Increased length of hospital stay in underweight and overweight patients at hospital admission: a controlled population study

Background: Reduced Lean tissue as well as high fat mass may be independent nutritional risk factors resulting in increased Length of hospital stay (LOS). This controlled population study (1707 patients, 1707 volunteers) aimed to evaluate the association between LOS in Geneva and Berlin patients at hospital admission and high fat mass index (FMI, kg/m(2)) and tow fat-free mass index (FFMI, kg/m(2)), and the respective value of body mass index (BMI) and of FFMI and FMI for nutritional assessment. Methods: Patients (891 men, 816 women) were prospectively recruited at hospital admission and compared to gender-, age- and height-matched healthy volunteers. Fat-free mass and fat mass, determined at admission by 50 kHz-bioelectrical impedance analysis, were expressed as indices (FFMI and FMI-kg/m(2)) to normalize for height. Patients were classified in four groups: normal, tow FFMI, high FMI, or Low FFMI and high FMI. Logistic regressions were used to determine the association between body composition and LOS. Results: Higher FMI and Lower FFMI were found in patients at hospital admission than in sex- and age-matched healthy volunteers. Low FFMI, high FMI, and Low FFMI/ high FMI combined, adjusted for age, were all significantly associated with Longer LOS (high FFMI: 1-5 days OR 2.4, Cl 2.0-2.9; 6-10 days OR 2.3, CI 1.8-3.0; greater than or equal to 11 days OR 2.8, CI 2.2-3.5); low FMI: 1-5 days OR 1.9, Cl 1.6-2.2; 6-10 days OR 2.7, Cl 2.0-3.5, greater than or equal to 11 days OR 2.1, Cl 1.7-2.7; low FFMI/high FMI: 1-5 days OR 7.8, Cl 5.3-11.4; 6-10 days OR 13.6, Cl 7.8-23.5, greater than or equal to11 days OR 11.8, Cl 7.0-19.8). Conclusion: Increased LOS is associated with adiposity (high FMI) and Low muscle mass (tow FFMI). The current study shows that both depletion of lean tissue and excess of fat mass negatively affect the LOS. Finally, we found that excess fat mass reduces the sensitivity of BMI to detect nutritional depletion. (C) 2004 Elsevier Ltd. All rights reserved.