* Department of Geriatric Medicine, Centre Hospitalier Regional de la Citadelle, Boulevard du 12ème de Ligne, 4000 Liège, Belgium. ** Department of Biostatistics, University of Liège, CHU Sart-Tilman, 4000 Liège, Belgium. Correspondence to: Claire Gazzotti, MD, Service de Gériatrie, CHR de la Citadelle, Bd. du 12eme de Ligne, 4000 Liège, Belgium. Tel: 32-4-3692575. Fax: 32-4-3694089 o E-mail : claire.gazzotti@chrcitadelle.be

Abstract: Objective: This study was undertaken to estimate the prevalence of malnutrition in elderly patients hospitalized with an acute illness, as well as to assess the clinical usefulness of standardized nutritional assessment upon admission by means of the MNA scale. Design: A prospective study. Setting: A large size regional university hospital. Subjects: There were 175 patients (113 women and 62 men) with a mean age of 79.7 + 8.5 years admitted for an acute problem. Death occurred in 11 patients (6.3%). Methods: Upon admission, demographic (age, gender, origin) and medical (disease, drugs) data were recorded for each patient; the MNA questionnaire (score: 0-30) was administered and Katz score (7-28) calculated. At hospital discharge, data included Katz score, outcome (death/survival), and destination. Results: The mean MNA score was 20.5 + 5.1 and the prevalence of severe malnutrition (MNA <17) was 21.7%. Further, 48.6% of elderly were at risk of malnutrition (MNA between 17 and 24). There was no association between MNA and age or gender, but underweight was a sign of low MNA values (P <.001). MNA scores were inversely related to Katz scores at both admission and hospital discharge (P < .001). Patients originating from nursing homes had a poorer nutritional status than those living at home (MNA: 18.4 against 22.3, P <.001). The number of drugs taken per patient (5.2 + 2.8) was found to be correlated with MNA (P = .049). MNA scores were on average significantly higher (P <.001) in survivors (20.9) than in nonsurvivors (14.1). Conclusions: The study clearly demonstrates the high prevalence of malnutrition and the clinical usefulness of the MNA scale in geriatric medicine. The MNA score upon admission reflects the patient's nutritional condition, degree of autonomy (Katz score), living conditions and current treatment. It is also predictive of patient's outcome (death or survival).

Key words : ageing, MNA, malnutrition, Katz score