The results of thymectomy on patients with generalised myasthenia gravis have been widely reported. However, there is no information on whether the experience of western countries can be generalised to the population of the Middle East. The purpose of this study was to evaluate the safety and efficacy of thymectomy in patients with myasthenia gravis in a Middle East patient population and to identify clinical and histopathological factors associated with improved long term outcome of surgery. In a prospective study, sixty three patients (aged 1.5-51 years) were treated in two university teaching hospitals between 1984 and 1991 and followed up for a mean of four years. Close communication was established with neurologists to obtain early referral. Radical anterior mediastinal dissection through a median sternotomy was performed in all patients. The response was evaluated by modified Osserman's classification. Eighteen patients achieved complete remission and a further 39 improved, producing an overall response rate of 90.5%. Patients with milder disease (stage II) had a higher response rate (97%) than those with more advanced disease (78%). Patients operated on with less than three years of symptoms had a better outcome (94%) than those with longer duration of preoperative symptoms, especially in non-thymomatous patients. Age and sex had no effect on the outcome. There was no effect on response rate if patients had hyperplastic or non-specific thymic histological findings, but patients with thymoma fared worse. These results are comparable with reports from the western world and represent the first prospective study from the Middle East. Thymectomy is indicated for all patients suffering from generalised myasthenia gravis soon after the diagnosis is made, regardless of age, stage, thymic pathology, and preoperative clinical status.