BACKGROUND: The solitary pulmonary nodule(SPN) presents a diagnostic dilemma to the physician and the patient. Many clinical characteristics(i.e. age, smoking history, prior history of malignancy) and radiological characteristics(i.e. size, calcification, growth rate, several findings of computed tomography) have been proposed to help to determine whether the SPN was benign or malignant. However, most of these diagnostic guidelines are based on the data collected before computed tomography(CT) has been introduced and lung cancer was not as common as these days. Moreover, it is not well established whether these guidelines from western populations could be applicable to Korean patients. METHODS: We had a retrospective analysis of the case records and radiographic findings in 114 patients presenting with SPN from Jan. 1994 to Feb. 1995 in Seoul National University Hospital, a tertiary referral hospital. RESULTS: We observed the following results ; (1) Out of 113 SPNs, the etiology was documented in 94 SPNs. There were 34 benign SPNs and 60 malignant SPNs. Among which, 49 SPNs were primary lung cancers and the most common histologic type was adenocarcinoma. (2) The average age of patients with benign and malignant SPNs was 49.7+/-12.0 and 58.1 +/-10.0 years, respectively(p=0.0004), and the malignant SPNs had a striking linear propensity to increase with age. (3) No significant difference in the history of smoking was noted between the patients with benign SPNs(13.0+/-17.6 pack-year) and those with malignant SPNs(18.6+/-25.1 pack-year) (p=0.2108). (4) 9 out of 10 patients with prior history of malignancy had malignant SPNs. 5 were new primary lung cancers with no relation to prior malignancy. (5) The average size of benign SPNs (3.01+/-1.20cm) and malignant SPNs(2.98+/- 0.97cm) was not significantly different(p=0.8937). (6) The volume doubling time could be calculated in 22 SPNs. 9 SPNs had the volume doubling time longer than 400 days. Out of these, 6 were malignant SPNs. (7) The CT findings suggesting malignancy included the lobulated or spiculated border, air-bronchogram, pleural tail, and lymphadenopathy. In contrast, calcification, central low attenuation, cavity with even thickness, well-marginated border, and perinodular micronodules were more suggestive for benign nodule. (8) The diagnostic yield of percutaneous needle aspiration and biopsy was 57.6%(19/33) of benign SPNs and 81.0%(47/58) of malignant SPNs. The diagnostic value of sputum analysis and bronchoscopic evaluations were relatively very low. (9) 42.3%(ll/26) of SPNs of undetermined etiology preoperatively turned out to be malignant after surgical resection. Overall, 75.4%(46/61) of surgically resected SPNs were malignant. CONCLUSIONS: We conclude that the likelihood of malignant SPN correlates the age of patient, prior history of malignancy, some CT findings including lobulated or spiculated border, air-bronchogram, pleural tail and lymphadenopathy. However, the history of smoking, the size of the nodule, and the volume doubling time are not helpful to determent whether the SPN is benign or malignant, which have been regarded as valuable clinical parameters previously. We suggest that aggressive diagnostic approach including surgical resection is necessary in patient with SPNs.