The objective of this study was to evaluate the utility of the determination of adenosine deaminase (ADA) level in pleural fluid for the differential diagnosis between tuberculous pleural effusion (TPE) and malignant pleural effusion (MPE) in Japan, a country with intermediate incidence of tuberculosis (TB). We retrospectively reviewed the clinical records of 435 patients with pleural effusion and investigated their pleural ADA levels as determined by an auto analyzer. ROC analysis was also performed. The study included patients with MPE (n＝188), TPE (n＝124), benign nontuberculous pleural effusion (n＝94), and pleural effusion of unknown etiology (n＝29). The median ADA level in the TPE group was 70.8U/L, which was significantly higher than that in any other groups (p＜0.05). The area under the curve (AUC) in ROC analysis was 0.895. With a cut-off level for ADA of 36U/L, the sensitivity, specificity, positive predictive value, and negative predictive value were 85.5%, 86.5%, 69.7%, and 93.6%, respectively. As many as 9% of patients with lung cancer and 15% of those with mesothelioma were false-positive with this ADA cutoff setting. Although the ADA activity in pleural fluid can help in the diagnosis of TPE, it should be noted that some cases of lung cancer or mesothelioma show high ADA activity in geographical regions with intermediate incidence of TB, in contrast to high prevalence areas.