Concord Dependency Seminars were previously published at

5 December 2009

The GREAT debate - what is the best drug for dependency?

Concord Seminar Series – Tues 1st December 2009.

Collegial debate: What is the drug of first choice for opioid dependence?

In this seminar we had four cases "for the sake of argument" as to what should be the drug of first choice for opioid dependence.

Richard Hallinan first argued the case for buprenorphine.

The first point is buprenorphine's similar efficacy and cost effectiveness compared with methadone. Systematic reviews show retention in treatment was superior for flexible MMT over flexible BMT dosing but no significant difference in opiate use; overall MMT was slightly more effective and less costly than BMT (Connock et al 2007; Mattick et al 2008, Cochrane).

Lower retention may reflect an advantage of buprenorphine, the ease of reductions & withdrawal. Buprenorphine had the highest cost effectiveness for inpatient or outpatient withdrawal management of the modalities examined in the NEPOD studies (Shanahan et al 2006); and the ease of reductions/withdrawal in short term dosing 1-4 weeks is endorsed in a Cochrane review (Gowing et al 2009). Buprenorphine has great flexibility as a "gateway" treatment, whether to abstinence, naltrexone (O’Connor et al 1997; Collins et 2005; Umbricht et 1999), or MMT.

The third major point is buprenorphine's greater margin of safety than methadone, reflected in low mortality rates in BMT in France despite liberal availability and unsupervised dosing for a number of years. “Indirect comparison of data from population cross-sectional studies suggests that mortality with BMT may be lower than that with MMT" (Connock et al 2007). In a 10 year longitudinal follow-up of participants in a randomized trial of MMT vs BMT there was no significant difference in mortality by treatment group (Gibson et al 2008). NSW data following up all MMT and BMT starts showed "Despite shorter retention in treatment, buprenorphine maintenance was not associated with higher risk of death." (Bell et al 2009).

Buprenorphine has fewer & less severe side effects and drug interactions compared with methadone. QT prolongation is not associated with buprenorphine, and pharmacokinetic interactions are less troublesome, partly because of the wide safety margin of buprenorphine itself. Sexual side effects appear to be lower with buprenorphine. Most importantly, sedation, especially with benzodiazepines, is less. (Lintzeris et al 2006, 2007)

A further argument in favour of buprenorphine as first line treatment is that a stepped care model starting with buprenorphine progressing to MMT has been shown to be non-inferior to standard MMT (Kakko et al 2007, 96 self-referred subjects randomised). Outcomes were virtually identical for retention in treatment and proportion of urine samples free of illicit drugs. Among completers of stepped therapy, 46% ultimately remained on buprenorphine/naloxone.

A number of RCTs show equivalence of second daily or thrice weekly dosing to daily dosing. Less frequent attendance has greater acceptability to patients, and there is reduced need for take-home or unsupervised doses, reduced risk of diversion, and potentially greater public and political acceptability of this treatment.

Finally there is the existence of a formulation of buprenorphine less attractive to injectors, buprenorphine/naloxone. “Most (68%) had tried the buprenorphine +naloxone combination IV, but 80% said they had a "bad" experience. Its street price was less than half that of buprenorphine alone.” (Alho et al 2006). “In the year after its introduction in Australia, BNX was injected less frequently and by fewer regular IDUs and clients compared with BPN” and compared with methadone. (Degenhardt et al 2009, IDRS, 2003-2007, adjusted for sales).

RH concluded: "What I am not saying: is that buprenorphine is "better". But why would you NOT choose as first line treatment a medication of similar efficacy and cost effectiveness, greater safety, better side effect profile, lower drug interactions, greater ease of reductions, and greater public acceptability than MMT, which gives equivalent results in a stepped care model to standard MMT? "


Prof Nicholas Lintzeris next argued why sustained release oxycodone (or indeed sustained release morphine, now available in twice daily and even once daily formulations) would be the ideal drug for opioid maintenance treatment. This is admittedly hypothetical since current NSW law only allows oxycodone hydrochloride for pain indications in non-dependent patients except under specific authority from PSB.

Prof Nicholas Lintzeris gave the case for oxycodone based on:

# Pharmacological principles (sustained delivery, avoiding peaks and troughs, possibly even superior in this respect to methadone which can fluctuate widely within a 24 hour period and sometimes needs twice daily dosing to avoid emergence of withdrawal symptoms).
# Side effects (there may be less constipation and sweating than with methadone)
# Overcoming demand problems (these medications are already preferred by many, if not most, opioid dependent people, on grounds of greater predictability and safety and lower cost, compared with heroin).
# Overcoming supply problems (any doctor can feel comfortable prescribing these medications, as they already do so for pain, provided they adhere to risk management strategies; the risks and difficulties associated with long half life agonists - accumulations and overdose - do not apply).
# Stigma (methadone has a stigma which discourages many people from entering maintenance treatments).
# Do-it-yourself injectables clinic.

There is an existing, albeit limited evidence base regarding the use of slow release oral morphine to treat opioid dependence (e.g. see Mitchell, White et al 2004 - below), demonstrating the such approaches are feasible. In Austria, oral morphine is a licensed and available treatment approach for opioid dependence, and accounts for approximately a third of treatment places. An important step forward however for more widespread uptake of this approach is development of abuse deterrent preparations that reduce the risk of injecting formulations.


Richard Hallinan then put the case for naltrexone. Naltrexone is the most direct path to abstinence, the gold standard treatment outcome. It has minimal side effects, no opioid side effects, minimal drug interactions, is non dependency forming, reduces opioid craving, and it’s easy to stop.

Naltrexone also has other therapeutic benefits including effectiveness in reducing alcohol consumption (Cochrane standard evidence), effectiveness for weight loss (especially in combination with bupropion) and effectiveness to reduce amphetamine use (Jayaram-Lindstrom et al 2008 RCT n=80, vs placebo).

Naltrexone ought to be first line treatment, however currently it is manifestly not so, neither as doctor or patient preference. Low demand and uptake reflects low patient enthusiasm, poor retention, and in Australia the cost, as the treatment is unsubsidised. There are also difficulties of induction, including the high cost of rapid opioid "detoxification" as the preliminary step.

There is a lack of RCT evidence supporting oral naltrexone: poor treatment retention, and modest heroin use reduction compared with placebo, which was non-significant with adequate concealment. (Minozzi et al 2006, Cochrane review). There was also greatly elevated overdose risk due to poor treatment retention. “The relative risk of death for oral naltrexone subjects was 7.4 times (high-risk period, p < 0.0001) or 2.8 times (low-risk period, p = 0.055) that of methadone subjects.” (Gibson and Degenhardt 2007, see also Digiusto et al 2004, NEPOD).

Naltrexone is safe and effective if people take it. As the problem is compliance with oral naltrexone, naltrexone sustained delivery formulations should be the answer. Unfortunately, large scale use of naltrexone implants began in Australia before preclinical, phase 1, 2 or 3 studies were carried out. Various formulations and devices have been offered in different places, and at different times, with some doubts about uniformity of manufacture and good manufacturing practice accreditation over the years.

Even nearly a decade later, the published literature leaves the critical reader in doubt about the rate of serious adverse effects, the efficacy of implants (duration of adequate blood levels and inter-individual variability in blood levels), the risk of problematic use of other substances, and long term outcomes.

Retrospective and naturalistic studies for efficacy and safety, however, suggest probably higher safety from implant formulations compared with oral naltrexone, and at least comparable safety (in terms of mortality) compared with MMT and BMT.

In Western Australia, naltrexone implant treatment was associated with reduced opiate overdoses compared with pretreatment, but increased sedative and other drug overdoses and other drug related hospital admissions (Hulse et al. 2005, Ngo et al. 2008). In Queensland, a retrospective study from a single treatment centre found lower crude mortality but no significant difference in standardized mortality rate for naltrexone implant compared to BMT (Reece 2007). Similarly Tait et al. (2007) found age standardized mortality rate ratio for naltrexone implant vs MMT to be 0.65 (95% CI = 0.12–1.2 NS).

There have been two RCTs of the “GoMedical” naltrexone implant. Kunøe et al 2009 (n=56, 180 days) found lower heroin use, fewer days’ use compared with “normal after care”. Hulse et al 2009 (vs oral Ntx, n=70, 180 days) found lower self reported opiate use, but no difference in urine test confirmed heroin use; other drug use, especially non-heroin opioid use, was poorly assessed. The average duration of adequate blood naltrexone level was substantially revised downwards, compared with previous studies. The good news was that in neither study were there any deaths in patients who received implants.

RH concluded: "What I am not saying is that naltrexone is 'better'. But everyone deserves a chance at abstinence before being committed to agonist therapy, which leaves patients physically dependent on opioids with their considerable side effects, as well as being subjected to intrusions and restrictions on their lives, which may be for many years with no good evidence for how to end the treatment. The future of naltrexone treatment for opioid dependence must be sustained release naltrexone preparations."


Andrew Byrne, Redfern Surgery.

Methadone is the drug of choice for opioid dependence. Forty years of research prove that this therapy is effective in retaining patients in treatment, reducing illicit drugs use, lowering viral transmission and eliminating almost 80% of overdose deaths. None of these benefits has been shown to the same degree for buprenorphine and thus it should not be recommended as first line drug unless there are contraindications for methadone. Further benefits for those taking methadone is that it is safe in pregnancy and lactation while it is also more widely available both in Australia and around the world.

A possible increase in the overdose rate in the first two weeks of treatment is balanced by high numbers of drop-outs in the first two weeks of buprenorphine treatment. A recent large study in Norway showed two deaths in the first month of treatment out of over 3000 episodes in a seven year period.

However, patients make the choice about which drug much more often than doctors do (see below).

Cost effectiveness is far more positive with methadone, being an inexpensive non-patented drug. Common side effects are limited to constipation and sweating. More serious side effects can be limited by dose adjustments and hormone replacement when testosterone levels are suppressed. QT prolongation is common but appears to be of no clinical significance in isolation. Torsade de pointes tachycardia has only been associated with multiple risk factors some of which coincide with long-term survivors of opioid use, prescribed medications, alcohol and illicit drug use. Methadone has not been demonstrated to cause torsade and it is possible that higher doses make this less likely to occur. It is generally agreed that as with other drugs having a dose response, there should be no arbitrary limit on methadone dose levels.

Some references were detailed for the Concord audience.

RCT buprenorphine/methadone.

 Ling W, Charuvastra C et al.
 Mattick RP, Ali R, White JM et al.
 Kristensen O, Espegren O et al.
 Kakko J, Grönbladh L, Svanborg KD et al.
 Pinto H, Rumball D et al.
 Kamien JB, Branstetter SA, Amass L.
 Petitjean S, Stohler R, Déglon JJ et al.
 Cochrane Database of Systematic Reviews

Reasons for choosing MMT.
 Cost (cheapest drug around)
 Safe in pregnancy
 Available most widely (Aussie pts travel!)
 Highest retention rate
 Lowest illicit drug use rates
 Safety when used under supervision
 Original and best known.

How many patients have already tried both methadone and buprenorphine? It seems that the majority of patients attending for opioid dependency treatment are now aware of both methadone and buprenorphine and most have a clear idea which is best for them. This is due to personal trial and error or else street talk of the properties of both. But is this all hypothetical? Doctors are in fact only rarely in a position to decide/advise between methadone and buprenorphine.

English GP study
 A pilot study for a randomized controlled and patient preference trial of buprenorphine versus methadone maintenance treatment in the management of opiate dependent patients. Dr Pinto in England was trying to prove that buprenorphine had benefits worth including on the NHS formulary for their area. Out of 42 subjects recruited, none agreed to be randomized. The reason was that all patients had a clear idea which drug they wanted and refused to consider alternatives in this trial setting.
Pinto H, Rumball D, et al. Journal of Substance Use 2008 13;2:73-82

World’s largest study on OTP.
 Burns L, Randall D, Hall WD, Law M, Butler T, Bell J, Degenhardt L. Opioid agonist pharmacotherapy in New South Wales from 1985 to 2006: patient characteristics and patterns and predictors of treatment retention. Addiction 2009 104;8:1363-1372

Burns et al. abstract quotes:

 “the hazard of leaving treatment was 1.9 times for those on buprenorphine relative to those on methadone”
 “This study has provided population-level evidence to suggest that retention in methadone and buprenorphine differ in routine clinical practice.” (retention in methadone was much better).

Things go better with methadone!

 Much lower diversion rates (anecdotally)
 No substantial problems with contaminated drug being injected (unlike buprenorphine).
 Less “revolving door” syndrome as with buprenorphine (see Burns’ latest study).

Best drug for addiction Rx.
 Do you want your patient to have the best chances of getting their lives together, staying in treatment and avoiding illicit drugs?
 Of course you do!
 Then you should recommend methadone as a first option when there is a choice.

Is this debate “for real”?
 Would we be having this debate about antibiotics, antidepressants or hypertension drugs?
 Of course not!
 **Treatment matching** is the way to go. [Which brings us to the case studies presented in the second half (see below)]
 Dependency medicine has a way to go, too!


Post Script

RH introduced several intentional errors/dubious claims in his presentations, some of which were identified by participants, and others swallowed whole, perhaps owing to the poker faced delivery.

There is no good evidence that naltrexone reduces opioid craving.
Evidence that sexual side effects are lower with buprenorphine is weak, and there is no research for women.
Second daily or thrice weekly buprenorphine dosing in clinical practice seems to suits only a minority.
Naltrexone alone, without bupropion, has small effectiveness for weight loss.
RCT evidence does not support claims for greater safety for buprenorphine (but as NL pointed out, safety for 3rd parties, especially children, is certainly greater for buprenorphine).
There is no evidence that reductions over a longer period are easier for buprenorphine than methadone, and 2 studies sometimes cited to support the claim short term ease of buprenorphine reductions compared with methadone compared a buprenorphine + carbamazepine regime to an methadone + carbamazepine regime. Carbamazepine is a powerful inducer of the metabolism of methadone. (Seifert et al 2002, 2005)


Concord Dependency Seminar November 2009, Treatment Matching - Case Studies

Kyle, a 19 yo heroin user, has been smoking heroin since age 16, daily for the last year, and recently he started injecting it. He used MDMA on weekends from age 16, seldom drinks alcohol or smokes THC, but smokes cigarettes. He has previously "dried out" several times using his mother’s Serepax, but stayed clean for only 3-10 days. He is a student at uni doing arts/law, and lives with his parents. He has maxed out credit card, has stolen money from his parents, used up all a private inheritance from his grandfather who died 3 years ago ($40,000).

Kyle feels he has lost control. He badly needs money, and feels he may have to work as a prostitute. He doesn’t want anyone to know he is using heroin, and is terrified his step-father (a public prosecutor) will find out. He can’t go away or do a rehab for this reason, and anyway he is in the middle of term with exams in 6 weeks. He asks for medications so he won’t have to use heroin.

What treatment is first choice for Kyle?

Comments: a show of hands gave majority support for buprenorphine for this patient, owing to his current predicament, and the short duration of his opioid dependence. He might use buprenorphine for withdrawal management and go onto maintenance if needed. Ross Colquhoun (who will present his experience with naltrexone implants in the Feb 2010 Concord seminar) felt this patient would do very well with naltrexone implant, but this would be too disruptive currently and he should be stabilised on buprenorphine first. Cost would probably also be a major problem.


Robbie is 41 yo, and has been a disability pensioner for 15 years for "mental health problems", including psychosis. He has been an involuntary inpatient for extended periods, including 6 months last year, and his community treatment order has ended. He is no longer getting depot flupenthixol. Two recent hospital admissions under the mental heath act were for aggression/violence. He comes asking for treatment because he has been hanging out, says he has been using $200/day of heroin (he says he has a neighbour who is a dealer). He says he was on MMT once before for 18 months but didn’t like it. Sometime he does inject BM methadone.

On examination he is in classic opioid withdrawal, and has plenty of old scars and recent tracks.

A call to the local mental health service confirms his psychotic episodes have often been precipitated by alcohol or, apparently, amphetamines. He also uses benzodiazepines and cannabis, according to previous history and urine toxicology.

What treatment is first choice for Robbie?

Comments: a show of hands gave majority support for methadone for this patient, owing to his psychotic illness and the antipsychotic properties of methadone. Retention in treatment might be an issue with buprenorphine given his previous poor compliance with antipsychotics. However, doubt was cast on the history of heroin use, as it seems unlikely a long standing disability pensioner for "mental health problems" would have the capacity and resources to afford $200/day. Induction onto methadone would have to be very cautious. It was noted that amphetamine intoxication could potentially be confused with opioid withdrawal. Naltrexone treatment was generally considered inappropriate.


Emma is 36, and a business analyst working for a merchant bank. She is single, works hard and plays hard, and loves having the freedom to party. But partying has got a bit out of control. She admits to near daily drinking (60-90g/day for several years), & occasional use of cocaine on the weekends. There is also frequent use of over-the-counter and prescribed codeine (40-120mg/day) and oral amphetamines as a "pick me up" during the week.

3 months ago Emma was introduced to heroin. She has recently injected it a few times and now is afraid she might be getting a habit. She wakes up craving heroin, can’t get it off her mind, and uses codeine to get through the day. She finds heroin really helps with her severe period pains and wonders about getting a hysterectomy. The heroin use worries her because she doesn’t want to be addicted. She despises junkies and is appalled by the idea of methadone.

What treatment is first choice for Emma?

Comments: the majority view was that alcohol was her primary problem, and her other drug problems (use of codeine and amphetamines for hangover) might be more easily controlled if she could achieve alcohol abstinence. An inpatient opioid/alcohol withdrawal management and a period of residential rehabilitation, or intensive outpatient counseling supported by alcohol pharmacotherapy would be the best option if she were prepared to try it.

However, if opioid maintenance was needed, a show of hands gave majority support for buprenorphine for this patient, owing to her resistance to methadone and the possibility that alcohol use might be less problematic if she were on a partial agonist. However the gynaecological problems need assessment and management, and her chronic pain might make a full opioid agonist necessary. NL argued that this is exactly the sort of patient who would do well with supervised dosing of a sustained release opioid such as oxycodone, which might be justified for her chronic pain, and which she would probably find more acceptable than methadone or buprenorphine. Ross Colquhoun felt this patient would do very well with naltrexone implant, which would also help with her alcohol problem.

Ref 1.
Mitchell TB, White JM, Somogyi AA, Bochner F. Slow-release oral morphine versus methadone: a crossover comparison of patient outcomes and acceptability as maintenance pharmacotherapies for opioid dependence. Addiction. 2004; 99:940-5