合并重症肌无力对胸腺瘤术后患者预后的影响

目的评价合并重症肌无力(Myasthenia gravis,MG)对胸腺瘤术后患者预后的影响,并对可能影响胸腺瘤术后患者预后的其他相关因素进行分析。方法回顾性分析2002年9月—2018年1月中国人民解放军总医院第八医学中心收治的187例胸腺瘤术后患者的临床资料,其中胸腺瘤合并重症肌无力患者139例(MG组),单纯胸腺瘤患者48例(非MG组)。采用1∶1最邻近匹配法对两组患者行倾向性评分匹配(Propensity score matching,PSM),使组间协变量达到均衡。采用Kaplan-Meier法计算生存率,Log-rank检验比较两组患者生存曲线的差异,采用Cox比例风险回归模型对总体胸腺瘤术后患者进行单因素和多因素分析。结果匹配前,MG组(n=139)与非MG组(n=48)患者的性别(χ^2=4.180,P=0.041)、确诊年龄(χ^2=8.590,P=0.003)、WHO组织学分型(χ^2=4.764,P=0.029)以及是否接受术后辅助化疗(χ^2=5.627,P=0.018)均存在统计学差异。两组患者的5年和10年生存率分别为92.4%和69.6%,76.9%和65.2%,两组生存曲线差异有统计学意义(P=0.002)。经PSM法匹配后,两组共44对患者成功匹配,两组患者临床基线资料差异均无统计学意义(P>0.05)。两组患者的5年和10年生存率分别为90.8%和67.3%,85.6%和62.5%,两组生存曲线差异有统计学意义(P=0.015)。总体胸腺瘤术后患者的单因素分析显示,Masaoka分期(HR=0.237,95%CI0.111~0.504,P<0.001)、切除状态(HR=0.250,95%CI0.096~0.654,P=0.005)、术后辅助化疗(HR=0.367,95%CI0.179~0.751,P=0.006)以及合并重症肌无力(HR=0.336,95%CI0.162~0.669,P=0.003)可能与胸腺瘤术后患者的生存预后相关。多因素分析结果显示,Masaoka分期是影响胸腺瘤术后患者生存的独立预后因素(HR=0.317,95%CI0.140~0.720,P=0.006),合并重症肌无力则为影响胸腺瘤术后患者生存的保护性因素(HR=0.445,95%CI0.204~0.971,P=0.042)。结论对于可手术的胸腺瘤患者,合并重症肌无力较不伴重症肌无力的患者生存预后更好。Masaoka分期是影响胸腺瘤患者生存的独立预后因素。

Effect of myasthenia gravis on prognosis of thymoma patients

Objective The aim of this study was to evaluate the effect of myasthenia gravis(MG)on the prognosis of patients with thymoma after surgery,and to analyze other related factors affecting the prognosis of patients with thymoma after surgery. Methods The clinical data of 187 thymoma patients in the 8th Medical Center of the Chinese People's Liberation Army from September 2002 to January 2018 were retrospectively analyzed.Among them,139 thymoma patients complicated with myasthenia gravis(MG group)and 48 patients with simple thymoma(non-MG group).The 1∶1 nearest neighbor matching method was used to perform propensity score matching(PSM)for the two groups to achieve equilibrium within covariance.Kaplan-Meier method was used to calculate the survival rate,and log-rank test was preformed to analyze the differences in survival rates between MG and non-MG groups.Cox proportional hazard regression model was used for univariate and multivariate analyses for overall thymoma patients after surgery. Results Before the results were matched,the gender(χ²=4.180,P=0.041),age of diagnosis(χ²=8.590,P=0.003),WHO histological subtype(χ²=4.764,P=0.029)and postoperative adjuvant chemotherapy(χ²=5.627,P=0.018)were significantly different between the MG group(n=139)and the non-MG group(n=48).The 5-year overall survival rates of patients in the MG group and non-MG group were 92.4% and 69.6%,respectively,and the difference was statistically significant(Log-rank test,P=0.002).After matching by the PSM method,a total of 44 of patients were successfully matched in the two groups,and there was no significantly difference in clinical baseline data between the MG and non-MG groups(P＞0.05).The 5-year overall survival rates of patients in the MG group and non-MG group patients were 90.8% and 67.3%,respectively,and the difference was statistically significant(Log-rank test,P=0.015).Univariate analysis for patients with overall thymoma surgery showed that Masaoka stage(HR=0.237,95% CI:0.111~0.504,P＜0.001),resection status(HR=0.250,95% CI:0.096~0.654,P=0.005),postoperative adjuvant chemotherapy(HR=0.367,95% CI:0.179~0.751,P=0.006)and MG(HR=0.336,95% CI:0.162~0.669,P=0.003)might be associated with survival prognosis of thymoma patients after surgery.Cox multivariate analysis showed that Masaoka stage was an independent prognostic factor affecting the survival of thymoma patients after surgery(HR=0.317,95% CI:0.140~0.720,P=0.006),and MG affected protective survival factors of thymoma patients after surgery(HR=0.445,95% CI:0.204~0.971,P=0.042). Conclusion For patients with operable thymoma,thymoma patients with MG have a better prognosis than those without MG.Masaoka stage is an independent prognostic factor that affects the survival of thymoma patients.