Intervention Review

You have full text access to this OnlineOpen article

Methadone at tapered doses for the management of opioid withdrawal

  1. Laura Amato1,*,
  2. Marina Davoli1,
  3. Silvia Minozzi1,
  4. Eliana Ferroni2,
  5. Robert Ali3,
  6. Marica Ferri4

Editorial Group: Cochrane Drugs and Alcohol Group

Published Online: 28 FEB 2013

Assessed as up-to-date: 20 JUL 2012

DOI: 10.1002/14651858.CD003409.pub4

How to Cite

Amato L, Davoli M, Minozzi S, Ferroni E, Ali R, Ferri M. Methadone at tapered doses for the management of opioid withdrawal. Cochrane Database of Systematic Reviews 2013, Issue 2. Art. No.: CD003409. DOI: 10.1002/14651858.CD003409.pub4.

Author Information

  1. 1

    Lazio Regional Health Service, Department of Epidemiology, Rome, Italy

  2. 2

    Public Health Agency of Lazio Region, Infectious Diseases Unit, Rome, Italy

  3. 3

    University of Adelaide, Discipline of Pharmacology, Adelaide, SA, Australia

  4. 4

    European Monitoring Centre for Drugs and Drug Addiction, Interventions, Best Practice and Scientific Partners, Lisbon, Portugal

*Laura Amato, Department of Epidemiology, Lazio Regional Health Service, Via di Santa Costanza, 53, Rome, 00198, Italy. l.amato@deplazio.it.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 28 FEB 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Bearn 1996

MethodsRandomised controlled trial. Setting: inpatient treatment.


Participants86 users of heroin, methadone or both; opioid dependent by DSM-IV, drug use confirmed by urine test. (1) 44, (2) 42; (1) 86%, (2) 74% male. 37/86 also used benzodiazepines. Mean duration opioid use 10.5 Y. Mean age 31.7 Y. Groups similar.
Excl. cr: major psychiatric or physical illness, pregnant or taking neuroleptic or antidepressant medication.


InterventionsStabilised on methadone (around 60 mg/day) for 3 days prior to detoxification, then: (1) Methadone, starting dose variable, tapered over 10 days. (2) Lofexidine,initial dose 0.6 mg/day until day 4, maintained at 2 mg/day for 3 days, then tapered over 3 days. Both drugs administered twice daily. Diazepam 3 days stabilisation then tapered over 21 days for those co dependent on benzodiazepines.
Scheduled duration of the study 20 days (10-day treatment program followed by 10 day-rehabilitation program). Country of origin: Europe (UK).


OutcomesCompletion rate as number completing 20 days treatment. Acceptability of the treatment as daily withdrawal score (graph) and as mean morning and evening daily blood pressure (graph) and number experiencing dizziness.


NotesSOWS (10 items, 0-3 severity) completed daily by participants.
Compliance corroborate by urine screening three times/week.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low risk"after the stabilisation period, ..patients randomly assigned to either methadone syrup and placebo tablets or placebo syrup and lofexidine tablet".

Blinding of outcome assessment (detection bias)
All outcomes
Low risk"a dedicated worker who did not have clinical contact with the patients had exclusive knowledge of urine drug screen.."

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Buydens-Branchey 2005

MethodsRandomised controlled trial. Setting: inpatient


Participants31 hospitalised heroin addicts (DSM IV), all males. 31 randomised, data presented on 29: 2 participants in the placebo group requested to discontinue. Age 48.3 years.12 Afroamerican, 10 Caucasian, 7 Hispanic. Mean age of starting regular heroin use: 24.6 years 15 iv; mean daily heroin use 0.62 g. 14 in the past had participated in MMT.
Incl.c: used heroin daily for at least the prior 6 months, using al least 2.5 g/week of heroin; physical dependence on opiates; urine samples positive for opiates; expressed willingness to participate in an RCT.
Excl. c: current or past psychiatric disorder; evidence of significant neurologic, gastrointestinal, hepatic, cardiovascular, renal, endocrine or haematologic disease; seropositive status for HIV.


Interventions(1) methadone, 8 participants; (2) placebo, 8 participants; (3) buspirone 30 mg, 8 participants; (4) buspirone 75 mg, 7 participants. Scheduled duration of the study 12 days. Country of origin: USA


OutcomesCompletion of treatment; Assessment of withdrawal symptoms (SOWS and OOWS)


Notes31 randomised, data presented on 29: 2 participants in the placebo group requested to discontinue.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod of allocation not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskstaff, participants blind to treatment

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to treatment

Incomplete outcome data (attrition bias)
All outcomes
High riskdata presented for the 29/31 participants who remained in the study

Camí 1985

MethodsControlled clinical trial. Setting: inpatient treatment, no phone calls or visitors. Detoxification preceded admission to drug-free therapeutic community.


Participants45 users of heroin, dependent by DSM-III-R. Of 30 who completed study, 24 male. (1) 15, (2) 15. Mean age 23.5 Y. Mean heroin use 4.2 Y. Mean previous supervised withdrawal attempts 1.8.


Interventions(1) Methadone 30-45 mg/day. Initial dose based on patient's weight and heroin consumed in last month.(2) Clonidine 0.9-1.35 mg/day, Both drugs given every 8 hours and tapered over 10 days. Flunitrazepam and acetylsalicylic acid as adjunct medications. Psychoterapeutic support for all. Naloxone challenges (0.4 mg ) on day of discharge. Scheduled duration of the study 8-10 days. Country of origin: Europe (Spain)


OutcomesAnalysis based on 30/45 who completed 12 days of treatment. Acceptability of the treatment as percentage of mean positive symptoms and as mean adverse effects and mean changes in heart rates 2/daily.


NotesWithdrawal rated daily by nurses (19 withdrawal signs, 17 adverse effects rated present/absent). Patients completed State-Trait Anxiety Inventory Questionnaire on days 1, 2, 3, 4, 7 & 10. Nurses measured blood pressure, heart rate and axillary temperature daily at 9 AM and 5 PM. Participants monitored by random urine screening.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskstated as double blind

Blinding of outcome assessment (detection bias)
All outcomes
Low riskstated as blind

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Dawe 1995

MethodsRandomised controlled trial. Setting: inpatient treatment.


Participants16 users of heroin, (1) 7 (2) 9. Mean age 29 Y., mean use of heroin 8 Y. Groups stated as similar.


Interventions(1) Methadone tapered by linear reduction, mean starting dose 57.2 (range 35-85) mg/day. (2) Clonidine oral, maximum 0.12 mg/day. (1) detoxification ward (2) behavioural psychotherapy ward. Scheduled duration of the study 13 days. Country of origin: Europe (UK)


OutcomesAcceptability of the treatment as min & max withdrawal scores; mean of withdrawal symptoms, negative and positive craving at time of maximum and minimum withdrawal. Drop-outs rates not reported.


NotesSymptom Checklist used to assess physical symptoms of opiate withdrawal. Craving Questionnaire.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskclinical staff and patients blind to treatment, blind maintained with placebo syrup and tablets

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot reported

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskresults on all randomised participants

Drummond 1989

MethodsRandomised controlled trial. Randomisation: Blindness: Placebos used to maintain blind. Setting: inpatient treatment, 3 hospitals involved.


Participants33 heroin users selected, 9 excluded, 24 treated; 85% injectors, mean dose 0.8 +/- 0.6 g/day. 54.2% male, most used cannabis regularly or occasionally, 3/24 used benzodiazepines regularly. Mean age 24.9 Y., mean duration of drug use 4.7 Y.; Excl. cr.= serious physical illness. Groups similar.


Interventions(1) Methadone, initial dose 20 mg/day plus more if needed. (2) Chlordiazepoxide, initial dose 200 mg/day plus more if needed. Patients chose rate of dose reduction, discharge 2 days after final dose. Scheduled duration of the study 14 days. Country of origin: Europe (UK).


OutcomesCompletion rate as rate of drop-outs and length of treatment. Acceptability of the treatment as mean total subjective and objective withdrawal scores and as mean heart rate, mean pupil size, mean temperature (all graph). Results at follow-up as urine screening, craving and mood measures, naltrexone compliance and relapse rate for 6 months.


NotesPatients rated expected withdrawal at entry, 16-item SMQ daily. In addiction Objective Opiate Withdrawal scale. Nurses recorded physiological measures & 10 items scale daily. Rating instruments completed by subjects & nurses. Random urine testing. Study across 3 hospitals. Rating reliability confirmed, training to ensure consistent application.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Low riskrandom allocation by pharmacist

Blinding of participants and personnel (performance bias)
All outcomes
Low riskstaff and patients blind to medication and urine screening results

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Gerra 2000

MethodsRandomised controlled trial. Setting: inpatient.


Participants98 heroin users by DSM-IV, 71 males, aged 18 to 36 years, use of heroin from 2 to 6 years. (1) 34 (2) 32 (3) 32, (1) 24 (2) 23 (3) 24 male.
Excl. c.: double dependence or prolonged use of drugs other than heroin, chronic physical disorders, psychosis, recent weight loss or obesity, endocrine-neopathies and immuno deficiencies.


InterventionsIntravenous heroin was administered to all participants until 12 hours before treatment.
(1) Methadone oral tapered from 40 mg to zero in 10 days.
(2) Clonidine iv 0.150 mg in 100 mL saline/three/morning and three/afternoon for 2 days, in the following 3 days, 0.150 mg three times/day. At 11 PM clonidine 0.150 mg orally every evening/5 days.
(3) Clonidine at the same doses and with the same procedures of (2) for 2 days, and oral 0.150 mg/3 on the third day; oxazepam 60 mg/day, oral baclofen 10 mg/3/day and ketoprofen 400 mg daily. During the first day of treatment naloxone injections until the full dose of 0.4 mg was attained and 5 mg orally of naltrexone syrup.
In the day 2, 50 mg of oral naltrexone. In (2) & (3) blood pressure was measured every 2 hours during detoxification procedure.
Scheduled duration of the study 10 days. Country of origin: Europe (Italy).


OutcomesAcceptability of the treatment as mean scores of withdrawal symptoms daily and negative and positive craving scores. Use of primary substance as percentage of positive urine controls. (All graph). Results at follow-up as rate of patients who accepted and continued naltrexone treatment (graph) and percentage of patients who relapsed in heroin dependence.
All participants were admitted to extended naltrexone treatment after detoxification.


NotesUrinary tests performed daily during detox. period. Withdrawal symptoms evaluated by the same observer daily (9 signs, severity 0-5). Craving Scale rated from patients at the beginning and following the treatment.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
High riskno information, apparently no blindness

Blinding of outcome assessment (detection bias)
All outcomes
High riskno information, apparently no blindness

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Howells 2002

MethodsRandomised controlled trial. Setting: inpatient treatment.


Participants76 heroin-dependent by DSM IV, were eligible, 68 treated. (1) 36 (2) 32. Mean age (1) 30.5 years, (2) 29.8; time from first use of heroin (1) 9.5 (2) 8.8. Past month use of other substances for all participants: benzodiazepine 67.6%, amphetamine, 5%, non prescribed methadone 5%, cocaine 1%, crack 2%. Excl c.: serious major psychiatric illness, serious physical illness.


Interventions(1) Methadone 30 mg/day 1, 25 mg/day days 2 and 3, 20 mg/day days 4 and 5, then tapered to 0 in 10 days. (2) Lofexidine 0.6 mg day 1, increased of 0.4 mg/day until day 4, 2 mg/day for three days, next 3 days dose tapered by 0.4 mg/day. Scheduled duration of the study 10 days. Country of origin: Europe (UK).


OutcomesCompletion rate as failure to complete detoxification. Acceptability of the treatment as withdrawal symptom severity, rates and timing of withdrawal.
Other: severity of psychological aspects of drug dependence.


NotesWithdrawal Problems Scale, Short Opiate Withdrawal Scale both self-rated daily. Severity of Dependency Scales. Hypotension, lying and sitting blood pressure, heart rate measured twice daily.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Low riskrandom allocation by pharmacist

Blinding of participants and personnel (performance bias)
All outcomes
Low riskstaff and patients blind to medication

Blinding of outcome assessment (detection bias)
All outcomes
Low riskstated as blind

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskresults on all randomised participants

Jiang 1993

MethodsRandomised controlled trial. Not all participants had entered treatment voluntarily. Setting: inpatient treatment in 5 different rehabilitation centres.


Participants200 heroin users, dependent by DSM-III-R. 100 allocated to each group. (1) 73% (2) 82% male. (1) 80 (2) 67 using orally only, others iv or iv and oral.
Men age (1) 24.9 (2) 24.7. No previous treatments (1) 79% (2) 63%. Duration of addiction (1) 16.1 (2) 15.2 months. At admission time since last drug intake (1) 8.7 hours(2) 10.7. No demographics differences. Excl. cr.: concurrent medical condition, infectious diseases, mental illnesses.


Interventions(1) Methadone, max days 1-2 then tapered and ceased after day 12; mean max dose day 2 = 21.6 mg.
(2) Clonidine, "sufficient" dose days 1-4, tapered days 5-8, ceased after day 11; mean max dose day 2 = 1.05 mg.
For both drugs initial dose dependent on body weight, physical condition, heroin intake previous week. Dose titrated against withdrawal and side effects. Scheduled duration of the study: 12 days. Country of origin: China


OutcomesAcceptability of the treatment as mean daily withdrawal score and as total scores of undesirable side effects. Other: score variation in Hamilton Anxiety test. No drop outs reported. Endpoint of naloxone challenge used for only half of participants.


NotesReport in Chinese, English translation obtained. Symptoms and vital signs assessed daily using Himmelsbach scale as guide; 21 designated symptoms and vital signs also assessed.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskno information on blinding

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskno information on blinding

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskresults on all randomised participants

Kleber 1985

MethodsRandomised controlled trial. Setting: outpatient treatment; component of multi-centre study.


Participants49 opiate addicts >- 6 mo. receiving methadone 20 mg/day or less, (1) 25 (2) 24, 76% males, mean age 29.5 Y., mean length of addiction 10.0 Y. Groups similar.


Interventions(1) Methadone, initial dose 20 mg/day, single daily oral dose tapered by 1 mg/day.
(2) Clonidine oral, initial dose 0.3 mg/day in 3 divided doses; by day 6, gradual increase to max 1 mg/day; from day 11, tapered by 20-25% per day. Chloral hydrate as adjunct medication. Scheduled duration of the study 30 days. Country of origin: USA.


OutcomesCompletion rate as number of drop-outs and percentages of success rates. Acceptability of the treatment as mean withdrawal scores at baseline and weeks 1-2-3-4; rates of withdrawal symptoms (graph), comparison of withdrawal characteristics of success and failure in the two groups, incidence and characteristics of side effects and number using sleep medication. Results at follow-up at 1, 3, 6 months as naloxone challenge rates. Other: Scores of Beck Depression inventory and of ASI.


NotesWithdrawal rated by nurses (24 items, 0-3 severity) and participants (31 items, 1-4 severity). Side effects rated by physicians and nurses. Successful detoxification, the main outcome, was defined as a) having 10 days following the last dose of study methadone in which no illicit opiate use is reported, or b) passing a naloxone test.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to treatment, blind maintained with placebo

Blinding of outcome assessment (detection bias)
All outcomes
Low riskstated as blind

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskITT analysis

Madlung-Kratzer 2009

MethodsRandomised controlled trial. Setting : inpatient at 3 psychiatric hospitals.


Participants202 patients male and female opioid dependents (age > 18 years) willing to undergo detoxification from maintenance therapy in order to reach abstinence.who (confirmed diagnosis of opioid addiction according to ICD-10 criteria) Incl.c: alcohol consumption of < 100 g/day during the last 4 weeks; reliable contraceptive methods (hormonal,non-hormonal) for female patients of childbearing potential. Occasional (but not daily) consumption of cocaine was acceptable. Exc criteria: Patients were excluded from the study if they had clinically significant somatic illness (except hepatitis), acute psychotic illnesses (i.e. known schizophrenia or major depression with suicidal intent) or known contraindications to morphine or methadone. Patients were also excluded if they had received maintenance treatment with other opioids (e.g. buprenorphine, codeine derivatives) or were unwilling to follow investigator instructions.


Interventions(1) SROM: N= 102; (2) methadone: N = 100 both tapered. Scheduled duration of the study: 16 days. Country of origin: Austria.


OutcomesCompletion rate, changes in signs and symptoms of opioid withdrawal [12-item German version of the Short Opioid Withdrawal Scale (SOWS)] [20] assessed
on days 0, 3, 7, 10, 14, 18 and 22 by patient self-rating; somatic and psychological symptoms [Symptom Checklist (SCL-90-R)] [21] assessed on days 0, 7, 14 and 22,from which global symptom scale scores were calculated; craving for heroin, alcohol, benzodiazepines, cocaine and cannabis (rated by patients on a visual analogue scale: 0 mm = no craving, 100 mm = most intense craving) assessed on days 0, 3, 7, 10, 14, 18 and 22; adverse events.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskcentral stratified randomisation

Allocation concealment (selection bias)Low riskrandom allocation by pharmacist

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to treatments

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskITT analysis

Salehi 2007

MethodsRandomised controlled trial. Setting: outpatient


Participants70 participants, all males, mean age 37 years; n. 60 married: 60; n. 33 elementary education, n. 26 high school, n.11 university degree. Duration of dependence mean 12.8 years. Groups similar. Exc cr: presence of any medical disease that prohibited using tramadol and methadone, taking extra medication, polysubstance dependence, presence of any major psychiatric disorder.


Interventions(1) methadone, starting dose 15 mg/day, 36 participants, ; (2) tramadol, starting dose 450 mg/day, 34 participants. Both groups were treated with 0.3 mg/day of clonidine and 10 to 30 mg/day oxazepam. Scheduled duration of the study 7 days. Country of origin: Iran


OutcomesCompletion of treatment, Withdrawal symptoms (SOWS)


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to medication

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to medication

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

San 1990

MethodsRandomised controlled trial. Randomisation: method of allocation not reported. Blindness: Setting: inpatient treatment.


Participants170 heroin dependent by DSM-III-R entered trial, analysis based on 90/170 who completed >-12 days of treatment. (1) 30, (2) 30, (3) 30. In each group around 80% male, mean age around 24 Y. and mean duration of opiate use around 5 Y. No differences between groups
Excl. c.: psychopathological antecedents before opioid addiction, signs cardiovascular diseases, previous participation in clinical trial. Country of origin: Europe (Spain)


InterventionsInitial dose of medication dependent on weight and heroin use in previous week. (1) Methadone, mean max dose 37.3 mg/day. (2) Clonidine, mean max dose 1.05 mg/day, (3) Guanfacine, mean max dose 3.58 mg/day. For all max dose given on days 2 & 3. Drugs tapered over 11 days. Benzodiazepines as adjunct medication as needed. Scheduled duration of the study 14 days.


OutcomesCompletion rate as mean duration in treatment, number completing detoxification, causes of failure. comparison of success and failure. Acceptability of the treatment as time course of withdrawal scores (graph), min & max withdrawal scores time, course of cardiovascular effects, mydriasis and other side effects (all graph). Other: Scores of EPQ and MMPI. Some data confounded by exclusion of early drop-outs.


NotesWithdrawal and side effects rated by observers. Participants completed psychometric evaluation (MMPI, State Trait Anxiety Inventory and Eysenck Personality Questionnaire)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to medication, placebo used to maintain blind.

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to medication

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

San 1992

MethodsRandomised controlled trial. Blindness: double blind. Setting: inpatient treatment.


Participants22 opioid dependent by DSM-III-R, using buprenorphine mean dose (1) 2.0 (2) 1.7 mg/day; route of assumption (1) 81.8% (2) 100% iv (1) 11 (2) 11 patients, 17 male, mean age (1) 28.0 (2) 29.7 Y. No differences in groups.
Excl. cr.: patients with methadone or heroin use detected by urine testing in 2 weeks prior to admission.


Interventions(1) Methadone max dose 20 mg/day, tapered over 9-11 days. (2) Placebo.
Scheduled duration of study 13 days Country of origin: Europe (Spain).


OutcomesCompletion rate as number who completed and numbers of patients who shifted from (2) to (1). Acceptability of the treatment as individual mean daily withdrawal scores in placebo group (graph).
Study confounded by 8/11 placebo-treated group being switched to methadone.


NotesOpiate Withdrawal Checklist (21 items, 0-3 severity) administered by nursing staff.
Data provided for placebo-treated patients only.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Low riskallocation by pharmacy prior to admission

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to medication

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to medication

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

San 1994

MethodsRandomised controlled trial. Setting: inpatient treatment. Trial in two phases.


Participants144 heroin dependent by DSM-III-R, mean heroin dose 656 mg/day; Group (3) introduced in phase 2. (1) 75 (2) 43 (3) 26, 102 male, mean age 27.1 Y., 52% HIV+. Stated no differences between groups. Excl. cr.: history of psychiatric disorders, liver dysfunction, cardiovascular diseases, other addiction, pregnancy.


InterventionsMethadone, 3 divided doses, initial dose based on body weight & heroin consumption, tapered over 8 days to (1) 10% (2&3) 50% of initial dose. From day 9: (1) Continued methadone taper, others switched to (2) 3 or (3)4 mg guanfacine. 59% given benzodiazepines 32% hypnotics as adjunct medication. Scheduled duration of the study 18 days. Country of origin: Europo (Spain)


OutcomesCompletion rate as percentage of participants remaining in the study (graph). Acceptability of the treatment as mean daily withdrawal scores (graph) and differences in blood pressure and heart rate. Other: mean dose diazepam, personality tests, patients' mood between groups.


NotesOpiate withdrawal symptoms were measured by means of the Opiate Withdrawal Checklist which was completed by nursing staff three times a day and the Opiate Withdrawal Syndrome which was self- completed by patients once a day. Nursing staff monitored heart rate and blood pressure daily. Urine screening days 1, 4, 7, 14, 17. Psychometric tests were performed in all participants.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Low riskallocation by pharmacy

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskdouble blind, treating doctor blind

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot stated if observer blind

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Seifert 2002

MethodsRandomised controlled trial. Randomisation: Blindness: Setting: inpatient treatment.


Participants26 opioid dependent (DSM IV criteria) and abused various additional drugs. (1) 12 (2) 14; Mean age (1) 31.8 (2) 31.1; Male (1) 9 (2) 13; Years of opioid abuse (1) 10.5 (2) 8.6. Excl cr: people who had participated in a structured drug research program within the previous 6 months or had active schizophrenia, active bipolar affective disorder, active hepatic disease, active cardiovascular disease, abnormal ECG, chronic obstructive pulmonary disease, pregnancy.


Interventions(1) Methadone tapered, starting dose 20 mg/day, last dose 2.5 mg/day. (2) Buprenorphine tapered starting dose 4 mg/day, last dose 0.4 mg/day. For both groups carbamazepine: days 1-6: 900 mg/day; days 7-10: 400 mg/day; days 11-14: 200 mg/day. Scheduled duration of the study 14 days. Country of origin: Europe (Germany).


OutcomesCompletion rate as number of non completer. Acceptability of the treatment as mean scores of SOWS and as scores of a visual analogue scale.


NotesSOWS self-rate and examiner rating using a visual analogue scale. Urine sample randomly once a week.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskdouble blind, placebo used to maintain blind

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot stated if observer blind

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Sorensen 1982

MethodsRandomised controlled trial. Setting: outpatient treatment.


Participants61 heroin dependent > 80/day. All male. 53% white, 36% Hispanic, 11% other. Mean age 28.9; 33% employed; 28% married; 57% arrested in last 2 Y.; 90% had prior treatment. (1A) 18 (1B) 15 (2A) 15 (2B) 13 . Groups similar on all except arrests in last 2 years. Excl. c.: life-threatening medical conditions.


Interventions(1A) Methadone 30 mg increasing to 40 mg, then tapered in 6 weeks, (2A) LAAM, doses to parallel methadone.,(1B) & (2B) same but active treatment only 3 weeks. For all, 1 week stabilisation. Scheduled duration of the study 21 days. Country of origin: USA


OutcomesCompletion rate as percentage of retention in treatment. Acceptability of the treatment as mean symptom discomfort index (graph). Use of primary substance of abuse as percentage of patients with urine samples positive for opiate (graph). Initial, stabilization and final ratings. Results at follow-up as number of abstinent > 1 day after detoxification and at follow-up at 3 months: number of abstinent, sought further treatment, enrolled in MMT.


NotesProfile of Mood State (POMS) completed wk prior treatment & day 14. Detoxification Symptom Scale (20 items) administered by researchers daily, reported as discomfort index combining frequency and severity. Two urine screens per week (random).


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskStratified by employment status

Allocation concealment (selection bias)Low riskrandom allocation by pharmacist

Blinding of participants and personnel (performance bias)
All outcomes
Low riskdouble blind, doses prepared by pharmacist; placebo used to maintain blind, staff and participants blind

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to medication

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskresults on 86% of participants available 12 weeks after intake

Steinmann 2007

MethodsRandomised controlled trial. Setting: inpatient


Participants39 opioid dependent, 31 males, mean age 27 years. EXC cr: previous detoxification treatments


Interventions(1) methadone 21 participants, starting dose 60 mg/day, tapered of 2.5-5 mg/day; (2) buprenorphine, 18 participants, starting dose 12-16 mg/day, tapered of 0.8-1.2 mg/day. Scheduled duration of the study 28 days. Country of origin: Europe (Germany)


OutcomesCompletion of treatment, withdrawal symptoms (OOWS), craving.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Unclear risknot reported

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Tennant 1975

MethodsRandomised controlled trial. Setting: outpatient treatment, daily clinic attendance for supervised dosing.


Participants72 heroin addict > 18 Y.; dependent by history, needle marks, positive urine test, observation of withdrawal symptoms.
(1) 36 (2) 36; (1) 80.6% (2) 77.7% male; (1) 50% (2) 56% white; mean age (1) 27.1 (2) 28.5 ; mean duration heroin use years (1) 7.8 (2) 9.1; mean current daily heroin use months (1) 8.8 (2) 7.0; (1) 5.0% (2) 3.2% urine positive for amphetamines or barbiturates during treatment . No differences between groups.


Interventions(1) Methadone, initial dose 24 mg daily tapered. (2) Propoxyphene napsylate, initial dose 800 mg daily, tapered. Scheduled duration of the study 21 days. Country of origin: USA.


OutcomesCompletion rate as number not completed treatment and mean days in treatment. Acceptability of the treatment as mean daily withdrawal scores (graph). Use of primary substance of abuse as percentage of patients with urine samples positive/negative during treatment.
Results at follow-up as at 1 month: patients abstinent, relapsed, entered in MMT.


NotesIntensity of withdrawal assessed daily using Himmelsbach scale (0-2+severity). 16 side effects assessed each day by same scoring system. Observed urine 2x weeks.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskstaff and patients blind; medication dispensed in identical capsules and placebos used to conceal tapering

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Umbricht 2003

MethodsRandomised controlled trial. Randomisation: Blindness: Setting: inpatient treatment.


Participants63 heroin dependent: (1) 21, (2) 21, (3) 21,. Mean age: (1) 40.0, (2) 39.6, (3) 40.0, ; Afro-American: (1) 21, (2) 20, (3) 21; Male (1) 9, (2) 15, (3) 14. Incl c.: current heroin dependent, HIV seropositivity, 18 years or more, hospitalisation for an acute medical illness. Excl c.: concurrent alcohol dependence, inability to give informed consent, acute psychosis or AIDS dementia, hypotension, bradycardia, coagulopathy or severe thrombocytopenia precluding intramuscular injections, enrolment in methadone maintenance treatment.


Interventions(1) Methadone orally once a day, 30 mg on day 1, 20 mg on day 2, 10 mg on day 3. (2) Buprenorphine intramuscularly 0.6 mg every 4 h on day 1, every 6 h on day 2, every 8 h on day 3. (3) Clonidine orally, a loading dose of 0.2 mg followed by 0.1 mg every 4 h on day 1, every 6 h on day 2 and every 8 h on day 3. Scheduled duration of the study 3 days. Country of origin: USA.


OutcomesCompletion rate as number who completed the study and number of drop-outs who voluntarily left the study. Acceptability of the treatment as mean participant- and observer-rated opioid withdrawal scores and pupil size. Craving.


NotesSOWS twice a day rated by participants, nurses rated opioid withdrawal scale three times a day.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskdouble blind, placebo used to maintain blind

Blinding of outcome assessment (detection bias)
All outcomes
Unclear risknot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Washton 1981

MethodsRandomised controlled trial. Setting: outpatient treatment, 3-5 clinic visits/week.


Participants26 patients, 19/26 MMT (15-30 mg/day), 7/26 illicit heroin and/or methadone, stabilised for 3 weeks on 15-30 mg/day methadone.
22 male, 18 white, 5 black, 3 Hispanic, mean age 31 (range 22-49) Y, mean duration of addiction 10 Y (range 3 mo-25 Y.) (1) 13 (2) 13. Groups stated as similar.
Excl. c.: evidence of serious medical or psychiatric illness.


Interventions(1) Methadone 20 mg/day reduced by 1 mg/day. (2) Clonidine, dose titrated against symptoms and side effects to max 1.2 mg/day. Scheduled duration of the study 30 days. Country of origin: USA


OutcomesCompletion rate as number completing detoxification. Results at follow-up as number initiating naltrexone maintenance treatment.


NotesRatings of withdrawal not reported.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to medication

Blinding of outcome assessment (detection bias)
All outcomes
Low riskinvestigators not informed of blood pressure measurements to avoid breaking blind

Incomplete outcome data (attrition bias)
All outcomes
Low riskresults on all randomised participants

Wright 2011

MethodsRandomised controlled trial. Setting: prison.


Participants289 prisoners, all male. Incl.c: 21–65 years old; using illicit opiates as confirmed by urine test; expressing a wish to detoxify and remain abstinent; willing to give informed consent; and remaining in custody for at least 28 days. Excl. c: contraindications to methadone or buprenorphine; medical conditions requiring emergency admission to hospital, thus precluding detoxification; currently undergoing detoxification from other addictive drugs whereby concurrent opiate detoxification would not be clinically indicated; and previously randomised into the trial.


Interventions(1) Methadone, n = 148, starting dose 30 mg (2) Buprenorphine, n = 141, starting dose 8 mg. Scheduled duration of the study: 20 days. Country of origin: UK.


OutcomesPost detoxification abstinence across time; completion rate.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskrandomisation sequence (with random block size) was generated using Microsoft Excel RAND function

Allocation concealment (selection bias)Low risksealed, opaque, consecutively numbered envelopes concealing the name of the allocated intervention were prepared by a researcher who had no contact with participants

Blinding of participants and personnel (performance bias)
All outcomes
Low riskparticipants and staff blind to medication

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to medication

Incomplete outcome data (attrition bias)
All outcomes
Low riskITT analysis

Yang 2006

MethodsRandomised controlled trial. Setting: inpatient.


Participants580 opioid dependents; (1) 278, (2) 302. Mean age: 23 years; Male 381.


Interventions(1) Methadone, starting dose 40-50 mg/day then tapered 20% per day. (2) Paidu capsules, starting dose 3-5 twice a day, then tapered. Scheduled duration of the study 10 days. Country of origin: China.


OutcomesWithdrawal symptoms (OOWS); Anxiety (Hamilton Anxiety Rating Scale).


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Low riskstaff and participants blind to treatment

Blinding of outcome assessment (detection bias)
All outcomes
Low riskoutcome assessors blind to treatment

Incomplete outcome data (attrition bias)
All outcomes
Unclear risknot clear

Zarghami 2012

MethodsRandomised controlled trial. Setting: inpatient.


Participants70 patients, all men; age range, 18–46 years, with a confirmed diagnosis of opioid dependence according to DSM-IV-TR criteria. Excl. c: clinically significant somatic illness (e.g., hepatitis, tuberculosis, acquired immune deficiency syndrome), a history of seizures, acute psychotic illnesses (e.g., known schizophrenia or major depression with suicidal intent), and using other substances except nicotine (e.g., other opioids, monoamine oxidase (MAO) inhibitors, doxepin, anti-spastic drugs, beta blockers, known inducers or inhibitors of CYP3A and CYP2D6, cannabinoids, and alcohol). No significant differences were found in baseline demographics and drug use history between the two patient cohorts.


Interventions(1) methadone, N = 35; (2) tramadol N = 35.

The dose reduction regimens were based on an oral dose of either 600 mg/day of tramadol (200 mg three times daily) or 60 mg/day of methadone (20 mg three times daily). These starting doses were maintained for three consecutive days under double-blind conditions. Thereafter, detoxifications were initiated by tapered dose reductions (20% every 2 days) over a period of 11 days to reach abstinence. At the end of second week, the medications were discontinued.


OutcomesWithdrawal scores; side effects


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskmethod not reported

Allocation concealment (selection bias)Unclear riskmethod of allocation not reported

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskmethod not reported

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskmethod not reported

Incomplete outcome data (attrition bias)
All outcomes
High risknine patients in the methadone group and five patients in the tramadol group were excluded from the study

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Albizu-Garcia 2012Excludes as study design was not in the inclusion criteria: survey report

Bakhshani 2008Excluded as type of intervention was not in the inclusion criteria: evaluation of the efficacy of transcutaneous electrical stimulation added or not to methadone

Bearn 1998Excluded as the study design was not in the inclusion criteria of the review: open design with a patient preference allocation

Bearn 2008Excluded as type of intervention was not in the inclusion criteria: auricular acupuncture as an adjunct to opiate detoxification treatment

Bell 2009Excluded as type of intervention was not in the inclusion criteria: investigate the pharmacokinetics and pharmacodynamics of orally administered methadone-naloxone

Bickel 1988Excluded as type of intervention was not in the inclusion criteria: length of treatment 90 days

Brewin 1989Excluded as the study design was not in the inclusion criteria: review article

Bux 1993Excluded as the study design was not in the inclusion criteria: no randomised controlled trial

Byrne 2006Excluded as the study design was not in the inclusion criteria: letter

Cameron 2006Excluded as type of intervention was not in the inclusion criteria: length of treatment 12 weeks

Critchlow 2006Excluded as the study design was not in the inclusion criteria: letter

Dawe 1991Excluded as type of intervention was not in the inclusion criteria: length of treatment 70 days

De Los Cobos 2000Excluded as the study design was not in the inclusion criteria: no randomised controlled trial

Deniker 1975Excluded as the study design was not in the inclusion criteria: no randomised controlled trial

Dijkstra 2010Excluded as type of intervention was not in the inclusion criteria: rapid detoxification with naltrexone

Ebner 2004Excluded as the study design was not in the inclusion criteria: no randomised controlled trial

Fulwiler 1979Excluded as the type of intervention was not in the inclusion criteria: two different modalities of tapering methadone (1) physician regulated, (2) self-regulated

Gerra 2004Excluded as the type of intervention was not in the inclusion criteria: methadone and buprenorphine both as maintenance treatments

Gerra 2007Excluded as type of intervention (maintenance) and type of participants (include also healthy participants ) were not in the inclusion criteria

Glasper 2008Excluded as the type of intervention was not in the inclusion criteria: both groups received methadone at different dosages to investigate influence of methadone doses on the Severity of Opiate Withdrawal Symptoms

Goldstein 1972Excluded as the study design was not in the inclusion criteria: theoretical and descriptive study

Gossop 1989AExcluded as the study design was not in the inclusion criteria: retrospective analysis

Green 1988Excluded as the type of intervention was not in the inclusion criteria: methadone tapered in both groups plus (1) 15-30 min interview with detailed information about the withdrawal regimen likely length and intensity of symptoms. (2) regular information about their admission and ward routine

Greenwald 2006Excluded as type of participants not in the inclusion criteria: volunteers no opioid dependent

Gruber 2008Excluded as the type of intervention was not in the inclusion criteria: methadone maintenance with standard or minimal counselling versus 21-day methadone
detoxification

Hall 1979Excluded as the type of intervention not in the inclusion criteria: methadone tapered in both groups plus (1) paid for drug-free urine 6 times during treatment and brief counselling weekly. (2) Paid $1 for each urine sample given

Hall 2008Excluded as the study design was not in the inclusion criteria: letter

Hasson 2007Excluded as type of intervention was not in the inclusion criteria: length of treatment 24 weeks

Highfield 2007Excluded as type of intervention was not in the inclusion criteria: length of treatment 120 days

Hser 2012Excluded as type of intervention was not in the inclusion criteria: MMT

Jaffe 1972Excluded as type of intervention was not in the inclusion criteria: length of treatment 15 weeks

JI 2007Excluded as type of intervention was not in the inclusion criteria: tapered methadone in both groups

Johnson 1992Excluded as type of intervention was not in the inclusion criteria: length of treatment 17 weeks

Kheirabadi 2008Excluded as type of intervention was not in the inclusion criteria: efficacy of gabapentin added to methadone

Krabbe 2003Excluded as the study design not in the inclusion criteria of the review: prospective clinical trial

Kristensen 2005Excluded as type of intervention was not in the inclusion criteria: length of treatment 26 weeks

Lal 1976Excluded as the type of intervention not in the inclusion criteria: two different modalities of tapering methadone (1) methadone tapered over 10 days (2) abrupt cessation day 11

Liu 2009Excludes as study design and type of intervention not in the inclusion criteria: review of randomised trials comparing acupuncture combined with opioid agonist treatment versus opioid agonists alone for treating symptoms of opioid withdrawal

Liu 2009aExcludes as study design and type of intervention not in the inclusion criteria: review of randomised trials comparing Chinese herbal medicine to either alpha2-adrenergic agonists or opioid agonists for heroin detoxification

Lobmaier 2010Excludes as study design not in the inclusion criteria: article that reviews the main pharmacotherapies that are currently being used to treat opioid addiction

Madden 1986Excluded as the type of intervention not in the inclusion criteria of the review: methadone tapered in both groups plus (1) "standard detoxification" (2) "cordial substitution"

Maddux 1980Excluded as the study design was not in the inclusion criteria: longitudinal study

Mannelli 2008Excluded as type of intervention not in the inclusion criteria: very low dose naltrexone addition in opioid detoxification

McCambridge 2006Excluded as type of intervention was not in the inclusion criteria: random allocation only for groups without methadone

McCaul 1984Excluded as the study design, the type of intervention not in the inclusion criteria: no RCT, three different modalities of tapering methadone, 6 weeks of treatment

Meader 2010Excluded as the study design was not in the inclusion criteria of the review: systematic review

Mintz 1975Excluded as the type of intervention not in the inclusion criteria: methadone maintenance patients were assigned to (1) decreasing dose or (2) continued methadone maintenance

Mitchell 2012Excluded as the study design was not in the inclusion criteria: the study compares the characteristics of patients entering methadone treatment vs. buprenorphine treatment to determine whether BT was attracting different types of patients

Mokhber 2008Excluded as the type of intervention not in the inclusion criteria: efficacy of totipalmate as an adjunct medication in heroin withdrawal

Neale 2005Excluded as study design not in the inclusion criteria: cross sectional data from a longitudinal study

O'Connor 1997Excluded as the type of intervention was not in the inclusion criteria: no methadone in the three detoxification protocols (clonidine, combined clonidine and naltrexone, buprenorphine)

Pjrek 2012Excluded as study design not in the inclusion criteria: naturalistic study

Rawson 1983Excluded as the type of intervention was not in the inclusion criteria: methadone tapered in both groups plus (1) with counselling sessions (2) without counselling

Reed 2007Excluded as study design not in the inclusion criteria: not RCT, allocation to detoxification condition was by patient choice

Reilly 1995Excluded as the study design was not in the inclusion criteria: not RCT

Sees 2000Excluded as the type of intervention was not in the inclusion criteria: (1) methadone maintenance treatment, (2) methadone tapered; outcomes at six months

Semba 2007Excluded as the study design was not in the inclusion criteria: not RCT

Shaygani 2009Excluded as the study design was not in the inclusion criteria: not RCT

Sheard 2006Excluded as the type of intervention not in the inclusion criteria: no methadone

Soyka 2009Excluded as the study design was not in the inclusion criteria: not RCT, open study

Stimmel 1982Excluded as the study design was not in the inclusion criteria: not RCT

Stotts 2012Excluded as type of intervention not in the inclusion criteria: study developed and tested an ACT-based opioid detoxification behavioral therapy study developed and tested an ACT-based opioid detoxification behavioural therapy

Strain 1993Excluded as type of intervention not in the inclusion criteria: the length of treatment (15 weeks) was too long

Strang 1990Excluded as the type of intervention was not in the inclusion criteria: two different modalities of tapered methadone on (1) linear (2) inverse exponential curve

Strang 1997Excluded as the type of intervention and the outcomes measures were not in the inclusion criteria: groups differed in duration of detoxification, intensity & duration of adjunct & follow-up care, no rating instruments used, no urinalysis reported, endpoint is vague

Sullivan 2004Excluded as study design not in the inclusion criteria: cross- sectional survey

Teesson 2006Excluded as the study design was not in the inclusion criteria: not RCT

Tennant 1978Excluded as type of intervention was not in the inclusion criteria: length of treatment 42 days

Van Beek-Verbeek 1983Excluded as the type of intervention was not in the inclusion criteria: methadone tapered in both groups plus (1) placebo (2) desglycinamide9-arginine 8-vasopressin

Veilleux 2010Excluded as the study design was not in the inclusion criteria: not RCT, review of opioid dependence treatment

Wang 1982Excluded as the type of intervention was not in the inclusion criteria: methadone not tapered

Yang 2008Excluded as the type of intervention was not in the inclusion criteria: the study examines the effects of levotetrahydropalmatine (l-THP) on reducing heroin craving and increasing the abstinence rate among heroin-dependent patients

Zeng 2005Excluded as type of intervention was not in the inclusion criteria: tapered methadone in both groups

Ziaadini 2011Excluded as the study design was not in the inclusion criteria: not RCT, cohort prospective study

 
Comparison 1. Tapered methadone versus any other treatment

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Completion of treatment161381Risk Ratio (M-H, Random, 95% CI)1.08 [0.97, 1.21]

 2 Number of participants abstinent at follow-up3386Risk Ratio (M-H, Random, 95% CI)0.98 [0.70, 1.37]

 
Comparison 2. Tapered methadone versus adrenergic agonists

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Completion of treatment7577Risk Ratio (M-H, Random, 95% CI)1.10 [0.91, 1.32]

 
Comparison 3. Tapered methadone versus other opioid agonists

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Completion of treatment methadone versus any other opioid agonist7695Risk Ratio (M-H, Random, 95% CI)1.10 [0.89, 1.37]

 2 Completion of treatment methadone versus buprenorphine4390Risk Ratio (M-H, Random, 95% CI)0.97 [0.69, 1.37]

 
Comparison 4. Tapered methadone versus anxiolytic

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Completion of treatment247Risk Ratio (M-H, Random, 95% CI)0.91 [0.47, 1.77]

 
Comparison 5. Tapered methadone versus placebo

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Completion of treatment238Risk Ratio (M-H, Fixed, 95% CI)1.95 [1.21, 3.13]

 
Summary of findings for the main comparison. Tapered methadone versus any other treatment for the management of opioid withdrawal

Tapered methadone versus any other treatment for the management of opioid withdrawal

Patient or population: patients with the management of opioid withdrawal
Settings: Inpatient and outpatient
Intervention: Tapered methadone versus any other treatment

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

ControlTapered methadone versus any other treatment

Completion of treatment
Objective
Follow-up: mean 30 days
Study populationRR 1.08
(0.97 to 1.21)
1381
(16 studies)
⊕⊕⊕⊕
high

547 per 1000591 per 1000
(531 to 662)

Moderate

505 per 1000545 per 1000
(490 to 611)

Number of participants abstinent at follow-up
Objective
Follow-up: mean 1.5 months
Study populationRR 0.98
(0.7 to 1.37)
386
(3 studies)
⊕⊕⊕⊕
high

255 per 1000250 per 1000
(179 to 350)

Moderate

267 per 1000262 per 1000
(187 to 366)

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 
Table 1. Withdrawal scales

AuthorName of ScalePublishedn° itemsn° scores

Bearn 1996Short Opiate Withdrawal Scale (Gossop 1990)yes104

Buydens-Branchey 2005Subjective Opiate Withdrawal Scale (Handelsman 1987)yes165

Buydens-Branchey 2005Objective Opiate Withdrawal Scale (Handelsman 1987)yes133

Camí 1985Abstinence Rating Scaleno17present/absent

Dawe 1995Symptom Checklist (Powell 1990)yes104

Drummond 1989Subjective Measures Questionnaireno16not reported

Drummond 1989Objective Opiate Withdrawal Scale (Himmelsbach 1942)yes104

Gerra 2000List of Withdrawal Symptoms (Gerra 1995)yes95

Howells 2002Withdrawal Problem Scale (Gossop 1990)yes204

Howells 2002Short Opiate Withdrawal Scale (Gossop 1990)yes84

Jiang 1993Himmelsbach Drug Withdrawal Symptoms Assessment Chart (Himmelsbach 1941)yes143

Kleber 1985Himmelsbach Drug Withdrawal Symptoms Assessment Chart (Himmelsbach 1941)yes143

Kleber 1985Self Rated Withdrawal Scale (Haertzen 1968)yes324

Kleber 1985Observer rating scale (Kolb 1938)yes103

Madlung-Kratzer 2009Short OpioidWithdrawal Scale [German version] (Gossop 1990)yes124

Salehi 2007Short Opioid Withdrawal Scale (Gossop 1990)yes164

San 1990Daily Abstinence Rating Scaleno21present/absent

San 1990Abstinence Signsno11not reported

San 1990Abstinence Symptomsno10not reported

San 1992Opiate Withdrawal Checklist (Schubert 1984)yes213

San 1994Opiate Withdrawal Checklist (Schubert 1984)yes213

San 1994Opiate Withdrawal Syndrome (Bradley 1987)yesnot reported11

Seifert 2002Short Opiate Witdrawal Scale (Gossop 1990)yesnot reportednot reported

Sorensen 1982Detoxification Symptom Scale (Fulwiler 1979)yes20not reported

Steinmann 2007Clinical Opiate Withdrawal Scale (Wesson 2003)yes11not reported

Steinmann 2007Witdrawal Syndrome Scaleno244

Tennant 1975Himmelsbach Drug Withdrawal Symptoms Assessment Chart (Himmelsbach 1941)yes143

Umbricht 2003Short Opiate Withdrawal Scale (Gossop 1990)yes104

Umbricht 2003Observer Opioid Withdrawal Scale (Peachey 1988)yes11not reported

Yang 2006Opiate Withdrawal Scalenonot reported4

Zarghami 2012Objective Opioid Withdrawal Scale (Handelsman 1987)yes133

 
Table 2. Withdrawal symptoms and side effects methadone versus adrenergic agonists

StudyWithdrawal symptomsSide effects

Bearn 1996Mean scores higher for (1) on days 13-21 and for (2) days 2-12. Peak score on (1) day 13 and (2) day 10.(2) 2 both female, experienced dizziness due to postural hypotension.

Camí 1985Muscular aching, flatulence and drowsiness more common in (1). Sleep disturbance & weeping in (2).1 of (2) had transferred loss of consciousness. (2) 4 (1) 1 experienced orthostatic hypotension. A graph shows a steady decline and similar magnitude in both groups regarding adverse effects, from days 1 to 6. After day 6, the scores for (1) increased, while in (2) continued to decline.

Dawe 1995Minimum withdrawal scores, mean (1) 25 (2) 33; maximum (1) 69 (2) 53. Time points minimum (1) day 1 (2) day 7, maximum (1) day 10 (2) day 2/3.Not reported.

Gerra 2000In (2) mean scores slightly lower but not significantly lower than (3). During the last 4 days of treatment and after the first 4 days after methadone discontinuation; (1) mean scores significantly higher (2) than (3).Mean daily blood pressure only for (2) & (3): no significant differences at any point. (2) 3 (3) 2 experienced side effects necessitate dose reduction.

Howells 2002Lowest daily score: mean (1) 49.4 (2) 50.0 ; Highest daily score: mean (1) 67.6 (2) 69.3; Total mean (1) 572.1 (2) 596.1Sitting blood pressure in (2) 4/32 lowest 61 mmHg, in (1) 3/36 lowest 80 mmHg, NS Depressive symptoms in 2 patients, one in each treatment group.

Jiang 1993On day 1 no significant differences in the two groups, between days 2 and 4 higher in (2), between days 8 and 12 lower in (2).The scores higher in (2) compared to (1) from days 1 to 7, the greatest difference occurred on day 2 of treatment when the mean side effects scores was (1) 1.9 and (2) 8.2. The score subsequently declined in both groups. Comparison of each undesirable side effect suggests that general tiredness, weakness in walking, dizziness in standing position, dry mouth and lethargy are most common occurrences in (2), while in (1) only a small numbers of patients experienced general tiredness, weakness in walking, dry mouth and lethargy . In (2) 89 experienced dizziness on standing, mostly in the second and third days of treatment. Anxiety scores were significantly lower in (2) by day 11.

Kleber 1985Mean withdrawal scores at baseline and weeks 1-2-3-4: at baseline no differences, during the first 2 weeks (1) 7.6, (2) 19.0, during the second two weeks (1) 13,0 (2) 12,0 ; Rates of withdrawal symptoms: (2) higher than (1); Comparison of withdrawal characteristics of success and failure in the groups were not statistically different.Incidence of side effects, rated by physicians and nurses: (1) 11,5 (2) 16,8. Characteristics of side effects: for the majority of symptoms there were no differences among the two groups. Number using sleep medications: (1) 70% (2) 63%; Scores of Beck Depression Inventory between successfully drug-free or not patients in the 2 groups were respectively: 54% and 17%. For the same groups, the percentage above the median for each ASI area were as follows: for medical area 62% and 46%, for employment 50% and 22%, for legal 35% and 22% , for family/social 31% and 28%, for psychological 31% and 28%.

San 1990Mean daily withdrawal scores (graph): Significant differences between (1) and (2) on days 2, 3, 4 and 5, higher in (1); between (1) and (3) on days 2, 3, 4 and 11, higher in (3); no significant differences between (2) and (3); the maximum mean was (1) 4.9, (2) 8.1, (3) 7.6 all on day 2.Hypotensive effect more intense in (2); changes in heart rate more pronounced in (2) than in (3); (1) and (3) most frequent side effects: asthenia, dry mouth, flushing, mental clouding, thirst. . Differences in personality tests and patient's mood: NS

San 1994Mean daily withdrawal scores (graph): from day 9 higher in (2) and (3) for three days and in (1) after day 11, statistically significant between (1) and (3) on days 10,11 higher in (3); and between (1) and (2) on days 11,15, higher in (1); on day 20 higher in (3) compared with (1) and (2).Differences in blood pressure and heart rate: for blood pressure, in (3) after day 13 grater decrease than in (1) and (2); for heart rate, in (3) bradycardia from day 9 with lowest rate on day 11, (62 beats per min); statistical significant difference between (3) compared with (1) and (2) in days 9, 10,11, 12,16, 17, 18, 19. Differences in personality tests and patient's mood: NS.

Umbricht 2003Mean withdrawal scores at baseline, after the first dose of medication and during the treatment no significant differences in the magnitude of the decrease in withdrawal between the groups.2 patients in (3) had to withdraw from the study because of decreased systolic blood pressure (< 90 mmHg) and bradycardia (HR < 50 BPM)

Washton 1981Major symptomatic complaints, specifically lethargy, restlessness and insomnia were identical for both groups. The main items contributing to scores in both groups were sleep problems, anxiety/nervousness, irritability, lack of energy, aches/pains and feeling cold. (2) reported symptoms during the 1° week of the study whereas (1) in the last week.Lethargy and sluggishness most consistent complaints in (2).

 
Table 3. Withdrawal symptoms and side effects methadone versus other opioid agonists

StudyWithdrawal symptomsSide effects

Madlung-Kratzer 2009Changes in signs and symptoms of opioid withdrawal [12-item German version of the Short Opioid Withdrawal Scale] assessed on days 0, 3, 7, 10, 14, 18 and 22 by patient self-rating.

At study entry signs and symptoms of withdrawal were mild but deteriorated steadily over time (day 0 versus day 22, P < 0.001). The only difference between the groups was found on day 18 (P = 0.022). All symptoms showed a homogeneous pattern of changes. (1) day 0 8.15 6.48 (7.00), day 22 16.00 7.81(15.00) P < 0.001; (2) day 0: 8.07 6.09 (7.00) day 22:18.32 8.98 (18.00) P < 0.001
The incidence of adverse events was low; 16 (16%) patients in the SRM group and 13 (13%) patients in the methadone group experienced at least one adverse event (c2 test, P = 0.586). Thirty of 45 (67%) of all adverse events were rated as being unrelated, nine (20%) as possibly related (SRM: six patients; methadone: three patients) and one (2%) (methadone group) as probably related to the study drug. The majority of adverse events (23 of 45) were gastrointestinal system disorders, such as
nausea (three), vomiting (10), dentalgia (five), followed by psychiatric disorders (seven of 45, e.g. dysphoria, agitation, depression, panic attacks).

Seifert 2002SOWS score (days 0-2): no differences; weeks 1-2 (2) fewer symptoms than (1); no differences in self-rating scalesNot assessed.

Sorensen 1982Mean symptom discomfort index initially declined then increased with drug taper. Initial, stabilisation and final ratings by graph: (1)18, 10, 15 (2) 16, 7, 16.1 overdose incident, possibly due to combination with alcohol.

Steinmann 2007Results in favour of buprenorphine only in the first day of treatmentNot assessed.

Tennant 1975Mean daily withdrawal scores (1) 6.6 (2) 9.6. Difference significant on days 8, 12, 15, 16, 17.Only one statistically significant difference. (1)6 [16.7%], (2)17 [47%] reported euphoria.

Umbricht 2003Participant and observer rating scales: after first-dose effect, further improvements were minimal, and overall mean scores during treatment were not significantly lower than scores after one treatment dose. The overall mean time averaged decreases ranged from-5.1 to -6.0 for OOWS and -3.3 to -4.7 for SOWS. No significant differences in the magnitude of the decrease in the withdrawal between the treatment groups. At no time during treatment did withdrawal scores exceed baseline scores for any individual.No major changes in blood pressure or heart rate during the observation time interval of the study. There was a trend (P = 0.06) toward myosis in the buprenorphine and methadone groups, consistent with a pharmacologic effect of treatment. Systolic blood pressure decreased significantly in the buprenorphine group.

Zarghami 2012Statistical analysis revealed that significant decreases (P ≤ .04) were found in the OOWS scores in both treatment methods up to day 14; no statistically significant difference was found between OOWS scores of the treatment methods at different intervals (P ≥ .1), except for day 6 of the study (P = .03) where results were in favour of tramadol.No significant differences were observed in side effects scores for dizziness, somnolence, ataxia, constipation, nausea, seizures, and respiratory depression between two treatment methods, except for perspiration and pain, which were significantly higher in tramadol (P = .02) and methadone (P = 0.01) treatment methods, respectively

 OOWS: Objective Opioid Withdrawal Scale; SOWS: Short Opioid Withdrawal Scale
 
Table 4. Withdrawal symptoms and side effects methadone versus anxiolytics

StudyWithdrawal symptomsSide effects

Buydens-Branchey 2005Subjective and Objective withdrawal scale: no significant differences between methadone and buspirone dosesNot assessed.

Drummond 1989Significant higher scores in the chlordiazepoxide group only on day 3; at the end of the study, the scores were higher in the methadone group but not statistically significant. Analysis of individual items in the OWS (Opiate Withdrawal Scale) failed to implicate any particular item as being responsible for the difference between the two groups.In methadone group relative bradycardia is more present in the first days of treatment and the difference with respect to the chlordiazepoxide group became statistically significant on days 4 and 7. As methadone was gradually withdrawn, the mean heart rate returned to a level comparable to the beginning period. Mean pupil size was less in methadone group during the treatment period and the difference was statistically significant on day 5, similarly mean temperature was lower in this group on day 3.

 
Table 5. Withdrawal symptoms and side effects methadone versus placebo

StudyWithdrawal symptomsSide effects

Buydens-Branchey 2005Subjective and Objective scales: symptoms in placebo group were significantly more pronouncedNot assessed

San 1992Higher scores in the placebo groups, with the most severe symptoms on day 1 to 15. 8/11 placebo-treated patients needed to be switched from placebo to methadone because the OWC (Opiate Withdrawal Checklist) daily score was > 15. With respect to features of the withdrawal syndrome in placebo patients, two stages were observed: anxiety, chills, gooseflesh, myalgia and weakness were the most common on day 1 to 5 and sleep disturbance persisted on day 6 to 13.Not assessed

 
Table 6. Withdrawal symptoms and side effects methadone versus paiduyangsheng

StudyWithdrawal symptomsSide effects

Yang 2006No differences between the two groupsNot assessed