BACKGROUND: The need for effective, long-term treatment for recurrent or chronic, treatment-resistant depression is well established. METHODS: This naturalistic follow-up describes outpatients with nonpsychotic major depressive (n = 185) or bipolar (I or II) disorder, depressed phase (n = 20) who initially received 10 weeks of active (n = 110) or sham vagus nerve stimulation (VNS) (n = 95). The initial active group received another 9 months, while the initial sham group received 12 months of VNS. Participants received antidepressant treatments and VNS, both of which could be adjusted. RESULTS: The primary analysis (repeated measures linear regression) revealed a significant reduction in 24-item Hamilton Rating Scale for Depression (HRSD(24)) scores (average improvement, .45 points [SE = .05] per month (p < .001). At exit, HRSD(24) response rate was 27.2% (55/202); remission rate (HRSD(24) < or = 9) was 15.8% (32/202). Montgomery Asberg Depression Rating Scale (28.2% [57/202]) and Clinical Global Impression-Improvement (34.0% [68/200]) showed similar response rates. Voice alteration, dyspnea, and neck pain were the most frequently reported adverse events. CONCLUSIONS: These 1-year open trial data found VNS to be well tolerated, suggesting a potential long-term, growing benefit in treatment-resistant depression, albeit in the context of changes in depression treatments. Comparative long-term data are needed to determine whether these benefits can be attributed to VNS.