The predictive value of GLIM criteria on clinical outcomes and responses to nutritional support in patients with neurocritical illnesses

Abstract

AbstractNeurocritically ill patients frequently exhibit coma, gastroparesis, and intense catabolism, leading to an increased risk of malnutrition. The Global Leadership Initiative on Malnutrition (GLIM) criteria for the diagnosis of malnutrition was created to achieve a consistent malnutrition diagnosis across diverse populations. This study aimed to validate the concurrent and predictive validity of GLIM criteria in patients with neurocritical illnesses. A total of 135 participants were followed from admission to the neurocritical unit (NCU) until discharge. Comparing GLIM criteria to the Subjective Global Assessment (SGA), sensitivity was 0.95 and specificity was 0.69. Predictive validity of GLIM criteria was assessed using a composite adverse clinical outcome, comprising mortality and various major complications. Adjusted hazard ratios for moderate and severe malnutrition were 2.86 (95% CI 1.45–5.67) and 3.88 (95% CI 1.51–9.94), respectively. Changes in indicators of nutritional status, including skeletal muscle mass and abdominal fat mass, within 7 days of admission were obtained for 61 participants to validate the predictive capability of the GLIM criteria for the patients’ response of standardized nutritional support. The GLIM criteria have a statistically significant predictive validity on changes in rectus femoris muscle thickness and midarm muscle circumference. In conclusion, the GLIM criteria demonstrate high sensitivity for diagnosing malnutrition in neurocritically ill patients and exhibit good predictive validity.