|
|
| |||||||||||||||||||||||||||||||||||||||||||||
Background Opioid dependence is a chronic, relapsing disorder with important public health implications.
Methods In a 17-week randomized study of 220 patients, we compared levomethadyl acetate (75 to 115 mg), buprenorphine (16 to 32 mg), and high-dose (60 to 100 mg) and low-dose (20 mg) methadone as treatments for opioid dependence. Levomethadyl acetate and buprenorphine were administered three times a week. Methadone was administered daily. Doses were individualized except in the group assigned to low-dose methadone. Patients with poor responses to treatment were switched to methadone.
Results There were 55 patients in each group; 51 percent completed the trial. The mean (±SE) number of days that a patient remained in the study was significantly higher for those receiving levomethadyl acetate (89±6), buprenorphine (96±4), and high-dose methadone (105±4) than for those receiving low-dose methadone (70±4, P<0.001). Continued participation in the study was also significantly more frequent among patients receiving high-dose methadone than among those receiving levomethadyl acetate (P=0.02). The percentage of patients with 12 or more consecutive opioid-negative urine specimens was 36 percent in the levomethadyl acetate group, 26 percent in the buprenorphine group, 28 percent in the high-dose methadone group, and 8 percent in the low-dose methadone group (P=0.005). At the time of their last report, patients reported on a scale of 0 to 100 that their drug problem had a mean severity of 35 with levomethadyl acetate, 34 with buprenorphine, 38 with high-dose methadone, and 53 with low-dose methadone (P=0.002).
Conclusions As compared with low-dose methadone, levomethadyl acetate, buprenorphine, and high-dose methadone substantially reduce the use of illicit opioids.
A clinical advantage of levomethadyl acetate and buprenorphine is the option of less-than-daily doses, which is made possible in the case of levomethadyl acetate by the long half-lives of its two active metabolites, nor-levomethadyl acetate and dinor-levomethadyl acetate. Similarly, clinical pharmacologic studies16,17 and controlled trials18,19,20,21 have supported the feasibility of less-than-daily doses of buprenorphine.
Several controlled trials have compared the efficacy of levomethadyl acetate22,23,24,25,26 or buprenorphine11,14,27,28,29 with that of methadone. We compared levomethadyl acetate, buprenorphine, and methadone as treatments for opioid dependence.
Methods
Subjects and Randomization
Two hundred twenty patients participated in this single-site, randomized, controlled study with four treatment groups (Figure 1). The eligibility criteria were an age of 21 to 55 years, a diagnosis of opioid dependence according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV),30 evidence of recent opioid use on toxicologic screening, the absence of serious medical or psychiatric illness requiring long-term medication, and negative results on a serum pregnancy test for women during screening. The study was approved by the local institutional review board, and all patients provided written informed consent.
|
Treatment Groups
Randomization occurred on the day of enrollment. The patients and clinic staff were unaware of treatment assignments and medication doses. The low-dose methadone group (the control group) received a fixed dose of 20 mg of methadone. However, the doses of levomethadyl acetate, buprenorphine, and high-dose methadone were adjusted within prespecified ranges to ensure that comparisons between drugs were based on individually optimized doses.
Levomethadyl Acetate
Levomethadyl acetate was administered at a dose of 75 to 115 mg on Mondays and Wednesdays (to approximate a dose of methadone of 60 to 100 mg daily); the Friday doses were 40 percent higher (to a maximal dose of 105 to 161 mg) to compensate for the longer interval before the next dose.32
Buprenorphine
Buprenorphine was administered at a dose of 16 to 32 mg on Mondays and Wednesdays (to approximate a dose of methadone of 60 to 100 mg daily)11; the Friday doses were 50 percent higher (24 to 48 mg).33
High-Dose and Low-Dose Methadone Groups
The doses for patients in the high-dose methadone group were in the upper range of doses generally used in clinical practice (60 to 100 mg). The dose in the low-dose control group (20 mg) was the minimally effective dose, according to the results of previous controlled trials.6,11
Procedures
The study had three phases: dose induction (weeks 1 and 2), maintenance (weeks 3 to 17), and disposition (weeks 18 to 28). On days of clinic attendance, all the patients received three solutions, only one of which contained active medication. Levomethadyl acetate and methadone were dispensed as oral solutions of different colors, and buprenorphine was dispensed sublingually as a 40 percent aqueous alcohol solution. Missed doses were treated according to preestablished blinded protocols developed by the investigators. Patients were discharged from the study if they were absent for five consecutive calendar days.
Dose Induction
Patients attended the clinic daily and received gradually increasing doses of medication. Patients assigned to levomethadyl acetate received 25 mg on day 1 and then began alternating between placebo and levomethadyl acetate, with 10-mg increases until the dose reached 75 mg. Buprenorphine was administered daily until day 7, starting at 4 mg on day 1 and increasing to 8 mg on days 2 through 7. On day 8, patients began alternating between placebo and 16-mg doses of buprenorphine. Patients assigned to high-dose methadone started with 20 mg, with 10-mg increases daily until the dose reached 60 mg. Patients assigned to low-dose methadone continued to receive 20 mg daily throughout the study.
Maintenance
Starting in week 3, patients attended the clinic three times a week, on Monday, Wednesday, and Friday. All patients received bottles of medication to take home for the other four days of the week.
Dose Increases
Patients receiving levomethadyl acetate, buprenorphine, or high-dose methadone could receive blinded increases in the dose starting in week 3 if they met preestablished criteria. Four dose increases were allowed, with one increase every two weeks. Doses were increased in 10-mg increments for levomethadyl acetate (on Monday and Wednesday, from 75 to 115 mg) and high-dose methadone (daily, from 60 to 100 mg), and in 4-mg increments for buprenorphine (on Monday and Wednesday, from 16 to 32 mg). Patients were evaluated weekly, and their doses were increased if they met two criteria: more than 83 percent attendance, with no Friday absences, and more than 33 percent of urine specimens opioid-positive during the previous week. The attendance criterion ensured that the patient received each intended dose before receiving further dose increases. The intent was to achieve individually optimized doses and to avoid confounding comparisons between drugs by potential differences in the adequacy of the dose.
Eighty percent of the patients in the levomethadyl acetate group, 96 percent of those in the buprenorphine group, and 91 percent of those in the high-dose methadone group met the criteria for dose increases. The mean maximal Monday and Wednesday doses were 100 mg of levomethadyl acetate, 27 mg of buprenorphine, 90 mg of high-dose methadone, and 20 mg of low-dose methadone. The average number of days required to achieve the maximal dose was 69 for levomethadyl acetate, 72 for buprenorphine, 68 for high-dose methadone, and 1 for low-dose methadone.
Rescue Treatment
A double-blind rescue procedure was implemented for patients with a poor response to treatment. Patients received rescue treatment if they attended more than 83 percent of clinic days, if they missed no Fridays, and if at least 66 percent of their urine specimens during the previous two weeks were opioid-positive. The patients also had either to request a dose increase or to report use of illicit opioids at more than 50 percent of the pretreatment level. The earliest they could receive rescue treatment was week 6 (for the low-dose methadone group) or week 13 (for all other groups). For rescue treatment, the patients were switched to an equivalent dose of methadone and followed clinically; they remained blinded to the treatment procedures, and data collection was continued, but data obtained after rescue treatment were excluded from analysis.
Take-Home Recall Procedure
To discourage methadone diversion and ensure compliance with medication, take-home bottles were randomly recalled. All take-home bottles scheduled for recall contained placebo. In a recall, patients were contacted by telephone and required to return their take-home bottles to the clinic; recalled bottles were replaced with a bottle containing placebo in the levomethadyl acetate and buprenorphine groups or with a bottle containing methadone (the scheduled dose) in the methadone groups.
Outcome Measures
The three primary outcome measures for assessing the efficacy of the medications were continued participation in the study (retention); opioid use, measured by the percentage of positive urine specimens, the degree of continuous abstinence from opioid use (defined by at least 12 consecutive opioid-free urine specimens), and the patients' own reports of frequency of use; and the patients' global ratings of the severity of their drug problem. Additional prespecified outcome measures were the percentage of cocaine-positive urine specimens, the duration of continuous abstinence from cocaine use, breath alcohol readings, side effects, and sex-related differences.
Statistical Analysis
Patient retention was calculated from the day of admission (first dose) to the day of study completion (119 days), the breaking of the dose code, the last administration of medication, or rescue. Kaplan–Meier estimates34 of mean retention times were compared among the groups by the Mantel–Cox log-rank test.35
Urine specimens were collected under observation on Monday, Wednesday, and Friday and analyzed on site for opioid and cocaine metabolites by the enzyme-multiplied immunoassay technique (Dade Behring Diagnostics, San Jose, Calif.). Findings of opioid or cocaine metabolites at a level of 300 ng or more per milliliter were considered positive results. Missed urine specimens were collected the following day (except Sundays). The percentage of missing samples was similar in all four groups (4.2 percent in the levomethadyl acetate group, 5.1 percent in the buprenorphine group, 3.0 percent in the high-dose methadone group, and 5.8 percent in the low-dose methadone group); these samples were considered positive for purposes of analysis. Breath alcohol levels (in grams per deciliter) were determined randomly on clinic days with a breath alcohol sensor (Alco-Sensor III, AlcoPro, Knoxville, Tenn.). The doses of medication were adjusted according to preestablished blinded protocols when the breath alcohol level was 0.05 g per deciliter or higher (26 doses were adjusted for this reason).
For each patient, we analyzed the weekly percentages of positive urine specimens separately for opioids and cocaine by multilevel analysis with an autoregression covariance structure36 for the week factor, using SAS Proc Mixed software (SAS Institute, Cary, N.C.). The restricted maximum-likelihood methods used in multilevel modeling have the flexibility to handle repeated-measures data sets with missing observations.37 We compensated for the potential effect of informative missing data by using study retention and the percentage of missed clinic visits as covariates.38 The percentages of patients who submitted at least 12 consecutive negative urine specimens were compared separately for opioids and cocaine by Fisher's exact test.39
The patients reported the frequency of opioid use each week, starting with the week before admission. Of 2993 scheduled reports, 196 were excluded from analysis because they were missing or were at least 3 SD from the mean: 6.8 percent of all scheduled reports for the levomethadyl acetate group, 7.4 percent for the buprenorphine group, 6.9 percent for the high-dose methadone group, and 8.5 percent for the low-dose methadone group. The data were analyzed according to the week with the multilevel analysis described above.
Every four weeks, the patients rated the severity of their drug problem from 0 (no drug problem) to 100 (worst ever). The data from the last assessment before completion of the study or discharge were analyzed by one-way analysis of variance.15 Fourteen patients dropped out before completing any severity assessment, and 14 of 206 patients missed their final assessment and were excluded from analysis (1.0 percent were missed in the levomethadyl acetate group, 1.5 percent in the buprenorphine group, 2.0 percent in the high-dose methadone group, and 2.4 percent in the low-dose methadone group).
All reported P values are two-tailed.
Results
Demographic Characteristics
There were no significant differences between groups in demographic characteristics or stratification variables (Table 1).
|
Patients were enrolled between January 17, 1996, and November 24, 1997. There were significant differences in study retention among the four groups. Pairwise comparisons showed significantly greater mean retention among patients receiving levomethadyl acetate (89±6 days), buprenorphine (96±4 days), or high-dose methadone (105±4 days) than among those receiving low-dose methadone (70±4 days, P< 0.001) (Figure 2A). The difference between patients taking high-dose methadone and those taking levomethadyl acetate was also significant (P=0.02), but not that between high-dose methadone and buprenorphine.
|
Opioid Use
Overall, patients receiving levomethadyl acetate, buprenorphine, and high-dose methadone submitted fewer opioid-positive urine specimens than did patients receiving low-dose methadone (Figure 2B and Table 2). The percentage of patients with at least 12 consecutive opioid-negative urine specimens differed significantly among groups (P=0.005), ranging from 36 percent in the levomethadyl acetate group to 8 percent in the low-dose methadone group (Table 2). There were no significant differences between groups in self-reported use of illicit opioids; however, each group did report a significant reduction in use over time (Figure 2C and Table 2). On a scale of 0 to 100, patients in the low-dose methadone group gave their drug problems the highest severity rating (mean, 53); the lowest rating was in the buprenorphine group (mean, 34) (Table 2).
|
Alcohol use was low at base line and throughout the study (only 3 percent of all breath samples had alcohol levels 
0.01 g per deciliter). Only one participant regularly tested positive for alcohol; all others showed sporadic use, if any. Patients receiving low-dose methadone were the least likely to abstain from cocaine use (Table 2).
Side effects were assessed every four weeks by an open-ended questionnaire and coded to the Coding Symbols for Thesaurus of Adverse Reaction Terms (COSTART) system.40 The percentage of patients reporting side effects was similar among groups. At least one side effect was reported by 55 percent of the patients receiving levomethadyl acetate, 49 percent of those receiving buprenorphine, 45 percent of those receiving high-dose methadone, and 40 percent of those receiving low-dose methadone. The most common was constipation (21 percent of all reports), followed by nausea (8 percent) and dry mouth (6 percent). No toxic interactions associated with illicit-drug use were observed in any of the groups. There were no significant effects of sex on any outcome measure.
Discussion
Levomethadyl acetate, buprenorphine, and high-dose methadone were all effective in treating opioid dependence and were superior on multiple measures to low-dose methadone. The percentage of patients retained at 17 weeks compared favorably with rates reported elsewhere for these medications.11,14,23,24,27,28,29 The proportion of opiate-positive urine specimens and self-reported opioid use decreased over time.
As compared with patients taking low-dose methadone, those taking levomethadyl acetate had a significantly higher rate of continuous abstinence from opioids, and those taking high-dose methadone and buprenorphine had a trend toward a higher rate of continuous abstinence. The greater effectiveness of levomethadyl acetate during maintenance therapy may reflect more stable blood levels with less variation in trough-to-peak concentrations.41 Levomethadyl acetate, however, was less effective than high-dose methadone in terms of patient retention, particularly during the dose-induction period. The high dropout rate among patients taking levomethadyl acetate may be explained, in part, by the longer period required to achieve the targeted maintenance dose (nine days, as compared with five for methadone)42 and the greater reinforcement provided by daily treatment with a full agonist (methadone). Our difficulty in retaining patients during the induction of levomethadyl acetate treatment corroborates earlier findings.25,43,44,45
The low level of retention in the low-dose methadone group at 17 weeks (20 percent) reflects primarily the substantial proportion of patients receiving rescue treatment; the sharp decline in retention observed after week 6 in this group coincided with the earliest time that patients could receive rescue treatment. The fact that a large proportion of patients receiving low-dose methadone continued to use illicit opioids while they were receiving maintenance therapy at a dose of 20 mg of methadone is consistent with other reports that low doses of methadone are less effective in retaining patients in treatment programs and reducing illicit opioid use.6,11,14,28 Nevertheless, the rescue procedure was successful, since 81 percent of such patients in the low-dose methadone group completed 17 weeks of treatment.
Most of the development and evaluation research on buprenorphine has been based on daily doses. Our study used thrice-weekly doses and found that outcomes were approximately equivalent to those with either daily methadone or thrice-weekly levomethadyl acetate. Thus, thrice-weekly buprenorphine may also offer greater convenience to patients and clinic staff.
Urinalysis and self-reporting are surrogate measures of drug-taking behavior that differ in their sensitivity in detecting abstinence. Urinalysis may overestimate drug use; opioid use can decline by 75 percent (for example, from four times to once daily) and still yield 100 percent opioid-positive urine specimens, and frequent testing can allow carryover between consecutive tests. Self-reported drug use is open to either over- or underreporting. However, even though quantitative methods for more accurate detection of new drug use are currently being developed,45 both urinalysis and self-reports are accepted surrogate measures of drug use.46,47
The strengths of our study include the use of doses based on clinical criteria, ensuring that comparisons between drugs were based on individually optimized doses for three of the randomly assigned treatment groups. The use of a fixed-dose control group allowed us to demonstrate differences in outcome measures and to analyze a heterogeneous population of heroin users. The rescue procedure allowed us to retain 87 percent of all patients who received rescue treatment and allowed 66 percent of the total population to remain in treatment. Our rescue procedure also addressed potential ethical concerns.48
In summary, levomethadyl acetate, buprenorphine, and high-dose methadone were more effective than low-dose methadone in reducing the use of illicit opioids. As compared with low-dose methadone, levomethadyl acetate produced the longest duration of continuous abstinence; buprenorphine administered three times weekly was similar to levomethadyl acetate in terms of study retention and was similar to high-dose methadone in terms of abstinence.
Dr. Johnson is or has been a consultant to Roxane Pharmaceuticals (manufacturer of levomethadyl acetate), Reckitt and Colman (manufacturer of buprenorphine), and Schering-Plough (distributor of buprenorphine).
Supported by grants (P50 DA05273, K02 DA00332, K05 DA00050, and Y01 DA40032) from the National Institute on Drug Abuse.
We are indebted to Daniel A. Bloch, Ph.D. (for assistance with statistics); Mary Beth Mudric (assessments); Sherry Thomas (counseling); Connie Lowery and Mary Blackburn (nursing); Iona Johnson and Robert Mullen (pharmacy); Russell Lum, Lisa Marsch, and Jennifer Rosenbaum (research); Tim Mudric and Linda Felch (statistics); and John Yingling and Lisa Schade (technical support).
Source Information
From the Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, 5510 Nathan Shock Dr., Baltimore, MD 21224, where reprint requests should be addressed to Dr. Johnson.
References
-acetylmethadol (LAAM) and active metabolites in plasma following chronic administration. J Anal Toxicol 1977;1:1-5.
| |||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
