ObjectiveTo analyze the clinical outcomes of extended thymectomy for myasthenia gravis (MG) patients under different surgical approaches, and to determine the factors affecting the prognosis of MG. MethodsThe MG patients who underwent extended thymectomy from January 2014 to March 2021 in our hospital were retrospectively collected. According to the surgical approach, they were divided into a subxiphoid group and an intercostal group, and the perioperative results and prognosis were compared between the two groups. A “good outcome” was defined as complete stable remission (CSR), pharmacological remission (PR) or minimal manifestations state (MMS); a “poor outcome” was defined as outcomes worse than MMS. Univariate and multivariate logistic regression analyses were performed to assess the factors associated with the good outcomes. ResultsA total of 187 MG patients were included in the study, including 82 males and 105 females, with a median age of 50 (36, 60) years. There were 134 patients in the intercostal group and 53 patients in the subxiphoid group. Compared with the intercostal group, although the operation time of the subxiphoid group was longer [200.0 (172.0, 232.0) min vs. 141.0 (118.0, 169.0) min, P<0.001], the intraoperative blood loss was less [10.0 (10.0, 20.0) mL vs. 20.0 (10.0, 50.0) mL, P<0.001], the postoperative hospital stay was shorter [3.0 (2.5, 4.0) d vs. 5.0 (3.0, 7.0) d, P<0.001], and the incidence of complications was lower [1 (1.9%) vs. 26 (19.4%), P=0.001]. A total of 159 (85.0%) patients were followed up for a median period of 46 (13, 99) months, with a good outcome rate of 90.6% and CSR rate of 33.3%. There were no statistical differences in PR, MMS or overall good outcome rates between the two groups (P>0.05). Multivariate logistic analysis showed that age≤50 years was an independent predictor for "good outcome" of MG patients. ConclusionExtended thymectomy via subxiphoid for MG is a safe, feasible and effective surgical approach.