Comparative efficacy and acceptability of antidepressant treatment in patients with post-stroke depression: a multiple-treatments meta-analysis

Abstract

BackgroundThe question of which antidepressants should be preferred for the treatment of post-stroke depression is controversial, and pairwise meta-analyses cannot provide a hierarchy of these interventions. We aimed to integrate the available evidence to create hierarchies of the comparative efficacy and acceptability of treatments.Methods We did a multiple-treatments meta-analysis (which uses both direct and indirect comparisons) of randomised controlled trials to compare antidepressants and placebo in the acute treatment of post-stroke depression. Electronic databases (MEDLINE, Embase, PsycINFO, Cochrane Central Register of Controlled Trials [CENTRAL], Web of Science) were searched up to Dec 31, 2016. We included only randomised controlled trials that compared antidepressants as monotherapy in patients with post-stroke depression. The patients had to have had a stroke diagnosed clinically or by CT scan or MRI and have been subsequently diagnosed with depression. We applied no inclusion restrictions on the basis of patient and study characteristics, such as age, sex, study dates, and class of antidepressants. The primary outcome was overall efficacy, defined as the mean change of the total depression score between baseline and 6 weeks. If 6-week data were not available, we used data from between 4 and 12 weeks (the datapoint closest to 6 weeks was given preference). The secondary outcome was acceptability, defined as risk of all-cause discontinuation. We tested for inconsistency with the design-by-treatment interaction model that provides a single inference, using the χ² test. This study has been registered with PROSPERO, number CRD42017054741. FindingsWe identified 12 suitable trials, with data from 707 participants. 11 antidepressants were assessed in the trials, including placebo. All drugs were significantly more effective than placebo apart from sertraline (odds ratio [OR] -0•61, 95% CI -1•47 to 0•25), nefiracetam (OR 0•51, -0•17 to 1•19), and fluoxetine (OR 0•46, -0•35 to 1•27). For efficacy, standardised mean differences, compared with placebo, varied from -6•54 (95% CI -8•42 to -4•65) for the best drug (reboxetine) to 0•51 (-0•17 to 1•19) for the worst drug (nefiracetam). Reboxetine, paroxetine, and doxepin were the most efficacious treatments (ranked in order), the cumulative probabilities of which were 100%, 85•7%, and 83•2%, respectively. For acceptability, ORs compared with placebo ranged from 0•09 (95% CI 0•00 to 1•83) for paroxetine to 3•42 (0•73 to 15•91) for citalopram. Patients assigned to paroxetine had significantly lower all-cause discontinuation than those assigned to doxepin (OR 0•04, 95% CI 0•00 to 0•73), citalopram (OR 0•03, 0•00 to 0•78), and fluoxetine (OR 0•04, 0•00 to 0•89); all other comparisons were not significant. Paroxetine (cumulative probability 92•4%), placebo (63•5%), and sertraline (57•3%) were the most acceptable treatments. Comparison-adjusted funnel plots for efficacy and acceptability were largely symmetrical. For acceptability (all-cause discontinuation), we found no e...

Comparison of Direct Endovascular Treatment Versus Thrombolysis and Rescue-endovascular Treatment for Patients with M1/M2 Occlusion: A Real-life Retrospective Study

“…Anxiety is a common problem in stroke patients, also after effective treatment. Up to 30% are affected [11]. This study did not put contrast on the treatment for anxiety and depression.…”