First Do No Harm … Reduction? FREE

Illicit opioid use exacts a tremendous toll in health and human suffering. In the United States, heroin and prescription opioid abuse and dependence affect more than 5 million persons, at great human and societal cost (1). Addiction is now understood to be a chronic relapsing condition with deep physiologic roots, and pharmacotherapies that effectively treat opioid dependence are acknowledged to be first-line treatment for this often devastating disorder (2).

Methadone is the medication most widely used to treat opioid dependence. Its efficacy has been established in diverse settings and across multiple outcomes, including reductions in mortality, HIV infection, incarceration, and costs. Extensive federal regulations govern methadone treatment, which is delivered to approximately 260 000 persons nationally by a heterogeneous mix of more than 1100 methadone treatment programs (3). Although mounting evidence supports integrating methadone treatment with medical and mental health services, most treatment programs provide minimal on-site medical care (4).

Two significant issues concerning methadone safety have emerged in recent years: An increase in methadone-related deaths and evidence that methadone is associated with prolongation of the QT interval. The rise in methadone-associated deaths seems in large measure to reflect multisubstance overdoses among persons who access methadone that was prescribed to treat pain, which reflects its growing use as an analgesic outside of methadone maintenance treatment programs (5). Although evidence of methadone's possible effect on the QT interval has been accumulating, it has remained unclear how to incorporate such information into clinical practice. To address this issue, the federal Center for Substance Abuse Treatment convened an expert panel. The findings and recommendations of some of the panel members are published in this issue (6). (Of note, federal guidelines related to the recommendations are currently undergoing field testing and may be refined.)

Using established criteria for determining the likelihood that a medication and side effects are causally related (7), the panel assembled data that ranged from experimental evidence of biological plausibility (methadone inhibits the hERG ion channel associated with QT prolongation) to prospective studies that demonstrate QT prolongation among patients who begin methadone treatment. In the aggregate, the data provide convincing evidence that methadone causes modest prolongation of the QT interval in many patients, with a small minority (approximately 2%) crossing a threshold associated with significantly increased risk for arrhythmia. The panel's summary of reports of torsade de pointes and sudden death in patients treated with methadone indicates that the incidence of these events is low but not zero and typically occurs among persons who receive exceptionally high doses of methadone or have other risk factors for dysrhythmia.

A central question thus emerges: Does electrocardiographic (ECG) screening have a role in methadone treatment? Here, the panel members wade into controversial territory by recommending that physicians obtain ECGs before treatment initiation, at 30 days, and annually thereafter in all patients receiving methadone treatment. Unfortunately, this suggested guideline ventures well beyond the evidence presented. The core issue in evaluating a screening test is whether the expected benefits of its widespread implementation outweigh the expected harms (8). However, the panel members had no trials or decision models to assist them in defining the relative benefits and harms of ECG screening. Rather than using their literature review to frame a process for acquiring such data, the panel members determined that a clinical practice guideline was in order and drafted and completed it during a 2-day meeting. Although other recent analyses of this topic have not recommended routine ECG screening (9 - 10), the panel members do not state why they reached a different conclusion.

Several key points demand consideration before routine ECG screening can be endorsed. Although many medications can cause QT prolongation, published trials documenting that torsade de pointes and sudden death can be averted by baseline ECG screening are lacking. The literature is replete, on the other hand, with examples of screening tests that seemed sensible when recommended yet whose disadvantages, once carefully considered, were found to negate if not outweigh associated benefits (11). Significant harms potentially associated with routine ECG screening of all patients entering methadone treatment include unnecessary evaluation, anxiety, treatment and costs associated with false-positive findings (red herrings), costs of implementation (which in many systems will be borne by patients), and the potential wrongful determination that some persons should not receive sorely needed opioid-dependence treatment. For the typical person who begins methadone treatment, the risk for death from torsade de pointes is likely to be substantially lower than that from competing causes of mortality associated with untreated opioid addiction (12 - 13). Although the panel members point out that buprenorphine may be a safer alternative for the estimated 2% of patients receiving methadone who are anticipated to benefit most from ECG screening, this newer medication is rarely prescribed by traditional methadone treatment programs and remains difficult to access in some areas (3). Some patients who have finally mustered the motivation to begin methadone maintenance only to be referred elsewhere for alternative treatment after an abnormal ECG will inevitably return to opioid use.

If ECG screening were demonstrated to be beneficial, important related questions require clarification through decision analysis studies or clinical research. The panel members provide no evidence to justify their selection of screening frequency (at baseline, 30 days, and annually) over alternate strategies, such as once after 30 days. Nor do we know whether baseline screening should be restricted to persons with existing risk factors (history of structural heart disease, arrhythmia, or syncope or use of other medications that may also prolong the QT interval). The merits of selecting a methadone dose of 100 mg/d as a trigger for more frequent ECG testing also requires more careful consideration. Methadone is too often prescribed at lower-than-effective doses; although this is improving, more than half of programs in the United States still typically offer doses below the mean effective range of 80 to 100 mg/d (14). Conveying to patients and programs that a 100-mg dose is associated with elevated risk may exert new downward pressure on dose levels and risk poorer treatment outcomes. Selection of an optimal cut-point for more intensive monitoring should be driven by modeling of the risks and benefits of alternative approaches.

In short, plausible arguments can be made for and against routine ECG screening in methadone treatment. But before guidelines are implemented, research methods, such as decision analysis, should be brought to bear to permit clear appreciation of the tradeoffs, benefits, and harms of alternative screening strategies—as has been done in similar circumstances (15). For too long, evidence-based approaches to structuring the delivery of methadone maintenance have been the exception (16). That situation is changing. This momentum must continue.

Hippocrates wrote, “As to diseases, make a habit of two things—to help, or at least to do no harm” (17). Before we adopt guidelines to detect QT prolongation and potentially prevent sudden death among patients receiving methadone, we need more evidence to help us walk the fine line between minimizing the harms of treatment and failing to treat for fear of causing harm.

Marc N. Gourevitch, MD, MPH

New York University School of Medicine

New York, NY 10016