The aim of this study is to determine the etiologic agents of pneumonia in liver recipients. Medical records of liver transplantations performed between January 2000 and January 2007 in our hospital were examined retrospectively. Pneumonia was defined as the presence of a new or progressive and persistent infiltrate on chest radiography associated with at least two of the following criteria as described: leukocyte count greater than 10.5x109/L or less than 3.5x109/L; temperature higher than 38C or lower than 36; new onset of purulent sputum, change of character of sputum or increased respiratory secretions; new onset of worsening cough, dyspnea, tachypnea, rales, or bronchial breath sounds and worsening gas exchange (i.e., oxygen desaturation, increased oxygen requirements or increased mechanical ventilatory support). One hundred and sixty-two liver transplantations were performed between January 2000 and January 2007. Twenty-three episodes of pneumonia were diagnosed in 19 (11.7%) patients. Eleven (47.9%) of the 23 episodes were seen in the first month after transplantation and were defined as nosocomial pneumonia. Two (8.6%) episodes were seen during the 1st and 6th months. Ten of the 23 (43.5%) episodes were seen in the late period (i.e. later than 6 months after transplantation). Etiologic agents were isolated in 6 (54.4%) of the 11 nosocomial infections: 1 Klebsiella pneumoniae, 1 Escherichia coli, 1 Pseudomonas aeruginosa, 1 Stenotrophomonas maltophilia, 1 Candida albicans, 1 non-albicans Candida. The two episodes seen during the 1st-6th months after transplantation were due to Aspergillus spp. and cytomegalovirus. Only one (10.0%) etiologic agent determined in the late period was Streptococcus pneumoniae. In conclusion, empirical antibacterial therapy is the cornerstone of management of pneumonia but determination of specific etiologic agents particularly in the immunocompromised patients is of great importance. The low percentage of isolated etiologic agents in the late period is probably due to the empiric use of antibiotics before the respiratory samples were taken.