considered adequate.¹⁴ Propensity scoring was considered but did not offer any advantages to logistic regression because we only had two observed covariates: CCI and age.

A total of 491 unique patients met our inclusion criteria, 177 patients had a preoperative specialty consult. Of these 177 patients, 24 patients had more than one consult; hence, the total number of consults was 201. Most of the consults were cardiology (159). Others were Infectious disease (11), Pulmonology (10), Neurology (7), and Miscellaneous (14, which included Nephrology, Gastroenterology, Hematology, and Oncology).

No significant differences were found between the consult and no-consult groups with respect to gender, race, body mass index, type of anesthesia, and day of the week of surgery. We did note that patients with a consult were older and had a significantly higher CCI and ASA score (Table 2).

Patients with preoperative specialty consults were older and had more comorbidities than patients without consults. Our findings suggest that consults contribute to delays to surgery and may lead to higher LOS and higher risk of 30-day readmission after controlling for age and comorbidities in a multivariate analysis. This observation is significant considering that consults were requested more frequently on patients with a higher comorbidity burden and included patients who did not get additional preoperative testing, suggesting that a delay from waiting for a consult alone may be deleterious. This was a unique observation in our study; prior studies examining this subject have attributed delays to additional testing and not consults alone. Even though most consult requests appear to be reasonable according to our criteria, the majority of recommendations were minor (72.9%), and 62.7% of consults resulted in no change in perioperative management. Major changes in perioperative management were noted in only 4.5% of patients.

Our finding that a majority of patients in the consult group had no significant change in perioperative management raises an important area of potential improvement in the care of hip fracture patients. We believe that narrowing indications for preoperative specialty consults may result in shorter TTS and LOS for this group of frail elderly patients without sacrificing the quality of care. Since all patients in our study were comanaged by hospitalists and patients without additional consults had similar or better outcomes, we believe that hospitalist physicians are well positioned to provide standardized comanagement to this patient group without additional consultation unless absolutely necessary.

The primary limitation of our study was that this was a retrospective case analysis. The designation of minor, moderate, or major recommendation was done after the consults were already completed, and it may not be possible to predict that a consult results in no change without it being actually performed. Additionally, our classification of recommendations is somewhat arbitrary and subjective; for example, some readers might argue that a medication change counts as a moderate recommendation. We rated a medication change to be minor as we believe that an experienced hospitalist may likely make such management decisions on their own, and if this is the only recommendation from a consult, it is not additional information critical to patient care. There may also be an “unmeasured complexity” noted by the admitting physician, which was not necessarily accounted for by multivariate analysis of age and CCI but one that led to higher mortality and readmissions. However, we feel that this “unmeasured complexity” is likely inconsequential as the vast majority of consults did not result in any change in management. We did adjust for covariates as noted, but some confounding by indication is likely to remain. Additionally, categorization of consult recommendations and consequent changes by one physician could be considered subjective. We did control for this by having another physician review the entire dataset and rate it independently for interrater reliability with excellent correlation and kappa, although these may be inflated to some degree because our chart review did not account for variability among chart extractors.

A prospective evaluation of a clinical protocol that delineates reasonable indications for a preoperative consult would be helpful to validate our findings. In our study, we noted moderate or major recommendations from a cardiologist only in cases where an active cardiac condition was suspected by the hospitalist requesting the consult; hence, limiting preoperative cardiology consults to active cardiac conditions may be a reasonable approach to evaluate in a prospective study.

In conclusion, a majority of preoperative specialty consults do not appear to meaningfully influence management and may indirectly increase morbidity by delaying surgery and extending hospital stays. Our data suggest that unless the patient is clinically unstable and likely to require active management by a consultant prior to hip fracture repair, consults may offer limited benefit. Appropriately standardized perioperative management of this patient group by hospitalist physicians appears to manage most hip fracture patients as effectively with faster TTS and shorter hospital LOS.

Acknowledgments

The authors would like to thank John Corradi, PhD (Research Department at Hartford Hospital) for his input in calculating the Charlson comorbidity index.^12,13

Hip fractures in the elderly are associated with significant morbidity and mortality.¹ These are typically fragility fractures since they are caused by mechanical forces that would ordinarily not result in a serious injury, such as a fall from or below standing level. The incidence of hip fractures in the United States is expected to increase as the population ages; estimates project 512,000 hip fractures with an associated cost of $16 billion annually by the year 2040.² Timely surgery is recommended for hip fracture patients as delayed surgery beyond 24 to 48 hours of presentation is associated with increased morbidity and mortality.^3-6 Time to surgery (TTS) has been shown to be the major potentially modifiable risk factor in the management of a hip fracture.⁷

Factors that have been noted to influence TTS include the American Society of Anesthesiologists’ (ASA) score, the day of the week of hospital admission, and preoperative testing.^8,9 Preoperative cardiology consultation and subsequent cardiac testing, in particular, can increase the TTS and length of stay (LOS) without changing perioperative management.^9,10 In our review of literature, we could not identify any studies specifically looking at the impact of preoperative specialty consults on short-term mortality or comparison of care provided by hospitalists alone versus additionally involving subspecialists such as cardiologists. To our knowledge, there are no studies that have categorized recommendations from a preoperative specialty consult as minor, moderate, or major.

Our study evaluated whether preoperative specialty consults meaningfully change management and influence outcomes for hip fracture patients. At our institution, all hip fracture patients are admitted to the hospitalist service and comanaged with the orthopedic team. The hospitalist physician performs the preoperative evaluation as part of the admission history and physical exam. Preoperative specialty consult(s), if needed, are requested only by the hospitalist team. A consultant such as a cardiologist provides input; however, final management decisions are coordinated by the hospitalist physician.

Study Design

We performed a retrospective cohort study of patients aged 50 years and older who underwent surgery for an isolated fragility fracture of the hip at Hartford Hospital, a level one trauma and tertiary care medical center, within the 24-month period from April 2015 to March 2017. Fragility hip fracture is defined as one occurring from a fall of a height of standing or less. A consult referred to a specialty or subspecialty consultation, other than hospital medicine, obtained prior to surgery. Patients with additional skeletal trauma and periprosthetic fractures were excluded. A total of 491 unique patients met the inclusion criteria, and data were obtained from chart review and an orthopedic surgery registry. The Hartford Hospital Institutional Review Board approved this study.

Our primary predictor was the presence or absence of a preoperative specialty consultation requested by the hospitalist. We also analyzed the following: covariates of demographics (age, sex, race), the ASA score, and severity of comorbidities using the Charlson comorbidity index (CCI) with a Quan modification;¹¹ “R program package, International Classification of Disease (ICD)”¹² was used to calculate the CCI using ICD-9 and ICD-10 diagnostic codes.

The primary outcome measures were TTS (measured in hours), LOS (measured in days), complications, and preoperative specialty consult resulting in a change in perioperative management. TTS was defined as the time elapsed from the presentation at the emergency department (ED) to surgery start. For transfer or direct admission patients, the time of admission was used in place of time of presentation. The measured complications included postoperative venous thromboembolic events, surgical site infection, myocardial infarction, stroke, and sepsis. Secondary outcome measures included 30-day mortality, readmission rate, and rate of return to OR. There were no elective or planned readmissions postoperatively on review of our institution’s orthopedic surgery registry.

Our team performed an extensive chart review including reviewing the admission note, consulting physician notes, and relevant test results. Our senior investigator (MK) then rated each preoperative specialty consult on appropriateness, the relative strength of the consultant’s recommendation, and resulting change in perioperative management. Cardiology consultations were deemed reasonable if a patient’s cardiac risk was considered elevated by the admitting physician or an active cardiac condition was present (suggestion of or clear evidence for acute coronary syndrome, acute congestive heart failure, uncontrolled arrhythmia, or symptomatic valvular disease). The determination of “elevated cardiac risk” was made, if admit note contained verbiage expressing concern for further evaluation for cardiac issues or words such as “high risk” or “elevated risk”. A specific guideline-based score such as the revised cardiac risk index was not consistently available in this retrospective chart review. A noncardiology consult was deemed reasonable only if it would have been warranted for the specific clinical situation­—for example, a neurology consult for an acute stroke or a pulmonary consult for acute respiratory failure. Consult recommendations or outcomes were rated as minor, moderate, or major (see Table 1 for detailed criteria). Some consults may generate more than one recommendation, in these cases, we determined that a major recommendation supersedes a moderate or minor recommendation and only one was counted in the final analysis. Next we determined if a consult recommendation led to a change in perioperative or therapeutic management, defined as a medication or dosage change, need to delay surgery to stabilize an unstable medical condition, invasive procedures (such as thoracentesis or cardiac catheterization) or change in postoperative monitoring. As a way of clarification, a consult may have a minor recommendation such as an EKG but if no other recommendations were given and there was no change in therapeutic management such as a medication change, this would be considered as a “no change”.

Statistical Analysis

The main analyses compared the two patient subgroups (with or without preoperative specialty consults) around outcome measures. Primary outcome measures were TTS, LOS, complications, and consult resulting in a change in perioperative management. Secondary outcome measures were 30-day readmit, return to OR, and mortality. A preliminary analysis was conducted to explore distributions for TTS and LOS. As expected, none met the assumptions of normality and were thus analyzed with Wilcoxon ranked-sum tests. The other outcomes were dichotomous and analyzed with chi-square tests of proportion or Fisher’s exact test when the expected cell frequencies were too low. Dichotomized variables for TTS (within 24 hours and 48 hours) and LOS (within five days, the median LOS for this cohort) were calculated and subsequently analyzed with additional chi-square tests of proportion or Fisher’s exact test¹³. To explore the effect of preoperative specialty consults independent of potential confounders, logistic regression analyses predicting each of the dichotomous outcomes were conducted with age and CCI used as predictors in addition to the main variable of whether or not there was a preoperative specialty consult. Since the CCI and ASA scores were highly intercorrelated, only the former was chosen for the multivariate analyses based on the consistent algorithm used to calculate CCI.

Additional analyses with the subgroup of patients with a preoperative specialty consult explored whether the consult was reasonable, the relative strength of resulting recommendation and whether it resulted in a change in management. The statistical approach used was the same as for the other dichotomous outcomes. All analyses used 0.05 as the level of statistical significance; SPSSv21 (IBM, Armonk, New York) was the statistical software used.

The sample size for this retrospective analysis was determined by the available number of patients meeting the inclusion criteria. An a priori power calculation was done to determine if the expected volume would be sufficient for the multivariate analysis; the presence of a complication was selected for calculation. Based on an expected volume of approximately 500 and an estimate of a 10% serious complication rate, it was determined that the sample could support the analysis of up to five predictor variables, sufficient for the main variable and four potential confounders; this was considered adequate.¹⁴ Propensity scoring was considered but did not offer any advantages to logistic regression because we only had two observed covariates: CCI and age.

A total of 491 unique patients met our inclusion criteria, 177 patients had a preoperative specialty consult. Of these 177 patients, 24 patients had more than one consult; hence, the total number of consults was 201. Most of the consults were cardiology (159). Others were Infectious disease (11), Pulmonology (10), Neurology (7), and Miscellaneous (14, which included Nephrology, Gastroenterology, Hematology, and Oncology).

No significant differences were found between the consult and no-consult groups with respect to gender, race, body mass index, type of anesthesia, and day of the week of surgery. We did note that patients with a consult were older and had a significantly higher CCI and ASA score (Table 2).

Patients with preoperative specialty consults were older and had more comorbidities than patients without consults. Our findings suggest that consults contribute to delays to surgery and may lead to higher LOS and higher risk of 30-day readmission after controlling for age and comorbidities in a multivariate analysis. This observation is significant considering that consults were requested more frequently on patients with a higher comorbidity burden and included patients who did not get additional preoperative testing, suggesting that a delay from waiting for a consult alone may be deleterious. This was a unique observation in our study; prior studies examining this subject have attributed delays to additional testing and not consults alone. Even though most consult requests appear to be reasonable according to our criteria, the majority of recommendations were minor (72.9%), and 62.7% of consults resulted in no change in perioperative management. Major changes in perioperative management were noted in only 4.5% of patients.

Our finding that a majority of patients in the consult group had no significant change in perioperative management raises an important area of potential improvement in the care of hip fracture patients. We believe that narrowing indications for preoperative specialty consults may result in shorter TTS and LOS for this group of frail elderly patients without sacrificing the quality of care. Since all patients in our study were comanaged by hospitalists and patients without additional consults had similar or better outcomes, we believe that hospitalist physicians are well positioned to provide standardized comanagement to this patient group without additional consultation unless absolutely necessary.

The primary limitation of our study was that this was a retrospective case analysis. The designation of minor, moderate, or major recommendation was done after the consults were already completed, and it may not be possible to predict that a consult results in no change without it being actually performed. Additionally, our classification of recommendations is somewhat arbitrary and subjective; for example, some readers might argue that a medication change counts as a moderate recommendation. We rated a medication change to be minor as we believe that an experienced hospitalist may likely make such management decisions on their own, and if this is the only recommendation from a consult, it is not additional information critical to patient care. There may also be an “unmeasured complexity” noted by the admitting physician, which was not necessarily accounted for by multivariate analysis of age and CCI but one that led to higher mortality and readmissions. However, we feel that this “unmeasured complexity” is likely inconsequential as the vast majority of consults did not result in any change in management. We did adjust for covariates as noted, but some confounding by indication is likely to remain. Additionally, categorization of consult recommendations and consequent changes by one physician could be considered subjective. We did control for this by having another physician review the entire dataset and rate it independently for interrater reliability with excellent correlation and kappa, although these may be inflated to some degree because our chart review did not account for variability among chart extractors.

A prospective evaluation of a clinical protocol that delineates reasonable indications for a preoperative consult would be helpful to validate our findings. In our study, we noted moderate or major recommendations from a cardiologist only in cases where an active cardiac condition was suspected by the hospitalist requesting the consult; hence, limiting preoperative cardiology consults to active cardiac conditions may be a reasonable approach to evaluate in a prospective study.

In conclusion, a majority of preoperative specialty consults do not appear to meaningfully influence management and may indirectly increase morbidity by delaying surgery and extending hospital stays. Our data suggest that unless the patient is clinically unstable and likely to require active management by a consultant prior to hip fracture repair, consults may offer limited benefit. Appropriately standardized perioperative management of this patient group by hospitalist physicians appears to manage most hip fracture patients as effectively with faster TTS and shorter hospital LOS.

Acknowledgments

The authors would like to thank John Corradi, PhD (Research Department at Hartford Hospital) for his input in calculating the Charlson comorbidity index.^12,13

An essential feature of cancer is its ability to evade the immune system. Multiple mechanisms are used for this purpose, including the disruption of antigen presentation and suppression of the immune response. The latter mechanism involves the activation of T-cell inhibition by recruiting regulatory T cells that weaken this response. Recent progress in understanding the ability of cancer to evade the immune system has paved the way to develop strategies to reverse this process and reactivate the immune system. Particularly, immune checkpoint signaling between T cells and tumor cells has been targeted with a new class of drug, immune checkpoint inhibitors. Immunotherapy has been an established and effective treatment in bladder cancer since 1976 when Morales and colleagues demonstrated that intravesical treatments with bacillus Calmette-Guérin can treat carcinoma in situ and prevent nonmuscle invasive urothelial cancer recurrence.^1,2 This treatment elicits a cytotoxic response via antigenic presentation by bladder tumor cells.

Cytotoxic T-lymphocyte-associated protein (CTLA)-4, programmed death-1 (PD-1) and programmed death-ligand-1 (PD-L1) are molecules that downregulate the immune response and are targets of therapeutic antibodies that have demonstrated clinical efficacy across a wide range of malignancies. Five such agents—pembrolizumab, atezolizumab, nivolumab, avelumab and durvalumab—were recently approved by the US Food and Drug Administration (FDA) for clinical use in patients with advanced urothelial cancers.³ This class of agents also has been approved for several other malignancies, most notably in melanoma, non-small cell lung cancer, and renal cell carcinoma.³

Immune Biology

CTLA-4 is expressed on activated CD4 and CD8 T cells and competes with CD28 on T cells to interact with the costimulatory B7 proteins on antigen presenting cells. The CD28/B7 interaction promotes T-cell activation and effector functions, and the CTLA-4/B7 interaction inhibits them. In addition, PD-1 is a receptor expressed on CD4 and CD8 T cells, T regulatory (Treg) cells, B cells and natural killer (NK) cells that interacts with its ligand PD-L1 to suppress the immune response. Urothelial cancer possesses features that make it an adequate target for immunotherapeutic agents. Primarily, it is characterized by a high-mutation load, which lends itself to an increased expression of immunogenic antigens on tumor cells.⁴

Immunotherapy Treatments in Cisplatin-Ineligible Patients

Cisplatin-based chemotherapy is the first-line treatment and standard of care in unresectable or metastatic urothelial cancer. However, many patients are unable to receive cisplatin secondary to renal dysfunction, poor performance status, or other comorbidities. Alternative cytotoxic therapies in the first-line setting such as carboplatin-based regimens are associated with inferior outcomes and poor tolerability. There is, therefore, a need for effective and well-tolerated therapies in cisplatin-ineligible patients (Table).

In the phase 2 Keynote-052 trial, 370 cisplatin-ineligible patients were treated with the anti-PD-1 antibody pembrolizumab 200 mg every 3 weeks for up to 2 years.⁵At a median follow-up of 9.5 months, the objective response rate (+ORR) was 29% for the entire cohort, with a 7% complete response (CR) rate, and a 22% partial response (PR) rate.⁵ The median duration of response had not been reached at the time of analysis. Responses were seen regardless of PD-L1 expression, although high response rates were noted in patients whose tumors had PD-L1 expression > 10%. Pembrolizumab had an acceptable tolerability profile in this population. The most common grade 3 or 4 treatment-related adverse event (AE) was fatigue at 2%; 5% of patients discontinued therapy due to treatment related AEs, whereas 17% of patients had immune-mediated AEs.⁵

Similarly, in a single-arm phase 2 trial, atezolizumab, an anti-PD-L1 antibody, dosed at 1,200 mg every 3 weeks was used as first-line therapy in 119 patients with advanced urothelial cancer who were cisplatin ineligible. At a median follow-up of 17 months, the ORR was 23%, with a 9% CR rate. The median duration of response had not been reached. Median progression free survival (PFS) was 2.7 months, whereas overall survival (OS) was 16 months. Eight percent of patients had an AE leading to treatment discontinuation, and 17% had immune-mediated AEs.⁶ Both pembrolizumab and atezolizumab were granted FDA approval in 2017 for patients with locally advanced or metastatic urothelial carcinoma who are not eligible for cisplatin-based chemotherapy.³

Immunotherapy Treatments After Progression With Cisplatin

Cytotoxic chemotherapy in the second-line setting with disease progression following platinum-based treatment has shown dismal responses, with a median OS of about 6 to 7 months.⁷ Immunotherapy provides an effective and a much-needed option in this scenario.

Five antibodies targeting the PD-1/PD-L1 pathway, pembrolizumab, nivolumab, atezolizumab, avelumab and durvalumab, have been granted FDA approval for patients who have progressed during or after platinum-based therapy (Table).³ In the phase 3 Keynote-045 trial, 542 patients were randomly assigned to receive either pembrolizumab 200 mg administered every 3 weeks or investigator’s choice chemotherapy (paclitaxel, docetaxel, or vinflunine).⁷ Median OS was 10.3 months in the pembrolizumab group and 7.4 months in the chemotherapy group (hazard ratio for death, 0.73; P = .002). Serious (grade 3 or above) treatment-related AEs were significantly less frequent with pembrolizumab (15% vs 49.4%).⁷ In a phase 2 trial, 270 patients were treated with nivolumab, a PD-1 inhibitor, at a dose of 3 mg/kg given every 2 weeks.⁸ The ORR was 19.6%, while the median OS for the entire cohort was 7 months. Responses were seen at all levels of PD-L1 expression, although in patients whose tumor expressed PD-L1 ≥1%, median OS was 11.3 months.⁸

It should be noted that in a large phase 3 trial comparing atezolizumab with chemotherapy in the second-line setting, ORR and OS were not statistically different between the 2 groups, although the duration of response was longer with atezolizumab.⁹ In early phase trials, avelumab and durvalumab, both PD-L1 inhibitors showed an ORR of about 17%, with higher ORR seen in patients with tumors positive for PD-L1 expression.^10,11 The AE profile of immune checkpoint inhibitors is relatively favorable in clinical trials. The American Society of Clinical Oncology and National Comprehensive Cancer Network have jointly published evidence-based guidelines for the management of their immune related AEs.¹²

Future Directions

Several challenges have emerged with immunotherapy treatments. One issue is the relatively low ORRs for immune checkpoint inhibitors, ranging from 13.4% to 24% depending on the trial. Therefore, there is a need to identify reliable biomarkers and selection criteria to predict their efficacy and improve patient selection. Although tumor PD-L1 expression has shown some usefulness in this setting, responses have been noted in patients whose tumors have low or no expression of PD-L1. This low predictive accuracy is caused by several factors, including PD-L1 intratumor expression heterogeneity, primary vs metastatic site PD-L1 expression heterogeneity, lack of consensus on which PD-L1 assays and which value cutoffs to use, and the differences seen in marker expression depending on the freshness of the tissue specimen.

Other predictive biomarkers with potential include tumor gene expression profiles/tumor mutational load, T-cell and B-cell signatures. The optimal imaging modality and timing of this imaging for response assessment also is uncertain. So-called tumor pseudo-progression seen on imaging after treatment with these agents as a result of the immune/inflammatory response to the tumor is now a well-recognized phenomenon, but it can be challenging to differentiate from true disease progression. Other challenges include deciding on which immune checkpoint inhibitor to use given a lack of head-to-head comparisons of these immunotherapeutic agents, finding the proper drug doses to maximize efficacy, as well as determining the optimal duration of treatment in patients with continued response to immunotherapy. Many oncologists continue these treatments for up to 2 years in the setting of a significant or complete response.

Conclusion

Immune checkpoint inhibitors have emerged as pivotal treatments for patients with advanced urothelial cancer who are unfit to receive cisplatin in the first-line setting or who experience disease progression after cisplatin-based chemotherapy. This field continues to expand at a rapid pace due to multiple ongoing clinical trials assessing these agents, whether alone, in combination with cytotoxic, targeted, radiation therapies, or with other immune checkpoint inhibitors, both in the advanced as well as the neoadjuvant/adjuvant settings.

An essential feature of cancer is its ability to evade the immune system. Multiple mechanisms are used for this purpose, including the disruption of antigen presentation and suppression of the immune response. The latter mechanism involves the activation of T-cell inhibition by recruiting regulatory T cells that weaken this response. Recent progress in understanding the ability of cancer to evade the immune system has paved the way to develop strategies to reverse this process and reactivate the immune system. Particularly, immune checkpoint signaling between T cells and tumor cells has been targeted with a new class of drug, immune checkpoint inhibitors. Immunotherapy has been an established and effective treatment in bladder cancer since 1976 when Morales and colleagues demonstrated that intravesical treatments with bacillus Calmette-Guérin can treat carcinoma in situ and prevent nonmuscle invasive urothelial cancer recurrence.^1,2 This treatment elicits a cytotoxic response via antigenic presentation by bladder tumor cells.

Cytotoxic T-lymphocyte-associated protein (CTLA)-4, programmed death-1 (PD-1) and programmed death-ligand-1 (PD-L1) are molecules that downregulate the immune response and are targets of therapeutic antibodies that have demonstrated clinical efficacy across a wide range of malignancies. Five such agents—pembrolizumab, atezolizumab, nivolumab, avelumab and durvalumab—were recently approved by the US Food and Drug Administration (FDA) for clinical use in patients with advanced urothelial cancers.³ This class of agents also has been approved for several other malignancies, most notably in melanoma, non-small cell lung cancer, and renal cell carcinoma.³

Immune Biology

CTLA-4 is expressed on activated CD4 and CD8 T cells and competes with CD28 on T cells to interact with the costimulatory B7 proteins on antigen presenting cells. The CD28/B7 interaction promotes T-cell activation and effector functions, and the CTLA-4/B7 interaction inhibits them. In addition, PD-1 is a receptor expressed on CD4 and CD8 T cells, T regulatory (Treg) cells, B cells and natural killer (NK) cells that interacts with its ligand PD-L1 to suppress the immune response. Urothelial cancer possesses features that make it an adequate target for immunotherapeutic agents. Primarily, it is characterized by a high-mutation load, which lends itself to an increased expression of immunogenic antigens on tumor cells.⁴

Immunotherapy Treatments in Cisplatin-Ineligible Patients

Cisplatin-based chemotherapy is the first-line treatment and standard of care in unresectable or metastatic urothelial cancer. However, many patients are unable to receive cisplatin secondary to renal dysfunction, poor performance status, or other comorbidities. Alternative cytotoxic therapies in the first-line setting such as carboplatin-based regimens are associated with inferior outcomes and poor tolerability. There is, therefore, a need for effective and well-tolerated therapies in cisplatin-ineligible patients (Table).

In the phase 2 Keynote-052 trial, 370 cisplatin-ineligible patients were treated with the anti-PD-1 antibody pembrolizumab 200 mg every 3 weeks for up to 2 years.⁵At a median follow-up of 9.5 months, the objective response rate (+ORR) was 29% for the entire cohort, with a 7% complete response (CR) rate, and a 22% partial response (PR) rate.⁵ The median duration of response had not been reached at the time of analysis. Responses were seen regardless of PD-L1 expression, although high response rates were noted in patients whose tumors had PD-L1 expression > 10%. Pembrolizumab had an acceptable tolerability profile in this population. The most common grade 3 or 4 treatment-related adverse event (AE) was fatigue at 2%; 5% of patients discontinued therapy due to treatment related AEs, whereas 17% of patients had immune-mediated AEs.⁵

Similarly, in a single-arm phase 2 trial, atezolizumab, an anti-PD-L1 antibody, dosed at 1,200 mg every 3 weeks was used as first-line therapy in 119 patients with advanced urothelial cancer who were cisplatin ineligible. At a median follow-up of 17 months, the ORR was 23%, with a 9% CR rate. The median duration of response had not been reached. Median progression free survival (PFS) was 2.7 months, whereas overall survival (OS) was 16 months. Eight percent of patients had an AE leading to treatment discontinuation, and 17% had immune-mediated AEs.⁶ Both pembrolizumab and atezolizumab were granted FDA approval in 2017 for patients with locally advanced or metastatic urothelial carcinoma who are not eligible for cisplatin-based chemotherapy.³

Immunotherapy Treatments After Progression With Cisplatin

Cytotoxic chemotherapy in the second-line setting with disease progression following platinum-based treatment has shown dismal responses, with a median OS of about 6 to 7 months.⁷ Immunotherapy provides an effective and a much-needed option in this scenario.

Five antibodies targeting the PD-1/PD-L1 pathway, pembrolizumab, nivolumab, atezolizumab, avelumab and durvalumab, have been granted FDA approval for patients who have progressed during or after platinum-based therapy (Table).³ In the phase 3 Keynote-045 trial, 542 patients were randomly assigned to receive either pembrolizumab 200 mg administered every 3 weeks or investigator’s choice chemotherapy (paclitaxel, docetaxel, or vinflunine).⁷ Median OS was 10.3 months in the pembrolizumab group and 7.4 months in the chemotherapy group (hazard ratio for death, 0.73; P = .002). Serious (grade 3 or above) treatment-related AEs were significantly less frequent with pembrolizumab (15% vs 49.4%).⁷ In a phase 2 trial, 270 patients were treated with nivolumab, a PD-1 inhibitor, at a dose of 3 mg/kg given every 2 weeks.⁸ The ORR was 19.6%, while the median OS for the entire cohort was 7 months. Responses were seen at all levels of PD-L1 expression, although in patients whose tumor expressed PD-L1 ≥1%, median OS was 11.3 months.⁸

It should be noted that in a large phase 3 trial comparing atezolizumab with chemotherapy in the second-line setting, ORR and OS were not statistically different between the 2 groups, although the duration of response was longer with atezolizumab.⁹ In early phase trials, avelumab and durvalumab, both PD-L1 inhibitors showed an ORR of about 17%, with higher ORR seen in patients with tumors positive for PD-L1 expression.^10,11 The AE profile of immune checkpoint inhibitors is relatively favorable in clinical trials. The American Society of Clinical Oncology and National Comprehensive Cancer Network have jointly published evidence-based guidelines for the management of their immune related AEs.¹²

Future Directions

Several challenges have emerged with immunotherapy treatments. One issue is the relatively low ORRs for immune checkpoint inhibitors, ranging from 13.4% to 24% depending on the trial. Therefore, there is a need to identify reliable biomarkers and selection criteria to predict their efficacy and improve patient selection. Although tumor PD-L1 expression has shown some usefulness in this setting, responses have been noted in patients whose tumors have low or no expression of PD-L1. This low predictive accuracy is caused by several factors, including PD-L1 intratumor expression heterogeneity, primary vs metastatic site PD-L1 expression heterogeneity, lack of consensus on which PD-L1 assays and which value cutoffs to use, and the differences seen in marker expression depending on the freshness of the tissue specimen.

Other predictive biomarkers with potential include tumor gene expression profiles/tumor mutational load, T-cell and B-cell signatures. The optimal imaging modality and timing of this imaging for response assessment also is uncertain. So-called tumor pseudo-progression seen on imaging after treatment with these agents as a result of the immune/inflammatory response to the tumor is now a well-recognized phenomenon, but it can be challenging to differentiate from true disease progression. Other challenges include deciding on which immune checkpoint inhibitor to use given a lack of head-to-head comparisons of these immunotherapeutic agents, finding the proper drug doses to maximize efficacy, as well as determining the optimal duration of treatment in patients with continued response to immunotherapy. Many oncologists continue these treatments for up to 2 years in the setting of a significant or complete response.

Conclusion

Immune checkpoint inhibitors have emerged as pivotal treatments for patients with advanced urothelial cancer who are unfit to receive cisplatin in the first-line setting or who experience disease progression after cisplatin-based chemotherapy. This field continues to expand at a rapid pace due to multiple ongoing clinical trials assessing these agents, whether alone, in combination with cytotoxic, targeted, radiation therapies, or with other immune checkpoint inhibitors, both in the advanced as well as the neoadjuvant/adjuvant settings.

An essential feature of cancer is its ability to evade the immune system. Multiple mechanisms are used for this purpose, including the disruption of antigen presentation and suppression of the immune response. The latter mechanism involves the activation of T-cell inhibition by recruiting regulatory T cells that weaken this response. Recent progress in understanding the ability of cancer to evade the immune system has paved the way to develop strategies to reverse this process and reactivate the immune system. Particularly, immune checkpoint signaling between T cells and tumor cells has been targeted with a new class of drug, immune checkpoint inhibitors. Immunotherapy has been an established and effective treatment in bladder cancer since 1976 when Morales and colleagues demonstrated that intravesical treatments with bacillus Calmette-Guérin can treat carcinoma in situ and prevent nonmuscle invasive urothelial cancer recurrence.^1,2 This treatment elicits a cytotoxic response via antigenic presentation by bladder tumor cells.

Cytotoxic T-lymphocyte-associated protein (CTLA)-4, programmed death-1 (PD-1) and programmed death-ligand-1 (PD-L1) are molecules that downregulate the immune response and are targets of therapeutic antibodies that have demonstrated clinical efficacy across a wide range of malignancies. Five such agents—pembrolizumab, atezolizumab, nivolumab, avelumab and durvalumab—were recently approved by the US Food and Drug Administration (FDA) for clinical use in patients with advanced urothelial cancers.³ This class of agents also has been approved for several other malignancies, most notably in melanoma, non-small cell lung cancer, and renal cell carcinoma.³

Immune Biology

CTLA-4 is expressed on activated CD4 and CD8 T cells and competes with CD28 on T cells to interact with the costimulatory B7 proteins on antigen presenting cells. The CD28/B7 interaction promotes T-cell activation and effector functions, and the CTLA-4/B7 interaction inhibits them. In addition, PD-1 is a receptor expressed on CD4 and CD8 T cells, T regulatory (Treg) cells, B cells and natural killer (NK) cells that interacts with its ligand PD-L1 to suppress the immune response. Urothelial cancer possesses features that make it an adequate target for immunotherapeutic agents. Primarily, it is characterized by a high-mutation load, which lends itself to an increased expression of immunogenic antigens on tumor cells.⁴

Immunotherapy Treatments in Cisplatin-Ineligible Patients

Cisplatin-based chemotherapy is the first-line treatment and standard of care in unresectable or metastatic urothelial cancer. However, many patients are unable to receive cisplatin secondary to renal dysfunction, poor performance status, or other comorbidities. Alternative cytotoxic therapies in the first-line setting such as carboplatin-based regimens are associated with inferior outcomes and poor tolerability. There is, therefore, a need for effective and well-tolerated therapies in cisplatin-ineligible patients (Table).

In the phase 2 Keynote-052 trial, 370 cisplatin-ineligible patients were treated with the anti-PD-1 antibody pembrolizumab 200 mg every 3 weeks for up to 2 years.⁵At a median follow-up of 9.5 months, the objective response rate (+ORR) was 29% for the entire cohort, with a 7% complete response (CR) rate, and a 22% partial response (PR) rate.⁵ The median duration of response had not been reached at the time of analysis. Responses were seen regardless of PD-L1 expression, although high response rates were noted in patients whose tumors had PD-L1 expression > 10%. Pembrolizumab had an acceptable tolerability profile in this population. The most common grade 3 or 4 treatment-related adverse event (AE) was fatigue at 2%; 5% of patients discontinued therapy due to treatment related AEs, whereas 17% of patients had immune-mediated AEs.⁵

Similarly, in a single-arm phase 2 trial, atezolizumab, an anti-PD-L1 antibody, dosed at 1,200 mg every 3 weeks was used as first-line therapy in 119 patients with advanced urothelial cancer who were cisplatin ineligible. At a median follow-up of 17 months, the ORR was 23%, with a 9% CR rate. The median duration of response had not been reached. Median progression free survival (PFS) was 2.7 months, whereas overall survival (OS) was 16 months. Eight percent of patients had an AE leading to treatment discontinuation, and 17% had immune-mediated AEs.⁶ Both pembrolizumab and atezolizumab were granted FDA approval in 2017 for patients with locally advanced or metastatic urothelial carcinoma who are not eligible for cisplatin-based chemotherapy.³

Immunotherapy Treatments After Progression With Cisplatin

Cytotoxic chemotherapy in the second-line setting with disease progression following platinum-based treatment has shown dismal responses, with a median OS of about 6 to 7 months.⁷ Immunotherapy provides an effective and a much-needed option in this scenario.

Five antibodies targeting the PD-1/PD-L1 pathway, pembrolizumab, nivolumab, atezolizumab, avelumab and durvalumab, have been granted FDA approval for patients who have progressed during or after platinum-based therapy (Table).³ In the phase 3 Keynote-045 trial, 542 patients were randomly assigned to receive either pembrolizumab 200 mg administered every 3 weeks or investigator’s choice chemotherapy (paclitaxel, docetaxel, or vinflunine).⁷ Median OS was 10.3 months in the pembrolizumab group and 7.4 months in the chemotherapy group (hazard ratio for death, 0.73; P = .002). Serious (grade 3 or above) treatment-related AEs were significantly less frequent with pembrolizumab (15% vs 49.4%).⁷ In a phase 2 trial, 270 patients were treated with nivolumab, a PD-1 inhibitor, at a dose of 3 mg/kg given every 2 weeks.⁸ The ORR was 19.6%, while the median OS for the entire cohort was 7 months. Responses were seen at all levels of PD-L1 expression, although in patients whose tumor expressed PD-L1 ≥1%, median OS was 11.3 months.⁸

It should be noted that in a large phase 3 trial comparing atezolizumab with chemotherapy in the second-line setting, ORR and OS were not statistically different between the 2 groups, although the duration of response was longer with atezolizumab.⁹ In early phase trials, avelumab and durvalumab, both PD-L1 inhibitors showed an ORR of about 17%, with higher ORR seen in patients with tumors positive for PD-L1 expression.^10,11 The AE profile of immune checkpoint inhibitors is relatively favorable in clinical trials. The American Society of Clinical Oncology and National Comprehensive Cancer Network have jointly published evidence-based guidelines for the management of their immune related AEs.¹²

Future Directions

Several challenges have emerged with immunotherapy treatments. One issue is the relatively low ORRs for immune checkpoint inhibitors, ranging from 13.4% to 24% depending on the trial. Therefore, there is a need to identify reliable biomarkers and selection criteria to predict their efficacy and improve patient selection. Although tumor PD-L1 expression has shown some usefulness in this setting, responses have been noted in patients whose tumors have low or no expression of PD-L1. This low predictive accuracy is caused by several factors, including PD-L1 intratumor expression heterogeneity, primary vs metastatic site PD-L1 expression heterogeneity, lack of consensus on which PD-L1 assays and which value cutoffs to use, and the differences seen in marker expression depending on the freshness of the tissue specimen.

Other predictive biomarkers with potential include tumor gene expression profiles/tumor mutational load, T-cell and B-cell signatures. The optimal imaging modality and timing of this imaging for response assessment also is uncertain. So-called tumor pseudo-progression seen on imaging after treatment with these agents as a result of the immune/inflammatory response to the tumor is now a well-recognized phenomenon, but it can be challenging to differentiate from true disease progression. Other challenges include deciding on which immune checkpoint inhibitor to use given a lack of head-to-head comparisons of these immunotherapeutic agents, finding the proper drug doses to maximize efficacy, as well as determining the optimal duration of treatment in patients with continued response to immunotherapy. Many oncologists continue these treatments for up to 2 years in the setting of a significant or complete response.

Conclusion

Immune checkpoint inhibitors have emerged as pivotal treatments for patients with advanced urothelial cancer who are unfit to receive cisplatin in the first-line setting or who experience disease progression after cisplatin-based chemotherapy. This field continues to expand at a rapid pace due to multiple ongoing clinical trials assessing these agents, whether alone, in combination with cytotoxic, targeted, radiation therapies, or with other immune checkpoint inhibitors, both in the advanced as well as the neoadjuvant/adjuvant settings.

Lung cancer is the leading cause of cancer death in the US, with 154 050 deaths in 2018.¹ There have been many attempts to reduce mortality of the disease through early diagnosis with use of computed tomography (CT). The National Lung Cancer Screening trial showed that screening high-risk populations with low-dose CT (LDCT) can reduce mortality.² However, implementing LDCT screening in the clinical setting has proven challenging, as illustrated by the VA Lung Cancer Screening Demonstration Project (LCSDP).³ A lung cancer diagnosis typically comprises several steps that require different medical specialties; this can lead to delays. In the LCSDP, the mean time to diagnosis was 137 days.³ There are no federal standards for timeliness of lung cancer diagnosis.

The nonprofit RAND Corporation is the only American research organization that has published guidelines specifying acceptable intervals for the diagnosis and treatment of lung cancer. In Quality of Care for Oncologic Conditions and HIV, RAND Corporation researchers propose management quality indicators: lung cancer diagnosis within 2 months of an abnormal radiologic study and treatment within 6 weeks of diagnosis.⁴ The Swedish Lung Cancer Study⁵ and the Canadian Strategy for Cancer Control⁶ both recommended a standard of about 30 days—half the time recommended by the RAND Corporation.

Bukhari and colleagues at the Dayton US Department of Veterans Affairs (VA) Medical Center (VAMC) conducted a quality improvement study that examined lung cancer diagnosis and management.7 They found the time (SD) from abnormal chest imaging to diagnosis was 35.5 (31.6) days. Of those veterans who received a lung cancer diagnosis, 89.2% had the diagnosis made within the 60 days recommended by the RAND Corporation. Although these results surpass those of the LCSDP, they can be exceeded.

Beyond the potential emotional distress of awaiting the final diagnosis of a lung lesion, a delay in diagnosis and treatment may adversely affect outcomes. LDCT screening has been shown to reduce mortality, which implies a link between survival and time to intervention. There is no published evidence that time to diagnosis in advanced stage lung cancer affects outcome. The National Cancer Database (NCDB) contains informtion on about 70% of the cancers diagnosed each year in the US.8 An analysis of 4984 patients with stage IA squamous cell lung cancer undergoing lobectomy from NCDB showed that earlier surgery was associated with an absolute decrease in 5-year mortality of 5% to 8%. 9 Hence, at least in early-stage disease, reduced time from initial suspect imaging to definitive treatment may improve survival.

A system that coordinates the requisite diagnostic steps and avoids delays should provide a significant improvement in patient care. The results of such an approach that utilized nurse navigators has been previously published. 10 Here, we present the results of a dedicated VA referral clinic with priority access to pulmonary consultation and procedures in place that are designed to expedite the diagnosis of potential lung cancer.

Methods

The John L. McClellan Memorial Veterans Hospital (JLMMVH) in Little Rock, Arkansas institutional review board approved this study, which was performed in accordance with the Declaration of Helsinki. Requirement for informed consent was waived, and patient confidentiality was maintained throughout.

We have developed a plan of care specifically to facilitate diagnosis and treatment of the large number of veterans referred to the JLMMVH Diagnostic Clinic for abnormal results of chest imaging. The clinic has priority access to same-day imaging and subspecialty consultation services. In the clinic, medical students and residents perform evaluations and a registered nurse (RN) manager coordinates care.

A Diagnostic Clinic consult for abnormal thoracic imaging immediately triggers an e-consult to an interventional pulmonologist (Figure). The RN manager and pulmonologist perform a joint review of records/imaging prior to scheduling, and the pulmonologist triages the patient. Triage options include follow-up imaging, bronchoscopy with endobronchial ultrasound (EBUS), endoscopic ultrasound (EUS), and CT-guided biopsy.

The RN manager then schedules a clinic visit that includes a medical evaluation by clinic staff and any indicated procedures on the same day. The interventional pulmonologist performs EBUS, EUS with the convex curvilinear bronchoscope, or both combined as indicated for diagnosis and staging. All procedures are performed in the JLMMVH bronchoscopy suite with standard conscious sedation using midazolam and fentanyl. Any other relevant procedures, such as pleural tap, also are performed at time of procedure. The pulmonologist and an attending pathologist interpret biopsies obtained in the bronchoscopy suite.

We performed a retrospective chart review of patients diagnosed with primary lung cancer through referral to the JLMMVH Diagnostic Clinic. The primary outcome was time from initial suspect chest imaging to cancer diagnosis. The study population consisted of patients referred for abnormal thoracic imaging between January 1, 2013 and December 31, 2016 and subsequently diagnosed with a primary lung cancer.

Subjects were excluded if (1) the patient was referred from outside our care network and a delay of > 10 days occurred between initial lesion imaging and referral; (2) the patient did not show up for appointments or chose to delay evaluation following referral; (3) biopsy demonstrated a nonlung primary cancer; and (4) serious intercurrent illness interrupted the diagnostic plan. In some cases, the radiologist or consulting pulmonologist had judged the lung lesion too small for immediate biopsy and recommended repeat imaging at a later date.

Patients were included in the study if the follow- up imaging led to a lung cancer diagnosis. However, because the interval between the initial imaging and the follow-up imaging in these patients did not represent a systems delay problem, the date of the scheduled follow-up abnormal imaging, which resulted in initiation of a potential cancer evaluation, served as the index suspect imaging date for this study.

Patient electronic medical records were reviewed and the following data were abstracted: date of the abnormal imaging that led to referral and time from abnormal chest X-ray to chest CT scan if applicable; date of referral and date of clinic visit; date of biopsy; date of lung cancer diagnosis; method of obtaining diagnostic specimen; lung cancer type and stage; type and date of treatment initiation or decision for supportive care only; and decision to seek further evaluation or care outside of our system.

All patients diagnosed with lung cancer during the study period were reviewed for inclusion, hence no required sample-size estimate was calculated. All outcomes were assessed as calendar days. The primary outcome was the time from the index suspect chest imaging study to the date of diagnosis of lung cancer. Prior to the initiation of our study, we chose this more stringent 30-day recommendation of the Canadian6 and Swedish5 studies as the comparator for our primary outcome, although data with respect to the 60-day Rand Corporation guidelines also are reported.⁴

Statistical Methods

The mean time to lung cancer diagnosis in our cohort was compared with this 30-day standard using a 2-sided Mann–Whitney U test. Normality of data distribution was determined using the Kolmogorov–Smirnov test. For statistical significance testing a P value of .05 was used. Statistical calculations were performed using R statistical software version 3.2.4. Secondary outcomes consisted of time from diagnosis to treatment; proportion of subjects diagnosed within 60 days; time from initial clinic visit to biopsy; and time from biopsy to diagnosis.

Results

Overall, 222 patients were diagnosed with a malignant lung lesion, of which 63 were excluded from analysis: 22 cancelled or did not appear for appointments, declined further evaluation, or completed evaluation outside of our network; 13 had the diagnosis made prior to Diagnostic Clinic visit; 13 proved to have a nonlung primary tumor presenting in the lung or mediastinal nodes; 12 were delayed > 10 days in referral from an outside network; and 3 had an intervening serious acute medical problem forcing delay in the diagnostic process.

Of the 159 included subjects, 154 (96.9%) were male, and the mean (SD) age was 67.6 (8.1) years. For 76 subjects, the abnormal chest X-ray and subsequent chest CT scan were performed the same day or the lung lesion had initially been noted on a CT scan. For 54 subjects, there was a delay of ≥ 1 week in obtaining a chest CT scan. The mean (SD) time from placement of the Diagnostic Clinic consultation by the primary care provider (PCP) or other provider and the initial Diagnostic Clinic visit was 6.3 (4.4) days. The mean (SD) time from suspect imaging to diagnosis (primary outcome) was 22.6(16.6) days.

The distribution of this outcome was nonnormal (Kolmogorov-Smirnov test P < .01). When compared with the standard of 30 days, the primary outcome of 22.6 days was significantly shorter (2-sided Mann–Whitney U test P < .01). Three-quarters (76.1%) of subjects were diagnosed within 30 days and 95.0% of subjects were diagnosed within 60 days of the initial imaging. For the 8 subjects diagnosed after 60 days, contributing factors included PCP delay in Diagnostic Clinic consultation, initial negative biopsy, delay in performance of chest CT scan prior to consultation, and outsourcing of positron emission tomography (PET) scans.

Overall, 57 (35.8%) of the subjects underwent biopsy on the day of their Diagnostic Clinic visit: 14 underwent CT-guided biopsy and 43 underwent EBUS/EUS. Within 2 days of the initial visit 106 subjects (66.7%) had undergone biopsy. The mean (SD) time from initial Diagnostic Clinic visit to biopsy was 6.3 (9.5) days. The mean (SD) interval was 1.8 (3.0) days for EBUS/ EUS and 11.3 (11.7) days for CT-guided biopsy. The mean (SD) interval from biopsy to diagnosis was 3.2 (6.2) days with 64 cases (40.3%) diagnosed the day of biopsy.

Excluding subjects whose treatment was delayed by patient choice or intercurrent illness, and those who left the VA system to seek treatment elsewhere (n = 21), 24 opted for palliative care, 5 died before treatment could be initiated, and 109 underwent treatment for their tumors (Table). The mean times (SD) from diagnosis to treatment were: chemotherapy alone 34.7 (25.3) days; chemoradiation 37.0 (22.8) days; surgery 44.3 (24.4) days; radiation therapy alone 47.9 (26.0) days. With respect to the RAND Corporation recommended diagnosis to treatment time, 60.9% of chemotherapy alone, 61.5% of chemoradiation, 66.7% of surgery, and 45.0% of radiation therapy alone treatments were initiated within the 6-week window.

Discussion

This retrospective case study demonstrates the effectiveness of a dedicated diagnostic clinic with priority EBUS/EUS access in diagnosing lung cancer within the VA system. Although there is no universally accepted quality standard for comparison, the RAND Corporation recommendation of 60 days from abnormal imaging to diagnosis and the Dayton VAMC published mean of 35.5 days are guideposts; however, the results from the Dayton VAMC may have been affected negatively by some subjects undergoing serial imaging for asymptomatic nodules. We chose a more stringent standard of 30 days as recommended by Swedish and Canadian task forces.

When diagnosing lung cancer, the overriding purpose of the Diagnostic Clinic is to minimize system delays. The method is to have as simple a task as possible for the PCP or other provider who identifies a lung nodule or mass and submits a single consultation request to the Diagnostic Clinic. Once this consultation is placed, the clinic RN manager oversees all further steps required for diagnosis and referral for treatment. The key factor in achieving a mean diagnosis time of 22.6 days is the cooperation between the RN manager and the interventional pulmonologist. When a consultation is received, the RN manager and pulmonologist review the data together and schedule the initial clinic visit; the goal is same-day biopsy, which is achieved in more than one-third of cases. Not all patients with a chest image suspected for lung cancer had it ordered by their PCP. For this reason, a Diagnostic Clinic consultation is available to all health care providers in our system. Many patients reach the clinic after the discovery of a suspect chest X-ray during an emergency department visit, a regularly scheduled subspecialty appointment, or during a preoperative evaluation.

The mean time from initial visit to biopsy was 1.8 days for EBUS/EUS compared with an interval of 11.3 days for CT-guided biopsy. This difference reflects the pulmonologist’s involvement in initial scheduling of Diagnostic Clinic patients. The ability of the pulmonologist to provide an accurate assessment of sample adequacy and a preliminary diagnosis at bedside, with concurrent confirmation by a staff pathologist, permitted the Diagnostic Clinic to inform 40.3% of patients of the finding of malignancy on the day of biopsy. A published comparison of the onsite review of biopsy material showed our pulmonologist and staff pathologists to be equally accurate in their interpretations.¹¹

Sources of Delays

While this study documents the shortest intervals from suspect imaging to diagnosis reported to date, it also identifies sources of system delay in diagnosing lung cancer that JLMMVH could further optimize. The first is the time from initial abnormal chest X-ray imaging to performance of the chest CT scan. On occasion, the index lung lesion is identified unexpectedly on an outpatient or emergency department chest CT scan. With greater use of LDCT lung cancer screening, the initial detection of suspect lesions by CT scanning will increase in the future. However, the PCP most often investigates a patient complaint with a standard chest X-ray that reveals a suspect nodule or mass. When ordered by the PCP as an outpatient test, scheduling of the follow-up chest CT scan is not given priority. More than a third of subjects experienced a delay ≥ 1 week in obtaining a chest CT scan ordered by the PCP; for 29 subjects the delay was ≥ 3weeks. At JLMMVH, the Diagnostic Clinic is given priority in scheduling CT scans. Hence, for suspect lung lesions, the chest CT scan, if not already obtained, is generally performed on the morning of the clinic visit. Educating the PCP to refer the patient immediately to the Diagnostic Clinic rather than waiting to obtain an outpatient chest CT scan may remove this source of unnecessary delay.

Scheduling a CT-guided fine needle aspiration of a lung lesion is another source of system delay. When the chest CT scan is available at the time of the Diagnostic Clinic referral, the clinic visit is scheduled for the earliest day a required CT-guided biopsy can be performed. However, the mean time of 11.3 days from initial Diagnostic Clinic visit to CT-guided biopsy is indicative of the backlog faced by the interventional radiologists.

Although infrequent, PET scans that are required before biopsy can lead to substantial delays. PET scans are performed at our university affiliate, and the joint VA-university lung tumor board sometimes generates requests for such scans prior to tissue diagnosis, yet another source of delay.

The time from referral receipt to the Diagnostic Clinic visit averaged 6.3 days. This delay usually was determined by the availability of the CT-guided biopsy or the dedicated interventional pulmonologist. Although other interventional pulmonologists at JLMMVH may perform the requisite diagnostic procedures, they are not always available for immediate review of imaging studies of referred patients nor can their schedules flexibly accommodate the number of patients seen in our clinic for evaluation.

Lung Cancer Diagnosis

Prompt diagnosis in the setting of a worrisome chest X-ray may help decrease patient anxiety, but does the clinic improve lung cancer treatment outcomes? Such improvement has been demonstrated only in stage IA squamous cell lung cancer.⁹ Of our study population, 37.7% had squamous cell carcinoma, and 85.5% had non-small cell lung cancer. Of those with non-small cell lung cancer, 28.9% had a clinical stage I tumor. Stage I squamous cell carcinoma, the type of tumor most likely to benefit from early diagnosis and treatment, was diagnosed in 11.3% of patients. With the increased application of LDCT screening, the proportion of veterans identified with early stage lung cancer may rise. The Providence VAMC in Rhode Island reported its results from instituting LDCT screening.¹² Prior to screening, 28% of patients diagnosed with lung cancer had a stage I tumor. Following the introduction of LDCT screening, 49% diagnosed by LDCT screening had a stage I tumor. Nearly a third of their patients diagnosed with lung cancer through LDCT screening had squamous cell tumor histology. Thus, we can anticipate an increasing number of veterans with early stage lung cancer who would benefit from timely diagnosis.

The JLMMVH is a referral center for the entire state of Arkansas. Quite a few of its referred patients come from a long distance, which may require overnight housing and other related travel expenses. Apart from any potential outcome benefit, the efficiencies of the system described herein include the minimization of extra trips, an inconvenience and cost to both patient and JLMMVH.

Although the primary task of the clinic is diagnosis, we also seek to facilitate timely treatment. Our lack of an on-site PET scanner and radiation therapy, resources present on-site at the Dayton VAMC, contribute to longer therapy wait times. The shortest mean wait time at JLMMVH is for chemotherapy alone (34.7 days), in part because the JLMMVH oncologists, performing initial consultations 2 to 3 times weekly in the Diagnostic Clinic, are more readily available than are our thoracic surgeons or radiation therapists. Yet overall, JLMMVH patients often face delay from the time of lung cancer diagnosis to initiation of treatment.

The Connecticut Veterans Affairs Healthcare System has published the results of changes in lung cancer management associated with a nurse navigator system.10 Prior to creating the position of cancer care coordinator, filled by an advanced practice RNs, the mean time from clinical suspicion of lung cancer to treatment was 117 days. After 4 years of such care navigation, this waiting time had decreased to 52.4 days. Associated with this dramatic improvement in overall waiting time were decreases in the turnaround time required for performance of CT and PET scans. With respect to this big picture view of lung cancer care, our Diagnostic Clinic serves as a model for the initial step of diagnosis. Coordination and streamlining of the various steps from diagnosis to definitive therapy shall require a more system-wide effort involving all the key players in cancer care.

Conclusion

We have developed a care pathway based in a dedicated diagnostic clinic and have been able to document the shortest interval from abnormality to diagnosis of lung cancer reported in the literature to date. Efficient functioning of this clinic is dependent upon the close cooperation between a full-time RN clinic manager and an interventional pulmonologist experienced in lung cancer management and able to interpret cytologic samples at the time of biopsy. Shortening the delay between diagnosis and definitive therapy remains a challenge and may benefit from the oncology nurse navigator model previously described within the VA system. ¹⁰

Lung cancer is the leading cause of cancer death in the US, with 154 050 deaths in 2018.¹ There have been many attempts to reduce mortality of the disease through early diagnosis with use of computed tomography (CT). The National Lung Cancer Screening trial showed that screening high-risk populations with low-dose CT (LDCT) can reduce mortality.² However, implementing LDCT screening in the clinical setting has proven challenging, as illustrated by the VA Lung Cancer Screening Demonstration Project (LCSDP).³ A lung cancer diagnosis typically comprises several steps that require different medical specialties; this can lead to delays. In the LCSDP, the mean time to diagnosis was 137 days.³ There are no federal standards for timeliness of lung cancer diagnosis.

The nonprofit RAND Corporation is the only American research organization that has published guidelines specifying acceptable intervals for the diagnosis and treatment of lung cancer. In Quality of Care for Oncologic Conditions and HIV, RAND Corporation researchers propose management quality indicators: lung cancer diagnosis within 2 months of an abnormal radiologic study and treatment within 6 weeks of diagnosis.⁴ The Swedish Lung Cancer Study⁵ and the Canadian Strategy for Cancer Control⁶ both recommended a standard of about 30 days—half the time recommended by the RAND Corporation.

Bukhari and colleagues at the Dayton US Department of Veterans Affairs (VA) Medical Center (VAMC) conducted a quality improvement study that examined lung cancer diagnosis and management.7 They found the time (SD) from abnormal chest imaging to diagnosis was 35.5 (31.6) days. Of those veterans who received a lung cancer diagnosis, 89.2% had the diagnosis made within the 60 days recommended by the RAND Corporation. Although these results surpass those of the LCSDP, they can be exceeded.

Beyond the potential emotional distress of awaiting the final diagnosis of a lung lesion, a delay in diagnosis and treatment may adversely affect outcomes. LDCT screening has been shown to reduce mortality, which implies a link between survival and time to intervention. There is no published evidence that time to diagnosis in advanced stage lung cancer affects outcome. The National Cancer Database (NCDB) contains informtion on about 70% of the cancers diagnosed each year in the US.8 An analysis of 4984 patients with stage IA squamous cell lung cancer undergoing lobectomy from NCDB showed that earlier surgery was associated with an absolute decrease in 5-year mortality of 5% to 8%. 9 Hence, at least in early-stage disease, reduced time from initial suspect imaging to definitive treatment may improve survival.

A system that coordinates the requisite diagnostic steps and avoids delays should provide a significant improvement in patient care. The results of such an approach that utilized nurse navigators has been previously published. 10 Here, we present the results of a dedicated VA referral clinic with priority access to pulmonary consultation and procedures in place that are designed to expedite the diagnosis of potential lung cancer.

Methods

The John L. McClellan Memorial Veterans Hospital (JLMMVH) in Little Rock, Arkansas institutional review board approved this study, which was performed in accordance with the Declaration of Helsinki. Requirement for informed consent was waived, and patient confidentiality was maintained throughout.

We have developed a plan of care specifically to facilitate diagnosis and treatment of the large number of veterans referred to the JLMMVH Diagnostic Clinic for abnormal results of chest imaging. The clinic has priority access to same-day imaging and subspecialty consultation services. In the clinic, medical students and residents perform evaluations and a registered nurse (RN) manager coordinates care.

A Diagnostic Clinic consult for abnormal thoracic imaging immediately triggers an e-consult to an interventional pulmonologist (Figure). The RN manager and pulmonologist perform a joint review of records/imaging prior to scheduling, and the pulmonologist triages the patient. Triage options include follow-up imaging, bronchoscopy with endobronchial ultrasound (EBUS), endoscopic ultrasound (EUS), and CT-guided biopsy.

The RN manager then schedules a clinic visit that includes a medical evaluation by clinic staff and any indicated procedures on the same day. The interventional pulmonologist performs EBUS, EUS with the convex curvilinear bronchoscope, or both combined as indicated for diagnosis and staging. All procedures are performed in the JLMMVH bronchoscopy suite with standard conscious sedation using midazolam and fentanyl. Any other relevant procedures, such as pleural tap, also are performed at time of procedure. The pulmonologist and an attending pathologist interpret biopsies obtained in the bronchoscopy suite.

We performed a retrospective chart review of patients diagnosed with primary lung cancer through referral to the JLMMVH Diagnostic Clinic. The primary outcome was time from initial suspect chest imaging to cancer diagnosis. The study population consisted of patients referred for abnormal thoracic imaging between January 1, 2013 and December 31, 2016 and subsequently diagnosed with a primary lung cancer.

Subjects were excluded if (1) the patient was referred from outside our care network and a delay of > 10 days occurred between initial lesion imaging and referral; (2) the patient did not show up for appointments or chose to delay evaluation following referral; (3) biopsy demonstrated a nonlung primary cancer; and (4) serious intercurrent illness interrupted the diagnostic plan. In some cases, the radiologist or consulting pulmonologist had judged the lung lesion too small for immediate biopsy and recommended repeat imaging at a later date.

Patients were included in the study if the follow- up imaging led to a lung cancer diagnosis. However, because the interval between the initial imaging and the follow-up imaging in these patients did not represent a systems delay problem, the date of the scheduled follow-up abnormal imaging, which resulted in initiation of a potential cancer evaluation, served as the index suspect imaging date for this study.

Patient electronic medical records were reviewed and the following data were abstracted: date of the abnormal imaging that led to referral and time from abnormal chest X-ray to chest CT scan if applicable; date of referral and date of clinic visit; date of biopsy; date of lung cancer diagnosis; method of obtaining diagnostic specimen; lung cancer type and stage; type and date of treatment initiation or decision for supportive care only; and decision to seek further evaluation or care outside of our system.

All patients diagnosed with lung cancer during the study period were reviewed for inclusion, hence no required sample-size estimate was calculated. All outcomes were assessed as calendar days. The primary outcome was the time from the index suspect chest imaging study to the date of diagnosis of lung cancer. Prior to the initiation of our study, we chose this more stringent 30-day recommendation of the Canadian6 and Swedish5 studies as the comparator for our primary outcome, although data with respect to the 60-day Rand Corporation guidelines also are reported.⁴

Statistical Methods

The mean time to lung cancer diagnosis in our cohort was compared with this 30-day standard using a 2-sided Mann–Whitney U test. Normality of data distribution was determined using the Kolmogorov–Smirnov test. For statistical significance testing a P value of .05 was used. Statistical calculations were performed using R statistical software version 3.2.4. Secondary outcomes consisted of time from diagnosis to treatment; proportion of subjects diagnosed within 60 days; time from initial clinic visit to biopsy; and time from biopsy to diagnosis.

Results

Overall, 222 patients were diagnosed with a malignant lung lesion, of which 63 were excluded from analysis: 22 cancelled or did not appear for appointments, declined further evaluation, or completed evaluation outside of our network; 13 had the diagnosis made prior to Diagnostic Clinic visit; 13 proved to have a nonlung primary tumor presenting in the lung or mediastinal nodes; 12 were delayed > 10 days in referral from an outside network; and 3 had an intervening serious acute medical problem forcing delay in the diagnostic process.

Of the 159 included subjects, 154 (96.9%) were male, and the mean (SD) age was 67.6 (8.1) years. For 76 subjects, the abnormal chest X-ray and subsequent chest CT scan were performed the same day or the lung lesion had initially been noted on a CT scan. For 54 subjects, there was a delay of ≥ 1 week in obtaining a chest CT scan. The mean (SD) time from placement of the Diagnostic Clinic consultation by the primary care provider (PCP) or other provider and the initial Diagnostic Clinic visit was 6.3 (4.4) days. The mean (SD) time from suspect imaging to diagnosis (primary outcome) was 22.6(16.6) days.

The distribution of this outcome was nonnormal (Kolmogorov-Smirnov test P < .01). When compared with the standard of 30 days, the primary outcome of 22.6 days was significantly shorter (2-sided Mann–Whitney U test P < .01). Three-quarters (76.1%) of subjects were diagnosed within 30 days and 95.0% of subjects were diagnosed within 60 days of the initial imaging. For the 8 subjects diagnosed after 60 days, contributing factors included PCP delay in Diagnostic Clinic consultation, initial negative biopsy, delay in performance of chest CT scan prior to consultation, and outsourcing of positron emission tomography (PET) scans.

Overall, 57 (35.8%) of the subjects underwent biopsy on the day of their Diagnostic Clinic visit: 14 underwent CT-guided biopsy and 43 underwent EBUS/EUS. Within 2 days of the initial visit 106 subjects (66.7%) had undergone biopsy. The mean (SD) time from initial Diagnostic Clinic visit to biopsy was 6.3 (9.5) days. The mean (SD) interval was 1.8 (3.0) days for EBUS/ EUS and 11.3 (11.7) days for CT-guided biopsy. The mean (SD) interval from biopsy to diagnosis was 3.2 (6.2) days with 64 cases (40.3%) diagnosed the day of biopsy.

Excluding subjects whose treatment was delayed by patient choice or intercurrent illness, and those who left the VA system to seek treatment elsewhere (n = 21), 24 opted for palliative care, 5 died before treatment could be initiated, and 109 underwent treatment for their tumors (Table). The mean times (SD) from diagnosis to treatment were: chemotherapy alone 34.7 (25.3) days; chemoradiation 37.0 (22.8) days; surgery 44.3 (24.4) days; radiation therapy alone 47.9 (26.0) days. With respect to the RAND Corporation recommended diagnosis to treatment time, 60.9% of chemotherapy alone, 61.5% of chemoradiation, 66.7% of surgery, and 45.0% of radiation therapy alone treatments were initiated within the 6-week window.

Discussion

This retrospective case study demonstrates the effectiveness of a dedicated diagnostic clinic with priority EBUS/EUS access in diagnosing lung cancer within the VA system. Although there is no universally accepted quality standard for comparison, the RAND Corporation recommendation of 60 days from abnormal imaging to diagnosis and the Dayton VAMC published mean of 35.5 days are guideposts; however, the results from the Dayton VAMC may have been affected negatively by some subjects undergoing serial imaging for asymptomatic nodules. We chose a more stringent standard of 30 days as recommended by Swedish and Canadian task forces.

When diagnosing lung cancer, the overriding purpose of the Diagnostic Clinic is to minimize system delays. The method is to have as simple a task as possible for the PCP or other provider who identifies a lung nodule or mass and submits a single consultation request to the Diagnostic Clinic. Once this consultation is placed, the clinic RN manager oversees all further steps required for diagnosis and referral for treatment. The key factor in achieving a mean diagnosis time of 22.6 days is the cooperation between the RN manager and the interventional pulmonologist. When a consultation is received, the RN manager and pulmonologist review the data together and schedule the initial clinic visit; the goal is same-day biopsy, which is achieved in more than one-third of cases. Not all patients with a chest image suspected for lung cancer had it ordered by their PCP. For this reason, a Diagnostic Clinic consultation is available to all health care providers in our system. Many patients reach the clinic after the discovery of a suspect chest X-ray during an emergency department visit, a regularly scheduled subspecialty appointment, or during a preoperative evaluation.

The mean time from initial visit to biopsy was 1.8 days for EBUS/EUS compared with an interval of 11.3 days for CT-guided biopsy. This difference reflects the pulmonologist’s involvement in initial scheduling of Diagnostic Clinic patients. The ability of the pulmonologist to provide an accurate assessment of sample adequacy and a preliminary diagnosis at bedside, with concurrent confirmation by a staff pathologist, permitted the Diagnostic Clinic to inform 40.3% of patients of the finding of malignancy on the day of biopsy. A published comparison of the onsite review of biopsy material showed our pulmonologist and staff pathologists to be equally accurate in their interpretations.¹¹

Sources of Delays

While this study documents the shortest intervals from suspect imaging to diagnosis reported to date, it also identifies sources of system delay in diagnosing lung cancer that JLMMVH could further optimize. The first is the time from initial abnormal chest X-ray imaging to performance of the chest CT scan. On occasion, the index lung lesion is identified unexpectedly on an outpatient or emergency department chest CT scan. With greater use of LDCT lung cancer screening, the initial detection of suspect lesions by CT scanning will increase in the future. However, the PCP most often investigates a patient complaint with a standard chest X-ray that reveals a suspect nodule or mass. When ordered by the PCP as an outpatient test, scheduling of the follow-up chest CT scan is not given priority. More than a third of subjects experienced a delay ≥ 1 week in obtaining a chest CT scan ordered by the PCP; for 29 subjects the delay was ≥ 3weeks. At JLMMVH, the Diagnostic Clinic is given priority in scheduling CT scans. Hence, for suspect lung lesions, the chest CT scan, if not already obtained, is generally performed on the morning of the clinic visit. Educating the PCP to refer the patient immediately to the Diagnostic Clinic rather than waiting to obtain an outpatient chest CT scan may remove this source of unnecessary delay.

Scheduling a CT-guided fine needle aspiration of a lung lesion is another source of system delay. When the chest CT scan is available at the time of the Diagnostic Clinic referral, the clinic visit is scheduled for the earliest day a required CT-guided biopsy can be performed. However, the mean time of 11.3 days from initial Diagnostic Clinic visit to CT-guided biopsy is indicative of the backlog faced by the interventional radiologists.

Although infrequent, PET scans that are required before biopsy can lead to substantial delays. PET scans are performed at our university affiliate, and the joint VA-university lung tumor board sometimes generates requests for such scans prior to tissue diagnosis, yet another source of delay.

The time from referral receipt to the Diagnostic Clinic visit averaged 6.3 days. This delay usually was determined by the availability of the CT-guided biopsy or the dedicated interventional pulmonologist. Although other interventional pulmonologists at JLMMVH may perform the requisite diagnostic procedures, they are not always available for immediate review of imaging studies of referred patients nor can their schedules flexibly accommodate the number of patients seen in our clinic for evaluation.

Lung Cancer Diagnosis

Prompt diagnosis in the setting of a worrisome chest X-ray may help decrease patient anxiety, but does the clinic improve lung cancer treatment outcomes? Such improvement has been demonstrated only in stage IA squamous cell lung cancer.⁹ Of our study population, 37.7% had squamous cell carcinoma, and 85.5% had non-small cell lung cancer. Of those with non-small cell lung cancer, 28.9% had a clinical stage I tumor. Stage I squamous cell carcinoma, the type of tumor most likely to benefit from early diagnosis and treatment, was diagnosed in 11.3% of patients. With the increased application of LDCT screening, the proportion of veterans identified with early stage lung cancer may rise. The Providence VAMC in Rhode Island reported its results from instituting LDCT screening.¹² Prior to screening, 28% of patients diagnosed with lung cancer had a stage I tumor. Following the introduction of LDCT screening, 49% diagnosed by LDCT screening had a stage I tumor. Nearly a third of their patients diagnosed with lung cancer through LDCT screening had squamous cell tumor histology. Thus, we can anticipate an increasing number of veterans with early stage lung cancer who would benefit from timely diagnosis.

The JLMMVH is a referral center for the entire state of Arkansas. Quite a few of its referred patients come from a long distance, which may require overnight housing and other related travel expenses. Apart from any potential outcome benefit, the efficiencies of the system described herein include the minimization of extra trips, an inconvenience and cost to both patient and JLMMVH.

Although the primary task of the clinic is diagnosis, we also seek to facilitate timely treatment. Our lack of an on-site PET scanner and radiation therapy, resources present on-site at the Dayton VAMC, contribute to longer therapy wait times. The shortest mean wait time at JLMMVH is for chemotherapy alone (34.7 days), in part because the JLMMVH oncologists, performing initial consultations 2 to 3 times weekly in the Diagnostic Clinic, are more readily available than are our thoracic surgeons or radiation therapists. Yet overall, JLMMVH patients often face delay from the time of lung cancer diagnosis to initiation of treatment.

The Connecticut Veterans Affairs Healthcare System has published the results of changes in lung cancer management associated with a nurse navigator system.10 Prior to creating the position of cancer care coordinator, filled by an advanced practice RNs, the mean time from clinical suspicion of lung cancer to treatment was 117 days. After 4 years of such care navigation, this waiting time had decreased to 52.4 days. Associated with this dramatic improvement in overall waiting time were decreases in the turnaround time required for performance of CT and PET scans. With respect to this big picture view of lung cancer care, our Diagnostic Clinic serves as a model for the initial step of diagnosis. Coordination and streamlining of the various steps from diagnosis to definitive therapy shall require a more system-wide effort involving all the key players in cancer care.

Conclusion

We have developed a care pathway based in a dedicated diagnostic clinic and have been able to document the shortest interval from abnormality to diagnosis of lung cancer reported in the literature to date. Efficient functioning of this clinic is dependent upon the close cooperation between a full-time RN clinic manager and an interventional pulmonologist experienced in lung cancer management and able to interpret cytologic samples at the time of biopsy. Shortening the delay between diagnosis and definitive therapy remains a challenge and may benefit from the oncology nurse navigator model previously described within the VA system. ¹⁰

Lung cancer is the leading cause of cancer death in the US, with 154 050 deaths in 2018.¹ There have been many attempts to reduce mortality of the disease through early diagnosis with use of computed tomography (CT). The National Lung Cancer Screening trial showed that screening high-risk populations with low-dose CT (LDCT) can reduce mortality.² However, implementing LDCT screening in the clinical setting has proven challenging, as illustrated by the VA Lung Cancer Screening Demonstration Project (LCSDP).³ A lung cancer diagnosis typically comprises several steps that require different medical specialties; this can lead to delays. In the LCSDP, the mean time to diagnosis was 137 days.³ There are no federal standards for timeliness of lung cancer diagnosis.

The nonprofit RAND Corporation is the only American research organization that has published guidelines specifying acceptable intervals for the diagnosis and treatment of lung cancer. In Quality of Care for Oncologic Conditions and HIV, RAND Corporation researchers propose management quality indicators: lung cancer diagnosis within 2 months of an abnormal radiologic study and treatment within 6 weeks of diagnosis.⁴ The Swedish Lung Cancer Study⁵ and the Canadian Strategy for Cancer Control⁶ both recommended a standard of about 30 days—half the time recommended by the RAND Corporation.

Bukhari and colleagues at the Dayton US Department of Veterans Affairs (VA) Medical Center (VAMC) conducted a quality improvement study that examined lung cancer diagnosis and management.7 They found the time (SD) from abnormal chest imaging to diagnosis was 35.5 (31.6) days. Of those veterans who received a lung cancer diagnosis, 89.2% had the diagnosis made within the 60 days recommended by the RAND Corporation. Although these results surpass those of the LCSDP, they can be exceeded.

Beyond the potential emotional distress of awaiting the final diagnosis of a lung lesion, a delay in diagnosis and treatment may adversely affect outcomes. LDCT screening has been shown to reduce mortality, which implies a link between survival and time to intervention. There is no published evidence that time to diagnosis in advanced stage lung cancer affects outcome. The National Cancer Database (NCDB) contains informtion on about 70% of the cancers diagnosed each year in the US.8 An analysis of 4984 patients with stage IA squamous cell lung cancer undergoing lobectomy from NCDB showed that earlier surgery was associated with an absolute decrease in 5-year mortality of 5% to 8%. 9 Hence, at least in early-stage disease, reduced time from initial suspect imaging to definitive treatment may improve survival.

A system that coordinates the requisite diagnostic steps and avoids delays should provide a significant improvement in patient care. The results of such an approach that utilized nurse navigators has been previously published. 10 Here, we present the results of a dedicated VA referral clinic with priority access to pulmonary consultation and procedures in place that are designed to expedite the diagnosis of potential lung cancer.

Methods

The John L. McClellan Memorial Veterans Hospital (JLMMVH) in Little Rock, Arkansas institutional review board approved this study, which was performed in accordance with the Declaration of Helsinki. Requirement for informed consent was waived, and patient confidentiality was maintained throughout.

We have developed a plan of care specifically to facilitate diagnosis and treatment of the large number of veterans referred to the JLMMVH Diagnostic Clinic for abnormal results of chest imaging. The clinic has priority access to same-day imaging and subspecialty consultation services. In the clinic, medical students and residents perform evaluations and a registered nurse (RN) manager coordinates care.

A Diagnostic Clinic consult for abnormal thoracic imaging immediately triggers an e-consult to an interventional pulmonologist (Figure). The RN manager and pulmonologist perform a joint review of records/imaging prior to scheduling, and the pulmonologist triages the patient. Triage options include follow-up imaging, bronchoscopy with endobronchial ultrasound (EBUS), endoscopic ultrasound (EUS), and CT-guided biopsy.

The RN manager then schedules a clinic visit that includes a medical evaluation by clinic staff and any indicated procedures on the same day. The interventional pulmonologist performs EBUS, EUS with the convex curvilinear bronchoscope, or both combined as indicated for diagnosis and staging. All procedures are performed in the JLMMVH bronchoscopy suite with standard conscious sedation using midazolam and fentanyl. Any other relevant procedures, such as pleural tap, also are performed at time of procedure. The pulmonologist and an attending pathologist interpret biopsies obtained in the bronchoscopy suite.

We performed a retrospective chart review of patients diagnosed with primary lung cancer through referral to the JLMMVH Diagnostic Clinic. The primary outcome was time from initial suspect chest imaging to cancer diagnosis. The study population consisted of patients referred for abnormal thoracic imaging between January 1, 2013 and December 31, 2016 and subsequently diagnosed with a primary lung cancer.

Subjects were excluded if (1) the patient was referred from outside our care network and a delay of > 10 days occurred between initial lesion imaging and referral; (2) the patient did not show up for appointments or chose to delay evaluation following referral; (3) biopsy demonstrated a nonlung primary cancer; and (4) serious intercurrent illness interrupted the diagnostic plan. In some cases, the radiologist or consulting pulmonologist had judged the lung lesion too small for immediate biopsy and recommended repeat imaging at a later date.

Patients were included in the study if the follow- up imaging led to a lung cancer diagnosis. However, because the interval between the initial imaging and the follow-up imaging in these patients did not represent a systems delay problem, the date of the scheduled follow-up abnormal imaging, which resulted in initiation of a potential cancer evaluation, served as the index suspect imaging date for this study.

Patient electronic medical records were reviewed and the following data were abstracted: date of the abnormal imaging that led to referral and time from abnormal chest X-ray to chest CT scan if applicable; date of referral and date of clinic visit; date of biopsy; date of lung cancer diagnosis; method of obtaining diagnostic specimen; lung cancer type and stage; type and date of treatment initiation or decision for supportive care only; and decision to seek further evaluation or care outside of our system.

All patients diagnosed with lung cancer during the study period were reviewed for inclusion, hence no required sample-size estimate was calculated. All outcomes were assessed as calendar days. The primary outcome was the time from the index suspect chest imaging study to the date of diagnosis of lung cancer. Prior to the initiation of our study, we chose this more stringent 30-day recommendation of the Canadian6 and Swedish5 studies as the comparator for our primary outcome, although data with respect to the 60-day Rand Corporation guidelines also are reported.⁴

Statistical Methods

The mean time to lung cancer diagnosis in our cohort was compared with this 30-day standard using a 2-sided Mann–Whitney U test. Normality of data distribution was determined using the Kolmogorov–Smirnov test. For statistical significance testing a P value of .05 was used. Statistical calculations were performed using R statistical software version 3.2.4. Secondary outcomes consisted of time from diagnosis to treatment; proportion of subjects diagnosed within 60 days; time from initial clinic visit to biopsy; and time from biopsy to diagnosis.

Results

Overall, 222 patients were diagnosed with a malignant lung lesion, of which 63 were excluded from analysis: 22 cancelled or did not appear for appointments, declined further evaluation, or completed evaluation outside of our network; 13 had the diagnosis made prior to Diagnostic Clinic visit; 13 proved to have a nonlung primary tumor presenting in the lung or mediastinal nodes; 12 were delayed > 10 days in referral from an outside network; and 3 had an intervening serious acute medical problem forcing delay in the diagnostic process.

Of the 159 included subjects, 154 (96.9%) were male, and the mean (SD) age was 67.6 (8.1) years. For 76 subjects, the abnormal chest X-ray and subsequent chest CT scan were performed the same day or the lung lesion had initially been noted on a CT scan. For 54 subjects, there was a delay of ≥ 1 week in obtaining a chest CT scan. The mean (SD) time from placement of the Diagnostic Clinic consultation by the primary care provider (PCP) or other provider and the initial Diagnostic Clinic visit was 6.3 (4.4) days. The mean (SD) time from suspect imaging to diagnosis (primary outcome) was 22.6(16.6) days.

The distribution of this outcome was nonnormal (Kolmogorov-Smirnov test P < .01). When compared with the standard of 30 days, the primary outcome of 22.6 days was significantly shorter (2-sided Mann–Whitney U test P < .01). Three-quarters (76.1%) of subjects were diagnosed within 30 days and 95.0% of subjects were diagnosed within 60 days of the initial imaging. For the 8 subjects diagnosed after 60 days, contributing factors included PCP delay in Diagnostic Clinic consultation, initial negative biopsy, delay in performance of chest CT scan prior to consultation, and outsourcing of positron emission tomography (PET) scans.

Overall, 57 (35.8%) of the subjects underwent biopsy on the day of their Diagnostic Clinic visit: 14 underwent CT-guided biopsy and 43 underwent EBUS/EUS. Within 2 days of the initial visit 106 subjects (66.7%) had undergone biopsy. The mean (SD) time from initial Diagnostic Clinic visit to biopsy was 6.3 (9.5) days. The mean (SD) interval was 1.8 (3.0) days for EBUS/ EUS and 11.3 (11.7) days for CT-guided biopsy. The mean (SD) interval from biopsy to diagnosis was 3.2 (6.2) days with 64 cases (40.3%) diagnosed the day of biopsy.

Excluding subjects whose treatment was delayed by patient choice or intercurrent illness, and those who left the VA system to seek treatment elsewhere (n = 21), 24 opted for palliative care, 5 died before treatment could be initiated, and 109 underwent treatment for their tumors (Table). The mean times (SD) from diagnosis to treatment were: chemotherapy alone 34.7 (25.3) days; chemoradiation 37.0 (22.8) days; surgery 44.3 (24.4) days; radiation therapy alone 47.9 (26.0) days. With respect to the RAND Corporation recommended diagnosis to treatment time, 60.9% of chemotherapy alone, 61.5% of chemoradiation, 66.7% of surgery, and 45.0% of radiation therapy alone treatments were initiated within the 6-week window.

Discussion

This retrospective case study demonstrates the effectiveness of a dedicated diagnostic clinic with priority EBUS/EUS access in diagnosing lung cancer within the VA system. Although there is no universally accepted quality standard for comparison, the RAND Corporation recommendation of 60 days from abnormal imaging to diagnosis and the Dayton VAMC published mean of 35.5 days are guideposts; however, the results from the Dayton VAMC may have been affected negatively by some subjects undergoing serial imaging for asymptomatic nodules. We chose a more stringent standard of 30 days as recommended by Swedish and Canadian task forces.

When diagnosing lung cancer, the overriding purpose of the Diagnostic Clinic is to minimize system delays. The method is to have as simple a task as possible for the PCP or other provider who identifies a lung nodule or mass and submits a single consultation request to the Diagnostic Clinic. Once this consultation is placed, the clinic RN manager oversees all further steps required for diagnosis and referral for treatment. The key factor in achieving a mean diagnosis time of 22.6 days is the cooperation between the RN manager and the interventional pulmonologist. When a consultation is received, the RN manager and pulmonologist review the data together and schedule the initial clinic visit; the goal is same-day biopsy, which is achieved in more than one-third of cases. Not all patients with a chest image suspected for lung cancer had it ordered by their PCP. For this reason, a Diagnostic Clinic consultation is available to all health care providers in our system. Many patients reach the clinic after the discovery of a suspect chest X-ray during an emergency department visit, a regularly scheduled subspecialty appointment, or during a preoperative evaluation.

The mean time from initial visit to biopsy was 1.8 days for EBUS/EUS compared with an interval of 11.3 days for CT-guided biopsy. This difference reflects the pulmonologist’s involvement in initial scheduling of Diagnostic Clinic patients. The ability of the pulmonologist to provide an accurate assessment of sample adequacy and a preliminary diagnosis at bedside, with concurrent confirmation by a staff pathologist, permitted the Diagnostic Clinic to inform 40.3% of patients of the finding of malignancy on the day of biopsy. A published comparison of the onsite review of biopsy material showed our pulmonologist and staff pathologists to be equally accurate in their interpretations.¹¹

Sources of Delays

While this study documents the shortest intervals from suspect imaging to diagnosis reported to date, it also identifies sources of system delay in diagnosing lung cancer that JLMMVH could further optimize. The first is the time from initial abnormal chest X-ray imaging to performance of the chest CT scan. On occasion, the index lung lesion is identified unexpectedly on an outpatient or emergency department chest CT scan. With greater use of LDCT lung cancer screening, the initial detection of suspect lesions by CT scanning will increase in the future. However, the PCP most often investigates a patient complaint with a standard chest X-ray that reveals a suspect nodule or mass. When ordered by the PCP as an outpatient test, scheduling of the follow-up chest CT scan is not given priority. More than a third of subjects experienced a delay ≥ 1 week in obtaining a chest CT scan ordered by the PCP; for 29 subjects the delay was ≥ 3weeks. At JLMMVH, the Diagnostic Clinic is given priority in scheduling CT scans. Hence, for suspect lung lesions, the chest CT scan, if not already obtained, is generally performed on the morning of the clinic visit. Educating the PCP to refer the patient immediately to the Diagnostic Clinic rather than waiting to obtain an outpatient chest CT scan may remove this source of unnecessary delay.

Scheduling a CT-guided fine needle aspiration of a lung lesion is another source of system delay. When the chest CT scan is available at the time of the Diagnostic Clinic referral, the clinic visit is scheduled for the earliest day a required CT-guided biopsy can be performed. However, the mean time of 11.3 days from initial Diagnostic Clinic visit to CT-guided biopsy is indicative of the backlog faced by the interventional radiologists.

Although infrequent, PET scans that are required before biopsy can lead to substantial delays. PET scans are performed at our university affiliate, and the joint VA-university lung tumor board sometimes generates requests for such scans prior to tissue diagnosis, yet another source of delay.

The time from referral receipt to the Diagnostic Clinic visit averaged 6.3 days. This delay usually was determined by the availability of the CT-guided biopsy or the dedicated interventional pulmonologist. Although other interventional pulmonologists at JLMMVH may perform the requisite diagnostic procedures, they are not always available for immediate review of imaging studies of referred patients nor can their schedules flexibly accommodate the number of patients seen in our clinic for evaluation.

Lung Cancer Diagnosis

Prompt diagnosis in the setting of a worrisome chest X-ray may help decrease patient anxiety, but does the clinic improve lung cancer treatment outcomes? Such improvement has been demonstrated only in stage IA squamous cell lung cancer.⁹ Of our study population, 37.7% had squamous cell carcinoma, and 85.5% had non-small cell lung cancer. Of those with non-small cell lung cancer, 28.9% had a clinical stage I tumor. Stage I squamous cell carcinoma, the type of tumor most likely to benefit from early diagnosis and treatment, was diagnosed in 11.3% of patients. With the increased application of LDCT screening, the proportion of veterans identified with early stage lung cancer may rise. The Providence VAMC in Rhode Island reported its results from instituting LDCT screening.¹² Prior to screening, 28% of patients diagnosed with lung cancer had a stage I tumor. Following the introduction of LDCT screening, 49% diagnosed by LDCT screening had a stage I tumor. Nearly a third of their patients diagnosed with lung cancer through LDCT screening had squamous cell tumor histology. Thus, we can anticipate an increasing number of veterans with early stage lung cancer who would benefit from timely diagnosis.

The JLMMVH is a referral center for the entire state of Arkansas. Quite a few of its referred patients come from a long distance, which may require overnight housing and other related travel expenses. Apart from any potential outcome benefit, the efficiencies of the system described herein include the minimization of extra trips, an inconvenience and cost to both patient and JLMMVH.

Although the primary task of the clinic is diagnosis, we also seek to facilitate timely treatment. Our lack of an on-site PET scanner and radiation therapy, resources present on-site at the Dayton VAMC, contribute to longer therapy wait times. The shortest mean wait time at JLMMVH is for chemotherapy alone (34.7 days), in part because the JLMMVH oncologists, performing initial consultations 2 to 3 times weekly in the Diagnostic Clinic, are more readily available than are our thoracic surgeons or radiation therapists. Yet overall, JLMMVH patients often face delay from the time of lung cancer diagnosis to initiation of treatment.

The Connecticut Veterans Affairs Healthcare System has published the results of changes in lung cancer management associated with a nurse navigator system.10 Prior to creating the position of cancer care coordinator, filled by an advanced practice RNs, the mean time from clinical suspicion of lung cancer to treatment was 117 days. After 4 years of such care navigation, this waiting time had decreased to 52.4 days. Associated with this dramatic improvement in overall waiting time were decreases in the turnaround time required for performance of CT and PET scans. With respect to this big picture view of lung cancer care, our Diagnostic Clinic serves as a model for the initial step of diagnosis. Coordination and streamlining of the various steps from diagnosis to definitive therapy shall require a more system-wide effort involving all the key players in cancer care.

Conclusion

We have developed a care pathway based in a dedicated diagnostic clinic and have been able to document the shortest interval from abnormality to diagnosis of lung cancer reported in the literature to date. Efficient functioning of this clinic is dependent upon the close cooperation between a full-time RN clinic manager and an interventional pulmonologist experienced in lung cancer management and able to interpret cytologic samples at the time of biopsy. Shortening the delay between diagnosis and definitive therapy remains a challenge and may benefit from the oncology nurse navigator model previously described within the VA system. ¹⁰

A growing body of evidence shows that electrosurgical smoke contains both harmful chemicals as well as live material, including blood particles, bacteria, and viruses.¹ Both human immunodeficiency virus and human papillomavirus have been identified in surgical smoke plumes, and bacterial colony growth has been demonstrated from electrosurgical smoke specimens, specifically Staphylococcus, Corynebacterium, and Neisseria species.^2-8 Treating 1 g of tissue with electrocoagulation produces chemical by-products equivalent to burning 6 unfiltered cigarettes,⁹ which is twice the amount of chemical by-products produced by CO₂ laser vaporization of the same quantity of tissue. It is a common misconception that electrosurgical smoke is less hazardous than smoke produced by ablative CO₂ procedures.⁹ Many chemicals are present in electrosurgical smoke plumes, including nitriles, benzenes, carbon monoxide, hydrogen cyanide, indoles, phenols, pyridine, pyrrole, styrene, toluene, and xylene.^10-12 In animal model studies of rat lungs exposed to surgical smoke, pathologic changes, including interstitial pneumonia, bronchiolitis, and emphysema, have been shown in a dose-dependent manner.^1,13-16 Diseases and symptoms linked to inhalation of electrosurgical smoke in humans include anemia, eye irritation, hypoxia, dizziness, nasopharyngeal lesions, vomiting, sneezing, throat irritation, and weakness.^1,8,17-19 A study of 153 dermatology residents found that more than 70% reported receiving no formal education on the hazards of electrosurgical smoke.²⁰ Approximately 45% were unaware if they had access to smoke evacuation in rooms where electrosurgery was performed. More than 76% were concerned with the infectious risk of electrosurgical smoke, and more than 71% were concerned with its potential carcinogenic risk.²⁰

We surveyed dermatologists who perform skin surgery as well as staff members with respect to their experiences with electrosurgical smoke and to observe any difference that information on the potential hazards of electrosurgical smoke may have on their attitudes and preferences.

Materials and Methods

Survey Instrument
We developed a REDCap survey consisting of 17 questions that was approved by the executive committees of the American College of Mohs Surgery and the American Society for Dermatologic Surgery for distribution to their dermatologist memberships. It was emailed to eligible participants using their mailing lists. Although the survey was sent directly to member physicians, it was recommended that they forward the survey to their clinical staff to complete.

After responding to an initial set of survey questions, respondents were informed that there is growing evidence of potential harms of inhalation of surgical smoke. They then were asked the same series of survey questions in light of this information.

Statistical Analysis
Statistical analysis of the survey responses was then completed, and free-text responses as a final question of the survey were assessed for themes. Preintervention responses of staff and clinicians noticing smoke and being bothered by smoke were assessed using proportions and 95% confidence interval (CI) estimates of the proportions. On most questions, respondents could answer on a scale of 1 to 10. Responses of 5 to 10 on noticing smoke and 5 to 10 on being bothered or troubled by the smoke smell were grouped for analyses. A cross-tabulation using the Bhapkar test for marginal homogeneity was used to assess if information presented on potential smoke hazards changed responses. A Cochran-Mantel-Haenszel test for ordinal responses was used to determine differences between surgeons and staff. A McNemar test was used to determine statistical significance of change in responses to cost. Statistical analysis was performed using SAS version 9.

Results

There was a total of 443 responses to our questionnaire. Two respondents answered that they did not work in an office where skin surgery was performed, and 4 respondents did not answer any questions and were therefore excluded, leaving a total of 437 responses (402 physicians and 35 staff members). A summary of the characteristics of the respondents is shown in the Table. Some respondents did not answer each question, leading to fewer than 437 answers for some questions.

Two hundred eighty-two respondents (64.5%) never or very rarely used smoke evacuation during skin surgical procedures, and only 85 (19.5%) used smoke evacuation with nearly every case. The remaining respondents sometimes used smoke evacuation (Figure 1).

Prior to being presented with the potential dangers of electrosurgical smoke and using a value of 5 to 10 to determine if respondents noticed smoke, 54.4% (95% CI, 49.5%-59.1%) did notice intraoperative smoke during procedures. Using a value of 5 to 10 to indicate if respondents were bothered or troubled by the smoke smell, 35.5% (95% CI, 31.0%-40.2%) were bothered or troubled by intraoperative smoke prior to potential hazards being presented.

Regarding acceptable increase in cost per procedure for smoke evacuation at baseline, 68.9% of respondents favored additional cost; 57.8% of respondents chose the lowest cost grouping of $1 to $30. After being presented with information about the potential harm of intraoperative smoke, the respondents in favor of additional cost increased to 71.5%, which was a small but statistically significant change (P=.0075)(Figure 2).

After being presented with the potential risks of electrosurgical smoke, 29 more respondents answered that they were severely bothered by electrosurgical smoke, whereas 45 fewer respondents selected that they were not bothered or troubled at all by electrosurgical smoke (Figure 3). This difference was statistically significant (P<.0001). Fifteen more respondents answered that they would be much more likely to choose to work at a practice with smoke evacuation once the potential harm of electrosurgical smoke was introduced, and 11 were somewhat more likely to choose a practice with smoke evacuation (P<.0001).

Comment

Developing evidence of health risks associated with electrosurgical smoke plumes has led to an increasing interest in the use of smoke protection or remediation tools during surgical procedures. High-filtration face masks and smoke-evacuation devices protect physicians, staff members, and patients, as well as improve the patient’s clinical experience.

Our study was designed to query dermatologists who perform skin surgery as well as staff members with respect to their experiences with electrosurgical smoke and to observe any difference that information on the potential hazards of electrosurgical smoke may have on their attitudes and preferences. We received 437 responses to our survey (Table). At baseline, 54.4% of respondents noticed and 35.5% were bothered or troubled by the smoke smell produced during skin electrosurgery. These data were intuitively associated in a statistically significant manner with the use of smoke evacuation for respondents; those respondents who more commonly used smoke evacuation were bothered less by electrosurgical smoke, and those respondents who used smoke evacuation less often were more likely to notice and be bothered by surgical smoke.

Once our respondents were presented with the potentially harmful effects of electrosurgical smoke, they became significantly more likely to be bothered by electrosurgical smoke and to want to work in a practice where smoke evacuation was available. This information, however, did not change respondents’ satisfaction with their work environment, and no statistically significant differences were noted between physicians and staff.

At baseline, 68.9% of respondents favored additional cost for smoke evacuation, with approximately 58% favoring the lowest cost category we presented ($1–$30). After being presented with information about the potential dangers of electrosurgical smoke, 71.5% were in favor of increased cost for smoke evacuation, which was a small but statistically significant increase.

The open-comment section of the survey provided interesting insight into the opinions of our respondents on smoke remediation. It is important to note that statistical analysis cannot be performed with these data, and firm generalizable conclusions cannot be drawn from them; however, they reveal topics that may guide further research and policy and certainly merit mention. Of 437 respondents, 108 left free-text comments. Twenty-six percent were categorized as unqualified proponents (in favor of smoke remediation) and 45% as qualified proponents (defined as an individual who verbalized a desire for smoke remediation but also cited a factor limiting their ability to use it, such as cost or staff availability). Only 12% were firmly against smoke remediation, while the remaining 17% did not comment discernibly for or against smoke remediation, indicating that a majority (71% of our comment section respondents) were in favor of some type of smoke remediation, especially if obstacles such as cost could be addressed. Only a small minority was firmly against smoke remediation.

The comments section of our survey highlighted some of the concerns that dermatologic surgeons and their staff have with electrosurgical smoke evacuation. Thirty percent cited cost as an obstacle to use of these devices, and several comments raised concern about increasing overhead and regulatory demands placed on practices. Many indicated that, without sufficient evidence of the harm caused by electrosurgical smoke, regulation that forces use of smoke remediation devices would represent a costly unfunded mandate. Others referenced the logistical challenges of smoke evacuation and the need for staff assistance. Newer smoke-evacuation wands built into cautery pens address much of this concern regarding logistical and staff challenges and further allow the evacuator tip to be located where it is most effective: 1 cm to 2 in from the point of cautery.^21,22

Additionally, 12% of commenters noted that their patients were bothered by the smell of electrosurgical smoke, which is a point that requires further research and is the focus of a current randomized trial at our institution (ClinicalTrials.gov Identifier NCT02958826).

Our current study is limited in that it is a survey and therefore is subject to response bias. Further, some may assert that the hazards of electrosurgical smoke are not settled science, and although we agree with this point on some level, the study aim was not to prove risk but rather to assess current attitudes and see if awareness of a potential risk influenced those attitudes. Additionally, most responses were from physicians—only 35 responses were from nonphysician staff—so it may be difficult to generalize the findings of this study to staff. The large number of physician respondents, however, can be seen as a strength, and the findings are likely much more generalizable to providers who routinely perform clinic-based surgical procedures involving electrosurgery.

Conclusion

Our study shows that most dermatologists who perform skin surgery notice and are bothered by the smoke produced by electrosurgery to at least some extent. When presented with the possibility that inhaling electrosurgical smoke may be harmful, dermatologists were more likely to be bothered by electrosurgical smoke, more likely to prefer a practice environment where smoke evacuation was available, and more likely to be willing to bear additional cost for smoke evacuation. The free-text comments on our survey highlighted that many dermatologic surgeons are proponents of smoke evacuation but have concerns about cost and potential regulatory challenges associated with smoke evacuation, especially if the potential risks are not settled science. Many logistical concerns for smoke evacuation are addressed with the use of integrated devices. More research is needed to determine the health effects of the surgical smoke we are exposed to daily and the optimal way to limit any risk.

Acknowledgment
The authors would like to thank Richard W. Madsen, PhD (Columbia, Missouri), biostatistician, for his valuable guidance in the statistical analysis of data, interpretation of results, and editorial support in finalizing the manuscript.

A growing body of evidence shows that electrosurgical smoke contains both harmful chemicals as well as live material, including blood particles, bacteria, and viruses.¹ Both human immunodeficiency virus and human papillomavirus have been identified in surgical smoke plumes, and bacterial colony growth has been demonstrated from electrosurgical smoke specimens, specifically Staphylococcus, Corynebacterium, and Neisseria species.^2-8 Treating 1 g of tissue with electrocoagulation produces chemical by-products equivalent to burning 6 unfiltered cigarettes,⁹ which is twice the amount of chemical by-products produced by CO₂ laser vaporization of the same quantity of tissue. It is a common misconception that electrosurgical smoke is less hazardous than smoke produced by ablative CO₂ procedures.⁹ Many chemicals are present in electrosurgical smoke plumes, including nitriles, benzenes, carbon monoxide, hydrogen cyanide, indoles, phenols, pyridine, pyrrole, styrene, toluene, and xylene.^10-12 In animal model studies of rat lungs exposed to surgical smoke, pathologic changes, including interstitial pneumonia, bronchiolitis, and emphysema, have been shown in a dose-dependent manner.^1,13-16 Diseases and symptoms linked to inhalation of electrosurgical smoke in humans include anemia, eye irritation, hypoxia, dizziness, nasopharyngeal lesions, vomiting, sneezing, throat irritation, and weakness.^1,8,17-19 A study of 153 dermatology residents found that more than 70% reported receiving no formal education on the hazards of electrosurgical smoke.²⁰ Approximately 45% were unaware if they had access to smoke evacuation in rooms where electrosurgery was performed. More than 76% were concerned with the infectious risk of electrosurgical smoke, and more than 71% were concerned with its potential carcinogenic risk.²⁰

We surveyed dermatologists who perform skin surgery as well as staff members with respect to their experiences with electrosurgical smoke and to observe any difference that information on the potential hazards of electrosurgical smoke may have on their attitudes and preferences.

Materials and Methods

Survey Instrument
We developed a REDCap survey consisting of 17 questions that was approved by the executive committees of the American College of Mohs Surgery and the American Society for Dermatologic Surgery for distribution to their dermatologist memberships. It was emailed to eligible participants using their mailing lists. Although the survey was sent directly to member physicians, it was recommended that they forward the survey to their clinical staff to complete.

After responding to an initial set of survey questions, respondents were informed that there is growing evidence of potential harms of inhalation of surgical smoke. They then were asked the same series of survey questions in light of this information.

Statistical Analysis
Statistical analysis of the survey responses was then completed, and free-text responses as a final question of the survey were assessed for themes. Preintervention responses of staff and clinicians noticing smoke and being bothered by smoke were assessed using proportions and 95% confidence interval (CI) estimates of the proportions. On most questions, respondents could answer on a scale of 1 to 10. Responses of 5 to 10 on noticing smoke and 5 to 10 on being bothered or troubled by the smoke smell were grouped for analyses. A cross-tabulation using the Bhapkar test for marginal homogeneity was used to assess if information presented on potential smoke hazards changed responses. A Cochran-Mantel-Haenszel test for ordinal responses was used to determine differences between surgeons and staff. A McNemar test was used to determine statistical significance of change in responses to cost. Statistical analysis was performed using SAS version 9.

Results

There was a total of 443 responses to our questionnaire. Two respondents answered that they did not work in an office where skin surgery was performed, and 4 respondents did not answer any questions and were therefore excluded, leaving a total of 437 responses (402 physicians and 35 staff members). A summary of the characteristics of the respondents is shown in the Table. Some respondents did not answer each question, leading to fewer than 437 answers for some questions.

Two hundred eighty-two respondents (64.5%) never or very rarely used smoke evacuation during skin surgical procedures, and only 85 (19.5%) used smoke evacuation with nearly every case. The remaining respondents sometimes used smoke evacuation (Figure 1).

Prior to being presented with the potential dangers of electrosurgical smoke and using a value of 5 to 10 to determine if respondents noticed smoke, 54.4% (95% CI, 49.5%-59.1%) did notice intraoperative smoke during procedures. Using a value of 5 to 10 to indicate if respondents were bothered or troubled by the smoke smell, 35.5% (95% CI, 31.0%-40.2%) were bothered or troubled by intraoperative smoke prior to potential hazards being presented.

Regarding acceptable increase in cost per procedure for smoke evacuation at baseline, 68.9% of respondents favored additional cost; 57.8% of respondents chose the lowest cost grouping of $1 to $30. After being presented with information about the potential harm of intraoperative smoke, the respondents in favor of additional cost increased to 71.5%, which was a small but statistically significant change (P=.0075)(Figure 2).

After being presented with the potential risks of electrosurgical smoke, 29 more respondents answered that they were severely bothered by electrosurgical smoke, whereas 45 fewer respondents selected that they were not bothered or troubled at all by electrosurgical smoke (Figure 3). This difference was statistically significant (P<.0001). Fifteen more respondents answered that they would be much more likely to choose to work at a practice with smoke evacuation once the potential harm of electrosurgical smoke was introduced, and 11 were somewhat more likely to choose a practice with smoke evacuation (P<.0001).

Comment

Developing evidence of health risks associated with electrosurgical smoke plumes has led to an increasing interest in the use of smoke protection or remediation tools during surgical procedures. High-filtration face masks and smoke-evacuation devices protect physicians, staff members, and patients, as well as improve the patient’s clinical experience.

Our study was designed to query dermatologists who perform skin surgery as well as staff members with respect to their experiences with electrosurgical smoke and to observe any difference that information on the potential hazards of electrosurgical smoke may have on their attitudes and preferences. We received 437 responses to our survey (Table). At baseline, 54.4% of respondents noticed and 35.5% were bothered or troubled by the smoke smell produced during skin electrosurgery. These data were intuitively associated in a statistically significant manner with the use of smoke evacuation for respondents; those respondents who more commonly used smoke evacuation were bothered less by electrosurgical smoke, and those respondents who used smoke evacuation less often were more likely to notice and be bothered by surgical smoke.

Once our respondents were presented with the potentially harmful effects of electrosurgical smoke, they became significantly more likely to be bothered by electrosurgical smoke and to want to work in a practice where smoke evacuation was available. This information, however, did not change respondents’ satisfaction with their work environment, and no statistically significant differences were noted between physicians and staff.

At baseline, 68.9% of respondents favored additional cost for smoke evacuation, with approximately 58% favoring the lowest cost category we presented ($1–$30). After being presented with information about the potential dangers of electrosurgical smoke, 71.5% were in favor of increased cost for smoke evacuation, which was a small but statistically significant increase.

The open-comment section of the survey provided interesting insight into the opinions of our respondents on smoke remediation. It is important to note that statistical analysis cannot be performed with these data, and firm generalizable conclusions cannot be drawn from them; however, they reveal topics that may guide further research and policy and certainly merit mention. Of 437 respondents, 108 left free-text comments. Twenty-six percent were categorized as unqualified proponents (in favor of smoke remediation) and 45% as qualified proponents (defined as an individual who verbalized a desire for smoke remediation but also cited a factor limiting their ability to use it, such as cost or staff availability). Only 12% were firmly against smoke remediation, while the remaining 17% did not comment discernibly for or against smoke remediation, indicating that a majority (71% of our comment section respondents) were in favor of some type of smoke remediation, especially if obstacles such as cost could be addressed. Only a small minority was firmly against smoke remediation.

The comments section of our survey highlighted some of the concerns that dermatologic surgeons and their staff have with electrosurgical smoke evacuation. Thirty percent cited cost as an obstacle to use of these devices, and several comments raised concern about increasing overhead and regulatory demands placed on practices. Many indicated that, without sufficient evidence of the harm caused by electrosurgical smoke, regulation that forces use of smoke remediation devices would represent a costly unfunded mandate. Others referenced the logistical challenges of smoke evacuation and the need for staff assistance. Newer smoke-evacuation wands built into cautery pens address much of this concern regarding logistical and staff challenges and further allow the evacuator tip to be located where it is most effective: 1 cm to 2 in from the point of cautery.^21,22

Additionally, 12% of commenters noted that their patients were bothered by the smell of electrosurgical smoke, which is a point that requires further research and is the focus of a current randomized trial at our institution (ClinicalTrials.gov Identifier NCT02958826).

Our current study is limited in that it is a survey and therefore is subject to response bias. Further, some may assert that the hazards of electrosurgical smoke are not settled science, and although we agree with this point on some level, the study aim was not to prove risk but rather to assess current attitudes and see if awareness of a potential risk influenced those attitudes. Additionally, most responses were from physicians—only 35 responses were from nonphysician staff—so it may be difficult to generalize the findings of this study to staff. The large number of physician respondents, however, can be seen as a strength, and the findings are likely much more generalizable to providers who routinely perform clinic-based surgical procedures involving electrosurgery.

Conclusion

Our study shows that most dermatologists who perform skin surgery notice and are bothered by the smoke produced by electrosurgery to at least some extent. When presented with the possibility that inhaling electrosurgical smoke may be harmful, dermatologists were more likely to be bothered by electrosurgical smoke, more likely to prefer a practice environment where smoke evacuation was available, and more likely to be willing to bear additional cost for smoke evacuation. The free-text comments on our survey highlighted that many dermatologic surgeons are proponents of smoke evacuation but have concerns about cost and potential regulatory challenges associated with smoke evacuation, especially if the potential risks are not settled science. Many logistical concerns for smoke evacuation are addressed with the use of integrated devices. More research is needed to determine the health effects of the surgical smoke we are exposed to daily and the optimal way to limit any risk.

Acknowledgment
The authors would like to thank Richard W. Madsen, PhD (Columbia, Missouri), biostatistician, for his valuable guidance in the statistical analysis of data, interpretation of results, and editorial support in finalizing the manuscript.

A growing body of evidence shows that electrosurgical smoke contains both harmful chemicals as well as live material, including blood particles, bacteria, and viruses.¹ Both human immunodeficiency virus and human papillomavirus have been identified in surgical smoke plumes, and bacterial colony growth has been demonstrated from electrosurgical smoke specimens, specifically Staphylococcus, Corynebacterium, and Neisseria species.^2-8 Treating 1 g of tissue with electrocoagulation produces chemical by-products equivalent to burning 6 unfiltered cigarettes,⁹ which is twice the amount of chemical by-products produced by CO₂ laser vaporization of the same quantity of tissue. It is a common misconception that electrosurgical smoke is less hazardous than smoke produced by ablative CO₂ procedures.⁹ Many chemicals are present in electrosurgical smoke plumes, including nitriles, benzenes, carbon monoxide, hydrogen cyanide, indoles, phenols, pyridine, pyrrole, styrene, toluene, and xylene.^10-12 In animal model studies of rat lungs exposed to surgical smoke, pathologic changes, including interstitial pneumonia, bronchiolitis, and emphysema, have been shown in a dose-dependent manner.^1,13-16 Diseases and symptoms linked to inhalation of electrosurgical smoke in humans include anemia, eye irritation, hypoxia, dizziness, nasopharyngeal lesions, vomiting, sneezing, throat irritation, and weakness.^1,8,17-19 A study of 153 dermatology residents found that more than 70% reported receiving no formal education on the hazards of electrosurgical smoke.²⁰ Approximately 45% were unaware if they had access to smoke evacuation in rooms where electrosurgery was performed. More than 76% were concerned with the infectious risk of electrosurgical smoke, and more than 71% were concerned with its potential carcinogenic risk.²⁰

We surveyed dermatologists who perform skin surgery as well as staff members with respect to their experiences with electrosurgical smoke and to observe any difference that information on the potential hazards of electrosurgical smoke may have on their attitudes and preferences.

Materials and Methods

Survey Instrument
We developed a REDCap survey consisting of 17 questions that was approved by the executive committees of the American College of Mohs Surgery and the American Society for Dermatologic Surgery for distribution to their dermatologist memberships. It was emailed to eligible participants using their mailing lists. Although the survey was sent directly to member physicians, it was recommended that they forward the survey to their clinical staff to complete.

After responding to an initial set of survey questions, respondents were informed that there is growing evidence of potential harms of inhalation of surgical smoke. They then were asked the same series of survey questions in light of this information.

Statistical Analysis
Statistical analysis of the survey responses was then completed, and free-text responses as a final question of the survey were assessed for themes. Preintervention responses of staff and clinicians noticing smoke and being bothered by smoke were assessed using proportions and 95% confidence interval (CI) estimates of the proportions. On most questions, respondents could answer on a scale of 1 to 10. Responses of 5 to 10 on noticing smoke and 5 to 10 on being bothered or troubled by the smoke smell were grouped for analyses. A cross-tabulation using the Bhapkar test for marginal homogeneity was used to assess if information presented on potential smoke hazards changed responses. A Cochran-Mantel-Haenszel test for ordinal responses was used to determine differences between surgeons and staff. A McNemar test was used to determine statistical significance of change in responses to cost. Statistical analysis was performed using SAS version 9.

Results

There was a total of 443 responses to our questionnaire. Two respondents answered that they did not work in an office where skin surgery was performed, and 4 respondents did not answer any questions and were therefore excluded, leaving a total of 437 responses (402 physicians and 35 staff members). A summary of the characteristics of the respondents is shown in the Table. Some respondents did not answer each question, leading to fewer than 437 answers for some questions.

Two hundred eighty-two respondents (64.5%) never or very rarely used smoke evacuation during skin surgical procedures, and only 85 (19.5%) used smoke evacuation with nearly every case. The remaining respondents sometimes used smoke evacuation (Figure 1).

Prior to being presented with the potential dangers of electrosurgical smoke and using a value of 5 to 10 to determine if respondents noticed smoke, 54.4% (95% CI, 49.5%-59.1%) did notice intraoperative smoke during procedures. Using a value of 5 to 10 to indicate if respondents were bothered or troubled by the smoke smell, 35.5% (95% CI, 31.0%-40.2%) were bothered or troubled by intraoperative smoke prior to potential hazards being presented.

Regarding acceptable increase in cost per procedure for smoke evacuation at baseline, 68.9% of respondents favored additional cost; 57.8% of respondents chose the lowest cost grouping of $1 to $30. After being presented with information about the potential harm of intraoperative smoke, the respondents in favor of additional cost increased to 71.5%, which was a small but statistically significant change (P=.0075)(Figure 2).

After being presented with the potential risks of electrosurgical smoke, 29 more respondents answered that they were severely bothered by electrosurgical smoke, whereas 45 fewer respondents selected that they were not bothered or troubled at all by electrosurgical smoke (Figure 3). This difference was statistically significant (P<.0001). Fifteen more respondents answered that they would be much more likely to choose to work at a practice with smoke evacuation once the potential harm of electrosurgical smoke was introduced, and 11 were somewhat more likely to choose a practice with smoke evacuation (P<.0001).

Comment

Developing evidence of health risks associated with electrosurgical smoke plumes has led to an increasing interest in the use of smoke protection or remediation tools during surgical procedures. High-filtration face masks and smoke-evacuation devices protect physicians, staff members, and patients, as well as improve the patient’s clinical experience.

Our study was designed to query dermatologists who perform skin surgery as well as staff members with respect to their experiences with electrosurgical smoke and to observe any difference that information on the potential hazards of electrosurgical smoke may have on their attitudes and preferences. We received 437 responses to our survey (Table). At baseline, 54.4% of respondents noticed and 35.5% were bothered or troubled by the smoke smell produced during skin electrosurgery. These data were intuitively associated in a statistically significant manner with the use of smoke evacuation for respondents; those respondents who more commonly used smoke evacuation were bothered less by electrosurgical smoke, and those respondents who used smoke evacuation less often were more likely to notice and be bothered by surgical smoke.

Once our respondents were presented with the potentially harmful effects of electrosurgical smoke, they became significantly more likely to be bothered by electrosurgical smoke and to want to work in a practice where smoke evacuation was available. This information, however, did not change respondents’ satisfaction with their work environment, and no statistically significant differences were noted between physicians and staff.

At baseline, 68.9% of respondents favored additional cost for smoke evacuation, with approximately 58% favoring the lowest cost category we presented ($1–$30). After being presented with information about the potential dangers of electrosurgical smoke, 71.5% were in favor of increased cost for smoke evacuation, which was a small but statistically significant increase.

The open-comment section of the survey provided interesting insight into the opinions of our respondents on smoke remediation. It is important to note that statistical analysis cannot be performed with these data, and firm generalizable conclusions cannot be drawn from them; however, they reveal topics that may guide further research and policy and certainly merit mention. Of 437 respondents, 108 left free-text comments. Twenty-six percent were categorized as unqualified proponents (in favor of smoke remediation) and 45% as qualified proponents (defined as an individual who verbalized a desire for smoke remediation but also cited a factor limiting their ability to use it, such as cost or staff availability). Only 12% were firmly against smoke remediation, while the remaining 17% did not comment discernibly for or against smoke remediation, indicating that a majority (71% of our comment section respondents) were in favor of some type of smoke remediation, especially if obstacles such as cost could be addressed. Only a small minority was firmly against smoke remediation.

The comments section of our survey highlighted some of the concerns that dermatologic surgeons and their staff have with electrosurgical smoke evacuation. Thirty percent cited cost as an obstacle to use of these devices, and several comments raised concern about increasing overhead and regulatory demands placed on practices. Many indicated that, without sufficient evidence of the harm caused by electrosurgical smoke, regulation that forces use of smoke remediation devices would represent a costly unfunded mandate. Others referenced the logistical challenges of smoke evacuation and the need for staff assistance. Newer smoke-evacuation wands built into cautery pens address much of this concern regarding logistical and staff challenges and further allow the evacuator tip to be located where it is most effective: 1 cm to 2 in from the point of cautery.^21,22

Additionally, 12% of commenters noted that their patients were bothered by the smell of electrosurgical smoke, which is a point that requires further research and is the focus of a current randomized trial at our institution (ClinicalTrials.gov Identifier NCT02958826).

Our current study is limited in that it is a survey and therefore is subject to response bias. Further, some may assert that the hazards of electrosurgical smoke are not settled science, and although we agree with this point on some level, the study aim was not to prove risk but rather to assess current attitudes and see if awareness of a potential risk influenced those attitudes. Additionally, most responses were from physicians—only 35 responses were from nonphysician staff—so it may be difficult to generalize the findings of this study to staff. The large number of physician respondents, however, can be seen as a strength, and the findings are likely much more generalizable to providers who routinely perform clinic-based surgical procedures involving electrosurgery.

Conclusion

Our study shows that most dermatologists who perform skin surgery notice and are bothered by the smoke produced by electrosurgery to at least some extent. When presented with the possibility that inhaling electrosurgical smoke may be harmful, dermatologists were more likely to be bothered by electrosurgical smoke, more likely to prefer a practice environment where smoke evacuation was available, and more likely to be willing to bear additional cost for smoke evacuation. The free-text comments on our survey highlighted that many dermatologic surgeons are proponents of smoke evacuation but have concerns about cost and potential regulatory challenges associated with smoke evacuation, especially if the potential risks are not settled science. Many logistical concerns for smoke evacuation are addressed with the use of integrated devices. More research is needed to determine the health effects of the surgical smoke we are exposed to daily and the optimal way to limit any risk.

Acknowledgment
The authors would like to thank Richard W. Madsen, PhD (Columbia, Missouri), biostatistician, for his valuable guidance in the statistical analysis of data, interpretation of results, and editorial support in finalizing the manuscript.

Approximately 17% to 25% of all softtissue sarcomas (STS) are liposarcomas, making liposarcoma the most common type of STS.¹ The 2013 World Health Organization (WHO) classification separates liposarcoma into 4 histologic subtypes: atypical lipomatous tumor/well-differentiated (ALT/ WDLPS), dedifferentiated (DDLPS), myxoid, and pleomorphic.² Each subtype has unique histology, morphology, and natural history. WDLPS and DDLPS are the most common histologic subtypes, comprising approximately 50% of all sarcomas that arise in the retroperitoneum.³ DDLPS represents 18% of all liposarcomas, making it the second most common subtype of liposarcoma.⁴

In 1979, DDLPS was first characterized.⁵ Most (90%) cases of DDLPS present de novo, whereas the other 10% transform from preexisting low-grade WDLPS.² DDLPSs are formed by an amplification of 12q14-15 involving the MDM2 gene.⁴ These malignancies most commonly present in the retroperitoneum as a large painless mass, consisting of both fatty and nonfatty components.² Primary site has been previously reported as a major prognostic factor for DDLPSs, with retroperitoneal DDLPSs demonstrating the worst prognosis.6 DDLPSs have a high risk of local recurrence, with some reports estimating recurrence rates approaching 40%.² Overall mortality at 5 years for DDLPS is estimated to be between 30% and 40%.⁴

Previous literature has determined that median income, race, health insurance, and facility type are related to survival outcomes for patients with DDLPS.7-9 When comparing the most common types of cancers, residents of poorer US counties consistently had a higher risk of mortality than residents in affluent US counties, and all racial minorities showed worse survival outcomes when compared with white patients.⁷ Differences in survival outcomes have been reported in patients attending different treatment facilities for other cancers including pancreatic cancers, glioblastomas, and oral cancers, with multiple studies concluding that academic and research programs are associated with the longest survival outcomes.^10-12 For many cancers, insurance status has been shown to be a significant prognostic factor, with private insurance typically resulting in the best prognosis.^8,9

The goal of this retrospective study was to assess the prognostic effects of socioeconomic variables on the overall survival (OS) probabilities in a large cohort of DDLPS patients in order to inform clinicians about a potentially at-risk population.

Method

The National Cancer Database (NCDB) was created by the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The NCDB is the largest cancer database in the US and includes data on almost 70% of US patients with cancer. CoC-accredited cancer programs add data on patients with cancer to the NCDB. The authors accessed the NCDB data through the use of the NCDB Participant Use File program.

Patients’ data from 2004 through 2015 were abstracted. Only patients with the International Classification of Diseases for Oncology histology code 8858, corresponding to DDLPS, were analyzed. Patients with other comorbid malignant tumors were excluded to accurately capture the true survival rates for DDLPS. Variables analyzed included age, sex, race, insurance status, treatment facility type, median household income by zip code, and percentage of adults in the patient’s zip code with no high school (HS) education.

Median survival, 5- and 10-year OS probabilities, and Kaplan-Meier survival curves were calculated for multiple variables, specifically race, insurance status, treatment facility type, median family income, and percentage of adults without a HS degree. Both 5- and 10-year OS probabilities were determined by race with the patients separated into white, African American, Asian, American Indian/Alaska Native (AI/AN), and Asian Indian or Pakistani groups. Our study categorized Chinese, Japanese, Filipino, Hmong, Korean, Vietnamese, Thai, Guamanian, Asian not otherwise specified, and other Asian ethnicity patients together into one collective Asian group. Insurance status was classified into Medicare, Medicaid, other government insurance, and private insurance groups. Other government insurance consisted of US Department of Veterans Affairs, Indian Health Service, Public Health Service, and other government health care programs. Further analysis could not be performed into the distribution of the other government insurance variable.

Facility types were divided into 4 groups: community, comprehensive community, academic/ research, and integrated network cancer treatment facilities. Median income quartiles and the percentage of adults with no high school degree were estimated by comparison of the patient’s zip code with US Census Bureau data. Median household income was separated into 4 groups, including lowest level of household income (< $38,000), low level of household income ($38,000 to $47,999), moderate level of household income ($48,000 to $62,999), and highest level of household income (≥ $63,000). The percentages of adults with no high school degree were divided into 4 groups: lowest level of HS education (≥ 21% ), low level of HS education (13.0% to 20.9%), moderate level of HS education (7.0% to 12.9%), and highest level of HS education (≤ 7%). The 5- and 10-year survival probabilities were calculated using the number of months between the date of diagnosis and the date of death or last known contact.

Continuous variables are presented as median and interquartile range (IQR) whereas categorical variables are presented as frequencies and proportion. IBM SPSS version 25.0 was used to produce Kaplan-Meier survival curves and descriptive statistics. This study used Kaplan- Meier survival tables and log-rank tests to analyze both the 5- and 10-year OS rates for the 5 variables listed above. This study also used a multivariable Cox regression model that accommodated the correlative nature of outcomes within facilities to study the association of the treatment facility type and other socioeconomic factors, while controlling for age, race (which was collapsed into 3 categories), sex, primary site, tumor stage, and treatment approaches. The proportional hazards assumption was individually checked for all pertinent variables. Any patient records that were missing data were excluded from the multivariable Cox regression model, which was analyzed with SAS version 9.4 (Cary, NC). P < 0.05 was used to indicate statistical significance for all analyses.

Results

Table 1 provides descriptive analysis for demographic characteristics of the 3573 patients including age, sex, and race. The median age at diagnosis was 64 years. There were 1073 more men (65%) than women (35%) in this analysis. Whites were the predominant racial category, comprising 87.7% of the patient population, followed by African Americans (6.5%) and Asians (2.5%).

Socioeconomic Variables

The largest proportion of the patient population (45.5%) had private insurance (Table 2). Medicare came in a close second covering almost 42.2% of the population, followed by Medicaid (5.0%), uninsured (2.8%), and other government insurance (1.5%). About half (53.7%) of the patients were treated at academic or research facilities, while the fewest number of patients (5.2%) underwent treatment at community cancer facilities. The largest percentage (36.6%) of patients lived in zip codes with the highest level of median household income, while 26.0% and 22.3% had moderate and low levels of income, respectively. About 14% of patients lived within an area of the lowest level of income. Similarly, almost 15% of patients lived in an area of lowest level of HS education. The greatest percentage of the patient population (34.5%) lived in a zip code with moderate level of HS education. Surgery was the most common treatment modality with 90.8% of the cohort undergoing surgery, while 35.4% and 16.5% were treated with radiation and chemotherapy, respectively (some patients received more than one type of treatment modality).

Survival Data

Survival data were available for 3112 patients. Kaplan-Meier survival curves were used to analyze OS according to insurance status, racial background, treatment facility type, median family income, and percentage of adults with no high school education. Overall 5- and 10- year OS probabilities were 51.5% and 34.8%, respectively, while the median OS (SD) was 63.57 (2.8) months (Table 3).

Private insurance showed significantly higher 5- and 10-year OS probabilities and median OS: 5-year OS was 61.2%, 10-year OS was 47.2%, and median survival (SD) was 101.2 (8.2) months compared with that of all other insurance groups (Medicare, Medicaid, other government insurance, and uninsured) (Figure 1). These other insurance types were fairly similar in their 5-year and median OS, but surprisingly, patients with no insurance had the second longest 10-year OS. The difference between the 5-year OS probabilities of private insurance compared with an average of the other insurances was 15.1%, which had almost doubled to 28.5% at 10 years, with a median OS difference of almost 5 years (56 months; data not shown).

Using the Kaplan-Meier survival curve, Asian Indians had the longest 5-year OS probability of 77.9% and African Americans had the longest 10-year OS probability of 40.6%. However, Asians as a group demonstrated the longest median (SD) OS outcome with 119.8 (47.8) months (Figure 2).

Overall, academic/research programs had the longest median OS and 5-year OS probability (SD) of 66.6 (4.5) months and 52.6%, respectively (Figure 3). Comprehensive community cancer programs and integrated network cancer programs had nearly identical 10-year OS rates (35.2% vs 35.1%, respectively). Community cancer programs had the worst 5- and 10-year OS probabilities (41.1% and 21.8%, respectively).

The top 2 income quartiles combined to demonstrate the longest median, 5-year, and 10-year OS probabilities and were very similar. Patients living in a zip code with the highest income level had the longest 5-year OS rates of 54.3%, while patients living in zip codes with a moderate income level had the longest 10-year OS at 39.3% and the longest median OS of about 71 months. Patients with the lowest level of median household income had the worst 5-year OS rates (48.3%) and a median (SD) OS of 53.4 (5.4) months (Figure 4).

A Kaplan-Meier curve for percentage of adults without a HS degree is displayed in Figure 5. Zip codes with the highest level of education had the longest 5-year OS rates and median (SD) OS of 55.3% and 70.9 (4.8) months, respectively. The longest 10-year OS outcomes at 38.1% were found in patients who lived in areas of low-education levels. The worst 5- and 10- year OS outcomes and median OS were found in the least educated zip codes.

Results from the Cox regression model of OS are displayed in Table 4. Race and ethnicity, zip code-level median household income, and zip code-level education were not associated with OS. Patients with no insurance had an increased risk of death (hazard ratio [HR], 1.84; 95% CI, 1.17-2.88; P < .01) when compared with patients with private insurance. Patients with other government insurance also had an increased risk of death (HR, 2.12; 95% CI, 1.27-3.54; P < .01) when compared with patients with private insurance while controlling for all other variables. Patients with Medicare had a decreased risk of death when compared with patients with other government insurance and no insurance (HR, 0.53; 95% CI, 0.31-0.92; P = .02 and HR, 0.62; 95% CI, 0.38-0.99; P = .05, respectively). Patients treated at academic centers had better OS when compared with patients treated at comprehensive treatment centers (HR, 0.77; 95% CI, 0.65-0.92;P < .01) and community treatment centers (HR, 0.62; 95% CI, 0.44-0.86; P < .01).

Discussion

This study is the largest study to date that specifically studies the type of treatment facilities and socioeconomic factors, including insurance status, race, income, and education, and how they affect survival of DDLPS. The overall 5- and 10-year OS probabilities for DDLPS in this study were 51.5% and 34.8%, respectively, with median OS of 63.6 months. These results were more encouraging than previous reports, which found a 5-year survival probability of 36.5% and a median OS of 45 months.^13,14

The largest age grouping was aged 61 to 80 years (48.9% of the cohort), and the median age at diagnosis was 64 years. DDLPSs most typically present between the ages of 50 and 70 years.¹⁵ Our cohort was 65% male. Previous studies have indicated that DDLPSs affect the sexes equally; however, another study showed a similar male predominance (68.8%) at the MD Anderson Cancer Center in Houston, Texas.^13,16

In our study, approximately 88% of patients were white, 6.5% were African American, and 2.5% were Asian, which differed from a previous study of 84 patients that had a 78.6% white, 4.8% Asian, and 1.2% African American patient population.¹⁴

Asian Indian or Pakistani patients had the best 5-year OS probability at 77.9%, followed by African American (57.2%), Asian (51.6%), AI/AN (51.4%), and white patients (50.9%). This trend had disappeared by 10 years and Asian, AI/AN, African American, and Asian Indian or Pakistani groups all demonstrated longer median OS than did white patients. In fact, Asian patients had the longest median OS at 119.8 months, which was almost double that of white patients with the lowest median OS of 61.2 months. This finding is contrary to previous studies, which reported that racial minorities typically had worse OS outcomes when compared with white patients in different types of cancer.^7,17 Notably, these findings were not statistically significant in our current study in the log-rank or multivariable analyses.

Private insurance was the most common form of insurance followed in decreasing order by Medicare, Medicaid, uninsured, and other government insurance. About 42% of the cohort had Medicare, which is a federally funded US insurance program designated for patients aged ≥ 65 years and certain younger patients with disabilities.

Patients with private insurance demonstrated the longest OS, essentially twice the median OS of all other insured groups at 101 months. Medicare had the worst 5-year OS probability and median OS of all groups. A previous study of 77 patients with DDLPS reported that patients aged > 65 years had reduced OS.13 Medicare patients in this study were older, with a mean and median age at DDLPS diagnosis of 71 and 72 years, respectively, while private insurance had a mean and median age at diagnosis of 56 and 57 years, respectively. Medicare inherently covers older patients and this age difference could account for the decrease in overall survival.

Improved OS for privately insured patients was most notable compared with the uninsured or patients with other government insurance. Uninsured patients had an 83.7% increased risk of mortality when compared with patients with private insurance. When compared with patients with private insurance, patients with other government insurance had an 111.5% increased risk of mortality. Comparing patients with Medicare vs patients with no insurance or other government insurance, there was a decreased risk of mortality of 38.5% and 46.6%, respectively. This decreased OS in patients with other government insurance could be related to the choice of treatment facility, because only 31% of the patients with other government insurance went to academic or research centers when compared with the 58.4% and 50.8% of patients with private and Medicare insurance treated there (data not shown). Such centers often have access to more advanced technology and protocols that may not be available at other treatment facilities.

A little more than half of the patients in the cohort went to an academic or research center for treatment (53.7%); comprehensive community cancer programs were the second most common treatment facility at 28%. Patients treated at academic or research centers demonstrated the best outcomes with a 5-year OS of 52.6%, followed in decreasing order by comprehensive community cancer programs (49.7%), integrated network cancer programs (48.8%), and community cancer programs (41.1%). In our patient cocohort, patients treated at an academic/research center had slightly decreased 10-year OS rates compared with those patients treated at a comprehensive community cancer program, although the median OS for the academic/research centers were still the highest of all treatment facilities.

Treatment options varied significantly by facility, and the number of patients treated surgically followed a similar trend, with 92% undergoing surgery as the primary treatment at academic or research programs compared with 89% at comprehensive cancer programs and 82.7% at community cancer programs (data not shown). Another potential explaination for differing OS outcomes across facilities is the surgical margin outcome. Surgeries performed at community cancer programs or comprehensive cancer programs resulted with no residual tumor in 36% and 40% of cases, respectively, whereas cases performed at academic or research programs resulted with no residual tumor in 47% of cases (data not shown). Regardless, multivariate analysis demonstrated a marked decrease in the chance of mortality when comparing treatment received at academic facility centers with that received at comprehensive cancer centers (22.9%) and community cancer centers (38.3%) (data not shown).

A recent study demonstrated improved outcomes for patients with retroperitoneal or extremity STS treated at high-volume treatment centers.¹⁸ Patients treated at high-volume centers were found to have an 8% decreased risk of death compared with patients treated at low-volume centers. Notably, they found highvolume academic centers demonstrated the strongest improvement in survival, while highvolume community centers showed decreased survival.¹⁸ Similarly, we found that patients treated at academic/research institutions had improved 5-year OS and greater median OS than did patients treated at community cancer programs or comprehensive community cancer programs.

The top 2 income quartiles (≥ $48,000) combined to demonstrate the longest median, 5-year, and 10-year OS and were fairly similar between the quartiles. Patients living in zip codes with a median income of $38,000 to $47,999 had the worst 5-year OS and median OS. The log-rank analysis showed statistical evidence of differences in survival associated with income, but within the context of the multivariable analysis, there was no remaining evidence of a difference.

The longest 5-year OS outcomes were seen in patients living in zip codes with the highest level of education (55.3%). However, the difference in OS was not statistically significant using either the log-rank analysis or multivariate analysis.

Limitations

This study has certain inherent limitations in using a retrospective design and a large database such as the NCDB. Many different pathologists at CoC-accredited cancer programs perform the pathology that contributes to the data in the NCDB. There was no pathological review of these findings, which could potentially introduce error into the findings of this study. With the NCDB, potential selection bias is possible because patients in the database are added only from CoC-accredited cancer programs. This risk is minimized because NCDB contains data on most newly diagnosed cancer patients in the US. Further potential risks, which are unable to be controlled for, include potential interobserver error and data that may be incompletely, improperly, or inaccurately recorded from the patients’ charts. Without patient-specific information regarding income and education, it is challenging to utilize zip codes to estimate socioeconomic status and educational level. Even though a patient may live in a zip code identified with specific economic and educational characteristics, that patient may not share those characteristics. Furthermore, patients with Medicare tend to be older than patients with other forms of insurance, which limits the significance of comparisons across insurance groups. A future SEER (Surveillance, Epidemiology, and End Results) program study to confirm this study’s results and the effects of socioeconomic variables on DDLPS would be an excellent followup study.

Conclusion

This study used a large cohort of patients with DDLPS to study the effects of treatment facility, insurance status, and socioeconomic variables on survival outcomes. Although insurance status, median household income, and treatment facility were associated with differences in median OS and 5- and 10-year OS probabilities, evidence for a difference remained for only insurance status and facility type within the context of a multivariable analysis irrespective of age, race, sex, insurance status, education, and median income. Patients with private insurance and Medicaid had a decreased risk of mortality compared with other government insurance and no insurance. Patients receiving treatment at academic research programs had the highest median and 5-year OS of 66.6 months and 52.6%, respectively. Patients receiving treatment at academic centers had improved survival outcomes with a decrease in mortality of 23% and 38% compared to comprehensive or community cancer programs.

Approximately 17% to 25% of all softtissue sarcomas (STS) are liposarcomas, making liposarcoma the most common type of STS.¹ The 2013 World Health Organization (WHO) classification separates liposarcoma into 4 histologic subtypes: atypical lipomatous tumor/well-differentiated (ALT/ WDLPS), dedifferentiated (DDLPS), myxoid, and pleomorphic.² Each subtype has unique histology, morphology, and natural history. WDLPS and DDLPS are the most common histologic subtypes, comprising approximately 50% of all sarcomas that arise in the retroperitoneum.³ DDLPS represents 18% of all liposarcomas, making it the second most common subtype of liposarcoma.⁴

In 1979, DDLPS was first characterized.⁵ Most (90%) cases of DDLPS present de novo, whereas the other 10% transform from preexisting low-grade WDLPS.² DDLPSs are formed by an amplification of 12q14-15 involving the MDM2 gene.⁴ These malignancies most commonly present in the retroperitoneum as a large painless mass, consisting of both fatty and nonfatty components.² Primary site has been previously reported as a major prognostic factor for DDLPSs, with retroperitoneal DDLPSs demonstrating the worst prognosis.6 DDLPSs have a high risk of local recurrence, with some reports estimating recurrence rates approaching 40%.² Overall mortality at 5 years for DDLPS is estimated to be between 30% and 40%.⁴

Previous literature has determined that median income, race, health insurance, and facility type are related to survival outcomes for patients with DDLPS.7-9 When comparing the most common types of cancers, residents of poorer US counties consistently had a higher risk of mortality than residents in affluent US counties, and all racial minorities showed worse survival outcomes when compared with white patients.⁷ Differences in survival outcomes have been reported in patients attending different treatment facilities for other cancers including pancreatic cancers, glioblastomas, and oral cancers, with multiple studies concluding that academic and research programs are associated with the longest survival outcomes.^10-12 For many cancers, insurance status has been shown to be a significant prognostic factor, with private insurance typically resulting in the best prognosis.^8,9

The goal of this retrospective study was to assess the prognostic effects of socioeconomic variables on the overall survival (OS) probabilities in a large cohort of DDLPS patients in order to inform clinicians about a potentially at-risk population.

Method

The National Cancer Database (NCDB) was created by the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The NCDB is the largest cancer database in the US and includes data on almost 70% of US patients with cancer. CoC-accredited cancer programs add data on patients with cancer to the NCDB. The authors accessed the NCDB data through the use of the NCDB Participant Use File program.

Patients’ data from 2004 through 2015 were abstracted. Only patients with the International Classification of Diseases for Oncology histology code 8858, corresponding to DDLPS, were analyzed. Patients with other comorbid malignant tumors were excluded to accurately capture the true survival rates for DDLPS. Variables analyzed included age, sex, race, insurance status, treatment facility type, median household income by zip code, and percentage of adults in the patient’s zip code with no high school (HS) education.

Median survival, 5- and 10-year OS probabilities, and Kaplan-Meier survival curves were calculated for multiple variables, specifically race, insurance status, treatment facility type, median family income, and percentage of adults without a HS degree. Both 5- and 10-year OS probabilities were determined by race with the patients separated into white, African American, Asian, American Indian/Alaska Native (AI/AN), and Asian Indian or Pakistani groups. Our study categorized Chinese, Japanese, Filipino, Hmong, Korean, Vietnamese, Thai, Guamanian, Asian not otherwise specified, and other Asian ethnicity patients together into one collective Asian group. Insurance status was classified into Medicare, Medicaid, other government insurance, and private insurance groups. Other government insurance consisted of US Department of Veterans Affairs, Indian Health Service, Public Health Service, and other government health care programs. Further analysis could not be performed into the distribution of the other government insurance variable.

Facility types were divided into 4 groups: community, comprehensive community, academic/ research, and integrated network cancer treatment facilities. Median income quartiles and the percentage of adults with no high school degree were estimated by comparison of the patient’s zip code with US Census Bureau data. Median household income was separated into 4 groups, including lowest level of household income (< $38,000), low level of household income ($38,000 to $47,999), moderate level of household income ($48,000 to $62,999), and highest level of household income (≥ $63,000). The percentages of adults with no high school degree were divided into 4 groups: lowest level of HS education (≥ 21% ), low level of HS education (13.0% to 20.9%), moderate level of HS education (7.0% to 12.9%), and highest level of HS education (≤ 7%). The 5- and 10-year survival probabilities were calculated using the number of months between the date of diagnosis and the date of death or last known contact.

Continuous variables are presented as median and interquartile range (IQR) whereas categorical variables are presented as frequencies and proportion. IBM SPSS version 25.0 was used to produce Kaplan-Meier survival curves and descriptive statistics. This study used Kaplan- Meier survival tables and log-rank tests to analyze both the 5- and 10-year OS rates for the 5 variables listed above. This study also used a multivariable Cox regression model that accommodated the correlative nature of outcomes within facilities to study the association of the treatment facility type and other socioeconomic factors, while controlling for age, race (which was collapsed into 3 categories), sex, primary site, tumor stage, and treatment approaches. The proportional hazards assumption was individually checked for all pertinent variables. Any patient records that were missing data were excluded from the multivariable Cox regression model, which was analyzed with SAS version 9.4 (Cary, NC). P < 0.05 was used to indicate statistical significance for all analyses.

Results

Table 1 provides descriptive analysis for demographic characteristics of the 3573 patients including age, sex, and race. The median age at diagnosis was 64 years. There were 1073 more men (65%) than women (35%) in this analysis. Whites were the predominant racial category, comprising 87.7% of the patient population, followed by African Americans (6.5%) and Asians (2.5%).

Socioeconomic Variables

The largest proportion of the patient population (45.5%) had private insurance (Table 2). Medicare came in a close second covering almost 42.2% of the population, followed by Medicaid (5.0%), uninsured (2.8%), and other government insurance (1.5%). About half (53.7%) of the patients were treated at academic or research facilities, while the fewest number of patients (5.2%) underwent treatment at community cancer facilities. The largest percentage (36.6%) of patients lived in zip codes with the highest level of median household income, while 26.0% and 22.3% had moderate and low levels of income, respectively. About 14% of patients lived within an area of the lowest level of income. Similarly, almost 15% of patients lived in an area of lowest level of HS education. The greatest percentage of the patient population (34.5%) lived in a zip code with moderate level of HS education. Surgery was the most common treatment modality with 90.8% of the cohort undergoing surgery, while 35.4% and 16.5% were treated with radiation and chemotherapy, respectively (some patients received more than one type of treatment modality).

Survival Data

Survival data were available for 3112 patients. Kaplan-Meier survival curves were used to analyze OS according to insurance status, racial background, treatment facility type, median family income, and percentage of adults with no high school education. Overall 5- and 10- year OS probabilities were 51.5% and 34.8%, respectively, while the median OS (SD) was 63.57 (2.8) months (Table 3).

Private insurance showed significantly higher 5- and 10-year OS probabilities and median OS: 5-year OS was 61.2%, 10-year OS was 47.2%, and median survival (SD) was 101.2 (8.2) months compared with that of all other insurance groups (Medicare, Medicaid, other government insurance, and uninsured) (Figure 1). These other insurance types were fairly similar in their 5-year and median OS, but surprisingly, patients with no insurance had the second longest 10-year OS. The difference between the 5-year OS probabilities of private insurance compared with an average of the other insurances was 15.1%, which had almost doubled to 28.5% at 10 years, with a median OS difference of almost 5 years (56 months; data not shown).

Using the Kaplan-Meier survival curve, Asian Indians had the longest 5-year OS probability of 77.9% and African Americans had the longest 10-year OS probability of 40.6%. However, Asians as a group demonstrated the longest median (SD) OS outcome with 119.8 (47.8) months (Figure 2).

Overall, academic/research programs had the longest median OS and 5-year OS probability (SD) of 66.6 (4.5) months and 52.6%, respectively (Figure 3). Comprehensive community cancer programs and integrated network cancer programs had nearly identical 10-year OS rates (35.2% vs 35.1%, respectively). Community cancer programs had the worst 5- and 10-year OS probabilities (41.1% and 21.8%, respectively).

The top 2 income quartiles combined to demonstrate the longest median, 5-year, and 10-year OS probabilities and were very similar. Patients living in a zip code with the highest income level had the longest 5-year OS rates of 54.3%, while patients living in zip codes with a moderate income level had the longest 10-year OS at 39.3% and the longest median OS of about 71 months. Patients with the lowest level of median household income had the worst 5-year OS rates (48.3%) and a median (SD) OS of 53.4 (5.4) months (Figure 4).

A Kaplan-Meier curve for percentage of adults without a HS degree is displayed in Figure 5. Zip codes with the highest level of education had the longest 5-year OS rates and median (SD) OS of 55.3% and 70.9 (4.8) months, respectively. The longest 10-year OS outcomes at 38.1% were found in patients who lived in areas of low-education levels. The worst 5- and 10- year OS outcomes and median OS were found in the least educated zip codes.

Results from the Cox regression model of OS are displayed in Table 4. Race and ethnicity, zip code-level median household income, and zip code-level education were not associated with OS. Patients with no insurance had an increased risk of death (hazard ratio [HR], 1.84; 95% CI, 1.17-2.88; P < .01) when compared with patients with private insurance. Patients with other government insurance also had an increased risk of death (HR, 2.12; 95% CI, 1.27-3.54; P < .01) when compared with patients with private insurance while controlling for all other variables. Patients with Medicare had a decreased risk of death when compared with patients with other government insurance and no insurance (HR, 0.53; 95% CI, 0.31-0.92; P = .02 and HR, 0.62; 95% CI, 0.38-0.99; P = .05, respectively). Patients treated at academic centers had better OS when compared with patients treated at comprehensive treatment centers (HR, 0.77; 95% CI, 0.65-0.92;P < .01) and community treatment centers (HR, 0.62; 95% CI, 0.44-0.86; P < .01).

Discussion

This study is the largest study to date that specifically studies the type of treatment facilities and socioeconomic factors, including insurance status, race, income, and education, and how they affect survival of DDLPS. The overall 5- and 10-year OS probabilities for DDLPS in this study were 51.5% and 34.8%, respectively, with median OS of 63.6 months. These results were more encouraging than previous reports, which found a 5-year survival probability of 36.5% and a median OS of 45 months.^13,14

The largest age grouping was aged 61 to 80 years (48.9% of the cohort), and the median age at diagnosis was 64 years. DDLPSs most typically present between the ages of 50 and 70 years.¹⁵ Our cohort was 65% male. Previous studies have indicated that DDLPSs affect the sexes equally; however, another study showed a similar male predominance (68.8%) at the MD Anderson Cancer Center in Houston, Texas.^13,16

In our study, approximately 88% of patients were white, 6.5% were African American, and 2.5% were Asian, which differed from a previous study of 84 patients that had a 78.6% white, 4.8% Asian, and 1.2% African American patient population.¹⁴

Asian Indian or Pakistani patients had the best 5-year OS probability at 77.9%, followed by African American (57.2%), Asian (51.6%), AI/AN (51.4%), and white patients (50.9%). This trend had disappeared by 10 years and Asian, AI/AN, African American, and Asian Indian or Pakistani groups all demonstrated longer median OS than did white patients. In fact, Asian patients had the longest median OS at 119.8 months, which was almost double that of white patients with the lowest median OS of 61.2 months. This finding is contrary to previous studies, which reported that racial minorities typically had worse OS outcomes when compared with white patients in different types of cancer.^7,17 Notably, these findings were not statistically significant in our current study in the log-rank or multivariable analyses.

Private insurance was the most common form of insurance followed in decreasing order by Medicare, Medicaid, uninsured, and other government insurance. About 42% of the cohort had Medicare, which is a federally funded US insurance program designated for patients aged ≥ 65 years and certain younger patients with disabilities.

Patients with private insurance demonstrated the longest OS, essentially twice the median OS of all other insured groups at 101 months. Medicare had the worst 5-year OS probability and median OS of all groups. A previous study of 77 patients with DDLPS reported that patients aged > 65 years had reduced OS.13 Medicare patients in this study were older, with a mean and median age at DDLPS diagnosis of 71 and 72 years, respectively, while private insurance had a mean and median age at diagnosis of 56 and 57 years, respectively. Medicare inherently covers older patients and this age difference could account for the decrease in overall survival.

Improved OS for privately insured patients was most notable compared with the uninsured or patients with other government insurance. Uninsured patients had an 83.7% increased risk of mortality when compared with patients with private insurance. When compared with patients with private insurance, patients with other government insurance had an 111.5% increased risk of mortality. Comparing patients with Medicare vs patients with no insurance or other government insurance, there was a decreased risk of mortality of 38.5% and 46.6%, respectively. This decreased OS in patients with other government insurance could be related to the choice of treatment facility, because only 31% of the patients with other government insurance went to academic or research centers when compared with the 58.4% and 50.8% of patients with private and Medicare insurance treated there (data not shown). Such centers often have access to more advanced technology and protocols that may not be available at other treatment facilities.

A little more than half of the patients in the cohort went to an academic or research center for treatment (53.7%); comprehensive community cancer programs were the second most common treatment facility at 28%. Patients treated at academic or research centers demonstrated the best outcomes with a 5-year OS of 52.6%, followed in decreasing order by comprehensive community cancer programs (49.7%), integrated network cancer programs (48.8%), and community cancer programs (41.1%). In our patient cocohort, patients treated at an academic/research center had slightly decreased 10-year OS rates compared with those patients treated at a comprehensive community cancer program, although the median OS for the academic/research centers were still the highest of all treatment facilities.

Treatment options varied significantly by facility, and the number of patients treated surgically followed a similar trend, with 92% undergoing surgery as the primary treatment at academic or research programs compared with 89% at comprehensive cancer programs and 82.7% at community cancer programs (data not shown). Another potential explaination for differing OS outcomes across facilities is the surgical margin outcome. Surgeries performed at community cancer programs or comprehensive cancer programs resulted with no residual tumor in 36% and 40% of cases, respectively, whereas cases performed at academic or research programs resulted with no residual tumor in 47% of cases (data not shown). Regardless, multivariate analysis demonstrated a marked decrease in the chance of mortality when comparing treatment received at academic facility centers with that received at comprehensive cancer centers (22.9%) and community cancer centers (38.3%) (data not shown).

A recent study demonstrated improved outcomes for patients with retroperitoneal or extremity STS treated at high-volume treatment centers.¹⁸ Patients treated at high-volume centers were found to have an 8% decreased risk of death compared with patients treated at low-volume centers. Notably, they found highvolume academic centers demonstrated the strongest improvement in survival, while highvolume community centers showed decreased survival.¹⁸ Similarly, we found that patients treated at academic/research institutions had improved 5-year OS and greater median OS than did patients treated at community cancer programs or comprehensive community cancer programs.

The top 2 income quartiles (≥ $48,000) combined to demonstrate the longest median, 5-year, and 10-year OS and were fairly similar between the quartiles. Patients living in zip codes with a median income of $38,000 to $47,999 had the worst 5-year OS and median OS. The log-rank analysis showed statistical evidence of differences in survival associated with income, but within the context of the multivariable analysis, there was no remaining evidence of a difference.

The longest 5-year OS outcomes were seen in patients living in zip codes with the highest level of education (55.3%). However, the difference in OS was not statistically significant using either the log-rank analysis or multivariate analysis.

Limitations

This study has certain inherent limitations in using a retrospective design and a large database such as the NCDB. Many different pathologists at CoC-accredited cancer programs perform the pathology that contributes to the data in the NCDB. There was no pathological review of these findings, which could potentially introduce error into the findings of this study. With the NCDB, potential selection bias is possible because patients in the database are added only from CoC-accredited cancer programs. This risk is minimized because NCDB contains data on most newly diagnosed cancer patients in the US. Further potential risks, which are unable to be controlled for, include potential interobserver error and data that may be incompletely, improperly, or inaccurately recorded from the patients’ charts. Without patient-specific information regarding income and education, it is challenging to utilize zip codes to estimate socioeconomic status and educational level. Even though a patient may live in a zip code identified with specific economic and educational characteristics, that patient may not share those characteristics. Furthermore, patients with Medicare tend to be older than patients with other forms of insurance, which limits the significance of comparisons across insurance groups. A future SEER (Surveillance, Epidemiology, and End Results) program study to confirm this study’s results and the effects of socioeconomic variables on DDLPS would be an excellent followup study.

Conclusion

This study used a large cohort of patients with DDLPS to study the effects of treatment facility, insurance status, and socioeconomic variables on survival outcomes. Although insurance status, median household income, and treatment facility were associated with differences in median OS and 5- and 10-year OS probabilities, evidence for a difference remained for only insurance status and facility type within the context of a multivariable analysis irrespective of age, race, sex, insurance status, education, and median income. Patients with private insurance and Medicaid had a decreased risk of mortality compared with other government insurance and no insurance. Patients receiving treatment at academic research programs had the highest median and 5-year OS of 66.6 months and 52.6%, respectively. Patients receiving treatment at academic centers had improved survival outcomes with a decrease in mortality of 23% and 38% compared to comprehensive or community cancer programs.

Approximately 17% to 25% of all softtissue sarcomas (STS) are liposarcomas, making liposarcoma the most common type of STS.¹ The 2013 World Health Organization (WHO) classification separates liposarcoma into 4 histologic subtypes: atypical lipomatous tumor/well-differentiated (ALT/ WDLPS), dedifferentiated (DDLPS), myxoid, and pleomorphic.² Each subtype has unique histology, morphology, and natural history. WDLPS and DDLPS are the most common histologic subtypes, comprising approximately 50% of all sarcomas that arise in the retroperitoneum.³ DDLPS represents 18% of all liposarcomas, making it the second most common subtype of liposarcoma.⁴

In 1979, DDLPS was first characterized.⁵ Most (90%) cases of DDLPS present de novo, whereas the other 10% transform from preexisting low-grade WDLPS.² DDLPSs are formed by an amplification of 12q14-15 involving the MDM2 gene.⁴ These malignancies most commonly present in the retroperitoneum as a large painless mass, consisting of both fatty and nonfatty components.² Primary site has been previously reported as a major prognostic factor for DDLPSs, with retroperitoneal DDLPSs demonstrating the worst prognosis.6 DDLPSs have a high risk of local recurrence, with some reports estimating recurrence rates approaching 40%.² Overall mortality at 5 years for DDLPS is estimated to be between 30% and 40%.⁴

Previous literature has determined that median income, race, health insurance, and facility type are related to survival outcomes for patients with DDLPS.7-9 When comparing the most common types of cancers, residents of poorer US counties consistently had a higher risk of mortality than residents in affluent US counties, and all racial minorities showed worse survival outcomes when compared with white patients.⁷ Differences in survival outcomes have been reported in patients attending different treatment facilities for other cancers including pancreatic cancers, glioblastomas, and oral cancers, with multiple studies concluding that academic and research programs are associated with the longest survival outcomes.^10-12 For many cancers, insurance status has been shown to be a significant prognostic factor, with private insurance typically resulting in the best prognosis.^8,9

The goal of this retrospective study was to assess the prognostic effects of socioeconomic variables on the overall survival (OS) probabilities in a large cohort of DDLPS patients in order to inform clinicians about a potentially at-risk population.

Method

The National Cancer Database (NCDB) was created by the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The NCDB is the largest cancer database in the US and includes data on almost 70% of US patients with cancer. CoC-accredited cancer programs add data on patients with cancer to the NCDB. The authors accessed the NCDB data through the use of the NCDB Participant Use File program.

Patients’ data from 2004 through 2015 were abstracted. Only patients with the International Classification of Diseases for Oncology histology code 8858, corresponding to DDLPS, were analyzed. Patients with other comorbid malignant tumors were excluded to accurately capture the true survival rates for DDLPS. Variables analyzed included age, sex, race, insurance status, treatment facility type, median household income by zip code, and percentage of adults in the patient’s zip code with no high school (HS) education.

Median survival, 5- and 10-year OS probabilities, and Kaplan-Meier survival curves were calculated for multiple variables, specifically race, insurance status, treatment facility type, median family income, and percentage of adults without a HS degree. Both 5- and 10-year OS probabilities were determined by race with the patients separated into white, African American, Asian, American Indian/Alaska Native (AI/AN), and Asian Indian or Pakistani groups. Our study categorized Chinese, Japanese, Filipino, Hmong, Korean, Vietnamese, Thai, Guamanian, Asian not otherwise specified, and other Asian ethnicity patients together into one collective Asian group. Insurance status was classified into Medicare, Medicaid, other government insurance, and private insurance groups. Other government insurance consisted of US Department of Veterans Affairs, Indian Health Service, Public Health Service, and other government health care programs. Further analysis could not be performed into the distribution of the other government insurance variable.

Facility types were divided into 4 groups: community, comprehensive community, academic/ research, and integrated network cancer treatment facilities. Median income quartiles and the percentage of adults with no high school degree were estimated by comparison of the patient’s zip code with US Census Bureau data. Median household income was separated into 4 groups, including lowest level of household income (< $38,000), low level of household income ($38,000 to $47,999), moderate level of household income ($48,000 to $62,999), and highest level of household income (≥ $63,000). The percentages of adults with no high school degree were divided into 4 groups: lowest level of HS education (≥ 21% ), low level of HS education (13.0% to 20.9%), moderate level of HS education (7.0% to 12.9%), and highest level of HS education (≤ 7%). The 5- and 10-year survival probabilities were calculated using the number of months between the date of diagnosis and the date of death or last known contact.

Continuous variables are presented as median and interquartile range (IQR) whereas categorical variables are presented as frequencies and proportion. IBM SPSS version 25.0 was used to produce Kaplan-Meier survival curves and descriptive statistics. This study used Kaplan- Meier survival tables and log-rank tests to analyze both the 5- and 10-year OS rates for the 5 variables listed above. This study also used a multivariable Cox regression model that accommodated the correlative nature of outcomes within facilities to study the association of the treatment facility type and other socioeconomic factors, while controlling for age, race (which was collapsed into 3 categories), sex, primary site, tumor stage, and treatment approaches. The proportional hazards assumption was individually checked for all pertinent variables. Any patient records that were missing data were excluded from the multivariable Cox regression model, which was analyzed with SAS version 9.4 (Cary, NC). P < 0.05 was used to indicate statistical significance for all analyses.

Results

Table 1 provides descriptive analysis for demographic characteristics of the 3573 patients including age, sex, and race. The median age at diagnosis was 64 years. There were 1073 more men (65%) than women (35%) in this analysis. Whites were the predominant racial category, comprising 87.7% of the patient population, followed by African Americans (6.5%) and Asians (2.5%).

Socioeconomic Variables

The largest proportion of the patient population (45.5%) had private insurance (Table 2). Medicare came in a close second covering almost 42.2% of the population, followed by Medicaid (5.0%), uninsured (2.8%), and other government insurance (1.5%). About half (53.7%) of the patients were treated at academic or research facilities, while the fewest number of patients (5.2%) underwent treatment at community cancer facilities. The largest percentage (36.6%) of patients lived in zip codes with the highest level of median household income, while 26.0% and 22.3% had moderate and low levels of income, respectively. About 14% of patients lived within an area of the lowest level of income. Similarly, almost 15% of patients lived in an area of lowest level of HS education. The greatest percentage of the patient population (34.5%) lived in a zip code with moderate level of HS education. Surgery was the most common treatment modality with 90.8% of the cohort undergoing surgery, while 35.4% and 16.5% were treated with radiation and chemotherapy, respectively (some patients received more than one type of treatment modality).

Survival Data

Survival data were available for 3112 patients. Kaplan-Meier survival curves were used to analyze OS according to insurance status, racial background, treatment facility type, median family income, and percentage of adults with no high school education. Overall 5- and 10- year OS probabilities were 51.5% and 34.8%, respectively, while the median OS (SD) was 63.57 (2.8) months (Table 3).

Private insurance showed significantly higher 5- and 10-year OS probabilities and median OS: 5-year OS was 61.2%, 10-year OS was 47.2%, and median survival (SD) was 101.2 (8.2) months compared with that of all other insurance groups (Medicare, Medicaid, other government insurance, and uninsured) (Figure 1). These other insurance types were fairly similar in their 5-year and median OS, but surprisingly, patients with no insurance had the second longest 10-year OS. The difference between the 5-year OS probabilities of private insurance compared with an average of the other insurances was 15.1%, which had almost doubled to 28.5% at 10 years, with a median OS difference of almost 5 years (56 months; data not shown).

Using the Kaplan-Meier survival curve, Asian Indians had the longest 5-year OS probability of 77.9% and African Americans had the longest 10-year OS probability of 40.6%. However, Asians as a group demonstrated the longest median (SD) OS outcome with 119.8 (47.8) months (Figure 2).

Overall, academic/research programs had the longest median OS and 5-year OS probability (SD) of 66.6 (4.5) months and 52.6%, respectively (Figure 3). Comprehensive community cancer programs and integrated network cancer programs had nearly identical 10-year OS rates (35.2% vs 35.1%, respectively). Community cancer programs had the worst 5- and 10-year OS probabilities (41.1% and 21.8%, respectively).

The top 2 income quartiles combined to demonstrate the longest median, 5-year, and 10-year OS probabilities and were very similar. Patients living in a zip code with the highest income level had the longest 5-year OS rates of 54.3%, while patients living in zip codes with a moderate income level had the longest 10-year OS at 39.3% and the longest median OS of about 71 months. Patients with the lowest level of median household income had the worst 5-year OS rates (48.3%) and a median (SD) OS of 53.4 (5.4) months (Figure 4).

A Kaplan-Meier curve for percentage of adults without a HS degree is displayed in Figure 5. Zip codes with the highest level of education had the longest 5-year OS rates and median (SD) OS of 55.3% and 70.9 (4.8) months, respectively. The longest 10-year OS outcomes at 38.1% were found in patients who lived in areas of low-education levels. The worst 5- and 10- year OS outcomes and median OS were found in the least educated zip codes.

Results from the Cox regression model of OS are displayed in Table 4. Race and ethnicity, zip code-level median household income, and zip code-level education were not associated with OS. Patients with no insurance had an increased risk of death (hazard ratio [HR], 1.84; 95% CI, 1.17-2.88; P < .01) when compared with patients with private insurance. Patients with other government insurance also had an increased risk of death (HR, 2.12; 95% CI, 1.27-3.54; P < .01) when compared with patients with private insurance while controlling for all other variables. Patients with Medicare had a decreased risk of death when compared with patients with other government insurance and no insurance (HR, 0.53; 95% CI, 0.31-0.92; P = .02 and HR, 0.62; 95% CI, 0.38-0.99; P = .05, respectively). Patients treated at academic centers had better OS when compared with patients treated at comprehensive treatment centers (HR, 0.77; 95% CI, 0.65-0.92;P < .01) and community treatment centers (HR, 0.62; 95% CI, 0.44-0.86; P < .01).

Discussion

This study is the largest study to date that specifically studies the type of treatment facilities and socioeconomic factors, including insurance status, race, income, and education, and how they affect survival of DDLPS. The overall 5- and 10-year OS probabilities for DDLPS in this study were 51.5% and 34.8%, respectively, with median OS of 63.6 months. These results were more encouraging than previous reports, which found a 5-year survival probability of 36.5% and a median OS of 45 months.^13,14

The largest age grouping was aged 61 to 80 years (48.9% of the cohort), and the median age at diagnosis was 64 years. DDLPSs most typically present between the ages of 50 and 70 years.¹⁵ Our cohort was 65% male. Previous studies have indicated that DDLPSs affect the sexes equally; however, another study showed a similar male predominance (68.8%) at the MD Anderson Cancer Center in Houston, Texas.^13,16

In our study, approximately 88% of patients were white, 6.5% were African American, and 2.5% were Asian, which differed from a previous study of 84 patients that had a 78.6% white, 4.8% Asian, and 1.2% African American patient population.¹⁴

Asian Indian or Pakistani patients had the best 5-year OS probability at 77.9%, followed by African American (57.2%), Asian (51.6%), AI/AN (51.4%), and white patients (50.9%). This trend had disappeared by 10 years and Asian, AI/AN, African American, and Asian Indian or Pakistani groups all demonstrated longer median OS than did white patients. In fact, Asian patients had the longest median OS at 119.8 months, which was almost double that of white patients with the lowest median OS of 61.2 months. This finding is contrary to previous studies, which reported that racial minorities typically had worse OS outcomes when compared with white patients in different types of cancer.^7,17 Notably, these findings were not statistically significant in our current study in the log-rank or multivariable analyses.

Private insurance was the most common form of insurance followed in decreasing order by Medicare, Medicaid, uninsured, and other government insurance. About 42% of the cohort had Medicare, which is a federally funded US insurance program designated for patients aged ≥ 65 years and certain younger patients with disabilities.

Patients with private insurance demonstrated the longest OS, essentially twice the median OS of all other insured groups at 101 months. Medicare had the worst 5-year OS probability and median OS of all groups. A previous study of 77 patients with DDLPS reported that patients aged > 65 years had reduced OS.13 Medicare patients in this study were older, with a mean and median age at DDLPS diagnosis of 71 and 72 years, respectively, while private insurance had a mean and median age at diagnosis of 56 and 57 years, respectively. Medicare inherently covers older patients and this age difference could account for the decrease in overall survival.

Improved OS for privately insured patients was most notable compared with the uninsured or patients with other government insurance. Uninsured patients had an 83.7% increased risk of mortality when compared with patients with private insurance. When compared with patients with private insurance, patients with other government insurance had an 111.5% increased risk of mortality. Comparing patients with Medicare vs patients with no insurance or other government insurance, there was a decreased risk of mortality of 38.5% and 46.6%, respectively. This decreased OS in patients with other government insurance could be related to the choice of treatment facility, because only 31% of the patients with other government insurance went to academic or research centers when compared with the 58.4% and 50.8% of patients with private and Medicare insurance treated there (data not shown). Such centers often have access to more advanced technology and protocols that may not be available at other treatment facilities.

A little more than half of the patients in the cohort went to an academic or research center for treatment (53.7%); comprehensive community cancer programs were the second most common treatment facility at 28%. Patients treated at academic or research centers demonstrated the best outcomes with a 5-year OS of 52.6%, followed in decreasing order by comprehensive community cancer programs (49.7%), integrated network cancer programs (48.8%), and community cancer programs (41.1%). In our patient cocohort, patients treated at an academic/research center had slightly decreased 10-year OS rates compared with those patients treated at a comprehensive community cancer program, although the median OS for the academic/research centers were still the highest of all treatment facilities.

Treatment options varied significantly by facility, and the number of patients treated surgically followed a similar trend, with 92% undergoing surgery as the primary treatment at academic or research programs compared with 89% at comprehensive cancer programs and 82.7% at community cancer programs (data not shown). Another potential explaination for differing OS outcomes across facilities is the surgical margin outcome. Surgeries performed at community cancer programs or comprehensive cancer programs resulted with no residual tumor in 36% and 40% of cases, respectively, whereas cases performed at academic or research programs resulted with no residual tumor in 47% of cases (data not shown). Regardless, multivariate analysis demonstrated a marked decrease in the chance of mortality when comparing treatment received at academic facility centers with that received at comprehensive cancer centers (22.9%) and community cancer centers (38.3%) (data not shown).

A recent study demonstrated improved outcomes for patients with retroperitoneal or extremity STS treated at high-volume treatment centers.¹⁸ Patients treated at high-volume centers were found to have an 8% decreased risk of death compared with patients treated at low-volume centers. Notably, they found highvolume academic centers demonstrated the strongest improvement in survival, while highvolume community centers showed decreased survival.¹⁸ Similarly, we found that patients treated at academic/research institutions had improved 5-year OS and greater median OS than did patients treated at community cancer programs or comprehensive community cancer programs.

The top 2 income quartiles (≥ $48,000) combined to demonstrate the longest median, 5-year, and 10-year OS and were fairly similar between the quartiles. Patients living in zip codes with a median income of $38,000 to $47,999 had the worst 5-year OS and median OS. The log-rank analysis showed statistical evidence of differences in survival associated with income, but within the context of the multivariable analysis, there was no remaining evidence of a difference.

The longest 5-year OS outcomes were seen in patients living in zip codes with the highest level of education (55.3%). However, the difference in OS was not statistically significant using either the log-rank analysis or multivariate analysis.

Limitations

This study has certain inherent limitations in using a retrospective design and a large database such as the NCDB. Many different pathologists at CoC-accredited cancer programs perform the pathology that contributes to the data in the NCDB. There was no pathological review of these findings, which could potentially introduce error into the findings of this study. With the NCDB, potential selection bias is possible because patients in the database are added only from CoC-accredited cancer programs. This risk is minimized because NCDB contains data on most newly diagnosed cancer patients in the US. Further potential risks, which are unable to be controlled for, include potential interobserver error and data that may be incompletely, improperly, or inaccurately recorded from the patients’ charts. Without patient-specific information regarding income and education, it is challenging to utilize zip codes to estimate socioeconomic status and educational level. Even though a patient may live in a zip code identified with specific economic and educational characteristics, that patient may not share those characteristics. Furthermore, patients with Medicare tend to be older than patients with other forms of insurance, which limits the significance of comparisons across insurance groups. A future SEER (Surveillance, Epidemiology, and End Results) program study to confirm this study’s results and the effects of socioeconomic variables on DDLPS would be an excellent followup study.

Conclusion

This study used a large cohort of patients with DDLPS to study the effects of treatment facility, insurance status, and socioeconomic variables on survival outcomes. Although insurance status, median household income, and treatment facility were associated with differences in median OS and 5- and 10-year OS probabilities, evidence for a difference remained for only insurance status and facility type within the context of a multivariable analysis irrespective of age, race, sex, insurance status, education, and median income. Patients with private insurance and Medicaid had a decreased risk of mortality compared with other government insurance and no insurance. Patients receiving treatment at academic research programs had the highest median and 5-year OS of 66.6 months and 52.6%, respectively. Patients receiving treatment at academic centers had improved survival outcomes with a decrease in mortality of 23% and 38% compared to comprehensive or community cancer programs.

Thyroid nodules are identified incidentally in 4% to 10% of the general population in the US.^1,2 Clinicians and patients often are concerned about potential malignancy when thyroid nodules are identified because 5% to 15% of nodules will be cancerous.¹ The most common form of cancer is papillary carcinoma followed by follicular carcinoma.² Initially, serum thyroid-stimulating hormone (TSH) levels and thyroid ultrasound are used to evaluate a thyroid nodule because both tests can reveal vital information about malignancy potential.³ Ultrasound characteristics, such as macrocalcifications, hypoechogenicity, absence of halo, increased vascularity, and irregular nodular margins, increase suspicion for malignancy and warrant further investigation.³

Ultrasound-guided fine-needle aspiration (FNA) is the modality of choice for evaluation of thyroid nodules with sensitivity and specificity > 90%.^2,4 Most patients receive a definitive diagnosis with this test; however, about 25% of cases are indeterminate based on the Bethesda System and require surgical investigation.³

Currently, it is well accepted clinical practice to refer all nodules > 4 cm for surgical intervention regardless of malignancy risk factors or the mass effect of the nodule.^3-6 The preference for surgery—rather than FNA—is because of the notable false negative rate with FNA in larger nodules; studies have described false negative rates for FNA close to 10%.^7,8 In contrast, Megwalu recently reported a FNA false negative rate of 0%.⁹

The risk of malignancy associated with nodule size has been researched for many years, but studies have produced conflicting results. In this retrospective cohort study, the authors compared malignancy rates between patients with nodules ≥ 3 cm and those with nodules < 3 cm.

Methods

The authors performed a retrospective chart review of the medical records of 329 patients presenting for thyroid nodule evaluation found on physical exam or incidentally identified with imaging at the Dayton Veteran Affairs Medical Center from January 2000 to May 2016. Data collection included sex, age, race, personal history of neck radiation treatment, family history of thyroid cancer, personal history of thyroid cancer, hot nodules/Graves disease, abnormal neck lymph nodes, and serum TSH levels. The authors looked for an association between TSH level and cancer. Hot thyroid nodules are known to have low risk of malignancy.

All patients aged 18 to 99 years with a thyroid nodule evaluated with FNA were included in the study. Patients were divided into 2 groups, those with nodules ≥ 3 cm and those with nodules < 3 cm. For nodules requiring subsequent biopsies, only the initial nodule biopsy was included in our study. The 3-cm cutoff was selected based on previous studies.^1,5,10 Patients who did not undergo a FNA study were excluded. Indications for surgery were positive FNA results, suspicious imaging, size of nodule, or patient preference.

Means and standard deviations are reported for continuous variables and counts and percentages for categorical variables. We used the Mann-Whitney test for comparisons involving continuous variables with 2 groups and the Kruskal-Wallis test for 4 groups. The chi-square test—corrected for continuity if necessary—was used to compare 2 categorical variables. We used multiple logistic regression to adjust for demographic and clinical variables other than nodule size that were related to malignancy. Inferences were made at the 0.05 level of significance.

Results

A total of 329 patients with thyroid nodules were identified: 236 were < 3 cm and 93 were ≥ 3 cm. The 2 groups differed on race, with more white patients in the < 3-cm nodule group (78% vs 67%, P = .036) (Table 1). 

Prevalence of cancer based on FNA in nodules < 3 cm was 6.4% (95% CI, 3.6%–10.3%) and nodules ≥ 3 cm was 8.6% (95% CI, 3.8%–16.2%; P = .23) (Table 2). 

When divided into 4 subgroups, cancer using FNA was found in 35.1% of nodules < 2 cm, 21.1% of nodules 2 cm to < 3 cm, 42.1% of nodules 3 cm to 4 cm, and 18.2% of nodules > 4 cm (P = .32) (Table 3). 

Surgical pathology results showed 17 cases of papillary carcinoma in nodules < 3 cm, whereas there were 9 cases of papillary carcinoma and 1 case of follicular carcinoma in nodules > 3 cm. When correlated with the cytology results, 10 cases were reported as benign, 11 were malignant, and 6 samples were non-diagnostic.

There were 30 nondiagnostic FNA samples: 7 patients had surgery, 19 were monitored with serial imaging, 2 were lost to follow-up, and 2 expired for other reasons. Of the 19 patients who were monitored with serial imaging, the nodules were stable and did not require repeat sampling.

Discussion

The authors found no relationship between thyroid nodule size and malignancy over a 16-year period in a veteran population, either with FNA or surgical pathology. The lack of relationship persists when adjusted for the only nonthyroid variable on which the 2 groups differed (race).

The finding of no relationship between larger thyroid nodule size and cancer is consistent with other studies. In a 10-year chart review of 695 patients at Walter Reed Army Medical Center, Burch and colleagues found a malignancy rate of 18.6% but no association between thyroid nodule size and malignancy.¹¹ They concluded that nodules ≥ 4 cm did not increase malignancy risk. In a 3-year retrospective study of 326 patients, Mangister and colleagues reported that the malignancy rate was higher in nodules < 3 cm (48.4%) compared with nodules ≥ 3 cm (33.3%).¹⁰ This study concluded that the malignancy potential of thyroid nodules peaked at 2 cm and decreased at > 3 cm. Kamran and colleagues reported a nonlinear relationship between nodule size and malignancy with a threshold of 2 cm, beyond which there was no increased risk of malignancy.¹

Conversely, in a prospective study Kuru and colleagues followed 571 patients who had undergone thyroidectomy and found that nodules ≥ 4 cm were associated with increased malignancy risk compared with nodules < 4 cm. However, with a cutoff of 3 cm there was no relationship.⁵ Discrepancies among studies might be because of variability in patient demographics and the prevalence of thyroid cancer in a specific institution. Although the majority of thyroid nodules are seen in females, the current study’s population was predominantly male and entirely veteran. Consequently, interpretation of these studies highlight the need to individualize clinical decision-making for each patient.

Limitations

This study has several limitations. It was conducted at a single institution with a group of veterans, which limits the ability to generalize its results to the general population. Second, data omissions are likely in retrospective chart reviews, and ensuring accuracy of data collection could be challenging. Third, all thyroid nodules found to be benign with cytology did not undergo surgical intervention to confirm the diagnosis; therefore, only 93 of 329 nodules were evaluated with the definitive diagnostic test. Therefore, selection bias was introduced into the nodule size comparisons when surgical intervention was used to measure the outcome. However, because false negative rates for FNA is low, likely few malignant nodules were missed. In addition, all patients with thyroid nodules are not referred for surgery because of potential complications.

Conclusion

This study strongly suggests there is no increased or decreased cancer risk for thyroid nodules ≥ 3 cm compared with those < 3 cm. Current clinical practice is to refer patients with larger nodules for surgical evaluation. In a large systemic review, Shin and colleagues reported higher pretest probability of malignancy in larger nodules and recommended consideration of surgical intervention for nodules > 3 cm because of false negatives and concerns for diagnostic inaccuracy with FNA.⁸ Although data were mixed, Shin and colleagues reported higher incidence of false negative FNA results in larger nodules.⁸ Given the authors’ findings and earlier conflicting results, the decision for surgical intervention cannot be made solely on nodule size and requires consideration of additional factors including FNA results, nodule characteristics, patient risk factors, and patient preference.

Thyroid nodules are identified incidentally in 4% to 10% of the general population in the US.^1,2 Clinicians and patients often are concerned about potential malignancy when thyroid nodules are identified because 5% to 15% of nodules will be cancerous.¹ The most common form of cancer is papillary carcinoma followed by follicular carcinoma.² Initially, serum thyroid-stimulating hormone (TSH) levels and thyroid ultrasound are used to evaluate a thyroid nodule because both tests can reveal vital information about malignancy potential.³ Ultrasound characteristics, such as macrocalcifications, hypoechogenicity, absence of halo, increased vascularity, and irregular nodular margins, increase suspicion for malignancy and warrant further investigation.³

Ultrasound-guided fine-needle aspiration (FNA) is the modality of choice for evaluation of thyroid nodules with sensitivity and specificity > 90%.^2,4 Most patients receive a definitive diagnosis with this test; however, about 25% of cases are indeterminate based on the Bethesda System and require surgical investigation.³

Currently, it is well accepted clinical practice to refer all nodules > 4 cm for surgical intervention regardless of malignancy risk factors or the mass effect of the nodule.^3-6 The preference for surgery—rather than FNA—is because of the notable false negative rate with FNA in larger nodules; studies have described false negative rates for FNA close to 10%.^7,8 In contrast, Megwalu recently reported a FNA false negative rate of 0%.⁹

The risk of malignancy associated with nodule size has been researched for many years, but studies have produced conflicting results. In this retrospective cohort study, the authors compared malignancy rates between patients with nodules ≥ 3 cm and those with nodules < 3 cm.

Methods

The authors performed a retrospective chart review of the medical records of 329 patients presenting for thyroid nodule evaluation found on physical exam or incidentally identified with imaging at the Dayton Veteran Affairs Medical Center from January 2000 to May 2016. Data collection included sex, age, race, personal history of neck radiation treatment, family history of thyroid cancer, personal history of thyroid cancer, hot nodules/Graves disease, abnormal neck lymph nodes, and serum TSH levels. The authors looked for an association between TSH level and cancer. Hot thyroid nodules are known to have low risk of malignancy.

All patients aged 18 to 99 years with a thyroid nodule evaluated with FNA were included in the study. Patients were divided into 2 groups, those with nodules ≥ 3 cm and those with nodules < 3 cm. For nodules requiring subsequent biopsies, only the initial nodule biopsy was included in our study. The 3-cm cutoff was selected based on previous studies.^1,5,10 Patients who did not undergo a FNA study were excluded. Indications for surgery were positive FNA results, suspicious imaging, size of nodule, or patient preference.

Means and standard deviations are reported for continuous variables and counts and percentages for categorical variables. We used the Mann-Whitney test for comparisons involving continuous variables with 2 groups and the Kruskal-Wallis test for 4 groups. The chi-square test—corrected for continuity if necessary—was used to compare 2 categorical variables. We used multiple logistic regression to adjust for demographic and clinical variables other than nodule size that were related to malignancy. Inferences were made at the 0.05 level of significance.

Results

A total of 329 patients with thyroid nodules were identified: 236 were < 3 cm and 93 were ≥ 3 cm. The 2 groups differed on race, with more white patients in the < 3-cm nodule group (78% vs 67%, P = .036) (Table 1). 

Prevalence of cancer based on FNA in nodules < 3 cm was 6.4% (95% CI, 3.6%–10.3%) and nodules ≥ 3 cm was 8.6% (95% CI, 3.8%–16.2%; P = .23) (Table 2). 

When divided into 4 subgroups, cancer using FNA was found in 35.1% of nodules < 2 cm, 21.1% of nodules 2 cm to < 3 cm, 42.1% of nodules 3 cm to 4 cm, and 18.2% of nodules > 4 cm (P = .32) (Table 3). 

Surgical pathology results showed 17 cases of papillary carcinoma in nodules < 3 cm, whereas there were 9 cases of papillary carcinoma and 1 case of follicular carcinoma in nodules > 3 cm. When correlated with the cytology results, 10 cases were reported as benign, 11 were malignant, and 6 samples were non-diagnostic.

There were 30 nondiagnostic FNA samples: 7 patients had surgery, 19 were monitored with serial imaging, 2 were lost to follow-up, and 2 expired for other reasons. Of the 19 patients who were monitored with serial imaging, the nodules were stable and did not require repeat sampling.

Discussion

The authors found no relationship between thyroid nodule size and malignancy over a 16-year period in a veteran population, either with FNA or surgical pathology. The lack of relationship persists when adjusted for the only nonthyroid variable on which the 2 groups differed (race).

The finding of no relationship between larger thyroid nodule size and cancer is consistent with other studies. In a 10-year chart review of 695 patients at Walter Reed Army Medical Center, Burch and colleagues found a malignancy rate of 18.6% but no association between thyroid nodule size and malignancy.¹¹ They concluded that nodules ≥ 4 cm did not increase malignancy risk. In a 3-year retrospective study of 326 patients, Mangister and colleagues reported that the malignancy rate was higher in nodules < 3 cm (48.4%) compared with nodules ≥ 3 cm (33.3%).¹⁰ This study concluded that the malignancy potential of thyroid nodules peaked at 2 cm and decreased at > 3 cm. Kamran and colleagues reported a nonlinear relationship between nodule size and malignancy with a threshold of 2 cm, beyond which there was no increased risk of malignancy.¹

Conversely, in a prospective study Kuru and colleagues followed 571 patients who had undergone thyroidectomy and found that nodules ≥ 4 cm were associated with increased malignancy risk compared with nodules < 4 cm. However, with a cutoff of 3 cm there was no relationship.⁵ Discrepancies among studies might be because of variability in patient demographics and the prevalence of thyroid cancer in a specific institution. Although the majority of thyroid nodules are seen in females, the current study’s population was predominantly male and entirely veteran. Consequently, interpretation of these studies highlight the need to individualize clinical decision-making for each patient.

Limitations

This study has several limitations. It was conducted at a single institution with a group of veterans, which limits the ability to generalize its results to the general population. Second, data omissions are likely in retrospective chart reviews, and ensuring accuracy of data collection could be challenging. Third, all thyroid nodules found to be benign with cytology did not undergo surgical intervention to confirm the diagnosis; therefore, only 93 of 329 nodules were evaluated with the definitive diagnostic test. Therefore, selection bias was introduced into the nodule size comparisons when surgical intervention was used to measure the outcome. However, because false negative rates for FNA is low, likely few malignant nodules were missed. In addition, all patients with thyroid nodules are not referred for surgery because of potential complications.

Conclusion

This study strongly suggests there is no increased or decreased cancer risk for thyroid nodules ≥ 3 cm compared with those < 3 cm. Current clinical practice is to refer patients with larger nodules for surgical evaluation. In a large systemic review, Shin and colleagues reported higher pretest probability of malignancy in larger nodules and recommended consideration of surgical intervention for nodules > 3 cm because of false negatives and concerns for diagnostic inaccuracy with FNA.⁸ Although data were mixed, Shin and colleagues reported higher incidence of false negative FNA results in larger nodules.⁸ Given the authors’ findings and earlier conflicting results, the decision for surgical intervention cannot be made solely on nodule size and requires consideration of additional factors including FNA results, nodule characteristics, patient risk factors, and patient preference.

Thyroid nodules are identified incidentally in 4% to 10% of the general population in the US.^1,2 Clinicians and patients often are concerned about potential malignancy when thyroid nodules are identified because 5% to 15% of nodules will be cancerous.¹ The most common form of cancer is papillary carcinoma followed by follicular carcinoma.² Initially, serum thyroid-stimulating hormone (TSH) levels and thyroid ultrasound are used to evaluate a thyroid nodule because both tests can reveal vital information about malignancy potential.³ Ultrasound characteristics, such as macrocalcifications, hypoechogenicity, absence of halo, increased vascularity, and irregular nodular margins, increase suspicion for malignancy and warrant further investigation.³

Ultrasound-guided fine-needle aspiration (FNA) is the modality of choice for evaluation of thyroid nodules with sensitivity and specificity > 90%.^2,4 Most patients receive a definitive diagnosis with this test; however, about 25% of cases are indeterminate based on the Bethesda System and require surgical investigation.³

Currently, it is well accepted clinical practice to refer all nodules > 4 cm for surgical intervention regardless of malignancy risk factors or the mass effect of the nodule.^3-6 The preference for surgery—rather than FNA—is because of the notable false negative rate with FNA in larger nodules; studies have described false negative rates for FNA close to 10%.^7,8 In contrast, Megwalu recently reported a FNA false negative rate of 0%.⁹

The risk of malignancy associated with nodule size has been researched for many years, but studies have produced conflicting results. In this retrospective cohort study, the authors compared malignancy rates between patients with nodules ≥ 3 cm and those with nodules < 3 cm.

Methods

The authors performed a retrospective chart review of the medical records of 329 patients presenting for thyroid nodule evaluation found on physical exam or incidentally identified with imaging at the Dayton Veteran Affairs Medical Center from January 2000 to May 2016. Data collection included sex, age, race, personal history of neck radiation treatment, family history of thyroid cancer, personal history of thyroid cancer, hot nodules/Graves disease, abnormal neck lymph nodes, and serum TSH levels. The authors looked for an association between TSH level and cancer. Hot thyroid nodules are known to have low risk of malignancy.

All patients aged 18 to 99 years with a thyroid nodule evaluated with FNA were included in the study. Patients were divided into 2 groups, those with nodules ≥ 3 cm and those with nodules < 3 cm. For nodules requiring subsequent biopsies, only the initial nodule biopsy was included in our study. The 3-cm cutoff was selected based on previous studies.^1,5,10 Patients who did not undergo a FNA study were excluded. Indications for surgery were positive FNA results, suspicious imaging, size of nodule, or patient preference.

Means and standard deviations are reported for continuous variables and counts and percentages for categorical variables. We used the Mann-Whitney test for comparisons involving continuous variables with 2 groups and the Kruskal-Wallis test for 4 groups. The chi-square test—corrected for continuity if necessary—was used to compare 2 categorical variables. We used multiple logistic regression to adjust for demographic and clinical variables other than nodule size that were related to malignancy. Inferences were made at the 0.05 level of significance.

Results

A total of 329 patients with thyroid nodules were identified: 236 were < 3 cm and 93 were ≥ 3 cm. The 2 groups differed on race, with more white patients in the < 3-cm nodule group (78% vs 67%, P = .036) (Table 1). 

Prevalence of cancer based on FNA in nodules < 3 cm was 6.4% (95% CI, 3.6%–10.3%) and nodules ≥ 3 cm was 8.6% (95% CI, 3.8%–16.2%; P = .23) (Table 2). 

When divided into 4 subgroups, cancer using FNA was found in 35.1% of nodules < 2 cm, 21.1% of nodules 2 cm to < 3 cm, 42.1% of nodules 3 cm to 4 cm, and 18.2% of nodules > 4 cm (P = .32) (Table 3). 

Surgical pathology results showed 17 cases of papillary carcinoma in nodules < 3 cm, whereas there were 9 cases of papillary carcinoma and 1 case of follicular carcinoma in nodules > 3 cm. When correlated with the cytology results, 10 cases were reported as benign, 11 were malignant, and 6 samples were non-diagnostic.

There were 30 nondiagnostic FNA samples: 7 patients had surgery, 19 were monitored with serial imaging, 2 were lost to follow-up, and 2 expired for other reasons. Of the 19 patients who were monitored with serial imaging, the nodules were stable and did not require repeat sampling.

Discussion

The authors found no relationship between thyroid nodule size and malignancy over a 16-year period in a veteran population, either with FNA or surgical pathology. The lack of relationship persists when adjusted for the only nonthyroid variable on which the 2 groups differed (race).

The finding of no relationship between larger thyroid nodule size and cancer is consistent with other studies. In a 10-year chart review of 695 patients at Walter Reed Army Medical Center, Burch and colleagues found a malignancy rate of 18.6% but no association between thyroid nodule size and malignancy.¹¹ They concluded that nodules ≥ 4 cm did not increase malignancy risk. In a 3-year retrospective study of 326 patients, Mangister and colleagues reported that the malignancy rate was higher in nodules < 3 cm (48.4%) compared with nodules ≥ 3 cm (33.3%).¹⁰ This study concluded that the malignancy potential of thyroid nodules peaked at 2 cm and decreased at > 3 cm. Kamran and colleagues reported a nonlinear relationship between nodule size and malignancy with a threshold of 2 cm, beyond which there was no increased risk of malignancy.¹

Conversely, in a prospective study Kuru and colleagues followed 571 patients who had undergone thyroidectomy and found that nodules ≥ 4 cm were associated with increased malignancy risk compared with nodules < 4 cm. However, with a cutoff of 3 cm there was no relationship.⁵ Discrepancies among studies might be because of variability in patient demographics and the prevalence of thyroid cancer in a specific institution. Although the majority of thyroid nodules are seen in females, the current study’s population was predominantly male and entirely veteran. Consequently, interpretation of these studies highlight the need to individualize clinical decision-making for each patient.

Limitations

This study has several limitations. It was conducted at a single institution with a group of veterans, which limits the ability to generalize its results to the general population. Second, data omissions are likely in retrospective chart reviews, and ensuring accuracy of data collection could be challenging. Third, all thyroid nodules found to be benign with cytology did not undergo surgical intervention to confirm the diagnosis; therefore, only 93 of 329 nodules were evaluated with the definitive diagnostic test. Therefore, selection bias was introduced into the nodule size comparisons when surgical intervention was used to measure the outcome. However, because false negative rates for FNA is low, likely few malignant nodules were missed. In addition, all patients with thyroid nodules are not referred for surgery because of potential complications.

Conclusion

This study strongly suggests there is no increased or decreased cancer risk for thyroid nodules ≥ 3 cm compared with those < 3 cm. Current clinical practice is to refer patients with larger nodules for surgical evaluation. In a large systemic review, Shin and colleagues reported higher pretest probability of malignancy in larger nodules and recommended consideration of surgical intervention for nodules > 3 cm because of false negatives and concerns for diagnostic inaccuracy with FNA.⁸ Although data were mixed, Shin and colleagues reported higher incidence of false negative FNA results in larger nodules.⁸ Given the authors’ findings and earlier conflicting results, the decision for surgical intervention cannot be made solely on nodule size and requires consideration of additional factors including FNA results, nodule characteristics, patient risk factors, and patient preference.

Cancer and cancer-related treatment can cause a myriad of adverse effects.^1,2 Early identification and management of these symptoms is paramount to the success of cancer treatment completion; however, clinic and telephonic strategies for addressing symptoms often result in delays in care.¹ New strategies for patient engagement in the management of cancer and treatment-related symptoms are needed.

The use of online self-management tools can result in improvement in symptoms, reduce cancer symptom distress, improve quality-of-life, and improve medication adherence.^3-9 A meta-analysis concluded that online interventions showed promise, but optimizing interventions would require additional research.¹⁰ Another meta-analysis found that online self-management was effective in managing several symptoms.¹¹ An e-health method of collecting patient self-reported symptoms has been found to be acceptable to patients and feasible for use.^12-14 We postulated that a mobile text messaging strategy may be an effective modality for augmenting symptom management for cancer patients in real time.

In the US Departmant of Veterans Affairs (VA), “Annie,” a self-care tool utilizing a text-messaging system has been implemented. Annie was developed modeling “Flo,” a messaging system in the United Kingdom that has been used for case management of chronic obstructive pulmonary disease, heart failure, stress incontinence, asthma, as a medication reminder tool, and to provide support for weight loss or post-operatively.^15-17 Using Annie in the US, veterans have the ability to receive and track health information. Use of the Annie program has demonstrated improved continuous positive airway pressure monitor utilization in veterans with traumatic brain injury.¹⁸ Other uses within the Veterans Health Administration (VHA) include assisting patients with anger management, liver disease, anxiety, asthma, diabetes, HIV, hypertension, weight loss, and smoking cessation.

Methods

The Hematology/Oncology division of the Minneapolis VA Healthcare System (MVAHCS) is a tertiary care facility that administers about 260 new chemotherapy regimens annually. The MVAHCS interdisciplinary hematology/oncology group initiated a quality improvement project to determine the feasibility, acceptability, and experience of tailoring the Annie tool for self-management of cancer symptoms. The group consisted of 2 physicians, 3 advanced practice registered nurses, 1 physician assistant, 2 registered nurses, and 2 Annie program team members.

We first created a symptom management pilot protocol as a result of multidisciplinary team discussions. Examples of discussion points for consideration included, but were not limited to, timing of texts, amount of information to ask for and provide, what potential symptoms to consider, and which patient population to pilot first.

Results

Through screening new patient consults or those referred for chemotherapy education, 15 male veterans enrolled in the symptom monitoring program over an 8 month period. There were additional patients who were not offered the program or chose not to participate; often due to not having texting capabilities on their phone or not liking the texting feature. The majority of those who participated in the program (n = 14) were enrolled at the start of Cycle 1; the other patient was enrolled at the start of Cycle 2. Patients were enrolled an average of 89 days (range 8-204). Average response rate was 84.2% (range 30-100%).

Although symptoms were not reviewed in real time, we reviewed responses to determine the utilization of the instructions given for the program. No veteran had 0 symptoms reported. There were numerous occurrences of a score of 1 or 2. Many of these patients had baseline symptoms due to their underlying cancer. A score of 3 or 4 on the system prompted the patient to call the clinic or go to the ED. Seven patients (some with multiple occurrences) were prompted to call; only 4 of these made the follow-up call to the clinic. All were offered a same day visit, but each declined. Only 1 patient reported a symptom on a day not prompted for that symptom. Symptoms that were reported are listed in order of frequency: fatigue, appetite loss, numbness, pain, mouth sore, and breathing difficulty. There were no visits to the ED.

Program Evaluation

An evaluation was conducted 30 to 60 days after program enrollment. We elicited feedback to determine who was reading and responding to the text message: the patient, a family member, or a caregiver; whether they found the prompts helpful and took action; how they felt about the number of texts; if they felt the program was helpful; and any other feedback that would improve the program. In general, the patients (8) answered the texts independently. In 4 cases, the spouse answered the texts, and 3 patients answered the texts together with their spouses. Most patients (11) found the amount of texting to be “just right.” However, 3 found it to be too many texts and 1 didn’t find the amount of texting to be enough.

Three veterans did not have enough symptoms to feel the program was of benefit to them, but they did feel it would have been helpful if they had been more symptomatic. One veteran recalled taking loperamide as needed, as a result of prompting. No veterans felt as though the texting feature was difficult to use; and overall, were very positive about the program. Several appreciated receiving messages that validated when they were doing well, and they felt empowered by self-management. One of the spouses was a registered nurse and found the information too basic to be of use.

Discussion

Initial evaluation of the program via survey found no technology challenges. Patients have been very positive about the program including ease of use, appreciation of messages that validated when they were doing well, empowerment of self-management, and some utilization of the texting advice for symptom management. Educational hyperlinks for constipation, fatigue, diarrhea, and nausea/vomiting were added after this evaluation, and patients felt that these additions provided a higher level of education.

Staff time for this intervention was minimal. A nurse navigator offered the texting program to the patient during chemotherapy education, along with some instructions, which generally took about 5 minutes. One of the Annie program staff enrolled the patient. From that point forward, this was a self-management tool, beyond checking to ensure that the patient was successful in starting the program and evaluating use for the purposes of this quality improvement project. This self-management tool did not replace any other mechanism that a patient would normally have in our department for seeking help for symptoms. The MVAHSC typical process for symptom management is to have patients call a 24/7 nurse line. If the triage nurse feels the symptoms are related to the patient’s cancer or cancer treatment, they are referred to the physician assistant who is assigned to take those calls and has the option to see the patient the same day. Patients could continue to call the nurse line or speak with providers at the next appointment at their discretion.

Conclusion

Although Annie has the option of using either text messaging or a mobile application, this project only utilized text messaging. The study by Basch and colleagues was the closest randomized trial we could identify to compare to our quality improvement intervention.⁵ The 2 main, distinct differences were that Basch and colleagues utilized online monitoring; and nurses were utilized to screen and intervene on responses, as appropriate.

The ability of our program to text patients without the use of an application or tablet, may enable more patients to participate due to ease of use. There would be no increased in expected workload for clinical staff, and may lead to decreased call burden. Since our program is automated, while still providing patients with the option to call and speak with a staff member as needed, this is a cost-effective, first-line option for symptom management for those experiencing cancer-related symptoms. We believe this text messaging tool can have system wide use and benefit throughout the VHA.

Cancer and cancer-related treatment can cause a myriad of adverse effects.^1,2 Early identification and management of these symptoms is paramount to the success of cancer treatment completion; however, clinic and telephonic strategies for addressing symptoms often result in delays in care.¹ New strategies for patient engagement in the management of cancer and treatment-related symptoms are needed.

The use of online self-management tools can result in improvement in symptoms, reduce cancer symptom distress, improve quality-of-life, and improve medication adherence.^3-9 A meta-analysis concluded that online interventions showed promise, but optimizing interventions would require additional research.¹⁰ Another meta-analysis found that online self-management was effective in managing several symptoms.¹¹ An e-health method of collecting patient self-reported symptoms has been found to be acceptable to patients and feasible for use.^12-14 We postulated that a mobile text messaging strategy may be an effective modality for augmenting symptom management for cancer patients in real time.

In the US Departmant of Veterans Affairs (VA), “Annie,” a self-care tool utilizing a text-messaging system has been implemented. Annie was developed modeling “Flo,” a messaging system in the United Kingdom that has been used for case management of chronic obstructive pulmonary disease, heart failure, stress incontinence, asthma, as a medication reminder tool, and to provide support for weight loss or post-operatively.^15-17 Using Annie in the US, veterans have the ability to receive and track health information. Use of the Annie program has demonstrated improved continuous positive airway pressure monitor utilization in veterans with traumatic brain injury.¹⁸ Other uses within the Veterans Health Administration (VHA) include assisting patients with anger management, liver disease, anxiety, asthma, diabetes, HIV, hypertension, weight loss, and smoking cessation.

Methods

The Hematology/Oncology division of the Minneapolis VA Healthcare System (MVAHCS) is a tertiary care facility that administers about 260 new chemotherapy regimens annually. The MVAHCS interdisciplinary hematology/oncology group initiated a quality improvement project to determine the feasibility, acceptability, and experience of tailoring the Annie tool for self-management of cancer symptoms. The group consisted of 2 physicians, 3 advanced practice registered nurses, 1 physician assistant, 2 registered nurses, and 2 Annie program team members.

We first created a symptom management pilot protocol as a result of multidisciplinary team discussions. Examples of discussion points for consideration included, but were not limited to, timing of texts, amount of information to ask for and provide, what potential symptoms to consider, and which patient population to pilot first.

Results

Through screening new patient consults or those referred for chemotherapy education, 15 male veterans enrolled in the symptom monitoring program over an 8 month period. There were additional patients who were not offered the program or chose not to participate; often due to not having texting capabilities on their phone or not liking the texting feature. The majority of those who participated in the program (n = 14) were enrolled at the start of Cycle 1; the other patient was enrolled at the start of Cycle 2. Patients were enrolled an average of 89 days (range 8-204). Average response rate was 84.2% (range 30-100%).

Although symptoms were not reviewed in real time, we reviewed responses to determine the utilization of the instructions given for the program. No veteran had 0 symptoms reported. There were numerous occurrences of a score of 1 or 2. Many of these patients had baseline symptoms due to their underlying cancer. A score of 3 or 4 on the system prompted the patient to call the clinic or go to the ED. Seven patients (some with multiple occurrences) were prompted to call; only 4 of these made the follow-up call to the clinic. All were offered a same day visit, but each declined. Only 1 patient reported a symptom on a day not prompted for that symptom. Symptoms that were reported are listed in order of frequency: fatigue, appetite loss, numbness, pain, mouth sore, and breathing difficulty. There were no visits to the ED.

Program Evaluation

An evaluation was conducted 30 to 60 days after program enrollment. We elicited feedback to determine who was reading and responding to the text message: the patient, a family member, or a caregiver; whether they found the prompts helpful and took action; how they felt about the number of texts; if they felt the program was helpful; and any other feedback that would improve the program. In general, the patients (8) answered the texts independently. In 4 cases, the spouse answered the texts, and 3 patients answered the texts together with their spouses. Most patients (11) found the amount of texting to be “just right.” However, 3 found it to be too many texts and 1 didn’t find the amount of texting to be enough.

Three veterans did not have enough symptoms to feel the program was of benefit to them, but they did feel it would have been helpful if they had been more symptomatic. One veteran recalled taking loperamide as needed, as a result of prompting. No veterans felt as though the texting feature was difficult to use; and overall, were very positive about the program. Several appreciated receiving messages that validated when they were doing well, and they felt empowered by self-management. One of the spouses was a registered nurse and found the information too basic to be of use.

Discussion

Initial evaluation of the program via survey found no technology challenges. Patients have been very positive about the program including ease of use, appreciation of messages that validated when they were doing well, empowerment of self-management, and some utilization of the texting advice for symptom management. Educational hyperlinks for constipation, fatigue, diarrhea, and nausea/vomiting were added after this evaluation, and patients felt that these additions provided a higher level of education.

Staff time for this intervention was minimal. A nurse navigator offered the texting program to the patient during chemotherapy education, along with some instructions, which generally took about 5 minutes. One of the Annie program staff enrolled the patient. From that point forward, this was a self-management tool, beyond checking to ensure that the patient was successful in starting the program and evaluating use for the purposes of this quality improvement project. This self-management tool did not replace any other mechanism that a patient would normally have in our department for seeking help for symptoms. The MVAHSC typical process for symptom management is to have patients call a 24/7 nurse line. If the triage nurse feels the symptoms are related to the patient’s cancer or cancer treatment, they are referred to the physician assistant who is assigned to take those calls and has the option to see the patient the same day. Patients could continue to call the nurse line or speak with providers at the next appointment at their discretion.

Conclusion

Although Annie has the option of using either text messaging or a mobile application, this project only utilized text messaging. The study by Basch and colleagues was the closest randomized trial we could identify to compare to our quality improvement intervention.⁵ The 2 main, distinct differences were that Basch and colleagues utilized online monitoring; and nurses were utilized to screen and intervene on responses, as appropriate.

The ability of our program to text patients without the use of an application or tablet, may enable more patients to participate due to ease of use. There would be no increased in expected workload for clinical staff, and may lead to decreased call burden. Since our program is automated, while still providing patients with the option to call and speak with a staff member as needed, this is a cost-effective, first-line option for symptom management for those experiencing cancer-related symptoms. We believe this text messaging tool can have system wide use and benefit throughout the VHA.

Cancer and cancer-related treatment can cause a myriad of adverse effects.^1,2 Early identification and management of these symptoms is paramount to the success of cancer treatment completion; however, clinic and telephonic strategies for addressing symptoms often result in delays in care.¹ New strategies for patient engagement in the management of cancer and treatment-related symptoms are needed.

The use of online self-management tools can result in improvement in symptoms, reduce cancer symptom distress, improve quality-of-life, and improve medication adherence.^3-9 A meta-analysis concluded that online interventions showed promise, but optimizing interventions would require additional research.¹⁰ Another meta-analysis found that online self-management was effective in managing several symptoms.¹¹ An e-health method of collecting patient self-reported symptoms has been found to be acceptable to patients and feasible for use.^12-14 We postulated that a mobile text messaging strategy may be an effective modality for augmenting symptom management for cancer patients in real time.

In the US Departmant of Veterans Affairs (VA), “Annie,” a self-care tool utilizing a text-messaging system has been implemented. Annie was developed modeling “Flo,” a messaging system in the United Kingdom that has been used for case management of chronic obstructive pulmonary disease, heart failure, stress incontinence, asthma, as a medication reminder tool, and to provide support for weight loss or post-operatively.^15-17 Using Annie in the US, veterans have the ability to receive and track health information. Use of the Annie program has demonstrated improved continuous positive airway pressure monitor utilization in veterans with traumatic brain injury.¹⁸ Other uses within the Veterans Health Administration (VHA) include assisting patients with anger management, liver disease, anxiety, asthma, diabetes, HIV, hypertension, weight loss, and smoking cessation.

Methods

The Hematology/Oncology division of the Minneapolis VA Healthcare System (MVAHCS) is a tertiary care facility that administers about 260 new chemotherapy regimens annually. The MVAHCS interdisciplinary hematology/oncology group initiated a quality improvement project to determine the feasibility, acceptability, and experience of tailoring the Annie tool for self-management of cancer symptoms. The group consisted of 2 physicians, 3 advanced practice registered nurses, 1 physician assistant, 2 registered nurses, and 2 Annie program team members.

We first created a symptom management pilot protocol as a result of multidisciplinary team discussions. Examples of discussion points for consideration included, but were not limited to, timing of texts, amount of information to ask for and provide, what potential symptoms to consider, and which patient population to pilot first.

Results

Through screening new patient consults or those referred for chemotherapy education, 15 male veterans enrolled in the symptom monitoring program over an 8 month period. There were additional patients who were not offered the program or chose not to participate; often due to not having texting capabilities on their phone or not liking the texting feature. The majority of those who participated in the program (n = 14) were enrolled at the start of Cycle 1; the other patient was enrolled at the start of Cycle 2. Patients were enrolled an average of 89 days (range 8-204). Average response rate was 84.2% (range 30-100%).

Although symptoms were not reviewed in real time, we reviewed responses to determine the utilization of the instructions given for the program. No veteran had 0 symptoms reported. There were numerous occurrences of a score of 1 or 2. Many of these patients had baseline symptoms due to their underlying cancer. A score of 3 or 4 on the system prompted the patient to call the clinic or go to the ED. Seven patients (some with multiple occurrences) were prompted to call; only 4 of these made the follow-up call to the clinic. All were offered a same day visit, but each declined. Only 1 patient reported a symptom on a day not prompted for that symptom. Symptoms that were reported are listed in order of frequency: fatigue, appetite loss, numbness, pain, mouth sore, and breathing difficulty. There were no visits to the ED.

Program Evaluation

An evaluation was conducted 30 to 60 days after program enrollment. We elicited feedback to determine who was reading and responding to the text message: the patient, a family member, or a caregiver; whether they found the prompts helpful and took action; how they felt about the number of texts; if they felt the program was helpful; and any other feedback that would improve the program. In general, the patients (8) answered the texts independently. In 4 cases, the spouse answered the texts, and 3 patients answered the texts together with their spouses. Most patients (11) found the amount of texting to be “just right.” However, 3 found it to be too many texts and 1 didn’t find the amount of texting to be enough.

Three veterans did not have enough symptoms to feel the program was of benefit to them, but they did feel it would have been helpful if they had been more symptomatic. One veteran recalled taking loperamide as needed, as a result of prompting. No veterans felt as though the texting feature was difficult to use; and overall, were very positive about the program. Several appreciated receiving messages that validated when they were doing well, and they felt empowered by self-management. One of the spouses was a registered nurse and found the information too basic to be of use.

Discussion

Initial evaluation of the program via survey found no technology challenges. Patients have been very positive about the program including ease of use, appreciation of messages that validated when they were doing well, empowerment of self-management, and some utilization of the texting advice for symptom management. Educational hyperlinks for constipation, fatigue, diarrhea, and nausea/vomiting were added after this evaluation, and patients felt that these additions provided a higher level of education.

Staff time for this intervention was minimal. A nurse navigator offered the texting program to the patient during chemotherapy education, along with some instructions, which generally took about 5 minutes. One of the Annie program staff enrolled the patient. From that point forward, this was a self-management tool, beyond checking to ensure that the patient was successful in starting the program and evaluating use for the purposes of this quality improvement project. This self-management tool did not replace any other mechanism that a patient would normally have in our department for seeking help for symptoms. The MVAHSC typical process for symptom management is to have patients call a 24/7 nurse line. If the triage nurse feels the symptoms are related to the patient’s cancer or cancer treatment, they are referred to the physician assistant who is assigned to take those calls and has the option to see the patient the same day. Patients could continue to call the nurse line or speak with providers at the next appointment at their discretion.

Conclusion

Although Annie has the option of using either text messaging or a mobile application, this project only utilized text messaging. The study by Basch and colleagues was the closest randomized trial we could identify to compare to our quality improvement intervention.⁵ The 2 main, distinct differences were that Basch and colleagues utilized online monitoring; and nurses were utilized to screen and intervene on responses, as appropriate.

The ability of our program to text patients without the use of an application or tablet, may enable more patients to participate due to ease of use. There would be no increased in expected workload for clinical staff, and may lead to decreased call burden. Since our program is automated, while still providing patients with the option to call and speak with a staff member as needed, this is a cost-effective, first-line option for symptom management for those experiencing cancer-related symptoms. We believe this text messaging tool can have system wide use and benefit throughout the VHA.

Central nervous system (CNS) lymphoma can be classified into 2 categories: primary CNS lymphoma (PCNSL), which includes disease limited to brain, eyes, spinal cord; and leptomeninges without coexisting or previous systemic lymphoma. Secondary CNS lymphoma (SCNSL) is essentially metastatic disease from a systemic primary site.¹ The focus of this case presentation is PCNSL, with an emphasis on imaging characteristics and differential diagnosis.

The median age at diagnosis for PCNSL is 65 years, and the overall incidence has been decreasing since the mid-1990s, likely related to the increased use of highly-active antiretroviral therapy (HAART) in patients with AIDS.^2,3 Although overall incidence has decreased, incidence in the elderly population has increased.⁴ Historically, PCNSL has been considered an AIDS-defining illness.⁵ These patients, among other immunocompromised patients, such as those on chronic immunosuppressive therapy, are at a higher risk for developing the malignancy.⁶

Clinical presentation varies because of the location of CNS involvement and may present with headache, mood or personality disturbances, or focal neurologic deficits. Seizures are less likely due to the tendency of PCNSL to spare gray matter. Initial workup generally includes a head computed tomography (CT) scan, as well as a contrast-enhanced magnetic resonance image (MRI), which may help direct clinicians to the appropriate diagnosis. However, there is significant overlap between the imaging characteristics of PCNSL and numerous other disease processes, including glioblastoma and demyelination. The imaging characteristics of PCNSL are considerably different depending on the patient’s immune status.⁷

This case illustrates a rare presentation of PCNSL in an immunocompetent patient whose MRI characteristics were seemingly more consistent with those seen in patients with immunodeficiency. The main differential diagnoses and key imaging characteristics, which may help obtain accurate diagnosis, will be discussed.

Case Presentation

A 72-year-old male veteran presented with a 2-month history of subjective weakness in his upper and lower extremities progressing to multiple falls at home. He had no significant medical history other than a thymectomy at age 15 for an enlarged thymus, which per patient report, was benign. An initial laboratory test that included vitamin B₁₂, folate, thyroid-stimulating hormone, complete blood cell count, and comprehensive metabolic panel, were unremarkable, with a white blood cell count of 8.5 K/uL. The initial neurologic evaluation did not show any focal neurologic deficits; however, during the initial hospital stay, the patient developed increasing lower extremity weakness on examination. A noncontrast CT head scan showed extensive nonspecific hypodensities within the periventricular white matter (Figure 1). A contrast-enhanced MRI showed enhancing lesions involving the corpus callosum, left cerebral peduncle, and right temporal lobe (Figures 2, 3, and 4). These lesions also exhibited significant restricted diffusion and a mild amount of surrounding vasogenic edema. The working diagnosis after the MRI included primary CNS lymphoma, multifocal glioblastoma, and tumefactive demyelinating disease. The patient was started on IV steroids and transferred for neurosurgical evaluation and biopsy at an outside hospital. The frontal lesion was biopsied, and the initial frozen section was consistent with lymphoma; a bone marrow biopsy was negative. The workup for immunodeficiency was unremarkable. Pathology revealed high-grade B-cell lymphoma, and the patient began a chemotherapy regimen.

Discussion

The workup of altered mental status, focal neurologic deficits, headaches, or other neurologic conditions often begins with a noncontrast CT scan. On CT, PCNSL generally appears isodense to hyperdense to gray matter, but appearance is variable. The often hyperdense appearance is attributable to the hypercellular nature of lymphoma. Many times, as in this case, CT may show only vague hypodensities, some of which may be associated with surrounding edema. This presentation is nonspecific and may be seen with advancing age due to changes of chronic microvascular ischemia as well as demyelination, other malignancies, and several other disease processes, both benign and malignant. After the initial CT scan, further workup requires evaluation with MRI. PCNSL exhibits restricted diffusion and variable signal intensity on T2-weighted imaging.

PCNSL is frequently centrally located within the periventricular white matter, often within the frontal lobe but can involve other lobes, the basal ganglia, brainstem, cerebellum, or less likely, the spinal canal.⁷ Contrary to primary CNS disease, secondary lymphoma within the CNS has been described classically as affecting a leptomeningeal (pia and arachnoid mater) distribution two-thirds of the time, with parenchymal involvement occurring in the other one-third of patients. A recent study by Malikova and colleagues found parenchymal involvement may be much more common than previously thought.¹ Leptomeningeal spread of disease often involves the cranial nerves, subependymal regions, spinal cord, or spinal nerve roots. Dural involvement in primary or secondary lymphoma is rare.

PCNSL nearly always shows enhancement. Linear enhancement along perivascular spaces is highly characteristic of PCNSL. The typical appearance of PCNSL associated with immunodeficiency varies from that seen in an otherwise immunocompetent patient. Patients with immunodeficiency usually have multifocal involvement, central necrosis leading to a ring enhancement appearance, and have more propensity for spontaneous hemorrhage.⁷ Immunocompetent patients are less likely to present with multifocal disease and rarely show ring enhancement. Also, spontaneous hemorrhage is rare in immunocompetent patients. In our case, extensive multifocal involvement was present, whereas typically immunocompetent patients will present with a solitary homogeneously enhancing parenchymal mass.

The primary differential for PCNSL includes malignant glioma, tumefactive multiple sclerosis, metastatic disease, and in an immunocompromised patient, toxoplasmosis. The degree of associated vasogenic edema and mass effect is generally lower in PCNSL than that of malignant gliomas and metastasis. Also, PCNSL tends to spare the cerebral cortex.⁸

Classically, PCNSL, malignant gliomas, and demyelinating disease have been considered the main differential for lesions that cross midline and involve both cerebral hemispheres. Lymphoma generally exhibits more restricted diffusion than malignant gliomas and metastasis, attributable to the highly cellular nature of lymphoma.⁷ Tumefactive multiple sclerosis is associated with relatively minimal mass effect for lesion size and exhibits less restricted diffusion values when compared to high grade gliomas and PCNSL. One fairly specific finding for tumefactive demyelinating lesions is incomplete rim enhancement.⁹ Unfortunately, an MRI is not reliable in differentiating these entities, and biopsy is required for definitive diagnosis. Many advancing imaging modalities may help provide the correct diagnosis of PCNSL, including diffusion-weighted and apparent diffusion coefficient imaging, diffusion tensor imaging, MR spectroscopy and PET imaging.⁷

Conclusion

With the increasing use of HAART, the paradigm of PCNSL is shifting toward one predominantly affecting immunocompetent patients. PCNSL should be considered in any patient with multiple enhancing CNS lesions, regardless of immune status. Several key imaging characteristics may help differentiate PCNSL and other disease processes; however, at this time, biopsy is recommended for definitive diagnosis.

Central nervous system (CNS) lymphoma can be classified into 2 categories: primary CNS lymphoma (PCNSL), which includes disease limited to brain, eyes, spinal cord; and leptomeninges without coexisting or previous systemic lymphoma. Secondary CNS lymphoma (SCNSL) is essentially metastatic disease from a systemic primary site.¹ The focus of this case presentation is PCNSL, with an emphasis on imaging characteristics and differential diagnosis.

The median age at diagnosis for PCNSL is 65 years, and the overall incidence has been decreasing since the mid-1990s, likely related to the increased use of highly-active antiretroviral therapy (HAART) in patients with AIDS.^2,3 Although overall incidence has decreased, incidence in the elderly population has increased.⁴ Historically, PCNSL has been considered an AIDS-defining illness.⁵ These patients, among other immunocompromised patients, such as those on chronic immunosuppressive therapy, are at a higher risk for developing the malignancy.⁶

Clinical presentation varies because of the location of CNS involvement and may present with headache, mood or personality disturbances, or focal neurologic deficits. Seizures are less likely due to the tendency of PCNSL to spare gray matter. Initial workup generally includes a head computed tomography (CT) scan, as well as a contrast-enhanced magnetic resonance image (MRI), which may help direct clinicians to the appropriate diagnosis. However, there is significant overlap between the imaging characteristics of PCNSL and numerous other disease processes, including glioblastoma and demyelination. The imaging characteristics of PCNSL are considerably different depending on the patient’s immune status.⁷

This case illustrates a rare presentation of PCNSL in an immunocompetent patient whose MRI characteristics were seemingly more consistent with those seen in patients with immunodeficiency. The main differential diagnoses and key imaging characteristics, which may help obtain accurate diagnosis, will be discussed.

Case Presentation

A 72-year-old male veteran presented with a 2-month history of subjective weakness in his upper and lower extremities progressing to multiple falls at home. He had no significant medical history other than a thymectomy at age 15 for an enlarged thymus, which per patient report, was benign. An initial laboratory test that included vitamin B₁₂, folate, thyroid-stimulating hormone, complete blood cell count, and comprehensive metabolic panel, were unremarkable, with a white blood cell count of 8.5 K/uL. The initial neurologic evaluation did not show any focal neurologic deficits; however, during the initial hospital stay, the patient developed increasing lower extremity weakness on examination. A noncontrast CT head scan showed extensive nonspecific hypodensities within the periventricular white matter (Figure 1). A contrast-enhanced MRI showed enhancing lesions involving the corpus callosum, left cerebral peduncle, and right temporal lobe (Figures 2, 3, and 4). These lesions also exhibited significant restricted diffusion and a mild amount of surrounding vasogenic edema. The working diagnosis after the MRI included primary CNS lymphoma, multifocal glioblastoma, and tumefactive demyelinating disease. The patient was started on IV steroids and transferred for neurosurgical evaluation and biopsy at an outside hospital. The frontal lesion was biopsied, and the initial frozen section was consistent with lymphoma; a bone marrow biopsy was negative. The workup for immunodeficiency was unremarkable. Pathology revealed high-grade B-cell lymphoma, and the patient began a chemotherapy regimen.

Discussion

The workup of altered mental status, focal neurologic deficits, headaches, or other neurologic conditions often begins with a noncontrast CT scan. On CT, PCNSL generally appears isodense to hyperdense to gray matter, but appearance is variable. The often hyperdense appearance is attributable to the hypercellular nature of lymphoma. Many times, as in this case, CT may show only vague hypodensities, some of which may be associated with surrounding edema. This presentation is nonspecific and may be seen with advancing age due to changes of chronic microvascular ischemia as well as demyelination, other malignancies, and several other disease processes, both benign and malignant. After the initial CT scan, further workup requires evaluation with MRI. PCNSL exhibits restricted diffusion and variable signal intensity on T2-weighted imaging.

PCNSL is frequently centrally located within the periventricular white matter, often within the frontal lobe but can involve other lobes, the basal ganglia, brainstem, cerebellum, or less likely, the spinal canal.⁷ Contrary to primary CNS disease, secondary lymphoma within the CNS has been described classically as affecting a leptomeningeal (pia and arachnoid mater) distribution two-thirds of the time, with parenchymal involvement occurring in the other one-third of patients. A recent study by Malikova and colleagues found parenchymal involvement may be much more common than previously thought.¹ Leptomeningeal spread of disease often involves the cranial nerves, subependymal regions, spinal cord, or spinal nerve roots. Dural involvement in primary or secondary lymphoma is rare.

PCNSL nearly always shows enhancement. Linear enhancement along perivascular spaces is highly characteristic of PCNSL. The typical appearance of PCNSL associated with immunodeficiency varies from that seen in an otherwise immunocompetent patient. Patients with immunodeficiency usually have multifocal involvement, central necrosis leading to a ring enhancement appearance, and have more propensity for spontaneous hemorrhage.⁷ Immunocompetent patients are less likely to present with multifocal disease and rarely show ring enhancement. Also, spontaneous hemorrhage is rare in immunocompetent patients. In our case, extensive multifocal involvement was present, whereas typically immunocompetent patients will present with a solitary homogeneously enhancing parenchymal mass.

The primary differential for PCNSL includes malignant glioma, tumefactive multiple sclerosis, metastatic disease, and in an immunocompromised patient, toxoplasmosis. The degree of associated vasogenic edema and mass effect is generally lower in PCNSL than that of malignant gliomas and metastasis. Also, PCNSL tends to spare the cerebral cortex.⁸

Classically, PCNSL, malignant gliomas, and demyelinating disease have been considered the main differential for lesions that cross midline and involve both cerebral hemispheres. Lymphoma generally exhibits more restricted diffusion than malignant gliomas and metastasis, attributable to the highly cellular nature of lymphoma.⁷ Tumefactive multiple sclerosis is associated with relatively minimal mass effect for lesion size and exhibits less restricted diffusion values when compared to high grade gliomas and PCNSL. One fairly specific finding for tumefactive demyelinating lesions is incomplete rim enhancement.⁹ Unfortunately, an MRI is not reliable in differentiating these entities, and biopsy is required for definitive diagnosis. Many advancing imaging modalities may help provide the correct diagnosis of PCNSL, including diffusion-weighted and apparent diffusion coefficient imaging, diffusion tensor imaging, MR spectroscopy and PET imaging.⁷

Conclusion

With the increasing use of HAART, the paradigm of PCNSL is shifting toward one predominantly affecting immunocompetent patients. PCNSL should be considered in any patient with multiple enhancing CNS lesions, regardless of immune status. Several key imaging characteristics may help differentiate PCNSL and other disease processes; however, at this time, biopsy is recommended for definitive diagnosis.

Central nervous system (CNS) lymphoma can be classified into 2 categories: primary CNS lymphoma (PCNSL), which includes disease limited to brain, eyes, spinal cord; and leptomeninges without coexisting or previous systemic lymphoma. Secondary CNS lymphoma (SCNSL) is essentially metastatic disease from a systemic primary site.¹ The focus of this case presentation is PCNSL, with an emphasis on imaging characteristics and differential diagnosis.

The median age at diagnosis for PCNSL is 65 years, and the overall incidence has been decreasing since the mid-1990s, likely related to the increased use of highly-active antiretroviral therapy (HAART) in patients with AIDS.^2,3 Although overall incidence has decreased, incidence in the elderly population has increased.⁴ Historically, PCNSL has been considered an AIDS-defining illness.⁵ These patients, among other immunocompromised patients, such as those on chronic immunosuppressive therapy, are at a higher risk for developing the malignancy.⁶

Clinical presentation varies because of the location of CNS involvement and may present with headache, mood or personality disturbances, or focal neurologic deficits. Seizures are less likely due to the tendency of PCNSL to spare gray matter. Initial workup generally includes a head computed tomography (CT) scan, as well as a contrast-enhanced magnetic resonance image (MRI), which may help direct clinicians to the appropriate diagnosis. However, there is significant overlap between the imaging characteristics of PCNSL and numerous other disease processes, including glioblastoma and demyelination. The imaging characteristics of PCNSL are considerably different depending on the patient’s immune status.⁷

This case illustrates a rare presentation of PCNSL in an immunocompetent patient whose MRI characteristics were seemingly more consistent with those seen in patients with immunodeficiency. The main differential diagnoses and key imaging characteristics, which may help obtain accurate diagnosis, will be discussed.

Case Presentation

A 72-year-old male veteran presented with a 2-month history of subjective weakness in his upper and lower extremities progressing to multiple falls at home. He had no significant medical history other than a thymectomy at age 15 for an enlarged thymus, which per patient report, was benign. An initial laboratory test that included vitamin B₁₂, folate, thyroid-stimulating hormone, complete blood cell count, and comprehensive metabolic panel, were unremarkable, with a white blood cell count of 8.5 K/uL. The initial neurologic evaluation did not show any focal neurologic deficits; however, during the initial hospital stay, the patient developed increasing lower extremity weakness on examination. A noncontrast CT head scan showed extensive nonspecific hypodensities within the periventricular white matter (Figure 1). A contrast-enhanced MRI showed enhancing lesions involving the corpus callosum, left cerebral peduncle, and right temporal lobe (Figures 2, 3, and 4). These lesions also exhibited significant restricted diffusion and a mild amount of surrounding vasogenic edema. The working diagnosis after the MRI included primary CNS lymphoma, multifocal glioblastoma, and tumefactive demyelinating disease. The patient was started on IV steroids and transferred for neurosurgical evaluation and biopsy at an outside hospital. The frontal lesion was biopsied, and the initial frozen section was consistent with lymphoma; a bone marrow biopsy was negative. The workup for immunodeficiency was unremarkable. Pathology revealed high-grade B-cell lymphoma, and the patient began a chemotherapy regimen.

Discussion

The workup of altered mental status, focal neurologic deficits, headaches, or other neurologic conditions often begins with a noncontrast CT scan. On CT, PCNSL generally appears isodense to hyperdense to gray matter, but appearance is variable. The often hyperdense appearance is attributable to the hypercellular nature of lymphoma. Many times, as in this case, CT may show only vague hypodensities, some of which may be associated with surrounding edema. This presentation is nonspecific and may be seen with advancing age due to changes of chronic microvascular ischemia as well as demyelination, other malignancies, and several other disease processes, both benign and malignant. After the initial CT scan, further workup requires evaluation with MRI. PCNSL exhibits restricted diffusion and variable signal intensity on T2-weighted imaging.

PCNSL is frequently centrally located within the periventricular white matter, often within the frontal lobe but can involve other lobes, the basal ganglia, brainstem, cerebellum, or less likely, the spinal canal.⁷ Contrary to primary CNS disease, secondary lymphoma within the CNS has been described classically as affecting a leptomeningeal (pia and arachnoid mater) distribution two-thirds of the time, with parenchymal involvement occurring in the other one-third of patients. A recent study by Malikova and colleagues found parenchymal involvement may be much more common than previously thought.¹ Leptomeningeal spread of disease often involves the cranial nerves, subependymal regions, spinal cord, or spinal nerve roots. Dural involvement in primary or secondary lymphoma is rare.

PCNSL nearly always shows enhancement. Linear enhancement along perivascular spaces is highly characteristic of PCNSL. The typical appearance of PCNSL associated with immunodeficiency varies from that seen in an otherwise immunocompetent patient. Patients with immunodeficiency usually have multifocal involvement, central necrosis leading to a ring enhancement appearance, and have more propensity for spontaneous hemorrhage.⁷ Immunocompetent patients are less likely to present with multifocal disease and rarely show ring enhancement. Also, spontaneous hemorrhage is rare in immunocompetent patients. In our case, extensive multifocal involvement was present, whereas typically immunocompetent patients will present with a solitary homogeneously enhancing parenchymal mass.

The primary differential for PCNSL includes malignant glioma, tumefactive multiple sclerosis, metastatic disease, and in an immunocompromised patient, toxoplasmosis. The degree of associated vasogenic edema and mass effect is generally lower in PCNSL than that of malignant gliomas and metastasis. Also, PCNSL tends to spare the cerebral cortex.⁸

Classically, PCNSL, malignant gliomas, and demyelinating disease have been considered the main differential for lesions that cross midline and involve both cerebral hemispheres. Lymphoma generally exhibits more restricted diffusion than malignant gliomas and metastasis, attributable to the highly cellular nature of lymphoma.⁷ Tumefactive multiple sclerosis is associated with relatively minimal mass effect for lesion size and exhibits less restricted diffusion values when compared to high grade gliomas and PCNSL. One fairly specific finding for tumefactive demyelinating lesions is incomplete rim enhancement.⁹ Unfortunately, an MRI is not reliable in differentiating these entities, and biopsy is required for definitive diagnosis. Many advancing imaging modalities may help provide the correct diagnosis of PCNSL, including diffusion-weighted and apparent diffusion coefficient imaging, diffusion tensor imaging, MR spectroscopy and PET imaging.⁷

Conclusion

With the increasing use of HAART, the paradigm of PCNSL is shifting toward one predominantly affecting immunocompetent patients. PCNSL should be considered in any patient with multiple enhancing CNS lesions, regardless of immune status. Several key imaging characteristics may help differentiate PCNSL and other disease processes; however, at this time, biopsy is recommended for definitive diagnosis.

Colorectal cancer is the second most common cause of cancer death in the US, with one-third of all colorectal cancers occurring within the rectum. Each year, an estimated 40000 Americans are diagnosed with rectal cancer (RC).^1,2 The prognosis and treatment of RC depends on both T and N stage at the time of diagnosis.^3-5 According to the most recent National Comprehensive Cancer Network guidelines from May 2019, patients with T1 to T2N0 tumors should undergo transanal or transabdominal surgery upfront, whereas patients with T3 to T4N0 or any TN1 to 2 should start with neoadjuvant therapy for better locoregional control, followed by surgery.⁶ Therefore, the appropriate management of RC requires adequate staging.

Endoscopic ultrasound (EUS), magnetic resonance imaging (MRI), and computed tomography (CT) are the imaging techniques currently used to stage RC. In a meta-analysis of 90 articles published between 1985 and 2002 that compared the 3 radiologic modalities, Bipat and colleagues found that MRI and EUS had a similar sensitivity of 94%, whereas the specificity of EUS (86%) was significantly higher than that of MRI (69%) for muscularis propria invasion.⁷ CT was performed only in a limited number of trials because CT was considered inadequate to assess early T stage. For perirectal tissue invasion, the sensitivity of EUS was statistically higher than that of CT and MRI imaging: 90% compared with 79% and 82%, respectively. The specificity estimates for EUS, CT, and MRI were comparable: 75%, 78%, and 76%, respectively. The respective sensitivity and specificity of the 3 imaging modalities to evaluate lymph nodes were also comparable: EUS, 67% and 78%; CT, 55% and 74%; and MRI, 66% and 76%.

The role of EUS in the diagnosis and treatment of RC has long been validated.^1,2-5 A meta-analysis of 42 studies involving 5039 patients found EUS to be highly accurate for differentiating various T stages.⁸ However, EUS cannot assess iliac and mesenteric lymph nodes or posterior tumor extension beyond endopelvic fascia in advanced RC. Notable heterogeneity was found among the studies in the meta-analyses with regard to the type of equipment used for staging, as well as the criteria used to assess the depth of penetration and nodal status. The recent introduction of phased-array coils and the development of T2-weighted fast spin sequences have improved the resolution of MRI. The MERCURY trial showed that extension of tumor to within 1 mm of the circumferential margin on high-resolution MRI correctly predicted margin involvement at the time of surgery in 92% of the patients.⁹ In the retrospective study by Balyasnikova and colleagues, MRI was found to correctly identify partial submucosal invasion and suitability for local excision in 89% of the cases.¹⁰

Therefore, both EUS and MRI are useful, more so than CT, in assessment of the depth of tumor invasion, nodal staging, and predicting the circumferential resection margin. The use of EUS, however, does not preclude the use of MRI, or vice versa. Rather, the 2 modalities can complement each other in staging and proper patient selection for treatment.¹¹

Despite data supporting the value of EUS in staging RC, its use is limited by a high degree of operator dependence and a substantial learning curve,^12-17 which may explain the low EUS accuracy observed in some reports.^7,13,15 Given the presence of recognized alternatives such as MRI, we decided to reevaluate EUS accuracy for the staging of RC outside high-volume specialized centers and prospective clinical trials.

Methods

A retrospective chart review was performed that included all consecutive patients undergoing rectal ultrasound from January 2011 to August 2015 at the US Department of Veterans Affairs Medical Center (VAMC) in Memphis, Tennessee. Sixty-five patients with short-stocked or sessile lesions < 15 cm from anal margin staged T2N0M0 or lower by endorectal ultrasound (ERUS) were included. The patients with neoplasms staged in excess of T2 or N0 were excluded from the study because treatment protocol dictates immediate neoadjuvant treatment, the administration of which would affect subsequent histopathology.

For the 37 patients included in the final analysis, ERUS results were compared with surgical pathology to ascertain accuracy. The resections were performed endoscopically or surgically with a goal of obtaining clear margins. The choice of procedure depended on size, shape, location, and depth of invasion. All patients underwent clinical and endoscopic surveillance with flexible sigmoidoscopy/EUS every 3 to 6 months for the first 2 years. We used 2 different gold standards for surveillance depending on the type of procedure performed to remove the lesion. A pathology report was the gold standard used for patients who underwent surgery. In patients who underwent endoscopic resection, we used the lack of recurrent disease, determined by normal endoscopic and endoscopic ultrasound examination, to signify complete endoscopic resection and therefore adequate staging as an early neoplasm.

Results

From January 2011 to August 2015, 65 rectal ultrasounds were performed. All EUS procedures were performed by 1 physician (C Ruben Tombazzi). All patients had previous endoscopic evaluation and tissue diagnoses. Twenty-eight patients were excluded: 18 had T3 or N1 disease, 2 had T2N0 but refused surgery, 2 had anal cancer, 3 patients with suspected cancer had benign nonneoplastic disease (2 radiation proctitis, 1 normal rectal wall), and 3 underwent EUS for benign tumors (1 ganglioneuroma and 2 lipomas).

Thirty-seven patients were included in the study, 3 of whom were staged as T2N0 and 34 as T1N0 or lower by EUS. All patients were men ranging in age from 43 to 73 years (mean, 59 years). All 37 patients underwent endoscopic or surgical resection of their early rectal neoplasm. The final pathologic evaluation of the specimens demonstrated 14 carcinoid tumors, 11 adenocarcinomas, 6 tubular adenomas with high-grade dysplasia, and 6 benign adenomas. The preoperative EUS staging was confirmed for all patients, with 100% sensitivity, specificity, and accuracy. None of the patients who underwent endoscopic or surgical transanal resection had recurrence, determined by normal endoscopic and endoscopic ultrasound appearance, during a mean of 32.6 months surveillance.

Discussion

EUS has long been a recognized method for T and N staging of RC.^1,3-5,7,8 Our data confirm that, in experienced hands, EUS is highly accurate in the staging of early rectal cancers.

The impact of EUS on the management of RC was demonstrated in a Mayo Clinic prospective blinded study.¹ In that cohort of 80 consecutive patients who had previously had a CT for staging, EUS altered patient management in about 30% of cases. The most common change precipatated by EUS was the indication for additional neoadjuvant treatment.

However, the results have not been as encouraging when ERUS is performed outside of strict research protocol. A multicenter, prospective, country-wide quality assurance study from > 300 German hospitals was designed to assess the diagnostic accuracy of EUS in RC.¹³ Of 29206 patients, 7096 underwent surgery, without neoadjuvant treatment, and were included in the final analysis. The correspondence of tumor invasion with histopathology was 64.7%, with understaging of 18% and overstaging of 17.3%.¹³ These numbers were better in hospitals with greater experience performing ERUS: 73% accuracy in the centers with a case load of > 30 cases per year compared with 63.2% accuracy for the centers with < 10 cases a year. Marusch and colleagues had previously demonstrated an EUS accuracy of 63.3% in a study of 1463 patients with RC in Germany.¹⁴ Another study based out of the UK had similar findings. Ashraf and colleagues performed a database analyses from 20 UK centers and identified 165 patients with RC who underwent ERUS and endoscopic microsurgery.¹⁵ Compared with histopathology, EUS had 57.1% sensitivity, 73% specificity, and 42.9% accuracy for T1 cancers; EUS accuracy was 50% for T2 and 58% for T3 tumors. The authors concluded that the general accuracy of EUS in determining stage was around 50%, the statistical equivalent of flipping a coin.

The low accuracy of EUS observed by German and British multicenter studies^13-15 was attributed to the difference that may exist in clinical trials at specialized centers compared with wider use of EUS in a community setting. As seen by our data, the Memphis VAMC is not a high-volume center for the treatment of RC. However, all our EUS procedures were performed and interpreted by a single operator (C. Ruben Tombazzi) with 18 years of EUS experience. We cannot conclude that no patient was overstaged, as patients receiving a stage of T3N0 or T > N0 received neoadjuvant treatment and were not included. However, we can conclude that no patient was understaged. All patients deemed to be T1 to T2N0 included in our study received accurate staging. Our results are consistent with the high accuracy of EUS reported from other centers with experience in diagnosis and treatment of RC.^1,3-5,17,18

Although EUS is accurate in differentiating T1 from T2 tumors, it cannot reliably differentiate T1 from T0 lesions. In one study, 57.6% of adenomas and 30.7% of carcinomas in situ were staged as T1 on EUS, while almost half of T1 cancers were interpreted as T0.¹⁷ This drawback is a well-known limitation of EUS; although, the misinterpretation does not affect treatment, as both T0 and T1 lesions can be treated successfully by local excision alone, which was the algorithm used for our patients. The choice of the specific procedure for local excision was left to the clinicians and included transanal endoscopic or surgical resections. At a mean follow-up of 32.6 months, none of the 37 patients who underwent endoscopic or surgical transanal resection had evidence of recurrent disease.

A limitation of EUS, or any other imaging modality, is differentiating tumor invasion from peritumoral inflammation. The inflammation can render images of tumor borders ill-defined and irregular, which hinders precise staging. However, the accurate identification of tumors with deep involvement of the submucosa (T1sm3) is of importance, because these tumors are more advanced than the superficial and intermediate T1 lesions (T1sm1 and T1sm2, respectively).

Patients with RC whose lesions are considered T1sm3 are at higher risk of harboring lymph node metastases.¹⁸ Nascimbeni and colleagues had shown that the invasion into the lower third of the submucosa (sm3) was an independent risk factor for lower cancer-free survival among patients with T1 RC.¹⁹We did not measure the distance of the tumor to muscular layer in our study, but we relied on EUS to predict the circumferential tumor margins and guide the surgical resection. Of the 11 patients with T1 rectal adenocarcinomas and the 6 patients with tubular adenoma with high-grade dysplasia, all treated by local excision, none developed a local or distant recurrence during follow-up.

Unlike rectal adenocarcinomas, the prognosis for carcinoid tumors correlates not only with the depth of invasion but also with the size of the tumor. The other adverse prognostic features include poor differentiation, high mitosis index, and lymphovascular invasion.²⁰

EUS had been shown to be highly accurate in determining the precise carcinoid tumor size, depth of invasion, and lymph node metastases.^20,21 In a study of 66 resected rectal carcinoid tumors by Ishii and colleagues, 57 lesions had a diameter of ≤ 10 mm and 9 lesions had a diameter of > 10 mm.²¹ All of the 57 carcinoid tumors with a diameter of ≤ 10 mm were confined to the submucosa. In contrast, 5 of the 9 lesions > 10 mm invaded the muscularis propria, 6 had a lymphovascular invasion, 4 were lymph node metastases, and 1 was a liver metastasis.

In our series, 4 of the 14 carcinoid tumors were > 10 mm but none were > 20 mm. None of the carcinoids with a diameter ≤ 10 mm invaded the muscularis propria. Of the 4 carcinoids > 10 mm, 1 was T2N0 and 3 were T1N0. All carcinoid tumors in our series were low grade and with low proliferation indexes, and all were treated successfully by local excision.

Conclusion

We believe our study shows that EUS can be highly accurate in staging rectal lesions, specifically lesions that are T1-T2N0, be they adenocarcinoma or carcinoid. Although we could not assess overstaging for lesions that were staged > T2 or > N0, we were able to determine no understaging for all of our patients. In experienced hands, EUS remains a highly accurate staging tool for early rectal carcinoma.

Colorectal cancer is the second most common cause of cancer death in the US, with one-third of all colorectal cancers occurring within the rectum. Each year, an estimated 40000 Americans are diagnosed with rectal cancer (RC).^1,2 The prognosis and treatment of RC depends on both T and N stage at the time of diagnosis.^3-5 According to the most recent National Comprehensive Cancer Network guidelines from May 2019, patients with T1 to T2N0 tumors should undergo transanal or transabdominal surgery upfront, whereas patients with T3 to T4N0 or any TN1 to 2 should start with neoadjuvant therapy for better locoregional control, followed by surgery.⁶ Therefore, the appropriate management of RC requires adequate staging.

Endoscopic ultrasound (EUS), magnetic resonance imaging (MRI), and computed tomography (CT) are the imaging techniques currently used to stage RC. In a meta-analysis of 90 articles published between 1985 and 2002 that compared the 3 radiologic modalities, Bipat and colleagues found that MRI and EUS had a similar sensitivity of 94%, whereas the specificity of EUS (86%) was significantly higher than that of MRI (69%) for muscularis propria invasion.⁷ CT was performed only in a limited number of trials because CT was considered inadequate to assess early T stage. For perirectal tissue invasion, the sensitivity of EUS was statistically higher than that of CT and MRI imaging: 90% compared with 79% and 82%, respectively. The specificity estimates for EUS, CT, and MRI were comparable: 75%, 78%, and 76%, respectively. The respective sensitivity and specificity of the 3 imaging modalities to evaluate lymph nodes were also comparable: EUS, 67% and 78%; CT, 55% and 74%; and MRI, 66% and 76%.

The role of EUS in the diagnosis and treatment of RC has long been validated.^1,2-5 A meta-analysis of 42 studies involving 5039 patients found EUS to be highly accurate for differentiating various T stages.⁸ However, EUS cannot assess iliac and mesenteric lymph nodes or posterior tumor extension beyond endopelvic fascia in advanced RC. Notable heterogeneity was found among the studies in the meta-analyses with regard to the type of equipment used for staging, as well as the criteria used to assess the depth of penetration and nodal status. The recent introduction of phased-array coils and the development of T2-weighted fast spin sequences have improved the resolution of MRI. The MERCURY trial showed that extension of tumor to within 1 mm of the circumferential margin on high-resolution MRI correctly predicted margin involvement at the time of surgery in 92% of the patients.⁹ In the retrospective study by Balyasnikova and colleagues, MRI was found to correctly identify partial submucosal invasion and suitability for local excision in 89% of the cases.¹⁰

Therefore, both EUS and MRI are useful, more so than CT, in assessment of the depth of tumor invasion, nodal staging, and predicting the circumferential resection margin. The use of EUS, however, does not preclude the use of MRI, or vice versa. Rather, the 2 modalities can complement each other in staging and proper patient selection for treatment.¹¹

Despite data supporting the value of EUS in staging RC, its use is limited by a high degree of operator dependence and a substantial learning curve,^12-17 which may explain the low EUS accuracy observed in some reports.^7,13,15 Given the presence of recognized alternatives such as MRI, we decided to reevaluate EUS accuracy for the staging of RC outside high-volume specialized centers and prospective clinical trials.

Methods

A retrospective chart review was performed that included all consecutive patients undergoing rectal ultrasound from January 2011 to August 2015 at the US Department of Veterans Affairs Medical Center (VAMC) in Memphis, Tennessee. Sixty-five patients with short-stocked or sessile lesions < 15 cm from anal margin staged T2N0M0 or lower by endorectal ultrasound (ERUS) were included. The patients with neoplasms staged in excess of T2 or N0 were excluded from the study because treatment protocol dictates immediate neoadjuvant treatment, the administration of which would affect subsequent histopathology.

For the 37 patients included in the final analysis, ERUS results were compared with surgical pathology to ascertain accuracy. The resections were performed endoscopically or surgically with a goal of obtaining clear margins. The choice of procedure depended on size, shape, location, and depth of invasion. All patients underwent clinical and endoscopic surveillance with flexible sigmoidoscopy/EUS every 3 to 6 months for the first 2 years. We used 2 different gold standards for surveillance depending on the type of procedure performed to remove the lesion. A pathology report was the gold standard used for patients who underwent surgery. In patients who underwent endoscopic resection, we used the lack of recurrent disease, determined by normal endoscopic and endoscopic ultrasound examination, to signify complete endoscopic resection and therefore adequate staging as an early neoplasm.

Results

From January 2011 to August 2015, 65 rectal ultrasounds were performed. All EUS procedures were performed by 1 physician (C Ruben Tombazzi). All patients had previous endoscopic evaluation and tissue diagnoses. Twenty-eight patients were excluded: 18 had T3 or N1 disease, 2 had T2N0 but refused surgery, 2 had anal cancer, 3 patients with suspected cancer had benign nonneoplastic disease (2 radiation proctitis, 1 normal rectal wall), and 3 underwent EUS for benign tumors (1 ganglioneuroma and 2 lipomas).

Thirty-seven patients were included in the study, 3 of whom were staged as T2N0 and 34 as T1N0 or lower by EUS. All patients were men ranging in age from 43 to 73 years (mean, 59 years). All 37 patients underwent endoscopic or surgical resection of their early rectal neoplasm. The final pathologic evaluation of the specimens demonstrated 14 carcinoid tumors, 11 adenocarcinomas, 6 tubular adenomas with high-grade dysplasia, and 6 benign adenomas. The preoperative EUS staging was confirmed for all patients, with 100% sensitivity, specificity, and accuracy. None of the patients who underwent endoscopic or surgical transanal resection had recurrence, determined by normal endoscopic and endoscopic ultrasound appearance, during a mean of 32.6 months surveillance.

Discussion

EUS has long been a recognized method for T and N staging of RC.^1,3-5,7,8 Our data confirm that, in experienced hands, EUS is highly accurate in the staging of early rectal cancers.

The impact of EUS on the management of RC was demonstrated in a Mayo Clinic prospective blinded study.¹ In that cohort of 80 consecutive patients who had previously had a CT for staging, EUS altered patient management in about 30% of cases. The most common change precipatated by EUS was the indication for additional neoadjuvant treatment.

However, the results have not been as encouraging when ERUS is performed outside of strict research protocol. A multicenter, prospective, country-wide quality assurance study from > 300 German hospitals was designed to assess the diagnostic accuracy of EUS in RC.¹³ Of 29206 patients, 7096 underwent surgery, without neoadjuvant treatment, and were included in the final analysis. The correspondence of tumor invasion with histopathology was 64.7%, with understaging of 18% and overstaging of 17.3%.¹³ These numbers were better in hospitals with greater experience performing ERUS: 73% accuracy in the centers with a case load of > 30 cases per year compared with 63.2% accuracy for the centers with < 10 cases a year. Marusch and colleagues had previously demonstrated an EUS accuracy of 63.3% in a study of 1463 patients with RC in Germany.¹⁴ Another study based out of the UK had similar findings. Ashraf and colleagues performed a database analyses from 20 UK centers and identified 165 patients with RC who underwent ERUS and endoscopic microsurgery.¹⁵ Compared with histopathology, EUS had 57.1% sensitivity, 73% specificity, and 42.9% accuracy for T1 cancers; EUS accuracy was 50% for T2 and 58% for T3 tumors. The authors concluded that the general accuracy of EUS in determining stage was around 50%, the statistical equivalent of flipping a coin.

The low accuracy of EUS observed by German and British multicenter studies^13-15 was attributed to the difference that may exist in clinical trials at specialized centers compared with wider use of EUS in a community setting. As seen by our data, the Memphis VAMC is not a high-volume center for the treatment of RC. However, all our EUS procedures were performed and interpreted by a single operator (C. Ruben Tombazzi) with 18 years of EUS experience. We cannot conclude that no patient was overstaged, as patients receiving a stage of T3N0 or T > N0 received neoadjuvant treatment and were not included. However, we can conclude that no patient was understaged. All patients deemed to be T1 to T2N0 included in our study received accurate staging. Our results are consistent with the high accuracy of EUS reported from other centers with experience in diagnosis and treatment of RC.^1,3-5,17,18

Although EUS is accurate in differentiating T1 from T2 tumors, it cannot reliably differentiate T1 from T0 lesions. In one study, 57.6% of adenomas and 30.7% of carcinomas in situ were staged as T1 on EUS, while almost half of T1 cancers were interpreted as T0.¹⁷ This drawback is a well-known limitation of EUS; although, the misinterpretation does not affect treatment, as both T0 and T1 lesions can be treated successfully by local excision alone, which was the algorithm used for our patients. The choice of the specific procedure for local excision was left to the clinicians and included transanal endoscopic or surgical resections. At a mean follow-up of 32.6 months, none of the 37 patients who underwent endoscopic or surgical transanal resection had evidence of recurrent disease.

A limitation of EUS, or any other imaging modality, is differentiating tumor invasion from peritumoral inflammation. The inflammation can render images of tumor borders ill-defined and irregular, which hinders precise staging. However, the accurate identification of tumors with deep involvement of the submucosa (T1sm3) is of importance, because these tumors are more advanced than the superficial and intermediate T1 lesions (T1sm1 and T1sm2, respectively).

Patients with RC whose lesions are considered T1sm3 are at higher risk of harboring lymph node metastases.¹⁸ Nascimbeni and colleagues had shown that the invasion into the lower third of the submucosa (sm3) was an independent risk factor for lower cancer-free survival among patients with T1 RC.¹⁹We did not measure the distance of the tumor to muscular layer in our study, but we relied on EUS to predict the circumferential tumor margins and guide the surgical resection. Of the 11 patients with T1 rectal adenocarcinomas and the 6 patients with tubular adenoma with high-grade dysplasia, all treated by local excision, none developed a local or distant recurrence during follow-up.

Unlike rectal adenocarcinomas, the prognosis for carcinoid tumors correlates not only with the depth of invasion but also with the size of the tumor. The other adverse prognostic features include poor differentiation, high mitosis index, and lymphovascular invasion.²⁰

EUS had been shown to be highly accurate in determining the precise carcinoid tumor size, depth of invasion, and lymph node metastases.^20,21 In a study of 66 resected rectal carcinoid tumors by Ishii and colleagues, 57 lesions had a diameter of ≤ 10 mm and 9 lesions had a diameter of > 10 mm.²¹ All of the 57 carcinoid tumors with a diameter of ≤ 10 mm were confined to the submucosa. In contrast, 5 of the 9 lesions > 10 mm invaded the muscularis propria, 6 had a lymphovascular invasion, 4 were lymph node metastases, and 1 was a liver metastasis.

In our series, 4 of the 14 carcinoid tumors were > 10 mm but none were > 20 mm. None of the carcinoids with a diameter ≤ 10 mm invaded the muscularis propria. Of the 4 carcinoids > 10 mm, 1 was T2N0 and 3 were T1N0. All carcinoid tumors in our series were low grade and with low proliferation indexes, and all were treated successfully by local excision.

Conclusion

We believe our study shows that EUS can be highly accurate in staging rectal lesions, specifically lesions that are T1-T2N0, be they adenocarcinoma or carcinoid. Although we could not assess overstaging for lesions that were staged > T2 or > N0, we were able to determine no understaging for all of our patients. In experienced hands, EUS remains a highly accurate staging tool for early rectal carcinoma.

Colorectal cancer is the second most common cause of cancer death in the US, with one-third of all colorectal cancers occurring within the rectum. Each year, an estimated 40000 Americans are diagnosed with rectal cancer (RC).^1,2 The prognosis and treatment of RC depends on both T and N stage at the time of diagnosis.^3-5 According to the most recent National Comprehensive Cancer Network guidelines from May 2019, patients with T1 to T2N0 tumors should undergo transanal or transabdominal surgery upfront, whereas patients with T3 to T4N0 or any TN1 to 2 should start with neoadjuvant therapy for better locoregional control, followed by surgery.⁶ Therefore, the appropriate management of RC requires adequate staging.

Endoscopic ultrasound (EUS), magnetic resonance imaging (MRI), and computed tomography (CT) are the imaging techniques currently used to stage RC. In a meta-analysis of 90 articles published between 1985 and 2002 that compared the 3 radiologic modalities, Bipat and colleagues found that MRI and EUS had a similar sensitivity of 94%, whereas the specificity of EUS (86%) was significantly higher than that of MRI (69%) for muscularis propria invasion.⁷ CT was performed only in a limited number of trials because CT was considered inadequate to assess early T stage. For perirectal tissue invasion, the sensitivity of EUS was statistically higher than that of CT and MRI imaging: 90% compared with 79% and 82%, respectively. The specificity estimates for EUS, CT, and MRI were comparable: 75%, 78%, and 76%, respectively. The respective sensitivity and specificity of the 3 imaging modalities to evaluate lymph nodes were also comparable: EUS, 67% and 78%; CT, 55% and 74%; and MRI, 66% and 76%.

The role of EUS in the diagnosis and treatment of RC has long been validated.^1,2-5 A meta-analysis of 42 studies involving 5039 patients found EUS to be highly accurate for differentiating various T stages.⁸ However, EUS cannot assess iliac and mesenteric lymph nodes or posterior tumor extension beyond endopelvic fascia in advanced RC. Notable heterogeneity was found among the studies in the meta-analyses with regard to the type of equipment used for staging, as well as the criteria used to assess the depth of penetration and nodal status. The recent introduction of phased-array coils and the development of T2-weighted fast spin sequences have improved the resolution of MRI. The MERCURY trial showed that extension of tumor to within 1 mm of the circumferential margin on high-resolution MRI correctly predicted margin involvement at the time of surgery in 92% of the patients.⁹ In the retrospective study by Balyasnikova and colleagues, MRI was found to correctly identify partial submucosal invasion and suitability for local excision in 89% of the cases.¹⁰

Therefore, both EUS and MRI are useful, more so than CT, in assessment of the depth of tumor invasion, nodal staging, and predicting the circumferential resection margin. The use of EUS, however, does not preclude the use of MRI, or vice versa. Rather, the 2 modalities can complement each other in staging and proper patient selection for treatment.¹¹

Despite data supporting the value of EUS in staging RC, its use is limited by a high degree of operator dependence and a substantial learning curve,^12-17 which may explain the low EUS accuracy observed in some reports.^7,13,15 Given the presence of recognized alternatives such as MRI, we decided to reevaluate EUS accuracy for the staging of RC outside high-volume specialized centers and prospective clinical trials.

Methods

A retrospective chart review was performed that included all consecutive patients undergoing rectal ultrasound from January 2011 to August 2015 at the US Department of Veterans Affairs Medical Center (VAMC) in Memphis, Tennessee. Sixty-five patients with short-stocked or sessile lesions < 15 cm from anal margin staged T2N0M0 or lower by endorectal ultrasound (ERUS) were included. The patients with neoplasms staged in excess of T2 or N0 were excluded from the study because treatment protocol dictates immediate neoadjuvant treatment, the administration of which would affect subsequent histopathology.

For the 37 patients included in the final analysis, ERUS results were compared with surgical pathology to ascertain accuracy. The resections were performed endoscopically or surgically with a goal of obtaining clear margins. The choice of procedure depended on size, shape, location, and depth of invasion. All patients underwent clinical and endoscopic surveillance with flexible sigmoidoscopy/EUS every 3 to 6 months for the first 2 years. We used 2 different gold standards for surveillance depending on the type of procedure performed to remove the lesion. A pathology report was the gold standard used for patients who underwent surgery. In patients who underwent endoscopic resection, we used the lack of recurrent disease, determined by normal endoscopic and endoscopic ultrasound examination, to signify complete endoscopic resection and therefore adequate staging as an early neoplasm.

Results

From January 2011 to August 2015, 65 rectal ultrasounds were performed. All EUS procedures were performed by 1 physician (C Ruben Tombazzi). All patients had previous endoscopic evaluation and tissue diagnoses. Twenty-eight patients were excluded: 18 had T3 or N1 disease, 2 had T2N0 but refused surgery, 2 had anal cancer, 3 patients with suspected cancer had benign nonneoplastic disease (2 radiation proctitis, 1 normal rectal wall), and 3 underwent EUS for benign tumors (1 ganglioneuroma and 2 lipomas).

Thirty-seven patients were included in the study, 3 of whom were staged as T2N0 and 34 as T1N0 or lower by EUS. All patients were men ranging in age from 43 to 73 years (mean, 59 years). All 37 patients underwent endoscopic or surgical resection of their early rectal neoplasm. The final pathologic evaluation of the specimens demonstrated 14 carcinoid tumors, 11 adenocarcinomas, 6 tubular adenomas with high-grade dysplasia, and 6 benign adenomas. The preoperative EUS staging was confirmed for all patients, with 100% sensitivity, specificity, and accuracy. None of the patients who underwent endoscopic or surgical transanal resection had recurrence, determined by normal endoscopic and endoscopic ultrasound appearance, during a mean of 32.6 months surveillance.

Discussion

EUS has long been a recognized method for T and N staging of RC.^1,3-5,7,8 Our data confirm that, in experienced hands, EUS is highly accurate in the staging of early rectal cancers.

The impact of EUS on the management of RC was demonstrated in a Mayo Clinic prospective blinded study.¹ In that cohort of 80 consecutive patients who had previously had a CT for staging, EUS altered patient management in about 30% of cases. The most common change precipatated by EUS was the indication for additional neoadjuvant treatment.

However, the results have not been as encouraging when ERUS is performed outside of strict research protocol. A multicenter, prospective, country-wide quality assurance study from > 300 German hospitals was designed to assess the diagnostic accuracy of EUS in RC.¹³ Of 29206 patients, 7096 underwent surgery, without neoadjuvant treatment, and were included in the final analysis. The correspondence of tumor invasion with histopathology was 64.7%, with understaging of 18% and overstaging of 17.3%.¹³ These numbers were better in hospitals with greater experience performing ERUS: 73% accuracy in the centers with a case load of > 30 cases per year compared with 63.2% accuracy for the centers with < 10 cases a year. Marusch and colleagues had previously demonstrated an EUS accuracy of 63.3% in a study of 1463 patients with RC in Germany.¹⁴ Another study based out of the UK had similar findings. Ashraf and colleagues performed a database analyses from 20 UK centers and identified 165 patients with RC who underwent ERUS and endoscopic microsurgery.¹⁵ Compared with histopathology, EUS had 57.1% sensitivity, 73% specificity, and 42.9% accuracy for T1 cancers; EUS accuracy was 50% for T2 and 58% for T3 tumors. The authors concluded that the general accuracy of EUS in determining stage was around 50%, the statistical equivalent of flipping a coin.

The low accuracy of EUS observed by German and British multicenter studies^13-15 was attributed to the difference that may exist in clinical trials at specialized centers compared with wider use of EUS in a community setting. As seen by our data, the Memphis VAMC is not a high-volume center for the treatment of RC. However, all our EUS procedures were performed and interpreted by a single operator (C. Ruben Tombazzi) with 18 years of EUS experience. We cannot conclude that no patient was overstaged, as patients receiving a stage of T3N0 or T > N0 received neoadjuvant treatment and were not included. However, we can conclude that no patient was understaged. All patients deemed to be T1 to T2N0 included in our study received accurate staging. Our results are consistent with the high accuracy of EUS reported from other centers with experience in diagnosis and treatment of RC.^1,3-5,17,18

Although EUS is accurate in differentiating T1 from T2 tumors, it cannot reliably differentiate T1 from T0 lesions. In one study, 57.6% of adenomas and 30.7% of carcinomas in situ were staged as T1 on EUS, while almost half of T1 cancers were interpreted as T0.¹⁷ This drawback is a well-known limitation of EUS; although, the misinterpretation does not affect treatment, as both T0 and T1 lesions can be treated successfully by local excision alone, which was the algorithm used for our patients. The choice of the specific procedure for local excision was left to the clinicians and included transanal endoscopic or surgical resections. At a mean follow-up of 32.6 months, none of the 37 patients who underwent endoscopic or surgical transanal resection had evidence of recurrent disease.

A limitation of EUS, or any other imaging modality, is differentiating tumor invasion from peritumoral inflammation. The inflammation can render images of tumor borders ill-defined and irregular, which hinders precise staging. However, the accurate identification of tumors with deep involvement of the submucosa (T1sm3) is of importance, because these tumors are more advanced than the superficial and intermediate T1 lesions (T1sm1 and T1sm2, respectively).

Patients with RC whose lesions are considered T1sm3 are at higher risk of harboring lymph node metastases.¹⁸ Nascimbeni and colleagues had shown that the invasion into the lower third of the submucosa (sm3) was an independent risk factor for lower cancer-free survival among patients with T1 RC.¹⁹We did not measure the distance of the tumor to muscular layer in our study, but we relied on EUS to predict the circumferential tumor margins and guide the surgical resection. Of the 11 patients with T1 rectal adenocarcinomas and the 6 patients with tubular adenoma with high-grade dysplasia, all treated by local excision, none developed a local or distant recurrence during follow-up.

Unlike rectal adenocarcinomas, the prognosis for carcinoid tumors correlates not only with the depth of invasion but also with the size of the tumor. The other adverse prognostic features include poor differentiation, high mitosis index, and lymphovascular invasion.²⁰

EUS had been shown to be highly accurate in determining the precise carcinoid tumor size, depth of invasion, and lymph node metastases.^20,21 In a study of 66 resected rectal carcinoid tumors by Ishii and colleagues, 57 lesions had a diameter of ≤ 10 mm and 9 lesions had a diameter of > 10 mm.²¹ All of the 57 carcinoid tumors with a diameter of ≤ 10 mm were confined to the submucosa. In contrast, 5 of the 9 lesions > 10 mm invaded the muscularis propria, 6 had a lymphovascular invasion, 4 were lymph node metastases, and 1 was a liver metastasis.

In our series, 4 of the 14 carcinoid tumors were > 10 mm but none were > 20 mm. None of the carcinoids with a diameter ≤ 10 mm invaded the muscularis propria. Of the 4 carcinoids > 10 mm, 1 was T2N0 and 3 were T1N0. All carcinoid tumors in our series were low grade and with low proliferation indexes, and all were treated successfully by local excision.

Conclusion

We believe our study shows that EUS can be highly accurate in staging rectal lesions, specifically lesions that are T1-T2N0, be they adenocarcinoma or carcinoid. Although we could not assess overstaging for lesions that were staged > T2 or > N0, we were able to determine no understaging for all of our patients. In experienced hands, EUS remains a highly accurate staging tool for early rectal carcinoma.