[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure 1.
CONSORT Diagram
CONSORT Diagram

Participant flowchart from screening through completion. ITT indicates intention-to-treat.

Figure 2.
Withdrawal Time Course
Withdrawal Time Course

Data represent mean peak ratings from the modified intention-to-treat (ITT) analyses as a function of study day. A, Clinical Opiate Withdrawal Scale (COWS) (an 11-item measure of opioid withdrawal symptoms with a possible score range of 0 to 48). B, Subjective Opiate Withdrawal Scale (SOWS) (a 16-item self-report measure of opioid withdrawal symptoms rated on a 5-point scale from 0 [not at all] to 4 [extremely], with a possible score range of 0 to 64). Group values were collapsed during the stabilization period (days −7 through −1) before randomization to study group. Study day 1 represents the first day of study medication dosing. Active dosing occurred during the taper period (days 1-7) and is designated with filled symbols; double-blind placebo dosing occurred during the post-taper period (days 8-14) and is designated with open symbols. Error bars represent SE.

Table 1.  
Tapering Schedulea
Tapering Schedulea
Table 2.  
Demographic and Drug Use Characteristics; Drug Identification Outcomes
Demographic and Drug Use Characteristics; Drug Identification Outcomes
Table 3.  
AUC Primary Outcomes
AUC Primary Outcomes
1.
Birnbaum  HG, White  AG, Schiller  M, Waldman  T, Cleveland  JM, Roland  CL.  Societal costs of prescription opioid abuse, dependence, and misuse in the United States.  Pain Med. 2011;12(4):657-667.PubMedGoogle ScholarCrossref
2.
Strassels  SA.  Economic burden of prescription opioid misuse and abuse.  J Manag Care Pharm. 2009;15(7):556-562.PubMedGoogle Scholar
3.
CDC Wonder. Compressed Mortality File. http://wonder.cdc.gov/mortsql.html. Updated 2014. Accessed June 1, 2017.
4.
Center for Behavioral Health Statistics and Quality. Treatment Episode Data Set 2002-2012. National Admissions to Substance Abuse Treatment Services. Rockville, MD: Department of Health and Human Services; 2014. HHS Publication No. (SMA) 14-4850. https://www.samhsa.gov/data/sites/default/files/TEDS2012N_Web.pdf.Accessed June 7, 2017.
5.
Bailey  GL, Herman  DS, Stein  MD.  Perceived relapse risk and desire for medication assisted treatment among persons seeking inpatient opiate detoxification.  J Subst Abuse Treat. 2013;45(3):302-305.PubMedGoogle ScholarCrossref
6.
Northrup  TF, Stotts  AL, Green  C,  et al.  Opioid withdrawal, craving, and use during and after outpatient buprenorphine stabilization and taper: a discrete survival and growth mixture model.  Addict Behav. 2015;41:20-28.PubMedGoogle ScholarCrossref
7.
Strobbe  S, Brower  KJ, Galen  LW.  Predicting completion of outpatient opioid detoxification with clonidine.  Am J Addict. 2003;12(3):260-269.PubMedGoogle ScholarCrossref
8.
Gowing  L, Farrell  M, Ali  R, White  JM. Alpha2-adrenergic agonists for the management of opioid withdrawal. Cochrane Database Syst Rev. 2016;(5):CD002024. Medline: 19370574
9.
Meader  N.  A comparison of methadone, buprenorphine and alpha2 adrenergic agonists for opioid detoxification: a mixed treatment comparison meta-analysis.  Drug Alcohol Depend. 2010;108(1-2):110-114.PubMedGoogle ScholarCrossref
10.
Jasinski  DR, Pevnick  JS, Griffith  JD.  Human pharmacology and abuse potential of the analgesic buprenorphine: a potential agent for treating narcotic addiction.  Arch Gen Psychiatry. 1978;35(4):501-516.PubMedGoogle ScholarCrossref
11.
Dunn  KE, Sigmon  SC, Strain  EC, Heil  SH, Higgins  ST.  The association between outpatient buprenorphine detoxification duration and clinical treatment outcomes: a review.  Drug Alcohol Depend. 2011;119(1-2):1-9.PubMedGoogle ScholarCrossref
12.
Nielsen  S, Larance  B, Degenhardt  L, Gowing  L, Kehler  C, Lintzeris  N.  Opioid agonist treatment for pharmaceutical opioid dependent people.  Cochrane Database Syst Rev. 2016;(5):CD011117.PubMedGoogle Scholar
13.
Hutchinson  E, Catlin  M, Andrilla  CH, Baldwin  LM, Rosenblatt  RA.  Barriers to primary care physicians prescribing buprenorphine.  Ann Fam Med. 2014;12(2):128-133.PubMedGoogle ScholarCrossref
14.
Molfenter  T, Sherbeck  C, Zehner  M,  et al.  Implementing buprenorphine in addiction treatment: payer and provider perspectives in Ohio.  Subst Abuse Treat Prev Policy. 2015;10(1):13.PubMedGoogle ScholarCrossref
15.
Schuman-Olivier  Z, Connery  H, Griffin  ML,  et al.  Clinician beliefs and attitudes about buprenorphine/naloxone diversion.  Am J Addict. 2013;22(6):574-580.PubMedGoogle ScholarCrossref
16.
Gillen  C, Haurand  M, Kobelt  DJ, Wnendt  S.  Affinity, potency and efficacy of tramadol and its metabolites at the cloned human mu-opioid receptor.  Naunyn Schmiedebergs Arch Pharmacol. 2000;362(2):116-121.PubMedGoogle ScholarCrossref
17.
Hennies  HH, Friderichs  E, Schneider  J.  Receptor binding, analgesic and antitussive potency of tramadol and other selected opioids.  Arzneimittelforschung. 1988;38(7):877-880.PubMedGoogle Scholar
18.
Raffa  RB, Friderichs  E, Reimann  W, Shank  RP, Codd  EE, Vaught  JL.  Opioid and nonopioid components independently contribute to the mechanism of action of tramadol, an “atypical” opioid analgesic.  J Pharmacol Exp Ther. 1992;260(1):275-285.PubMedGoogle Scholar
19.
Gibson  TP.  Pharmacokinetics, efficacy, and safety of analgesia with a focus on tramadol HCl.  Am J Med. 1996;101(1A):47S-53S.PubMedGoogle ScholarCrossref
20.
O’Connor  EC, Mead  AN.  Tramadol acts as a weak reinforcer in the rat self-administration model, consistent with its low abuse liability in humans.  Pharmacol Biochem Behav. 2010;96(3):279-286.PubMedGoogle ScholarCrossref
21.
Preston  KL, Jasinski  DR, Testa  M.  Abuse potential and pharmacological comparison of tramadol and morphine.  Drug Alcohol Depend. 1991;27(1):7-17.PubMedGoogle ScholarCrossref
22.
Lofwall  MR, Walsh  SL, Bigelow  GE, Strain  EC.  Modest opioid withdrawal suppression efficacy of oral tramadol in humans.  Psychopharmacology (Berl). 2007;194(3):381-393.PubMedGoogle ScholarCrossref
23.
Lofwall  MR, Babalonis  S, Nuzzo  PA, Siegel  A, Campbell  C, Walsh  SL.  Efficacy of extended-release tramadol for treatment of prescription opioid withdrawal: a two-phase randomized controlled trial.  Drug Alcohol Depend. 2013;133(1):188-197.PubMedGoogle ScholarCrossref
24.
Sobey  PW, Parran  TV  Jr, Grey  SF, Adelman  CL, Yu  J.  The use of tramadol for acute heroin withdrawal: a comparison to clonidine.  J Addict Dis. 2003;22(4):13-25.PubMedGoogle ScholarCrossref
25.
Chattopadhyay  S, Singh  OP, Bhattacharyya  A, Sen  S, Roy  P, Debnath  S.  Tramadol versus clonidine in management of heroin withdrawal.  Asian J Psychiatr. 2010;3(4):237-239.PubMedGoogle ScholarCrossref
26.
Tamaskar  R, Parran  TV  Jr, Heggi  A, Brateanu  A, Rabb  M, Yu  J.  Tramadol versus buprenorphine for the treatment of opiate withdrawal: a retrospective cohort control study.  J Addict Dis. 2003;22(4):5-12.PubMedGoogle ScholarCrossref
27.
Threlkeld  M, Parran  TV, Adelman  CA, Grey  SF, Yu  J.  Tramadol versus buprenorphine for the management of acute heroin withdrawal: a retrospective matched cohort controlled study.  Am J Addict. 2006;15(2):186-191.PubMedGoogle ScholarCrossref
28.
Zarghami  M, Masoum  B, Shiran  MR.  Tramadol versus methadone for treatment of opiate withdrawal: a double-blind, randomized, clinical trial.  J Addict Dis. 2012;31(2):112-117.PubMedGoogle ScholarCrossref
29.
Ziaaddini  H, Ziaaddini  A, Asghari  N, Nakhaee  N, Eslami  M.  Trial of tramadol plus gabapentin for opioid detoxification.  Iran Red Crescent Med J. 2014;17(1):e18202.PubMedGoogle ScholarCrossref
30.
Wesson  DR, Ling  W.  The Clinical Opiate Withdrawal Scale (COWS).  J Psychoactive Drugs. 2003;35(2):253-259.PubMedGoogle ScholarCrossref
31.
Gan  SH, Ismail  R, Wan Adnan  WA, Zulmi  W.  Impact of CYP2D6 genetic polymorphism on tramadol pharmacokinetics and pharmacodynamics.  Mol Diagn Ther. 2007;11(3):171-181.PubMedGoogle ScholarCrossref
32.
Lassen  D, Damkier  P, Brøsen  K.  The pharmacogenetics of tramadol.  Clin Pharmacokinet. 2015;54(8):825-836.PubMedGoogle ScholarCrossref
33.
Handelsman  L, Cochrane  KJ, Aronson  MJ, Ness  R, Rubinstein  KJ, Kanof  PD.  Two new rating scales for opiate withdrawal.  Am J Drug Alcohol Abuse. 1987;13(3):293-308.PubMedGoogle ScholarCrossref
34.
Morley  KI, Ferris  JA, Winstock  AR, Lynskey  MT.  Polysubstance use and misuse or abuse of prescription opioid analgesics: a multi-level analysis of international data.  Pain. 2017;158(6):1138-1144.PubMedGoogle ScholarCrossref
35.
Sigmon  SC, Dunn  KE, Badger  GJ, Heil  SH, Higgins  ST.  Brief buprenorphine detoxification for the treatment of prescription opioid dependence: a pilot study.  Addict Behav. 2009;34(3):304-311.PubMedGoogle ScholarCrossref
36.
Fiellin  DA, Schottenfeld  RS, Cutter  CJ, Moore  BA, Barry  DT, O’Connor  PG.  Primary care-based buprenorphine taper vs maintenance therapy for prescription opioid dependence: a randomized clinical trial.  JAMA Intern Med. 2014;174(12):1947-1954.PubMedGoogle ScholarCrossref
37.
Ravndal  E, Amundsen  EJ.  Mortality among drug users after discharge from inpatient treatment: an 8-year prospective study.  Drug Alcohol Depend. 2010;108(1-2):65-69.PubMedGoogle ScholarCrossref
38.
Kosten  TR, Rounsaville  BJ, Kleber  HD.  Comparison of clinician ratings to self reports of withdrawal during clonidine detoxification of opiate addicts.  Am J Drug Alcohol Abuse. 1985;11(1-2):1-10.PubMedGoogle ScholarCrossref
Original Investigation
September 2017

Efficacy of Tramadol Extended-Release for Opioid WithdrawalA Randomized Clinical Trial

Author Affiliations
  • 1Behavioral Pharmacology Research Unit, Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland
JAMA Psychiatry. 2017;74(9):885-893. doi:10.1001/jamapsychiatry.2017.1838
Key Points

Question  What is the relative efficacy of tramadol hydrochloride extended-release vs clonidine or buprenorphine hydrochloride in supervised opioid withdrawal settings?

Findings  This randomized clinical trial shows that tramadol extended-release suppressed withdrawal more than clonidine and comparably to buprenorphine during a double-blind taper and did not produce a delayed onset of opioid withdrawal after the taper.

Meaning  Tramadol extended-release is a Schedule IV medication that is available generically and can suppress withdrawal comparably to buprenorphine and superior to clonidine, suggesting that tramadol extended-release may have clinical value for the indication of opioid withdrawal suppression.

Abstract

Importance  Opioid use disorder (OUD) is a significant public health problem. Supervised withdrawal (ie, detoxification) from opioids using clonidine or buprenorphine hydrochloride is a widely used treatment.

Objective  To evaluate whether tramadol hydrochloride extended-release (ER), an approved analgesic with opioid and nonopioid mechanisms of action and low abuse potential, is effective for use in supervised withdrawal settings.

Design, Setting, and Participants  A randomized clinical trial was conducted in a residential research setting with 103 participants with OUD. Participants’ treatment was stabilized with morphine, 30 mg, administered subcutaneously 4 times daily. A 7-day taper using clonidine (n = 36), tramadol ER (n = 36), or buprenorphine (n = 31) was then instituted, and patients were crossed-over to double-blind placebo during a post-taper period. The study was conducted from October 25, 2010, to June 23, 2015.

Main Outcomes and Measures  Retention, withdrawal symptom management, concomitant medication utilization, and naltrexone induction. Results were analyzed over time and using area under the curve for the intention-to-treat and completer groups.

Results  Of the 103 participants, 88 (85.4%) were men and 43 (41.7%) were white; mean (SD) age was 28.9 (10.4) years. Buprenorphine participants (28 [90.3%]) were significantly more likely to be retained at the end of the taper compared with clonidine participants (22 [61.1%]); tramadol ER retention was intermediate and did not differ significantly from that of the other groups (26 [72.2%]; χ2 = 8.5, P = .01). Time-course analyses of withdrawal revealed significant effects of phase (taper, post taper) for the Clinical Opiate Withdrawal Scale (COWS) score (taper mean, 5.19 [SE, .26]; post-taper mean, 3.97 [SE, .23]; F2,170 = 3.6, P = .03) and Subjective Opiate Withdrawal Scale (SOWS) score (taper mean,8.81 [SE, .40]; post-taper mean, 4.14 [SE, .30]; F2,170 = 15.7, P < .001), but no group effects or group × phase interactions. Analyses of area under the curve of SOWS total scores showed significant reductions (F2,159 = 17.7, P < .001) in withdrawal severity between the taper and post-taper periods for clonidine (taper mean, 13.1; post-taper mean, 3.2; P < .001) and tramadol ER (taper mean, 7.4; post-taper mean, 2.8; P = .03), but not buprenorphine (taper mean, 6.4; post-taper mean, 7.4). Use of concomitant medication increased significantly (F2,159 = 30.7,  P < .001) from stabilization to taper in the clonidine (stabilization mean, 0.64 [SE, .05]; taper mean, 1.54 [SE, .10]; P < .001) and tramadol ER (stabilization mean, 0.53 [SE, .05]; taper mean, 1.19 [SE, .09]; P = .003) groups and from stabilization to post taper in the buprenorphine group (stabilization mean, 0.46 [SE, .05] post-taper mean, 1.17 [SE, .09]; P = .006), suggesting higher withdrawal for those groups during those periods. Naltrexone initiation was voluntary and the percentage of participants choosing naltrexone therapy within the clonidine (8 [22.2%]), tramadol ER (7 [19.4%]), or buprenorphine (3 [9.7%]) groups did not differ significantly (χ2 = 2.5, P = .29).

Conclusions and Relevance  The results of this trial suggest that tramadol ER is more effective than clonidine and comparable to buprenorphine in reducing opioid withdrawal symptoms during a residential tapering program. Data support further examination of tramadol ER as a method to manage opioid withdrawal symptoms.

Trial Registration  Clinicaltrials.gov Identifier: NCT01188421

Introduction

Quiz Ref IDOpioid use disorder (OUD) is a significant public health problem that costs $8 billion annually in the United States1,2 and contributes to unprecedented levels of overdose deaths.3 The most widely accessed treatment for OUD is medically supervised withdrawal (ie, detoxification),4 a brief and expedient treatment that can accommodate more patients per year compared with agonist maintenance treatments5 and is often a patient’s first treatment exposure.

Quiz Ref IDFailure to adequately manage opioid withdrawal symptoms contributes to withdrawal treatment attrition.6,7 Opioid withdrawal is primarily managed by 2 widely used medications. The first of these is clonidine, an unscheduled adrenergic agonist with low abuse liability that suppresses opioid withdrawal more than placebo8 but less than opioid agonists.9 Clonidine must be administered several times per day and can produce hypotension and sedation.8 The second medication is buprenorphine hydrochloride, a partial μ and opioid receptor like-1 receptor agonist and κ antagonist.10 Buprenorphine suppresses opioid withdrawal comparably to methadone and superior to clonidine11,12 and is effective with once-daily dosing.10 However, buprenorphine has known abuse potential and requires special authorization for use in several countries, including the United States, where it is a Schedule III medication that may be prescribed only for OUD treatment by physicians with specialized waivers; this limitation has slowed its adoption.13-15

Tramadol hydrochloride is a promising alternative option for effective OUD treatment. Tramadol is a mild to moderate opioid agonist with low affinity for the μ, κ, and δ opioid receptors and an active metabolite with μ agonist activity.16-18 The extended-release (ER) formulation of tramadol has a long elimination half-life (6-10 hours) that supports once-daily dosing.19 Tramadol is a Schedule IV drug that has less abuse liability than other opioid agonists,20,21 and accumulating evidence supports the evaluation of tramadol for opioid withdrawal management. Retrospective reviews, human laboratory studies, and randomized trials report that tramadol suppresses opioid withdrawal symptoms more than placebo22,23 and clonidine24,25 and is comparable to buprenorphine26,27 and methadone28,29 without increasing positive drug effects.22,23

The present study compared the relative efficacy of clonidine, tramadol ER, and buprenorphine during a randomized, double-blind, double-dummy, placebo-controlled opioid withdrawal procedure. The primary study aim was to compare opioid withdrawal suppression ratings during and after a taper; secondary aims evaluated retention and voluntary induction to naltrexone therapy.

Methods
Participants

Participants (N = 106) were recruited between October 25, 2010, and June 23, 2015. Eligible individuals were between age 18 and 60 years, met DSM-IV criteria for opioid dependence, provided a urine sample that tested positive on a qualitative test for opiates (morphine standard) and/or oxycodone or had evidence of opioid withdrawal by the end of the screening visit, and had no significant medical and/or psychiatric illnesses. Participants were excluded for pregnancy, hypotension, physical dependence on alcohol and/or benzodiazepines that required treatment, history of seizures, known allergies to study medications, or current enrollment in opioid agonist treatment. Three participants were randomized but left before receiving any study medications and so were omitted from analyses. The final sample size was 103 individuals with OUD (Figure 1). The protocol for the trial is available in Supplement 1.

The Johns Hopkins University Institutional Review Board approved the study, and all participants provided written informed consent. Participants received financial compensation.

Design
General Design

This was a 26- to 28-day residential study separated into 3 phases (eFigure 1 in Supplement 2). Double-blind, double-dummy medication was prepared by a research pharmacist who had no participant interaction, and participants and staff were blinded to participant group assignment. Withdrawal assessments were completed 7 times daily; details are provided below.

Morphine Stabilization: Phase 1

Participants’ treatment was stabilized with subcutaneous injections of morphine, 30 mg, administered 4 times daily for a 7- to 10-day period. Placebo was substituted for the last 2 morphine doses on the final stabilization day to allow for potential transition to buprenorphine.

Naloxone Challenge: Phase 1

Quiz Ref IDA naloxone challenge was conducted during the morphine phase to determine the level of physical dependence for group stratification. On the challenge day, participants received a morning dose of morphine, completed baseline withdrawal ratings, and then received a 0.4-mg intramuscular dose of naloxone. Withdrawal ratings were collected at 15-minute intervals for 3 hours, and morphine was administered at the session’s end.

Randomization to Study Group

The research pharmacist randomized participants (1:1:1) to receive clonidine, tramadol ER, or buprenorphine, using an urn randomization procedure with the following stratification variables: sex, race (African American/other), CYP2D6 genotype (extensive/heterozygous, intermediate, ultraextensive poor; Quest Diagnostics), and peak Clinical Opiate Withdrawal Scale (COWS) score rating30 (an 11-item measure of opioid withdrawal symptoms with a possible score range of 0 to 48 and higher levels indicating more severe withdrawal) with stratification of ≥9 vs <9 from the naloxone challenge session. The CYP2D6 genotype was included because tramadol is differentially metabolized by polymorphisms of this enzyme.31,32

Taper: Phase 2

Beginning on the first taper day (ie, day 1), participants ingested 1 capsule orally 4 times a day that contained either clonidine, tramadol hydrochloride, or placebo, and four 2-mg–sized sublingual tablets once daily that contained either buprenorphine hydrochloride or placebo, depending on the study group. Study medications were tapered down in a double-blind, double-dummy fashion during days 1 to 7 (Table 1).

Post Taper: Phase 3

Beginning on the first post-taper day (ie, day 8), all participants were crossed over to placebo capsules and tablets in a double-blind, double-dummy manner for the duration of the study. A second naloxone challenge session was conducted 3 to 5 days before discharge to verify successful withdrawal from opioids. Participants who chose naltrexone were provided a 30-day supply on discharge. All participants were referred for follow-up treatment.

Measures

Study staff completed the COWS.30 Participants completed the Subjective Opiate Withdrawal Scale (SOWS),33 a 16-item self-report measure of opioid withdrawal symptoms rated on a 5-point scale from 0 (not at all) to 4 (extremely), with a possible score range of 0 to 64. Beginning on day 1, participants also completed a daily drug identification questionnaire to identify which study medication (eg, clonidine, tramadol ER, buprenorphine) they believed they were receiving. Pupil diameter was assessed via a pupilometer (NeurOptics Inc).

Adverse events (AEs) were evaluated throughout each day and rated for study medication association and severity (mild, moderate, and severe). The study did not differentiate opioid withdrawal symptoms from AEs.

Medications

Each study day, participants received oral capsules containing clonidine, tramadol ER, or placebo and sublingual tablets containing buprenorphine or placebo. Active dosing began on study day 1, and participants did not receive more than 1 active taper medication. Dose schedules are reported in Table 1.

The following medications were available for concomitant administration by patient request during the taper and post-taper phases: milk of magnesia, antacid, loperamide hydrochloride, simethicone, ibuprofen, acetaminophen, hydroxyzine hydrochloride, and/or diphenhydramine hydrochloride. The total number of medications consumed daily (range, 0-8) was recorded for each participant.

All participants were informed about naltrexone at the beginning of the study. Participants who completed the second naloxone challenge with no evidence of withdrawal and requested naltrexone were inducted into 50 mg orally daily and were provided a 30-day supply on study discharge.

Statistical Analysis

This study hypothesized that both tramadol ER and buprenorphine would suppress opioid withdrawal symptoms greater than clonidine and that buprenorphine would suppress withdrawal more than tramadol ER. A power analysis based on a previous nonrandomized comparison of clonidine and tramadol,24 assuming an α level of .05 with 80% power to detect a moderate effect size (0.29), determined that 40 participants per group would detect between-group differences. Owing to the expiration of placebo buprenorphine tablets, recruitment was terminated after 103 participants (85.8% of the intended completer sample).

Baseline demographic and drug use characteristics were compared across groups using the χ2 test for dichotomous variables and 1-factor models for continuous variables. Retention was evaluated as being enrolled on the final day of the taper and the study; each retention outcome was compared across groups separately as a dichotomous variable with generalized estimating equation analyses, using site as a covariate. To assess the fidelity of study blinding, the mean percentage of days during the taper phase that participants correctly identified their medication on the Drug Identification Questionnaire was compared using 1-factor models. Withdrawal outcomes were analyzed by medication group and study phase; the post-taper period was defined as the first 7 days of placebo dosing. Since COWS, SOWS, and pupil diameter measurements were collected 7 times daily, the peak and nadir values that represented the most severe rating were identified for each participant for each study day. The time course of COWS, SOWS, pupil diameter, and concomitant medications by group, phase (stabilization, taper, post taper), and group × phase interactions were compared with Proc Mixed analyses. Area under the curve (AUC) was also derived for each participant and modeled as a function of medication group, phase, and group × phase interactions. Tukey tests were used for post hoc analyses of between-group differences to detect an effect of the study medication across groups, and between-phase differences were used to detect an effect within the same medication group across phases. The percentage of participants who chose to be inducted onto naltrexone was compared across groups with generalized estimating equation analyses, using site as a covariate. Adverse events are described descriptively. All withdrawal data were evaluated with modified (participants who received ≥1 study medication dose) intention-to-treat (ITT) and completers (defined as remaining enrolled on the final taper day) analyses. Missing data were treated as missing, α level was set to .05, and models were conducted using Proc Mixed in SAS, version 9.3 (SAS Institute).

Results
Participants

Of the 103 participants, 88 (85.4%) were men and 43 (41.7%) were white; mean (SD) age was 28.9 (10.4) years. A total of 96 (93.3%) and 49 (47.9%) participants reported past 30-day use of heroin and prescription opioids, respectively. There were no significant between-group demographic differences (Table 2). A main effect of group on retention (χ2 = 8.5, P = .01) revealed that buprenorphine participants (28 [90.3%]) were significantly more likely to be enrolled on the final taper day than clonidine participants (22 [61.1%], P = .01); tramadol ER participants did not differ significantly from either group (26 [72.2%]). There were no between-group differences in remaining enrolled on the final study day for clonidine (21 [58.3%]), tramadol ER (21 [58.3%]), or buprenorphine (18 [58.1%]) participants. The study blinding was relatively well maintained (Table 2). The buprenorphine group correctly identified their treatment medication more frequently than those receiving clonidine or tramadol ER.

Observer-Rated Outcomes

Mean peak COWS total score ratings as a function of study phase are displayed for the ITT group in Figure 2A. Analyses revealed a significant main effect of study phase (F2,170 = 3.6, P = .03). Ratings increased at the start of the taper and then decreased for all groups and diverged during the post-taper phase (Figure 2A). A significant main effect of phase (P = .03) and group × phase interaction (P < .001) was observed in the ITT COWS AUC analyses (Table 3).

The COWS time-course analyses of completers did not reveal any significant main effects or interactions. The AUC analyses of completers revealed a significant main effect of phase (P < .001) and group × phase interaction (P < .001) (Table 3).

Patient-Reported Outcomes

The pattern observed on the SOWS total score was similar to that observed for the COWS (Figure 2B). The ITT time-course analyses revealed a main effect of phase (F2,170 = 15.7, P < .001), and the AUC analyses revealed a significant main effect of phase (P < .001) and a group × phase interaction (P < .001) (Table 3). Quiz Ref IDPost hoc between-group testing of the AUC outcomes indicated that clonidine participants experienced significantly more severe withdrawal symptoms compared with tramadol ER (P = .02) and buprenorphine (P < .001) participants during the taper but not post-taper phases (eFigure 2A in Supplement 2). Post hoc between-phase testing of AUC outcomes revealed that both clonidine (P < .001) and tramadol ER (P = .03), but not buprenorphine, participants reported significant reductions in withdrawal symptoms between the taper and post-taper phases (eFigure 2B in Supplement 2; Table 3).

Analyses of SOWS total scores among completers revealed a significant main effect of phase in the time course (F2,143 = 13.0, P < .001) and significant main effect of phase (P < .001) and group × phase interactions (P < .001) in the AUC analyses. Consistent with the ITT results, post hoc between-phase testing of completer AUC outcomes indicated that both the clonidine (P < .001) and tramadol ER (P = .03), but not buprenorphine, participants experienced significant reductions in withdrawal symptoms between the taper and post-taper phases (eFigure 2).

Pupil Diameter

Intention-to-treat time-course analyses revealed a significant main effect of phase (F2,170 = 21.4, P < .001) and group × phase interaction (F4,170 = 3.5, P = .01) on pupil diameter. Post hoc testing demonstrated that pupil diameter size increased slightly between the stabilization and the taper (P < .001) and post-taper (P < .001) phases within the clonidine and tramadol ER groups; the buprenorphine group experienced increased pupil diameters only in the post-taper period. The AUC analyses of pupil diameter among ITT participants revealed significant main effects of phase (P < .001) and a group × phase interaction (P < .001) (Table 3).

Time-course analyses of completers also revealed a significant main effect of phase (F2,143 = 27.3, P < .001) and a group × phase interaction (F4,143 = 3.8, P = .005) on pupil diameter in the same directions as the ITT analyses. The AUC analyses of pupil diameter among completers also revealed a significant main effect of phase and a group × phase interaction (Table 3).

Concomitant Medications

The number of concomitant medications used was relatively low overall. The ITT time-course analyses revealed different patterns of use as a function of phase (F2,158 = 25.7, P < .001), but not group. Post hoc testing indicated that the clonidine (P < .001) and tramadol ER (P = .003) groups used significantly more concomitant medications during the taper phase, whereas buprenorphine participants used significantly more concomitant medications during the post-taper phase (P = .006), compared with the stabilization phase. A nonsignificant trend was also observed whereby clonidine participants consumed more medications compared with buprenorphine participants during the taper phase (P = .07). Overall, participants in each group consumed, on average, between 0 and 2 medications per day during the taper and post-taper phases. The completer results followed a similar pattern; Table 3 summarizes the results of the AUC concomitant medication analyses.

Naltrexone Induction

No significant between-group difference was observed in the percentage of participants from the clonidine (8 [22.2%]), tramadol ER (7 [19.4%]), or buprenorphine (3 [9.7%]) groups who were eligible for and voluntarily began naltrexone treatment (χ2 = 2.5, P = .29).

Adverse Events

A full list of AEs is presented in the eTable in Supplement 2. Approximately 50.0% to 55.6% of participants in each experimental group reported an AE, and no serious AEs were reported. The most common AEs were symptoms of opioid withdrawal, including muscle aches and pains (46 participants [44.7%]), rhinorrhea (45 [44.7%]), diarrhea (42 [40.0%]), headache (41 [39.8%]), and insomnia (41 [39.8%]).

Discussion

This study compared tramadol ER with clonidine and buprenorphine for opioid withdrawal symptom management in patients with OUD. Results show that tramadol ER produced retention and withdrawal suppression greater than clonidine and comparable to buprenorphine during a 7-day taper and was associated with less severe withdrawal symptoms relative to buprenorphine in the post-taper period. No significant between-group differences in naltrexone induction were identified, although induction was not mandatory. These data empirically support the use and further evaluation of tramadol ER for medically supervised opioid withdrawal.

Tramadol ER produced results that were similar to those of buprenorphine in several study outcomes, which supports its further evaluation for use in opioid withdrawal symptom management. Results are tempered by the high rates of relapse generally associated with supervised withdrawal relative to maintenance treatments, and epidemiologic evidence that tramadol is abusable.34 Furthermore, in the United States, tramadol ER was recently moved from an unscheduled medication to Schedule IV, which adds complexity in its use for opioid withdrawal suppression. Additional studies evaluating tramadol ER’s effectiveness in patients with more severe withdrawal than what was seen here, as well as outpatient treatment settings and in combination with extended care to mitigate the risk of overdose, are necessary to validate use of this medication as an OUD treatment.

Quiz Ref IDBetween the taper and post-taper phases, SOWS withdrawal ratings decreased among clonidine and tramadol ER participants but remained the same or slightly increased for buprenorphine participants; this outcome was mirrored with increases in concomitant medications during these periods. Together, these findings suggest that buprenorphine participants experience continued withdrawal following the removal of buprenorphine. Although a delayed onset of withdrawal has been previously reported following buprenorphine withdrawal,11,35 it is notable that this pattern was not observed in the other groups. Up to 70% of patients in primary care undergoing buprenorphine withdrawal relapse within 30 days,36 preventing induction to naltrexone and increasing the risk of overdose due to changes in tolerance.37 These data call for further research into the contribution that withdrawal symptoms in the days following removal of buprenorphine may have on opioid relapse and transition to naltrexone. Participants treated with tramadol ER in this study did not report an increase in withdrawal symptoms in the post-taper period, suggesting that tramadol ER should be evaluated as a method to facilitate patient transition to naltrexone or agonist maintenance more generally.

The doses of tramadol ER used here (up to 600 mg/d) are higher than recommended for the treatment of chronic pain (300 mg/d). Although these increased doses may be appropriate for patients with high levels of opioid tolerance, additional research establishing effective doses in samples with varying levels of opioid physical dependence is needed, as higher doses increase the risk for non-opioid–related AEs. Tramadol is also contraindicated for concomitant use with medications that induce/inhibit CYP3A4 and CYP2D6 enzymes, and coadministration could result in increased serum levels of tramadol and its metabolite in a manner that could increase the risk of opioid agonist and other AEs.

Limitations

Both extended inpatient and opioid withdrawal studies make participant recruitment challenging, and this study also had limited inpatient research resources that resulted in the study being underpowered. This limitation could explain why the COWS results trended but did not reach the same significance as the SOWS results. However, the clonidine and buprenorphine group outcomes are consistent with those of previous studies,9 and self-reported withdrawal symptoms are better indicators of patient retention in treatment than observer ratings,38 which increases the value of the SOWS results. In addition, withdrawal symptom levels in this study were mild, although they are consistent with those observed in other taper studies.11 Differences in taper completion rates between the tramadol ER (72.2%) and buprenorphine (90.3%) groups, as well as voluntary naltrexone induction, may also have reached statistical significance with more participants. Finally, a primarily male sample and lack of specific information regarding past 30-day use of other illicit drugs and alcohol also limit the study and its conclusions.

Conclusions

Tramadol ER produced the same level of treatment retention and opioid withdrawal symptom suppression as buprenorphine. Tramadol ER is a Schedule IV medication that can be administered once a day and has lower abuse liability than other opioid agonists. These data suggest that tramadol ER is a promising and valuable medication for the management of opioid withdrawal in patients undergoing treatment for OUD. Future studies should evaluate whether relapse varies following supervised withdrawal with tramadol ER vs other medications and whether tramadol ER can be used to transition patients to naltrexone treatment.

Back to top
Article Information

Accepted for Publication: May 16, 2017.

Corresponding Author: Kelly E. Dunn, PhD, Behavioral Pharmacology Research Unit, Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, 5510 Nathan Shock Dr, Baltimore, MD 21224 (kdunn9@jhmi.edu).

Correction: This article was corrected on September 6, 2017, for errors in the abstract and an error in the discussion section; this article was corrected on February 14, 2018, to fix an error in the Key Points and the Conclusions of the article.

Published Online: July 12, 2017. doi:10.1001/jamapsychiatry.2017.1838

Author Contributions: Dr Dunn had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Strain.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Dunn, Strain.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Dunn, Strain.

Obtained funding: Strain.

Administrative, technical, or material support: All authors.

Supervision: Tompkins, Strain.

Conflict of Interest Disclosures: Dr Strain has been a paid consultant for Indivior Pharmaceuticals and is a paid advisory board member for the Egalet Corporation, The Oak Group, and Pinney Associates. Dr Tompkins has received medication supplies from Indivior Inc for an investigator-initiated research protocol; has been a paid consultant with Astra-Zeneca and Theravance; and is site principal investigator for a multisite clinical trial funded by Alkermes. No other disclosures are reported.

Funding/Support: This study was funded by National Institute on Drug Abuse grant R01DA-018125 (Dr Strain), with additional salary support provided by grants R01DA035246, R01DA040644 (Dr Dunn), K23DA029609 (Dr Tompkins), and T32DA007209 (Dr Bigelow). Buprenorphine sublingual tablets and placebo tablets were provided by an unrestricted, unsolicited investigator-initiated grant from Reckitt Benckiser Pharmaceuticals.

Role of the Funder/Sponsor: The National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Reckitt Benckiser Pharmaceuticals had no role in the study design, collection, analysis, and interpretation of data or in the writing of the manuscript, but reviewed the report for scientific accuracy.

Additional Contributions: Paul Nuzzo, MS (University of Kentucky), helped with data analyses and Jessica Sides, BA, and Hye Jeong Han, BA (conducted at The Johns Hopkins University), managed daily study activities; they received financial compensation for these contributions. The Behavioral Pharmacology Research Unit and Bayview Clinical Research Unit staff provided many varied and valuable contributions.

References
1.
Birnbaum  HG, White  AG, Schiller  M, Waldman  T, Cleveland  JM, Roland  CL.  Societal costs of prescription opioid abuse, dependence, and misuse in the United States.  Pain Med. 2011;12(4):657-667.PubMedGoogle ScholarCrossref
2.
Strassels  SA.  Economic burden of prescription opioid misuse and abuse.  J Manag Care Pharm. 2009;15(7):556-562.PubMedGoogle Scholar
3.
CDC Wonder. Compressed Mortality File. http://wonder.cdc.gov/mortsql.html. Updated 2014. Accessed June 1, 2017.
4.
Center for Behavioral Health Statistics and Quality. Treatment Episode Data Set 2002-2012. National Admissions to Substance Abuse Treatment Services. Rockville, MD: Department of Health and Human Services; 2014. HHS Publication No. (SMA) 14-4850. https://www.samhsa.gov/data/sites/default/files/TEDS2012N_Web.pdf.Accessed June 7, 2017.
5.
Bailey  GL, Herman  DS, Stein  MD.  Perceived relapse risk and desire for medication assisted treatment among persons seeking inpatient opiate detoxification.  J Subst Abuse Treat. 2013;45(3):302-305.PubMedGoogle ScholarCrossref
6.
Northrup  TF, Stotts  AL, Green  C,  et al.  Opioid withdrawal, craving, and use during and after outpatient buprenorphine stabilization and taper: a discrete survival and growth mixture model.  Addict Behav. 2015;41:20-28.PubMedGoogle ScholarCrossref
7.
Strobbe  S, Brower  KJ, Galen  LW.  Predicting completion of outpatient opioid detoxification with clonidine.  Am J Addict. 2003;12(3):260-269.PubMedGoogle ScholarCrossref
8.
Gowing  L, Farrell  M, Ali  R, White  JM. Alpha2-adrenergic agonists for the management of opioid withdrawal. Cochrane Database Syst Rev. 2016;(5):CD002024. Medline: 19370574
9.
Meader  N.  A comparison of methadone, buprenorphine and alpha2 adrenergic agonists for opioid detoxification: a mixed treatment comparison meta-analysis.  Drug Alcohol Depend. 2010;108(1-2):110-114.PubMedGoogle ScholarCrossref
10.
Jasinski  DR, Pevnick  JS, Griffith  JD.  Human pharmacology and abuse potential of the analgesic buprenorphine: a potential agent for treating narcotic addiction.  Arch Gen Psychiatry. 1978;35(4):501-516.PubMedGoogle ScholarCrossref
11.
Dunn  KE, Sigmon  SC, Strain  EC, Heil  SH, Higgins  ST.  The association between outpatient buprenorphine detoxification duration and clinical treatment outcomes: a review.  Drug Alcohol Depend. 2011;119(1-2):1-9.PubMedGoogle ScholarCrossref
12.
Nielsen  S, Larance  B, Degenhardt  L, Gowing  L, Kehler  C, Lintzeris  N.  Opioid agonist treatment for pharmaceutical opioid dependent people.  Cochrane Database Syst Rev. 2016;(5):CD011117.PubMedGoogle Scholar
13.
Hutchinson  E, Catlin  M, Andrilla  CH, Baldwin  LM, Rosenblatt  RA.  Barriers to primary care physicians prescribing buprenorphine.  Ann Fam Med. 2014;12(2):128-133.PubMedGoogle ScholarCrossref
14.
Molfenter  T, Sherbeck  C, Zehner  M,  et al.  Implementing buprenorphine in addiction treatment: payer and provider perspectives in Ohio.  Subst Abuse Treat Prev Policy. 2015;10(1):13.PubMedGoogle ScholarCrossref
15.
Schuman-Olivier  Z, Connery  H, Griffin  ML,  et al.  Clinician beliefs and attitudes about buprenorphine/naloxone diversion.  Am J Addict. 2013;22(6):574-580.PubMedGoogle ScholarCrossref
16.
Gillen  C, Haurand  M, Kobelt  DJ, Wnendt  S.  Affinity, potency and efficacy of tramadol and its metabolites at the cloned human mu-opioid receptor.  Naunyn Schmiedebergs Arch Pharmacol. 2000;362(2):116-121.PubMedGoogle ScholarCrossref
17.
Hennies  HH, Friderichs  E, Schneider  J.  Receptor binding, analgesic and antitussive potency of tramadol and other selected opioids.  Arzneimittelforschung. 1988;38(7):877-880.PubMedGoogle Scholar
18.
Raffa  RB, Friderichs  E, Reimann  W, Shank  RP, Codd  EE, Vaught  JL.  Opioid and nonopioid components independently contribute to the mechanism of action of tramadol, an “atypical” opioid analgesic.  J Pharmacol Exp Ther. 1992;260(1):275-285.PubMedGoogle Scholar
19.
Gibson  TP.  Pharmacokinetics, efficacy, and safety of analgesia with a focus on tramadol HCl.  Am J Med. 1996;101(1A):47S-53S.PubMedGoogle ScholarCrossref
20.
O’Connor  EC, Mead  AN.  Tramadol acts as a weak reinforcer in the rat self-administration model, consistent with its low abuse liability in humans.  Pharmacol Biochem Behav. 2010;96(3):279-286.PubMedGoogle ScholarCrossref
21.
Preston  KL, Jasinski  DR, Testa  M.  Abuse potential and pharmacological comparison of tramadol and morphine.  Drug Alcohol Depend. 1991;27(1):7-17.PubMedGoogle ScholarCrossref
22.
Lofwall  MR, Walsh  SL, Bigelow  GE, Strain  EC.  Modest opioid withdrawal suppression efficacy of oral tramadol in humans.  Psychopharmacology (Berl). 2007;194(3):381-393.PubMedGoogle ScholarCrossref
23.
Lofwall  MR, Babalonis  S, Nuzzo  PA, Siegel  A, Campbell  C, Walsh  SL.  Efficacy of extended-release tramadol for treatment of prescription opioid withdrawal: a two-phase randomized controlled trial.  Drug Alcohol Depend. 2013;133(1):188-197.PubMedGoogle ScholarCrossref
24.
Sobey  PW, Parran  TV  Jr, Grey  SF, Adelman  CL, Yu  J.  The use of tramadol for acute heroin withdrawal: a comparison to clonidine.  J Addict Dis. 2003;22(4):13-25.PubMedGoogle ScholarCrossref
25.
Chattopadhyay  S, Singh  OP, Bhattacharyya  A, Sen  S, Roy  P, Debnath  S.  Tramadol versus clonidine in management of heroin withdrawal.  Asian J Psychiatr. 2010;3(4):237-239.PubMedGoogle ScholarCrossref
26.
Tamaskar  R, Parran  TV  Jr, Heggi  A, Brateanu  A, Rabb  M, Yu  J.  Tramadol versus buprenorphine for the treatment of opiate withdrawal: a retrospective cohort control study.  J Addict Dis. 2003;22(4):5-12.PubMedGoogle ScholarCrossref
27.
Threlkeld  M, Parran  TV, Adelman  CA, Grey  SF, Yu  J.  Tramadol versus buprenorphine for the management of acute heroin withdrawal: a retrospective matched cohort controlled study.  Am J Addict. 2006;15(2):186-191.PubMedGoogle ScholarCrossref
28.
Zarghami  M, Masoum  B, Shiran  MR.  Tramadol versus methadone for treatment of opiate withdrawal: a double-blind, randomized, clinical trial.  J Addict Dis. 2012;31(2):112-117.PubMedGoogle ScholarCrossref
29.
Ziaaddini  H, Ziaaddini  A, Asghari  N, Nakhaee  N, Eslami  M.  Trial of tramadol plus gabapentin for opioid detoxification.  Iran Red Crescent Med J. 2014;17(1):e18202.PubMedGoogle ScholarCrossref
30.
Wesson  DR, Ling  W.  The Clinical Opiate Withdrawal Scale (COWS).  J Psychoactive Drugs. 2003;35(2):253-259.PubMedGoogle ScholarCrossref
31.
Gan  SH, Ismail  R, Wan Adnan  WA, Zulmi  W.  Impact of CYP2D6 genetic polymorphism on tramadol pharmacokinetics and pharmacodynamics.  Mol Diagn Ther. 2007;11(3):171-181.PubMedGoogle ScholarCrossref
32.
Lassen  D, Damkier  P, Brøsen  K.  The pharmacogenetics of tramadol.  Clin Pharmacokinet. 2015;54(8):825-836.PubMedGoogle ScholarCrossref
33.
Handelsman  L, Cochrane  KJ, Aronson  MJ, Ness  R, Rubinstein  KJ, Kanof  PD.  Two new rating scales for opiate withdrawal.  Am J Drug Alcohol Abuse. 1987;13(3):293-308.PubMedGoogle ScholarCrossref
34.
Morley  KI, Ferris  JA, Winstock  AR, Lynskey  MT.  Polysubstance use and misuse or abuse of prescription opioid analgesics: a multi-level analysis of international data.  Pain. 2017;158(6):1138-1144.PubMedGoogle ScholarCrossref
35.
Sigmon  SC, Dunn  KE, Badger  GJ, Heil  SH, Higgins  ST.  Brief buprenorphine detoxification for the treatment of prescription opioid dependence: a pilot study.  Addict Behav. 2009;34(3):304-311.PubMedGoogle ScholarCrossref
36.
Fiellin  DA, Schottenfeld  RS, Cutter  CJ, Moore  BA, Barry  DT, O’Connor  PG.  Primary care-based buprenorphine taper vs maintenance therapy for prescription opioid dependence: a randomized clinical trial.  JAMA Intern Med. 2014;174(12):1947-1954.PubMedGoogle ScholarCrossref
37.
Ravndal  E, Amundsen  EJ.  Mortality among drug users after discharge from inpatient treatment: an 8-year prospective study.  Drug Alcohol Depend. 2010;108(1-2):65-69.PubMedGoogle ScholarCrossref
38.
Kosten  TR, Rounsaville  BJ, Kleber  HD.  Comparison of clinician ratings to self reports of withdrawal during clonidine detoxification of opiate addicts.  Am J Drug Alcohol Abuse. 1985;11(1-2):1-10.PubMedGoogle ScholarCrossref
×