ESPEN 2021 Late Breaking Abstracts

Topic: Nutritional assessment

Abstract Submission Identifier: ESPEN21-LB-1891

CONCURRENT AND PREDICTIVE VALIDITY OF GLIM CRITERIA AND AND/ASPEN CRITERIA FOR MALNUTRITION DIAGNOSIS AMONG HOSPITALIZED CARDIAC PATIENTS

V. V Sundar^*, ¹, O. Shu Hwa ¹, M. E. P.M Easaw ¹, W. C. Siew Swee ¹

¹INTERNATIONAL MEDICAL UNIVERSITY, Kuala Lumpur, Malaysia

Rationale: Malnutrition is prevalent among hospitalised cardiac patients due to multiple co-morbidities and inflammation conditions. In recent decade, the Global Leadership Initiative on Malnutrition (GLIM) and the Academy of Nutrition and Dietetics and American Society for Parenteral and Enteral Nutrition (AND/ASPEN) have proposed respective criteria for malnutrition diagnosis, however studies on their validity are limited. This study aimed to identify the concurrent and predictive validity of GLIM and AND/ASPEN criteria for malnutrition diagnosis among hospitalised cardiac patients.

Methods: Prospective study was conducted among adult/elderly hospitalised patients with cardiac diseases in general wards. Nutrition status was assessed with GLIM criteria, AND/ASPEN criteria and Subjective Global Assessment (SGA) within 48 hours of admission. Quality of life (QOL) was evaluated using the Medical Outcomes Short Form 12 (SF-12) questionnaire.  Patients were followed up to gather outcomes of length of hospital stay (LOS) and unplanned readmission, incidence of infection and mortality 90-days after discharge. Concurrent and predictive validity were determined.

Results: A total of 276 patients (61±11 years, 76.4% males) were evaluated. The prevalence of malnutrition was 41.7%, 54.0% and 58.3% by GLIM, AND/ASPEN and SGA, respectively. GLIM showed substantial accuracy (AUC=0.77, 95%CI 0.70-0.82; 63.4% sensitivity; 88.7% specificity) and fair agreement with SGA (k=0.49). AND/ASPEN had good accuracy (AUC=0.96, 95%CI 0.93-0.98; 92.5% sensitivity; 100.0% specificity) and good agreement with SGA (k=0.91). After adjusting for confounders, malnutrition diagnosed by GLIM decreased QOL by 2.1 (95%CI 1.25-3.52) times, increased the risk of readmission by 5.2 (95%CI 3.00-9.30) times, infection by 3.7 (95%CI 2.12-6.70) times and mortality by 5.1 (95%CI 1.37-19.14) times. Malnutrition diagnosed by AND/ASPEN increased the risk of readmission by 32.4 (95%CI 14.45-73.01) times, infection by 13.9 (95%CI 6.58-29.59) times and mortality by 5.3 (95%CI 1.10-26.11) times.  

Conclusion: Both the GLIM and AND/ASPEN criteria demonstrated good validity in diagnosing malnutrition and predicting clinical outcomes of hospitalised cardiac patients.

References: de van der Schueren MA. et al. Clinical Nutrition. 2020 Sep;39(9):2872-80.

Disclosure of Interest: V. V Sundar: None Declared, O. Shu Hwa: None Declared, M. P.M Easaw : None Declared, W. Siew Swee Grant / Research Support from: International Medical University

Keywords: GLIM criteria, hospitalized patients, malnutrition, validation