Bronchoalveolar lavage (BAL) and transbronchial biopsy (TBB) frequently are performed in the investigation of immunocompromised patients with lung disorders. The risk-benefit ratio of TBB currently is debated, since several authors have found that the less invasive BAL may provide as much information as TBB, with the avoidance of some biopsy-related side effects. We retrospectively evaluated 157 instances of bronchoscopy carried out on 142 immunocompromised patients, with both BAL and TBB performed in every case. Immunosuppressant conditions were HIV infection (79), hematologic malignancies (36), and antirejection therapy in renal transplant recipients (27). Transbronchial biopsy provided a diagnostic yield significantly higher than that obtained by BAL in all categories investigated; diagnostic rates were 77.3% for TBB and 47.6% for BAL (p < 0.001) in patients with HIV infection, 55 and 20% (p < 0.001) in patients with hematologic malignancies, and 57.5 and 27.2% (p < 0.001) in renal transplant recipients. Looking at the whole series, the diagnostic rates of TBB and BAL were 67.5 and 36.3%, respectively (p < 0.001), with a total additional yield of 33% provided by TBB, while in only 2% of cases BAL gave rise to diagnostic information not achieved by TBB. Considering that side effects followed TBB at a negligible rate (2.5%), we believe that TBB should be routinely carried out in these patients once the diagnostic strategy has been oriented to bronchoscopy.