The Impact of Body-Weight Components on Forced Spirometry in Healthy Italians E.I. Mohamed, 1,3 C. Maiolo, 1 L. Iacopino, 1 M. Pepe, 1 N. Di Daniele, 2 and A. De Lorenzo 1 1 Division of Human Physiology, University of Rome ``Tor Vergata,'' Via Montpellier 1, 00133 Rome, Italy 2 Division of Internal Medicine, University of Rome ``Tor Vergata,'' Via Montpellier 1, 00133 Rome, Italy 3 Biophysics Department, Medical Research Institute, University of Alexandria, Alexandria, Egypt Abstract. Many studies have investigated lung function in relation to age and height among Caucasians, however, most of these studies did not consider the individual components of body weight. The objective of the present study was to study the eect of body weight components [bone-free lean body mass BF- LBM), bone mineral content BMC), and fat mass FM)] measured by dual x-ray absorptiometry DXA) on the lung-function variables FVC, FEV1, and PEF) and to derive prediction equations for these variables in healthy adult Italians. Dynamic spirometric tests and body composition analysis by DXA were performed on 58 nonsmoking males, mean age SE) 26.72  1.98 years and BMI 25.51  0.64 kg/m 2 , and 60 nonsmoking females matched for age and BMI 29.61  1.65 years and 26.45  1.05 kg/m 2 , respectively). Bivariate linear regression analysis showed the variables age, height, BF-LBM, BMC, and the interaction term BF-LBM*Height, but not weight and FM, to corre- late signi®cantly with lung-function variables for males and for females sepa- rately. Multiple linear regression analysis showed that sex, age, height, and BF-LBM*Height were signi®cantly associated with FVC, FEV1, and PEF. The prediction equations developed for FVC, FEV1, and PEF on the basis of the independent variables i.e. sex, age y), height m), and BF-LBM*Height kg á m) had a signi®cantly higher cumulative correlation coecient and a lower SEE compared with those based on age and height only. The present report suggests that the BF-LBM, expressed independently from height, can be considered for predicting lung-function variables. Lung 2002) 180:149±159 DOI: 10.1007/s004080000089 Correspondence to: De Lorenzo, Department of Neurosciences, Faculty of Medicine and Sur- gery, University of Rome ``Tor Vergata'', Via Montpellier 1, 00133 Rome, Italy, E-mail: delor- enzo@uniroma2.it