How To Determine If A Patient Is Better Off With Buprenorphine

Transcription

1 Drug and Alcohol Dependence 86 (2007) Evaluation of the use of buprenorphine for opioid withdrawal in an emergency department M.L. Berg a, U. Idrees b,, R. Ding c, S.A. Nesbit b, H.K. Liang d, M.L. McCarthy c a University of Illinois-Chicago, Edward Hospital, Department of Pharmacy, 801 S. Washington Street, Naperville, IL 60540, United States b The Johns Hopkins Hospital, Department of Pharmacy, 600 North Wolfe Street, Carnegie 180, Baltimore, MD , United States c The Johns Hopkins University, Department of Emergency Medicine, Davis Building, Suite 3220, 5801 Smith Avenue, Baltimore, MD 21209, United States d The Johns Hopkins University, Department of Emergency Medicine, 600 North Wolfe Street, Marburg B-180, Baltimore, MD 21287, United States Received 5 May 2006; received in revised form 27 June 2006; accepted 28 June 2006 Abstract Objectives: To examine the use of buprenorphine for the treatment of opioid withdrawal (OW) in an emergency department (ED) setting. Methods: The medical records of all adult patients who presented to the study ED during a 10 week period for OW were abstracted. Subjects were categorized as receiving buprenorphine, symptomatic treatment or no pharmacologic treatment for their OW. The three groups were compared by patient and service characteristics, withdrawal symptoms and outcomes. Results: Of the 11,019 patients who presented to the ED during the 10 week study period, 158 (1.4%) were eligible. Subjects were more likely to receive buprenorphine (56%) compared to symptomatic treatment only (26%) or no pharmacologic treatment (18%). Subjects who received buprenorphine were more likely to have a history of suicide ideation (34% versus 12% p < 0.05) compared to subjects who received symptomatic treatment(s) and were less likely to present with a gastrointestinal complaint (9% versus 25% p < 0.05). Subjects who received buprenorphine were less likely to return to the same ED within 30 days for a drug-related visit (8%) compared to those who received symptomatic treatment (17%) (p < 0.05). Conclusions: Buprenorphine was a common treatment for OW in this ED without any documented adverse outcomes. Given that it did not result in an increase in drug-related return ED visits and its proven efficacy in other settings, a prospective evaluation of its potential value to ED patients who present with OW is warranted Elsevier Ireland Ltd. All rights reserved. Keywords: Buprenorphine; Opiate withdrawal; Opioid withdrawal; Heroin; Emergency department; Prescription opioids 1. Introduction In the United States, there are approximately 800,000 chronic heroin users and 3.5 million people who misuse prescription drugs (Fiellin et al., 2004). Opioid addiction is associated with significant adverse health outcomes such as transmission of HIV or hepatitis C, development of endocarditis and osteomyelitis, and increased mortality, especially among persons between the ages of 25 and 44 years old (Krantz and Mehler, 2004; Substance Abuse and Mental Health Services Administration, 2004a). Users often suffer from withdrawal symptoms as part of their physiologic dependence on opioids and may seek emergency department (ED) care for relief. In 2003, approximately Corresponding author. Tel.: ; fax: address: uidrees1@jhmi.edu (U. Idrees). 47,000 of all drug-related ED visits involved heroin and almost 25,000 visits involved an unspecified opiate (Substance Abuse and Mental Health Services Administration, 2004b). In the ED, treatment of opioid withdrawal (OW) primarily consists of symptomatic relief, including medications such as antiemetics, muscle relaxants, benzodiazepines, and clonidine (Krantz and Mehler, 2004; DiPaula et al., 2002). Buprenorphine, a mixed opioid agonist/antagonist, is used for opioid detoxification and for treatment of OW symptoms in inpatient and outpatient settings. Advantages of buprenorphine over full opioid agonists, such as methadone, are that it is associated with less respiratory depression at higher doses and causes fewer OW symptoms upon discontinuation. In addition, it has been shown to have improved efficacy in relieving subjective and objective withdrawal symptoms and is associated with fewer adverse effects when compared to other symptomatic treatments (Nigam et al., 1993; Lintzeris et al., 2002; O Connor et al., 1997; Janiri et /$ see front matter 2006 Elsevier Ireland Ltd. All rights reserved. doi: /j.drugalcdep

2 240 M.L. Berg et al. / Drug and Alcohol Dependence 86 (2007) al., 1994). The potential disadvantages of using buprenorphine in the ED setting include a possible increase in the incidence of patients returning to the ED requesting the medication and any adverse effects associated with the medication, primarily the precipitation of OW in patients who have recently used a full opioid agonist. The purpose of this pilot study was to examine the use of buprenorphine in an academic ED that treats a significant number of patients with opioid dependence. More specifically, we compared those who received buprenorphine for treatment of OW symptoms versus those who did not by patient and service characteristics as well as the number and type of OW symptoms documented. In addition, we examined differences in discharge disposition, ED length of stay, and frequency of drug-related return visits to the ED between subjects who received buprenorphine and those who did not. 2. Methods 2.1. Study design We conducted a retrospective chart review to identify patients with OW during a 10 week period. We determined whether there were significant differences between patients who received buprenorphine versus symptomatic treatment only or buprenorphine versus no pharmacologic treatment by patient or service characteristics, types of withdrawal symptoms and patient outcomes. The study protocol was approved by the hospital s Institutional Review Board Study setting and population The study setting is an urban ED within a large academic teaching hospital that serves as a regional level 1 trauma center and a state-designated cardiac consultation center. The study ED primarily treats a socioeconomically disadvantaged population who live in the surrounding community. The annual census of the study ED is approximately 60,000 visits Study protocol The medical records of all patients who presented to the adult ED between November 15, 2004 and January 31, 2005 were reviewed and patients were considered eligible if they met the following criteria: (1) 18 years of age or older at time of ED visit; (2) documentation of signs of OW or of OW occurring during the ED visit; and (3) a history of opioid abuse documented in the medical record. All of the medical records for each patient who presented during the study period were reviewed manually by one of the authors (MB), a pharmacy resident, to ensure consistency of the data abstraction process. A paper copy of the ED record is maintained in the ED for approximately 2 months before being sent to the medical record department. Each chart was screened for eligibility by viewing the chief complaints recorded by the ED triage nurse and the ED physician, the past medical history recorded by the ED triage nurse and the ED physician, and the discharge diagnoses recorded by the ED physician. An ED record is typically three pages. On average, it took approximately 2 h to screen all patients charts on a daily basis during the study period Measurements This study relied on medical record and registration data. The following types of information were abstracted from each subject s chart: (1) patient characteristics; (2) service characteristics; (3) presence of OW symptoms; (4) pharmacologic treatment(s) received; and (5) patient outcomes. Patient characteristics included age, gender, race, insurance status, drug abuse history, history of psychiatric conditions, and presenting complaint. The service characteristics collected included triage level, mode of arrival, and date and time of arrival. The presence of all OW symptoms that are included in two validated OW scales: the Short Opioid Withdrawal Scale (SOWS) and the Objective Opioid Withdrawal Scale (OOWS) were abstracted prior to pharmacologic treatment (Gossop, 1990; Peachey and Lei, 1988). In addition, any precipitation of OW symptoms that occurred following receipt of buprenorphine were also abstracted. All treatments rendered for OW were documented as well as the timing of each treatment. For patients who received buprenorphine, all doses were given parenterally because the oral formulations of buprenorphine were not available through our hospital formulary during the study period. The following outcomes were recorded on all subjects: precipitation of withdrawal after administration of buprenorphine, ED discharge disposition, ED length of stay, and return drug-related ED visits. ED length of stay was calculated as the difference between the triage time and the time of ED discharge. The electronic patient registration database was used to identify any subjects who returned to the same ED within 30 days of their index visit. The charts of all subjects who had a return ED visit within 30 days of their index visit were reviewed to determine whether any of the return visits were drug-related (as defined by the initial study criteria) Data analysis First, we classified each subject into one of three drug treatment groups: (1) buprenorphine (with or without symptomatic treatment); (2) symptomatic treatment(s) only; or (3) no pharmacologic treatment. A chi-square test was used to compare the frequency distribution of patient characteristics, service characteristics, OW symptoms, and patient outcomes between those who received buprenorphine and each of the other two treatment groups. ED length of stay was compared using a median test due to its skewed distribution. For all comparisons, differences were considered statistically significant if the p-value associated with the test statistic was Results During the study period, a total of 11,019 adult patients presented to the ED and 158 (1.4%) subjects met the inclusion criteria. Table 1 displays the types of pharmacologic treatment that subjects received in the ED. Eighty-eight subjects (56%) received buprenorphine, 42 (27%) received it alone and 46 (29%) received it in combination with other drug treatments. Of the 46 subjects who received both types of therapy, 39 (85%) received symptomatic treatment prior to buprenorphine. All doses of buprenorphine were administered by the intramuscular or intravenous route. Seventy out of the 88 patients (80%) who received buprenorphine, received a dose of 0.3 mg (range of mg). Table 1 Percent distribution of treatment received (N = 158) Treatment N % Buprenorphine Buprenorphine alone Buprenorphine + symptomatic treatment Symptomatic treatment(s) only a Dicyclomine Antiemetics 14 9 Benzodiazepines 9 6 Other Opioids 9 6 Clonidine 8 5 Others 2 1 No pharmacologic treatment a Some patients received more than one symptomatic treatment so the numbers and percents are greater than the total.

3 Table 2 Frequency distribution of patient and service characteristics by treatment group M.L. Berg et al. / Drug and Alcohol Dependence 86 (2007) Characteristic Overall (N = 158) (%) Buprenorphine (N = 88) (%) Symptomatic treatment (N = 41) (%) No pharmacologic treatment (N = 29) (%) Age (years) Male gender Race White African American Insurance status Private Medicaid/medicare Uninsured Past alcohol abuse Past tobacco abuse Past cocaine abuse History of psychiatric condition * None Suicide ideation Other Presenting complaint * Gastrointestinal Psychiatric Other Arrival mode Ambulance Family Police Self Triage level *, Weekday Triage time 07:00 14: :00 22: :00 06: * Significant difference between Buprenorphine group and symptomatic treatment group. Triage level definitions: level 1 are the sickest patients who require immediate treatment; level 2 are those who require care within 30 min; level 3 are those who require care within 1 h; and level 4 do not require care within a specified time period. Forty-one subjects (26%) received one or more medications for symptom relief. Of those who only received symptomatic treatment, 18 subjects received dicyclomine and 14 received an antiemetic. The antiemetics given in the ED for OW symptoms were promethazine, metoclopramide and dolasetron. Twentynine subjects (18%) did not receive any pharmacologic treatment in the ED for their OW symptoms. Table 2 displays the patient and service characteristics of the study sample by treatment group. The mean age of the study sample was 39 years old (S.D. = 8.6). One hundred and one subjects (64%) were male, 120 (76%) were African American and 101 (64%) were uninsured. There were no significant differences between those who received buprenorphine versus the other two treatment groups in terms of demographic characteristics such as age, gender, race, or insurance status. Thirty of the subjects (34%) who received buprenorphine had a history of suicide ideation compared to five subjects (12%) who only received symptomatic treatment(s) (p = 0.02). Only 10 subjects (9%) who received buprenorphine presented with a gastrointestinal chief complaint compared to 16 subjects (25%) who received symptomatic treatment only (p 0.05). Subjects who were categorized as emergent (triage level 1) were significantly more likely to receive buprenorphine compared to

4 242 M.L. Berg et al. / Drug and Alcohol Dependence 86 (2007) Table 3 Number and presence of opioid withdrawal symptoms by treatment type Characteristic Overall (N = 158) (%) Buprenorphine (N = 88) (%) Symptomatic treatment (N = 41) (%) No pharmacologic treatment (N = 29) (%) Number of symptoms None symptoms >3 symptoms Nausea * Vomiting * Complaint of feeling sick Goose flesh Sweating Restlessness/insomnia Tremor Lacrimation Nasal congestion Yawning Abdominal cramping * Feeling of coldness Uncontrollable shivering Muscle aches Twitching Feeling of heart pounding Heart rate > 100 beats per minute) Systolic blood pressure > 140 mmhg Diarrhea * General pain * Significant difference between Buprenorphine group and symptomatic treatment group. subjects who were categorized as non-urgent (triage levels 3 4) (p = 0.01). Table 3 presents the number and types of symptoms documented in subjects medical record before pharmacologic treatment (if any) by treatment group. Subjects who received buprenorphine were significantly less likely to have documentation of gastrointestinal symptoms such as nausea (n = 19, 22%), vomiting (n = 13, 15%), abdominal cramping (n = 26, 30%) or diarrhea (n = 6, 7%) compared to subjects who received symptomatic treatment only (n = 18, 44%; n = 15, 37%; n = 21, 51%; and n = 8, 20%, respectively) (p 0.05). As displayed in Table 4, 66 of the subjects (75%) who received buprenorphine were admitted to an inpatient service (75%) compared with 11 subjects (27%) who received symp- Table 4 Frequency distribution of ED disposition, length of stay (LOS) and drug-related ED visits by treatment group Outcome Overall (N = 158) (%) Buprenorphine (N = 88) (%) Symptomatic treatment (N = 41) (%) Disposition * Against medical advice Home Inpatient LOS for admitted patients (h) < > Median * LOS for discharge patients (h) < > Median (h) # Drug-related return ED visit within 30 days * * Significant difference between buprenorphine group and symptomatic treatment group. # Significant difference between buprenorphine group vs. symptomatic treatment group and buprenorphine group vs. no treatment group. No pharmacologic treatment (N = 29) (%)

5 M.L. Berg et al. / Drug and Alcohol Dependence 86 (2007) tomatic treatment (p < 0.001). Among the 88 subjects who were admitted, the 66 subjects who received buprenorphine had a shorter ED length of stay (median = 9.1 h) compared to the 11 subjects who only received symptomatic treatment (median = 11.2 h) (p = 0.05). In contrast, among the 66 subjects discharged, the 21 subjects who received buprenorphine had a significantly longer ED length of stay (median 12.6) compared to the 30 who received symptomatic treatment only (median 6.6 h) (p = 0.01) or to the 15 subjects who had no pharmacologic treatment (median = 4.6 h) (p < 0.01). Only 7 subjects (8%) who received buprenorphine returned to the same ED for a drug-related visit within 1 month of the index visit compared to 7 subjects (17%) who received symptomatic treatment only (p 0.05). Of the 16 subjects with a return drug-related visit, only 4 received buprenorphine during their return visit. Finally, there was no documentation of precipitation of OW among any of the study subjects who received buprenorphine in the ED. 4. Discussion We conducted this pilot study to examine the use of buprenorphine in an academic ED setting. We found that buprenorphine is commonly used in our ED to alleviate OW symptoms. Patients who received buprenorphine were significantly more likely to be admitted, have a history of suicide ideation and be acutely ill according to their triage level. There were no clear advantages to buprenorphine over other treatment approaches in terms of symptom management or ED length of stay, but there were also no major disadvantages. Specifically there was no increase in the number of drug-related return ED visits or precipitation of OW. Buprenorphine has been shown to be effective for opioid detoxification and OW symptom management in outpatient and inpatient settings. A study conducted by Lintzeris et al. demonstrated that patients who received buprenorphine experienced less discomfort during withdrawal, less heroin use, and increased post-withdrawal treatment retention compared to patients who received symptomatic medications (clonidine with or without antiemetics, benzodiazepines, muscle relaxants and/or ibuprofen) in managing heroin withdrawal in an ambulatory setting (Lintzeris et al., 2002). O Connor et al. compared buprenorphine to clonidine alone or clonidine with naltrexone for heroin detoxification in an ambulatory care setting (O Connor et al., 1997). All regimens were successful at detoxifying patients. However, patients who received buprenorphine experienced less severe withdrawal symptoms than those assigned to the other two groups. Several studies have also demonstrated the benefit of buprenorphine in the inpatient setting. Janiri et al., demonstrated in their study that intramuscular buprenorphine is superior to clonidine in controlling subjective and objective withdrawal symptoms in the inpatient methadone detoxification setting (Janiri et al., 1994). Buprenorphine has also been shown to be effective for OW symptom management. A study, conducted by Welsh et al., found that intravenous buprenorphine was an effective and safe treatment option for controlling OW symptoms in medically ill hospitalized patients (Welsh et al., 2002). Buprenorphine decreased withdrawal symptoms in all patients. It also did not appear to cause any respiratory depression or give any patients a euphoric feeling. Contrary to the studies cited above, we did not find that buprenorphine was more effective at relieving OW symptoms compared to symptomatic treatments, but this is probably a limitation of our study design rather than a true finding. This study was based upon chart review and providers may not have documented all OW symptoms. They did not rate the severity of the symptoms they documented and they did not document the monitoring of symptoms before and after treatment, so this study was not able to adequately evaluate the impact of buprenorphine on OW symptoms compared to other treatments. There were relatively few differences noted between patients who received buprenorphine and those who received symptomatic treatment or no pharmacologic treatment. Patients who received buprenorphine were more likely to be admitted. It may be that providers feel more comfortable administering buprenorphine, a controlled substance, if they know the patient will be admitted and will have access to a supervised opioid detoxification regimen with buprenorphine. In addition, it may have been easier to admit patients who received buprenorphine because the hospital has a unit that specifically treats patients who have both a substance abuse problem and a psychiatric condition and a higher proportion of patients who received buprenorphine presented with a psychiatric complaint. A higher proportion of patients who received buprenorphine had the highest acuity level but this also could have to do with a psychiatric chief complaint because a complaint such as suicide ideation is considered emergent. In contrast, patients who presented with gastrointestinal withdrawal symptoms were more likely to receive symptomatic treatment than buprenorphine. For patients with primarily gastrointestinal complaints, providers may be less likely to prescribe buprenorphine because they have effective alternatives available to treat these specific symptoms. Buprenorphine did not appear to have a consistent impact on ED length of stay. Among patients who were admitted, the ED length of stay was significantly shorter for patients who received buprenorphine compared to those who received symptomatic treatment. However, among patients who were discharged, length of stay was significantly longer for those who received buprenorphine than those who received symptomatic treatment or no pharmacologic treatment. There did not appear to be any difference in day of week or time of day that the patients presented by treatment group which could confound ED length of stay. Contrary to our expectations, patients who received buprenorphine were not more likely to return to our ED within 30 days for a drug-related visit. However, patients who received buprenorphine were more likely to be admitted, so they received more supervised treatment for OW and opioid detoxification compared to patients who were symptomatically treated in the ED and then discharged.

6 244 M.L. Berg et al. / Drug and Alcohol Dependence 86 (2007) Because of its partial opioid antagonist activity, the prescribing information for buprenorphine includes a warning that its use in the physically opioid dependent individual may result in withdrawal effects if given shortly after the administration of a full opioid agonist (Buprenex product information, 2005). However, in our study, there was no documentation of precipitation of withdrawal in any patients who received buprenorphine, which is likely because our inclusion criteria restricted patients to those who were already in OW. The use of buprenorphine in the ED does not appear to precipitate withdrawal, but an assessment of its safety in this setting warrants further evaluation Limitations This was a retrospective pilot study with limitations. As mentioned previously, the study relied on documentation available in the medical record, and this was particularly problematic for the assessment of drug treatments for OW symptoms. Second, the sample size of the study was small, but this was regarded as a pilot effort to determine whether a prospective study is warranted. Third, we used ED length of stay as a surrogate for time to relief of symptoms. However, length of stay can be affected by many variables, so this may not have been appropriate. Finally, we did not complete any follow-up on the study subjects to determine if the patients who were admitted were continued on their buprenorphine and their respective success rates for detoxification. 5. Conclusions This study provides preliminary data on the use of buprenorphine in an ED setting. Buprenorphine has been found to be effective for both opioid detoxification and management of OW symptoms in other settings. A significant proportion of our patients with OW received buprenorphine without any documented adverse consequences. Given that it did not result in an increase in drug-related return ED visits and its proven efficacy in other settings, a prospective evaluation of its potential value to ED patients who present with OW is warranted. Acknowledgements This research project was supported by internal funds only. The authors would like to acknowledge Yngvild Olsen, MD, MPH for providing insightful comments on the manuscript. References Buprenex (Buprenorphine) Injection (Product Information), Reckitt Benckiser Pharmaceuticals Inc., Richmond, VA. DiPaula, B.A., Schwartz, R., Montoya, I.D., Barrett, D., Tang, C., Heroin detoxification with buprenorphine on an inpatient psychiatric unit. J. Subst. Abuse Treat. 23, Fiellin, D.A., Kleber, H., Trumble-Hejduk, J.G., McLellan, A.T., Kosten, T.R., Consensus statement on office-based treatment of opioid dependence using buprenorphine. J. Subst. Abuse Treat. 27, Gossop, M., The development of a short opiate withdrawal scale (SOWS). Addict. Behav. 15, Janiri, L., Mannelli, P., Persico, A.M., Serretti, A., Tempesta, E., Opiate detoxification of methadone maintenance patients using lefetamine, clonidine, and buprenorphine. Drug Alcohol Depend. 36, Krantz, M.J., Mehler, P.S., Treating Opioid Dependence: Growing Implications for Primary Care. Arch. Intern. Med. 164, Lintzeris, N., Bell, J., Bammer, G., Jolley, D.J., Rushworth, L., A randomized controlled trial of buprenorphine in the management of short-term ambulatory heroin withdrawal. Addiction 97, Nigam, A.K., Ray, R., Tripathi, B.M., Buprenorphine in opiate withdrawal: a comparison with clonidine. J. Subst. Abuse Treat. 10, O Connor, P.G., Carroll, K.M., Shi, J.M., Schottenfeld, R.S., Kosten, T.R., Rounsaville, B.J., Three methods of opioid detoxification in a primary care setting. Ann. Intern. Med. 127 (7), Peachey, J.E., Lei, H., Assessment of opioid dependence with naloxone. Br. J. Addict. 83, Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 2002, 2004a. Mortality Data from the Drug Abuse Warning Network, DAWN series D-25, DHHS Publication No. (SMA) Rockville, MD. Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 2003, 2004b. Drug Abuse Warning Network, 2003: Interim National Estimates of Drug-Related Emergency Department Visits. DAWN series D- 26, DHHS Publication No. (SMA) Rockville, MD. Welsh, C.J., Suman, M., Cohen, A., Broyles, L., Bennett, M., Weintraub, E., The use of intravenous buprenorphine for the treatment of opioid withdrawal in medically ill hospitalized patients. Am. J. Addict. 11,