The mechanics of gas flow in endotracheal (ET) tubes have been evaluated extensively in vitro under static and dynamic conditions. Previous bedside determinations of respiratory system mechanics in patients with acute respiratory failure have been based on assumptions derived from in vitro measurements without direct measurement of in vivo ET tube resistance (RET). We hypothesized that the RET measured in vivo would be greater than those values obtained in vitro when peak flow rates and ET tube size were held constant. We measured airflow, volume (pneumotachograph), esophageal pressure (nasogastric-esophageal balloon), and airway pressure (airway catheter) in 10 patients intubated with no. 8.0 orotracheal tubes. We also studied the static and dynamic flow-pressure relationships for five different sizes of ET tubes in vitro (artificial lung) (6.0, 7.0, 7.5, 8.0, and 8.5). The static and dynamic values of RET and the Rohrer coefficients of linear and nonlinear resistance (K1 and K2) were similar to values previously reported in the literature. Although there was considerable individual variation, values of RET measured in vivo were generally higher than those derived from in vitro measurements at both peak flow rates tested, perhaps because of secretions, head or neck position, tube deformation, or increased turbulence. We conclude that ET tubes contribute significantly to total airflow resistance and that RET is often significantly greater than indicated by in vitro studies. Estimates of work of breathing in critically ill patients must take into consideration the contribution of in vivo RET on total pulmonary system resistance.