Clinical Review

Homeless individuals and IV drug users are susceptible to hepatitis A, B, and C infections, and co-infection with these diseases may complicate treatment and result in poor medical outcomes.¹ Vaccination offers the best protection against hepatitis A and B, particularly among high-risk populations.^2,3 Immunization against hepatitis A and B is of even greater importance for patients with hepatitis C, because there is no specific hepatitis C vaccine, and concomitant infections of B with C are damaging to the liver.⁴

Veterans have a rate of hepatitis C infection that is 3 times that of the general population.⁵ Some evidence exists that veterans with serious mental illness (SMI) have a higher rate of hepatitis C infection relative to patients without SMI. Co-occurring substance abuse may add another layer of vulnerability to hepatitis C infection, particularly for homeless veterans.^5-7

Mental Health and Primary Care Integration

Substance abuse and dual-diagnosis treatment programs (ie, those programs that treat both substance abuse and co-occurring serious mental health problems, such as bipolar disorder, severe major depressive disorder, psychotic disorders, and posttraumatic stress disorder [PTSD]) that have integrated mental health and primary care into their treatment programs may offer a window of opportunity for risk-reducing interventions. These interventions include testing and education of patients regarding infectious diseases, such as viral hepatitis and HIV, and completion of the hepatitis A/B immunization series.

The James A. Haley Veterans’ Hospital (JAHVH) in Tampa, Florida, has demonstrated some limited success in the past with integrating a standard dosing schedule for hepatitis A/B vaccination into its substance abuse treatment program (SATP), though recent evidence points to more promising results using an accelerated regimen as indicated by a high completion rate for hepatitis B vaccination in a methadone clinic.^8,9 A relatively low proportion of SATPs in the U.S. provide testing, education, or vaccination for hepatitis A and B, especially considering the public health importance of controlling these diseases in the substance abusing populations.^10,11

Related: Combination Pill Approved for HCV

In 1999, a primary care team was added to the alcohol and drug abuse treatment program at JAHVH.In 2005, the nurses in the program began scheduling vaccinations and screening patients for medical and psychiatric issues, pain, hypertension, diabetes, hepatitis C, alcohol use, depression, PTSD, prostate and colorectal cancers.¹² Such a multidisciplinary approach provides many treatment advantages for patients and may save lives.¹³

Even with a multidisciplinary approach, the nurses found it difficult to provide adequate hepatitis A/B immunization within the 3- to 6-week intensive SATP, because standard immunization dosing regimens are spread over 6 months.¹⁴ As with all types of immunizations, long dosing schedules may reduce patient adherence and result in inadequate seroprotection.¹⁵ Thus, there is a need to provide a completed immunization series in a more expeditious fashion, and an accelerated dosing regimen makes that possible.^15,16

Hepatitis A/B Vaccination

Twinrix (GlaxoSmithKline, Brentford, United Kingdom) is a vaccine that provides dual immunization for hepatitis A and B. Whereas the standard vaccination schedule takes 6 months to complete, the accelerated dosing schedule can be used to complete the first 3 doses in less than a month. The accelerated dosing schedule was incorporated into the JAHVH clinic to capture as many patients as possible in the 3- to 6-week time frame: The first dose is administered and followed by a second dose 7 days later. The third dose is administered 21 to 30 days after the first dose. Twelve months after the first dose, a booster dose is given. 

After the first 3 accelerated doses, > 98% of patients show a sustained immune response to hepatitis A, and > 63% demonstrate immunity to hepatitis B. If a 12-month booster injection is given, 100% of patients may receive immunity to hepatitis A and > 96% may have immunity to hepatitis B.¹⁶ Another study of the combined vaccine showed even greater seroprotection for hepatitis A and B after only 1 month, 100% and 82%, respectively.¹⁷

Related: Viral Hepatitis Awareness

This JAHVH retrospective feasibility study describes a risk-reduction program for hepatitis A/B prevention that was implemented within a 3- to 4-week intensive outpatient SATP and a 6-week dual-diagnosis treatment program. The study includes the development and implementation of the program, designed to vaccinate patients using the accelerated Twinrix schedule. To ascertain the feasibility of this vaccination approach, historical medical records were used to describe and examine the vaccination initiation and follow-up rates of the treatment program participants who received the hepatitis A/B immunization series during their intensive SATP.

Study Design

A retrospective review of medical records was conducted for all participants who were admitted to the intensive JAHVH SATP between October 1, 2008, and September 30, 2009. This study was reviewed and approved by the JAHVH research and development committee and its associated University of South Florida institutional review board. Informed consent to participate was not obtained, because the study was retrospective.

Patient Identification and Education

All program participants were offered testing for HIV and hepatitis A, B, and C. Program participants were educated about hepatitis and HIV transmission, as well as about the long-term effects of continued substance abuse on the progression of hepatitis C. Education about hepatitis, HIV, and substance abuse was provided in a group setting by a member of the program’s nursing staff. One-on-one risk education counseling was also provided when requested or otherwise indicated.

Laboratory testing was performed following each participant’s initial physical examination (within 3 to 5 days of program admission), and the nursing staff reviewed the results before vaccination. Explanation of laboratory results and an individualized immunization regimen were provided to each participant. On review of participants’ laboratory results, those with seroconversion of both hepatitis A and B were not given the combined immunization. Participants who had seroconversion of hepatitis A were offered the hepatitis B vaccination series, and vice versa.

Immunization Process

Participants who lacked prior immunization for hepatitis A and B and had no seroconversion of either hepatitis A or B were offered vaccination. Some patients declined vaccination, even though they were eligible. Their reasons were not formally assessed.

Related: Nivolumab Approved for Expanded Indication

Patients who accepted the vaccination were given the accelerated regimen.¹⁶ Participants were educated on the importance of compliance with the vaccination series and provided with follow-up immunization dates and a reminder for the 1-year booster vaccine. The immunizations were ordered by the program’s primary care NP and administered by a licensed practical nurse. The nurse who administered the injections took responsibility for scheduling the patients for all their subsequent injections, including the 1-year booster.

Follow-up Care

If the third injection was not completed before discharge, patients were given a follow-up appointment with the nurse if they remained in the JAHVH service area. If they were leaving the area, they were given instructions on how to follow-up at another VA facility to continue their immunization schedule. A note was written in the electronic medical record documenting their abbreviated hepatitis A/B immunization schedule, which could be accessed by other providers at other VA facilities. Patients who did not show up for any follow-up appointments (third injection or the 1-year booster injection) were contacted and reminded about the importance of completing the immunization series and to schedule an appointment.

Statistical Analysis

All data were analyzed using IBM Statistical Package for the Social Sciences (IBM SPSS, Armonk, New York) with a focus on identifying differences between vaccination-eligible patients (n = 269) who did (n = 128) and did not (n = 141) initiate the immunization schedule during the treatment program. Chi-square and Fisher exact tests were used to assess statistical differences in initiation of the immunization schedule related to categoric variables (ie, marital status, race, history of IV drug abuse, cigarette smoking status, housing status, legal status, history of combat, having a psychiatric or medical diagnosis, and program track). Independent sample t tests were used to test for differences between these 2 groups on the continuous variables, including age, number of previous treatment programs, Global Assessment of Functioning score, severity of smoking dependence as measured by the Fagerström Test for Nicotine Dependence, and the Addiction Severity Index scales.^18-20

Results

The sample consisted of 284 successive admissions to an intensive outpatient program for veterans with substance use disorders. About one-third of the patients were homeless at the time of admission to the treatment, and 87% required contracted housing while completing treatment for reasons related to lack of housing, transportation, clinical necessity, or a combination of those factors (Table 1). The most common substance problems were alcohol and cocaine dependence, and 21% (n = 59) of the patients acknowledged a history of IV drug use during their initial psychiatric evaluation. Seventy percent were dually diagnosed with some other Axis I disorder, and 40% had a history of serious mental illness. More than one-fourth (n = 77) of the patients admitted to the intensive outpatient SATP were seropositive for hepatitis A, B and/or C, and the most common hepatitis diagnosis was hepatitis C (n = 71).

Accelerated Immunization Regimen

Patients were eligible to receive the accelerated vaccination schedule only if they had no prior immunization for hepatitis A or B and if they had no seroconversion for either hepatitis A or B. Six people had hepatitis B alone, 7 had hepatitis B and C, 1 had hepatitis A and C, and 1 had all 3 (Table 2). Thus, 15 participants were ineligible to receive the accelerated hepatitis A/B immunization. Chi-square, Fisher exact, and independent sample t tests showed that among those who were vaccination-eligible (269), there were no significant differences in any of the demographic or clinical characteristics between those who initiated the vaccination schedule and those who did not. Among those who completed the first 3 vaccine injections, those who received the 1-year booster injection (54) did not differ (on any demographic or clinical variables) from those who did not (58).

Nearly half (48%) of all the eligible patients admitted to the program began the accelerated immunization schedule for hepatitis A and B.  Of those, 88% completed the first 3 injections in the series. Among the patients who received the first 3 injections, 48% received the 1-year booster injection—a 20% completion rate for the vaccination-eligible sample overall (Table 3).

Of the 74 patients who did not complete their vaccinations once initiating the accelerated schedule, the most common reason identified was that the patient moved away (37), or no reason could be identified (33). It was uncommon for a patient not to complete the vaccination schedule because of terminating treatment prematurely (4).

Compared with the vaccine-eligible patients without hepatitis C (207), patients with hepatitis C were less likely to receive any vaccination injections (Table 3). Specifically, 51% of the vaccination-eligible patients who did not have hepatitis C began the vaccination regimen. However, only 22 patients with hepatitis C, or
35% of all vaccination-eligible patients with hepatitis C, began the vaccination regimen. Patients with hepatitis C were also less likely than those without hepatitis C to complete the first 3 injections of the vaccination series once they had initiated it (77%, vs 90%, respectively). This difference continued to be apparent at the time of the 12-month booster injection. Only 35% of vaccine-eligible individuals with hepatitis C received the 12-month booster injection, whereas 51% of vaccination-eligible individuals without hepatitis C received the 12-month booster injection. As with the sample overall, the most common reason patients with hepatitis C did not complete the vaccination regimen was because they moved away (9), followed by no identified reason (5), and premature termination of treatment (2).

Discussion

Individuals abusing alcohol and drugs have an increased vulnerability for infectious diseases, and homeless veterans with substance use disorders may be at a particularly heightened risk.^21,22 This study describes a sample of veterans, many were homeless and most were dually diagnosed, in an intensive outpatient SATP that offered an accelerated dosing regimen for hepatitis A and B vaccination. Almost half (48%) of the vaccination-eligible patients began the accelerated regimen for hepatitis A/B vaccination. Moreover, 88% of those who started the vaccination regimen received the first 3 injections of the series, thus possibly conferring substantial immunity to hepatitis A and B and demonstrating the feasibility of an accelerated vaccination schedule in an intensive outpatient SATP.

It is especially important to demonstrate the successful integration of a hepatitis screening and immunization program within a SATP, given that many such programs do not offer screening or immunization for hepatitis, even though substance abusers are disproportionately affected by the disease and contribute greatly to the ongoing hepatitis epidemic.^10,11 This study’s results were in line with another study of rapid vaccination for hepatitis B in IV drug users being treated in a methadone clinic, where 83% of the vaccination initiators completed the first 3 injections of the series.⁹

Unvaccinated Patients

The treatment team in the current study seemed to be less effective at reaching the subset of vaccination-eligible veterans with hepatitis C (almost one-quarter of the sample) in order to administer the accelerated vaccination schedule, as indicated by the lower rate of vaccination initiation as well as a lower rate of completion of the vaccination series among those patients. This replicates a finding from another study that also indicated a low rate of hepatitis A and B vaccination among patients with hepatitis C.²³ Only 35% of the vaccination-eligible patients with hepatitis C in the current study initiated the vaccination series, compared with 51% of the patients without hepatitis C. However, the rate of completion of the first 3 injections of the series in the hepatitis C group was respectably high (77%), especially given the high relapse rate and psychosocial instability of individuals with addictive disorders. Initiation seems to be a bigger obstacle than completion of at least the first 3 injections of the vaccination series in both patients with and without hepatitis C.

The study investigators did not formally assess the reasons that more than half the patients in the study did not begin the vaccination series, but anecdotal evidence from the nurses indicated that many patients were afraid of needles. In addition, other patients felt that they simply did not need the vaccination. Some also insisted that they had already had the vaccination despite a blood test showing no evidence for either hepatitis A or B immunization.

Although the nursing team provided group and individual risk-based education as well as information about the effects of continued substance abuse on hepatitis C, it is possible that patients still underestimated their own risk of hepatitis infection and its consequences, or perhaps the information was simply not retained.²⁴

Patient Education

A recent study showed that there is a positive relationship between the amount of hepatitis counseling received and knowledge of hepatitis.²⁵ Possibly, increased intensity of education efforts may make an impact on initiation rates. Encouragingly, there is also evidence that prompting people to predict their future vaccination behavior may increase vaccination initiation rates despite a high-degree of short-term barriers, such as perceived pain or inconvenience.²⁶ A brief intervention to induce people to formulate their future intentions would be relatively easy to incorporate into a vaccination program, and the study team is considering options for this to improve vaccination initiation rates.

Patients can expect to achieve substantial immunity from hepatitis A and, to a lesser degree, hepatitis B after completing the first 3 injections of the series, although the best seroprotection from both is obtained by completing the 12-month booster injection as well.¹⁷ Overall, about half of all patients who completed the first 3 injections returned for the booster shot, but only 35% of the patients with hepatitis C did so. The most common known cause of any patient not receiving the booster was movement out of the geographic area. However, much of the time the investigators were unable to determine the reasons patients did not return for the booster shot.

Medication adherence is a difficult problem with vaccination in high-risk samples, although Stitzer and colleagues found a significant improvement in follow-up for a 6-month vaccination protocol by using monetary incentives.²⁷ In addition to ensuring medication adherence, it would also be of value for future immunization efforts to include testing to assess whether seroconversion has occurred once the vaccinations are complete, which is the ultimate measure of the success of a vaccination program. Most patients in the current study did not receive such testing at the completion of their vaccination schedules, and thus, seroconversion rates could not be determined. However, existing studies suggest high rates of seroprotection after the first 3 doses of the combined vaccine.^10,17

Limitations

The retrospective nature of the study is its most significant limitation. Any conclusions about the results must be made with caution. However, this design allowed for a naturalistic and potentially generalizable investigation into the application of a vaccination program in a real-world treatment setting. As such, the investigators were able to demonstrate the feasibility of conducting a rapid vaccination program within a 3- to 6-week SATP.

The retrospective nature of the study also limited a full investigation into the reasons behind the lack of vaccination initiation and vaccination noncompletion among the study’s treatment population, especially with regard to the follow-up booster injection. Initial statistical comparisons of initiators and noninitiators and completers and noncompleters showed no significant statistical differences between the groups. Future prospective designs should take into account the need to successfully initiate and complete vaccinations for all eligible patients and include assessment measures to determine the specific reasons that patients did not initiate or complete their vaccinations.

Conclusions

Many patients began and completed the accelerated vaccination schedule for hepatitis A and B in the context of a 3- to 6-week SATP at JAHVH. The overall vaccination rate, including the 12-month booster injection, was one-fifth of the entire vaccination-eligible sample. Additionally, 88% of the vaccination-eligible patients who began the vaccination schedule (or 42% of the whole sample) completed at least the first 3 doses, which may confer substantial immunity from hepatitis A and B. For reasons not entirely clear, a little less than half the vaccination-eligible patients began the vaccination schedule, and only about 50% of those returned to receive their 12-month booster injection. Future prospective studies may be able to determine barriers to both the initiation of and adherence to the vaccination protocol.

The results of this study are also a testament to having primary care nursing staff available and actively involved in the care of patients in a SATP. It seems likely that additional interventions might be needed for outreach to and retention of patients in need of vaccination for hepatitis A and B, and particularly those patients with hepatitis C. It is important to find ways to increase the rates of 12-month booster vaccinations, both for veterans who continue to receive services at JAHVH and for those who transfer care to other VA facilities. Finally, testing to confirm serologic immunity to hepatitis A and hepatitis B would be the next step in the effort to eliminate the risk of hepatitis A and hepatitis B and minimize additional harm for those with hepatitis C in the population receiving treatment for addictive disorders.

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Homeless individuals and IV drug users are susceptible to hepatitis A, B, and C infections, and co-infection with these diseases may complicate treatment and result in poor medical outcomes.¹ Vaccination offers the best protection against hepatitis A and B, particularly among high-risk populations.^2,3 Immunization against hepatitis A and B is of even greater importance for patients with hepatitis C, because there is no specific hepatitis C vaccine, and concomitant infections of B with C are damaging to the liver.⁴

Veterans have a rate of hepatitis C infection that is 3 times that of the general population.⁵ Some evidence exists that veterans with serious mental illness (SMI) have a higher rate of hepatitis C infection relative to patients without SMI. Co-occurring substance abuse may add another layer of vulnerability to hepatitis C infection, particularly for homeless veterans.^5-7

Mental Health and Primary Care Integration

Substance abuse and dual-diagnosis treatment programs (ie, those programs that treat both substance abuse and co-occurring serious mental health problems, such as bipolar disorder, severe major depressive disorder, psychotic disorders, and posttraumatic stress disorder [PTSD]) that have integrated mental health and primary care into their treatment programs may offer a window of opportunity for risk-reducing interventions. These interventions include testing and education of patients regarding infectious diseases, such as viral hepatitis and HIV, and completion of the hepatitis A/B immunization series.

The James A. Haley Veterans’ Hospital (JAHVH) in Tampa, Florida, has demonstrated some limited success in the past with integrating a standard dosing schedule for hepatitis A/B vaccination into its substance abuse treatment program (SATP), though recent evidence points to more promising results using an accelerated regimen as indicated by a high completion rate for hepatitis B vaccination in a methadone clinic.^8,9 A relatively low proportion of SATPs in the U.S. provide testing, education, or vaccination for hepatitis A and B, especially considering the public health importance of controlling these diseases in the substance abusing populations.^10,11

Related: Combination Pill Approved for HCV

In 1999, a primary care team was added to the alcohol and drug abuse treatment program at JAHVH.In 2005, the nurses in the program began scheduling vaccinations and screening patients for medical and psychiatric issues, pain, hypertension, diabetes, hepatitis C, alcohol use, depression, PTSD, prostate and colorectal cancers.¹² Such a multidisciplinary approach provides many treatment advantages for patients and may save lives.¹³

Even with a multidisciplinary approach, the nurses found it difficult to provide adequate hepatitis A/B immunization within the 3- to 6-week intensive SATP, because standard immunization dosing regimens are spread over 6 months.¹⁴ As with all types of immunizations, long dosing schedules may reduce patient adherence and result in inadequate seroprotection.¹⁵ Thus, there is a need to provide a completed immunization series in a more expeditious fashion, and an accelerated dosing regimen makes that possible.^15,16

Hepatitis A/B Vaccination

Twinrix (GlaxoSmithKline, Brentford, United Kingdom) is a vaccine that provides dual immunization for hepatitis A and B. Whereas the standard vaccination schedule takes 6 months to complete, the accelerated dosing schedule can be used to complete the first 3 doses in less than a month. The accelerated dosing schedule was incorporated into the JAHVH clinic to capture as many patients as possible in the 3- to 6-week time frame: The first dose is administered and followed by a second dose 7 days later. The third dose is administered 21 to 30 days after the first dose. Twelve months after the first dose, a booster dose is given. 

After the first 3 accelerated doses, > 98% of patients show a sustained immune response to hepatitis A, and > 63% demonstrate immunity to hepatitis B. If a 12-month booster injection is given, 100% of patients may receive immunity to hepatitis A and > 96% may have immunity to hepatitis B.¹⁶ Another study of the combined vaccine showed even greater seroprotection for hepatitis A and B after only 1 month, 100% and 82%, respectively.¹⁷

Related: Viral Hepatitis Awareness

This JAHVH retrospective feasibility study describes a risk-reduction program for hepatitis A/B prevention that was implemented within a 3- to 4-week intensive outpatient SATP and a 6-week dual-diagnosis treatment program. The study includes the development and implementation of the program, designed to vaccinate patients using the accelerated Twinrix schedule. To ascertain the feasibility of this vaccination approach, historical medical records were used to describe and examine the vaccination initiation and follow-up rates of the treatment program participants who received the hepatitis A/B immunization series during their intensive SATP.

Study Design

A retrospective review of medical records was conducted for all participants who were admitted to the intensive JAHVH SATP between October 1, 2008, and September 30, 2009. This study was reviewed and approved by the JAHVH research and development committee and its associated University of South Florida institutional review board. Informed consent to participate was not obtained, because the study was retrospective.

Patient Identification and Education

All program participants were offered testing for HIV and hepatitis A, B, and C. Program participants were educated about hepatitis and HIV transmission, as well as about the long-term effects of continued substance abuse on the progression of hepatitis C. Education about hepatitis, HIV, and substance abuse was provided in a group setting by a member of the program’s nursing staff. One-on-one risk education counseling was also provided when requested or otherwise indicated.

Laboratory testing was performed following each participant’s initial physical examination (within 3 to 5 days of program admission), and the nursing staff reviewed the results before vaccination. Explanation of laboratory results and an individualized immunization regimen were provided to each participant. On review of participants’ laboratory results, those with seroconversion of both hepatitis A and B were not given the combined immunization. Participants who had seroconversion of hepatitis A were offered the hepatitis B vaccination series, and vice versa.

Immunization Process

Participants who lacked prior immunization for hepatitis A and B and had no seroconversion of either hepatitis A or B were offered vaccination. Some patients declined vaccination, even though they were eligible. Their reasons were not formally assessed.

Related: Nivolumab Approved for Expanded Indication

Patients who accepted the vaccination were given the accelerated regimen.¹⁶ Participants were educated on the importance of compliance with the vaccination series and provided with follow-up immunization dates and a reminder for the 1-year booster vaccine. The immunizations were ordered by the program’s primary care NP and administered by a licensed practical nurse. The nurse who administered the injections took responsibility for scheduling the patients for all their subsequent injections, including the 1-year booster.

Follow-up Care

If the third injection was not completed before discharge, patients were given a follow-up appointment with the nurse if they remained in the JAHVH service area. If they were leaving the area, they were given instructions on how to follow-up at another VA facility to continue their immunization schedule. A note was written in the electronic medical record documenting their abbreviated hepatitis A/B immunization schedule, which could be accessed by other providers at other VA facilities. Patients who did not show up for any follow-up appointments (third injection or the 1-year booster injection) were contacted and reminded about the importance of completing the immunization series and to schedule an appointment.

Statistical Analysis

All data were analyzed using IBM Statistical Package for the Social Sciences (IBM SPSS, Armonk, New York) with a focus on identifying differences between vaccination-eligible patients (n = 269) who did (n = 128) and did not (n = 141) initiate the immunization schedule during the treatment program. Chi-square and Fisher exact tests were used to assess statistical differences in initiation of the immunization schedule related to categoric variables (ie, marital status, race, history of IV drug abuse, cigarette smoking status, housing status, legal status, history of combat, having a psychiatric or medical diagnosis, and program track). Independent sample t tests were used to test for differences between these 2 groups on the continuous variables, including age, number of previous treatment programs, Global Assessment of Functioning score, severity of smoking dependence as measured by the Fagerström Test for Nicotine Dependence, and the Addiction Severity Index scales.^18-20

Results

The sample consisted of 284 successive admissions to an intensive outpatient program for veterans with substance use disorders. About one-third of the patients were homeless at the time of admission to the treatment, and 87% required contracted housing while completing treatment for reasons related to lack of housing, transportation, clinical necessity, or a combination of those factors (Table 1). The most common substance problems were alcohol and cocaine dependence, and 21% (n = 59) of the patients acknowledged a history of IV drug use during their initial psychiatric evaluation. Seventy percent were dually diagnosed with some other Axis I disorder, and 40% had a history of serious mental illness. More than one-fourth (n = 77) of the patients admitted to the intensive outpatient SATP were seropositive for hepatitis A, B and/or C, and the most common hepatitis diagnosis was hepatitis C (n = 71).

Accelerated Immunization Regimen

Patients were eligible to receive the accelerated vaccination schedule only if they had no prior immunization for hepatitis A or B and if they had no seroconversion for either hepatitis A or B. Six people had hepatitis B alone, 7 had hepatitis B and C, 1 had hepatitis A and C, and 1 had all 3 (Table 2). Thus, 15 participants were ineligible to receive the accelerated hepatitis A/B immunization. Chi-square, Fisher exact, and independent sample t tests showed that among those who were vaccination-eligible (269), there were no significant differences in any of the demographic or clinical characteristics between those who initiated the vaccination schedule and those who did not. Among those who completed the first 3 vaccine injections, those who received the 1-year booster injection (54) did not differ (on any demographic or clinical variables) from those who did not (58).

Nearly half (48%) of all the eligible patients admitted to the program began the accelerated immunization schedule for hepatitis A and B.  Of those, 88% completed the first 3 injections in the series. Among the patients who received the first 3 injections, 48% received the 1-year booster injection—a 20% completion rate for the vaccination-eligible sample overall (Table 3).

Of the 74 patients who did not complete their vaccinations once initiating the accelerated schedule, the most common reason identified was that the patient moved away (37), or no reason could be identified (33). It was uncommon for a patient not to complete the vaccination schedule because of terminating treatment prematurely (4).

Compared with the vaccine-eligible patients without hepatitis C (207), patients with hepatitis C were less likely to receive any vaccination injections (Table 3). Specifically, 51% of the vaccination-eligible patients who did not have hepatitis C began the vaccination regimen. However, only 22 patients with hepatitis C, or
35% of all vaccination-eligible patients with hepatitis C, began the vaccination regimen. Patients with hepatitis C were also less likely than those without hepatitis C to complete the first 3 injections of the vaccination series once they had initiated it (77%, vs 90%, respectively). This difference continued to be apparent at the time of the 12-month booster injection. Only 35% of vaccine-eligible individuals with hepatitis C received the 12-month booster injection, whereas 51% of vaccination-eligible individuals without hepatitis C received the 12-month booster injection. As with the sample overall, the most common reason patients with hepatitis C did not complete the vaccination regimen was because they moved away (9), followed by no identified reason (5), and premature termination of treatment (2).

Discussion

Individuals abusing alcohol and drugs have an increased vulnerability for infectious diseases, and homeless veterans with substance use disorders may be at a particularly heightened risk.^21,22 This study describes a sample of veterans, many were homeless and most were dually diagnosed, in an intensive outpatient SATP that offered an accelerated dosing regimen for hepatitis A and B vaccination. Almost half (48%) of the vaccination-eligible patients began the accelerated regimen for hepatitis A/B vaccination. Moreover, 88% of those who started the vaccination regimen received the first 3 injections of the series, thus possibly conferring substantial immunity to hepatitis A and B and demonstrating the feasibility of an accelerated vaccination schedule in an intensive outpatient SATP.

It is especially important to demonstrate the successful integration of a hepatitis screening and immunization program within a SATP, given that many such programs do not offer screening or immunization for hepatitis, even though substance abusers are disproportionately affected by the disease and contribute greatly to the ongoing hepatitis epidemic.^10,11 This study’s results were in line with another study of rapid vaccination for hepatitis B in IV drug users being treated in a methadone clinic, where 83% of the vaccination initiators completed the first 3 injections of the series.⁹

Unvaccinated Patients

The treatment team in the current study seemed to be less effective at reaching the subset of vaccination-eligible veterans with hepatitis C (almost one-quarter of the sample) in order to administer the accelerated vaccination schedule, as indicated by the lower rate of vaccination initiation as well as a lower rate of completion of the vaccination series among those patients. This replicates a finding from another study that also indicated a low rate of hepatitis A and B vaccination among patients with hepatitis C.²³ Only 35% of the vaccination-eligible patients with hepatitis C in the current study initiated the vaccination series, compared with 51% of the patients without hepatitis C. However, the rate of completion of the first 3 injections of the series in the hepatitis C group was respectably high (77%), especially given the high relapse rate and psychosocial instability of individuals with addictive disorders. Initiation seems to be a bigger obstacle than completion of at least the first 3 injections of the vaccination series in both patients with and without hepatitis C.

The study investigators did not formally assess the reasons that more than half the patients in the study did not begin the vaccination series, but anecdotal evidence from the nurses indicated that many patients were afraid of needles. In addition, other patients felt that they simply did not need the vaccination. Some also insisted that they had already had the vaccination despite a blood test showing no evidence for either hepatitis A or B immunization.

Although the nursing team provided group and individual risk-based education as well as information about the effects of continued substance abuse on hepatitis C, it is possible that patients still underestimated their own risk of hepatitis infection and its consequences, or perhaps the information was simply not retained.²⁴

Patient Education

A recent study showed that there is a positive relationship between the amount of hepatitis counseling received and knowledge of hepatitis.²⁵ Possibly, increased intensity of education efforts may make an impact on initiation rates. Encouragingly, there is also evidence that prompting people to predict their future vaccination behavior may increase vaccination initiation rates despite a high-degree of short-term barriers, such as perceived pain or inconvenience.²⁶ A brief intervention to induce people to formulate their future intentions would be relatively easy to incorporate into a vaccination program, and the study team is considering options for this to improve vaccination initiation rates.

Patients can expect to achieve substantial immunity from hepatitis A and, to a lesser degree, hepatitis B after completing the first 3 injections of the series, although the best seroprotection from both is obtained by completing the 12-month booster injection as well.¹⁷ Overall, about half of all patients who completed the first 3 injections returned for the booster shot, but only 35% of the patients with hepatitis C did so. The most common known cause of any patient not receiving the booster was movement out of the geographic area. However, much of the time the investigators were unable to determine the reasons patients did not return for the booster shot.

Medication adherence is a difficult problem with vaccination in high-risk samples, although Stitzer and colleagues found a significant improvement in follow-up for a 6-month vaccination protocol by using monetary incentives.²⁷ In addition to ensuring medication adherence, it would also be of value for future immunization efforts to include testing to assess whether seroconversion has occurred once the vaccinations are complete, which is the ultimate measure of the success of a vaccination program. Most patients in the current study did not receive such testing at the completion of their vaccination schedules, and thus, seroconversion rates could not be determined. However, existing studies suggest high rates of seroprotection after the first 3 doses of the combined vaccine.^10,17

Limitations

The retrospective nature of the study is its most significant limitation. Any conclusions about the results must be made with caution. However, this design allowed for a naturalistic and potentially generalizable investigation into the application of a vaccination program in a real-world treatment setting. As such, the investigators were able to demonstrate the feasibility of conducting a rapid vaccination program within a 3- to 6-week SATP.

The retrospective nature of the study also limited a full investigation into the reasons behind the lack of vaccination initiation and vaccination noncompletion among the study’s treatment population, especially with regard to the follow-up booster injection. Initial statistical comparisons of initiators and noninitiators and completers and noncompleters showed no significant statistical differences between the groups. Future prospective designs should take into account the need to successfully initiate and complete vaccinations for all eligible patients and include assessment measures to determine the specific reasons that patients did not initiate or complete their vaccinations.

Conclusions

Many patients began and completed the accelerated vaccination schedule for hepatitis A and B in the context of a 3- to 6-week SATP at JAHVH. The overall vaccination rate, including the 12-month booster injection, was one-fifth of the entire vaccination-eligible sample. Additionally, 88% of the vaccination-eligible patients who began the vaccination schedule (or 42% of the whole sample) completed at least the first 3 doses, which may confer substantial immunity from hepatitis A and B. For reasons not entirely clear, a little less than half the vaccination-eligible patients began the vaccination schedule, and only about 50% of those returned to receive their 12-month booster injection. Future prospective studies may be able to determine barriers to both the initiation of and adherence to the vaccination protocol.

The results of this study are also a testament to having primary care nursing staff available and actively involved in the care of patients in a SATP. It seems likely that additional interventions might be needed for outreach to and retention of patients in need of vaccination for hepatitis A and B, and particularly those patients with hepatitis C. It is important to find ways to increase the rates of 12-month booster vaccinations, both for veterans who continue to receive services at JAHVH and for those who transfer care to other VA facilities. Finally, testing to confirm serologic immunity to hepatitis A and hepatitis B would be the next step in the effort to eliminate the risk of hepatitis A and hepatitis B and minimize additional harm for those with hepatitis C in the population receiving treatment for addictive disorders.

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Homeless individuals and IV drug users are susceptible to hepatitis A, B, and C infections, and co-infection with these diseases may complicate treatment and result in poor medical outcomes.¹ Vaccination offers the best protection against hepatitis A and B, particularly among high-risk populations.^2,3 Immunization against hepatitis A and B is of even greater importance for patients with hepatitis C, because there is no specific hepatitis C vaccine, and concomitant infections of B with C are damaging to the liver.⁴

Veterans have a rate of hepatitis C infection that is 3 times that of the general population.⁵ Some evidence exists that veterans with serious mental illness (SMI) have a higher rate of hepatitis C infection relative to patients without SMI. Co-occurring substance abuse may add another layer of vulnerability to hepatitis C infection, particularly for homeless veterans.^5-7

Mental Health and Primary Care Integration

Substance abuse and dual-diagnosis treatment programs (ie, those programs that treat both substance abuse and co-occurring serious mental health problems, such as bipolar disorder, severe major depressive disorder, psychotic disorders, and posttraumatic stress disorder [PTSD]) that have integrated mental health and primary care into their treatment programs may offer a window of opportunity for risk-reducing interventions. These interventions include testing and education of patients regarding infectious diseases, such as viral hepatitis and HIV, and completion of the hepatitis A/B immunization series.

The James A. Haley Veterans’ Hospital (JAHVH) in Tampa, Florida, has demonstrated some limited success in the past with integrating a standard dosing schedule for hepatitis A/B vaccination into its substance abuse treatment program (SATP), though recent evidence points to more promising results using an accelerated regimen as indicated by a high completion rate for hepatitis B vaccination in a methadone clinic.^8,9 A relatively low proportion of SATPs in the U.S. provide testing, education, or vaccination for hepatitis A and B, especially considering the public health importance of controlling these diseases in the substance abusing populations.^10,11

Related: Combination Pill Approved for HCV

In 1999, a primary care team was added to the alcohol and drug abuse treatment program at JAHVH.In 2005, the nurses in the program began scheduling vaccinations and screening patients for medical and psychiatric issues, pain, hypertension, diabetes, hepatitis C, alcohol use, depression, PTSD, prostate and colorectal cancers.¹² Such a multidisciplinary approach provides many treatment advantages for patients and may save lives.¹³

Even with a multidisciplinary approach, the nurses found it difficult to provide adequate hepatitis A/B immunization within the 3- to 6-week intensive SATP, because standard immunization dosing regimens are spread over 6 months.¹⁴ As with all types of immunizations, long dosing schedules may reduce patient adherence and result in inadequate seroprotection.¹⁵ Thus, there is a need to provide a completed immunization series in a more expeditious fashion, and an accelerated dosing regimen makes that possible.^15,16

Hepatitis A/B Vaccination

Twinrix (GlaxoSmithKline, Brentford, United Kingdom) is a vaccine that provides dual immunization for hepatitis A and B. Whereas the standard vaccination schedule takes 6 months to complete, the accelerated dosing schedule can be used to complete the first 3 doses in less than a month. The accelerated dosing schedule was incorporated into the JAHVH clinic to capture as many patients as possible in the 3- to 6-week time frame: The first dose is administered and followed by a second dose 7 days later. The third dose is administered 21 to 30 days after the first dose. Twelve months after the first dose, a booster dose is given. 

After the first 3 accelerated doses, > 98% of patients show a sustained immune response to hepatitis A, and > 63% demonstrate immunity to hepatitis B. If a 12-month booster injection is given, 100% of patients may receive immunity to hepatitis A and > 96% may have immunity to hepatitis B.¹⁶ Another study of the combined vaccine showed even greater seroprotection for hepatitis A and B after only 1 month, 100% and 82%, respectively.¹⁷

Related: Viral Hepatitis Awareness

This JAHVH retrospective feasibility study describes a risk-reduction program for hepatitis A/B prevention that was implemented within a 3- to 4-week intensive outpatient SATP and a 6-week dual-diagnosis treatment program. The study includes the development and implementation of the program, designed to vaccinate patients using the accelerated Twinrix schedule. To ascertain the feasibility of this vaccination approach, historical medical records were used to describe and examine the vaccination initiation and follow-up rates of the treatment program participants who received the hepatitis A/B immunization series during their intensive SATP.

Study Design

A retrospective review of medical records was conducted for all participants who were admitted to the intensive JAHVH SATP between October 1, 2008, and September 30, 2009. This study was reviewed and approved by the JAHVH research and development committee and its associated University of South Florida institutional review board. Informed consent to participate was not obtained, because the study was retrospective.

Patient Identification and Education

All program participants were offered testing for HIV and hepatitis A, B, and C. Program participants were educated about hepatitis and HIV transmission, as well as about the long-term effects of continued substance abuse on the progression of hepatitis C. Education about hepatitis, HIV, and substance abuse was provided in a group setting by a member of the program’s nursing staff. One-on-one risk education counseling was also provided when requested or otherwise indicated.

Laboratory testing was performed following each participant’s initial physical examination (within 3 to 5 days of program admission), and the nursing staff reviewed the results before vaccination. Explanation of laboratory results and an individualized immunization regimen were provided to each participant. On review of participants’ laboratory results, those with seroconversion of both hepatitis A and B were not given the combined immunization. Participants who had seroconversion of hepatitis A were offered the hepatitis B vaccination series, and vice versa.

Immunization Process

Participants who lacked prior immunization for hepatitis A and B and had no seroconversion of either hepatitis A or B were offered vaccination. Some patients declined vaccination, even though they were eligible. Their reasons were not formally assessed.

Related: Nivolumab Approved for Expanded Indication

Patients who accepted the vaccination were given the accelerated regimen.¹⁶ Participants were educated on the importance of compliance with the vaccination series and provided with follow-up immunization dates and a reminder for the 1-year booster vaccine. The immunizations were ordered by the program’s primary care NP and administered by a licensed practical nurse. The nurse who administered the injections took responsibility for scheduling the patients for all their subsequent injections, including the 1-year booster.

Follow-up Care

If the third injection was not completed before discharge, patients were given a follow-up appointment with the nurse if they remained in the JAHVH service area. If they were leaving the area, they were given instructions on how to follow-up at another VA facility to continue their immunization schedule. A note was written in the electronic medical record documenting their abbreviated hepatitis A/B immunization schedule, which could be accessed by other providers at other VA facilities. Patients who did not show up for any follow-up appointments (third injection or the 1-year booster injection) were contacted and reminded about the importance of completing the immunization series and to schedule an appointment.

Statistical Analysis

All data were analyzed using IBM Statistical Package for the Social Sciences (IBM SPSS, Armonk, New York) with a focus on identifying differences between vaccination-eligible patients (n = 269) who did (n = 128) and did not (n = 141) initiate the immunization schedule during the treatment program. Chi-square and Fisher exact tests were used to assess statistical differences in initiation of the immunization schedule related to categoric variables (ie, marital status, race, history of IV drug abuse, cigarette smoking status, housing status, legal status, history of combat, having a psychiatric or medical diagnosis, and program track). Independent sample t tests were used to test for differences between these 2 groups on the continuous variables, including age, number of previous treatment programs, Global Assessment of Functioning score, severity of smoking dependence as measured by the Fagerström Test for Nicotine Dependence, and the Addiction Severity Index scales.^18-20

Results

The sample consisted of 284 successive admissions to an intensive outpatient program for veterans with substance use disorders. About one-third of the patients were homeless at the time of admission to the treatment, and 87% required contracted housing while completing treatment for reasons related to lack of housing, transportation, clinical necessity, or a combination of those factors (Table 1). The most common substance problems were alcohol and cocaine dependence, and 21% (n = 59) of the patients acknowledged a history of IV drug use during their initial psychiatric evaluation. Seventy percent were dually diagnosed with some other Axis I disorder, and 40% had a history of serious mental illness. More than one-fourth (n = 77) of the patients admitted to the intensive outpatient SATP were seropositive for hepatitis A, B and/or C, and the most common hepatitis diagnosis was hepatitis C (n = 71).

Accelerated Immunization Regimen

Patients were eligible to receive the accelerated vaccination schedule only if they had no prior immunization for hepatitis A or B and if they had no seroconversion for either hepatitis A or B. Six people had hepatitis B alone, 7 had hepatitis B and C, 1 had hepatitis A and C, and 1 had all 3 (Table 2). Thus, 15 participants were ineligible to receive the accelerated hepatitis A/B immunization. Chi-square, Fisher exact, and independent sample t tests showed that among those who were vaccination-eligible (269), there were no significant differences in any of the demographic or clinical characteristics between those who initiated the vaccination schedule and those who did not. Among those who completed the first 3 vaccine injections, those who received the 1-year booster injection (54) did not differ (on any demographic or clinical variables) from those who did not (58).

Nearly half (48%) of all the eligible patients admitted to the program began the accelerated immunization schedule for hepatitis A and B.  Of those, 88% completed the first 3 injections in the series. Among the patients who received the first 3 injections, 48% received the 1-year booster injection—a 20% completion rate for the vaccination-eligible sample overall (Table 3).

Of the 74 patients who did not complete their vaccinations once initiating the accelerated schedule, the most common reason identified was that the patient moved away (37), or no reason could be identified (33). It was uncommon for a patient not to complete the vaccination schedule because of terminating treatment prematurely (4).

Compared with the vaccine-eligible patients without hepatitis C (207), patients with hepatitis C were less likely to receive any vaccination injections (Table 3). Specifically, 51% of the vaccination-eligible patients who did not have hepatitis C began the vaccination regimen. However, only 22 patients with hepatitis C, or
35% of all vaccination-eligible patients with hepatitis C, began the vaccination regimen. Patients with hepatitis C were also less likely than those without hepatitis C to complete the first 3 injections of the vaccination series once they had initiated it (77%, vs 90%, respectively). This difference continued to be apparent at the time of the 12-month booster injection. Only 35% of vaccine-eligible individuals with hepatitis C received the 12-month booster injection, whereas 51% of vaccination-eligible individuals without hepatitis C received the 12-month booster injection. As with the sample overall, the most common reason patients with hepatitis C did not complete the vaccination regimen was because they moved away (9), followed by no identified reason (5), and premature termination of treatment (2).

Discussion

Individuals abusing alcohol and drugs have an increased vulnerability for infectious diseases, and homeless veterans with substance use disorders may be at a particularly heightened risk.^21,22 This study describes a sample of veterans, many were homeless and most were dually diagnosed, in an intensive outpatient SATP that offered an accelerated dosing regimen for hepatitis A and B vaccination. Almost half (48%) of the vaccination-eligible patients began the accelerated regimen for hepatitis A/B vaccination. Moreover, 88% of those who started the vaccination regimen received the first 3 injections of the series, thus possibly conferring substantial immunity to hepatitis A and B and demonstrating the feasibility of an accelerated vaccination schedule in an intensive outpatient SATP.

It is especially important to demonstrate the successful integration of a hepatitis screening and immunization program within a SATP, given that many such programs do not offer screening or immunization for hepatitis, even though substance abusers are disproportionately affected by the disease and contribute greatly to the ongoing hepatitis epidemic.^10,11 This study’s results were in line with another study of rapid vaccination for hepatitis B in IV drug users being treated in a methadone clinic, where 83% of the vaccination initiators completed the first 3 injections of the series.⁹

Unvaccinated Patients

The treatment team in the current study seemed to be less effective at reaching the subset of vaccination-eligible veterans with hepatitis C (almost one-quarter of the sample) in order to administer the accelerated vaccination schedule, as indicated by the lower rate of vaccination initiation as well as a lower rate of completion of the vaccination series among those patients. This replicates a finding from another study that also indicated a low rate of hepatitis A and B vaccination among patients with hepatitis C.²³ Only 35% of the vaccination-eligible patients with hepatitis C in the current study initiated the vaccination series, compared with 51% of the patients without hepatitis C. However, the rate of completion of the first 3 injections of the series in the hepatitis C group was respectably high (77%), especially given the high relapse rate and psychosocial instability of individuals with addictive disorders. Initiation seems to be a bigger obstacle than completion of at least the first 3 injections of the vaccination series in both patients with and without hepatitis C.

The study investigators did not formally assess the reasons that more than half the patients in the study did not begin the vaccination series, but anecdotal evidence from the nurses indicated that many patients were afraid of needles. In addition, other patients felt that they simply did not need the vaccination. Some also insisted that they had already had the vaccination despite a blood test showing no evidence for either hepatitis A or B immunization.

Although the nursing team provided group and individual risk-based education as well as information about the effects of continued substance abuse on hepatitis C, it is possible that patients still underestimated their own risk of hepatitis infection and its consequences, or perhaps the information was simply not retained.²⁴

Patient Education

A recent study showed that there is a positive relationship between the amount of hepatitis counseling received and knowledge of hepatitis.²⁵ Possibly, increased intensity of education efforts may make an impact on initiation rates. Encouragingly, there is also evidence that prompting people to predict their future vaccination behavior may increase vaccination initiation rates despite a high-degree of short-term barriers, such as perceived pain or inconvenience.²⁶ A brief intervention to induce people to formulate their future intentions would be relatively easy to incorporate into a vaccination program, and the study team is considering options for this to improve vaccination initiation rates.

Patients can expect to achieve substantial immunity from hepatitis A and, to a lesser degree, hepatitis B after completing the first 3 injections of the series, although the best seroprotection from both is obtained by completing the 12-month booster injection as well.¹⁷ Overall, about half of all patients who completed the first 3 injections returned for the booster shot, but only 35% of the patients with hepatitis C did so. The most common known cause of any patient not receiving the booster was movement out of the geographic area. However, much of the time the investigators were unable to determine the reasons patients did not return for the booster shot.

Medication adherence is a difficult problem with vaccination in high-risk samples, although Stitzer and colleagues found a significant improvement in follow-up for a 6-month vaccination protocol by using monetary incentives.²⁷ In addition to ensuring medication adherence, it would also be of value for future immunization efforts to include testing to assess whether seroconversion has occurred once the vaccinations are complete, which is the ultimate measure of the success of a vaccination program. Most patients in the current study did not receive such testing at the completion of their vaccination schedules, and thus, seroconversion rates could not be determined. However, existing studies suggest high rates of seroprotection after the first 3 doses of the combined vaccine.^10,17

Limitations

The retrospective nature of the study is its most significant limitation. Any conclusions about the results must be made with caution. However, this design allowed for a naturalistic and potentially generalizable investigation into the application of a vaccination program in a real-world treatment setting. As such, the investigators were able to demonstrate the feasibility of conducting a rapid vaccination program within a 3- to 6-week SATP.

The retrospective nature of the study also limited a full investigation into the reasons behind the lack of vaccination initiation and vaccination noncompletion among the study’s treatment population, especially with regard to the follow-up booster injection. Initial statistical comparisons of initiators and noninitiators and completers and noncompleters showed no significant statistical differences between the groups. Future prospective designs should take into account the need to successfully initiate and complete vaccinations for all eligible patients and include assessment measures to determine the specific reasons that patients did not initiate or complete their vaccinations.

Conclusions

Many patients began and completed the accelerated vaccination schedule for hepatitis A and B in the context of a 3- to 6-week SATP at JAHVH. The overall vaccination rate, including the 12-month booster injection, was one-fifth of the entire vaccination-eligible sample. Additionally, 88% of the vaccination-eligible patients who began the vaccination schedule (or 42% of the whole sample) completed at least the first 3 doses, which may confer substantial immunity from hepatitis A and B. For reasons not entirely clear, a little less than half the vaccination-eligible patients began the vaccination schedule, and only about 50% of those returned to receive their 12-month booster injection. Future prospective studies may be able to determine barriers to both the initiation of and adherence to the vaccination protocol.

The results of this study are also a testament to having primary care nursing staff available and actively involved in the care of patients in a SATP. It seems likely that additional interventions might be needed for outreach to and retention of patients in need of vaccination for hepatitis A and B, and particularly those patients with hepatitis C. It is important to find ways to increase the rates of 12-month booster vaccinations, both for veterans who continue to receive services at JAHVH and for those who transfer care to other VA facilities. Finally, testing to confirm serologic immunity to hepatitis A and hepatitis B would be the next step in the effort to eliminate the risk of hepatitis A and hepatitis B and minimize additional harm for those with hepatitis C in the population receiving treatment for addictive disorders.

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Orthopedics and the sports medicine subspecialty are continually evolving fields that depend on research investigation and publication to further knowledge and advance practice. Research has produced new findings that have changed the way we practice sports medicine. In this review, we identify the most widely referenced sports medicine topics and articles, which we believe by their permeative presence in the literature have made lasting contributions to the field.

Many factors can be used to quantify the influence of an academic article on the practice of medicine. Citation analysis is one method that reflects the impact of a publication on the academic medical community.^1-3 Total citations record the number of times a journal article has been credited by another study. Therefore, citation count indirectly highlights the articles that are widespread, relevant, and that form the foundation for other investigations on the topic. Related to the impact of the article is the impact of the journal that published the study. We examined journals by impact factor, a score based on the mean number of citations a published article received during the preceding 2 years.

Similar analyses have been performed of publication history in orthopedics and other medical fields. Investigators have examined which historical articles were the most influential in orthopedics as a whole,⁴ pediatric orthopedics,^5,6 shoulder surgery,⁷ and arthroscopy.⁸ This influence has also been studied in general surgery,⁹ otolaryngology,¹⁰ plastic surgery,¹¹ dermatology,¹² critical care,¹³ and other disciplines. To our knowledge, the present study is the first bibliometric analysis of the highest-impact articles in orthopedic sports medicine.

Our goal was to identify the 100 articles that have had the highest impact on the clinical orthopedic sports medicine literature. We hypothesized that the most widely recognized articles would be from the highest-impact journals and may also have earlier publication dates. We describe the topics and objectives of these articles to highlight the sports medicine areas on which most research has focused during the past century.

Materials and Methods

Our bibliometric analysis used the Thomson Reuters Web of Knowledge, which consists of all publications from 1900 to the present. This research modality ranks journal articles by frequency of citation. Similar analyses have identified the most often cited articles in pediatric orthopedics,⁵ shoulder surgery,⁷ and arthroscopy.⁸ In our analysis, we included the top 25 journals by impact factor in the field of sports medicine, as rated by the Journal Citation Reports database. Within the highest-impact journals, we sorted all articles by those most often cited, and read them all to identify which ones discuss conditions commonly encountered in the clinical practice of sports medicine. We focused on clinical articles only and therefore excluded related basic science and cadaveric biomechanical studies. The 100 most cited articles were then further evaluated by primary author, journal of publication, institution, country of origin, year of publication, topic, and total number of citations. One-way analysis of variance (ANOVA) and linear regression analyses were used to determine if publication date correlated with mean number of citations.

Results

Eighty authors wrote the top 100 articles in sports medicine, and each publication garnered several hundred citations, ranging from 229 to 1629 with a mean of 408 (Table 1^14-113). Most of these articles were written in the past 3 decades, with equal distribution from the 1980s, 1990s, and 2000s (Figure 1A). We ran a linear regression to determine if publication date correlated with higher number of citations by virtue of longer time available for citation. The analysis poorly modeled the variability (R² = 0.05), revealing no correlation between number of citations and publication date. Further, 1-way ANOVA found no significant difference between the number of citations per decade, F(5, 93) = 1.60, P = .17 (Figure 1B). Despite this finding, the oldest cited article, written by Fairbank³⁹ in 1948, ranked high (position 7). Of these top 100 publications, the most recent, written by Knutsen and colleagues⁶⁹ in 2007, ranked in the second half at position 66.

Seven journals published the top 100 articles, with the American volume of the Journal of Bone and Joint Surgery publishing nearly half (44%) (Table 2). In second place, with 28 articles, was the American Journal of Sports Medicine, followed by the British volume of the Journal of Bone and Joint Surgery, with 10 articles.

Thirty different topics were investigated in this collection of articles, encompassing nearly every major research area of sports medicine. There was a heavy emphasis on anterior cruciate ligament (ACL) injury and reconstruction, knee rating systems, rotator cuff reconstruction, and chondrocyte transplantation (Table 3).

In several cases, an author contributed more than 1 classic article. In fact, 31 of the top 100 articles were by an individual who had coauthored 2 or more of the publications on this list. The researchers with the largest number of first-authored articles were Noyes^88-92 (5 articles), Neer^81-84 (4 articles), and Rowe,^102-104 Daniel,^35-37 Peterson,^97-99 and Hewett^52-54 (3 articles each) (Table 4^{17,19,21-24,29-31,35-37,42,44,45,52-54,58,61-65,69,70,72,74,80-84,87-92,97-99,101-105,107,109,110,113}). Articles from authors with multiple publications had a common topic.

Last, these articles originated from a number of different countries and institutions. Of the 15 source countries (Figure 2), the United States contributed the most (61 articles). Other countries had prominent representation: Sweden and Switzerland (8 each), United Kingdom (5), and Canada, France, and Norway (3 each). These articles originated from 69 universities, hospitals, and clinics; 21 institutions had 2 or more articles (Table 5). The 5 institutions with the highest number of articles were Hospital for Special Surgery, University of Bern, Columbia College of Physicians and Surgeons/Columbia-Presbyterian Medical Center, Cincinnati Sports Medicine and Orthopaedic Center, and Massachusetts General Hospital.

Discussion

Several trends can be ascertained from analyzing the top 100 clinical articles cited in sports medicine. The 5 most frequent topics discussed were ACL injury and reconstruction, knee rating systems for injury and function, rotator cuff reconstruction, chondrocyte transplantation, and femoroacetabular impingement (Table 3). Of those 5 topics, only ACL injury and reconstruction falls within the top 10 most common orthopedic surgical procedures performed in the United States reported by one analysis.¹¹⁴ The most common orthopedic surgical procedure, knee arthroscopy, ranks 10th of all topics covered by the top 100 articles, whereas the second most common procedure, shoulder arthroscopy, was not discussed by any of those 100 articles. Also notable is the high frequency of knee rating system studies, which correlates well with the fact that 4 of the most common orthopedic surgical procedures are knee procedures. The prevalence of rating system articles reflects the importance of and need for accurate methods in the diagnosis of injuries in sports medicine.

The most cited sports medicine article was written by Insall and colleagues⁶² in 1989, more than 2 decades ago. In this article, “Rationale of the Knee Society Clinical Rating System,” they reported on a rigorous system that rates knee function and ability to walk and climb stairs. The second most cited article, “A Clinical Method of Functional Assessment of the Shoulder,” was written in 1987 by Constant and Murley.³² This article discusses another rating system but offers a functional assessment of the shoulder that is highly reproducible and time-efficient. “Rating Systems in the Evaluation of Knee Ligament Injuries,” the third most cited article, was written in 1985 by Tegner and Lysholm.¹¹³ This article details the complexities and variable uses of different knee ligament injury rating systems. These top 3 articles were all published in Clinical Orthopaedics and Related Research. In addition, all 3 discussed rating systems, reinforcing the need for accurate scoring systems to standardize the diagnosis of injury across the field of orthopedics and qualify outcomes after injury.

A number of studies have introduced physical examination findings, clinical tests, and rating systems used in the clinical setting of sports medicine (and named after the contributing authors). For example, the Neer sign⁸² and the Hawkins-Kennedy test⁵¹ are used to determine shoulder impingement. In knee ligament injuries, the Tegner knee activity score¹¹³ complements other functional scores (eg, Lysholm knee score⁷⁴). For grading joint cartilage breakdown, the Outerbridge classification system⁹⁶ is commonly used. The Fairbank test³⁹ is used to gauge knee instability. In evaluating fatty degeneration of rotator cuff muscles through computed tomography scans, the Goutallier classification⁴⁷ is used. Other metrics, such as the Knee Injury and Osteoarthritis Outcome Score, introduced by Roos and colleagues,¹⁰¹ measure knee injury and osteoarthritis. In other scenarios, studies have improved on surgical techniques—for example, the Neer open modification⁸⁴ of the Bankart procedure. Many of these rating systems and named clinical findings are so ingrained in the practice and vernacular of orthopedics that it is possible they are in fact undercited in the literature.

As in other bibliometric analyses, one concession made here was to credit the first author listed for making the primary contribution to an article. As a result of journal variability and inconsistency, we were precluded from analyzing senior authors. When analyzed for authorship at any position, 3 of the top authors (Table 4) showed contributions to additional articles in the top 100 list. Noyes was listed as last author on 2 other articles,^52,54 raising his total to 7. Daniel was listed as second author on 1 additional article,¹⁰⁵ and Beck was listed as third author on 1 other article,⁴² raising their totals to 4 and 3, respectively.

A criticism of bibliometric analysis is its use of number of citations as an accurate measure of academic contribution. However, other methods for measuring the productivity and impact of researchers (eg, the recently developed Hirsch Index) have their own drawbacks,^115,116 including being able to compare authors only at the same point in their careers and self-citation. It is important to note that our analyses focused strictly on publications related to clinical sports medicine, with the exclusion of basic science and cadaveric biomechanical studies.

Through bibliometric citation analysis, we have identified the authors who have made lasting contributions to the field of sports medicine, and we have highlighted the publications that have been cited by hundreds to thousands of authors. This list identifies trends within the articles that have become “classic,” by nature of their deep permeation into subsequent sports medicine literature, and offers guidance for trainees interested in studying the most high-yield sports medicine literature. Given that 69 institutions in 15 countries conducted these studies, we have also shown that orthopedic research can be readily disseminated internationally. Last, our study provides a thorough overview of the sports medicine literature over the past century and provides a strong framework for future research in our field.

Orthopedics and the sports medicine subspecialty are continually evolving fields that depend on research investigation and publication to further knowledge and advance practice. Research has produced new findings that have changed the way we practice sports medicine. In this review, we identify the most widely referenced sports medicine topics and articles, which we believe by their permeative presence in the literature have made lasting contributions to the field.

Many factors can be used to quantify the influence of an academic article on the practice of medicine. Citation analysis is one method that reflects the impact of a publication on the academic medical community.^1-3 Total citations record the number of times a journal article has been credited by another study. Therefore, citation count indirectly highlights the articles that are widespread, relevant, and that form the foundation for other investigations on the topic. Related to the impact of the article is the impact of the journal that published the study. We examined journals by impact factor, a score based on the mean number of citations a published article received during the preceding 2 years.

Similar analyses have been performed of publication history in orthopedics and other medical fields. Investigators have examined which historical articles were the most influential in orthopedics as a whole,⁴ pediatric orthopedics,^5,6 shoulder surgery,⁷ and arthroscopy.⁸ This influence has also been studied in general surgery,⁹ otolaryngology,¹⁰ plastic surgery,¹¹ dermatology,¹² critical care,¹³ and other disciplines. To our knowledge, the present study is the first bibliometric analysis of the highest-impact articles in orthopedic sports medicine.

Our goal was to identify the 100 articles that have had the highest impact on the clinical orthopedic sports medicine literature. We hypothesized that the most widely recognized articles would be from the highest-impact journals and may also have earlier publication dates. We describe the topics and objectives of these articles to highlight the sports medicine areas on which most research has focused during the past century.

Materials and Methods

Our bibliometric analysis used the Thomson Reuters Web of Knowledge, which consists of all publications from 1900 to the present. This research modality ranks journal articles by frequency of citation. Similar analyses have identified the most often cited articles in pediatric orthopedics,⁵ shoulder surgery,⁷ and arthroscopy.⁸ In our analysis, we included the top 25 journals by impact factor in the field of sports medicine, as rated by the Journal Citation Reports database. Within the highest-impact journals, we sorted all articles by those most often cited, and read them all to identify which ones discuss conditions commonly encountered in the clinical practice of sports medicine. We focused on clinical articles only and therefore excluded related basic science and cadaveric biomechanical studies. The 100 most cited articles were then further evaluated by primary author, journal of publication, institution, country of origin, year of publication, topic, and total number of citations. One-way analysis of variance (ANOVA) and linear regression analyses were used to determine if publication date correlated with mean number of citations.

Results

Eighty authors wrote the top 100 articles in sports medicine, and each publication garnered several hundred citations, ranging from 229 to 1629 with a mean of 408 (Table 1^14-113). Most of these articles were written in the past 3 decades, with equal distribution from the 1980s, 1990s, and 2000s (Figure 1A). We ran a linear regression to determine if publication date correlated with higher number of citations by virtue of longer time available for citation. The analysis poorly modeled the variability (R² = 0.05), revealing no correlation between number of citations and publication date. Further, 1-way ANOVA found no significant difference between the number of citations per decade, F(5, 93) = 1.60, P = .17 (Figure 1B). Despite this finding, the oldest cited article, written by Fairbank³⁹ in 1948, ranked high (position 7). Of these top 100 publications, the most recent, written by Knutsen and colleagues⁶⁹ in 2007, ranked in the second half at position 66.

Seven journals published the top 100 articles, with the American volume of the Journal of Bone and Joint Surgery publishing nearly half (44%) (Table 2). In second place, with 28 articles, was the American Journal of Sports Medicine, followed by the British volume of the Journal of Bone and Joint Surgery, with 10 articles.

Thirty different topics were investigated in this collection of articles, encompassing nearly every major research area of sports medicine. There was a heavy emphasis on anterior cruciate ligament (ACL) injury and reconstruction, knee rating systems, rotator cuff reconstruction, and chondrocyte transplantation (Table 3).

In several cases, an author contributed more than 1 classic article. In fact, 31 of the top 100 articles were by an individual who had coauthored 2 or more of the publications on this list. The researchers with the largest number of first-authored articles were Noyes^88-92 (5 articles), Neer^81-84 (4 articles), and Rowe,^102-104 Daniel,^35-37 Peterson,^97-99 and Hewett^52-54 (3 articles each) (Table 4^{17,19,21-24,29-31,35-37,42,44,45,52-54,58,61-65,69,70,72,74,80-84,87-92,97-99,101-105,107,109,110,113}). Articles from authors with multiple publications had a common topic.

Last, these articles originated from a number of different countries and institutions. Of the 15 source countries (Figure 2), the United States contributed the most (61 articles). Other countries had prominent representation: Sweden and Switzerland (8 each), United Kingdom (5), and Canada, France, and Norway (3 each). These articles originated from 69 universities, hospitals, and clinics; 21 institutions had 2 or more articles (Table 5). The 5 institutions with the highest number of articles were Hospital for Special Surgery, University of Bern, Columbia College of Physicians and Surgeons/Columbia-Presbyterian Medical Center, Cincinnati Sports Medicine and Orthopaedic Center, and Massachusetts General Hospital.

Discussion

Several trends can be ascertained from analyzing the top 100 clinical articles cited in sports medicine. The 5 most frequent topics discussed were ACL injury and reconstruction, knee rating systems for injury and function, rotator cuff reconstruction, chondrocyte transplantation, and femoroacetabular impingement (Table 3). Of those 5 topics, only ACL injury and reconstruction falls within the top 10 most common orthopedic surgical procedures performed in the United States reported by one analysis.¹¹⁴ The most common orthopedic surgical procedure, knee arthroscopy, ranks 10th of all topics covered by the top 100 articles, whereas the second most common procedure, shoulder arthroscopy, was not discussed by any of those 100 articles. Also notable is the high frequency of knee rating system studies, which correlates well with the fact that 4 of the most common orthopedic surgical procedures are knee procedures. The prevalence of rating system articles reflects the importance of and need for accurate methods in the diagnosis of injuries in sports medicine.

The most cited sports medicine article was written by Insall and colleagues⁶² in 1989, more than 2 decades ago. In this article, “Rationale of the Knee Society Clinical Rating System,” they reported on a rigorous system that rates knee function and ability to walk and climb stairs. The second most cited article, “A Clinical Method of Functional Assessment of the Shoulder,” was written in 1987 by Constant and Murley.³² This article discusses another rating system but offers a functional assessment of the shoulder that is highly reproducible and time-efficient. “Rating Systems in the Evaluation of Knee Ligament Injuries,” the third most cited article, was written in 1985 by Tegner and Lysholm.¹¹³ This article details the complexities and variable uses of different knee ligament injury rating systems. These top 3 articles were all published in Clinical Orthopaedics and Related Research. In addition, all 3 discussed rating systems, reinforcing the need for accurate scoring systems to standardize the diagnosis of injury across the field of orthopedics and qualify outcomes after injury.

A number of studies have introduced physical examination findings, clinical tests, and rating systems used in the clinical setting of sports medicine (and named after the contributing authors). For example, the Neer sign⁸² and the Hawkins-Kennedy test⁵¹ are used to determine shoulder impingement. In knee ligament injuries, the Tegner knee activity score¹¹³ complements other functional scores (eg, Lysholm knee score⁷⁴). For grading joint cartilage breakdown, the Outerbridge classification system⁹⁶ is commonly used. The Fairbank test³⁹ is used to gauge knee instability. In evaluating fatty degeneration of rotator cuff muscles through computed tomography scans, the Goutallier classification⁴⁷ is used. Other metrics, such as the Knee Injury and Osteoarthritis Outcome Score, introduced by Roos and colleagues,¹⁰¹ measure knee injury and osteoarthritis. In other scenarios, studies have improved on surgical techniques—for example, the Neer open modification⁸⁴ of the Bankart procedure. Many of these rating systems and named clinical findings are so ingrained in the practice and vernacular of orthopedics that it is possible they are in fact undercited in the literature.

As in other bibliometric analyses, one concession made here was to credit the first author listed for making the primary contribution to an article. As a result of journal variability and inconsistency, we were precluded from analyzing senior authors. When analyzed for authorship at any position, 3 of the top authors (Table 4) showed contributions to additional articles in the top 100 list. Noyes was listed as last author on 2 other articles,^52,54 raising his total to 7. Daniel was listed as second author on 1 additional article,¹⁰⁵ and Beck was listed as third author on 1 other article,⁴² raising their totals to 4 and 3, respectively.

A criticism of bibliometric analysis is its use of number of citations as an accurate measure of academic contribution. However, other methods for measuring the productivity and impact of researchers (eg, the recently developed Hirsch Index) have their own drawbacks,^115,116 including being able to compare authors only at the same point in their careers and self-citation. It is important to note that our analyses focused strictly on publications related to clinical sports medicine, with the exclusion of basic science and cadaveric biomechanical studies.

Through bibliometric citation analysis, we have identified the authors who have made lasting contributions to the field of sports medicine, and we have highlighted the publications that have been cited by hundreds to thousands of authors. This list identifies trends within the articles that have become “classic,” by nature of their deep permeation into subsequent sports medicine literature, and offers guidance for trainees interested in studying the most high-yield sports medicine literature. Given that 69 institutions in 15 countries conducted these studies, we have also shown that orthopedic research can be readily disseminated internationally. Last, our study provides a thorough overview of the sports medicine literature over the past century and provides a strong framework for future research in our field.

Orthopedics and the sports medicine subspecialty are continually evolving fields that depend on research investigation and publication to further knowledge and advance practice. Research has produced new findings that have changed the way we practice sports medicine. In this review, we identify the most widely referenced sports medicine topics and articles, which we believe by their permeative presence in the literature have made lasting contributions to the field.

Many factors can be used to quantify the influence of an academic article on the practice of medicine. Citation analysis is one method that reflects the impact of a publication on the academic medical community.^1-3 Total citations record the number of times a journal article has been credited by another study. Therefore, citation count indirectly highlights the articles that are widespread, relevant, and that form the foundation for other investigations on the topic. Related to the impact of the article is the impact of the journal that published the study. We examined journals by impact factor, a score based on the mean number of citations a published article received during the preceding 2 years.

Similar analyses have been performed of publication history in orthopedics and other medical fields. Investigators have examined which historical articles were the most influential in orthopedics as a whole,⁴ pediatric orthopedics,^5,6 shoulder surgery,⁷ and arthroscopy.⁸ This influence has also been studied in general surgery,⁹ otolaryngology,¹⁰ plastic surgery,¹¹ dermatology,¹² critical care,¹³ and other disciplines. To our knowledge, the present study is the first bibliometric analysis of the highest-impact articles in orthopedic sports medicine.

Our goal was to identify the 100 articles that have had the highest impact on the clinical orthopedic sports medicine literature. We hypothesized that the most widely recognized articles would be from the highest-impact journals and may also have earlier publication dates. We describe the topics and objectives of these articles to highlight the sports medicine areas on which most research has focused during the past century.

Materials and Methods

Our bibliometric analysis used the Thomson Reuters Web of Knowledge, which consists of all publications from 1900 to the present. This research modality ranks journal articles by frequency of citation. Similar analyses have identified the most often cited articles in pediatric orthopedics,⁵ shoulder surgery,⁷ and arthroscopy.⁸ In our analysis, we included the top 25 journals by impact factor in the field of sports medicine, as rated by the Journal Citation Reports database. Within the highest-impact journals, we sorted all articles by those most often cited, and read them all to identify which ones discuss conditions commonly encountered in the clinical practice of sports medicine. We focused on clinical articles only and therefore excluded related basic science and cadaveric biomechanical studies. The 100 most cited articles were then further evaluated by primary author, journal of publication, institution, country of origin, year of publication, topic, and total number of citations. One-way analysis of variance (ANOVA) and linear regression analyses were used to determine if publication date correlated with mean number of citations.

Results

Eighty authors wrote the top 100 articles in sports medicine, and each publication garnered several hundred citations, ranging from 229 to 1629 with a mean of 408 (Table 1^14-113). Most of these articles were written in the past 3 decades, with equal distribution from the 1980s, 1990s, and 2000s (Figure 1A). We ran a linear regression to determine if publication date correlated with higher number of citations by virtue of longer time available for citation. The analysis poorly modeled the variability (R² = 0.05), revealing no correlation between number of citations and publication date. Further, 1-way ANOVA found no significant difference between the number of citations per decade, F(5, 93) = 1.60, P = .17 (Figure 1B). Despite this finding, the oldest cited article, written by Fairbank³⁹ in 1948, ranked high (position 7). Of these top 100 publications, the most recent, written by Knutsen and colleagues⁶⁹ in 2007, ranked in the second half at position 66.

Seven journals published the top 100 articles, with the American volume of the Journal of Bone and Joint Surgery publishing nearly half (44%) (Table 2). In second place, with 28 articles, was the American Journal of Sports Medicine, followed by the British volume of the Journal of Bone and Joint Surgery, with 10 articles.

Thirty different topics were investigated in this collection of articles, encompassing nearly every major research area of sports medicine. There was a heavy emphasis on anterior cruciate ligament (ACL) injury and reconstruction, knee rating systems, rotator cuff reconstruction, and chondrocyte transplantation (Table 3).

In several cases, an author contributed more than 1 classic article. In fact, 31 of the top 100 articles were by an individual who had coauthored 2 or more of the publications on this list. The researchers with the largest number of first-authored articles were Noyes^88-92 (5 articles), Neer^81-84 (4 articles), and Rowe,^102-104 Daniel,^35-37 Peterson,^97-99 and Hewett^52-54 (3 articles each) (Table 4^{17,19,21-24,29-31,35-37,42,44,45,52-54,58,61-65,69,70,72,74,80-84,87-92,97-99,101-105,107,109,110,113}). Articles from authors with multiple publications had a common topic.

Last, these articles originated from a number of different countries and institutions. Of the 15 source countries (Figure 2), the United States contributed the most (61 articles). Other countries had prominent representation: Sweden and Switzerland (8 each), United Kingdom (5), and Canada, France, and Norway (3 each). These articles originated from 69 universities, hospitals, and clinics; 21 institutions had 2 or more articles (Table 5). The 5 institutions with the highest number of articles were Hospital for Special Surgery, University of Bern, Columbia College of Physicians and Surgeons/Columbia-Presbyterian Medical Center, Cincinnati Sports Medicine and Orthopaedic Center, and Massachusetts General Hospital.

Discussion

Several trends can be ascertained from analyzing the top 100 clinical articles cited in sports medicine. The 5 most frequent topics discussed were ACL injury and reconstruction, knee rating systems for injury and function, rotator cuff reconstruction, chondrocyte transplantation, and femoroacetabular impingement (Table 3). Of those 5 topics, only ACL injury and reconstruction falls within the top 10 most common orthopedic surgical procedures performed in the United States reported by one analysis.¹¹⁴ The most common orthopedic surgical procedure, knee arthroscopy, ranks 10th of all topics covered by the top 100 articles, whereas the second most common procedure, shoulder arthroscopy, was not discussed by any of those 100 articles. Also notable is the high frequency of knee rating system studies, which correlates well with the fact that 4 of the most common orthopedic surgical procedures are knee procedures. The prevalence of rating system articles reflects the importance of and need for accurate methods in the diagnosis of injuries in sports medicine.

The most cited sports medicine article was written by Insall and colleagues⁶² in 1989, more than 2 decades ago. In this article, “Rationale of the Knee Society Clinical Rating System,” they reported on a rigorous system that rates knee function and ability to walk and climb stairs. The second most cited article, “A Clinical Method of Functional Assessment of the Shoulder,” was written in 1987 by Constant and Murley.³² This article discusses another rating system but offers a functional assessment of the shoulder that is highly reproducible and time-efficient. “Rating Systems in the Evaluation of Knee Ligament Injuries,” the third most cited article, was written in 1985 by Tegner and Lysholm.¹¹³ This article details the complexities and variable uses of different knee ligament injury rating systems. These top 3 articles were all published in Clinical Orthopaedics and Related Research. In addition, all 3 discussed rating systems, reinforcing the need for accurate scoring systems to standardize the diagnosis of injury across the field of orthopedics and qualify outcomes after injury.

A number of studies have introduced physical examination findings, clinical tests, and rating systems used in the clinical setting of sports medicine (and named after the contributing authors). For example, the Neer sign⁸² and the Hawkins-Kennedy test⁵¹ are used to determine shoulder impingement. In knee ligament injuries, the Tegner knee activity score¹¹³ complements other functional scores (eg, Lysholm knee score⁷⁴). For grading joint cartilage breakdown, the Outerbridge classification system⁹⁶ is commonly used. The Fairbank test³⁹ is used to gauge knee instability. In evaluating fatty degeneration of rotator cuff muscles through computed tomography scans, the Goutallier classification⁴⁷ is used. Other metrics, such as the Knee Injury and Osteoarthritis Outcome Score, introduced by Roos and colleagues,¹⁰¹ measure knee injury and osteoarthritis. In other scenarios, studies have improved on surgical techniques—for example, the Neer open modification⁸⁴ of the Bankart procedure. Many of these rating systems and named clinical findings are so ingrained in the practice and vernacular of orthopedics that it is possible they are in fact undercited in the literature.

As in other bibliometric analyses, one concession made here was to credit the first author listed for making the primary contribution to an article. As a result of journal variability and inconsistency, we were precluded from analyzing senior authors. When analyzed for authorship at any position, 3 of the top authors (Table 4) showed contributions to additional articles in the top 100 list. Noyes was listed as last author on 2 other articles,^52,54 raising his total to 7. Daniel was listed as second author on 1 additional article,¹⁰⁵ and Beck was listed as third author on 1 other article,⁴² raising their totals to 4 and 3, respectively.

A criticism of bibliometric analysis is its use of number of citations as an accurate measure of academic contribution. However, other methods for measuring the productivity and impact of researchers (eg, the recently developed Hirsch Index) have their own drawbacks,^115,116 including being able to compare authors only at the same point in their careers and self-citation. It is important to note that our analyses focused strictly on publications related to clinical sports medicine, with the exclusion of basic science and cadaveric biomechanical studies.

Through bibliometric citation analysis, we have identified the authors who have made lasting contributions to the field of sports medicine, and we have highlighted the publications that have been cited by hundreds to thousands of authors. This list identifies trends within the articles that have become “classic,” by nature of their deep permeation into subsequent sports medicine literature, and offers guidance for trainees interested in studying the most high-yield sports medicine literature. Given that 69 institutions in 15 countries conducted these studies, we have also shown that orthopedic research can be readily disseminated internationally. Last, our study provides a thorough overview of the sports medicine literature over the past century and provides a strong framework for future research in our field.

DOES PEANUT EXPOSURE INCREASE ALLERGY RISK?
Du Toit G, Roberts G, Sayre PH, et al; LEAP Study Team. Randomized trial of peanut consumption in infants at risk for peanut allergy. N Engl J Med. 2015;372(9):803-813. doi: 10.1056/NEJMoa1414850.

Among children at high risk for peanut allergy, early introduction of peanuts significantly reduced the risk for allergy, according to a randomized controlled trial of 640 infants with severe eczema, egg allergy, or both.

Participants ages 4 to 11 months who were high risk—based on severe atopy or allergy to eggs—were tested for sensitivity to peanut extract at baseline and then assigned to either avoid or consume peanuts.

At 60 months of age, skin allergy testing was repeated; the resulting prevalence of peanut allergy was as follows

A greater percentage of the consumption group showed an increase in levels of peanut-specific IgG4 antibody; the avoidance group, elevated titers of peanut specific IgE antibody.

COMMENTARY
The prevalence of peanut allergy in the United States has increased from 0.4% in 1997 to more than 2% in 2010.¹ In 2000, the American Academy of Pediatrics recommended peanut avoidance until age 3 in children at high risk for atopic disease; then in 2008, based on emerging evidence that early introduction of allergenic foods, including peanuts, may decrease the development of allergy, the recommendations were retracted.^2,3 The present study, using a randomized trial design, confirms the paradigm-changing hypothesis that early introduction of allergenic foods decreases the subsequent development of food allergies. The authors of the accompanying editorial state that these data are incontrovertible and that children at high risk for peanut allergy should, under supervision of an allergist, be tested and if skin-prick–­negative for peanut allergy, be started on a peanut protein–containing diet.¹—NS

1. Gruchalla RS, Sampson HA. Preventing peanut allergy through early consumption ready for prime time? N Engl J Med. 2015;372:875-876.
2. Du Toit G, Katz Y, Sasieni P, et al. Early consumption of peanuts in infancy is associated with a low prevalence of peanut allergy. J Allergy Clin Immunol. 2008;122:984-991.
3. Katz Y, Rajuan N, Goldberg MR, et al. Early exposure to cow’s milk protein is protective against IgE-mediated cow’s milk protein allergy. J Allergy Clin Immunol. 2010;126:77-82. 

Continue for Anaphylaxis guideline from AAAAI/ACAAI >>

ANAPHYLAXIS GUIDELINE FROM AAAAI/ACAAI
Campbell RL, Li JT, Nicklas RA, Sadosty AT. Emergency department diagnosis and treatment of anaphylaxis: a practice parameter. Ann Allergy Asthma Immunol. 2014;113(6):599-608. doi: 10.1016/j.anai.2014.10.007.

The American Academy of Allergy, Asthma & Immunology and the American College of Allergy, Asthma, and Immunology guideline for emergency department diagnosis and treatment of anaphylaxis includes the following recommendations
• Carefully and immediately triage and monitor patients who have signs and symptoms of ­anaphylaxis in preparation for epinephrine administration.
• First-line treatment for patients experiencing anaphylaxis is epinephrine. It should be administered intramuscularly in the anterolateral thigh immediately after the diagnosis is made. Epinephrine can be administered every 5 to 15 minutes as needed to control symptoms.
• Do not substitute for epinephrine in the treatment of anaphylaxis. Antihistamines and corticosteroids can be administered in conjunction with epinephrine, not in place of it.
• Determine whether the patient has risk factors for severe and potentially fatal anaphylaxis, such as delayed administration of epinephrine, asthma, a history of biphasic reactions, or cardiovascular disease.
• For anaphylaxis patients with bronchospasms, administer a β-agonist.
• Patients should be observed for 4 to 8 hours (longer for those with a history of risk factors for severe anaphylaxis).
• Refer patients to an allergist–immunologist upon discharge.

COMMENTARY
Identification of anaphylaxis requires judgment. Abbreviated criteria for anaphylaxis include essentially two organ systems of involvement: skin manifestations of pruritus, flushing, hives, or angioedema; respiratory manifestations of wheezing or stridor; decreased blood pressure; and GI symptoms of vomiting, cramping abdominal pain, or diarrhea. Quick identification and treatment of anaphylaxis is important, as the median time to respiratory or cardiac arrest in food-induced anaphylaxis is only 30 minutes. All offices should stock epinephrine in an obvious place; you may consider use of a prefilled pen, so it is easy to find and to give the correct dose. —NS

Continue for Updated practice parameter: Diagnosing and treating food allergies >>

UPDATED PRACTICE PARAMETER: DIAGNOSING AND TREATING FOOD ALLERGIES
Sampson HA, Aceves S, Bock SA, et al. Food allergy: a practice parameter update—2014. J Allergy Clin Immunol. 2014. pii: S0091-6749(14)00672-1. doi: 10.1016/j.jaci.2014.05.013. [Epub ahead of print]

Over the past decade, health care providers have been confronted with a growing number of patients with suspected food allergies, but data supporting an increase in confirmed allergy cases is limited.

The 2014 practice parameter update on food allergies from the American Academy of Allergy, Asthma & Immunology advises clinicians to keep in mind that self-reported food allergy is more common than proven food allergy; that allergy is more common in children and in patients with other atopic conditions; and that the majority of allergic reactions are from peanuts, tree nuts, fish, shellfish, milk, eggs, wheat, soy, and seeds.

The update of the 2006 guidelines includes 64 new summary statements. Highlights include
• Clinicians should advise patients about the risk for cross-reactions from foods similar to their allergens, such as other tree nuts, vertebrate fish, crustaceans, or milk from cows, goats, or other mammals.
• Patients with a seafood allergy should be advised that they are not at increased risk for a reaction to radiocontrast media.
• Patients with food allergies do not need to be concerned about eating genetically modified foods (GMO), due to current FDA screening requirements to rule out allergenicity.
• Patients with chronic idiopathic urticaria or hyperactivity/attention-deficit disorder should not routinely be advised to avoid food additives.
• Patients with asthma should not routinely be advised to avoid sulfates, unless they have had a previous reaction to them.

The guidelines also include a new section on diagnosing and treating non-IgE-mediated food allergies, such as food-protein–induced enterocolitis syndrome, allergic proctocolitis, enteropathy, eosinophilic esophagitis, and gastroenteritis. 

COMMENTARY
Food allergies are a commonly encountered and difficult area of practice for those of us in primary care. They can be life threatening; yet many patients who are concerned about the possibility of food allergy do not actually have a food allergy, so making an accurate diagnosis is important. Allergic evaluation starts with a careful history, and then laboratory testing can begin with specific IgE testing to foods suspected to have caused the clinical reaction of concern. The IgE results need to be carefully interpreted in light of the clinical context in which they were ordered. Oral food challenge can be helpful in addition to IgE testing. Consultation with an allergy-immunology specialist is often helpful as well. —NS

DOES PEANUT EXPOSURE INCREASE ALLERGY RISK?
Du Toit G, Roberts G, Sayre PH, et al; LEAP Study Team. Randomized trial of peanut consumption in infants at risk for peanut allergy. N Engl J Med. 2015;372(9):803-813. doi: 10.1056/NEJMoa1414850.

Among children at high risk for peanut allergy, early introduction of peanuts significantly reduced the risk for allergy, according to a randomized controlled trial of 640 infants with severe eczema, egg allergy, or both.

Participants ages 4 to 11 months who were high risk—based on severe atopy or allergy to eggs—were tested for sensitivity to peanut extract at baseline and then assigned to either avoid or consume peanuts.

At 60 months of age, skin allergy testing was repeated; the resulting prevalence of peanut allergy was as follows

A greater percentage of the consumption group showed an increase in levels of peanut-specific IgG4 antibody; the avoidance group, elevated titers of peanut specific IgE antibody.

COMMENTARY
The prevalence of peanut allergy in the United States has increased from 0.4% in 1997 to more than 2% in 2010.¹ In 2000, the American Academy of Pediatrics recommended peanut avoidance until age 3 in children at high risk for atopic disease; then in 2008, based on emerging evidence that early introduction of allergenic foods, including peanuts, may decrease the development of allergy, the recommendations were retracted.^2,3 The present study, using a randomized trial design, confirms the paradigm-changing hypothesis that early introduction of allergenic foods decreases the subsequent development of food allergies. The authors of the accompanying editorial state that these data are incontrovertible and that children at high risk for peanut allergy should, under supervision of an allergist, be tested and if skin-prick–­negative for peanut allergy, be started on a peanut protein–containing diet.¹—NS

1. Gruchalla RS, Sampson HA. Preventing peanut allergy through early consumption ready for prime time? N Engl J Med. 2015;372:875-876.
2. Du Toit G, Katz Y, Sasieni P, et al. Early consumption of peanuts in infancy is associated with a low prevalence of peanut allergy. J Allergy Clin Immunol. 2008;122:984-991.
3. Katz Y, Rajuan N, Goldberg MR, et al. Early exposure to cow’s milk protein is protective against IgE-mediated cow’s milk protein allergy. J Allergy Clin Immunol. 2010;126:77-82. 

Continue for Anaphylaxis guideline from AAAAI/ACAAI >>

ANAPHYLAXIS GUIDELINE FROM AAAAI/ACAAI
Campbell RL, Li JT, Nicklas RA, Sadosty AT. Emergency department diagnosis and treatment of anaphylaxis: a practice parameter. Ann Allergy Asthma Immunol. 2014;113(6):599-608. doi: 10.1016/j.anai.2014.10.007.

The American Academy of Allergy, Asthma & Immunology and the American College of Allergy, Asthma, and Immunology guideline for emergency department diagnosis and treatment of anaphylaxis includes the following recommendations
• Carefully and immediately triage and monitor patients who have signs and symptoms of ­anaphylaxis in preparation for epinephrine administration.
• First-line treatment for patients experiencing anaphylaxis is epinephrine. It should be administered intramuscularly in the anterolateral thigh immediately after the diagnosis is made. Epinephrine can be administered every 5 to 15 minutes as needed to control symptoms.
• Do not substitute for epinephrine in the treatment of anaphylaxis. Antihistamines and corticosteroids can be administered in conjunction with epinephrine, not in place of it.
• Determine whether the patient has risk factors for severe and potentially fatal anaphylaxis, such as delayed administration of epinephrine, asthma, a history of biphasic reactions, or cardiovascular disease.
• For anaphylaxis patients with bronchospasms, administer a β-agonist.
• Patients should be observed for 4 to 8 hours (longer for those with a history of risk factors for severe anaphylaxis).
• Refer patients to an allergist–immunologist upon discharge.

COMMENTARY
Identification of anaphylaxis requires judgment. Abbreviated criteria for anaphylaxis include essentially two organ systems of involvement: skin manifestations of pruritus, flushing, hives, or angioedema; respiratory manifestations of wheezing or stridor; decreased blood pressure; and GI symptoms of vomiting, cramping abdominal pain, or diarrhea. Quick identification and treatment of anaphylaxis is important, as the median time to respiratory or cardiac arrest in food-induced anaphylaxis is only 30 minutes. All offices should stock epinephrine in an obvious place; you may consider use of a prefilled pen, so it is easy to find and to give the correct dose. —NS

Continue for Updated practice parameter: Diagnosing and treating food allergies >>

UPDATED PRACTICE PARAMETER: DIAGNOSING AND TREATING FOOD ALLERGIES
Sampson HA, Aceves S, Bock SA, et al. Food allergy: a practice parameter update—2014. J Allergy Clin Immunol. 2014. pii: S0091-6749(14)00672-1. doi: 10.1016/j.jaci.2014.05.013. [Epub ahead of print]

Over the past decade, health care providers have been confronted with a growing number of patients with suspected food allergies, but data supporting an increase in confirmed allergy cases is limited.

The 2014 practice parameter update on food allergies from the American Academy of Allergy, Asthma & Immunology advises clinicians to keep in mind that self-reported food allergy is more common than proven food allergy; that allergy is more common in children and in patients with other atopic conditions; and that the majority of allergic reactions are from peanuts, tree nuts, fish, shellfish, milk, eggs, wheat, soy, and seeds.

The update of the 2006 guidelines includes 64 new summary statements. Highlights include
• Clinicians should advise patients about the risk for cross-reactions from foods similar to their allergens, such as other tree nuts, vertebrate fish, crustaceans, or milk from cows, goats, or other mammals.
• Patients with a seafood allergy should be advised that they are not at increased risk for a reaction to radiocontrast media.
• Patients with food allergies do not need to be concerned about eating genetically modified foods (GMO), due to current FDA screening requirements to rule out allergenicity.
• Patients with chronic idiopathic urticaria or hyperactivity/attention-deficit disorder should not routinely be advised to avoid food additives.
• Patients with asthma should not routinely be advised to avoid sulfates, unless they have had a previous reaction to them.

The guidelines also include a new section on diagnosing and treating non-IgE-mediated food allergies, such as food-protein–induced enterocolitis syndrome, allergic proctocolitis, enteropathy, eosinophilic esophagitis, and gastroenteritis. 

COMMENTARY
Food allergies are a commonly encountered and difficult area of practice for those of us in primary care. They can be life threatening; yet many patients who are concerned about the possibility of food allergy do not actually have a food allergy, so making an accurate diagnosis is important. Allergic evaluation starts with a careful history, and then laboratory testing can begin with specific IgE testing to foods suspected to have caused the clinical reaction of concern. The IgE results need to be carefully interpreted in light of the clinical context in which they were ordered. Oral food challenge can be helpful in addition to IgE testing. Consultation with an allergy-immunology specialist is often helpful as well. —NS

DOES PEANUT EXPOSURE INCREASE ALLERGY RISK?
Du Toit G, Roberts G, Sayre PH, et al; LEAP Study Team. Randomized trial of peanut consumption in infants at risk for peanut allergy. N Engl J Med. 2015;372(9):803-813. doi: 10.1056/NEJMoa1414850.

Among children at high risk for peanut allergy, early introduction of peanuts significantly reduced the risk for allergy, according to a randomized controlled trial of 640 infants with severe eczema, egg allergy, or both.

Participants ages 4 to 11 months who were high risk—based on severe atopy or allergy to eggs—were tested for sensitivity to peanut extract at baseline and then assigned to either avoid or consume peanuts.

At 60 months of age, skin allergy testing was repeated; the resulting prevalence of peanut allergy was as follows

A greater percentage of the consumption group showed an increase in levels of peanut-specific IgG4 antibody; the avoidance group, elevated titers of peanut specific IgE antibody.

COMMENTARY
The prevalence of peanut allergy in the United States has increased from 0.4% in 1997 to more than 2% in 2010.¹ In 2000, the American Academy of Pediatrics recommended peanut avoidance until age 3 in children at high risk for atopic disease; then in 2008, based on emerging evidence that early introduction of allergenic foods, including peanuts, may decrease the development of allergy, the recommendations were retracted.^2,3 The present study, using a randomized trial design, confirms the paradigm-changing hypothesis that early introduction of allergenic foods decreases the subsequent development of food allergies. The authors of the accompanying editorial state that these data are incontrovertible and that children at high risk for peanut allergy should, under supervision of an allergist, be tested and if skin-prick–­negative for peanut allergy, be started on a peanut protein–containing diet.¹—NS

1. Gruchalla RS, Sampson HA. Preventing peanut allergy through early consumption ready for prime time? N Engl J Med. 2015;372:875-876.
2. Du Toit G, Katz Y, Sasieni P, et al. Early consumption of peanuts in infancy is associated with a low prevalence of peanut allergy. J Allergy Clin Immunol. 2008;122:984-991.
3. Katz Y, Rajuan N, Goldberg MR, et al. Early exposure to cow’s milk protein is protective against IgE-mediated cow’s milk protein allergy. J Allergy Clin Immunol. 2010;126:77-82. 

Continue for Anaphylaxis guideline from AAAAI/ACAAI >>

ANAPHYLAXIS GUIDELINE FROM AAAAI/ACAAI
Campbell RL, Li JT, Nicklas RA, Sadosty AT. Emergency department diagnosis and treatment of anaphylaxis: a practice parameter. Ann Allergy Asthma Immunol. 2014;113(6):599-608. doi: 10.1016/j.anai.2014.10.007.

The American Academy of Allergy, Asthma & Immunology and the American College of Allergy, Asthma, and Immunology guideline for emergency department diagnosis and treatment of anaphylaxis includes the following recommendations
• Carefully and immediately triage and monitor patients who have signs and symptoms of ­anaphylaxis in preparation for epinephrine administration.
• First-line treatment for patients experiencing anaphylaxis is epinephrine. It should be administered intramuscularly in the anterolateral thigh immediately after the diagnosis is made. Epinephrine can be administered every 5 to 15 minutes as needed to control symptoms.
• Do not substitute for epinephrine in the treatment of anaphylaxis. Antihistamines and corticosteroids can be administered in conjunction with epinephrine, not in place of it.
• Determine whether the patient has risk factors for severe and potentially fatal anaphylaxis, such as delayed administration of epinephrine, asthma, a history of biphasic reactions, or cardiovascular disease.
• For anaphylaxis patients with bronchospasms, administer a β-agonist.
• Patients should be observed for 4 to 8 hours (longer for those with a history of risk factors for severe anaphylaxis).
• Refer patients to an allergist–immunologist upon discharge.

COMMENTARY
Identification of anaphylaxis requires judgment. Abbreviated criteria for anaphylaxis include essentially two organ systems of involvement: skin manifestations of pruritus, flushing, hives, or angioedema; respiratory manifestations of wheezing or stridor; decreased blood pressure; and GI symptoms of vomiting, cramping abdominal pain, or diarrhea. Quick identification and treatment of anaphylaxis is important, as the median time to respiratory or cardiac arrest in food-induced anaphylaxis is only 30 minutes. All offices should stock epinephrine in an obvious place; you may consider use of a prefilled pen, so it is easy to find and to give the correct dose. —NS

Continue for Updated practice parameter: Diagnosing and treating food allergies >>

UPDATED PRACTICE PARAMETER: DIAGNOSING AND TREATING FOOD ALLERGIES
Sampson HA, Aceves S, Bock SA, et al. Food allergy: a practice parameter update—2014. J Allergy Clin Immunol. 2014. pii: S0091-6749(14)00672-1. doi: 10.1016/j.jaci.2014.05.013. [Epub ahead of print]

Over the past decade, health care providers have been confronted with a growing number of patients with suspected food allergies, but data supporting an increase in confirmed allergy cases is limited.

The 2014 practice parameter update on food allergies from the American Academy of Allergy, Asthma & Immunology advises clinicians to keep in mind that self-reported food allergy is more common than proven food allergy; that allergy is more common in children and in patients with other atopic conditions; and that the majority of allergic reactions are from peanuts, tree nuts, fish, shellfish, milk, eggs, wheat, soy, and seeds.

The update of the 2006 guidelines includes 64 new summary statements. Highlights include
• Clinicians should advise patients about the risk for cross-reactions from foods similar to their allergens, such as other tree nuts, vertebrate fish, crustaceans, or milk from cows, goats, or other mammals.
• Patients with a seafood allergy should be advised that they are not at increased risk for a reaction to radiocontrast media.
• Patients with food allergies do not need to be concerned about eating genetically modified foods (GMO), due to current FDA screening requirements to rule out allergenicity.
• Patients with chronic idiopathic urticaria or hyperactivity/attention-deficit disorder should not routinely be advised to avoid food additives.
• Patients with asthma should not routinely be advised to avoid sulfates, unless they have had a previous reaction to them.

The guidelines also include a new section on diagnosing and treating non-IgE-mediated food allergies, such as food-protein–induced enterocolitis syndrome, allergic proctocolitis, enteropathy, eosinophilic esophagitis, and gastroenteritis. 

COMMENTARY
Food allergies are a commonly encountered and difficult area of practice for those of us in primary care. They can be life threatening; yet many patients who are concerned about the possibility of food allergy do not actually have a food allergy, so making an accurate diagnosis is important. Allergic evaluation starts with a careful history, and then laboratory testing can begin with specific IgE testing to foods suspected to have caused the clinical reaction of concern. The IgE results need to be carefully interpreted in light of the clinical context in which they were ordered. Oral food challenge can be helpful in addition to IgE testing. Consultation with an allergy-immunology specialist is often helpful as well. —NS

CASE 1 › Steve J, age 43, comes to your clinic looking uncharacteristically glum. He was recently downsized from his job and misses his former colleagues. His job loss has caused a financial strain for his family, and he admits to crying in the shower when he thinks about how his life has turned out. Mr. J tells you that he’s gotten a part-time job, but he’s already called in sick several times. On those sick days he “stayed in bed all day and slept.” He says that when he does go to work, he rarely interacts with his coworkers and his concentration is poor. He tells you he wakes up early in the morning on most days and cannot return to sleep, despite being “tired all the time.” He denies suicidal ideation. Mr. J has never felt this way before, which is what prompted his visit today, but he thinks it is “weak to take a pill to feel better.”

What nonpharmacologic options can you offer him?

CASE 2 › Kerri S is a 27-year-old mother of 2 who comes to your clinic to establish care. She tells you about a recent recurrence of depressed mood, which she feels is due to the stress of moving to the area. She is experiencing sleep-onset insomnia and concentration lapses. Her appetite is poor (self-reported 8-lb weight loss in 2 months) and she lacks the motivation to engage in her daily activities, saying, “I wouldn’t even get out of bed if my kids didn’t need me.” She notes that she is constantly irritable and has completely lost her sex drive. Unlike her prior depressive episode, she has not had any suicidal thoughts. Mrs. S was previously successfully treated with paroxetine, 20 mg/d, but she is not interested in restarting her medication because she is still breastfeeding her toddler.

Are there evidence-based options for her care that do not include medication?

Major depressive disorder (MDD) is widespread and often disabling, affecting nearly 8% of people ages 12 and older at any given time.¹ Thus, it’s crucial to be familiar with the diverse array of evidence-based treatment options from which patients can choose. Although medications are an essential treatment option for patients with severe depression, their value for patients with mild to moderate depression is often limited.² In addition, when antidepressants aren’t combined with psychosocial interventions, discontinuing them is associated with relapse.³

Fortunately, research has found that certain nonpharmacologic interventions—including psychotherapies, somatic therapies, and dietary supplements—can have either therapeutic or adjunctive benefits for treating depression, and can be provided in ways that are time- and cost-effective. This article reviews the evidence supporting several options in each of these treatment categories.

Evidence backs several types of psychotherapy

Several recent meta-analyses suggest that a variety of psychotherapeutic treatments may hold promise for your patients with depression.^4,5 When analyses were limited to larger studies in order to decrease the risk of bias, cognitive behavioral therapy (CBT), interpersonal therapy (IPT), and problem-solving therapy (PST) all resulted in moderate to large improvement in depressive symptoms when compared to wait-list controls.⁴ These findings were echoed in a recent systematic review/meta-analysis that focused on depressed primary care patients. Linde et al⁵ found that the number needed to treat (NNT) to achieve one response (≥50% reduction in score on a depression scale) using any type of psychotherapy was 10, and the NNT to achieve one remission (scoring below a predefined score on a depression scale) was 15.

Psychotherapy can be effective when provided in individual and group settings,⁶ as well as via telephone, the Internet, or software programs.⁷ (For a list of self-help, computerized, and Internet-based resources, see TABLE W1 below.)

CBT has been studied for several decades and there’s strong evidence for its efficacy.⁶ Recent investigations have suggested that CBT delivered in less resource-intensive modes (such as via computer program, Internet, telephone, or videoconferencing) can be as effective as face-to-face CBT.^6,8 CBT has been shown to be helpful for a wide range of patients,⁶ improves outcomes over standard primary care treatment,⁹ and provides a useful adjunct to medication in treatment-resistant severe depression.¹⁰

Behavioral activation (BA), which generally is included as a component of CBT, has received support as an independent treatment, and may produce therapeutic results similar to CBT¹¹ and PST (which we’ll discuss in a bit).¹² The core components of BA are scheduling pleasant activities and increasing the patient’s positive interactions with his or her environment by decreasing avoidance, withdrawal, and inactivity.¹¹ Compared to CBT, BA is easier for clinicians to learn and incorporate into primary care visits, and it may be especially useful as an adjunctive or first-step intervention in outpatient clinics.¹¹ Like CBT, BA can be effective in diverse patient groups^13,14 and can be provided using novel delivery modes, such as via the Internet.¹⁵

Compared to cognitive behavioral therapy, behavioral activation is easier for clinicians to learn and incorporate into primary care visits.

IPT is a supportive, structured, brief therapy (12-16 visits) that focuses on helping patients identify and solve current situation- and relationship-based problems that stem from or contribute to their depression.¹⁶ Enhancing the patient’s interpersonal communication—including improving social skills, assertiveness, and appropriate expression of anger—is typically a component of IPT. Like CBT, IPT has been found to be effective for treating depression when administered in person, in group therapy, or via the phone or Internet, and across a broad age range.^17-19

PST involves teaching patients a structured problem-solving process to decrease interpersonal strain and improve positive life experiences.²⁰ Patients are taught to define their problem, generate and evaluate multiple solutions for it, implement a plan for the solution, and evaluate the results. In addition to being used to successfully treat adults,^4,5 PST has been adapted effectively to treat adolescents¹⁶ and older adults.¹⁸

Somatic therapies are also an option

Exercise has long been considered a possible depression treatment due to its activity on endorphin, monoamine, and cortisol levels and via increased social and general activity. A 2013 Cochrane review of 39 randomized control trials (RCTs; N=2326) assessed whether exercise was effective for treating depression in adults.²¹ Thirty-five trials found a moderate effect size when specifically comparing exercise to no treatment or control interventions. The effect size was reduced, however, when analyses were restricted to trials with the highest methodological quality. There was no statistically significant difference when exercise was compared to pharmacologic treatment or psychotherapy.

Although the amount of research is meager, small but statistically significant improvements have also been found for older adults²² and children/adolescents.²³ There is no consensus on the type, frequency, or intensity of exercise needed to achieve benefit. However, because nearly all studies for all age groups have found that exercise has no adverse psychological effects and substantial positive physical effects, exercise should be recommended to all patients with depression unless contraindicated.

Yoga (both exercise-based and meditation-based) has been evaluated both as a sole treatment and as an adjunctive treatment for depression. Several studies have supported the impact of yoga, particularly in pregnant women,²⁴ although the evidence for its efficacy is inconsistent, with yoga frequently failing to improve upon the outcome of waitlist control.²⁵ The evidence for meditation and mindfulness is more consistently positive, with these interventions equaling or exceeding “treatment as usual,” other psychotherapies, and antidepressants in numerous RCTs.²⁵

Electroconvulsive therapy (ECT) has a substantial evidence base supporting its efficacy.²⁶ ECT has been used for decades, although stigma, cardiac and memory risks, and risks of anesthesia often limit its use. Benefits of ECT include a rapid response relative to pharmacotherapy (>50% of patients respond by the end of the first week of ECT)²⁷ and a strong response in older patients.²⁸

More than half of patients receiving electroconvulsive therapy respond to treatment by the end of the first week.

In repetitive transcranial magnetic stimulation (rTMS), electromagnetic coils are placed on a patient’s head to deliver electromagnetic pulses that stimulate areas of the brain that regulate mood. Although rTMS is not widely available, a growing body of evidence supports its use for treating depression, including a meta-analysis of 34 RCTs that included 1383 patients.²⁹ A multisite RCT (N=190) that was not industry-funded reported a 15% response rate and 60% maintenance of remission at 3 months (NNT=12).³⁰ Although ECT is more effective than rTMS, rTMS appears useful for treatment-resistant depression, and can be used as an adjunctive treatment.^29,31

Dietary supplements may be best used as adjuncts

St. John’s wort (Hypericum perforatum), which contains 2 bioactive ingredients (hyperforin and hypericin), has been effectively used to treat depression.³² A 2008 Cochrane review that was limited to high-quality trials involving patients meeting Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria for depression identified 29 trials (N=5489), of which 18 involved comparisons with placebo and 17 with standard antidepressants.³³ Patients’ depression was rated mild to moderate in 19 studies and moderate to severe in 9 studies. Trials examined 4 to 12 weeks of treatment with Hypericum extracts. This study (and several published since) provides strong clinical evidence supporting the efficacy of St. John’s wort for mild to moderate depression. There is insufficient evidence for its use for severe major depression.³³TABLE 1 contains dosing information for St. John’s wort and other supplements used to treat depression.^34-36 

S-adenosyl-L-methionine (SAMe). In a 2003 systematic review,³⁷ 1600 mg/d of oral SAMe was found to significantly benefit patients with depression in 4 of 5 studies, as did parenteral SAMe (7 of 7 trials). Another review of 48 studies found SAMe was safe and effective for depression.³⁸ SAMe has been proposed for use alone or in combination with an antidepressant.

Low folate levels have been associated with a less robust response to antidepressants in patients with major depressive disorder.

Folate and folic acid. Low folate levels have been associated with a less robust response to antidepressants in patients with MDD,³⁹ and higher folate levels appear to be associated with better antidepressant response.⁴⁰ A 2003 Cochrane review suggested folate might have a role in treating depression.³⁹ A 2009 study found folate supplementation could reduce depressive symptoms for patients with normal baseline folate levels as well as those with low levels.⁴¹ Although the evidence is equivocal, folate augmentation may enhance antidepressant efficacy or improve response/remission rates.^41,42

It seems reasonable to check folate levels in depressed patients, and address deficiencies by instructing patients to increase their dietary intake of folate or to take supplements. Augmenting antidepressants with folate appears to be low-risk and possibly helpful in maintaining remission.

Omega-3 fatty acids. There is substantial evidence that omega-3 fatty acids can prevent and treat depression.^43,44 Recent meta-analyses support the use of omega-3 fatty acids as monotherapy and augmentation, but only formulations that contain a high eicosapentaenoic acid (EPA) to docosahexaenoic acid (DHA) ratio (EPA/DHA 2:1).^45,46 Omega-3 supplementation has been used with positive results in older adults, children,⁴⁷ pregnant women,⁴⁸ and women with postpartum depression.⁴⁹ Although initial research into omega-3 treatment of depression appears promising, augmentation of standard antidepressant therapy may be a good conservative option.

Use a validated tool to monitor response to treatment

You can enhance outcomes for your patients with depression if you schedule routine follow-up visits with them to gauge adherence to recommendations, monitor response to treatment, and increase the intensity of care when response is inadequate.⁵⁰ The most important aspect of monitoring response is to use a standardized instrument that quantifies symptoms at every visit.

The PHQ-9—which is free—has been validated for depression screening and monitoring of treatment response in primary care patients.

The Patient Health Questionnaire 9-item depression assessment (PHQ-9)—which is free—has been validated for depression screening and monitoring of treatment response in primary care patients.⁵¹ A decrease of 5 points on the PHQ-9 is the minimum considered to be clinically significant.⁵² Other well-validated, although lengthier, self-report depression assessment and monitoring instruments include the Beck Depression Inventory-revised and the Zung Depression Scale.

CASE 1 › Mr. J is not enjoying his new job or engaging with new coworkers to replace the positive social experiences he had at his previous job. Together, you set a goal of increasing social involvement by having him make plans to see at least one friend per weekend. Because he indicates that he is unlikely to follow through with a therapy referral, you encourage him to try an online CBT program, start an exercise regimen, or take a SAMe supplement. Mr. Jackson agrees to try the CBT and exercise (moderate intensity, 30 minutes 3-4 times per week), but does not want to take SAMe. He agrees to an assessment of his folate levels, which are normal.

Mr. J starts the online CBT program, which reinforces the exercise and social activity prescription you provided. He establishes a regular exercise routine with a good friend. After one month, his mood has started to improve and he has added regular participation in a hobby (woodworking), as well as volunteer work, which he finds fulfilling. You plan to continue monitoring his depression and his adherence to the treatment plan.

CASE 2 › The recent move has decreased Mrs. S’s interactions with family and long-time friends. Because she had previously expressed interest in exercise, you encourage her to join a local “Mommy and Me” exercise and support group for mothers of toddlers. She is willing to participate in psychotherapy, so you provide a referral to a local therapist with expertise in IPT. You also discuss with Mrs. S the possible benefits of omega-3 fatty acid supplementation, which appears to be safe during breastfeeding.³⁴

Mrs. S begins therapy and exercise classes, but can’t motivate herself to continue either of these activities. She becomes discouraged because she’s unable to easily find an omega-3 fatty acid supplement with the ratio you specified (EPA/DHA 2:1). When you see her 2 weeks later, her depression has worsened.

Because you are concerned her suicidality will return, you revisit the pros and cons of taking an antidepressant. Although small amounts of antidepressants can be passed from mother to infant via breastmilk, the amount varies by specific medication, as do the potential risks. Mrs. S decides to resume taking paroxetine 20 mg/d and eventually, once her motivation improves, she’s able to add psychotherapy and exercise to her maintenance/relapse prevention regimen. After you discuss with her the possibility that B vitamin supplementation may assist in maintenance of remission, she adds L-methylfolate 7.5 mg/day to her regimen.

CORRESPONDENCE 
Michele M. Larzelere, PhD; LSUHSC Department of Family Medicine; 200 W. Esplanade Avenue, Suite 409; Kenner, LA 70065; [email protected]

CASE 1 › Steve J, age 43, comes to your clinic looking uncharacteristically glum. He was recently downsized from his job and misses his former colleagues. His job loss has caused a financial strain for his family, and he admits to crying in the shower when he thinks about how his life has turned out. Mr. J tells you that he’s gotten a part-time job, but he’s already called in sick several times. On those sick days he “stayed in bed all day and slept.” He says that when he does go to work, he rarely interacts with his coworkers and his concentration is poor. He tells you he wakes up early in the morning on most days and cannot return to sleep, despite being “tired all the time.” He denies suicidal ideation. Mr. J has never felt this way before, which is what prompted his visit today, but he thinks it is “weak to take a pill to feel better.”

What nonpharmacologic options can you offer him?

CASE 2 › Kerri S is a 27-year-old mother of 2 who comes to your clinic to establish care. She tells you about a recent recurrence of depressed mood, which she feels is due to the stress of moving to the area. She is experiencing sleep-onset insomnia and concentration lapses. Her appetite is poor (self-reported 8-lb weight loss in 2 months) and she lacks the motivation to engage in her daily activities, saying, “I wouldn’t even get out of bed if my kids didn’t need me.” She notes that she is constantly irritable and has completely lost her sex drive. Unlike her prior depressive episode, she has not had any suicidal thoughts. Mrs. S was previously successfully treated with paroxetine, 20 mg/d, but she is not interested in restarting her medication because she is still breastfeeding her toddler.

Are there evidence-based options for her care that do not include medication?

Major depressive disorder (MDD) is widespread and often disabling, affecting nearly 8% of people ages 12 and older at any given time.¹ Thus, it’s crucial to be familiar with the diverse array of evidence-based treatment options from which patients can choose. Although medications are an essential treatment option for patients with severe depression, their value for patients with mild to moderate depression is often limited.² In addition, when antidepressants aren’t combined with psychosocial interventions, discontinuing them is associated with relapse.³

Fortunately, research has found that certain nonpharmacologic interventions—including psychotherapies, somatic therapies, and dietary supplements—can have either therapeutic or adjunctive benefits for treating depression, and can be provided in ways that are time- and cost-effective. This article reviews the evidence supporting several options in each of these treatment categories.

Evidence backs several types of psychotherapy

Several recent meta-analyses suggest that a variety of psychotherapeutic treatments may hold promise for your patients with depression.^4,5 When analyses were limited to larger studies in order to decrease the risk of bias, cognitive behavioral therapy (CBT), interpersonal therapy (IPT), and problem-solving therapy (PST) all resulted in moderate to large improvement in depressive symptoms when compared to wait-list controls.⁴ These findings were echoed in a recent systematic review/meta-analysis that focused on depressed primary care patients. Linde et al⁵ found that the number needed to treat (NNT) to achieve one response (≥50% reduction in score on a depression scale) using any type of psychotherapy was 10, and the NNT to achieve one remission (scoring below a predefined score on a depression scale) was 15.

Psychotherapy can be effective when provided in individual and group settings,⁶ as well as via telephone, the Internet, or software programs.⁷ (For a list of self-help, computerized, and Internet-based resources, see TABLE W1 below.)

CBT has been studied for several decades and there’s strong evidence for its efficacy.⁶ Recent investigations have suggested that CBT delivered in less resource-intensive modes (such as via computer program, Internet, telephone, or videoconferencing) can be as effective as face-to-face CBT.^6,8 CBT has been shown to be helpful for a wide range of patients,⁶ improves outcomes over standard primary care treatment,⁹ and provides a useful adjunct to medication in treatment-resistant severe depression.¹⁰

Behavioral activation (BA), which generally is included as a component of CBT, has received support as an independent treatment, and may produce therapeutic results similar to CBT¹¹ and PST (which we’ll discuss in a bit).¹² The core components of BA are scheduling pleasant activities and increasing the patient’s positive interactions with his or her environment by decreasing avoidance, withdrawal, and inactivity.¹¹ Compared to CBT, BA is easier for clinicians to learn and incorporate into primary care visits, and it may be especially useful as an adjunctive or first-step intervention in outpatient clinics.¹¹ Like CBT, BA can be effective in diverse patient groups^13,14 and can be provided using novel delivery modes, such as via the Internet.¹⁵

Compared to cognitive behavioral therapy, behavioral activation is easier for clinicians to learn and incorporate into primary care visits.

IPT is a supportive, structured, brief therapy (12-16 visits) that focuses on helping patients identify and solve current situation- and relationship-based problems that stem from or contribute to their depression.¹⁶ Enhancing the patient’s interpersonal communication—including improving social skills, assertiveness, and appropriate expression of anger—is typically a component of IPT. Like CBT, IPT has been found to be effective for treating depression when administered in person, in group therapy, or via the phone or Internet, and across a broad age range.^17-19

PST involves teaching patients a structured problem-solving process to decrease interpersonal strain and improve positive life experiences.²⁰ Patients are taught to define their problem, generate and evaluate multiple solutions for it, implement a plan for the solution, and evaluate the results. In addition to being used to successfully treat adults,^4,5 PST has been adapted effectively to treat adolescents¹⁶ and older adults.¹⁸

Somatic therapies are also an option

Exercise has long been considered a possible depression treatment due to its activity on endorphin, monoamine, and cortisol levels and via increased social and general activity. A 2013 Cochrane review of 39 randomized control trials (RCTs; N=2326) assessed whether exercise was effective for treating depression in adults.²¹ Thirty-five trials found a moderate effect size when specifically comparing exercise to no treatment or control interventions. The effect size was reduced, however, when analyses were restricted to trials with the highest methodological quality. There was no statistically significant difference when exercise was compared to pharmacologic treatment or psychotherapy.

Although the amount of research is meager, small but statistically significant improvements have also been found for older adults²² and children/adolescents.²³ There is no consensus on the type, frequency, or intensity of exercise needed to achieve benefit. However, because nearly all studies for all age groups have found that exercise has no adverse psychological effects and substantial positive physical effects, exercise should be recommended to all patients with depression unless contraindicated.

Yoga (both exercise-based and meditation-based) has been evaluated both as a sole treatment and as an adjunctive treatment for depression. Several studies have supported the impact of yoga, particularly in pregnant women,²⁴ although the evidence for its efficacy is inconsistent, with yoga frequently failing to improve upon the outcome of waitlist control.²⁵ The evidence for meditation and mindfulness is more consistently positive, with these interventions equaling or exceeding “treatment as usual,” other psychotherapies, and antidepressants in numerous RCTs.²⁵

Electroconvulsive therapy (ECT) has a substantial evidence base supporting its efficacy.²⁶ ECT has been used for decades, although stigma, cardiac and memory risks, and risks of anesthesia often limit its use. Benefits of ECT include a rapid response relative to pharmacotherapy (>50% of patients respond by the end of the first week of ECT)²⁷ and a strong response in older patients.²⁸

More than half of patients receiving electroconvulsive therapy respond to treatment by the end of the first week.

In repetitive transcranial magnetic stimulation (rTMS), electromagnetic coils are placed on a patient’s head to deliver electromagnetic pulses that stimulate areas of the brain that regulate mood. Although rTMS is not widely available, a growing body of evidence supports its use for treating depression, including a meta-analysis of 34 RCTs that included 1383 patients.²⁹ A multisite RCT (N=190) that was not industry-funded reported a 15% response rate and 60% maintenance of remission at 3 months (NNT=12).³⁰ Although ECT is more effective than rTMS, rTMS appears useful for treatment-resistant depression, and can be used as an adjunctive treatment.^29,31

Dietary supplements may be best used as adjuncts

St. John’s wort (Hypericum perforatum), which contains 2 bioactive ingredients (hyperforin and hypericin), has been effectively used to treat depression.³² A 2008 Cochrane review that was limited to high-quality trials involving patients meeting Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria for depression identified 29 trials (N=5489), of which 18 involved comparisons with placebo and 17 with standard antidepressants.³³ Patients’ depression was rated mild to moderate in 19 studies and moderate to severe in 9 studies. Trials examined 4 to 12 weeks of treatment with Hypericum extracts. This study (and several published since) provides strong clinical evidence supporting the efficacy of St. John’s wort for mild to moderate depression. There is insufficient evidence for its use for severe major depression.³³TABLE 1 contains dosing information for St. John’s wort and other supplements used to treat depression.^34-36 

S-adenosyl-L-methionine (SAMe). In a 2003 systematic review,³⁷ 1600 mg/d of oral SAMe was found to significantly benefit patients with depression in 4 of 5 studies, as did parenteral SAMe (7 of 7 trials). Another review of 48 studies found SAMe was safe and effective for depression.³⁸ SAMe has been proposed for use alone or in combination with an antidepressant.

Low folate levels have been associated with a less robust response to antidepressants in patients with major depressive disorder.

Folate and folic acid. Low folate levels have been associated with a less robust response to antidepressants in patients with MDD,³⁹ and higher folate levels appear to be associated with better antidepressant response.⁴⁰ A 2003 Cochrane review suggested folate might have a role in treating depression.³⁹ A 2009 study found folate supplementation could reduce depressive symptoms for patients with normal baseline folate levels as well as those with low levels.⁴¹ Although the evidence is equivocal, folate augmentation may enhance antidepressant efficacy or improve response/remission rates.^41,42

It seems reasonable to check folate levels in depressed patients, and address deficiencies by instructing patients to increase their dietary intake of folate or to take supplements. Augmenting antidepressants with folate appears to be low-risk and possibly helpful in maintaining remission.

Omega-3 fatty acids. There is substantial evidence that omega-3 fatty acids can prevent and treat depression.^43,44 Recent meta-analyses support the use of omega-3 fatty acids as monotherapy and augmentation, but only formulations that contain a high eicosapentaenoic acid (EPA) to docosahexaenoic acid (DHA) ratio (EPA/DHA 2:1).^45,46 Omega-3 supplementation has been used with positive results in older adults, children,⁴⁷ pregnant women,⁴⁸ and women with postpartum depression.⁴⁹ Although initial research into omega-3 treatment of depression appears promising, augmentation of standard antidepressant therapy may be a good conservative option.

Use a validated tool to monitor response to treatment

You can enhance outcomes for your patients with depression if you schedule routine follow-up visits with them to gauge adherence to recommendations, monitor response to treatment, and increase the intensity of care when response is inadequate.⁵⁰ The most important aspect of monitoring response is to use a standardized instrument that quantifies symptoms at every visit.

The PHQ-9—which is free—has been validated for depression screening and monitoring of treatment response in primary care patients.

The Patient Health Questionnaire 9-item depression assessment (PHQ-9)—which is free—has been validated for depression screening and monitoring of treatment response in primary care patients.⁵¹ A decrease of 5 points on the PHQ-9 is the minimum considered to be clinically significant.⁵² Other well-validated, although lengthier, self-report depression assessment and monitoring instruments include the Beck Depression Inventory-revised and the Zung Depression Scale.

CASE 1 › Mr. J is not enjoying his new job or engaging with new coworkers to replace the positive social experiences he had at his previous job. Together, you set a goal of increasing social involvement by having him make plans to see at least one friend per weekend. Because he indicates that he is unlikely to follow through with a therapy referral, you encourage him to try an online CBT program, start an exercise regimen, or take a SAMe supplement. Mr. Jackson agrees to try the CBT and exercise (moderate intensity, 30 minutes 3-4 times per week), but does not want to take SAMe. He agrees to an assessment of his folate levels, which are normal.

Mr. J starts the online CBT program, which reinforces the exercise and social activity prescription you provided. He establishes a regular exercise routine with a good friend. After one month, his mood has started to improve and he has added regular participation in a hobby (woodworking), as well as volunteer work, which he finds fulfilling. You plan to continue monitoring his depression and his adherence to the treatment plan.

CASE 2 › The recent move has decreased Mrs. S’s interactions with family and long-time friends. Because she had previously expressed interest in exercise, you encourage her to join a local “Mommy and Me” exercise and support group for mothers of toddlers. She is willing to participate in psychotherapy, so you provide a referral to a local therapist with expertise in IPT. You also discuss with Mrs. S the possible benefits of omega-3 fatty acid supplementation, which appears to be safe during breastfeeding.³⁴

Mrs. S begins therapy and exercise classes, but can’t motivate herself to continue either of these activities. She becomes discouraged because she’s unable to easily find an omega-3 fatty acid supplement with the ratio you specified (EPA/DHA 2:1). When you see her 2 weeks later, her depression has worsened.

Because you are concerned her suicidality will return, you revisit the pros and cons of taking an antidepressant. Although small amounts of antidepressants can be passed from mother to infant via breastmilk, the amount varies by specific medication, as do the potential risks. Mrs. S decides to resume taking paroxetine 20 mg/d and eventually, once her motivation improves, she’s able to add psychotherapy and exercise to her maintenance/relapse prevention regimen. After you discuss with her the possibility that B vitamin supplementation may assist in maintenance of remission, she adds L-methylfolate 7.5 mg/day to her regimen.

CORRESPONDENCE 
Michele M. Larzelere, PhD; LSUHSC Department of Family Medicine; 200 W. Esplanade Avenue, Suite 409; Kenner, LA 70065; [email protected]

CASE 1 › Steve J, age 43, comes to your clinic looking uncharacteristically glum. He was recently downsized from his job and misses his former colleagues. His job loss has caused a financial strain for his family, and he admits to crying in the shower when he thinks about how his life has turned out. Mr. J tells you that he’s gotten a part-time job, but he’s already called in sick several times. On those sick days he “stayed in bed all day and slept.” He says that when he does go to work, he rarely interacts with his coworkers and his concentration is poor. He tells you he wakes up early in the morning on most days and cannot return to sleep, despite being “tired all the time.” He denies suicidal ideation. Mr. J has never felt this way before, which is what prompted his visit today, but he thinks it is “weak to take a pill to feel better.”

What nonpharmacologic options can you offer him?

CASE 2 › Kerri S is a 27-year-old mother of 2 who comes to your clinic to establish care. She tells you about a recent recurrence of depressed mood, which she feels is due to the stress of moving to the area. She is experiencing sleep-onset insomnia and concentration lapses. Her appetite is poor (self-reported 8-lb weight loss in 2 months) and she lacks the motivation to engage in her daily activities, saying, “I wouldn’t even get out of bed if my kids didn’t need me.” She notes that she is constantly irritable and has completely lost her sex drive. Unlike her prior depressive episode, she has not had any suicidal thoughts. Mrs. S was previously successfully treated with paroxetine, 20 mg/d, but she is not interested in restarting her medication because she is still breastfeeding her toddler.

Are there evidence-based options for her care that do not include medication?

Major depressive disorder (MDD) is widespread and often disabling, affecting nearly 8% of people ages 12 and older at any given time.¹ Thus, it’s crucial to be familiar with the diverse array of evidence-based treatment options from which patients can choose. Although medications are an essential treatment option for patients with severe depression, their value for patients with mild to moderate depression is often limited.² In addition, when antidepressants aren’t combined with psychosocial interventions, discontinuing them is associated with relapse.³

Fortunately, research has found that certain nonpharmacologic interventions—including psychotherapies, somatic therapies, and dietary supplements—can have either therapeutic or adjunctive benefits for treating depression, and can be provided in ways that are time- and cost-effective. This article reviews the evidence supporting several options in each of these treatment categories.

Evidence backs several types of psychotherapy

Several recent meta-analyses suggest that a variety of psychotherapeutic treatments may hold promise for your patients with depression.^4,5 When analyses were limited to larger studies in order to decrease the risk of bias, cognitive behavioral therapy (CBT), interpersonal therapy (IPT), and problem-solving therapy (PST) all resulted in moderate to large improvement in depressive symptoms when compared to wait-list controls.⁴ These findings were echoed in a recent systematic review/meta-analysis that focused on depressed primary care patients. Linde et al⁵ found that the number needed to treat (NNT) to achieve one response (≥50% reduction in score on a depression scale) using any type of psychotherapy was 10, and the NNT to achieve one remission (scoring below a predefined score on a depression scale) was 15.

Psychotherapy can be effective when provided in individual and group settings,⁶ as well as via telephone, the Internet, or software programs.⁷ (For a list of self-help, computerized, and Internet-based resources, see TABLE W1 below.)

CBT has been studied for several decades and there’s strong evidence for its efficacy.⁶ Recent investigations have suggested that CBT delivered in less resource-intensive modes (such as via computer program, Internet, telephone, or videoconferencing) can be as effective as face-to-face CBT.^6,8 CBT has been shown to be helpful for a wide range of patients,⁶ improves outcomes over standard primary care treatment,⁹ and provides a useful adjunct to medication in treatment-resistant severe depression.¹⁰

Behavioral activation (BA), which generally is included as a component of CBT, has received support as an independent treatment, and may produce therapeutic results similar to CBT¹¹ and PST (which we’ll discuss in a bit).¹² The core components of BA are scheduling pleasant activities and increasing the patient’s positive interactions with his or her environment by decreasing avoidance, withdrawal, and inactivity.¹¹ Compared to CBT, BA is easier for clinicians to learn and incorporate into primary care visits, and it may be especially useful as an adjunctive or first-step intervention in outpatient clinics.¹¹ Like CBT, BA can be effective in diverse patient groups^13,14 and can be provided using novel delivery modes, such as via the Internet.¹⁵

Compared to cognitive behavioral therapy, behavioral activation is easier for clinicians to learn and incorporate into primary care visits.

IPT is a supportive, structured, brief therapy (12-16 visits) that focuses on helping patients identify and solve current situation- and relationship-based problems that stem from or contribute to their depression.¹⁶ Enhancing the patient’s interpersonal communication—including improving social skills, assertiveness, and appropriate expression of anger—is typically a component of IPT. Like CBT, IPT has been found to be effective for treating depression when administered in person, in group therapy, or via the phone or Internet, and across a broad age range.^17-19

PST involves teaching patients a structured problem-solving process to decrease interpersonal strain and improve positive life experiences.²⁰ Patients are taught to define their problem, generate and evaluate multiple solutions for it, implement a plan for the solution, and evaluate the results. In addition to being used to successfully treat adults,^4,5 PST has been adapted effectively to treat adolescents¹⁶ and older adults.¹⁸

Somatic therapies are also an option

Exercise has long been considered a possible depression treatment due to its activity on endorphin, monoamine, and cortisol levels and via increased social and general activity. A 2013 Cochrane review of 39 randomized control trials (RCTs; N=2326) assessed whether exercise was effective for treating depression in adults.²¹ Thirty-five trials found a moderate effect size when specifically comparing exercise to no treatment or control interventions. The effect size was reduced, however, when analyses were restricted to trials with the highest methodological quality. There was no statistically significant difference when exercise was compared to pharmacologic treatment or psychotherapy.

Although the amount of research is meager, small but statistically significant improvements have also been found for older adults²² and children/adolescents.²³ There is no consensus on the type, frequency, or intensity of exercise needed to achieve benefit. However, because nearly all studies for all age groups have found that exercise has no adverse psychological effects and substantial positive physical effects, exercise should be recommended to all patients with depression unless contraindicated.

Yoga (both exercise-based and meditation-based) has been evaluated both as a sole treatment and as an adjunctive treatment for depression. Several studies have supported the impact of yoga, particularly in pregnant women,²⁴ although the evidence for its efficacy is inconsistent, with yoga frequently failing to improve upon the outcome of waitlist control.²⁵ The evidence for meditation and mindfulness is more consistently positive, with these interventions equaling or exceeding “treatment as usual,” other psychotherapies, and antidepressants in numerous RCTs.²⁵

Electroconvulsive therapy (ECT) has a substantial evidence base supporting its efficacy.²⁶ ECT has been used for decades, although stigma, cardiac and memory risks, and risks of anesthesia often limit its use. Benefits of ECT include a rapid response relative to pharmacotherapy (>50% of patients respond by the end of the first week of ECT)²⁷ and a strong response in older patients.²⁸

More than half of patients receiving electroconvulsive therapy respond to treatment by the end of the first week.

In repetitive transcranial magnetic stimulation (rTMS), electromagnetic coils are placed on a patient’s head to deliver electromagnetic pulses that stimulate areas of the brain that regulate mood. Although rTMS is not widely available, a growing body of evidence supports its use for treating depression, including a meta-analysis of 34 RCTs that included 1383 patients.²⁹ A multisite RCT (N=190) that was not industry-funded reported a 15% response rate and 60% maintenance of remission at 3 months (NNT=12).³⁰ Although ECT is more effective than rTMS, rTMS appears useful for treatment-resistant depression, and can be used as an adjunctive treatment.^29,31

Dietary supplements may be best used as adjuncts

St. John’s wort (Hypericum perforatum), which contains 2 bioactive ingredients (hyperforin and hypericin), has been effectively used to treat depression.³² A 2008 Cochrane review that was limited to high-quality trials involving patients meeting Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria for depression identified 29 trials (N=5489), of which 18 involved comparisons with placebo and 17 with standard antidepressants.³³ Patients’ depression was rated mild to moderate in 19 studies and moderate to severe in 9 studies. Trials examined 4 to 12 weeks of treatment with Hypericum extracts. This study (and several published since) provides strong clinical evidence supporting the efficacy of St. John’s wort for mild to moderate depression. There is insufficient evidence for its use for severe major depression.³³TABLE 1 contains dosing information for St. John’s wort and other supplements used to treat depression.^34-36 

S-adenosyl-L-methionine (SAMe). In a 2003 systematic review,³⁷ 1600 mg/d of oral SAMe was found to significantly benefit patients with depression in 4 of 5 studies, as did parenteral SAMe (7 of 7 trials). Another review of 48 studies found SAMe was safe and effective for depression.³⁸ SAMe has been proposed for use alone or in combination with an antidepressant.

Low folate levels have been associated with a less robust response to antidepressants in patients with major depressive disorder.

Folate and folic acid. Low folate levels have been associated with a less robust response to antidepressants in patients with MDD,³⁹ and higher folate levels appear to be associated with better antidepressant response.⁴⁰ A 2003 Cochrane review suggested folate might have a role in treating depression.³⁹ A 2009 study found folate supplementation could reduce depressive symptoms for patients with normal baseline folate levels as well as those with low levels.⁴¹ Although the evidence is equivocal, folate augmentation may enhance antidepressant efficacy or improve response/remission rates.^41,42

It seems reasonable to check folate levels in depressed patients, and address deficiencies by instructing patients to increase their dietary intake of folate or to take supplements. Augmenting antidepressants with folate appears to be low-risk and possibly helpful in maintaining remission.

Omega-3 fatty acids. There is substantial evidence that omega-3 fatty acids can prevent and treat depression.^43,44 Recent meta-analyses support the use of omega-3 fatty acids as monotherapy and augmentation, but only formulations that contain a high eicosapentaenoic acid (EPA) to docosahexaenoic acid (DHA) ratio (EPA/DHA 2:1).^45,46 Omega-3 supplementation has been used with positive results in older adults, children,⁴⁷ pregnant women,⁴⁸ and women with postpartum depression.⁴⁹ Although initial research into omega-3 treatment of depression appears promising, augmentation of standard antidepressant therapy may be a good conservative option.

Use a validated tool to monitor response to treatment

You can enhance outcomes for your patients with depression if you schedule routine follow-up visits with them to gauge adherence to recommendations, monitor response to treatment, and increase the intensity of care when response is inadequate.⁵⁰ The most important aspect of monitoring response is to use a standardized instrument that quantifies symptoms at every visit.

The PHQ-9—which is free—has been validated for depression screening and monitoring of treatment response in primary care patients.

The Patient Health Questionnaire 9-item depression assessment (PHQ-9)—which is free—has been validated for depression screening and monitoring of treatment response in primary care patients.⁵¹ A decrease of 5 points on the PHQ-9 is the minimum considered to be clinically significant.⁵² Other well-validated, although lengthier, self-report depression assessment and monitoring instruments include the Beck Depression Inventory-revised and the Zung Depression Scale.

CASE 1 › Mr. J is not enjoying his new job or engaging with new coworkers to replace the positive social experiences he had at his previous job. Together, you set a goal of increasing social involvement by having him make plans to see at least one friend per weekend. Because he indicates that he is unlikely to follow through with a therapy referral, you encourage him to try an online CBT program, start an exercise regimen, or take a SAMe supplement. Mr. Jackson agrees to try the CBT and exercise (moderate intensity, 30 minutes 3-4 times per week), but does not want to take SAMe. He agrees to an assessment of his folate levels, which are normal.

Mr. J starts the online CBT program, which reinforces the exercise and social activity prescription you provided. He establishes a regular exercise routine with a good friend. After one month, his mood has started to improve and he has added regular participation in a hobby (woodworking), as well as volunteer work, which he finds fulfilling. You plan to continue monitoring his depression and his adherence to the treatment plan.

CASE 2 › The recent move has decreased Mrs. S’s interactions with family and long-time friends. Because she had previously expressed interest in exercise, you encourage her to join a local “Mommy and Me” exercise and support group for mothers of toddlers. She is willing to participate in psychotherapy, so you provide a referral to a local therapist with expertise in IPT. You also discuss with Mrs. S the possible benefits of omega-3 fatty acid supplementation, which appears to be safe during breastfeeding.³⁴

Mrs. S begins therapy and exercise classes, but can’t motivate herself to continue either of these activities. She becomes discouraged because she’s unable to easily find an omega-3 fatty acid supplement with the ratio you specified (EPA/DHA 2:1). When you see her 2 weeks later, her depression has worsened.

Because you are concerned her suicidality will return, you revisit the pros and cons of taking an antidepressant. Although small amounts of antidepressants can be passed from mother to infant via breastmilk, the amount varies by specific medication, as do the potential risks. Mrs. S decides to resume taking paroxetine 20 mg/d and eventually, once her motivation improves, she’s able to add psychotherapy and exercise to her maintenance/relapse prevention regimen. After you discuss with her the possibility that B vitamin supplementation may assist in maintenance of remission, she adds L-methylfolate 7.5 mg/day to her regimen.

CORRESPONDENCE 
Michele M. Larzelere, PhD; LSUHSC Department of Family Medicine; 200 W. Esplanade Avenue, Suite 409; Kenner, LA 70065; [email protected]

A 6-year-old girl was brought to the emergency department (ED) by her mother after the child bumped her head while playing. While the physician examined the child’s head, the mother remarked that her daughter had recently developed bruises that appeared suddenly and only after minor, if any, known trauma. The ED physician determined that the child’s bump to the head was nothing to worry about, attributed the bruising to the child being a “healthy, active 6-year-old,” and sent her home.

Two days later, the child was brought to our office because the mother was still concerned about her daughter’s easy bruising. The mother pointed out ecchymosis scattered across her daughter’s extremities and torso. The child denied any pain or other complaints, including any active or recurrent bleeding. Upon further questioning, the mother mentioned that her daughter had recovered from a coldlike illness several weeks earlier.

THE DIAGNOSIS
We ordered a complete blood count (CBC) and peripheral smear, which were normal except for the platelet count, which was 7,000/μL (normal, 150,000-450,000/μL). Based on the child’s easy bruising and isolated thrombocytopenia, we diagnosed immune thrombocytopenia, which is also known as idiopathic thrombocytopenic purpura (ITP).

DISCUSSION
In ITP, autoantibodies are directed against platelets, leading to their sequestration and destruction in the spleen and a resultant drop in platelet count.¹ Children with ITP typically present between the ages of 2 and 10, with a peak incidence between 2 and 5.² The incidence is estimated to be as high as 8 per 100,000 children.³ However, this estimate primarily reflects symptomatic children, and the true incidence of childhood ITP may be much higher because asymptomatic children may not be brought in to see a doctor.

For the majority of patients, ITP resolves within three months. However, for 20% to 30% of patients, thrombocytopenia will last beyond six months, with or without treatment.⁴ In 1% of cases, patients will have a recurrence of ITP.³

In addition to easy bruising, nearly all patients who present with possible ITP will complain of cutaneous bleeding, typically a nose bleed or bleeding in the oral cavity.² Upon questioning, 60% of patients will report a history of recent infection.⁴ Not surprisingly, bleeding severity correlates inversely with platelet count; severe bleeding is seen in patients with a platelet count < 10,000/μL.

While rare, the more worrisome complications include intracranial hemorrhage (incidence, 0.1% to 0.8%) and other serious hemorrhages that would require transfusion (estimated incidence, 2.9%).²

Vast differential seen in child bruising
When a child presents with bruising, perform a thorough history, including birth and prenatal course, as well as a physical to exclude other potential causes, such as physical abuse, use of herbal remedies or other natural supplements that may not be disclosed as medication, or even environmental exposure. When bruising is present in a child who has isolated thrombocytopenia, the diagnosis of ITP may be straightforward. However, many conditions share thrombocytopenia in their disease process and should be considered in the differential diagnosis of a child who you suspect may have ITP.
• Suspect physical abuse in a bruised child who does not have thrombocytopenia, whose mood is flat or depressed, or who has experienced recurrent injuries or bruising.
• Leukemia, particularly acute lymphoblastic leukemia (ALL), the predominant leukemia found in children, should be ruled out as well. Symptoms that may distinguish a child with ALL from one with ITP include fever, weight loss, and joint pain, as well as signs such as lymphadenopathy, hepatosplenomegaly, anemia, and leukocytosis. A peripheral smear may be ordered to help confirm or exclude a diagnosis of ALL, should any of the above be present in a child with thrombocytopenia.⁵ It may show lymphoblasts and/or atypical cells in a patient with ALL.⁵
• Infections should also be included in a differential when a patient is suspected of having ITP, particularly if he or she has systemic symptoms. Viral infections that may cause thrombocytopenia include mononucleosis, dengue virus, human herpesvirus-6, and HIV.^6,7

ITP often follows an infection, and the incidence of ITP may be higher during winter months, when infections are more common. However, infection may not always be the cause of ITP. Sepsis may also lead to thrombocytopenia, but a child with sepsis would present very differently from one who has only ITP. A septic child would present with acute illness and signs and symptoms of severe systemic illness, such as high fever, altered mental status, tachycardia, pallor, diaphoresis, and hypotension.
• Drug-induced thrombocytopenia (DIT) should be considered in any child who is taking or recently took a medication that may cause thrombocytopenia. Medications that can cause thrombocytopenia include heparin, quinine, vancomycin, trimethoprim-sulfamethoxazole, rifampin, carbamazepine, phenytoin, piperacillin, linezolid, and valproic acid.8 The measles, mumps, and rubella vaccine also can cause thrombocytopenia.8 A careful medication history may determine if the child is at risk for DIT.
• To narrow the differential, obtain a CBC and peripheral smear when evaluating a patient you suspect may have ITP5 (strength of recommendation [SOR]: A). A CBC will determine the patient’s platelet count, and a peripheral smear should be obtained to exclude other possible diagnoses.⁵

If there are any questions regarding the results of a peripheral smear, it may be necessary to perform a bone marrow aspiration. This, however, is not usually necessary in an otherwise typical case of ITP.9 Bone marrow aspiration may, however, be necessary to reevaluate the initial diagnosis for a child who does not respond to treatment for ITP.

Continue for corticosteroids, IVIg are usually effective >>

Corticosteroids, IVIg are usually effective
The first step in treating a patient with ITP is to limit the risk for further injury or bleeding by stopping NSAIDs or ending participation in contact sports^2,9 (SOR: C). The next step is to determine if pharmacologic therapy is warranted.

Medication, if necessary, is the mainstay of treatment for patients with ITP, particularly those experiencing significant bleeding.² Corticosteroids, intravenous (IV) immunoglobulin (IVIg), and IV Rho(D) immune globulin (also known as anti-D) are the medications typically used to treat a child with ITP, depending on availability of the drugs, bleeding or bleeding risk, and convenience of dosing. For example, corticosteroids can be used orally or IV, whereas IVIg and IV Rho(D) may not be readily available in some treatment settings.

Corticosteroids have been shown to more rapidly increase platelet count compared to placebo and appear to have a dose-related effect.^10,11 Oral prednisone can be dosed at 1 to 2 mg/kg/d for 14 days and then tapered over the course of one week^10,11 or one may prescribe 4 mg/kg/d for four days.^10,11 IV methylprednisolone typically is given at 30 mg/kg/d for three to four days.⁹

IVIg may have greater efficacy than corticosteroids in treating ITP, but it may also cause adverse effects, including nausea, headache, and fever. IVIg can be administered as a single 800 to 1,000 mg/kg dose, or as a daily 400 mg/kg dose form five days; higher doses should be reserved for patients with severe bleeding.¹²

If ITP persists despite the use of corticosteroids or IVIg, IV Rho(D) Ig may be used in patients with Rho(D)-positive blood at a single dose of 25 to 50 μg/kg, with additional doses administered on separate days as required to elevate platelet count. However, only Rho(D)-positive patients are eligible for anti-D treatment.

The response rates/times and adverse effects of common treatments for ITP are summarized in the Table.⁹ A small randomized study found that oral methylprednisolone 30 mg/kg/d for three days followed by 20 mg/kg/d for an additional four days was comparable to IVIg 0.4 g/kg/d for five days.¹¹ A different study that compared oral methylprednisolone (30 mg/kg/d or 50 mg/kg/d for seven days) and IVIg (0.5 g/kg/d for five days) found no difference in outcomes among the three treatments.¹³ One advantage, though, of IVIg is that it can be administered as a single IV dose, rather than multiple doses over several weeks, as is the case with oral prednisone.^9,11-13
• Follow platelet counts closely. Patients with ITP should have their platelet counts monitored at least once weekly and as often as twice weekly. The frequency of monitoring may be tapered depending on an individual patient’s response to treatment and the severity of the thrombocytopenia.¹⁴
• We referred our patient to a nearby children’s hospital, where a repeat CBC  showed her platelets had decreased to 3,000/μL. She received a six-hour infusion of IVIg and was discharged with instructions to have her CBC closely monitored. Her platelets remained stable until four weeks later, when they decreased from 102,000/μL to 71,000/μL. She received a second infusion of IVIg as an outpatient.

Soon after, she presented to our ED with a headache, nausea, and fever of 102°F. CT of her head was normal; a repeat CBC showed no elevation in white blood cells, but her hemoglobin had decreased from 11.9 g/dL to 9.7 g/dL. (Her platelets were 254,000/μL.) The patient’s complaints were likely adverse effects of the IVIg. The CBC abnormalities, fever, headache, and malaise resolved shortly thereafter, and the patient remains asymptomatic with no recurrence of ITP.

Continue for the takeaway >>

THE TAKEAWAY
Suspect ITP in a child who bruises easily and who also has thrombocytopenia. Order a CBC and peripheral blood smear to rule out other potential illnesses. Pharmacotherapy, if needed, typically consists of an oral or IV corticosteroid or IVIg; IV Rho(D) Ig may be used in Rho(D)-positive patients who don’t respond to other treatments. Patients with ITP should have their platelet count monitored at least once weekly until platelets have increased to 150,000/μL or higher. Frequency of monitoring may be reduced as the clinical picture improves and the patient remains stable. More frequent monitoring may be necessary based on severity, complications, and response to treatment.

REFERENCES
1. Johnsen J. Pathogenesis in immune thrombocytopenia: new insights. Hematology Am Soc Hematol Educ Program. 2012;2012:306-312.
2. Kühne T, Buchanan GR, Zimmerman S, et al; Intercontinental Childhood ITP Study Group. A prospective comparative study of 2540 infants and children with newly diagnosed idiopathic thrombocytopenic purpura (ITP) from the Intercontinental Childhood ITP Study Group. J Pediatr. 2003;143:605-608.
3. Kurtzberg J, Stockman JA III. Idiopathic autoimmune thrombocytopenic purpura. Adv Pediatr. 1994;41:111-134.
4. Zeller B, Rajantie J, Hedlund-Treutiger I, et al. Childhood idiopathic thrombocytopenic purpura in the Nordic countries: epidemiology and predictors of chronic disease. Acta Paediatr. 2005;94:178-184.
5. Margolin JF, Steuber CP, Poplack DG. Acute lymphoblastic leukemia. In: Pizzo PA, Poplack DG, eds. Principles and Practice of Pediatric Oncology. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001:317-321.
6. Hashimoto H, Maruyama H, Fujimoto K, et al. Hematologic findings associated with thrombocytopenia during the acute phase of exanthem subitum confirmed by primary human herpesvirus-6 infection. J Pediatr Hematol Oncol. 2002;24:211-214.
7. La Russa VF, Innis BL. Mechanisms of dengue virus-induced bone marrow suppression. Baillieres Clin Haematol. 1995;8:249-270.
8. Aster RH, Curtis BR, McFarland JG, et al. Drug-induced immune thrombocytopenia: pathogenesis, diagnosis, and management. Thromb Haemost. 2009;7:911-918.
9. Provan D, Stasi R, Newland AC, et al. International consensus report on the investigation and management of primary immune thrombocytopenia. Blood. 2010;115:168-186.
10. Bellucci S, Charpak Y, Chastang C, et al. Low doses v conventional doses of corticoids in immune thrombocytopenic purpura (ITP): results of a randomized clinical trial in 160 children, 223 adults. Blood. 1988;71:1165-1169.
11. Ozsoylu S, Sayli TR, Oztürk G. Oral megadose methylprednisolone versus intravenous immunoglobulin for acute childhood idiopathic thrombocytopenic purpura. Pediatr Hematol Oncol. 1993;10:317-321.
12. Beck CE, Nathan PC, Parkin PC, et al. Corticosteroids versus intravenous immune globulin for the treatment of acute immune thrombocytopenic purpura in children: a systematic review and meta-analysis of randomized controlled trials. J Pediatr. 2005;147:521-527.
13. Albayrak D, Islek I, Kalaycí AG, et al. Acute immune thrombocytopenic purpura: a comparative study of very high oral doses of methylprednisolone and intravenously administered immune globulin. J Pediatr. 1994;125(6 pt 1):1004-1007.
14. Tarantino MD, Madden RM, Fennewald DL, et al. Treatment of childhood acute immune thrombocytopenic purpura with anti-D immune globulin or pooled immune globulin. J Pediatr. 1999;134:21-26.

A 6-year-old girl was brought to the emergency department (ED) by her mother after the child bumped her head while playing. While the physician examined the child’s head, the mother remarked that her daughter had recently developed bruises that appeared suddenly and only after minor, if any, known trauma. The ED physician determined that the child’s bump to the head was nothing to worry about, attributed the bruising to the child being a “healthy, active 6-year-old,” and sent her home.

Two days later, the child was brought to our office because the mother was still concerned about her daughter’s easy bruising. The mother pointed out ecchymosis scattered across her daughter’s extremities and torso. The child denied any pain or other complaints, including any active or recurrent bleeding. Upon further questioning, the mother mentioned that her daughter had recovered from a coldlike illness several weeks earlier.

THE DIAGNOSIS
We ordered a complete blood count (CBC) and peripheral smear, which were normal except for the platelet count, which was 7,000/μL (normal, 150,000-450,000/μL). Based on the child’s easy bruising and isolated thrombocytopenia, we diagnosed immune thrombocytopenia, which is also known as idiopathic thrombocytopenic purpura (ITP).

DISCUSSION
In ITP, autoantibodies are directed against platelets, leading to their sequestration and destruction in the spleen and a resultant drop in platelet count.¹ Children with ITP typically present between the ages of 2 and 10, with a peak incidence between 2 and 5.² The incidence is estimated to be as high as 8 per 100,000 children.³ However, this estimate primarily reflects symptomatic children, and the true incidence of childhood ITP may be much higher because asymptomatic children may not be brought in to see a doctor.

For the majority of patients, ITP resolves within three months. However, for 20% to 30% of patients, thrombocytopenia will last beyond six months, with or without treatment.⁴ In 1% of cases, patients will have a recurrence of ITP.³

In addition to easy bruising, nearly all patients who present with possible ITP will complain of cutaneous bleeding, typically a nose bleed or bleeding in the oral cavity.² Upon questioning, 60% of patients will report a history of recent infection.⁴ Not surprisingly, bleeding severity correlates inversely with platelet count; severe bleeding is seen in patients with a platelet count < 10,000/μL.

While rare, the more worrisome complications include intracranial hemorrhage (incidence, 0.1% to 0.8%) and other serious hemorrhages that would require transfusion (estimated incidence, 2.9%).²

Vast differential seen in child bruising
When a child presents with bruising, perform a thorough history, including birth and prenatal course, as well as a physical to exclude other potential causes, such as physical abuse, use of herbal remedies or other natural supplements that may not be disclosed as medication, or even environmental exposure. When bruising is present in a child who has isolated thrombocytopenia, the diagnosis of ITP may be straightforward. However, many conditions share thrombocytopenia in their disease process and should be considered in the differential diagnosis of a child who you suspect may have ITP.
• Suspect physical abuse in a bruised child who does not have thrombocytopenia, whose mood is flat or depressed, or who has experienced recurrent injuries or bruising.
• Leukemia, particularly acute lymphoblastic leukemia (ALL), the predominant leukemia found in children, should be ruled out as well. Symptoms that may distinguish a child with ALL from one with ITP include fever, weight loss, and joint pain, as well as signs such as lymphadenopathy, hepatosplenomegaly, anemia, and leukocytosis. A peripheral smear may be ordered to help confirm or exclude a diagnosis of ALL, should any of the above be present in a child with thrombocytopenia.⁵ It may show lymphoblasts and/or atypical cells in a patient with ALL.⁵
• Infections should also be included in a differential when a patient is suspected of having ITP, particularly if he or she has systemic symptoms. Viral infections that may cause thrombocytopenia include mononucleosis, dengue virus, human herpesvirus-6, and HIV.^6,7

ITP often follows an infection, and the incidence of ITP may be higher during winter months, when infections are more common. However, infection may not always be the cause of ITP. Sepsis may also lead to thrombocytopenia, but a child with sepsis would present very differently from one who has only ITP. A septic child would present with acute illness and signs and symptoms of severe systemic illness, such as high fever, altered mental status, tachycardia, pallor, diaphoresis, and hypotension.
• Drug-induced thrombocytopenia (DIT) should be considered in any child who is taking or recently took a medication that may cause thrombocytopenia. Medications that can cause thrombocytopenia include heparin, quinine, vancomycin, trimethoprim-sulfamethoxazole, rifampin, carbamazepine, phenytoin, piperacillin, linezolid, and valproic acid.8 The measles, mumps, and rubella vaccine also can cause thrombocytopenia.8 A careful medication history may determine if the child is at risk for DIT.
• To narrow the differential, obtain a CBC and peripheral smear when evaluating a patient you suspect may have ITP5 (strength of recommendation [SOR]: A). A CBC will determine the patient’s platelet count, and a peripheral smear should be obtained to exclude other possible diagnoses.⁵

If there are any questions regarding the results of a peripheral smear, it may be necessary to perform a bone marrow aspiration. This, however, is not usually necessary in an otherwise typical case of ITP.9 Bone marrow aspiration may, however, be necessary to reevaluate the initial diagnosis for a child who does not respond to treatment for ITP.

Continue for corticosteroids, IVIg are usually effective >>

Corticosteroids, IVIg are usually effective
The first step in treating a patient with ITP is to limit the risk for further injury or bleeding by stopping NSAIDs or ending participation in contact sports^2,9 (SOR: C). The next step is to determine if pharmacologic therapy is warranted.

Medication, if necessary, is the mainstay of treatment for patients with ITP, particularly those experiencing significant bleeding.² Corticosteroids, intravenous (IV) immunoglobulin (IVIg), and IV Rho(D) immune globulin (also known as anti-D) are the medications typically used to treat a child with ITP, depending on availability of the drugs, bleeding or bleeding risk, and convenience of dosing. For example, corticosteroids can be used orally or IV, whereas IVIg and IV Rho(D) may not be readily available in some treatment settings.

Corticosteroids have been shown to more rapidly increase platelet count compared to placebo and appear to have a dose-related effect.^10,11 Oral prednisone can be dosed at 1 to 2 mg/kg/d for 14 days and then tapered over the course of one week^10,11 or one may prescribe 4 mg/kg/d for four days.^10,11 IV methylprednisolone typically is given at 30 mg/kg/d for three to four days.⁹

IVIg may have greater efficacy than corticosteroids in treating ITP, but it may also cause adverse effects, including nausea, headache, and fever. IVIg can be administered as a single 800 to 1,000 mg/kg dose, or as a daily 400 mg/kg dose form five days; higher doses should be reserved for patients with severe bleeding.¹²

If ITP persists despite the use of corticosteroids or IVIg, IV Rho(D) Ig may be used in patients with Rho(D)-positive blood at a single dose of 25 to 50 μg/kg, with additional doses administered on separate days as required to elevate platelet count. However, only Rho(D)-positive patients are eligible for anti-D treatment.

The response rates/times and adverse effects of common treatments for ITP are summarized in the Table.⁹ A small randomized study found that oral methylprednisolone 30 mg/kg/d for three days followed by 20 mg/kg/d for an additional four days was comparable to IVIg 0.4 g/kg/d for five days.¹¹ A different study that compared oral methylprednisolone (30 mg/kg/d or 50 mg/kg/d for seven days) and IVIg (0.5 g/kg/d for five days) found no difference in outcomes among the three treatments.¹³ One advantage, though, of IVIg is that it can be administered as a single IV dose, rather than multiple doses over several weeks, as is the case with oral prednisone.^9,11-13
• Follow platelet counts closely. Patients with ITP should have their platelet counts monitored at least once weekly and as often as twice weekly. The frequency of monitoring may be tapered depending on an individual patient’s response to treatment and the severity of the thrombocytopenia.¹⁴
• We referred our patient to a nearby children’s hospital, where a repeat CBC  showed her platelets had decreased to 3,000/μL. She received a six-hour infusion of IVIg and was discharged with instructions to have her CBC closely monitored. Her platelets remained stable until four weeks later, when they decreased from 102,000/μL to 71,000/μL. She received a second infusion of IVIg as an outpatient.

Soon after, she presented to our ED with a headache, nausea, and fever of 102°F. CT of her head was normal; a repeat CBC showed no elevation in white blood cells, but her hemoglobin had decreased from 11.9 g/dL to 9.7 g/dL. (Her platelets were 254,000/μL.) The patient’s complaints were likely adverse effects of the IVIg. The CBC abnormalities, fever, headache, and malaise resolved shortly thereafter, and the patient remains asymptomatic with no recurrence of ITP.

Continue for the takeaway >>

THE TAKEAWAY
Suspect ITP in a child who bruises easily and who also has thrombocytopenia. Order a CBC and peripheral blood smear to rule out other potential illnesses. Pharmacotherapy, if needed, typically consists of an oral or IV corticosteroid or IVIg; IV Rho(D) Ig may be used in Rho(D)-positive patients who don’t respond to other treatments. Patients with ITP should have their platelet count monitored at least once weekly until platelets have increased to 150,000/μL or higher. Frequency of monitoring may be reduced as the clinical picture improves and the patient remains stable. More frequent monitoring may be necessary based on severity, complications, and response to treatment.

REFERENCES
1. Johnsen J. Pathogenesis in immune thrombocytopenia: new insights. Hematology Am Soc Hematol Educ Program. 2012;2012:306-312.
2. Kühne T, Buchanan GR, Zimmerman S, et al; Intercontinental Childhood ITP Study Group. A prospective comparative study of 2540 infants and children with newly diagnosed idiopathic thrombocytopenic purpura (ITP) from the Intercontinental Childhood ITP Study Group. J Pediatr. 2003;143:605-608.
3. Kurtzberg J, Stockman JA III. Idiopathic autoimmune thrombocytopenic purpura. Adv Pediatr. 1994;41:111-134.
4. Zeller B, Rajantie J, Hedlund-Treutiger I, et al. Childhood idiopathic thrombocytopenic purpura in the Nordic countries: epidemiology and predictors of chronic disease. Acta Paediatr. 2005;94:178-184.
5. Margolin JF, Steuber CP, Poplack DG. Acute lymphoblastic leukemia. In: Pizzo PA, Poplack DG, eds. Principles and Practice of Pediatric Oncology. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001:317-321.
6. Hashimoto H, Maruyama H, Fujimoto K, et al. Hematologic findings associated with thrombocytopenia during the acute phase of exanthem subitum confirmed by primary human herpesvirus-6 infection. J Pediatr Hematol Oncol. 2002;24:211-214.
7. La Russa VF, Innis BL. Mechanisms of dengue virus-induced bone marrow suppression. Baillieres Clin Haematol. 1995;8:249-270.
8. Aster RH, Curtis BR, McFarland JG, et al. Drug-induced immune thrombocytopenia: pathogenesis, diagnosis, and management. Thromb Haemost. 2009;7:911-918.
9. Provan D, Stasi R, Newland AC, et al. International consensus report on the investigation and management of primary immune thrombocytopenia. Blood. 2010;115:168-186.
10. Bellucci S, Charpak Y, Chastang C, et al. Low doses v conventional doses of corticoids in immune thrombocytopenic purpura (ITP): results of a randomized clinical trial in 160 children, 223 adults. Blood. 1988;71:1165-1169.
11. Ozsoylu S, Sayli TR, Oztürk G. Oral megadose methylprednisolone versus intravenous immunoglobulin for acute childhood idiopathic thrombocytopenic purpura. Pediatr Hematol Oncol. 1993;10:317-321.
12. Beck CE, Nathan PC, Parkin PC, et al. Corticosteroids versus intravenous immune globulin for the treatment of acute immune thrombocytopenic purpura in children: a systematic review and meta-analysis of randomized controlled trials. J Pediatr. 2005;147:521-527.
13. Albayrak D, Islek I, Kalaycí AG, et al. Acute immune thrombocytopenic purpura: a comparative study of very high oral doses of methylprednisolone and intravenously administered immune globulin. J Pediatr. 1994;125(6 pt 1):1004-1007.
14. Tarantino MD, Madden RM, Fennewald DL, et al. Treatment of childhood acute immune thrombocytopenic purpura with anti-D immune globulin or pooled immune globulin. J Pediatr. 1999;134:21-26.

A 6-year-old girl was brought to the emergency department (ED) by her mother after the child bumped her head while playing. While the physician examined the child’s head, the mother remarked that her daughter had recently developed bruises that appeared suddenly and only after minor, if any, known trauma. The ED physician determined that the child’s bump to the head was nothing to worry about, attributed the bruising to the child being a “healthy, active 6-year-old,” and sent her home.

Two days later, the child was brought to our office because the mother was still concerned about her daughter’s easy bruising. The mother pointed out ecchymosis scattered across her daughter’s extremities and torso. The child denied any pain or other complaints, including any active or recurrent bleeding. Upon further questioning, the mother mentioned that her daughter had recovered from a coldlike illness several weeks earlier.

THE DIAGNOSIS
We ordered a complete blood count (CBC) and peripheral smear, which were normal except for the platelet count, which was 7,000/μL (normal, 150,000-450,000/μL). Based on the child’s easy bruising and isolated thrombocytopenia, we diagnosed immune thrombocytopenia, which is also known as idiopathic thrombocytopenic purpura (ITP).

DISCUSSION
In ITP, autoantibodies are directed against platelets, leading to their sequestration and destruction in the spleen and a resultant drop in platelet count.¹ Children with ITP typically present between the ages of 2 and 10, with a peak incidence between 2 and 5.² The incidence is estimated to be as high as 8 per 100,000 children.³ However, this estimate primarily reflects symptomatic children, and the true incidence of childhood ITP may be much higher because asymptomatic children may not be brought in to see a doctor.

For the majority of patients, ITP resolves within three months. However, for 20% to 30% of patients, thrombocytopenia will last beyond six months, with or without treatment.⁴ In 1% of cases, patients will have a recurrence of ITP.³

In addition to easy bruising, nearly all patients who present with possible ITP will complain of cutaneous bleeding, typically a nose bleed or bleeding in the oral cavity.² Upon questioning, 60% of patients will report a history of recent infection.⁴ Not surprisingly, bleeding severity correlates inversely with platelet count; severe bleeding is seen in patients with a platelet count < 10,000/μL.

While rare, the more worrisome complications include intracranial hemorrhage (incidence, 0.1% to 0.8%) and other serious hemorrhages that would require transfusion (estimated incidence, 2.9%).²

Vast differential seen in child bruising
When a child presents with bruising, perform a thorough history, including birth and prenatal course, as well as a physical to exclude other potential causes, such as physical abuse, use of herbal remedies or other natural supplements that may not be disclosed as medication, or even environmental exposure. When bruising is present in a child who has isolated thrombocytopenia, the diagnosis of ITP may be straightforward. However, many conditions share thrombocytopenia in their disease process and should be considered in the differential diagnosis of a child who you suspect may have ITP.
• Suspect physical abuse in a bruised child who does not have thrombocytopenia, whose mood is flat or depressed, or who has experienced recurrent injuries or bruising.
• Leukemia, particularly acute lymphoblastic leukemia (ALL), the predominant leukemia found in children, should be ruled out as well. Symptoms that may distinguish a child with ALL from one with ITP include fever, weight loss, and joint pain, as well as signs such as lymphadenopathy, hepatosplenomegaly, anemia, and leukocytosis. A peripheral smear may be ordered to help confirm or exclude a diagnosis of ALL, should any of the above be present in a child with thrombocytopenia.⁵ It may show lymphoblasts and/or atypical cells in a patient with ALL.⁵
• Infections should also be included in a differential when a patient is suspected of having ITP, particularly if he or she has systemic symptoms. Viral infections that may cause thrombocytopenia include mononucleosis, dengue virus, human herpesvirus-6, and HIV.^6,7

ITP often follows an infection, and the incidence of ITP may be higher during winter months, when infections are more common. However, infection may not always be the cause of ITP. Sepsis may also lead to thrombocytopenia, but a child with sepsis would present very differently from one who has only ITP. A septic child would present with acute illness and signs and symptoms of severe systemic illness, such as high fever, altered mental status, tachycardia, pallor, diaphoresis, and hypotension.
• Drug-induced thrombocytopenia (DIT) should be considered in any child who is taking or recently took a medication that may cause thrombocytopenia. Medications that can cause thrombocytopenia include heparin, quinine, vancomycin, trimethoprim-sulfamethoxazole, rifampin, carbamazepine, phenytoin, piperacillin, linezolid, and valproic acid.8 The measles, mumps, and rubella vaccine also can cause thrombocytopenia.8 A careful medication history may determine if the child is at risk for DIT.
• To narrow the differential, obtain a CBC and peripheral smear when evaluating a patient you suspect may have ITP5 (strength of recommendation [SOR]: A). A CBC will determine the patient’s platelet count, and a peripheral smear should be obtained to exclude other possible diagnoses.⁵

If there are any questions regarding the results of a peripheral smear, it may be necessary to perform a bone marrow aspiration. This, however, is not usually necessary in an otherwise typical case of ITP.9 Bone marrow aspiration may, however, be necessary to reevaluate the initial diagnosis for a child who does not respond to treatment for ITP.

Continue for corticosteroids, IVIg are usually effective >>

Corticosteroids, IVIg are usually effective
The first step in treating a patient with ITP is to limit the risk for further injury or bleeding by stopping NSAIDs or ending participation in contact sports^2,9 (SOR: C). The next step is to determine if pharmacologic therapy is warranted.

Medication, if necessary, is the mainstay of treatment for patients with ITP, particularly those experiencing significant bleeding.² Corticosteroids, intravenous (IV) immunoglobulin (IVIg), and IV Rho(D) immune globulin (also known as anti-D) are the medications typically used to treat a child with ITP, depending on availability of the drugs, bleeding or bleeding risk, and convenience of dosing. For example, corticosteroids can be used orally or IV, whereas IVIg and IV Rho(D) may not be readily available in some treatment settings.

Corticosteroids have been shown to more rapidly increase platelet count compared to placebo and appear to have a dose-related effect.^10,11 Oral prednisone can be dosed at 1 to 2 mg/kg/d for 14 days and then tapered over the course of one week^10,11 or one may prescribe 4 mg/kg/d for four days.^10,11 IV methylprednisolone typically is given at 30 mg/kg/d for three to four days.⁹

IVIg may have greater efficacy than corticosteroids in treating ITP, but it may also cause adverse effects, including nausea, headache, and fever. IVIg can be administered as a single 800 to 1,000 mg/kg dose, or as a daily 400 mg/kg dose form five days; higher doses should be reserved for patients with severe bleeding.¹²

If ITP persists despite the use of corticosteroids or IVIg, IV Rho(D) Ig may be used in patients with Rho(D)-positive blood at a single dose of 25 to 50 μg/kg, with additional doses administered on separate days as required to elevate platelet count. However, only Rho(D)-positive patients are eligible for anti-D treatment.

The response rates/times and adverse effects of common treatments for ITP are summarized in the Table.⁹ A small randomized study found that oral methylprednisolone 30 mg/kg/d for three days followed by 20 mg/kg/d for an additional four days was comparable to IVIg 0.4 g/kg/d for five days.¹¹ A different study that compared oral methylprednisolone (30 mg/kg/d or 50 mg/kg/d for seven days) and IVIg (0.5 g/kg/d for five days) found no difference in outcomes among the three treatments.¹³ One advantage, though, of IVIg is that it can be administered as a single IV dose, rather than multiple doses over several weeks, as is the case with oral prednisone.^9,11-13
• Follow platelet counts closely. Patients with ITP should have their platelet counts monitored at least once weekly and as often as twice weekly. The frequency of monitoring may be tapered depending on an individual patient’s response to treatment and the severity of the thrombocytopenia.¹⁴
• We referred our patient to a nearby children’s hospital, where a repeat CBC  showed her platelets had decreased to 3,000/μL. She received a six-hour infusion of IVIg and was discharged with instructions to have her CBC closely monitored. Her platelets remained stable until four weeks later, when they decreased from 102,000/μL to 71,000/μL. She received a second infusion of IVIg as an outpatient.

Soon after, she presented to our ED with a headache, nausea, and fever of 102°F. CT of her head was normal; a repeat CBC showed no elevation in white blood cells, but her hemoglobin had decreased from 11.9 g/dL to 9.7 g/dL. (Her platelets were 254,000/μL.) The patient’s complaints were likely adverse effects of the IVIg. The CBC abnormalities, fever, headache, and malaise resolved shortly thereafter, and the patient remains asymptomatic with no recurrence of ITP.

Continue for the takeaway >>

THE TAKEAWAY
Suspect ITP in a child who bruises easily and who also has thrombocytopenia. Order a CBC and peripheral blood smear to rule out other potential illnesses. Pharmacotherapy, if needed, typically consists of an oral or IV corticosteroid or IVIg; IV Rho(D) Ig may be used in Rho(D)-positive patients who don’t respond to other treatments. Patients with ITP should have their platelet count monitored at least once weekly until platelets have increased to 150,000/μL or higher. Frequency of monitoring may be reduced as the clinical picture improves and the patient remains stable. More frequent monitoring may be necessary based on severity, complications, and response to treatment.

REFERENCES
1. Johnsen J. Pathogenesis in immune thrombocytopenia: new insights. Hematology Am Soc Hematol Educ Program. 2012;2012:306-312.
2. Kühne T, Buchanan GR, Zimmerman S, et al; Intercontinental Childhood ITP Study Group. A prospective comparative study of 2540 infants and children with newly diagnosed idiopathic thrombocytopenic purpura (ITP) from the Intercontinental Childhood ITP Study Group. J Pediatr. 2003;143:605-608.
3. Kurtzberg J, Stockman JA III. Idiopathic autoimmune thrombocytopenic purpura. Adv Pediatr. 1994;41:111-134.
4. Zeller B, Rajantie J, Hedlund-Treutiger I, et al. Childhood idiopathic thrombocytopenic purpura in the Nordic countries: epidemiology and predictors of chronic disease. Acta Paediatr. 2005;94:178-184.
5. Margolin JF, Steuber CP, Poplack DG. Acute lymphoblastic leukemia. In: Pizzo PA, Poplack DG, eds. Principles and Practice of Pediatric Oncology. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001:317-321.
6. Hashimoto H, Maruyama H, Fujimoto K, et al. Hematologic findings associated with thrombocytopenia during the acute phase of exanthem subitum confirmed by primary human herpesvirus-6 infection. J Pediatr Hematol Oncol. 2002;24:211-214.
7. La Russa VF, Innis BL. Mechanisms of dengue virus-induced bone marrow suppression. Baillieres Clin Haematol. 1995;8:249-270.
8. Aster RH, Curtis BR, McFarland JG, et al. Drug-induced immune thrombocytopenia: pathogenesis, diagnosis, and management. Thromb Haemost. 2009;7:911-918.
9. Provan D, Stasi R, Newland AC, et al. International consensus report on the investigation and management of primary immune thrombocytopenia. Blood. 2010;115:168-186.
10. Bellucci S, Charpak Y, Chastang C, et al. Low doses v conventional doses of corticoids in immune thrombocytopenic purpura (ITP): results of a randomized clinical trial in 160 children, 223 adults. Blood. 1988;71:1165-1169.
11. Ozsoylu S, Sayli TR, Oztürk G. Oral megadose methylprednisolone versus intravenous immunoglobulin for acute childhood idiopathic thrombocytopenic purpura. Pediatr Hematol Oncol. 1993;10:317-321.
12. Beck CE, Nathan PC, Parkin PC, et al. Corticosteroids versus intravenous immune globulin for the treatment of acute immune thrombocytopenic purpura in children: a systematic review and meta-analysis of randomized controlled trials. J Pediatr. 2005;147:521-527.
13. Albayrak D, Islek I, Kalaycí AG, et al. Acute immune thrombocytopenic purpura: a comparative study of very high oral doses of methylprednisolone and intravenously administered immune globulin. J Pediatr. 1994;125(6 pt 1):1004-1007.
14. Tarantino MD, Madden RM, Fennewald DL, et al. Treatment of childhood acute immune thrombocytopenic purpura with anti-D immune globulin or pooled immune globulin. J Pediatr. 1999;134:21-26.

Overview

Palliative medicine in the ED represents a paradigm shift for the emergency physician (EP)—from identifying and stabilizing acute medical and surgical conditions to providing symptomatic comfort care to a dying patient. When the ED became the “safety net” for patients who have serious, life-limiting illnesses,^1-3 it also became the most frequent place where such care is initially sought⁴—although not considered an ideal place to begin such care.

In one study, approximately 40% of dying patients presented to the ED during their final 2 weeks of life.⁵ With the ED becoming more recognized as a location for palliative care, the EP plays a key role in the care of these patients. The 2013 Model of the Clinical Practice of Emergency Medi­cine explicitly lists palliative medicine within the EP’s scope of practice.⁶ Further support for providing palliative care in emergency medicine includes the cosponsorship of Hospice and Palliative Medicine subspecialty board certification by the American Board of Emergency Medicine in 2008. Finally, palliative care medicine principles have been endorsed in the “Choosing Wisely” initiative of the American College of Emergency Physicians.

Essential Palliative Care Skills

Quest et al⁷ have identified the following 12 primary palliative care skills in which every EP should be competent:

1. Assessment of illness trajectory;
2. Determination of prognosis;
3. Communication of bad news;
4. Interpretation and formation of an advance care plan;
5. Allowance of family presence during resuscitation;
6. Symptom management (both pain and nonpain);
7. Withholding and withdrawal of life-sustaining treatments;
8. Management of imminently dying patients;
9. Identification and implementation of hospice and palliative care plans;
10. Understanding of ethical and legal issues pertinent to end-of-life care;
11. Display of spiritual and cultural competency; and
12. Management of the dying child.

Although all of the above are important skills, this paper focuses on the symptom management of pain and nonpain (skill 6) in patients presenting to the ED with a life-limiting illness. The evidence base for these treatments is limited due to the many methodological challenges faced when studying symptoms in patients who are at end of life.

Pharmacologic Management of Symptoms

Recent research has found that symptom burden is high at end of life. Despite the increase in attention to these patients and their needs, symptoms including pain, depression, and delirium have repeatedly increased between 1998 and 2010.⁸ A 2013 study recommended that a minimum of four classes of medications be considered for patients who are at end of life: opioid (for pain); benzodiazepine (for anxiety); antipsychotic (for delirium and nausea); and antimuscarinic (for excessive secretions).⁹ The role and indications for each of these drug classes will be discussed.

Palliative Care Intervention

Though EPs frequently request specialty and subspecialty consultation for ED patients, they usually do not consider a palliative care medicine consult for the dying patient. Palliative care medicine utilizes an interdisciplinary, collaborative, team-based approach to decrease the pain and suffering of patients with advanced illness.¹⁰

Benefits from early palliative care intervention in the ED include improved symptom management, improved patient and family satisfaction, improved outcomes, decreased length of stay, less use of intensive care units, and less costs.⁴

Pain Management

Pain is one of the most devastating symptoms that a patient can experience, and its management is an integral component of palliative care medicine. Initial evaluation must include appropriate assessment of the pain and its impact on a patient’s function and quality of life.

The general approach to pain management follows the World Health Organization pain ladder. For mild to moderate pain, step 1 begins with acetaminophen or a nonsteroidal anti-inflammatory drug (NSAID), with or without an adjuvant such as an antidepressant or anticonvulsant. If pain persists, step 2 involves the addition of an opioid. For moderate to severe pain, step 3 involves the addition of stronger opioids, such as hydromorphone, morphine, and oxycodone. Typically, a patient with a serious, life-limiting illness who presents to the ED for help will likely require treatment with strong opioids (step 3).

Opioids

In patients requiring step 3 management, opioids are the primary medication used to manage pain. An initial equivalent dose of morphine 5 mg intravenously (IV) is appropriate in an opioid-naïve patient. The adage of “starting low and going slow” is important to follow; however, an important corollary is “…and use enough.” If a patient’s pain is not controlled with initial dosages, additional bolus doses of 50% to 100% increments will be necessary. Because opioids do not have a ceiling effect, it is important to understand that dosages may seem very high for some patients compared to others. In this population, ensuring baseline pain control, with either an oral long-acting formulation or a continuous IV infusion, is important.¹¹

Difficulties clinicians have in determining opioids for the management of pain are multifactorial. One consideration may be the growing public concern for prescription opioid abuse, potentially creating resistance to appropriate use of opioids by clinicians who fear legal or regulatory push back.

General principles in managing severe pain in the opioid-tolerant patient include the following: (1) calculating the morphine equivalent as a daily 24-hour dose; (2) determining the breakthrough dose, which is usually 10% to 15% of the calculated daily dose; (3) titrating doses upward if pain is not controlled, or if more than three breakthrough doses are being required daily; and (4) reducing the calculated conversion dose of a new opioid 25% to 50% when converting to a different opioid.¹²

The EP is frequently required to convert a patient’s oral opioid analgesic to an equivalent IV dose, and hydromorphone and morphine are the two most commonly used. The Table provides an approximation for this conversion.

Equianalgesic Dosing

Equianalgesic dosing is an important concept in pain management, especially for those patients already receiving opioids. There is great variation in the analgesic potency of the different opioids. The dose at which two opioids provide equivalent pain relief is the equianalgesic dose. Usually, this is standardized to 10 mg of parenteral morphine.¹³ Unfortunately, it is not uncommon for patients to be undertreated when switched to another opioid.

Nonpain Symptom Management

Nonpain symptoms that all EPs must know how to manage include constipation, dyspnea, nausea/vomiting, the so-called death rattle, and terminal delirium. In one study of reasons for ED visits by palliative care patients, the most common were dyspnea (26%), nausea/vomiting/constipation (17%), and uncontrolled pain (15%).¹⁴

Constipation

The most important adverse effect of opioids—one that does not improve or change during treatment—is constipation. Constipation in general—not just associated with opioids—has been ranked as one of the 10 most bothersome symptoms in the palliative care population, leading to discomfort, decreased quality of life, and potential small bowel obstruction or perforation.¹⁵ Unless contraindicated, a gastrointestinal stimulant such as senna, or an osmotic laxative such as lactulose, must be prescribed whenever an opioid is initiated. As the author (Galicia-Castillo) often notes, “The hand that writes the prescription for an opioid should be the hand that writes an Rx for a bowel regimen, or it becomes the hand that disimpacts the patient.”

The most recent Cochrane Review for the management of constipation in the palliative care population did not show any differences in the effectiveness among three commonly used laxatives: senna, docusate, and lactulose. This review did not evaluate polyethylene glycol, which is also commonly used.¹⁶ The addition of stool softeners, bisacodyl and nightly prune juice can also be helpful.¹⁰

Dyspnea

Dyspnea, the subjective feeling of breathing discomfort, is a common end-of-life complaint. Similar to pain, self-report is required for adequate assessment of dyspnea. Treatment recommendations include opioids, anxiolytics, and oxygen therapy.¹⁸ Opioids are the most widely studied treatment for dyspnea, demonstrating reduction in breathlessness in patients who have a variety of conditions, such as advanced chronic obstructive pulmonary disease, interstitial lung disease, cancer, and chronic heart failure.¹⁹

While many of the benefits of opioids are widely recognized and understood, the manner in which they improve symptoms of dyspnea is less well known. In addition, the evidence of effectiveness is limited to oral or parenteral morphine and fentanyl, and nebulized opioids have not been well studied. Oxygen treatments have been shown to reduce dyspnea in patients who suffer from hypoxemia; however, no benefit was found for patients who had only mild or no hypoxemia. A majority of dying patients did not experience a change in respiratory comfort after their supplemental oxygen was withdrawn. In these cases, when administration of oxygen is unnecessary, it may potentially introduce further discomfort to end-of-life patients by causing nasal dryness and impaired mobility.²⁰

The use of benzodiazepines as the primary medication to manage dyspnea is unfounded, but may provide some benefit when used in conjunction with opioids.¹¹ When indicated, a longer-acting agent (eg, clonazepam, with an initial starting does of 0.25 mg orally every 12 hours) may be used.⁴

Nausea and Vomiting

Nausea and vomiting have been reported by 16% to 68% of patients who had life-limiting illness, such as cancer, heart failure, renal failure, or acquired immunodeficiency syndrome.²¹ The etiology of nausea and vomiting is multifactorial in a dying patient. Assessment and treatment has been based on understanding how neurotransmitters are involved in the “emetic pathway,”²² but other pathways, such as a cytokine-mediated model of cancer symptoms, may also be important.²³

Nonpharmacologic methods to utilize include avoidance of environmental stimuli, such as fatty, spicy, and salted foods; use of relaxation and distraction; and massage.²² Several medication classes have been utilized to treat nausea and vomiting: prokinetic agents (metoclopramide 10 mg three to four times a day, 30 minutes prior to meals and bedtime); dopamine receptor antagonists (haloperidol 1.5-5 mg two to three times a day); antihistaminic agents (promethazine 25 mg orally or IV every 4-6 hours, with a maximum dose of 100 mg/d); and selective 5 hydroxytryptamine-3 receptor antagonists (ondansetron 4-8 mg once or twice a day). Other agents that have been utilized include corticosteroids, benzodiazepines, octreotide, and cannabinoids.²²

Procedures such as percutaneous endoscopic gastrostomy placement, nasogastric tube placement, and stenting may be necessary for patients who have advanced disease caused by a mechanical obstruction.²²

Death Rattle

The death rattle occurs when secretions accumulate in the pharynx and/or airways when swallowing and cough mechanisms are no longer intact.²⁴ This phenomenon occurs in 23% to 92% of dying patients.²⁵ Generally, death occurs within 48 hours for about 75% of such patients.²⁶ The noise that results from this process is usually more disturbing for those visiting the patients than to the patient themselves. Conservative measures to employ include placing patients on their sides to facilitate secretion drainage and to minimize upper airway sounds, gentle oral and pharyngeal suctioning, and limiting fluid input.¹¹

One recent study reviewing the use of the anticholinergics atropine, scopolamine, and hyoscine demonstrated similar efficacy among the three drugs. Dosages used in this study included atropine 0.5 mg as a subcutaneous bolus, followed by 3 mg every 24 hours subcutaneously; scopolamine as a 0.25 mg subcutaneous bolus, followed by 1.5 mg every 24 hours IV or by subcutaneous infusion; and hyoscine 20 mg as a subcutaneous bolus followed by 60 mg every 24 hours IV or subcutaneous infusion. Glycopyrrolate is often used in the cognitively intact patient, as it does not cross the blood-brain barrier; however, supply concerns at the time of the study prevented a review of its efficiency.²⁷ All of these medications are also available in oral and transdermal formulations.

Terminal Delirium

Delirium is a common complication for patients nearing the end of life, affecting as many as 88% of dying patients.²⁸ It is characterized by an acute onset of cognitive impairment that may manifest as either a hyperactive or hypoactive state. Causes for terminal delirium are multifactorial. Initially, management should include prevention strategies, such as frequently orientating the patient, maintenance of day-night cycles, provision of adequate sleep, and minimization of sensory overload.¹¹ When pharmacological therapy is required to improve quality of life, a neuroleptic medication, namely haloperidol, should be used initially. The addition of a benzodiazepine may help if the initial treatments are ineffective, or if sedation is desired.²⁸

Summary

Emergency physicians have a unique opportunity to improve the quality of life for patients suffering serious illness, especially those who are actively dying. The management of pain and nonpain symptoms in patients who are at end of life, is a particularly important skill for every EP. If available, a consultation with a palliative care medicine consultant may improve both short- and long-term patient care.

Dr Galicia-Castillo is the Sue Faulkner Scribner professor of geriatrics at the Eastern Virginia Medical School Glennan Center for Geriatrics and Gerontology, and Medical Director for Palliative Care Medicine at Sentara Norfolk General Hospital. Dr Counselman is the distinguished professor and chairman of the department of emergency medicine at Eastern Virginia Medical School, Norfolk; and a physician at Emergency Physicians of Tidewater, Norfolk, Virginia. He is also the associate editor in chief of EMERGENCY MEDICINE editorial board.

Overview

Palliative medicine in the ED represents a paradigm shift for the emergency physician (EP)—from identifying and stabilizing acute medical and surgical conditions to providing symptomatic comfort care to a dying patient. When the ED became the “safety net” for patients who have serious, life-limiting illnesses,^1-3 it also became the most frequent place where such care is initially sought⁴—although not considered an ideal place to begin such care.

In one study, approximately 40% of dying patients presented to the ED during their final 2 weeks of life.⁵ With the ED becoming more recognized as a location for palliative care, the EP plays a key role in the care of these patients. The 2013 Model of the Clinical Practice of Emergency Medi­cine explicitly lists palliative medicine within the EP’s scope of practice.⁶ Further support for providing palliative care in emergency medicine includes the cosponsorship of Hospice and Palliative Medicine subspecialty board certification by the American Board of Emergency Medicine in 2008. Finally, palliative care medicine principles have been endorsed in the “Choosing Wisely” initiative of the American College of Emergency Physicians.

Essential Palliative Care Skills

Quest et al⁷ have identified the following 12 primary palliative care skills in which every EP should be competent:

1. Assessment of illness trajectory;
2. Determination of prognosis;
3. Communication of bad news;
4. Interpretation and formation of an advance care plan;
5. Allowance of family presence during resuscitation;
6. Symptom management (both pain and nonpain);
7. Withholding and withdrawal of life-sustaining treatments;
8. Management of imminently dying patients;
9. Identification and implementation of hospice and palliative care plans;
10. Understanding of ethical and legal issues pertinent to end-of-life care;
11. Display of spiritual and cultural competency; and
12. Management of the dying child.

Although all of the above are important skills, this paper focuses on the symptom management of pain and nonpain (skill 6) in patients presenting to the ED with a life-limiting illness. The evidence base for these treatments is limited due to the many methodological challenges faced when studying symptoms in patients who are at end of life.

Pharmacologic Management of Symptoms

Recent research has found that symptom burden is high at end of life. Despite the increase in attention to these patients and their needs, symptoms including pain, depression, and delirium have repeatedly increased between 1998 and 2010.⁸ A 2013 study recommended that a minimum of four classes of medications be considered for patients who are at end of life: opioid (for pain); benzodiazepine (for anxiety); antipsychotic (for delirium and nausea); and antimuscarinic (for excessive secretions).⁹ The role and indications for each of these drug classes will be discussed.

Palliative Care Intervention

Though EPs frequently request specialty and subspecialty consultation for ED patients, they usually do not consider a palliative care medicine consult for the dying patient. Palliative care medicine utilizes an interdisciplinary, collaborative, team-based approach to decrease the pain and suffering of patients with advanced illness.¹⁰

Benefits from early palliative care intervention in the ED include improved symptom management, improved patient and family satisfaction, improved outcomes, decreased length of stay, less use of intensive care units, and less costs.⁴

Pain Management

Pain is one of the most devastating symptoms that a patient can experience, and its management is an integral component of palliative care medicine. Initial evaluation must include appropriate assessment of the pain and its impact on a patient’s function and quality of life.

The general approach to pain management follows the World Health Organization pain ladder. For mild to moderate pain, step 1 begins with acetaminophen or a nonsteroidal anti-inflammatory drug (NSAID), with or without an adjuvant such as an antidepressant or anticonvulsant. If pain persists, step 2 involves the addition of an opioid. For moderate to severe pain, step 3 involves the addition of stronger opioids, such as hydromorphone, morphine, and oxycodone. Typically, a patient with a serious, life-limiting illness who presents to the ED for help will likely require treatment with strong opioids (step 3).

Opioids

In patients requiring step 3 management, opioids are the primary medication used to manage pain. An initial equivalent dose of morphine 5 mg intravenously (IV) is appropriate in an opioid-naïve patient. The adage of “starting low and going slow” is important to follow; however, an important corollary is “…and use enough.” If a patient’s pain is not controlled with initial dosages, additional bolus doses of 50% to 100% increments will be necessary. Because opioids do not have a ceiling effect, it is important to understand that dosages may seem very high for some patients compared to others. In this population, ensuring baseline pain control, with either an oral long-acting formulation or a continuous IV infusion, is important.¹¹

Difficulties clinicians have in determining opioids for the management of pain are multifactorial. One consideration may be the growing public concern for prescription opioid abuse, potentially creating resistance to appropriate use of opioids by clinicians who fear legal or regulatory push back.

General principles in managing severe pain in the opioid-tolerant patient include the following: (1) calculating the morphine equivalent as a daily 24-hour dose; (2) determining the breakthrough dose, which is usually 10% to 15% of the calculated daily dose; (3) titrating doses upward if pain is not controlled, or if more than three breakthrough doses are being required daily; and (4) reducing the calculated conversion dose of a new opioid 25% to 50% when converting to a different opioid.¹²

The EP is frequently required to convert a patient’s oral opioid analgesic to an equivalent IV dose, and hydromorphone and morphine are the two most commonly used. The Table provides an approximation for this conversion.

Equianalgesic Dosing

Equianalgesic dosing is an important concept in pain management, especially for those patients already receiving opioids. There is great variation in the analgesic potency of the different opioids. The dose at which two opioids provide equivalent pain relief is the equianalgesic dose. Usually, this is standardized to 10 mg of parenteral morphine.¹³ Unfortunately, it is not uncommon for patients to be undertreated when switched to another opioid.

Nonpain Symptom Management

Nonpain symptoms that all EPs must know how to manage include constipation, dyspnea, nausea/vomiting, the so-called death rattle, and terminal delirium. In one study of reasons for ED visits by palliative care patients, the most common were dyspnea (26%), nausea/vomiting/constipation (17%), and uncontrolled pain (15%).¹⁴

Constipation

The most important adverse effect of opioids—one that does not improve or change during treatment—is constipation. Constipation in general—not just associated with opioids—has been ranked as one of the 10 most bothersome symptoms in the palliative care population, leading to discomfort, decreased quality of life, and potential small bowel obstruction or perforation.¹⁵ Unless contraindicated, a gastrointestinal stimulant such as senna, or an osmotic laxative such as lactulose, must be prescribed whenever an opioid is initiated. As the author (Galicia-Castillo) often notes, “The hand that writes the prescription for an opioid should be the hand that writes an Rx for a bowel regimen, or it becomes the hand that disimpacts the patient.”

The most recent Cochrane Review for the management of constipation in the palliative care population did not show any differences in the effectiveness among three commonly used laxatives: senna, docusate, and lactulose. This review did not evaluate polyethylene glycol, which is also commonly used.¹⁶ The addition of stool softeners, bisacodyl and nightly prune juice can also be helpful.¹⁰

Dyspnea

Dyspnea, the subjective feeling of breathing discomfort, is a common end-of-life complaint. Similar to pain, self-report is required for adequate assessment of dyspnea. Treatment recommendations include opioids, anxiolytics, and oxygen therapy.¹⁸ Opioids are the most widely studied treatment for dyspnea, demonstrating reduction in breathlessness in patients who have a variety of conditions, such as advanced chronic obstructive pulmonary disease, interstitial lung disease, cancer, and chronic heart failure.¹⁹

While many of the benefits of opioids are widely recognized and understood, the manner in which they improve symptoms of dyspnea is less well known. In addition, the evidence of effectiveness is limited to oral or parenteral morphine and fentanyl, and nebulized opioids have not been well studied. Oxygen treatments have been shown to reduce dyspnea in patients who suffer from hypoxemia; however, no benefit was found for patients who had only mild or no hypoxemia. A majority of dying patients did not experience a change in respiratory comfort after their supplemental oxygen was withdrawn. In these cases, when administration of oxygen is unnecessary, it may potentially introduce further discomfort to end-of-life patients by causing nasal dryness and impaired mobility.²⁰

The use of benzodiazepines as the primary medication to manage dyspnea is unfounded, but may provide some benefit when used in conjunction with opioids.¹¹ When indicated, a longer-acting agent (eg, clonazepam, with an initial starting does of 0.25 mg orally every 12 hours) may be used.⁴

Nausea and Vomiting

Nausea and vomiting have been reported by 16% to 68% of patients who had life-limiting illness, such as cancer, heart failure, renal failure, or acquired immunodeficiency syndrome.²¹ The etiology of nausea and vomiting is multifactorial in a dying patient. Assessment and treatment has been based on understanding how neurotransmitters are involved in the “emetic pathway,”²² but other pathways, such as a cytokine-mediated model of cancer symptoms, may also be important.²³

Nonpharmacologic methods to utilize include avoidance of environmental stimuli, such as fatty, spicy, and salted foods; use of relaxation and distraction; and massage.²² Several medication classes have been utilized to treat nausea and vomiting: prokinetic agents (metoclopramide 10 mg three to four times a day, 30 minutes prior to meals and bedtime); dopamine receptor antagonists (haloperidol 1.5-5 mg two to three times a day); antihistaminic agents (promethazine 25 mg orally or IV every 4-6 hours, with a maximum dose of 100 mg/d); and selective 5 hydroxytryptamine-3 receptor antagonists (ondansetron 4-8 mg once or twice a day). Other agents that have been utilized include corticosteroids, benzodiazepines, octreotide, and cannabinoids.²²

Procedures such as percutaneous endoscopic gastrostomy placement, nasogastric tube placement, and stenting may be necessary for patients who have advanced disease caused by a mechanical obstruction.²²

Death Rattle

The death rattle occurs when secretions accumulate in the pharynx and/or airways when swallowing and cough mechanisms are no longer intact.²⁴ This phenomenon occurs in 23% to 92% of dying patients.²⁵ Generally, death occurs within 48 hours for about 75% of such patients.²⁶ The noise that results from this process is usually more disturbing for those visiting the patients than to the patient themselves. Conservative measures to employ include placing patients on their sides to facilitate secretion drainage and to minimize upper airway sounds, gentle oral and pharyngeal suctioning, and limiting fluid input.¹¹

One recent study reviewing the use of the anticholinergics atropine, scopolamine, and hyoscine demonstrated similar efficacy among the three drugs. Dosages used in this study included atropine 0.5 mg as a subcutaneous bolus, followed by 3 mg every 24 hours subcutaneously; scopolamine as a 0.25 mg subcutaneous bolus, followed by 1.5 mg every 24 hours IV or by subcutaneous infusion; and hyoscine 20 mg as a subcutaneous bolus followed by 60 mg every 24 hours IV or subcutaneous infusion. Glycopyrrolate is often used in the cognitively intact patient, as it does not cross the blood-brain barrier; however, supply concerns at the time of the study prevented a review of its efficiency.²⁷ All of these medications are also available in oral and transdermal formulations.

Terminal Delirium

Delirium is a common complication for patients nearing the end of life, affecting as many as 88% of dying patients.²⁸ It is characterized by an acute onset of cognitive impairment that may manifest as either a hyperactive or hypoactive state. Causes for terminal delirium are multifactorial. Initially, management should include prevention strategies, such as frequently orientating the patient, maintenance of day-night cycles, provision of adequate sleep, and minimization of sensory overload.¹¹ When pharmacological therapy is required to improve quality of life, a neuroleptic medication, namely haloperidol, should be used initially. The addition of a benzodiazepine may help if the initial treatments are ineffective, or if sedation is desired.²⁸

Summary

Emergency physicians have a unique opportunity to improve the quality of life for patients suffering serious illness, especially those who are actively dying. The management of pain and nonpain symptoms in patients who are at end of life, is a particularly important skill for every EP. If available, a consultation with a palliative care medicine consultant may improve both short- and long-term patient care.

Dr Galicia-Castillo is the Sue Faulkner Scribner professor of geriatrics at the Eastern Virginia Medical School Glennan Center for Geriatrics and Gerontology, and Medical Director for Palliative Care Medicine at Sentara Norfolk General Hospital. Dr Counselman is the distinguished professor and chairman of the department of emergency medicine at Eastern Virginia Medical School, Norfolk; and a physician at Emergency Physicians of Tidewater, Norfolk, Virginia. He is also the associate editor in chief of EMERGENCY MEDICINE editorial board.

Overview

Palliative medicine in the ED represents a paradigm shift for the emergency physician (EP)—from identifying and stabilizing acute medical and surgical conditions to providing symptomatic comfort care to a dying patient. When the ED became the “safety net” for patients who have serious, life-limiting illnesses,^1-3 it also became the most frequent place where such care is initially sought⁴—although not considered an ideal place to begin such care.

In one study, approximately 40% of dying patients presented to the ED during their final 2 weeks of life.⁵ With the ED becoming more recognized as a location for palliative care, the EP plays a key role in the care of these patients. The 2013 Model of the Clinical Practice of Emergency Medi­cine explicitly lists palliative medicine within the EP’s scope of practice.⁶ Further support for providing palliative care in emergency medicine includes the cosponsorship of Hospice and Palliative Medicine subspecialty board certification by the American Board of Emergency Medicine in 2008. Finally, palliative care medicine principles have been endorsed in the “Choosing Wisely” initiative of the American College of Emergency Physicians.

Essential Palliative Care Skills

Quest et al⁷ have identified the following 12 primary palliative care skills in which every EP should be competent:

1. Assessment of illness trajectory;
2. Determination of prognosis;
3. Communication of bad news;
4. Interpretation and formation of an advance care plan;
5. Allowance of family presence during resuscitation;
6. Symptom management (both pain and nonpain);
7. Withholding and withdrawal of life-sustaining treatments;
8. Management of imminently dying patients;
9. Identification and implementation of hospice and palliative care plans;
10. Understanding of ethical and legal issues pertinent to end-of-life care;
11. Display of spiritual and cultural competency; and
12. Management of the dying child.

Although all of the above are important skills, this paper focuses on the symptom management of pain and nonpain (skill 6) in patients presenting to the ED with a life-limiting illness. The evidence base for these treatments is limited due to the many methodological challenges faced when studying symptoms in patients who are at end of life.

Pharmacologic Management of Symptoms

Recent research has found that symptom burden is high at end of life. Despite the increase in attention to these patients and their needs, symptoms including pain, depression, and delirium have repeatedly increased between 1998 and 2010.⁸ A 2013 study recommended that a minimum of four classes of medications be considered for patients who are at end of life: opioid (for pain); benzodiazepine (for anxiety); antipsychotic (for delirium and nausea); and antimuscarinic (for excessive secretions).⁹ The role and indications for each of these drug classes will be discussed.

Palliative Care Intervention

Though EPs frequently request specialty and subspecialty consultation for ED patients, they usually do not consider a palliative care medicine consult for the dying patient. Palliative care medicine utilizes an interdisciplinary, collaborative, team-based approach to decrease the pain and suffering of patients with advanced illness.¹⁰

Benefits from early palliative care intervention in the ED include improved symptom management, improved patient and family satisfaction, improved outcomes, decreased length of stay, less use of intensive care units, and less costs.⁴

Pain Management

Pain is one of the most devastating symptoms that a patient can experience, and its management is an integral component of palliative care medicine. Initial evaluation must include appropriate assessment of the pain and its impact on a patient’s function and quality of life.

The general approach to pain management follows the World Health Organization pain ladder. For mild to moderate pain, step 1 begins with acetaminophen or a nonsteroidal anti-inflammatory drug (NSAID), with or without an adjuvant such as an antidepressant or anticonvulsant. If pain persists, step 2 involves the addition of an opioid. For moderate to severe pain, step 3 involves the addition of stronger opioids, such as hydromorphone, morphine, and oxycodone. Typically, a patient with a serious, life-limiting illness who presents to the ED for help will likely require treatment with strong opioids (step 3).

Opioids

In patients requiring step 3 management, opioids are the primary medication used to manage pain. An initial equivalent dose of morphine 5 mg intravenously (IV) is appropriate in an opioid-naïve patient. The adage of “starting low and going slow” is important to follow; however, an important corollary is “…and use enough.” If a patient’s pain is not controlled with initial dosages, additional bolus doses of 50% to 100% increments will be necessary. Because opioids do not have a ceiling effect, it is important to understand that dosages may seem very high for some patients compared to others. In this population, ensuring baseline pain control, with either an oral long-acting formulation or a continuous IV infusion, is important.¹¹

Difficulties clinicians have in determining opioids for the management of pain are multifactorial. One consideration may be the growing public concern for prescription opioid abuse, potentially creating resistance to appropriate use of opioids by clinicians who fear legal or regulatory push back.

General principles in managing severe pain in the opioid-tolerant patient include the following: (1) calculating the morphine equivalent as a daily 24-hour dose; (2) determining the breakthrough dose, which is usually 10% to 15% of the calculated daily dose; (3) titrating doses upward if pain is not controlled, or if more than three breakthrough doses are being required daily; and (4) reducing the calculated conversion dose of a new opioid 25% to 50% when converting to a different opioid.¹²

The EP is frequently required to convert a patient’s oral opioid analgesic to an equivalent IV dose, and hydromorphone and morphine are the two most commonly used. The Table provides an approximation for this conversion.

Equianalgesic Dosing

Equianalgesic dosing is an important concept in pain management, especially for those patients already receiving opioids. There is great variation in the analgesic potency of the different opioids. The dose at which two opioids provide equivalent pain relief is the equianalgesic dose. Usually, this is standardized to 10 mg of parenteral morphine.¹³ Unfortunately, it is not uncommon for patients to be undertreated when switched to another opioid.

Nonpain Symptom Management

Nonpain symptoms that all EPs must know how to manage include constipation, dyspnea, nausea/vomiting, the so-called death rattle, and terminal delirium. In one study of reasons for ED visits by palliative care patients, the most common were dyspnea (26%), nausea/vomiting/constipation (17%), and uncontrolled pain (15%).¹⁴

Constipation

The most important adverse effect of opioids—one that does not improve or change during treatment—is constipation. Constipation in general—not just associated with opioids—has been ranked as one of the 10 most bothersome symptoms in the palliative care population, leading to discomfort, decreased quality of life, and potential small bowel obstruction or perforation.¹⁵ Unless contraindicated, a gastrointestinal stimulant such as senna, or an osmotic laxative such as lactulose, must be prescribed whenever an opioid is initiated. As the author (Galicia-Castillo) often notes, “The hand that writes the prescription for an opioid should be the hand that writes an Rx for a bowel regimen, or it becomes the hand that disimpacts the patient.”

The most recent Cochrane Review for the management of constipation in the palliative care population did not show any differences in the effectiveness among three commonly used laxatives: senna, docusate, and lactulose. This review did not evaluate polyethylene glycol, which is also commonly used.¹⁶ The addition of stool softeners, bisacodyl and nightly prune juice can also be helpful.¹⁰

Dyspnea

Dyspnea, the subjective feeling of breathing discomfort, is a common end-of-life complaint. Similar to pain, self-report is required for adequate assessment of dyspnea. Treatment recommendations include opioids, anxiolytics, and oxygen therapy.¹⁸ Opioids are the most widely studied treatment for dyspnea, demonstrating reduction in breathlessness in patients who have a variety of conditions, such as advanced chronic obstructive pulmonary disease, interstitial lung disease, cancer, and chronic heart failure.¹⁹

While many of the benefits of opioids are widely recognized and understood, the manner in which they improve symptoms of dyspnea is less well known. In addition, the evidence of effectiveness is limited to oral or parenteral morphine and fentanyl, and nebulized opioids have not been well studied. Oxygen treatments have been shown to reduce dyspnea in patients who suffer from hypoxemia; however, no benefit was found for patients who had only mild or no hypoxemia. A majority of dying patients did not experience a change in respiratory comfort after their supplemental oxygen was withdrawn. In these cases, when administration of oxygen is unnecessary, it may potentially introduce further discomfort to end-of-life patients by causing nasal dryness and impaired mobility.²⁰

The use of benzodiazepines as the primary medication to manage dyspnea is unfounded, but may provide some benefit when used in conjunction with opioids.¹¹ When indicated, a longer-acting agent (eg, clonazepam, with an initial starting does of 0.25 mg orally every 12 hours) may be used.⁴

Nausea and Vomiting

Nausea and vomiting have been reported by 16% to 68% of patients who had life-limiting illness, such as cancer, heart failure, renal failure, or acquired immunodeficiency syndrome.²¹ The etiology of nausea and vomiting is multifactorial in a dying patient. Assessment and treatment has been based on understanding how neurotransmitters are involved in the “emetic pathway,”²² but other pathways, such as a cytokine-mediated model of cancer symptoms, may also be important.²³

Nonpharmacologic methods to utilize include avoidance of environmental stimuli, such as fatty, spicy, and salted foods; use of relaxation and distraction; and massage.²² Several medication classes have been utilized to treat nausea and vomiting: prokinetic agents (metoclopramide 10 mg three to four times a day, 30 minutes prior to meals and bedtime); dopamine receptor antagonists (haloperidol 1.5-5 mg two to three times a day); antihistaminic agents (promethazine 25 mg orally or IV every 4-6 hours, with a maximum dose of 100 mg/d); and selective 5 hydroxytryptamine-3 receptor antagonists (ondansetron 4-8 mg once or twice a day). Other agents that have been utilized include corticosteroids, benzodiazepines, octreotide, and cannabinoids.²²

Procedures such as percutaneous endoscopic gastrostomy placement, nasogastric tube placement, and stenting may be necessary for patients who have advanced disease caused by a mechanical obstruction.²²

Death Rattle

The death rattle occurs when secretions accumulate in the pharynx and/or airways when swallowing and cough mechanisms are no longer intact.²⁴ This phenomenon occurs in 23% to 92% of dying patients.²⁵ Generally, death occurs within 48 hours for about 75% of such patients.²⁶ The noise that results from this process is usually more disturbing for those visiting the patients than to the patient themselves. Conservative measures to employ include placing patients on their sides to facilitate secretion drainage and to minimize upper airway sounds, gentle oral and pharyngeal suctioning, and limiting fluid input.¹¹

One recent study reviewing the use of the anticholinergics atropine, scopolamine, and hyoscine demonstrated similar efficacy among the three drugs. Dosages used in this study included atropine 0.5 mg as a subcutaneous bolus, followed by 3 mg every 24 hours subcutaneously; scopolamine as a 0.25 mg subcutaneous bolus, followed by 1.5 mg every 24 hours IV or by subcutaneous infusion; and hyoscine 20 mg as a subcutaneous bolus followed by 60 mg every 24 hours IV or subcutaneous infusion. Glycopyrrolate is often used in the cognitively intact patient, as it does not cross the blood-brain barrier; however, supply concerns at the time of the study prevented a review of its efficiency.²⁷ All of these medications are also available in oral and transdermal formulations.

Terminal Delirium

Delirium is a common complication for patients nearing the end of life, affecting as many as 88% of dying patients.²⁸ It is characterized by an acute onset of cognitive impairment that may manifest as either a hyperactive or hypoactive state. Causes for terminal delirium are multifactorial. Initially, management should include prevention strategies, such as frequently orientating the patient, maintenance of day-night cycles, provision of adequate sleep, and minimization of sensory overload.¹¹ When pharmacological therapy is required to improve quality of life, a neuroleptic medication, namely haloperidol, should be used initially. The addition of a benzodiazepine may help if the initial treatments are ineffective, or if sedation is desired.²⁸

Summary

Emergency physicians have a unique opportunity to improve the quality of life for patients suffering serious illness, especially those who are actively dying. The management of pain and nonpain symptoms in patients who are at end of life, is a particularly important skill for every EP. If available, a consultation with a palliative care medicine consultant may improve both short- and long-term patient care.

Dr Galicia-Castillo is the Sue Faulkner Scribner professor of geriatrics at the Eastern Virginia Medical School Glennan Center for Geriatrics and Gerontology, and Medical Director for Palliative Care Medicine at Sentara Norfolk General Hospital. Dr Counselman is the distinguished professor and chairman of the department of emergency medicine at Eastern Virginia Medical School, Norfolk; and a physician at Emergency Physicians of Tidewater, Norfolk, Virginia. He is also the associate editor in chief of EMERGENCY MEDICINE editorial board.