Fat mass protects hospitalized elderly persons against morbidity and mortality 1–3 Olivier Bouillanne, Claire Dupont-Belmont, Phasaro Hay, Brigitte Hamon-Vilcot, Luc Cynober, and Christian Aussel ABSTRACT Background: It is well established that the combination of protein- energy malnutrition and low body mass index (BMI) increases the risk of death in elderly patients, but recent studies indicate that the risk of death may decrease with higher body weight. However, these studies did not perform direct, separate, and reliable measurements of fat and lean mass by using a reference technique. Objective: Our objective was to evaluate the relation between body composition, based on the 4-compartment model, and morbidity and mortality in hospitalized elderly patients. Design: This prospective study enrolled 125 elderly patients eval- uated at admission for body composition on the basis of BMI plus fat mass, lean mass, appendicular skeletal muscle mass, and body cell mass indexes (calculated as the ratio of the mass to the square of the height) measured by X-ray absorptiometry and bioelectrical impedance analysis. Outcomes were assessed 6 mo later by using a score system that takes into account complications (pressure ul- cers and/or infections) and 6-mo mortality. Results: The fat mass index correlated positively (r = 0.19 and P = 0.023 adjusted for sex; r = 0.18 and P = 0.043 adjusted for sex, albuminemia, and C-reactive protein) with outcome score (1: death, 2: complications, 3: no complications). There was no significant correlation between outcome score and BMI, lean mass, appendic- ular skeletal muscle mass, and body cell mass indexes. Conclusions: This study clearly indicates that the generally ac- cepted rule that overweight is associated with morbidity and mor- tality does not apply to hospitalized elderly patients, for whom fat mass is associated with a decreased risk of adverse events. Am J Clin Nutr 2009;90:505–10. INTRODUCTION Protein-energy malnutrition is known to be highly prevalent in geriatric patients (1) and to be associated with an increased risk of mortality and morbidity (infections and pressure ulcers) in hospitalized elderly patients (1). However, the relation between overweight and mortality may vary with age. In middle-aged adults, the relation between body mass index (BMI; in kg/m 2 ) and overall mortality is U-shaped, with increased risk in the lowest and highest percentiles (2–4). In contrast, studies in elderly persons (5–7) suggest that the risk of mortality decreases with increasing body weight. A study in hospitalized middle-aged and elderly patients found that for patients aged ,65 y, the relation between BMI and mortality was U-shaped (5), whereas for patients aged .75 y, there was an inverse relation between mortality risk and BMI, with the highest mortality rates at the lowest BMIs. Similarly, a study in community-based elderly subjects found that those with a BMI .30 had the lowest mortality (6). Another study in community-dwelling elderly subjects also found that mortality rates were lower in obese subjects and higher in underweight subjects than in normal- weight subjects (7). Because fat mass (FM) forms the main energy stores of the body, it may confer survival advantages in elderly patients experiencing catabolic events, such as infectious complications. However, the studies cited did not carry out di- rect separate measures of fat and lean soft tissue mass (LM), which could not be discriminated by BMI. Clearly, the BMI of elderly patients can be affected by confounding factors, eg, normal or low BMI due to loss of LM with age (sarcopenia), compensated to varying degrees by an increased FM. Our working hypothesis was that body fat may be one of the com- ponents predicting malnutrition-related risks of mortality and morbidity in hospitalized elderly patients. To the best of our knowledge, this was the first study to evaluate the relation between FM, which was assessed by using the gold-standard dual-energy X-ray absorptiometry (DXA) technique (8), and morbidity and mortality in elderly patients. As comorbid factors, we focused our attention on infection and pressure ulcers be- cause 1) they are hospital-acquired malnutrition-related com- plications, 2) they are highly prevalent in elderly hospitalized patients, and 3) they increase the risk of mortality and morbidity. SUBJECTS AND METHODS Subjects Our geriatric hospital is affiliated with the Assistance Publique– Ho ˆpitaux de Paris and has close ties with Paris-area university hospitals specializing in acute care. One hundred forty-eight 1 From Services de Ge ´rontologie 2 (OB, CD-B, PH, and BH-V) and Biologie (CA), Ho ˆpital Emile-Roux, Assistance Publique-Ho ˆpitaux de Paris, Limeil-Bre ´vannes, France; Laboratoire de Biologie de la Nutrition, Univer- site ´ Paris Descartes, Faculte ´ de Pharmacie, Paris, France (OB, LC, and CA); and Service de Biochimie, Ho ˆpitaux Cochin et Ho ˆtel-Dieu, Assistance Publique–Ho ˆpitaux de Paris, Limeil-Bre ´vannes, France (LC). 2 Supported by Emile-Roux Hospital. 3 Address reprint requests and correspondence to O Bouillanne, Service de Ge ´rontologie 2, Ho ˆpital Emile-Roux, (Assistance Publique–Ho ˆpitaux de Paris), 94456 Limeil-Bre ´vannes, France. E-mail: olivier.bouillanne@erx.aphp.fr. Received March 23, 2009. Accepted for publication June 20, 2009. First published online July 29, 2009; doi: 10.3945/ajcn.2009.27819. Am J Clin Nutr 2009;90:505–10. Printed in USA. Ó 2009 American Society for Nutrition 505 Downloaded from https://academic.oup.com/ajcn/article/90/3/505/4597021 by guest on 12 March 2022