Review: A Pilot for a Randomized Patient Preference Trial of Buprenorphine versus Methadone Maintenance Treatment

Byrne Review: A pilot study for a randomized controlled and patient preference trial of buprenorphine versus methadone maintenance treatment in the management of opiate dependent patients
Byrne Comments
September 25, 2008

Pinto H, Rumball D, Maskrey V, Holland R. Journal of Substance Use 2008 13;2:73-82

This pilot study report demonstrates the ethical and practical differences between the British and American approach to drug treatment and research. The first author told me that they were trying to prove that they could obtain as good or better results using buprenorphine when compared with methadone in order to force their local NHS formulary to include it. Hence they attempted to randomise subjects applying for opioid prescription to methadone or buprenorphine and then follow progress. However, the first and probably most important finding of this study was that not one single patient of almost 50 presenting to their service over a six month period agreed to this randomisation. Apparently, each patient already had a clear preference for buprenorphine or methadone. Note that September 25 combination buprenorphine did not rate a mention in this context despite most doses being non-supervised.

Of those who agreed to be followed for this study, 22 chose methadone and 20 buprenorphine. Of those opting for methadone, 80% had had a previous script for the drug. Only 30% of those choosing buprenorphine had had a previous prescription for that drug (and 40% had tried methadone previously). Thus methadone choice was largely based on previous experience while buprenorphine mostly on second hand information. Consistent with the literature they report: “At 6 months more methadone patients were retained (68 vs. 55% for buprenorphine) …”. There was one ‘cross-over’ patient from each group, each ending the trial on the alternative medication.

Despite no randomisation and no significant differences found between those followed “open-label” over 6 months, these authors make a spectacular reversal of both the above ‘trend’, numerous randomised controlled trials and a Cochrane summary, based on slightly different baselines for methadone against buprenorphine subjects. “As a pilot this study lacked power but the results suggest that, in practice, in the UK, buprenorphine may be more able to retain patients in treatment, suppress illicit opiate use and improve functioning [despite that not happening in RCT elsewhere]. Given the significantly higher cost of buprenorphine a larger study is needed to answer these questions.”

Even if one showed significant differences between methadone and buprenorphine outcomes, this would not “favour” one or other drug, both of which are highly effective in a substantial proportion of heroin addicts presenting for treatment. Further, because patients already know what they want, discussion about which drug has a better retention rate or ability to suppress illicit drug use is almost academic. This may be the first reported series of buprenorphine subjects who had all been offered methadone initially as a choice. The finding of comparable results is greatly reassuring for those of us who prescribe buprenorphine regularly.

These authors take another unreferenced ‘dig’ by stating that methadone “causes a degree of persisting intoxication (which can limit the users’ ability to function) … and has a prolonged abstinence syndrome in withdrawal, leading some to suggest that it prolongs dependence.” Thus they perpetuate the myth that methadone is a sedative drug and buprenorphine is not. They base this purely on anecdotal reports that certain patients feel more energy on buprenorphine after having been on methadone. The reverse may be true for certain patients. It is well known that when stabilised, patients on methadone can drive, operate machinery and look after children perfectly safely. If Dr Pinto has patients reporting sedation on methadone then he might consider lower or split doses as recommended by Payte and others.

We have known for 15 years that buprenorphine can obtain results almost as favourable as methadone. It seems that buprenorphine can lead to increased early drop-outs, possibly due to a lack of agonist reinforcement and/or inadequate doses (Kakko used an average of about 30mg daily). It is hardly surprising that some do better on buprenorphine, even though it is clear that rather more will always do better on methadone in general (see Kakko’s classic study in which most buprenorphine-started patients ended up on methadone ‘rescue’). The lack of toxicity in overdose for buprenorphine must be a major factor in a country like England where more than 90% of opiate substitution doses are apparently still non-supervised (‘take-aways’).

Regarding price, generic buprenorphine is now available in Europe and at certain dose levels should be comparable in price with methadone. I understand that it is not an expensive drug to manufacture.

In America most research has been performed in a situation where treatment is in extremely short supply and any offer to join funded drug research, even where placebo is a possible offering, is generally taken up promptly by illicit drug users. Many of us have found ethical flaws in this environment, where “choice” is really taken out of the equation, like offering a ‘choice’ of food in a famine, or for prison settings. None can be considered a genuine volunteer when the alternative to being in a trial is to receive no treatment at all (even though this is apparently the norm for 6 out of 7 American addicts currently).

Comments by Andrew Byrne.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Dr Andrew Byrne MB BS (Syd) FAChAM (RACP)

Dependency Medicine,

75 Redfern Street, Redfern,

New South Wales, 2016, Australia

Email - ajbyrne@ozemail.com.au

Tel (61 - 2) 9319 5524 Fax 9318 0631

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Surgery web page: http://www.redfernclinic.com/

Byrne Review: QTc Interval Prolongation and Opioid Addiction Therapy.

Byrne Review: QTc Interval Prolongation and Opioid Addiction Therapy
Byrne Comments
September 19, 2008

Baker WA, Krantz MJ. Archives Internal Medicine (letter) 2008 168;14:1502

In this letter Baker and Krantz applaud Wedam et al’s ‘commendable’ “addition to the medical literature” which found QT prolongation in 23% of methadone subjects. In fact these researchers found much the same as Janet Lipski’s report from 1973 (34%). Thus it is clear that a substantial minority of methadone patients have modest prolongation of QT intervals of uncertain significance.

Baker and Krantz then express the unsurprising deduction that if a lower cut-off for QT prolongation were used, a correspondingly higher proportion of prolongation subjects will result (up to 50%). The clinical significance of these observations, if any, is not detailed. Indeed, not a single subject in the entire study had a QT interval (corrected) of 500ms or more which is considered by some authorities to be the clinically significant cut-off for risk of tachycardia. And none developed tachycardia.

Krantz has wrongly cited Lipski’s study as being (at least in part) to investigate sudden deaths in methadone patients. In fact, Lipski and colleagues were concerned about heroin deaths and the QT findings in methadone patients were incidental. When this mistake was pointed out in Lancet, Krantz in reply ignored the matter and turned to increasing deaths of non-addiction treatment (pain) cases in America more recently (Mehler 2007).

Specifically, Krantz writes that Lipski’s study was aimed at investigating “a perceived increase in the risk of sudden death in heroin addicts, even in those successfully treated with methadone” in 1973 in New York (Krantz MJ 2006). From the title and opening lines it is clear that Lipski and colleagues were actually concerned over deaths in heroin users where sometimes blood levels were not particularly high, an issue raised by long time city coroner Dr Milton Helpern (see ref). They performed cardiographs in new methadone applicants who had used street heroin in the previous hours or day(s) to see if there were any electrical disturbances which might explain the sudden deaths in heroin users.

Unlike in his first paper on this subject, Krantz now treats QT prolongation in isolation as if it were a complication in itself, avoiding discussion of torsades or other cardiac events. And like most of the other studies, Wedam and colleagues report no symptomatic heart disease whatever and, most importantly, no sudden deaths. Over a period of 30 years of intensive, supervised prescription of methadone only one single case of arrhythmia was reported to the FDA (Pearson). So while torsades or ventricular fibrillation occur in methadone patients, the prevalence must be very low and possibly near to the rate in the general population (estimated to be 1 in 2000 - see Smith W).

This laudatory letter is just another ‘weak link’ in the case put by Krantz that due to potential cardiac complications, methadone treatment needs to be reviewed, especially when used in high doses. He recently noted ‘with consternation’ (hardly a scientific term) that higher doses were being used in addiction clinics in his own state, something most public health specialists would have applauded (D’Aunno). Many rigorous studies indicate that higher doses protect dependency patients from significant and measurable events (overdose, viral disease, criminal behaviour, depression, etc). Dole’s seminal study of methadone treatment reported up to 180mg daily with excellent results and no serious side effects.

To my knowledge there have still been no reported series of confirmed case reports of cardiac complications (eg. torsades-de-pointes) in patients receiving methadone treatment for addiction under existing guidelines. Krantz emphasises the proven benefits of MMT yet at the same time questions this established, evidence-based practice, emphasising a difficult therapeutic “trade-off”. Yet this is before there is documented evidence of a problem existing and despite 30 years of careful research showing safety and effectiveness. Further, there has been no analysis of the costs and benefits of any suggested alternative strategies such as low dose methadone or transfer to buprenorphine. Krantz has also apparently had minimal input from dependency specialists. Few non-cardiologists would be so bold regarding advice on heart treatments.

Dr Krantz's widely publicized views have had the consequence of denigrating methadone as a treatment for both addiction and chronic pain. I can find no evidence suggesting that any cases of cardiac complications resulting from methadone treatment have been avoided as a result. The literature reveals only flimsy and conflicting evidence - most of it retrospective - of an association of methadone and cardiac events. Some have reported a lower cardiac risk (Marmor 2004).

Krantz's articles, letters and personal communications show a clear 'disconnect' between his earlier findings and the sentiments he has expressed more recently. While his earlier findings relate mostly to pain patients, he has strongly targeted an addiction treatment audience for his communications and further research (eg. Krantz 2007). His advice, which has changed over the years, has involved (1) the need (or otherwise) for a cardiograph before starting treatment, (2) the avoidance of methadone where possible and (3) the avoidance of high doses where methadone is used. His other recommendations are purely generic in recommending careful history and physical on each patient and acting on any evidence of cardiac risk. While this is important, it is probably no more important than acting on evidence of infectious disease, overdose risk, liver disease, allergies, mental illness, etc, etc.

The potential risk of cardiac side effects must be balanced against the unequivocal benefits of methadone maintenance to those who need and want it for their condition. There need be no ‘competition’ with buprenorphine which is an excellent alternative with certain benefit and certain disadvantages in the opiate treatment population.

Comments by Andrew Byrne.

References:

Wedam EF, Bigelow GE, Johnson RE, Nuzzo PA, Haigney MCP. QT-Interval Effects of Methadone, Levomethadyl, and Buprenorphine in a Randomized Trial. Arch Intern Med 2007 167;22:2469-2473

Krantz MJ, Mehler PS. QTc prolongation: methadone's efficacy-safety paradox. Lancet 2006 368:556-557

Byrne A, Stimmel B. Methadone and QTc prolongation. Lancet 2007 369:366

Mehler PS, Krantz MJ. Authors’ reply. Methadone and QTc prolongation. Lancet 2007 369:366-7

Smith WM. Cardiac repolarisation: the long and short of it. MJA 2008 188;12:688-689

Lipski J, Stimmel B, Donoso E. The effect of heroin and multiple drug abuse on the electrocardiogram. American Heart J 1973 86:663-8

Pearson EC, Woosley RL. QT prolongation and torsades de pointes among methadone users: reports to the FDA spontaneous reporting system. Pharmcoepidemiol Drug Saf. 2005 14;11:747-753

D'Aunno T, Folz-Murphy N, Lin X. Changes in Methadone Treatment Practices: Results from a Panel Study, 1988 - 1995. American Journal of Drug and Alcohol Abuse 1999 25;4:681-700

Dole VP, Nyswander ME. A medical treatment for diacetylmorphine (heroin) addiction. J Amer Med Assoc 1965;193:646-50 '193(8) 80-84'

Helpern M. Fatalities from narcotic addiction in New York City: incidence, circumstances and pathologic findings. Human Pathology 1972 2:13-21

Krantz MJ, Rowan SB, Schmittner J, Bucher Bartelson B. Physician Awareness of the Cardiac Effects of Methadone: Results of a National Survey. Journal of Addictive Diseases 2007 26;4:79-85

Marmor M, Penn A, Widmer K, Levin R, Maslansky R. Coronary artery disease and opioid use. Am J Cardiol. 2004;93:1295-1297

Leavitt S. Methadone: facts and fiction. Available from: ATForum.com.

Review: A Randomized Clinical Trial of Methadone Maintenance for Prisoners: Findings at 6 Months Post-release

Byrne Review: A randomized clinical trial of methadone maintenance for prisoners: findings at 6 months post-release
Byrne Comments
September 19, 2008

Gordon MS, Kinlock TW, Schwartz RP, O'Grady KE. Addiction 2008 103;8:1333-1342

These researchers found, predictably, that offering methadone maintenance treatment (MMT) to released prisoners with a history of opiate addiction was feasible, safe and effective, just like it is in the community generally when done according to established guidelines. They compared counselling with/without MMT, finding less heroin use and less criminal activity at 6 months after release in those offered MMT. Treatment retention was also higher.

This is yet another example of American clinical practice which is decades behind other countries. And this is despite heroin addiction has been accepted as a ‘brain disease’ by the White House and methadone/agonist treatments are now approved in every state. However, for those in the US prison system these maxims do not apply for some reason. Note that this study was not published in an American journal.

Almost uniquely, in New South Wales, prisoners have had access to methadone treatment for over 20 years. It was initially introduced in the 1980s as a pre-release measure to address the high rate of overdoses in that group. There is now a copious world literature on the subject, largely very positive. Methadone for prisoners has now been introduced in many other jurisdictions, although rarely ‘across the board’ as occurs in New South Wales.

Thus a trial which gave some subjects no access to such treatment would be unethical, unnecessary and cruel in a ‘normal’ jurisdiction. Yet in America, despite a large drug budget and constitutional protections, denying prisoners appropriate treatment seems to be ‘business as usual’. These researchers should be commended for trying to buck the trend. They write that there is an “urgent treatment need” and one only hopes that something is done for the sake of the prisoners, their families and the general community where the adverse consequences currently must be enormous.

Comments by Andrew Byrne ..

Surgery web page: http://www.redfernclinic.com/

Opera blog: http://www.redfernclinic.com/opera/critique/blog/

New York in 2008: http://ajbtravels.blogspot.com/

New York in 1922: http://bpresent.com/harry/code/10b_bowery.php

Review: A Community-Based Evaluation of Sudden Death Associated with Therapeutic Levels of Methadone

Byrne Review: A Community-Based Evaluation of Sudden Death Associated with Therapeutic Levels of Methadone
Byrne Comments
September 5, 2008

Chugh SS, Socoteanu C, Reinier K, Waltz J, Jui J, Gunson K. American Journal of Medicine 2008 121, 66-71

In my view this study draws a long pseudo-scientific bow, implicating methadone treatment in cardiac effects without as much as one confirmed case description of an arrhythmia in a methadone maintenance subject in the catchment area.

The study selected all 183 subjects who died of sudden death in Portland, Oregon over a 4 year period, grouping them according to whether they had ‘therapeutic’ levels of methadone (up to 1.0mg/L) or not. They excluded cases of proven poisoning from methadone (11 cases) or other recreational drugs (32 cases - all apparently in the methadone group). They also excluded those who did not have a full autopsy 7/29 in study group and 5/111 in the ‘control’ group (a rather large difference). The authors found that there was identifiable structural heart disease in 23% of the methadone group and 60% of the ‘control’ group. This was mostly coronary artery disease and/or ventricular hypertrophy.

Of the 22 cases with therapeutic levels of methadone, 55% were prescribed the drug for pain, 14% took it ‘recreationally’, 18% unknown and 14% for ‘opioid withdrawal’ - note that none were reported to be on methadone maintenance treatment (although 3 patients, 14%, were apparently prescribed the drug for addiction purposes). We are not informed of the quality and availability of addiction services in the area but I understand there are a number of large licensed methadone clinics in Portland. We are not told the details of the 3 addiction cases who died.

The authors claim their data show an association between methadone at ‘therapeutic’ levels and sudden death. They speculate about cardiac conduction, prolonged QT interval and ventricular tachycardia despite no reported cardiographic evidence pre-mortem. The world literature they cite only appears to contain 28 cases of methadone related tachycardia, mostly non-fatal (see below). A more likely cause of death from methadone than arrhythmia is respiratory depression, even at ‘normal’ blood levels. The authors note this in their discussion but still maintain that their study provides evidence, albeit indirect, for arrhythmias.

I find the methodology of this paper bewildering. If this study were duplicated for any other drug, eg. insulin, thiazides, aspirin, fluoxetine or lipid lowering drugs, such conclusions on the cause of death would be considered laughable in my view.

The main weakness of the study is the attempt to link cases in whom no cause of death could be found with a very rare side effect of methadone, in the face of other possible causes of death. This is especially hard to understand in the majority of cases prescribed the drug for pain relief (55%) as opposed to methadone maintenance patients treated for addiction (~14%). The second weakness of the study related to the 'therapeutic' levels of methadone. They seem to believe (1) that therapeutic levels imply therapeutic doses and (2) that these do not cause respiratory depression and death. Neither of these assumptions is correct and no reference is given to support them, making the paper faulty and its conclusions even more questionable.

Two cited series of prolonged QT interval and/or torsades tachycardia cases (Krantz and Pearson) show a minority of patients treated under addiction program protocols (and in whom there was only one single death). Krantz’s series had an average dose of about four times the usual mean dose in dependency treatment (397mg daily – and none died). Ellen Pearson’s series of 59 FDA reports nationally included only 14 likely dependency patients of whom only one died (a patient on a sub-therapeutic dose of 29mg daily). Ehret’s series was a retrospective study and patients took an average of 4 additional drugs (range 0-14).

Other citations given are single case reports and/or do not involve methadone maintenance patients but are patients with a variety of serious medical conditions for which methadone was used for pain management or else were overdose cases. There are only a few isolated torsades reports in dependency cases over the last 40 years and unexplained sudden death is either unreported or else exceedingly rare. It is certainly not a major public health issue as claimed by Krantz.

In order to be quite certain about all this, having been in New York for several weeks recently I met up with numerous doctors involved in methadone treatment of tens of thousands of patients over a long period. Not one of these could cite a single case of an addiction patient developing torsades in their experience. This is indirect proof to my mind that an easily diagnosed condition (torsades tachycardia) with a mortality of about 15% is no public health problem but a clinical rarity.

Colorado cardiologist Dr Mori Krantz wrote the seminal paper of this subject and while he writes candidly of the various proven benefits of traditional methadone treatment, his clear implication is to avoid methadone treatment if possible and beware of high doses when it is the chosen treatment. Yet most addiction guidelines advise higher dose methadone for better outcomes, with the usual safeguards. Also, most authorities recommend methadone in pregnancy and in those with the most severe dependency and/or co-existing mental illness. Yet Krantz has still not provided a series of detailed individual case studies of this syndrome occurring in regular addiction treatment subjects. And this is despite giving advice about how we should treat such cases.

It is intriguing that Krantz took funding for a large survey of methadone clinic staff awareness, despite the lack of evidence (including his own) of it being a significant problem in such patients when compared to pain management cases. If torsades tachycardia is indeed associated with methadone, it does not appear to occur to any measurable extent in methadone maintenance patients. Most cases have been taking high doses (>300mg daily) and/or were taking multiple drugs with serious medical or metabolic problems. Recent alarming upsurges of methadone-related deaths in America do not involve clinic populations but seem to be associated with the increased popularity of the drug as an analgesic used by doctors and patients who may be unfamiliar with its very long half-life and consequent overdose potential.

Comments by Andrew Byrne ..

Lancet letters

References:

Krantz MJ, Lewkowiez L, Hays H, Woodroffe MA, D. Robertson AD, Mehler PS. Torsade de Pointes Associated with Very-High-Dose Methadone. Ann Intern Med. (2002) 137:501-504

Pearson EC, Woosley RL. QT prolongation and torsades de pointes among methadone users: reports to the FDA spontaneous reporting system. Pharmcoepidemiol Drug Saf. 2005 14;11:747-753

Grönbladh L, Öhlund LS, Gunne LM. Mortality in heroin addiction: impact of methadone treatment. Acta Psychiatr Scand (1990) 82:223-227