Abstract
Malnutrition (MN) is a highly prevalent condition in the elderly. It is associated with functional impairment, disability, frailty, and sarcopenia. The aim was to analyze the capacity of GLIM and ESPEN criteria to diagnose MN in a sample of institutionalized psychogeriatric patients. Clinical and anthropometric data were collected in a cross-sectional study. Patients' frailty, dependence, functional capacity, MNA, hand-grip strength (HS), and sarcopenia were evaluated. Body composition (BC) was estimated by conventional bioimpedance analysis. MN diagnosis was established using the ESPEN and the GLIM criteria based on fat-free mass index (GLIM-FFMI), appendicular skeletal muscle mass index (GLIM-ASMMI), skeletal muscle mass index (GLIM-SMMI), and HS (mGLIM). Ninety-two patients (57.6% men; mean age: 79.4 years) were studied. Depending on the diagnosis criteria, MN prevalence was between 25% (ESPEN) and 41.3% (GLIM-SMMI). Agreement between ESPEN and all GLIM criteria was poor, but it was excellent between all GLIM criteria (kappa > 0.8). Phenotypic criteria carried more weight in the diagnosis of MN than etiological ones. Depending on the parameter used, the prevalence of reduced muscle mass was notably different. Differences in BMI, BC, inflammation, and albumin are detected by the GLIM-FFMI criteria in the MN and non-MN subjects. Also, this criterion is the only one that identified differences in phase angle (PhA) between these groups. In the elderly, PhA can be very useful to monitor nutritional status.