Computed tomography (CT) assessment of positive end-expiratory pressure (PEEP)-induced alveolar recruitment is classically achieved by quantifying the decrease in nonaerated lung parenchyma on a single juxtadiaphragmatic section (Gattinoni's method). This approach ignores the alveolar recruitment occurring in poorly aerated lung areas and may not reflect the alveolar recruitment of the entire lung. This study describes a new CT method in which PEEP-induced alveolar recruitment is computed as the volume of gas penetrating in poorly and nonaerated lung regions following PEEP. In 16 patients with acute respiratory distress syndrome a thoracic spiral CT scan was performed in ZEEP and PEEP 15 cm H(2)O. According to the new method, PEEP induced a 119% increase in functional residual capacity (FRC). PEEP-induced alveolar recruitment was 499 +/- 279 ml whereas distension and overdistension of previously aerated lung areas were 395 +/- 382 ml and 28 +/- 6 ml, respectively. The alveolar recruitment according to Gattinoni's method was 26 +/- 24 g and no correlation was found between both methods. A significant correlation was found between PEEP-induced alveolar recruitment and increase in Pa(O(2)) only when recruitment was assessed by the new method (Rho = 0.76, p = 0.003), suggesting that it may be more accurate than Gattinoni's method.