A best evidence topic in thoracic surgery was written according to a structured protocol. The question addressed was: Does repeat thymectomy improve symptoms in patients with refractory myasthenia gravis after thymectomy? A total of 189 papers were found using the reported search, of which seven represented the best evidence to answer the clinical question. The outcome measures included operative mortality and morbidity, as well as long-term remission rate. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. All the studies were small (4-21 patients), retrospective, single institutional case series. There was considerable heterogeneity in the studies. The interval between the first and second operation ranged from less than a year to over 10 years. The operative approach of the initial operation included transcervical, trans-sternal and substernal approaches. The maximal medical therapy received by the patients prior to reoperation varied from anticholinesterase alone to cytotoxic therapy and regular plasmapheresis. The severity of symptoms ranged from Osserman Class IIa to V. The operative approach to re-thymectomy included resternotomy, thoracoscopy and a combination of both. There was no perioperative mortality. One study reported injury to the innominate vein at resternotomy in 3 (14.3%) patients. One study reported myasthenic crisis in 2 patients in the postoperative period. Only one study reported complete remission in 2 patients. In general, however, 52-95% of patients reported some improvement. There was no consistent, objective measure of improvement in these studies. We conclude that repeat thymectomy for patients with refractory myasthenia gravis after previous thymectomy is safe especially for patients whose first procedure was transcervical. Complete remissions are rare but, in these small series, 60-70% of patients report improvement. Clinical improvement appears to be associated with the presence of residual thymic tissue at the second operation, but these cannot be reliably identified on preoperative imaging. Patient selection remains driven by symptoms.