Review of: Aaronson ST et al, Am J Psychiatry 2017;174(7):640–648

Treatment-resistant depression (TRD) is typically defined as a major depression that fails to remit after at least 2 trials of 2 different classes of antidepressants. Other than electroconvulsive therapy (ECT), there remain few evidence-based biological treatment options for TRD.

In 2005, the FDA approved vagus nerve stimulation (VNS), where a small stimulator device is surgically implanted in the chest, with three small electrodes wrapped around the vagus nerve. The device was originally approved for use with treatment-refractory epilepsy, although the approval was very controversial due to the poor quality of the data. TCPR was unconvinced that VNS had shown any evidence of being more effective for depression than sham treatment (see TCPR, January 2006). As a condition of approval, the FDA required post-marketing surveillance, and so the Treatment-Resistant Depression Registry was established.

The authors of this study, a 5-year longitudinal observational study conducted at 61 separate U.S. sites, used the registry to follow the clinical course and outcome of 2 large groups of patients diagnosed with TRD. One group received adjunctive VNS, and the other group received treatment as usual (TAU).

The patients could select their treatment—VNS or TAU—and 795 patients were included (495 patients in the VNS arm and 301 in the TAU arm). All patients had previously failed 4 or more treatments, with an average of 8.2 failed treatments. Response was defined as a decrease of > 50% in baseline Montgomery-Åsberg Depression Rating Scale (MADRS) score, and remission was based on a MADRS score ≤ 9.

The adjunctive VNS group had better clinical outcomes than the TAU group, including a significantly higher 5-year cumulative response rate (67.6% compared to 40.9%, p < 0.001), and a significantly higher remission rate (43.3% compared to 25.7%, p < 0.001).

TCPR’s Take

This study suggests that VNS is effective for TRD, but this treatment does not work quickly. Differences did not emerge until 6–9 months after treatment. Further, the study design had many limitations. Patients were not randomly assigned to treatment groups, there was no sham “placebo” comparison, and neither patients nor researchers were blinded to treatment. In addition, the study was funded by Cyberonics, the manufacturer of the device. While commercial funding does not necessarily imply that a study’s results are invalid, it does behoove us to give the results extra scrutiny.

VNS is a complicated surgical procedure requiring a large investment in both time and money. Whether the potential benefits are worth the costs must be weighed individually for each patient. This study does suggest, however, that VNS is a potentially useful treatment for a small group of patients with treatment-refractory depression.