Jonathan Bortinger, MD

Clinical question: What are the risk factors identified at the onset of illness that are associated with recurrent C. diff infection?

Background: After initial infection, 10%-25% of patients experience recurrent C. diff infection (CDI). The identification of patients at high risk of recurrence would be beneficial for therapeutic decision making.

Study design: Retrospective cohort study.

Setting: Large, urban, academic medical center.

Synopsis: Authors included 4,196 patients with an initial infection, defined by a positive C. diff toxin assay and unformed stools. A repeat positive toxin within 42 days of completing treatment for the initial infection represented recurrent CDI. Multiple characteristics were examined to identify risks of recurrent infection, including demographics and those related to acute and chronic disease. A logistic regression model was used to identify risk factors for recurrence.

Four hundred twenty-five patients (10.1%) had recurrent infection. Age, fluoroquinolone and high-risk antibiotic use, community-acquired healthcare-associated infection, multiple hospitalizations, and gastric acid suppression were found to predict recurrent infection through multivariate analysis.

Limitations of the study included potential confounding, use of observational data, and generalizability, given the urban academic medical center setting. This prediction model differs from previously developed models in that it identifies factors present at the onset of infection.

Bottom line: Multiple factors identified at the onset of illness can predict CDI recurrence.

Citation: Zilberberg MD, Reske K, Olsen M, Yan Y, Dubberke ER. Development and validation of a recurrent Clostridium difficile risk-prediction model. J Hosp Med. 2014;9(7):418-423.

Clinical question: What are the risk factors identified at the onset of illness that are associated with recurrent C. diff infection?

Background: After initial infection, 10%-25% of patients experience recurrent C. diff infection (CDI). The identification of patients at high risk of recurrence would be beneficial for therapeutic decision making.

Study design: Retrospective cohort study.

Setting: Large, urban, academic medical center.

Synopsis: Authors included 4,196 patients with an initial infection, defined by a positive C. diff toxin assay and unformed stools. A repeat positive toxin within 42 days of completing treatment for the initial infection represented recurrent CDI. Multiple characteristics were examined to identify risks of recurrent infection, including demographics and those related to acute and chronic disease. A logistic regression model was used to identify risk factors for recurrence.

Four hundred twenty-five patients (10.1%) had recurrent infection. Age, fluoroquinolone and high-risk antibiotic use, community-acquired healthcare-associated infection, multiple hospitalizations, and gastric acid suppression were found to predict recurrent infection through multivariate analysis.

Limitations of the study included potential confounding, use of observational data, and generalizability, given the urban academic medical center setting. This prediction model differs from previously developed models in that it identifies factors present at the onset of infection.

Bottom line: Multiple factors identified at the onset of illness can predict CDI recurrence.

Citation: Zilberberg MD, Reske K, Olsen M, Yan Y, Dubberke ER. Development and validation of a recurrent Clostridium difficile risk-prediction model. J Hosp Med. 2014;9(7):418-423.

Clinical question: What are the risk factors identified at the onset of illness that are associated with recurrent C. diff infection?

Background: After initial infection, 10%-25% of patients experience recurrent C. diff infection (CDI). The identification of patients at high risk of recurrence would be beneficial for therapeutic decision making.

Study design: Retrospective cohort study.

Setting: Large, urban, academic medical center.

Synopsis: Authors included 4,196 patients with an initial infection, defined by a positive C. diff toxin assay and unformed stools. A repeat positive toxin within 42 days of completing treatment for the initial infection represented recurrent CDI. Multiple characteristics were examined to identify risks of recurrent infection, including demographics and those related to acute and chronic disease. A logistic regression model was used to identify risk factors for recurrence.

Four hundred twenty-five patients (10.1%) had recurrent infection. Age, fluoroquinolone and high-risk antibiotic use, community-acquired healthcare-associated infection, multiple hospitalizations, and gastric acid suppression were found to predict recurrent infection through multivariate analysis.

Limitations of the study included potential confounding, use of observational data, and generalizability, given the urban academic medical center setting. This prediction model differs from previously developed models in that it identifies factors present at the onset of infection.

Bottom line: Multiple factors identified at the onset of illness can predict CDI recurrence.

Citation: Zilberberg MD, Reske K, Olsen M, Yan Y, Dubberke ER. Development and validation of a recurrent Clostridium difficile risk-prediction model. J Hosp Med. 2014;9(7):418-423.

Background

Hospitalists commonly order red blood cell (RBC) transfusion as a therapy for patients with anemia resulting from a variety of clinical conditions. There has been lack of consensus on when to transfuse, because patients with anemia frequently have multiple co-morbidities, including coronary artery disease and congestive heart failure, which may influence their ability to tolerate a potentially ischemic state related to anemia or to accommodate volume fluctuations related to transfusion.

Furthermore, RBC transfusions are not without inherent risk. Life-threatening transfusion reactions occur in approximately seven per million transfused blood components, and transfusion-associated circulatory overload (TACO) can develop in one in 100 transfusions.¹

Recently published guidelines provide recommendations for management of hemodynamically stable adults with anemia.

Guideline Update

The AABB published guidelines in the Annals of Internal Medicine in 2012 addressing RBC transfusion thresholds.¹ The updated guideline makes a recommendation that clinicians utilize a restrictive transfusion strategy. Transfusion is strongly recommended for ICU patients with hemoglobin ≤7g/dL. In post-operative surgical patients and for post-operative patients with symptomatic anemia, transfusion is recommended for hemoglobin ≤8g/dL. The authors also made a weak recommendation to transfuse for hemoglobin ≤8g/dL or for symptoms in hospitalized hemodynamically stable patients with preexisting cardiovascular disease.

These recommendations draw from past literature, along with two more recent trials examining liberal or restrictive transfusion thresholds. The newer trials increased the total number of patients studied by nearly one third compared with prior reviews.^2,3 The authors also incorporated recently published systematic reviews in their analysis.

Although the definition of a restrictive transfusion threshold varied across trials, including hemoglobin ≤7g/dL and ≤8g/dL, the authors used the pooled data to provide several recommendations in the new guideline. Of note, the pooled data was underpowered to detect up to a twofold increase in risk of myocardial infarction in patients in the restrictive strategy group.¹

There were insufficient data for the authors to recommend for or against a restrictive transfusion strategy in patients with acute coronary syndrome, based on very low quality evidence.

Finally, the authors recommended that symptoms and hemoglobin level should both be used in determining transfusion criteria, based on low quality of evidence.

Analysis

The current AABB guidelines have two primary differences from earlier guidelines. First, the AABB authors used GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) methodology to formalize evidence-based practice in their analysis of the literature. The authors purposely used the GRADE methodology to systematically evaluate the quality of the evidence base and explicitly state the strength of the recommendation for a particular transfusion threshold.⁴

Second, the AABB guidelines incorporated data from the more recently published FOCUS (Functional Outcomes in Cardiovascular patients Undergoing Surgical repair of hip fracture) and TRACS (Transfusion Requirements After Cardiac Surgery) trials, resulting in a stronger recommendation supporting the use of a restrictive transfusion strategy in non-ICU and post-operative patients. The findings of the FOCUS trial are especially applicable to hospitalists, because many patients who undergo hip fracture repair are directly cared for or are co-managed by hospitalists.

The current guidelines built upon previous guidelines that advocated a restrictive strategy (hemoglobin ≤7g/dL) in hemodynamically stable, critically ill adult patients.⁵ In general, restrictive transfusion strategy led to nearly 40% fewer patients receiving transfusion compared with the use of a liberal transfusion strategy.1 No additional harm to patients was evidenced in the restrictive transfusion group, though the trials were not designed to answer this question; moreover, there was no statistically significant difference in mortality or functional outcome between the two groups.

The authors of the current AABB guidelines recognized the importance of replicating the current findings in a more diverse patient population. An area where further study is indicated is in the use of specific transfusion thresholds in patients with acute coronary syndrome. These guidelines did not clarify whether or not there is a physiologic difference between use of different restrictive transfusion thresholds such as <8g/dL and <7g/dL.

The authors of the AABB guidelines also commented that performing a future trial to compare RBC transfusion for symptoms vs. hemoglobin “trigger” would be useful; however, they recognized that this may not be feasible due to the need to blind providers in the trial to hemoglobin values. Various society guidelines currently call for different transfusion thresholds or do not make a specific recommendation at all.¹

Key Takeaways for Hospitalists

For the vast majority of medical patients, hospitalists can safely use a restrictive RBC transfusion threshold (≤7g/dL or ≤8g/dL), which can lead to a significant decrease in RBC transfusions without adversely affecting overall mortality.

Drs. Bortinger and Carbo are hospitalists at Beth Israel Deaconess Medical Center in Boston.

References

1. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB. Ann Inter Med. 2012;157(1):49-58.
2. Carson AL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;367(26):2453-2462.
3. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA. 2010;304(14):1559-1567.
4. Carson JL, Carless PA, Herbert PC. Transfusion threshold and other strategies for guiding allogenic red blood cell transfusion. Cochrane Database Syst Rev. 2012;CD002042.
5. Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med. 2009;37(12):3124-3157.

Background

Hospitalists commonly order red blood cell (RBC) transfusion as a therapy for patients with anemia resulting from a variety of clinical conditions. There has been lack of consensus on when to transfuse, because patients with anemia frequently have multiple co-morbidities, including coronary artery disease and congestive heart failure, which may influence their ability to tolerate a potentially ischemic state related to anemia or to accommodate volume fluctuations related to transfusion.

Furthermore, RBC transfusions are not without inherent risk. Life-threatening transfusion reactions occur in approximately seven per million transfused blood components, and transfusion-associated circulatory overload (TACO) can develop in one in 100 transfusions.¹

Recently published guidelines provide recommendations for management of hemodynamically stable adults with anemia.

Guideline Update

The AABB published guidelines in the Annals of Internal Medicine in 2012 addressing RBC transfusion thresholds.¹ The updated guideline makes a recommendation that clinicians utilize a restrictive transfusion strategy. Transfusion is strongly recommended for ICU patients with hemoglobin ≤7g/dL. In post-operative surgical patients and for post-operative patients with symptomatic anemia, transfusion is recommended for hemoglobin ≤8g/dL. The authors also made a weak recommendation to transfuse for hemoglobin ≤8g/dL or for symptoms in hospitalized hemodynamically stable patients with preexisting cardiovascular disease.

These recommendations draw from past literature, along with two more recent trials examining liberal or restrictive transfusion thresholds. The newer trials increased the total number of patients studied by nearly one third compared with prior reviews.^2,3 The authors also incorporated recently published systematic reviews in their analysis.

Although the definition of a restrictive transfusion threshold varied across trials, including hemoglobin ≤7g/dL and ≤8g/dL, the authors used the pooled data to provide several recommendations in the new guideline. Of note, the pooled data was underpowered to detect up to a twofold increase in risk of myocardial infarction in patients in the restrictive strategy group.¹

There were insufficient data for the authors to recommend for or against a restrictive transfusion strategy in patients with acute coronary syndrome, based on very low quality evidence.

Finally, the authors recommended that symptoms and hemoglobin level should both be used in determining transfusion criteria, based on low quality of evidence.

Analysis

The current AABB guidelines have two primary differences from earlier guidelines. First, the AABB authors used GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) methodology to formalize evidence-based practice in their analysis of the literature. The authors purposely used the GRADE methodology to systematically evaluate the quality of the evidence base and explicitly state the strength of the recommendation for a particular transfusion threshold.⁴

Second, the AABB guidelines incorporated data from the more recently published FOCUS (Functional Outcomes in Cardiovascular patients Undergoing Surgical repair of hip fracture) and TRACS (Transfusion Requirements After Cardiac Surgery) trials, resulting in a stronger recommendation supporting the use of a restrictive transfusion strategy in non-ICU and post-operative patients. The findings of the FOCUS trial are especially applicable to hospitalists, because many patients who undergo hip fracture repair are directly cared for or are co-managed by hospitalists.

The current guidelines built upon previous guidelines that advocated a restrictive strategy (hemoglobin ≤7g/dL) in hemodynamically stable, critically ill adult patients.⁵ In general, restrictive transfusion strategy led to nearly 40% fewer patients receiving transfusion compared with the use of a liberal transfusion strategy.1 No additional harm to patients was evidenced in the restrictive transfusion group, though the trials were not designed to answer this question; moreover, there was no statistically significant difference in mortality or functional outcome between the two groups.

The authors of the current AABB guidelines recognized the importance of replicating the current findings in a more diverse patient population. An area where further study is indicated is in the use of specific transfusion thresholds in patients with acute coronary syndrome. These guidelines did not clarify whether or not there is a physiologic difference between use of different restrictive transfusion thresholds such as <8g/dL and <7g/dL.

The authors of the AABB guidelines also commented that performing a future trial to compare RBC transfusion for symptoms vs. hemoglobin “trigger” would be useful; however, they recognized that this may not be feasible due to the need to blind providers in the trial to hemoglobin values. Various society guidelines currently call for different transfusion thresholds or do not make a specific recommendation at all.¹

Key Takeaways for Hospitalists

For the vast majority of medical patients, hospitalists can safely use a restrictive RBC transfusion threshold (≤7g/dL or ≤8g/dL), which can lead to a significant decrease in RBC transfusions without adversely affecting overall mortality.

Drs. Bortinger and Carbo are hospitalists at Beth Israel Deaconess Medical Center in Boston.

References

1. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB. Ann Inter Med. 2012;157(1):49-58.
2. Carson AL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;367(26):2453-2462.
3. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA. 2010;304(14):1559-1567.
4. Carson JL, Carless PA, Herbert PC. Transfusion threshold and other strategies for guiding allogenic red blood cell transfusion. Cochrane Database Syst Rev. 2012;CD002042.
5. Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med. 2009;37(12):3124-3157.

Background

Hospitalists commonly order red blood cell (RBC) transfusion as a therapy for patients with anemia resulting from a variety of clinical conditions. There has been lack of consensus on when to transfuse, because patients with anemia frequently have multiple co-morbidities, including coronary artery disease and congestive heart failure, which may influence their ability to tolerate a potentially ischemic state related to anemia or to accommodate volume fluctuations related to transfusion.

Furthermore, RBC transfusions are not without inherent risk. Life-threatening transfusion reactions occur in approximately seven per million transfused blood components, and transfusion-associated circulatory overload (TACO) can develop in one in 100 transfusions.¹

Recently published guidelines provide recommendations for management of hemodynamically stable adults with anemia.

Guideline Update

The AABB published guidelines in the Annals of Internal Medicine in 2012 addressing RBC transfusion thresholds.¹ The updated guideline makes a recommendation that clinicians utilize a restrictive transfusion strategy. Transfusion is strongly recommended for ICU patients with hemoglobin ≤7g/dL. In post-operative surgical patients and for post-operative patients with symptomatic anemia, transfusion is recommended for hemoglobin ≤8g/dL. The authors also made a weak recommendation to transfuse for hemoglobin ≤8g/dL or for symptoms in hospitalized hemodynamically stable patients with preexisting cardiovascular disease.

These recommendations draw from past literature, along with two more recent trials examining liberal or restrictive transfusion thresholds. The newer trials increased the total number of patients studied by nearly one third compared with prior reviews.^2,3 The authors also incorporated recently published systematic reviews in their analysis.

Although the definition of a restrictive transfusion threshold varied across trials, including hemoglobin ≤7g/dL and ≤8g/dL, the authors used the pooled data to provide several recommendations in the new guideline. Of note, the pooled data was underpowered to detect up to a twofold increase in risk of myocardial infarction in patients in the restrictive strategy group.¹

There were insufficient data for the authors to recommend for or against a restrictive transfusion strategy in patients with acute coronary syndrome, based on very low quality evidence.

Finally, the authors recommended that symptoms and hemoglobin level should both be used in determining transfusion criteria, based on low quality of evidence.

Analysis

The current AABB guidelines have two primary differences from earlier guidelines. First, the AABB authors used GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) methodology to formalize evidence-based practice in their analysis of the literature. The authors purposely used the GRADE methodology to systematically evaluate the quality of the evidence base and explicitly state the strength of the recommendation for a particular transfusion threshold.⁴

Second, the AABB guidelines incorporated data from the more recently published FOCUS (Functional Outcomes in Cardiovascular patients Undergoing Surgical repair of hip fracture) and TRACS (Transfusion Requirements After Cardiac Surgery) trials, resulting in a stronger recommendation supporting the use of a restrictive transfusion strategy in non-ICU and post-operative patients. The findings of the FOCUS trial are especially applicable to hospitalists, because many patients who undergo hip fracture repair are directly cared for or are co-managed by hospitalists.

The current guidelines built upon previous guidelines that advocated a restrictive strategy (hemoglobin ≤7g/dL) in hemodynamically stable, critically ill adult patients.⁵ In general, restrictive transfusion strategy led to nearly 40% fewer patients receiving transfusion compared with the use of a liberal transfusion strategy.1 No additional harm to patients was evidenced in the restrictive transfusion group, though the trials were not designed to answer this question; moreover, there was no statistically significant difference in mortality or functional outcome between the two groups.

The authors of the current AABB guidelines recognized the importance of replicating the current findings in a more diverse patient population. An area where further study is indicated is in the use of specific transfusion thresholds in patients with acute coronary syndrome. These guidelines did not clarify whether or not there is a physiologic difference between use of different restrictive transfusion thresholds such as <8g/dL and <7g/dL.

The authors of the AABB guidelines also commented that performing a future trial to compare RBC transfusion for symptoms vs. hemoglobin “trigger” would be useful; however, they recognized that this may not be feasible due to the need to blind providers in the trial to hemoglobin values. Various society guidelines currently call for different transfusion thresholds or do not make a specific recommendation at all.¹

Key Takeaways for Hospitalists

For the vast majority of medical patients, hospitalists can safely use a restrictive RBC transfusion threshold (≤7g/dL or ≤8g/dL), which can lead to a significant decrease in RBC transfusions without adversely affecting overall mortality.

Drs. Bortinger and Carbo are hospitalists at Beth Israel Deaconess Medical Center in Boston.

References

1. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB. Ann Inter Med. 2012;157(1):49-58.
2. Carson AL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;367(26):2453-2462.
3. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA. 2010;304(14):1559-1567.
4. Carson JL, Carless PA, Herbert PC. Transfusion threshold and other strategies for guiding allogenic red blood cell transfusion. Cochrane Database Syst Rev. 2012;CD002042.
5. Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med. 2009;37(12):3124-3157.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. High-dose vs. low-dose clopidogrel after cardiac stenting
2. Rates of overdiagnosis of PE with CTPA
3. Outcomes of hospitalists with PAs or residents
4. White coats and MRSA
5. Correlation of vital signs and pain
6. Rate of asymptomatic perioperative MI
7. Relationship of opioid prescription patterns and overdose
8. Interdisciplinary rounds and rates of adverse events

High-Dose Clopidogrel Is Not Superior to Standard-Dose Clopidogrel in Patients with High On-Treatment Platelet Activity after Percutaneous Corona

Clinical question: In patients with high on-treatment platelet activity, does the use of high-dose clopidogrel after percutaneous coronary intervention (PCI) decrease the risk of cardiovascular events?

Background: In patients receiving clopidogrel, high platelet reactivity after PCI is associated with an increase in cardiovascular events. At present, treatments targeted at this population are not well-defined.

Study design: Randomized, double-blind, active-control trial.

Setting: Eighty-three centers in North America.

Synopsis: Researchers randomized 2,214 patients with drug-eluting stents to receive either high-dose clopidogrel (600 mg initial dose, 150 mg daily thereafter) or standard-dose clopidogrel (no additional loading dose, 75 mg daily). At six months, the primary endpoint of death from cardiovascular causes, nonfatal myocardial infarction, or stent thrombosis was no different in the two groups (2.3% in the high-dose group versus 2.3% in the standard-dose group; hazard ratio 1.01).

Bottom line: High-dose clopidogrel adds no benefit over standard-dose clopidogrel in patients with high platelet reactivity who have undergone PCI with drug-eluting stent placement.

Citation: Price MJ, Berger PB, Teirstein PS, et al. Standard- vs. high-dose clopidogrel based on platelet function testing after percutaneous coronary intervention: the GRAVITAS randomized trial. JAMA. 2011;305(11):1097-1105.

Computed Tomographic Pulmonary Angiography (CTPA) Is Associated with Overdiagnosis and Overtreatment of Pulmonary Embolism (PE)

Clinical question: Is the use of CTPA associated with increased incidence of PE and increased complications from anticoagulation treatment?

Background: CTPA is a sensitive, noninvasive test for diagnosing PE that could have a drawback: identifying potentially clinically unimportant (small) pulmonary emboli that subsequently are treated. Overtreatment might be associated with patient harm due to increased complications of anticoagulation therapy.

Study design: Time-trend analysis of PE between the pre-CTPA period (1993 to 1998) and the post-CTPA period (1998 to 2006).

Setting: Nongovernmental U.S. hospitals.

Synopsis: The Nationwide Inpatient Sample and Multiple Cause-of-Death databases were used to determine national estimates of hospitalization for PE, along with morbidity and mortality from PE.

The age-adjusted analysis revealed a statistically significant increase in the incidence of PE diagnosis after introduction of CTPA (to 112 per 100,000 from 62 per 100,000), with minimal change in overall PE mortality. This was accompanied by a substantial reduction in PE case-fatality rate, the rate of hospital deaths among patients with a diagnosis of pulmonary embolism.

Availability of CTPA was associated with a significant increase in anticoagulation complication rates (to 5.3 per 100,000 from 3.1 per 100,000), including statistically significant increases in gastrointestinal hemorrhage and secondary thrombocytopenia, and a trend toward higher rates of intracranial hemorrhage.

Bottom line: Introduction of CTPA was associated with changes suggestive of overdiagnosis (increased incidence, relatively unchanged mortality) and overtreatment (increased complication rates) of PE, but it remains unknown which small PEs are clinically significant.

Citation: Wiener RS, Schwartz LM, Woloshin S. Time trends in pulmonary embolism in the United States: evidence of overdiagnosis. Arch Intern Med. 2011;171(9):831-837.

Hospitalist-Physician Assistant Teams Associated with Longer Length of Stay, No Change in Mortality, Readmission Rates

Clinical question: Do length of stay (LOS), hospital mortality, or readmission rate change if hospitalists and physician assistants, or the traditional resident-hospitalist teams, provide the patient care?

Background: Resident work-hour limitations require new models of care for hospitalized patients. Many academic medical centers have hired physician assistants to work with hospitalists to provide care. Little is known about how these models affect such outcomes as LOS, inpatient mortality rates, and readmission rates.

Study design: Retrospective cohort.

Setting: A 430-bed urban academic medical center in Milwaukee.

Synopsis: Administrative data were gathered on 9,681 patients admitted to the general medical service. Of those enrolled, 2,171 were cared for by a hospitalist-physician assistant (H-PA) team, while resident-hospitalist teams cared for 7,510 patients. Patient assignment was dependent on time of admission but not on patient complexity. Patients admitted overnight after the resident team capped were assigned to the H-PA team the next morning, resulting in increased transitions of care for the H-PA team.

Adjusted analyses revealed a 6.45% increase in LOS for the H-PA team compared with the resident team. Charges, inpatient mortality, and readmission rates at seven, 14, and 30 days were unchanged. Subgroup analyses revealed smaller differences in LOS for H-PA teams and resident-hospitalist teams with the same hospitalist (LOS 5.44% higher, P=0.081).

Conclusions from this study are limited due to lack of randomization of assignment, the retrospective design, and the use of administrative data at one institution.

Bottom line: Hospitalist-PA teams might result in a slightly increased LOS compared with the traditional resident teams; however, inpatient mortality and readmission rates are similar.

Citation: Singh S, Fletcher KE, Schapira MM, et al. A comparison of outcomes of general medical inpatient care provided by a hospitalist-physician assistant model vs a traditional resident-based model. J Hosp Med. 2011;6:122-130.

Washing White Coats Does Not Lower MRSA Bacterial Contamination

Clinical question: Are clean, short-sleeved uniforms less likely to carry MRSA than regularly laundered long-sleeved white coats?

Background: Studies have shown that bacteria frequently colonize in physician garments. However, evidence that short-sleeved garments or newly laundered garments are less likely to be contaminated has been lacking. Despite the paucity of evidence, the British Department of Health barred the use of traditional white coats and long-sleeved garments in 2007.

Study design: Prospective, randomized, controlled trial.

Setting: Urban U.S. hospital.

Synopsis: Study authors randomized 100 internal-medicine residents and hospitalists to their own long-sleeved white coats or freshly laundered short-sleeved uniforms from August 2008 to November 2009. Swabs were taken from the sleeves of the white coats or uniform, the breast pocket, and the volar wrist surface of the dominant hand. Swabs were cultured for MRSA and for general colony count.

Results showed no significant difference in colony counts or MRSA colonization in any of the sites tested between the newly laundered uniforms and the white coats. Additionally, there was no effect in relation to the frequency of laundering the white coats. Notably, within three hours of donning freshly laundered uniforms, bacterial counts approached 50% of the total bacterial counts seen at eight hours.

Bottom line: Laundering of uniforms does not affect MRSA colonization rate or general bacterial burden on physician uniforms or skin surfaces, though the effect on nosocomial infection has not been established.

Citation: Burden M, Cervantes L, Weed D, Keniston A, Price CS, Albert RK. Newly cleaned physician uniforms and infrequently washed white coats have similar rates of bacterial contamination after an 8-hour workday: a randomized controlled trial. J Hosp Med. 2011;6:177-182.

Self-Reported Pain Severity Does Not Correlate with Heart Rate or Blood Pressure Measurements in Pre-Hospital Setting

Clinical question: Do measured vital signs, including heart rate, blood pressure, and respiratory rate, correlate with the degree of self-reported pain?

Background: Because pain often can be associated with alterations in autonomic tone, it has been hypothesized that alterations in vital signs will occur in patients who report pain.

Study design: Retrospective cohort study.

Setting: Pre-hospital in Melbourne, Australia.

Synopsis: The authors reviewed all ambulance patient care records for patients age >14 years with a Glasgow Coma Score (GCS) >12 transported to a hospital during a seven-day period in 2005. Patients were selected for analysis if their patient care record included an initial assessment of pain severity, as measured by a numeric rating scale (NRS), in which patients rate their pain from 0 to 10.

More than half of the 3,357 patients transported by paramedics during the period were included in this analysis (n=1286). There was no correlation between heart rate or systolic blood pressure with the degree of self-reported pain. Although an increased respiratory rate was statistically correlated with a higher rating of pain, this relationship was not clinically significant, as each one-point increase in the pain rating scale was associated with a 0.16-breaths-per-minute increase in the respiratory rate.

Limitations included the large number of records excluded from analysis because pain was not evaluated, as well as numerous unmeasured confounders, including active disease processes such as sepsis, that were not accounted for.

Bottom line: Severity of pain did not correlate with heart rate or systolic blood pressure in the pre-hospital setting.

Citation: Lord B, Woollard M. The reliability of vital signs in estimating pain severity among adult patients treated by paramedics. Emerg Med J. 2011;28:147-150.

Asymptomatic Perioperative Myocardial Infarction Is Common in Patients Undergoing Noncardiac Surgery

Clinical question: In patients undergoing noncardiac surgery, what is the incidence and clinical characteristics of perioperative myocardial infarction (MI)?

Background: Though millions of patients experience perioperative MI after noncardiac surgery, little is known about the characteristics and outcomes of these patients.

Study design: Cohort study.

Setting: One hundred ninety centers in 23 countries.

Synopsis: Using data from the 8,351 patients in the POISE (PeriOperative ISchemic Evaluation) trial, this study showed that perioperative MI occurred in 5% of patients; 65% were asymptomatic. Patients who experienced postoperative MI were older and had more cardiovascular risk factors when compared to those who did not. The 30-day mortality was higher in patients with a perioperative MI (11.6%) compared with those who did not (2.2%); the presence or absence of ischemic symptoms was not associated with mortality rate.

Of the 8.3% of patients who experienced an elevation in cardiac biomarkers but who did not meet the definition of MI, there was an increased risk of nonfatal cardiac arrest and nonacute coronary revascularization. Those in the highest quartile also had increased 30-day mortality.

Bottom line: Given the high proportion of asymptomatic MIs and isolated elevations in cardiac biomarkers and the association between these events and increased risk of death, hospitalists should consider routine monitoring of troponin in at-risk patients undergoing noncardiac surgery.

Citation: Devereaux PJ, Xavier D, Pogue J, et al. Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: a cohort study. Ann Intern Med. 2011;154(8):523-528.

Patients Prescribed Higher Opioid Doses Are at Increased Risk of Opioid Overdose Death

Clinical question: What is the association between opioid prescribing patterns and fatal opioid overdose?

Background: In the past 10 years, the rate of fatal overdose from opioid prescription for pain has more than doubled. Little is known about how the indications (substance abuse disorders, cancer-related pain, chronic pain, acute pain), maximal daily dose, and scheduling (standing, as-needed, or both) of opioid prescriptions relate to this increased risk.

Study design: Case-cohort study.

Setting: Veterans Health Administration (VHA) patients.

Synopsis: The VHA’s National Patient Care Database was used to randomly select a cohort of 154,684 nonhospice/nonpalliative-care patients who were prescribed opioids from 2004 to 2008. They were compared with 750 patients who were treated with prescription opioids who died from opioid overdose during this time.

Fatal opioid overdose was a rare event (0.04%), but risk increased with higher prescribed maximum daily morphine dose-equivalence, especially when greater than or equal to 50 mg/day in all subgroups (substance abuse, acute and chronic pain, and cancer). Fatal overdoses were higher in middle-aged white men with acute or chronic pain, substance abuse disorders, and other psychiatric illness. Patients with cancer were at increased risk of fatal overdose if they were prescribed as-needed opioids alone.

Treatment with both as-needed and standing opioids did not statistically affect risk of overdose death in any subgroup.

Bottom line: Although rare, risk of fatal opioid overdose in patients prescribed opiate medication increases with higher maximum prescribed daily dose.

Citation: Bohnert AS, Valenstein M, Bair M, et al. Association between opioid prescribing patterns and opioid overdose-related deaths. JAMA. 2011; 305:1315-1321.

Structured Interdisciplinary Rounds on Medical Teaching Unit Significantly Decrease Adverse Events

Clinical question: Do structured interdisciplinary rounds have an impact on the rate of adverse events?

Background: Many preventable adverse events occurring during hospitalization can be attributed to communication failures. Structured interdisciplinary rounds provide a format as well as a forum for team members to discuss patient care. Prior studies demonstrated improvements in collaboration; whether this translates to better patient care is not known.

Study design: Retrospective cohort using historic and concurrent control.

Setting: Tertiary-care teaching hospital in Chicago.

Synopsis: Structured interdisciplinary rounds, led by a nurse manager and medical director, and including nurses, residents, pharmacists, social workers, and case managers, were implemented on a medical teaching unit. New patients were discussed using a structured communication tool; existing patients were discussed in an unstructured format. Medical records were abstracted for 370 patients hospitalized after implementation of the intervention, equally divided between intervention and control units. One hundred eighty-five patients hospitalized on the intervention unit prior to the implementation of rounds served as a historic control.

Patients in the intervention unit had significantly lower rates of total adverse events (3.9 per 100 patient days in the intervention, compared with 7.2 and 7.7 per 100 patient days for the concurrent and historic control units, respectively), and preventable adverse events (0.9 per 100 patient days, compared with 2.8 and 2.1 per 100 patient days for the concurrent and historic controls, respectively).

Limitations of the study include lack of blinding of the medical record, slightly different patient populations in intervention and control groups, and the one-hospital setting, which could limit generalizability.

Bottom line: Structured interdisciplinary rounds might serve to improve communication between nurses, pharmacists, and physicians, resulting in decreases in adverse events.

Citation: O’Leary KJ, Buck R, Fligiel HM, et al. Structured interdisciplinary rounds in a medical teaching unit: improving patient safety. Arch Intern Med. 2011;171(7):678-684.

Pediatric HM Literature

Short-Course Antibiotic Therapy Effective for Bacterial Meningitis

Clinical question: Is five days of parenteral ceftriaxone as effective as 10 days for the treatment of bacterial meningitis in children?

Background: Morbidity and mortality in bacterial meningitis remain high, particularly in developing countries. Antibiotics are effective treatment, yet the optimal duration of treatment remains uncertain. Some data support a shorter duration of treatment (three to five days).

Study design: Multicountry, double-blind, placebo-controlled, randomized equivalence study.

Setting: Ten pediatric referral hospitals in Bangladesh, Egypt, Malawi, Pakistan, and Vietnam.

Synopsis: Children aged two months to 12 years with bacterial meningitis (due to Haemophilus influenza, Streptococcus pneumonia, Neisseria meningitidis, or culture-negative with indicative cerebrospinal fluid findings) and without complicating medical conditions were enrolled at participating centers. All children received 80 mg/kg to 100 mg/kg of parenteral ceftriaxone daily and a repeat lumbar puncture 48 to 72 hours after initiation of therapy.

Ultimately, 1,004 children without resistant organisms, persistently positive cultures, or suppurative complications were randomized on day five of therapy to placebo or continuance of ceftriaxone for five more days.

No bacteriologic failures (primary endpoint) were evident with either five or 10 days of treatment.

In addition, no statistically significant differences were found between the groups with respect to clinical treatment failure, hearing loss, neurological sequelae, or death. Secondary analysis by organism revealed similar results.

The primary limitation of this study is that it occurred in developing countries with a fair incidence of H. influenzae meningitis and a low rate of third-generation cephalosporin resistance.

However, pneumococcal and meningococcal disease remained prominent, and this study suggests that clinically stable patients might be treated with a shorter course of parenteral ceftriaxone therapy than currently is recommended.

Bottom line: Five days of ceftriaxone is as effective as 10 days for uncomplicated bacterial meningitis in children.

Citation: Molyneux E, Nizami SQ, Saha S, et al. 5 versus 10 days of treatment with ceftriaxone for bacterial meningitis in children: a double-blind randomised equivalence study. Lancet. 2011;377:1837-1845.

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. High-dose vs. low-dose clopidogrel after cardiac stenting
2. Rates of overdiagnosis of PE with CTPA
3. Outcomes of hospitalists with PAs or residents
4. White coats and MRSA
5. Correlation of vital signs and pain
6. Rate of asymptomatic perioperative MI
7. Relationship of opioid prescription patterns and overdose
8. Interdisciplinary rounds and rates of adverse events

High-Dose Clopidogrel Is Not Superior to Standard-Dose Clopidogrel in Patients with High On-Treatment Platelet Activity after Percutaneous Corona

Clinical question: In patients with high on-treatment platelet activity, does the use of high-dose clopidogrel after percutaneous coronary intervention (PCI) decrease the risk of cardiovascular events?

Background: In patients receiving clopidogrel, high platelet reactivity after PCI is associated with an increase in cardiovascular events. At present, treatments targeted at this population are not well-defined.

Study design: Randomized, double-blind, active-control trial.

Setting: Eighty-three centers in North America.

Synopsis: Researchers randomized 2,214 patients with drug-eluting stents to receive either high-dose clopidogrel (600 mg initial dose, 150 mg daily thereafter) or standard-dose clopidogrel (no additional loading dose, 75 mg daily). At six months, the primary endpoint of death from cardiovascular causes, nonfatal myocardial infarction, or stent thrombosis was no different in the two groups (2.3% in the high-dose group versus 2.3% in the standard-dose group; hazard ratio 1.01).

Bottom line: High-dose clopidogrel adds no benefit over standard-dose clopidogrel in patients with high platelet reactivity who have undergone PCI with drug-eluting stent placement.

Citation: Price MJ, Berger PB, Teirstein PS, et al. Standard- vs. high-dose clopidogrel based on platelet function testing after percutaneous coronary intervention: the GRAVITAS randomized trial. JAMA. 2011;305(11):1097-1105.

Computed Tomographic Pulmonary Angiography (CTPA) Is Associated with Overdiagnosis and Overtreatment of Pulmonary Embolism (PE)

Clinical question: Is the use of CTPA associated with increased incidence of PE and increased complications from anticoagulation treatment?

Background: CTPA is a sensitive, noninvasive test for diagnosing PE that could have a drawback: identifying potentially clinically unimportant (small) pulmonary emboli that subsequently are treated. Overtreatment might be associated with patient harm due to increased complications of anticoagulation therapy.

Study design: Time-trend analysis of PE between the pre-CTPA period (1993 to 1998) and the post-CTPA period (1998 to 2006).

Setting: Nongovernmental U.S. hospitals.

Synopsis: The Nationwide Inpatient Sample and Multiple Cause-of-Death databases were used to determine national estimates of hospitalization for PE, along with morbidity and mortality from PE.

The age-adjusted analysis revealed a statistically significant increase in the incidence of PE diagnosis after introduction of CTPA (to 112 per 100,000 from 62 per 100,000), with minimal change in overall PE mortality. This was accompanied by a substantial reduction in PE case-fatality rate, the rate of hospital deaths among patients with a diagnosis of pulmonary embolism.

Availability of CTPA was associated with a significant increase in anticoagulation complication rates (to 5.3 per 100,000 from 3.1 per 100,000), including statistically significant increases in gastrointestinal hemorrhage and secondary thrombocytopenia, and a trend toward higher rates of intracranial hemorrhage.

Bottom line: Introduction of CTPA was associated with changes suggestive of overdiagnosis (increased incidence, relatively unchanged mortality) and overtreatment (increased complication rates) of PE, but it remains unknown which small PEs are clinically significant.

Citation: Wiener RS, Schwartz LM, Woloshin S. Time trends in pulmonary embolism in the United States: evidence of overdiagnosis. Arch Intern Med. 2011;171(9):831-837.

Hospitalist-Physician Assistant Teams Associated with Longer Length of Stay, No Change in Mortality, Readmission Rates

Clinical question: Do length of stay (LOS), hospital mortality, or readmission rate change if hospitalists and physician assistants, or the traditional resident-hospitalist teams, provide the patient care?

Background: Resident work-hour limitations require new models of care for hospitalized patients. Many academic medical centers have hired physician assistants to work with hospitalists to provide care. Little is known about how these models affect such outcomes as LOS, inpatient mortality rates, and readmission rates.

Study design: Retrospective cohort.

Setting: A 430-bed urban academic medical center in Milwaukee.

Synopsis: Administrative data were gathered on 9,681 patients admitted to the general medical service. Of those enrolled, 2,171 were cared for by a hospitalist-physician assistant (H-PA) team, while resident-hospitalist teams cared for 7,510 patients. Patient assignment was dependent on time of admission but not on patient complexity. Patients admitted overnight after the resident team capped were assigned to the H-PA team the next morning, resulting in increased transitions of care for the H-PA team.

Adjusted analyses revealed a 6.45% increase in LOS for the H-PA team compared with the resident team. Charges, inpatient mortality, and readmission rates at seven, 14, and 30 days were unchanged. Subgroup analyses revealed smaller differences in LOS for H-PA teams and resident-hospitalist teams with the same hospitalist (LOS 5.44% higher, P=0.081).

Conclusions from this study are limited due to lack of randomization of assignment, the retrospective design, and the use of administrative data at one institution.

Bottom line: Hospitalist-PA teams might result in a slightly increased LOS compared with the traditional resident teams; however, inpatient mortality and readmission rates are similar.

Citation: Singh S, Fletcher KE, Schapira MM, et al. A comparison of outcomes of general medical inpatient care provided by a hospitalist-physician assistant model vs a traditional resident-based model. J Hosp Med. 2011;6:122-130.

Washing White Coats Does Not Lower MRSA Bacterial Contamination

Clinical question: Are clean, short-sleeved uniforms less likely to carry MRSA than regularly laundered long-sleeved white coats?

Background: Studies have shown that bacteria frequently colonize in physician garments. However, evidence that short-sleeved garments or newly laundered garments are less likely to be contaminated has been lacking. Despite the paucity of evidence, the British Department of Health barred the use of traditional white coats and long-sleeved garments in 2007.

Study design: Prospective, randomized, controlled trial.

Setting: Urban U.S. hospital.

Synopsis: Study authors randomized 100 internal-medicine residents and hospitalists to their own long-sleeved white coats or freshly laundered short-sleeved uniforms from August 2008 to November 2009. Swabs were taken from the sleeves of the white coats or uniform, the breast pocket, and the volar wrist surface of the dominant hand. Swabs were cultured for MRSA and for general colony count.

Results showed no significant difference in colony counts or MRSA colonization in any of the sites tested between the newly laundered uniforms and the white coats. Additionally, there was no effect in relation to the frequency of laundering the white coats. Notably, within three hours of donning freshly laundered uniforms, bacterial counts approached 50% of the total bacterial counts seen at eight hours.

Bottom line: Laundering of uniforms does not affect MRSA colonization rate or general bacterial burden on physician uniforms or skin surfaces, though the effect on nosocomial infection has not been established.

Citation: Burden M, Cervantes L, Weed D, Keniston A, Price CS, Albert RK. Newly cleaned physician uniforms and infrequently washed white coats have similar rates of bacterial contamination after an 8-hour workday: a randomized controlled trial. J Hosp Med. 2011;6:177-182.

Self-Reported Pain Severity Does Not Correlate with Heart Rate or Blood Pressure Measurements in Pre-Hospital Setting

Clinical question: Do measured vital signs, including heart rate, blood pressure, and respiratory rate, correlate with the degree of self-reported pain?

Background: Because pain often can be associated with alterations in autonomic tone, it has been hypothesized that alterations in vital signs will occur in patients who report pain.

Study design: Retrospective cohort study.

Setting: Pre-hospital in Melbourne, Australia.

Synopsis: The authors reviewed all ambulance patient care records for patients age >14 years with a Glasgow Coma Score (GCS) >12 transported to a hospital during a seven-day period in 2005. Patients were selected for analysis if their patient care record included an initial assessment of pain severity, as measured by a numeric rating scale (NRS), in which patients rate their pain from 0 to 10.

More than half of the 3,357 patients transported by paramedics during the period were included in this analysis (n=1286). There was no correlation between heart rate or systolic blood pressure with the degree of self-reported pain. Although an increased respiratory rate was statistically correlated with a higher rating of pain, this relationship was not clinically significant, as each one-point increase in the pain rating scale was associated with a 0.16-breaths-per-minute increase in the respiratory rate.

Limitations included the large number of records excluded from analysis because pain was not evaluated, as well as numerous unmeasured confounders, including active disease processes such as sepsis, that were not accounted for.

Bottom line: Severity of pain did not correlate with heart rate or systolic blood pressure in the pre-hospital setting.

Citation: Lord B, Woollard M. The reliability of vital signs in estimating pain severity among adult patients treated by paramedics. Emerg Med J. 2011;28:147-150.

Asymptomatic Perioperative Myocardial Infarction Is Common in Patients Undergoing Noncardiac Surgery

Clinical question: In patients undergoing noncardiac surgery, what is the incidence and clinical characteristics of perioperative myocardial infarction (MI)?

Background: Though millions of patients experience perioperative MI after noncardiac surgery, little is known about the characteristics and outcomes of these patients.

Study design: Cohort study.

Setting: One hundred ninety centers in 23 countries.

Synopsis: Using data from the 8,351 patients in the POISE (PeriOperative ISchemic Evaluation) trial, this study showed that perioperative MI occurred in 5% of patients; 65% were asymptomatic. Patients who experienced postoperative MI were older and had more cardiovascular risk factors when compared to those who did not. The 30-day mortality was higher in patients with a perioperative MI (11.6%) compared with those who did not (2.2%); the presence or absence of ischemic symptoms was not associated with mortality rate.

Of the 8.3% of patients who experienced an elevation in cardiac biomarkers but who did not meet the definition of MI, there was an increased risk of nonfatal cardiac arrest and nonacute coronary revascularization. Those in the highest quartile also had increased 30-day mortality.

Bottom line: Given the high proportion of asymptomatic MIs and isolated elevations in cardiac biomarkers and the association between these events and increased risk of death, hospitalists should consider routine monitoring of troponin in at-risk patients undergoing noncardiac surgery.

Citation: Devereaux PJ, Xavier D, Pogue J, et al. Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: a cohort study. Ann Intern Med. 2011;154(8):523-528.

Patients Prescribed Higher Opioid Doses Are at Increased Risk of Opioid Overdose Death

Clinical question: What is the association between opioid prescribing patterns and fatal opioid overdose?

Background: In the past 10 years, the rate of fatal overdose from opioid prescription for pain has more than doubled. Little is known about how the indications (substance abuse disorders, cancer-related pain, chronic pain, acute pain), maximal daily dose, and scheduling (standing, as-needed, or both) of opioid prescriptions relate to this increased risk.

Study design: Case-cohort study.

Setting: Veterans Health Administration (VHA) patients.

Synopsis: The VHA’s National Patient Care Database was used to randomly select a cohort of 154,684 nonhospice/nonpalliative-care patients who were prescribed opioids from 2004 to 2008. They were compared with 750 patients who were treated with prescription opioids who died from opioid overdose during this time.

Fatal opioid overdose was a rare event (0.04%), but risk increased with higher prescribed maximum daily morphine dose-equivalence, especially when greater than or equal to 50 mg/day in all subgroups (substance abuse, acute and chronic pain, and cancer). Fatal overdoses were higher in middle-aged white men with acute or chronic pain, substance abuse disorders, and other psychiatric illness. Patients with cancer were at increased risk of fatal overdose if they were prescribed as-needed opioids alone.

Treatment with both as-needed and standing opioids did not statistically affect risk of overdose death in any subgroup.

Bottom line: Although rare, risk of fatal opioid overdose in patients prescribed opiate medication increases with higher maximum prescribed daily dose.

Citation: Bohnert AS, Valenstein M, Bair M, et al. Association between opioid prescribing patterns and opioid overdose-related deaths. JAMA. 2011; 305:1315-1321.

Structured Interdisciplinary Rounds on Medical Teaching Unit Significantly Decrease Adverse Events

Clinical question: Do structured interdisciplinary rounds have an impact on the rate of adverse events?

Background: Many preventable adverse events occurring during hospitalization can be attributed to communication failures. Structured interdisciplinary rounds provide a format as well as a forum for team members to discuss patient care. Prior studies demonstrated improvements in collaboration; whether this translates to better patient care is not known.

Study design: Retrospective cohort using historic and concurrent control.

Setting: Tertiary-care teaching hospital in Chicago.

Synopsis: Structured interdisciplinary rounds, led by a nurse manager and medical director, and including nurses, residents, pharmacists, social workers, and case managers, were implemented on a medical teaching unit. New patients were discussed using a structured communication tool; existing patients were discussed in an unstructured format. Medical records were abstracted for 370 patients hospitalized after implementation of the intervention, equally divided between intervention and control units. One hundred eighty-five patients hospitalized on the intervention unit prior to the implementation of rounds served as a historic control.

Patients in the intervention unit had significantly lower rates of total adverse events (3.9 per 100 patient days in the intervention, compared with 7.2 and 7.7 per 100 patient days for the concurrent and historic control units, respectively), and preventable adverse events (0.9 per 100 patient days, compared with 2.8 and 2.1 per 100 patient days for the concurrent and historic controls, respectively).

Limitations of the study include lack of blinding of the medical record, slightly different patient populations in intervention and control groups, and the one-hospital setting, which could limit generalizability.

Bottom line: Structured interdisciplinary rounds might serve to improve communication between nurses, pharmacists, and physicians, resulting in decreases in adverse events.

Citation: O’Leary KJ, Buck R, Fligiel HM, et al. Structured interdisciplinary rounds in a medical teaching unit: improving patient safety. Arch Intern Med. 2011;171(7):678-684.

Pediatric HM Literature

Short-Course Antibiotic Therapy Effective for Bacterial Meningitis

Clinical question: Is five days of parenteral ceftriaxone as effective as 10 days for the treatment of bacterial meningitis in children?

Background: Morbidity and mortality in bacterial meningitis remain high, particularly in developing countries. Antibiotics are effective treatment, yet the optimal duration of treatment remains uncertain. Some data support a shorter duration of treatment (three to five days).

Study design: Multicountry, double-blind, placebo-controlled, randomized equivalence study.

Setting: Ten pediatric referral hospitals in Bangladesh, Egypt, Malawi, Pakistan, and Vietnam.

Synopsis: Children aged two months to 12 years with bacterial meningitis (due to Haemophilus influenza, Streptococcus pneumonia, Neisseria meningitidis, or culture-negative with indicative cerebrospinal fluid findings) and without complicating medical conditions were enrolled at participating centers. All children received 80 mg/kg to 100 mg/kg of parenteral ceftriaxone daily and a repeat lumbar puncture 48 to 72 hours after initiation of therapy.

Ultimately, 1,004 children without resistant organisms, persistently positive cultures, or suppurative complications were randomized on day five of therapy to placebo or continuance of ceftriaxone for five more days.

No bacteriologic failures (primary endpoint) were evident with either five or 10 days of treatment.

In addition, no statistically significant differences were found between the groups with respect to clinical treatment failure, hearing loss, neurological sequelae, or death. Secondary analysis by organism revealed similar results.

The primary limitation of this study is that it occurred in developing countries with a fair incidence of H. influenzae meningitis and a low rate of third-generation cephalosporin resistance.

However, pneumococcal and meningococcal disease remained prominent, and this study suggests that clinically stable patients might be treated with a shorter course of parenteral ceftriaxone therapy than currently is recommended.

Bottom line: Five days of ceftriaxone is as effective as 10 days for uncomplicated bacterial meningitis in children.

Citation: Molyneux E, Nizami SQ, Saha S, et al. 5 versus 10 days of treatment with ceftriaxone for bacterial meningitis in children: a double-blind randomised equivalence study. Lancet. 2011;377:1837-1845.

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. High-dose vs. low-dose clopidogrel after cardiac stenting
2. Rates of overdiagnosis of PE with CTPA
3. Outcomes of hospitalists with PAs or residents
4. White coats and MRSA
5. Correlation of vital signs and pain
6. Rate of asymptomatic perioperative MI
7. Relationship of opioid prescription patterns and overdose
8. Interdisciplinary rounds and rates of adverse events

High-Dose Clopidogrel Is Not Superior to Standard-Dose Clopidogrel in Patients with High On-Treatment Platelet Activity after Percutaneous Corona

Clinical question: In patients with high on-treatment platelet activity, does the use of high-dose clopidogrel after percutaneous coronary intervention (PCI) decrease the risk of cardiovascular events?

Background: In patients receiving clopidogrel, high platelet reactivity after PCI is associated with an increase in cardiovascular events. At present, treatments targeted at this population are not well-defined.

Study design: Randomized, double-blind, active-control trial.

Setting: Eighty-three centers in North America.

Synopsis: Researchers randomized 2,214 patients with drug-eluting stents to receive either high-dose clopidogrel (600 mg initial dose, 150 mg daily thereafter) or standard-dose clopidogrel (no additional loading dose, 75 mg daily). At six months, the primary endpoint of death from cardiovascular causes, nonfatal myocardial infarction, or stent thrombosis was no different in the two groups (2.3% in the high-dose group versus 2.3% in the standard-dose group; hazard ratio 1.01).

Bottom line: High-dose clopidogrel adds no benefit over standard-dose clopidogrel in patients with high platelet reactivity who have undergone PCI with drug-eluting stent placement.

Citation: Price MJ, Berger PB, Teirstein PS, et al. Standard- vs. high-dose clopidogrel based on platelet function testing after percutaneous coronary intervention: the GRAVITAS randomized trial. JAMA. 2011;305(11):1097-1105.

Computed Tomographic Pulmonary Angiography (CTPA) Is Associated with Overdiagnosis and Overtreatment of Pulmonary Embolism (PE)

Clinical question: Is the use of CTPA associated with increased incidence of PE and increased complications from anticoagulation treatment?

Background: CTPA is a sensitive, noninvasive test for diagnosing PE that could have a drawback: identifying potentially clinically unimportant (small) pulmonary emboli that subsequently are treated. Overtreatment might be associated with patient harm due to increased complications of anticoagulation therapy.

Study design: Time-trend analysis of PE between the pre-CTPA period (1993 to 1998) and the post-CTPA period (1998 to 2006).

Setting: Nongovernmental U.S. hospitals.

Synopsis: The Nationwide Inpatient Sample and Multiple Cause-of-Death databases were used to determine national estimates of hospitalization for PE, along with morbidity and mortality from PE.

The age-adjusted analysis revealed a statistically significant increase in the incidence of PE diagnosis after introduction of CTPA (to 112 per 100,000 from 62 per 100,000), with minimal change in overall PE mortality. This was accompanied by a substantial reduction in PE case-fatality rate, the rate of hospital deaths among patients with a diagnosis of pulmonary embolism.

Availability of CTPA was associated with a significant increase in anticoagulation complication rates (to 5.3 per 100,000 from 3.1 per 100,000), including statistically significant increases in gastrointestinal hemorrhage and secondary thrombocytopenia, and a trend toward higher rates of intracranial hemorrhage.

Bottom line: Introduction of CTPA was associated with changes suggestive of overdiagnosis (increased incidence, relatively unchanged mortality) and overtreatment (increased complication rates) of PE, but it remains unknown which small PEs are clinically significant.

Citation: Wiener RS, Schwartz LM, Woloshin S. Time trends in pulmonary embolism in the United States: evidence of overdiagnosis. Arch Intern Med. 2011;171(9):831-837.

Hospitalist-Physician Assistant Teams Associated with Longer Length of Stay, No Change in Mortality, Readmission Rates

Clinical question: Do length of stay (LOS), hospital mortality, or readmission rate change if hospitalists and physician assistants, or the traditional resident-hospitalist teams, provide the patient care?

Background: Resident work-hour limitations require new models of care for hospitalized patients. Many academic medical centers have hired physician assistants to work with hospitalists to provide care. Little is known about how these models affect such outcomes as LOS, inpatient mortality rates, and readmission rates.

Study design: Retrospective cohort.

Setting: A 430-bed urban academic medical center in Milwaukee.

Synopsis: Administrative data were gathered on 9,681 patients admitted to the general medical service. Of those enrolled, 2,171 were cared for by a hospitalist-physician assistant (H-PA) team, while resident-hospitalist teams cared for 7,510 patients. Patient assignment was dependent on time of admission but not on patient complexity. Patients admitted overnight after the resident team capped were assigned to the H-PA team the next morning, resulting in increased transitions of care for the H-PA team.

Adjusted analyses revealed a 6.45% increase in LOS for the H-PA team compared with the resident team. Charges, inpatient mortality, and readmission rates at seven, 14, and 30 days were unchanged. Subgroup analyses revealed smaller differences in LOS for H-PA teams and resident-hospitalist teams with the same hospitalist (LOS 5.44% higher, P=0.081).

Conclusions from this study are limited due to lack of randomization of assignment, the retrospective design, and the use of administrative data at one institution.

Bottom line: Hospitalist-PA teams might result in a slightly increased LOS compared with the traditional resident teams; however, inpatient mortality and readmission rates are similar.

Citation: Singh S, Fletcher KE, Schapira MM, et al. A comparison of outcomes of general medical inpatient care provided by a hospitalist-physician assistant model vs a traditional resident-based model. J Hosp Med. 2011;6:122-130.

Washing White Coats Does Not Lower MRSA Bacterial Contamination

Clinical question: Are clean, short-sleeved uniforms less likely to carry MRSA than regularly laundered long-sleeved white coats?

Background: Studies have shown that bacteria frequently colonize in physician garments. However, evidence that short-sleeved garments or newly laundered garments are less likely to be contaminated has been lacking. Despite the paucity of evidence, the British Department of Health barred the use of traditional white coats and long-sleeved garments in 2007.

Study design: Prospective, randomized, controlled trial.

Setting: Urban U.S. hospital.

Synopsis: Study authors randomized 100 internal-medicine residents and hospitalists to their own long-sleeved white coats or freshly laundered short-sleeved uniforms from August 2008 to November 2009. Swabs were taken from the sleeves of the white coats or uniform, the breast pocket, and the volar wrist surface of the dominant hand. Swabs were cultured for MRSA and for general colony count.

Results showed no significant difference in colony counts or MRSA colonization in any of the sites tested between the newly laundered uniforms and the white coats. Additionally, there was no effect in relation to the frequency of laundering the white coats. Notably, within three hours of donning freshly laundered uniforms, bacterial counts approached 50% of the total bacterial counts seen at eight hours.

Bottom line: Laundering of uniforms does not affect MRSA colonization rate or general bacterial burden on physician uniforms or skin surfaces, though the effect on nosocomial infection has not been established.

Citation: Burden M, Cervantes L, Weed D, Keniston A, Price CS, Albert RK. Newly cleaned physician uniforms and infrequently washed white coats have similar rates of bacterial contamination after an 8-hour workday: a randomized controlled trial. J Hosp Med. 2011;6:177-182.

Self-Reported Pain Severity Does Not Correlate with Heart Rate or Blood Pressure Measurements in Pre-Hospital Setting

Clinical question: Do measured vital signs, including heart rate, blood pressure, and respiratory rate, correlate with the degree of self-reported pain?

Background: Because pain often can be associated with alterations in autonomic tone, it has been hypothesized that alterations in vital signs will occur in patients who report pain.

Study design: Retrospective cohort study.

Setting: Pre-hospital in Melbourne, Australia.

Synopsis: The authors reviewed all ambulance patient care records for patients age >14 years with a Glasgow Coma Score (GCS) >12 transported to a hospital during a seven-day period in 2005. Patients were selected for analysis if their patient care record included an initial assessment of pain severity, as measured by a numeric rating scale (NRS), in which patients rate their pain from 0 to 10.

More than half of the 3,357 patients transported by paramedics during the period were included in this analysis (n=1286). There was no correlation between heart rate or systolic blood pressure with the degree of self-reported pain. Although an increased respiratory rate was statistically correlated with a higher rating of pain, this relationship was not clinically significant, as each one-point increase in the pain rating scale was associated with a 0.16-breaths-per-minute increase in the respiratory rate.

Limitations included the large number of records excluded from analysis because pain was not evaluated, as well as numerous unmeasured confounders, including active disease processes such as sepsis, that were not accounted for.

Bottom line: Severity of pain did not correlate with heart rate or systolic blood pressure in the pre-hospital setting.

Citation: Lord B, Woollard M. The reliability of vital signs in estimating pain severity among adult patients treated by paramedics. Emerg Med J. 2011;28:147-150.

Asymptomatic Perioperative Myocardial Infarction Is Common in Patients Undergoing Noncardiac Surgery

Clinical question: In patients undergoing noncardiac surgery, what is the incidence and clinical characteristics of perioperative myocardial infarction (MI)?

Background: Though millions of patients experience perioperative MI after noncardiac surgery, little is known about the characteristics and outcomes of these patients.

Study design: Cohort study.

Setting: One hundred ninety centers in 23 countries.

Synopsis: Using data from the 8,351 patients in the POISE (PeriOperative ISchemic Evaluation) trial, this study showed that perioperative MI occurred in 5% of patients; 65% were asymptomatic. Patients who experienced postoperative MI were older and had more cardiovascular risk factors when compared to those who did not. The 30-day mortality was higher in patients with a perioperative MI (11.6%) compared with those who did not (2.2%); the presence or absence of ischemic symptoms was not associated with mortality rate.

Of the 8.3% of patients who experienced an elevation in cardiac biomarkers but who did not meet the definition of MI, there was an increased risk of nonfatal cardiac arrest and nonacute coronary revascularization. Those in the highest quartile also had increased 30-day mortality.

Bottom line: Given the high proportion of asymptomatic MIs and isolated elevations in cardiac biomarkers and the association between these events and increased risk of death, hospitalists should consider routine monitoring of troponin in at-risk patients undergoing noncardiac surgery.

Citation: Devereaux PJ, Xavier D, Pogue J, et al. Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: a cohort study. Ann Intern Med. 2011;154(8):523-528.

Patients Prescribed Higher Opioid Doses Are at Increased Risk of Opioid Overdose Death

Clinical question: What is the association between opioid prescribing patterns and fatal opioid overdose?

Background: In the past 10 years, the rate of fatal overdose from opioid prescription for pain has more than doubled. Little is known about how the indications (substance abuse disorders, cancer-related pain, chronic pain, acute pain), maximal daily dose, and scheduling (standing, as-needed, or both) of opioid prescriptions relate to this increased risk.

Study design: Case-cohort study.

Setting: Veterans Health Administration (VHA) patients.

Synopsis: The VHA’s National Patient Care Database was used to randomly select a cohort of 154,684 nonhospice/nonpalliative-care patients who were prescribed opioids from 2004 to 2008. They were compared with 750 patients who were treated with prescription opioids who died from opioid overdose during this time.

Fatal opioid overdose was a rare event (0.04%), but risk increased with higher prescribed maximum daily morphine dose-equivalence, especially when greater than or equal to 50 mg/day in all subgroups (substance abuse, acute and chronic pain, and cancer). Fatal overdoses were higher in middle-aged white men with acute or chronic pain, substance abuse disorders, and other psychiatric illness. Patients with cancer were at increased risk of fatal overdose if they were prescribed as-needed opioids alone.

Treatment with both as-needed and standing opioids did not statistically affect risk of overdose death in any subgroup.

Bottom line: Although rare, risk of fatal opioid overdose in patients prescribed opiate medication increases with higher maximum prescribed daily dose.

Citation: Bohnert AS, Valenstein M, Bair M, et al. Association between opioid prescribing patterns and opioid overdose-related deaths. JAMA. 2011; 305:1315-1321.

Structured Interdisciplinary Rounds on Medical Teaching Unit Significantly Decrease Adverse Events

Clinical question: Do structured interdisciplinary rounds have an impact on the rate of adverse events?

Background: Many preventable adverse events occurring during hospitalization can be attributed to communication failures. Structured interdisciplinary rounds provide a format as well as a forum for team members to discuss patient care. Prior studies demonstrated improvements in collaboration; whether this translates to better patient care is not known.

Study design: Retrospective cohort using historic and concurrent control.

Setting: Tertiary-care teaching hospital in Chicago.

Synopsis: Structured interdisciplinary rounds, led by a nurse manager and medical director, and including nurses, residents, pharmacists, social workers, and case managers, were implemented on a medical teaching unit. New patients were discussed using a structured communication tool; existing patients were discussed in an unstructured format. Medical records were abstracted for 370 patients hospitalized after implementation of the intervention, equally divided between intervention and control units. One hundred eighty-five patients hospitalized on the intervention unit prior to the implementation of rounds served as a historic control.

Patients in the intervention unit had significantly lower rates of total adverse events (3.9 per 100 patient days in the intervention, compared with 7.2 and 7.7 per 100 patient days for the concurrent and historic control units, respectively), and preventable adverse events (0.9 per 100 patient days, compared with 2.8 and 2.1 per 100 patient days for the concurrent and historic controls, respectively).

Limitations of the study include lack of blinding of the medical record, slightly different patient populations in intervention and control groups, and the one-hospital setting, which could limit generalizability.

Bottom line: Structured interdisciplinary rounds might serve to improve communication between nurses, pharmacists, and physicians, resulting in decreases in adverse events.

Citation: O’Leary KJ, Buck R, Fligiel HM, et al. Structured interdisciplinary rounds in a medical teaching unit: improving patient safety. Arch Intern Med. 2011;171(7):678-684.

Pediatric HM Literature

Short-Course Antibiotic Therapy Effective for Bacterial Meningitis

Clinical question: Is five days of parenteral ceftriaxone as effective as 10 days for the treatment of bacterial meningitis in children?

Background: Morbidity and mortality in bacterial meningitis remain high, particularly in developing countries. Antibiotics are effective treatment, yet the optimal duration of treatment remains uncertain. Some data support a shorter duration of treatment (three to five days).

Study design: Multicountry, double-blind, placebo-controlled, randomized equivalence study.

Setting: Ten pediatric referral hospitals in Bangladesh, Egypt, Malawi, Pakistan, and Vietnam.

Synopsis: Children aged two months to 12 years with bacterial meningitis (due to Haemophilus influenza, Streptococcus pneumonia, Neisseria meningitidis, or culture-negative with indicative cerebrospinal fluid findings) and without complicating medical conditions were enrolled at participating centers. All children received 80 mg/kg to 100 mg/kg of parenteral ceftriaxone daily and a repeat lumbar puncture 48 to 72 hours after initiation of therapy.

Ultimately, 1,004 children without resistant organisms, persistently positive cultures, or suppurative complications were randomized on day five of therapy to placebo or continuance of ceftriaxone for five more days.

No bacteriologic failures (primary endpoint) were evident with either five or 10 days of treatment.

In addition, no statistically significant differences were found between the groups with respect to clinical treatment failure, hearing loss, neurological sequelae, or death. Secondary analysis by organism revealed similar results.

The primary limitation of this study is that it occurred in developing countries with a fair incidence of H. influenzae meningitis and a low rate of third-generation cephalosporin resistance.

However, pneumococcal and meningococcal disease remained prominent, and this study suggests that clinically stable patients might be treated with a shorter course of parenteral ceftriaxone therapy than currently is recommended.

Bottom line: Five days of ceftriaxone is as effective as 10 days for uncomplicated bacterial meningitis in children.

Citation: Molyneux E, Nizami SQ, Saha S, et al. 5 versus 10 days of treatment with ceftriaxone for bacterial meningitis in children: a double-blind randomised equivalence study. Lancet. 2011;377:1837-1845.

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

Clinical question: Is dalteparin better than unfractionated heparin at preventing venous thromboembolism (VTE) in ICU patients?

Background: VTE is an important health problem for critically ill patients, but prevention is possible, with both unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) superior to placebo in previous studies. Studies comparing UFH and LMWH in ICU patients have been inconclusive thus far.

Study design: Randomized controlled trial.

Setting: Sixty-seven ICUs in six countries from 2006 to 2010.

Synopsis: Researchers randomized 3,746 patients who met the enrollment criteria to either the LMWH dalteparin 5,000 units daily or UFH 5,000 IU twice daily. The drug was held if major bleeding occurred or the patient developed thrombocytopenia concerning heparin-induced thrombocytopenia (HIT). Patients were followed until discharge or death. VTE was evaluated by ultrasound two days after ICU admission and then twice weekly.

There was no difference in incidence of VTE in patients receiving dalteparin versus UFH [5.1% vs. 5.8%, HR 0.92 (CI 0.68-1.23), P=0.57]. Fewer pulmonary emboli occurred in the dalteparin group (1.3% vs. 2.3%, HR 0.51, P=0.01). There was no difference in major bleeding or HIT between groups.

Bottom line: Dalteparin and UFH were equally effective at preventing proximal VTE in ICU patients, but dalteparin prevented more pulmonary emboli.

Citation: The PROTECT investigators for the Canadian Critical Care Trials Group and the Australian and New Zealand Intensive Care Society Clinical Trials Group. Dalteparin versus unfractionated heparin in critically ill patients. N Engl J Med. 2010;364(14):1305-1314.

For more physician reviews of HM-related research, visit our website.

Clinical question: Is dalteparin better than unfractionated heparin at preventing venous thromboembolism (VTE) in ICU patients?

Background: VTE is an important health problem for critically ill patients, but prevention is possible, with both unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) superior to placebo in previous studies. Studies comparing UFH and LMWH in ICU patients have been inconclusive thus far.

Study design: Randomized controlled trial.

Setting: Sixty-seven ICUs in six countries from 2006 to 2010.

Synopsis: Researchers randomized 3,746 patients who met the enrollment criteria to either the LMWH dalteparin 5,000 units daily or UFH 5,000 IU twice daily. The drug was held if major bleeding occurred or the patient developed thrombocytopenia concerning heparin-induced thrombocytopenia (HIT). Patients were followed until discharge or death. VTE was evaluated by ultrasound two days after ICU admission and then twice weekly.

There was no difference in incidence of VTE in patients receiving dalteparin versus UFH [5.1% vs. 5.8%, HR 0.92 (CI 0.68-1.23), P=0.57]. Fewer pulmonary emboli occurred in the dalteparin group (1.3% vs. 2.3%, HR 0.51, P=0.01). There was no difference in major bleeding or HIT between groups.

Bottom line: Dalteparin and UFH were equally effective at preventing proximal VTE in ICU patients, but dalteparin prevented more pulmonary emboli.

Citation: The PROTECT investigators for the Canadian Critical Care Trials Group and the Australian and New Zealand Intensive Care Society Clinical Trials Group. Dalteparin versus unfractionated heparin in critically ill patients. N Engl J Med. 2010;364(14):1305-1314.

For more physician reviews of HM-related research, visit our website.

Clinical question: Is dalteparin better than unfractionated heparin at preventing venous thromboembolism (VTE) in ICU patients?

Background: VTE is an important health problem for critically ill patients, but prevention is possible, with both unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) superior to placebo in previous studies. Studies comparing UFH and LMWH in ICU patients have been inconclusive thus far.

Study design: Randomized controlled trial.

Setting: Sixty-seven ICUs in six countries from 2006 to 2010.

Synopsis: Researchers randomized 3,746 patients who met the enrollment criteria to either the LMWH dalteparin 5,000 units daily or UFH 5,000 IU twice daily. The drug was held if major bleeding occurred or the patient developed thrombocytopenia concerning heparin-induced thrombocytopenia (HIT). Patients were followed until discharge or death. VTE was evaluated by ultrasound two days after ICU admission and then twice weekly.

There was no difference in incidence of VTE in patients receiving dalteparin versus UFH [5.1% vs. 5.8%, HR 0.92 (CI 0.68-1.23), P=0.57]. Fewer pulmonary emboli occurred in the dalteparin group (1.3% vs. 2.3%, HR 0.51, P=0.01). There was no difference in major bleeding or HIT between groups.

Bottom line: Dalteparin and UFH were equally effective at preventing proximal VTE in ICU patients, but dalteparin prevented more pulmonary emboli.

Citation: The PROTECT investigators for the Canadian Critical Care Trials Group and the Australian and New Zealand Intensive Care Society Clinical Trials Group. Dalteparin versus unfractionated heparin in critically ill patients. N Engl J Med. 2010;364(14):1305-1314.

For more physician reviews of HM-related research, visit our website.