Opioid addiction agonist therapy and the QT prolongation phenomenon: state of the science and evolving research questions

Opioid agents have been used for addiction treatment for more than half a century. Levomethadyl acetate (LAAM) and methadone have been associated with risks of QT interval prolongation and arrhythmias, but little consensus exists regarding the strategy for their management. Critical studies assessing the costs and benefits of electrocardiographic screening are needed. In 1990, Monahan et al. reported that the widely used antihistamine, terfenadine (Seldane), prolonged the duration of the re-polarization phase of the electrocardiogram (ECG), represented by the QT interval, and caused the life-threatening ventricular arrhythmia known as ‘torsades de pointes’ (TdP) 1. Concern regarding the cardiac effects of non-cardiac drugs spread, resulting in increased scrutiny of ECG data and reports of arrhythmias. In 2001, Deamer et al. published a case series of TdP associated with high-dose levomethadyl acetate (LAAM) 2. In 2002, Krantz et al. published a case series of 17 individuals with methadone-associated TdP 3. In-vitro studies by Katchman et al. found that LAAM and methadone are unique among opioids in their tendency to block a critical potassium conductance at concentrations approaching those achieved clinically 4. Subsequent studies have confirmed that methadone can cause TdP and significantly prolongs the QT interval 5-9. In a prospective investigation of 177 consecutive cases of sudden death, 72 (41%) were found to have detectable methadone on toxicology examination, while only 12 of 177 subjects (7%) had other opioids, suggesting that methadone is associated with a particular propensity for sudden death 10. Most recently, data from the Drug Abuse Warning Network (DAWN) database show that, on a morphine-equivalent adjusted basis, methadone is 50–100 times more likely to result in death than buprenorphine, an opioid with little evidence of QT prolongation 11. Questioning the importance of QT prolongation, Anchersen and colleagues from Norway examined the prevalence of QT prolongation in treated addicts as well as the causes of death in 90 of 2382 subjects followed-up on out-patient methadone 12. The authors confirmed that there is a dose-related increase in the heart rate-correct QT (QTc) in methadone-treated individuals, and found that 4% had a QTc greater than 500 ms, a level considered to be associated with significant TdP risk. Despite finding QTc prolongation, they concluded that post-mortem examination or clinical history excluded TdP as the cause of death in all but four of 90 individuals. However, while methadone may have a low TdP incidence in Norway, the same cannot be said of the United States, where it is the most commonly reported drug associated with TdP in the Food and Drug Administration database between 2004 and 2011 13. Based on the preceding considerations, we believe that methadone presents significant arrhythmic risk, but agree that legitimate questions remain regarding the best way to manage these risks. Krantz et al. have previously recommended a harm-reduction strategy focusing on informed consent, risk evaluation and universal ECG-guided decision-making when prescribing methadone 14. These recommendations reflect the standard approach used by cardiologists when prescribing QT-prolonging drugs 15, but were rejected by the US Center for Substance Abuse Treatment (CSAT). Instead of advocating universal ECG screening, a preferred approach was based on identifying ‘high-risk’ individuals using the patient's historical data 16. Under this approach, screening ECGs would be reserved for those ‘with significant risk factors for QT prolongation, including a history of cardiac arrhythmia or prolonged QT interval; symptoms suggestive of arrhythmia, such as episodes of syncope, dizzy spells, palpitations, or seizures; medication history; family history of premature death or any other historical information suggestive of a possible cardiac arrhythmia’. 16 This is an approach that has never been implemented, to our knowledge, and which assumes that clinicians can identify subjects at risk for TdP by history; it was not favored by the majority of the committee. Against universal ECG screening, the dissenters raised ‘concerns over the level of resources involved in implementing routine ECGs and the absence of clear evidence for the effectiveness of the practice in achieving meaningful reductions in methadone-associated cardiac events’. In our opinion, raising concerns about the costs of ECG screening is perfectly fair, but requiring that ‘clear evidence of … reductions in methadone-associated cardiac events’ be established before advocating a practice intended to reduce iatrogenic events is not. A randomized controlled trial powered to detect the impact of ECG screening on the incidence of TdP (or mortality) would need thousands of patients 17 and, arguably, would be unethical based on the present standard of care for the management of QT-prolonging drugs. It would be well to remember that no randomized, controlled trial testing whether methadone reduces mortality has been performed, nor would it be humane to attempt such a study. Requiring ECG screening to satisfy a level of evidence higher than that required for the use of methadone is disingenuous. Most importantly, we need a study testing whether ECG screening can be performed in a manner that is acceptable to patients, affordable and has an impact on their management. If it is the case that requiring a screening ECG results in refusal of therapy or has no effect on subsequent treatment, then this approach, which at present is our best option, is effectively foreclosed. In particular, we need to know whether ECG-guided dose reduction, substitution with buprenorphine or use of the R-isomer of methadone shortens the QT interval and reduces adverse events. While a study that was powered to evaluate mortality would be impossibly large, a more modest study using implanted ECG monitors (which have battery lives lasting 3 years and record all significant arrhythmias) could be performed in a far smaller population. It is estimated that TdP is fatal in only 10% of cases, and a primary end-point of asymptomatic or non-lethal arrhythmias would require a fraction of the number of subjects. Ultimately, of course, one would really like to know how many deaths are due to arrhythmias versus respiratory depression or other factors. This question seems unlikely to be answered satisfactorily until a study using ambulatory ECG monitoring is performed or a pathological finding specific for arrhythmic death is found. Until then, one must keep an open mind, examining with skepticism overstated claims of both the benefits of ECG screening and the safety of methadone therapy. Studies examining the feasibility, cost and utility of ECG screening need to be performed. The authors have no conflicts of interest to declare.