CONTEXT: Essential hyperhidrosis is a frequent disorder causing significant functional impairment. The advent and development of video-assisted thoracoscopic techniques now allows thoracic sympathectomy to be carried out precisely and safety with good results and minimal morbidity. OBJECTIVE: To assess the impact of video-assisted thoracic sympathectomy in patients diagnosed as presenting severe and disabling hyperhidrosis. TYPE OF STUDY: This was a longitudinal study of the clinical course of all hyperhidrosis cases selected for surgery between May 1999 and January 2003. SETTING: Division of Thoracic Surgery, Universidade Federal de São Paulo (UNIFESP). PARTICIPANTS: 743 patients with surgery indicated due to palmar hyperhidrosis (49.8%), palmar-axillary hyperhidrosis (38.1%), craniofacial hyperhidrosis (8.9%) or isolated axillary hyperhidrosis (2.8%). PROCEDURES: Video-thoracoscopic sympathectomy was performed, isolating the second thoracic ganglion (T2) in all patients, with additional sympathectomy of T3 and T4 if necessary. MAIN MEASUREMENTS:The clinical course was followed up via questionnaires, phone calls, letters and statements. Simple questions were asked regarding the disappearance of symptoms and presence and intensity of compensatory sweating. RESULTS: The surgery was regarded as efficient in all cases of palmar hyperhidrosis. In the craniofacial hyperhidrosis cases, partial recurrence of the symptoms occurred in 2 cases (3.0%). Partial recurrence or persistence of symptoms occurred in 20% of the patients with predominantly axillary symptomatology. The compensatory sweating was considered disagreeable or uncomfortable by about 30% of the patients, but it only reached the level of regretting the operation for 3% of them. This occurred more frequently in patients with axillary hyperhidrosis. Ten cases of complications occurred. CONCLUSION: Thoracoscopic sympathectomy provides very good results in most patients, with a very low complication rate. However, the assessment of surgical results using conventional methods is imprecise and inaccurate. Different methodology, including quality of life assessment, must be used for comparing results and providing objective data on the results of this operation.