Bupropion confirmed not to cause weight gain in a real-world clinical population

COMMENTARY ON: Blumenthal SR, Castro VM, Clements CC, et al. An electronic health records study of long-term weight gain following antidepressant use. JAMA Psychiatry 2014;71:889–96.

What is already known on this topic

The risk of weight gain is a major concern for patients initiating antidepressant treatment. However, whether specific medications are more or less likely to cause weight gain is unclear. The association is not a simple cause–effect relationship since many factors can contribute to weight gain, including depression itself.¹ A 2010 systematic review suggested that amitriptyline, mirtazapine and paroxetine were associated with a greater risk of weight gain than other antidepressants.² In contrast, some weight loss was found to occur with fluoxetine and bupropion. Weight loss associated with bupropion was observed during acute and maintenance treatment whereas that for fluoxetine was seen only during acute treatment.² If weight loss occurs during long-term bupropion treatment, this medication may be a preferred option for some patients. However, reflecting the broader literature, most studies in this systematic review were short-term trials (4–12 weeks of follow-up). There were only three bupropion trials with >12 weeks of follow-up and only two of these were depression trials. Only one study was rated as having ‘good’ quality. Generalisability to real-world settings and realistic durations of treatment are therefore questionable.

Methods of the study

Blumenthal and colleagues report a pharmacoepidemiological analysis using electronic health records from the Partners HealthCare electronic record, a database containing data from hospitals as well as from community and specialty outpatient clinics that comprise the Partners HealthCare system in Boston, Massachusetts. New users of various antidepressants (defined as those without prior recorded treatment) between the ages of 18 and 65 were identified. Entries recording body weight, from routine assessments made during clinical visits, were also identified. A mixed model was used to estimate the rate of weight gain over 1 year, with citalopram serving as a reference group. The models adjusted for baseline weight, change in weight over 6 months preceding the date of antidepressant exposure, age, sex, race/ethnicity, public versus private insurance, diagnosis, psychotherapy and antipsychotic treatment. The analysis sought to determine whether various medications were associated with weight changes. Statistical tests compared weight changes associated with each medication to citalopram. By using real-world data and a relatively long follow-up interval, this study addressed a gap in the prior literature.

What this paper adds

• Bupropion, nortriptyline and amitriptyline were associated with significantly less weight gain than citalopram.

• The magnitude of change was modest. Linear predictions from the mixed models suggested weight gain of approximately 0.25–0.75% for most antidepressants, whereas that for bupropion was near zero.

Limitations

• The electronic health records database included over four million patients, with over 480 000 patients having an antidepressant prescription. However, less than 20 000 were included in the study. An eligibility criterion required that patients be prescribed a sufficient quantity of antidepressant to last at least 3 months. This may have resulted in exclusion of patients who discontinued their antidepressant soon after its initiation due to weight gain.

• Amitriptyline and nortriptyline were used to treat pain more often than depression. Such indication typically uses lower doses than depression. Dosages were not addressed in this study. The results do not indicate that amitriptyline or nortriptyline causes less weight gain than citalopram when used at antidepressant dosages.

What next in research

• This study, and many prior ones, tried to isolate the independent effect of various antidepressants from the effect of various other potential determinants of weight gain. This approach can help to inform the selection of antidepressant treatment. The literature is consistent in identifying bupropion as an antidepressant that is less likely to cause weight gain than other antidepressants.² However, there appears to be considerable heterogeneity in the effects of these medications. Rather than attempting to control for various determinants of weight gain, future studies should explore this heterogeneity by examining interactions between various medications and other aspects of the treatment context (eg, indication for use, effectiveness, lifestyle factors, dosage and history).

Do these results change your practices and why?

The finding suggesting that nortriptyline and/or amitriptyline may cause less weight gain than citalopram is unlikely, in isolation, to change prescribing practice. The literature has been inconsistent on such comparisons, but the bulk of evidence indicates that these medications are associated with substantial weight gain. While this study suggests the rate of weight gain may be less with these medications than with citalopram, weight gain was still evident (approximately 0.25% of body weight over 12 months) for amitriptyline and nortriptyline. The magnitude of difference in weight gain, depending on duration of use, may or may not be clinically significant. The indication for use of amitriptyline and nortriptyline was often for pain, suggesting that the use was probably at low dosages.

The finding that weight gain does not occur with bupropion helps confirm that this feature of the medication can be factored into risk–benefit decisions and can play a role in the selection of antidepressant medications. Additional considerations may include the efficacy profile of alternative medications and adverse effects other than weight gain. Given the small differences between medications, considerations such as lifestyle modification and dietary counselling may ultimately be more important than the choice of medication.