OBJECTIVES The goal of this study was to identify the relationship between clinical characteristics and the occurrence of postoperative myasthenia gravis (PMG) in patients with thymomas and to further identify the relationship between PMG and prognosis. METHODS Thymoma patients who had surgery at the First Affiliated Hospital of Sun Yat-sen University between July 2004 and July 2016 were reviewed and those who had no previous symptoms of myasthenia gravis were selected for further investigation. In total, 229 patients were included in the study; their clinical characteristics were gathered and analysed. RESULTS Among the 229 patients, 19 (8.3%) had PMG. The time between the operation and the onset of myasthenia gravis was 134 days on average (range 2–730 days). Patients experiencing PMG showed a lower rate of complete thymoma resection (73.7% vs 91.4%; P = 0.014) and total thymectomy (63.2% vs 82.9%; P = 0.035) compared with those who did not. Univariable and multivariable logistic regression revealed that thymomectomy [odds ratio (OR) 2.81, 95% confidence interval (CI) 1.02–7.77; P = 0.047] and incomplete tumour resection (OR 3.79, 95% CI 1.20–11.98; P = 0.023) were associated with the occurrence of PMG. Multivariable Cox regression showed that the PMG was not related to overall survival (P = 0.087). CONCLUSIONS This study revealed that incomplete tumour resection and thymomectomy were independent risk factors for PMG in thymoma patients with no previous history of myasthenia gravis.
