Every third to sixth patient with medical diseases receives antidepressants, but regulatory trials typically exclude comorbid medical diseases. Meta-analyses of antidepressants have shown small to medium effect sizes, but generalizability to clinical settings is unclear, where medical comorbidity is highly prevalent. To perform an umbrella systematic review of the meta-analytic evidence and meta-analysis of the efficacy and safety of antidepressant use in populations with medical diseases and comorbid depression. PubMed and EMBASE were searched from inception until March 31, 2023, for systematic reviews with or without meta-analyses of randomized clinical trials (RCTs) examining the efficacy and safety of antidepressants for treatment or prevention of comorbid depression in any medical disease. Meta-analyses of placebo- or active-controlled RCTs studying antidepressants for depression in individuals with medical diseases. Data extraction and quality assessment using A Measurement Tool for the Assessment of Multiple Systematic Reviews (AMSTAR-2 and AMSTAR-Content) were performed by pairs of independent reviewers following PRISMA guidelines. When several meta-analyses studied the same medical disease, the largest meta-analysis was included. Random-effects meta-analyses pooled data on the primary outcome (efficacy), key secondary outcomes (acceptability and tolerability), and additional secondary outcomes (response and remission). Antidepressant efficacy presented as standardized mean differences (SMDs) and tolerability (discontinuation for adverse effects) and acceptability (all-cause discontinuation) presented as risk ratios (RRs). Of 6587 references, 176 systematic reviews were identified in 43 medical diseases. Altogether, 52 meta-analyses in 27 medical diseases were included in the evidence synthesis (mean [SD] AMSTAR-2 quality score, 9.3 [3.1], with a maximum possible of 16; mean [SD] AMSTAR-Content score, 2.4 [1.9], with a maximum possible of 9). Across medical diseases (23 meta-analyses), antidepressants improved depression vs placebo (SMD, 0.42 [95% CI, 0.30-0.54]; I2 = 76.5%), with the largest SMDs for myocardial infarction (SMD, 1.38 [95% CI, 0.82-1.93]), functional chest pain (SMD, 0.87 [95% CI, 0.08-1.67]), and coronary artery disease (SMD, 0.83 [95% CI, 0.32-1.33]) and the smallest for low back pain (SMD, 0.06 [95% CI, 0.17-0.39]) and traumatic brain injury (SMD, 0.08 [95% CI, -0.28 to 0.45]). Antidepressants showed worse acceptability (24 meta-analyses; RR, 1.17 [95% CI, 1.02-1.32]) and tolerability (18 meta-analyses; RR, 1.39 [95% CI, 1.13-1.64]) compared with placebo. Antidepressants led to higher rates of response (8 meta-analyses; RR, 1.54 [95% CI, 1.14-1.94]) and remission (6 meta-analyses; RR, 1.43 [95% CI, 1.25-1.61]) than placebo. Antidepressants more likely prevented depression than placebo (7 meta-analyses; RR, 0.43 [95% CI, 0.33-0.53]). The results of this umbrella systematic review of meta-analyses found that antidepressants are effective and safe in treating and preventing depression in patients with comorbid medical disease. However, few large, high-quality RCTs exist in most medical diseases.
Read full abstract