Malnutrition is common in patients with acute stroke and may increase susceptibility to stroke-associated pneumonia (SAP). This study aimed to assess the predictive value of the Mini Nutritional Assessment Short Form (MNA-SF) for SAP and to compare its performance with laboratory and anthropometric nutritional indicators. In this prospective cohort study, 317 patients hospitalized with acute stroke were enrolled. Nutritional status was evaluated at admission using MNA-SF, routine laboratory markers, and anthropometric measurements. SAP occurring within 7 days of stroke onset was the primary outcome. Multivariable logistic regression models were applied to examine independent associations between nutritional indicators and SAP. Predictive performance was evaluated using receiver operating characteristic (ROC) curve analysis. SAP occurred in 86 patients (27.1%). Patients at risk of malnutrition (MNA-SF score 8–11) had a significantly higher risk of SAP compared with those with normal nutritional status (adjusted OR up to 3.94, 95% CI 1.26–12.32), independent of demographic, clinical, laboratory, and anthropometric factors. An MNA-SF cutoff score of 11 demonstrated good discriminative ability for SAP prediction (AUROC 0.84, 95% CI 0.79–0.88), outperforming blood-based nutritional biomarkers and anthropometric measures. The negative predictive value was 90.6%. Lower MNA-SF scores are strongly associated with the development of stroke-associated pneumonia. Compared with commonly used nutritional indicators, MNA-SF provides superior predictive performance and represents a simple, non-invasive tool for early identification of stroke patients at high risk for SAP.
Liao et al. (Fri,) studied this question.