Clinical response and facial emotion recognition following repetitive transcranial magnetic stimulation in treatment-resistant bipolar depression

INTRODUCTION: Repetitive transcranial magnetic stimulation (rTMS) is increasingly used in treatment-resistant depression, but its relationship with facial emotion recognition in bipolar depression remains unclear. This study examined clinical outcomes and changes in facial emotion recognition observed over an rTMS treatment course in patients with treatment-resistant bipolar depression. METHODS: We conducted a naturalistic observational study using pseudonymized routinely collected clinical data from adults with bipolar depression who underwent rTMS. Depressive symptoms, baseline neurocognitive measures, and facial emotion recognition outcomes were analyzed. Between-group comparisons were performed for responders versus non-responders and remitters versus non-remitters. Spearman correlations examined associations between HDRS-17 reduction and facial emotion recognition changes. Because multiple emotion-specific correlations were tested, false discovery rate correction was applied. RESULTS: Among the included patients, 116 were classified as clinical responders and 67 as non-responders. Baseline characteristics were similar between groups, except that responders showed higher baseline fear recognition accuracy. Across the sample, facial emotion recognition accuracy increased and reaction times decreased over the treatment course, with the largest accuracy gain observed for sadness recognition. However, improvements in facial emotion recognition were not significantly associated with reductions in depressive symptoms after correction for multiple comparisons. In exploratory logistic regression analyses, higher post-treatment sadness recognition accuracy was associated with lower odds of clinical response, although the overall model showed limited explanatory power. CONCLUSION: In this naturalistic observational sample of patients with treatment-resistant bipolar depression receiving rTMS, improvements in facial emotion recognition were observed over the treatment course, but these changes were not clearly associated with clinical response. Controlled studies are needed to determine whether these changes are attributable to rTMS and to clarify their clinical significance.