Outpatient rotations to methadone

Cancer has published a paper about rotation to methadone in the outpatient setting for cancer pain. It's a notable paper because very little has been published about this practice (although I think it's widespread in the hospice and palliative care world) - most of the published reports (mostly case series/retrospective chart reviews) have been in inpatients, where close monitoring for safety (and efficacy) can occur. So, while people are doing it, there hasn't been much support in the research literature about, at the very least, the safety of this practice.

This study is a retrospective chart review of a single center's (MD Anderson) experience with this. The paper includes data about patients initiated on methadone (as the first strong opioid), but I'll focus on the rotation data here. The patients (N=89, mean age 58, median baseline oral morphine equivalent daily dose [MEDD] of 100 mg) were rotated to methadone as outpatients, and had median first follow up at 13 days, and second at 37 days. The rotation protocol is not specified in great detail (and they imply there is not a set protocol for this group - instead more of general parameters that are adjusted based on the physician's discretion). It seems generally a start/stop strategy (discontinue prior opioids, initiate methadone the same day) was used. Morphine to methadone ratios that were used were 5:1 for MEDD less than 90 mg/day; 8:1 for 91 to 300 mg/day; and 12:1 for MEDDs over 301 mg/day. What is done with the breakthrough medication is not specified. It's important to note that the range of baseline MEDD was 60-185 mg: these patients were being rotated to methadone at moderate morphine doses, not once patients were taking many hundreds (or thousands) of mg a day.

This was a retrospective chart review, of one group's real-life practice, and the criteria for evaluating the reason for rotation to methadone, as well as its success, were based on the research team's best efforts at abstracting from chart data. They tried to identify why patients were rotated (inadequate analgesia, non-methadone opioid side effects, or both) and then if the rotation was successful upon follow-up (pain improvement greater than 30% or 2/10 on a 11-point rating scale, reduction in the side effects which prompted the rotation, etc.). If these targets were clearly met the rotation was labeled as success; if the methadone was discontinued, the patient was admitted to the hospital for pain or methadone related side effects, or if the patient was lost to follow-up, the rotation was labeled a failure. Everything else was labeled a 'partial success' (indicating, at least, the patient tolerated the rotation adequately to continue methadone and not be admitted with side effects).

The major data on success were positive: 47% had a 'complete success', 38% a partial success, and 15% a 'failure.'*** The best rates of success were for patients rotated only for side effects, although this was only 5 patients. Median methadone doses were 15 mg/day at the first follow up and 18 mg/day at the 2nd. They did plot baseline MEDD with stable methadone dose, and confirmed the previous findings that baseline MEDD is associated with MEDD:methadone ratio (e.g. the higher the baseline MEDD, the higher the ratio, and the lower relative amount of methadone patients need). And in fact the median MEDD:methadone ratio they identified after patients were stable were 5:1 and 8:1 for MEDDs less than 90, and greater than 90 mg daily, respectively (which is of course their conversion ratio, suggesting that these are in fact reasonable ratios for patients with MEDDs less than 200 mg).

Most interesting is that they calculated MEDD:methadone ratios for patients who were rotated for pain vs. those rotated for side effects (with or without pain). In the 'side effect' patients the MEDD:methadone ratio was 9:1, and 6:1 for the 'pain' patients (ie the 'side effect' patients needed less methadone, relative to their baseline opioid dose, than the 'pain' patients).

In discussing the failures, they indicate these were due to non-efficacy, and don't discuss any toxicities.

All of this is good, and represents some of the first data (with a decent N for a study of this type) supporting the safety of outpatient methadone rotations for pain. I doubt a prospective study (even just a natural history one, with modest outcomes of safety at one month) is coming anytime soon. It's confirmatory of the broad trend in previously published chart reviews that methadone rotations are usually (for over 3/4 patients) helpful, although rigorously controlling for the reason for methadone rotation has never occurred and all the data we have is essentially descriptive and based on real-life clinician practice (which has both benefits and drawbacks in interpreting this).

The major caveat here is that this is for patients on moderate MEDDs - less than 200 mg a day. One cannot conclude from this that the practice is safe/advisable for patients on significantly higher doses (ie in the 500 mg and up range). Which raises the question of whether we should routinely be rotating patients to methadone once they're on, say, MEDDs of 100 mg or more, and who have uncontrolled pain or side effects. This research suggests only, of course, that the rotation to methadone in this setting can be successful, not that the patient was better off for it (as opposed to continuing up-titration of the prior opioid assuming the rotation wasn't for side effects, or rotating to a different opioid such as fentanyl). Trying to answer this in a trial would be difficult, but not impossible, and you do kind of wish there was a methadone lobby to support such research. My guess is that there's a wide variation of practice out there, and I'd be curious as to what others do - when, why you switch to methadone - a simple mg cut off (ie propose it to a patient once they're at a certain MEDD regardless of how they're doing) or later? I tend to do it later, but am not convinced that is best: we have no data either way.

(***Addendum Jan 29, 2010: I had originally put in incorrect numbers for the % of complete or partial successes for the rotation; a commenter pointed this out and I've changed the numbers here for posterity. See the comments.)