Fiberoptic bronchoscopy (FOB) with bronchoalveolar lavage (BAL) and transbronchial biopsies (TBB) is a widely used method to detect respiratory infections and to differentiate them from other postoperative complications in lung transplant (LTX) recipients, but the usefulness of surveillance FOBs is not yet established. The aim of this study was to evaluate the usefulness of FOB in the diagnosis and surveillance of infections in LTX recipients. We reviewed all the consecutive 609 FOBs performed on 40 lung or heart-LTX recipients between February 1994 and November 2002. The overall diagnostic yield was 115/190 (61%) and 43/282 (15%) for clinically indicated and surveillance FOBs respectively (P < 0.001). Infection was established by bronchoscopic samples in 96/190 (50.5.%) of the clinically indicated FOBs and 34/282 (12.1%) of the surveillance FOBs (P < 0.001). The diagnostic yield of the clinically indicated FOBs was highest (72%) from 1 to 6 months post-transplant (P = 0.04). Pneumocystis carinii was detected in 23 (4.9%) of the bronchoscopic specimens and 15 (65%) of the P. carinii infections were detected during adequate chemoprophylaxis. To conclude, in LTX recipients clinically indicated FOB has a good diagnostic yield in detecting infections and other postoperative complications, whereas the information received from surveillance FOB has remained less significant. With current prophylaxis and screening strategies FOB is still required to detect P. carinii infections.