The solitary pulmonary nodule often presents a diagnostic challenge to the specialist because the nature of the nodule is often indeterminate at the end of the usual diagnostic process, and operation frequently is required before a definite diagnosis can be made. We have conducted a randomized, prospective trial to evaluate the diagnostic efficacy of video-assisted thoracic surgery versus muscle-sparing lateral thoracotomy. Between January 1991 and May 1994, 44 patients suffering from solitary pulmonary nodule were divided at random into two groups: the nodule was removed in 22 cases by video-assisted thoracic surgery and in 22 cases by lateral thoracotomy. Nineteen wedge resections, 1 segmentectomy, and 2 lobectomies were performed in the first group and 13 wedge resections, 8 segmentectomies, and 1 lobectomy in the second group. An "access" thoracotomy had to be performed in 5 patients in the video-assisted thoracic surgery group. The operating room time was 97.2 +/- 32.9 minutes in the video-assisted thoracic surgery group and 130.5 +/- 14 minutes in the lateral thoracotomy group (p > 0.05). In both groups a final diagnosis was made in 100% of cases. The postoperative hospital stay was 4.6 +/- 1.08 days in the video-assisted thoracic surgery group and 7.8 +/- 0.89 days in the lateral thoracotomy group (p < 0.01). Pain was evaluated on a visual analogue scale; the scores were 26.5 +/- 11.6 in the video-assisted thoracic surgery group and 48.3 +/- 12.8 in the lateral thoracotomy group (p < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)