Can cardiac resynchronization therapy improve cognitive function? A systematic review.

INTRODUCTION Cognitive impairment (CI) comprises a measurable deficit of different cognitive domains (memory, attention, problem solving, and motor speed), and has a high prevalence among congestive heart failure (CHF) patients. Only a few pilot studies have investigated the effects of cardiac resynchronization therapy (CRT) on cognitive performance. The purpose of this systematic review is to outline and evaluate results of published studies that assess the impact of CRT on neuropsychological function in CHF. METHODS Electronic databases were searched for articles containing the following terms: CRT, cognition, cognitive, and neurocognitive. A data extraction was performed according to our objective from each study. Effect sizes were computed using Hedges' g. The within-group formula was used for cohort studies with a pre-post design, while the between-group formula was used for studies that compared independent groups. Multiple outcomes were combined in domain-specific synthetic scores as well as in a global score for each study, and a fixed-effect model was used to estimate the summary effects. RESULTS Only three studies met criteria for inclusion in the analysis. The results of these studies were discordant and methodological limitations were identified. The meta-analysis of cognitive outcomes showed a nonsignificant overall effect (Hedges' g = 0.131, 95% confidence interval: -0.16 to 0.422), while the summary effects on executive functioning and attention reached statistical significance (Hedges' g = 0.374, 95% confidence interval: 0.085-0.662 and Hedges' g = 0.343, 95% confidence interval: 0.051-0.635, respectively). CONCLUSION CI and related negative consequences have been largely documented in patients with heart failure but very few studies have assessed the plausible benefits of CRT on patients' cognitive function. Despite the statistical significance of the domain-specific pooled effects, their validity and clinical relevance is lacking due to methodological limitations.