METHODS: Electronic databases were searched up to March 2024 for randomized controlled trials (RCTs) recruiting adults with FD. Data of overall symptoms improved between the antidepressants and placebo groups was pooled to obtain risk ratio (RR) employing the random-effects model. The effect of random errors was evaluated with TSA. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was used to assess the certainty of evidence. Analyses were performed using STATA version 16.0.

RESULTS: Nine RCTs with 924 patients met the eligible criteria. The RRs of FD symptoms improving with any antidepressants, tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors were (n = 9, RR = 1.30 [95% CI, 1.02-1.67]), (n = 5, RR = 1.41 [95% CI, 1.07-1.85]) and (n = 2, RR = 0.97 [95% CI, 0.72-1.29]), respectively. TSA demonstrated conclusive evidence for the beneficial effect of TCAs. The number needed to treat (NNT) with any depressants and TCAs were 11 (95% CI, 7-36) and 6 (95% CI, 4-15), respectively. The certainty of the evidence for an effect of TCAs was that of moderate GRADE quality. The benefit, however, was limited to the western population (n = 3, RR = 1.43 [95% CI, 1.04-1.96]) and did not extend to the Asian population (n = 2, RR = 1.32 [95% CI, 0.75-2.32]). Conversely, antidepressant-using patients experienced adverse events more frequently. However, no statistically significant association was found between TCAs and any adverse events (n = 3; RR = 1.36 [95% CI, 0.91-2.04]).