In 243 of 945 patients having lung percutaneous needle biopsy, we retrospectively evaluated 10 independent variables that potentially influence the incidence of pneumothorax (PNX): age, sex, lesion size, number of needle passes, needle intrapulmonary route, distance of the lesion from the chest wall, FEV1, FVC, residual volume, and total lung capacity percent predicted. The subjects considered (mean age, 61 years; 192 men and 51 women) performed respiratory function tests within 1 year of the procedure. The sample was also subdivided into four functional groups: restricted patients, obstructed ones, obstructed with alveolar hyperinflation patients, and normal subjects. The variables significantly correlated to PNX were the length of the needle intraparenchymal route (p<0.05) and the distance of the lesion from chest wall greater than 0 cm (p<0.01). None of the functional parameters was determinant in predicting PNX occurrence. Bronchial obstruction was not significantly associated with a higher risk of PNX. Nevertheless, when alveolar hyperinflation was associated with bronchial obstruction, the risk increased significantly (odds ratio, 2.38).