22 February 2011

Possibly the last case report of torsade in a methadone patient. Now, how to treat it?

Torsade de Pointes due to Methadone Use in a Patient with HIV and Hepatitis C Coinfection. John J, Amley X, Bombino G, Gitelis C, Topi B, Hollander G, Ghosh J. Cardiol Res Pract. 2010; 2010: 524764

Dear Colleagues,

This case history is instructive although it is not novel. To call this torsade "due to methadone" in the title is highly misleading and contrary to the paper’s data.

The daily methadone dose was low at 40mg (and ‘recently increased’); the patient was aged 50 and was also prescribed three antiviral agents and two antibiotics for HIV. These factors made the patient easily identifiable as a high risk case for numerous medical complications including torsade de pointes tachycardia. Importantly, one of the HIV drugs used has been reported elsewhere to cause torsade (atazanavir, as reported by Dabiesingh; Ly and others). Hence older age, atazanavir and HIV are thus all known to be specific risk factors for developing QT problems. There is also some evidence that HCV may be an independent risk factor for QT prolongation (Norden). From the literature I have searched it seems that most HIV subjects with torsade de pointes are not taking methadone.

Unfortunately, details of this patient’s dependency management are scanty and it seems that there was limited if any addiction specialty input. Note that the patient survived and even 23 days after the methadone was ceased still had QT prolongation and was considered to be in need of a pacemaker (ICD). The reader is not informed, but methadone may have been substituted by another drug, such as morphine. Either way, it would seem to indicate that the methadone was NOT the cause of the electrical perturbations in this patient, despite the alarmist title.

The authors concede that a dose level of 40mg (increased two months before the episode following two years of no illicit drug use) is very unlikely to be a significant causative factor for QT prolongation. The methadone may even have delayed the onset of torsade, for example, by keeping the patient from other cardio-toxic drugs including alcohol, cocaine and amphetamine. It is paradoxical that this very patient was administered another QT prolonging drug, the anti-arrhythmic amiodarone during resuscitation. This drug is similar to methadone in that it can reportedly cause QT prolongation but rarely induces torsade.

These authors cite Adelaide author Athanasos but do so rather selectively. While he does indeed describe U-waves, of uncertain clinical significance, his study found no cardiologic consequences of low to medium doses of methadone, the latter findings appear to be ignored in the current paper. It is intriguing that these authors would repeat Krantz and colleagues’ controversial and unproven advice (Ann Int Med) about performing annual ECGs on methadone patients (see responses in Annals, 4 against, none in favour). This very case report is yet another example in which such serial ECGs were performed and yet torsade occurred regardless. It seems that in America performing a test is sometimes seen as a defence regardless of its utility or otherwise.

That this patient is alive at 50 after having such serious complications of IV drug use as HCV, HIV and hepatoma is indeed a credit to the methadone and medical care received at this institution bearing the name of one of the most innovative doctors of all time, Maimonides (see his treatises on haemorrhoids, digestion, chicken soup and cohabitation).

Reports of torsade in methadone patients are not usually published any more, probably because the details are similar to over 100 other such reports - it would be like describing a 'routine' case of appendicitis. Furthermore, there have apparently been no fatalities from confirmed torsade to date to my best knowledge, at least in the past 15 years (an American text states that it should have a mortality of zero). According to the Journal's web site these authors would have paid over US$500 to have this vignette made public.

If nothing more, the report should be a wake-up call to all who treat middle aged opioid dependent patients and remind us to take greatest care when anti-virals, anti-fungals, antibiotics, etc are being prescribed, especially by others who may not be familiar with the dependency side of their patient’s treatment. The size of the methadone dose appears to be of little relevance (Krantz’s original series had 6 of 9 subjects taking between 65mg and 125mg, mean 96mg daily; Pearson reports one case at 29mg daily).

Comments by Andrew Byrne ..

Link to article: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3021856/

Correction for Jones article summary: I stated in an early draft that there was no fetal loss but in fact there were two such events, both in the methadone group. This was promptly corrected on the web site and I apologise for the mistake. Dr Jones has also pointed out that the trend for greater attrition in the buprenorphine group did not reach statistical significance. To my mind the authors have still not justified their comment that buprenorphine is now a ‘first line option’. Just because they showed no difference in retention/attrition, this does not prove that the two treatments are equivalent therapeutically. I believe that both drugs are extremely useful in opioid dependency and that methadone remains the gold standard (O’Connor 2010 JAMA) although it does not suit everyone or every situation.

15 February 2011

Like randomised trials, patient choosing methadone also have much better retention.

The SUMMIT Trial: A field comparison of buprenorphine versus methadone maintenance treatment. Pinto H, Maskrey V, Swift L, Rumball D, Wagle A, Holland R. Journal of Substance Abuse Treatment, 2010 39;4:340-352

Dear Reader,

This naturalistic description and follow up study from a drug service in England shows comparative retention rates in opiate maintenance patients with exhaustive comparisons between the two groups. The current study followed a previous abortive attempt at randomisation of the groups which failed to attract any subjects, so determined were they all on their preferred medication.

These authors enrolled 361 eligible applicants for opioid maintenance of whom 227 (63%) chose methadone with the rest opting for (pure) buprenorphine (37%).

Despite the methadone patients having ‘more severe substance abuse and psychiatric and physical problems’ their retention rate was substantially higher than the buprenorphine patients (70% vs. 43% at 6 months).

Only seven of 134 patients swapped from buprenorphine to methadone early in the trial (apparently none swapped the other way around). Of the remaining buprenorphine drop-outs 10 were planned detox episodes; 40 failed to attend; 13 were detained by police; 10 were involuntary discharges. Relating to buprenorphine retention the authors report that the only significant difference on multivariate analysis in the many features/opinions solicited from the subjects was a ‘belief as to whether buprenorphine blocked the effect of heroin’. While this belief is not soundly based, it might be a helpful street myth, keeping some patients away from dangerous drugs. However, it could also give some users a false sense of security in a high risk overdose situation.

The trial patients were prescribed a mean maximum daily dose for buprenorphine of 12mg (r 4-20mg); and for methadone of 73mg (r 10-170mg). There were two deaths in the methadone group, one in the induction period and another while on a stable dose of methadone caused by overdose of multiple depressants.

As in most such longitudinal studies the proportion of subjects successfully reducing their dose to zero was low - around 7% for buprenorphine and under 1% for those choosing methadone during a six month period. This is consistent with other research showing about 4% of opioid dependent people become abstinent each year (and this is regardless of the type of treatment - see Thorley’s seminal work from the UK).

Crucial to the results of this trial was the finding that 10% of the total said that they would not have attended if methadone were the only drug available. The derived figure of buprenorphine patients who stated that they would not have attended for methadone in the absence of an alternative was 28%. This is a novel finding to my knowledge but fits with clinical experience in other areas that greater choice yields greater patient acceptance, better enrolments, compliance (adherence) and retention. Apparently English health authorities are required to justify the expense of each medication and so this provides strong evidence in that regard which was one of the aims of the authors (see their previous paper).

Since these authors used pure buprenorphine SL tablets readers should be cautious about extrapolating these findings to buprenorphine-naloxone (Suboxone) … it may or may not be the case that buprenorphine-naloxone would have achieved similar results to pure buprenorphine on which most of the fundamental research has been performed to date (RCT, safety, etc). Another reason to favour pure buprenorphine is that it is available in 0.4mg tablets (in Australia). In our own practice we have at least a quarter of our patients taking increments requiring 0.4mg tablets.

While there must be end-points to any research trial, it would be instructive to know how many of the drop-outs had re-registered in treatment (if any). While the group was not able to examine this issue, other research would indicate largely negative results including a high mortality. Dr Bell investigated this from official enrolment statistics in New South Wales some years ago [Bell J, Burrell T, Indig D, Gilmour S. Cycling in and out of treatment; participation in methadone treatment in NSW, 1990–2002. Drug and Alcohol Dependence 2006 81; 1: 55-61]. Such research, as well as clinical experience with buprenorphine shows that there is indeed a ‘revolving door’ or ‘frequent flyer’ syndrome with this treatment, at least in comparison to methadone.

Because there was no randomisation to the intervention, this study does not give us scientific information about the different drugs. However, it does inform us for the first time I believe that the most seriously dependent patients are more likely to choose methadone and that they have a better retention rate in so doing. Less seriously dependent patients are more likely to choose buprenorphine and are more likely to drop out of treatment (and some successfully withdraw from treatment). So unsurprisingly, outcomes are a function of both the patient group and the treatment chosen. While this information is not a major benefit in individual clinical decision making, this study at the very least provides an excellent “barometer” by which we might measure our own practices regarding the proportion taking each drug and the dose ranges used.

Comments by Andrew Byrne ..