Cognitive impairment in malnourished heart failure patients

Abstract Background Malnutrition is common in heart failure (HF) and is associated with poor outcomes, including increased mortality and hospitalizations. However, its impact on cognitive function in patients with HF is less well understood. The Geriatric Nutritional Risk Index (GNRI) is a validated marker of nutritional status that has been associated with adverse clinical outcomes in various populations. Purpose The objective of this study was to examine whether low GNRI (high nutrition-related risk) is associated with worse cognitive performance in patients hospitalized for HF. Methods We analyzed data from 384 patients hospitalized for HF between March 2014 and February 2025. Cognitive function was assessed using the Montreal Cognitive Assessment (MoCA), the Symbol Digit Modalities Test (SDMT), and Trail Making Test A (TMT-A). GNRI was categorized into high risk of malnutrition-related morbidity corresponding to GNRI 82, and all others (GNRI ≥ 82). Multivariable linear regression models were used to evaluate the association between GNRI and cognitive test performance, adjusting for age, sex, systolic blood pressure, education level (≥12 years vs. 12 years), and history of stroke. Results A total of 384 patients (mean age 73.8 (±12.6) years, 68.2% men) had complete cognitive assessments and covariate data. Patients with high nutritional risk (GNRI 82) performed significantly worse on MoCA (beta -1.57, 95% CI -3.10 to -0.04, p=0.045) and SDMT (beta -4.07, 95% CI -8.00 to -0.15, p=0.045) compared to those without high nutritional risk (Table 1). These associations remained significant after adjusting for covariates. In contrast, GNRI was not significantly associated with performance on TMT-A (mean difference -0.03, 95% CI -0.25 to 0.20, p=0.80). Conclusions Our findings indicate that malnutrition, as assessed by GNRI, is associated with lower cognitive performance in HF patients, particularly in global cognition (MoCA) and processing speed (SDMT). These results suggest that nutritional status should be considered when evaluating cognitive impairment in HF. However, GNRI was not associated with performance on TMT-A, suggesting that executive function and psychomotor speed may be less affected in this population. Alternatively, TMT-A may not be an optimal instrument for assessing cognitive impairment in hospitalized HF patients due to factors such as frailty, motor impairments, or fluctuating attention. Future research is needed to explore the underlying mechanisms linking malnutrition and cognitive decline in HF patients, as well as the suitability of different cognitive tests in this clinical context.Table 1.