Lung transplantation has become an acceptable therapeutic option for end-stage pulmonary diseases. The most common causes of long-term mortality after transplantation are infections and obliterative bronchiolitis (OB). While acute rejection has been shown to be associated with an increased risk of development of OB, cytomegalovirus (CMV) pneumonitis is more controversial as a risk factor for OB. Surveillance bronchoscopies are therefore advocated as a method of detecting silent episodes of CMV pneumonitis or acute rejection. We performed 226 bronchoscopies in 43 lung transplant recipients over 34 months. One hundred fifty-seven of the 226 bronchoscopies were performed according to a surveillance protocol. Acute rejection was diagnosed if lung histologic study revealed grade 2 to 4 rejection or if prompt reversal of clinical deterioration occurred after initiation of pulse steroid therapy. CMV pneumonitis was diagnosed when transbronchial biopsy histologic specimens revealed evidence of CMV inclusion bodies, or when CMV was recovered on BAL fluid in the presence of allograft deterioration. The proportion of patients who were free from any episode of acute rejection or CMV pneumonitis after transplantation was determined by Kaplan-Meier analysis. Twenty-one percent of our transplant recipients were free from acute rejection or CMV pneumonitis after a mean follow-up of 13 months. All patients who had acute rejection or CMV pneumonitis had the initial episode in the first 4 months after transplantation. Patients free of acute rejection or CMV pneumonitis 4 months after transplantation continued to be event free for the duration of follow-up. Our data suggest that surveillance bronchoscopy can be aborted in patients who are free from acute rejection or CMV pneumonitis by 4 months after transplantation. The role of surveillance bronchoscopy in decreasing the incidence of OB or improving survival can be determined only by future randomized prospective trials.