Antidepressants decrease symptoms of IBS; benefit of psychological therapies is less clear

Reference

Ford AC, Lacy BE, Harris LA, Quigley EMM, Moayyedi P. Effect of antidepressants and psychological therapies in irritable bowel syndrome: an updated systematic review and meta-analysis. Am J Gastroenterol 2019;114(1):21-39.

Synopsis

These authors systematically searched several databases, a clinical trial registry, and abstracts of conference proceedings to identify randomized trials of antidepressants or psychological therapies in the treatment of adults with IBS. The trials of antidepressants had to include a placebo comparator, while the psychological therapy trials could include placebo, symptom monitoring, or usual care. Two of the authors independently evaluated papers for inclusion and assessed the risk of bias. They resolved disagreements by consensus, and ultimately included 53 trials. The studies ranged in size from 15 to 172 patients; 17 studies compared antidepressants with placebo, 35 compared psychological therapies with control therapy or usual care, and 1 compared both psychological therapy and antidepressants with placebo. The main outcome was a dichotomous assessment of any improvement in global IBS symptoms or abdominal pain. The studies that evaluated drugs included 1127 patients. Only 4 were at low risk of bias. Overall, 57% of actively treated patients had global improvement compared with 34% of placebo-treated patients (number needed to treat [NNT] = 5; 95% CI 4 - 6). Additionally, 52% of actively treated patients had improvement in abdominal pain compared with 27% of placebo-treated patients (NNT = 4; 3 - 7). There was statistically significant heterogeneity among these outcomes. Most of the improvement in abdominal pain was from tricyclic antidepressants. Only 8 studies reported on adverse events, finding that 36% of actively treated patients had adverse events compared with 21% of placebo-treated patients (number needed to treat to harm = 9; 5 - 21). The studies that evaluated psychological therapies included 2487 patients and assessed several treatments including cognitive behavioral therapy (CBT), relaxation therapy, hypnotherapy, Internet-based CBT, psychotherapy, mindfulness-based meditation, stress management, or various combinations of these. Other than a lack of masking, the authors report these studies were generally at low risk of bias. Overall, 48% of patients receiving psychological therapies had global improvement compared with 24% of control patients (NNT = 5; 4 - 5). Among individual therapies, the following were effective: CBT (NNT = 4; 3 - 9), relaxation (NNT = 6; 3 - 60), multicomponent therapy (NNT = 4; 3 - 7), hypnotherapy (NNT = 5; 3.5 - 10), and dynamic psychotherapy (NNT = 4; 2 - 20). Overall, the authors found significant heterogeneity as well as evidence for publication bias among the studies, so the final word is not out!