Data from 756 myasthenic patients were analyzed for diagnostic criteria, clinical aspects, and therapeutic approaches. The patients were followed up at our institution from 1981 to 2001. Clinical evaluation was performed according to the myasthenia gravis score adopted at our clinic. Clinical features of each patient (comprising demographic, clinical, neurophysiological, immunological, radiological, and surgical data, as well as serial myasthenia gravis scores) were filed in a relational database containing more than 7000 records. Clinical efficacy and variables influencing outcome were assessed by life-table methods and Cox proportional hazards regression analysis. Complete stable remission, as defined by the Task Force of the Medical Scientific Advisory Board of the Myasthenia Gravis Foundation of America, was the end point for good prognosis. Four hundred and ninety-nine patients (66%) were female and 257 (34%) were male. Mean follow-up was 55.1 +/- 48.1 months. Onset of symptoms peaked in the third decade in females, whereas the male distribution was bimodal with peaks in the third and sixth decades. Modality of myasthenia gravis presentation was as follows: ocular, 39.3%; generalized, 28.5%; bulbar, 31.3%; and respiratory, 0.8%. Thymectomy was carried out on 63.7% of our patients by different approaches: (1) transcervical; (2) transsternal; (3) video-thoracoscopic mini-invasive surgery. The last approach has been preferentially used in more recent years and accounted for 62.4% of the thymectomized myasthenia gravis population. Univariate analysis and Kaplan-Meier analysis showed that variables such as sex (female), age at onset (below 40 years), thymectomy, and histological diagnosis of thymic hyperplasia were significantly associated with complete stable remission, whereas on multivariate analysis only age at onset below 40 years and thymectomy were confirmed.