Validation of GLIM Criteria for Diagnosis of Malnutrition in Stroke Survivors

Norah Alshammari¹Alanoud Aladel¹Mahmoud Desoky²Danyah Althunyyan¹Hissah Altimyat¹Fatimah Alsoqeah³Mahmoud M. A. Abulmeaty^1*

• ¹Community Health Sciences Department, College of Applied Medical Sciences, King Saud University, Riyadh, Saudi Arabia
• ²Internal Medicine Department, Sultan Bin Abdulaziz Humanitarian City, Riyadh, Riyadh, Saudi Arabia
• ³Clinical Nutrition Department, Sultan Bin Abdulaziz Humanitarian City, Riyadh, Saudi Arabia

Background/Objectives: Nutritional assessment is essential for delivering optimal care and achieving the best possible outcomes for stroke survivors. The Global Leadership Initiative on Malnutrition (GLIM) proposed new consensus criteria for diagnosing malnutrition in 2018. These criteria are anticipated to effectively predict significant outcomes related to malnutrition. This study aims to validate the GLIM criteria as a diagnostic tool for malnutrition among Saudi stroke survivors, comparing it with the subjective global assessment (SGA). Methods: A cross-sectional study was conducted involving 104 adult stroke survivors. Nutritional risk was first evaluated using the Nutrition Risk Screening 2002 (NRS-2002) as an initial step of the GLIM criteria, followed by diagnosis using both the GLIM criteria and the SGA. The level of agreement between the tools was assessed using the kappa coefficient (κ) statistics. Additionally, the area under the receiver operating characteristic curve (AUC-ROC) analysis was performed to determine the sensitivity, specificity, positive predictive value, and negative predictive value of the tools, thereby evaluating their accuracy. Results: A total of 104 stroke survivors were evaluated, with an average age of 61.0 years (interquartile range (IQR): 15 years), and 73.1% were men. According to the GLIM criteria, malnutrition was identified in 47.1% of the survivors, while the SGA indicated malnutrition in 51.9%. The GLIM criteria demonstrated acceptable performance, with an AUC of 0.819 (95% CI: 0.734 – 0.905), a sensitivity of 78.2%, and a specificity of 85.7%. The level of agreement between the two tools was substantial (κ = 0.635). Conclusion: The GLIM criteria for diagnosing malnutrition showed good criterion validity and appear to be a reliable approach for assessing nutritional status among stroke survivors.