Hybrid systematic review and network meta-analysis of randomized controlled trials of interventions for depressive symptoms in patients with coronary artery disease

OBJECTIVE: Depression is common in patients with coronary artery disease (CAD) and is associated with poor outcomes. Although different treatments are available, it is unclear which are best or most acceptable to patients, so we conducted a network meta-analysis of evidence from randomized controlled trials (RCTs) of different depression treatments to ascertain relative efficacy. METHODS: We searched for systematic reviews of RCTs of depression treatments in CAD and updated these with a comprehensive search for recent individual RCTs. RCTs comparing depression treatments (pharmacological, psychotherapeutic, combined pharmacological/psychotherapeutic, exercise, collaborative care) were included. Primary outcomes were acceptability (dropout rate) and change in depressive symptoms 8-weeks post-treatment commencement. Change in 26-week depression and mortality were secondary outcomes. Frequentist, random effects network meta-analysis was used to synthesize the evidence and evidence quality was evaluated following GRADE recommendations. RESULTS: Thirty-three RCTs (7240 participants) provided analyzable data. All treatments were equally acceptable. At 8-weeks, combination therapy (1 study), exercise (1 study), and antidepressants (10 studies) yielded the strongest effects versus comparators. At 26-weeks, antidepressants were consistently effective, but psychotherapy was only effective versus usual care. There were no differences in treatment groups for mortality. GRADE ratings ranged from very low to low. CONCLUSIONS: Overall, the evidence was limited and biased. While all treatments for post-CAD depression were equally acceptable, antidepressants have the most robust evidence base and should be the first-line treatment. Combinations of antidepressants and psychotherapy, along with exercise, could be more effective than antidepressants alone, but require further rigorous, multi-arm intervention trials. SYSTEMATIC REVIEW REGISTRATION: CRD42018108293 (PROSPERO)