• M Planas, S Audivert, C Pérez-Portabella, R Burgos, C Puiggrós, J M Casanelles, and J Rosselló.
• Nutritional Support Unit, University Hospital Vall d'Hebron, Passeig Vall d'Hebron 119-129, Barcelona 08035, Spain. mplanas@hg.vhebron.es
• Clin Nutr. 2004 Oct 1; 23 (5): 1016-24.

BackgroundAlthough malnutrition in hospitalized patients is generally associated with increasing morbidity and mortality, it is yet a widely unknown problem in hospitals.ObjectivesThe aim of this study was to assess the nutritional status of patients admitted to a university-affiliated hospital in Spain using anthropometry measurements and the Subjective Global Assessment (SGA) technique.MethodsWe enrolled 400 patients selected at random using a computer software program. The primary end-point was nutritional status determined within 48 h of admission by anthropometric data (body mass index, triceps skinfold thickness, and upper arm muscle circumference) and by the SGA technique. Using anthropometric data, patients were considered to have normonutrition or malnutrition. Those with malnutrition, were subdivided in patients with low (undernutrition) or high (overnutrition) body weight. Through SGA patients were classified as having normonutrition or malnutrition (moderate and severe). Secondary end-points were hospital length of stay (LOS), mortality, and readmissions (total and non-elective readmissions) over the next 6 months. Overall population, patients scheduled admitted, patients admitted from emergency room, and those with any cancer were individually analyzed.ResultsThe frequency of malnutrition varied from 72.7% assessed by anthropometry (undernutrition in 26.7% and overnutrition in 46.0%), to 46% using SGA. Malnutrition was not related to the type of admission neither to the diagnosis of cancer. Of 400 patients analyzed, two patients died (0.5%). Using SGA, LOS was significantly higher in patients with malnutrition vs. those with normonutrition, in the overall population and in patients scheduled admitted, and there were more total and non-elective readmissions in patients with malnutrition than in patients with normonutrition in the overall population, in patients scheduled admitted and in those with cancer. When we used anthropometric data, LOS was superior in undernutrition compared to normo and overnutrition in scheduled admitted patients alone. Although there were more total readmissions in undernutrition than in normonutrition and overnutrition in overall population, no significant differences were observed with the non-elective readmission rate.ConclusionsA high prevalence of malnutrition was found in this study. At a time, a high prevalence of overnutrition was observed. Anthropometric data and SGA technique are not concordant, reflecting the limitations of markers of nutritional status. While with SGA malnutrition was detected in patients with normal to high BMI, with anthropometry overnutrition was diagnosed. SGA seems to be more accurate than anthropometry to anticipate hospital LOS and readmission rate. Due to the increased LOS and readmission rates found in patients with malnutrition, further steps among health care professionals are warranted to identify and control them.

26 keywords...

hide…