The isolation perfusion model, including transbronchial ventilation of human lung, offers the possibility to study pharmacological interactions under physiological conditions. In view of the increasing importance of targeted therapy of lung diseases, this model of perfusion might attract major interest, particularly, in lung cancer. Our study investigated physiological, histological, and immunohistochemical alterations of lung and tumor tissue during isolated perfusion of lung lobectomy specimens to explore potential limitations of this model. Right after resection, 16 human lung resection specimens for primary lung cancer were isolated, ventilated, and perfused under physiological conditions with a modified Krebs-Henseleit solution over a period of 10, 60, 90, 120, and 240 min. Perfusion pressure, pH, lung weight gain, and histological edema formation were measured continuously before and during perfusion. After perfusion, lung and tumor tissue was investigated by hematoxylin-and-eosin stained sections. Immunohistochemistry of NADH, PECAM-1, angiotensin-converting-enzyme and NF-kappabeta were performed to determine lung tissue viability and changes at the endothelial layer. We found that perfusion up to 120 min could be performed with completely stable physiological conditions. Beyond that time span, edema formation and weight gain of the resection specimen started and were followed by an increase in inspiratory pressure and pulmonary artery pressure. Perfusion of more than 4 h led to a significant edema formation in lung tissue accompanied by loss of viability and significant histological alterations. We conclude that isolated ventilation and perfusion of human lung resections within the setup chosen is reliable for pharmacological studies up to a period of 120 min. Thereafter, edema formation and endothelial damage develop and limit the interpretation and reliability of drug delivery studies.