Abstract 4372843: Cognitive Impairment in Heart Failure: A Systematic Review and Meta-Analysis of Prevalence and Predictors

Background: Cognitive impairment (CI) is increasingly recognized as a critical comorbidity in patients with heart failure (HF), contributing to medication nonadherence, readmissions, and mortality. However, its true prevalence and determinants remain unclear. We conducted a systematic review and meta-analysis to quantify the prevalence of CI in HF and identify key clinical predictors. Methods: We followed PRISMA guidelines and searched PubMed, EMBASE, Scopus, and Cochrane databases through April 2024 using terms: (“cognitive dysfunction” OR “cognitive impairment”) AND (“heart failure” OR “cardiac dysfunction”). Studies were included if they: (1) enrolled adult HF patients; (2) assessed CI using validated tools (e.g., MoCA, MMSE); and (3) reported prevalence or correlations with clinical variables. Two reviewers independently extracted data and assessed risk of bias. Pooled prevalence estimates and odds ratios (ORs) were calculated using a random-effects model (DerSimonian–Laird method). Results: Of 1,147 screened articles, 26 met inclusion criteria (n=6,328 HF patients; mean age 69±9 years). The pooled prevalence of CI in HF was 42% (95% CI: 36–48%, I = 72%). Subgroup analysis revealed higher prevalence in HF with reduced EF (HFrEF) (48%) compared to preserved EF (HFpEF) (35%, p<0.01). Severe LV dysfunction (LVEF <30%) was associated with increased risk of CI (OR 1.92; 95% CI: 1.43–2.59), as were elevated NT-proBNP (OR 1.71) and age >75 (OR 2.14). Cognitive domains most affected included executive function and visuospatial processing. Funnel plot asymmetry was minimal (Egger’s p = 0.11). Conclusion: CI affects over 40% of HF patients, with greater burden in HFrEF and elderly subgroups. Clinical predictors such as reduced LVEF, age, and NT-proBNP may aid early identification. These findings highlight the need for routine cognitive screening and integration of cognitive assessment into HF care models to improve adherence and outcomes.