个体化预测育龄期子宫肌瘤患者宫腔镜术后复发风险列线图模型构建与验证

Construction and validation of a nomogram model for individualized prediction of postoperative recurrence risk of hysteros⁃copy for uterine fibroids of childbearing age

目的探讨个体化预测育龄期子宫肌瘤患者宫腔镜术后复发风险列线图模型的构建与验证。方法选取2016年1月至2018年4月在苏州九龙医院行宫腔镜下子宫肌瘤剔除术的育龄期患者289例,术后3年内定期随访,最终纳入随访资料齐全的236例患者为研究对象。收集所有纳入患者的临床资料,根据随访期间患者是否复发,分为复发组(n=49)和未复发组(n=187)。通过单因素和多因素logistic回归分析,确定影响术后复发的危险因素,随后建立列线图模型并进行验证。结果纳入的236例患者中,3年内复发率为20.76%(49/236)。单因素分析显示,两组患者宫腔黏连、肌瘤数目、肌瘤分型、肌瘤直径、血清雌激素受体、孕酮受体、B淋巴细胞瘤-2基因水平水平比较,差异有统计学意义(P<0.05)。多因素logistic回归分析显示,宫腔黏连、肌瘤数目、肌瘤分型、肌瘤直径、ER、PR、Bcl-2水平均为育龄期子宫肌瘤患者宫腔镜术后复发的独立危险因素(P<0.05)。根据上述7项独立危险因素构建列线图模型及验证发现,列线图模型预测育龄期子宫肌瘤患者术后复发风险的AUC为0.971(95%CI:0.951~0.990)、灵敏度为0.898、特异度为0.936;Hosmer-Lemeshow偏差性检验显示,χ2=2.866(P=0.943);Bootstrap内部验证显示,该列线图模型校准曲线的MAE为0.015。结论育龄期子宫肌瘤患者术后复发危险因素构建列线图模型具有较好的区分度、校准度及预测能力,可为宫腔镜下子宫肌瘤剔除术后复发高危患者筛查和干预方案提供参考。

Objective To explore the construction and verification of a nomogram model for individualized prediction of postoperative recurrence risk of hysteroscopy for uterine fibroids of childbearing age.Methods A total of 289 patients of childbearing age who underwent hysteroscopic myomectomy in Department of Gynecology,Suzhou Kowloon Hospital from January 2016 to April 2018 were selected and fol⁃lowed up regularly within three years after the operation.Finally,236 patients with complete follow-up data were included as the research ob⁃jects.The clinical data of all included patients were collected and divided into recurrence group(n=49)and non-recurrence group(n=187)according to whether the patients recurred during the follow-up period.Through univariate and multivariate logistic regression analysis,the risk factors affecting postoperative recurrence were determined,and then a nomogram model was established and verified.Results Among the 236 patients included,the recurrence rate within three years was 20.76%(49/236).Univariate analysis showed that there were statistically significant differences between the two groups of patients in terms of uterine adhesions,fibroids number,types of fibroids,fi⁃broids diameter,ER,PR,and Bcl-2 levels(P<0.05).Multivariate logistic regression analysis showed that intrauterine adhesions,number of fibroids,types of fibroids,fibroids diameter,ER,PR,and Bcl-2 levels were all independent influencing factors for postoperative recur⁃rence of hysteroscopy in uterine fibroids of childbearing age(P<0.05).According to the above seven independent influencing factors,a no⁃mogram model was constructed and verified,and it was found that the AUC of this nomogram model predicting the risk of recurrence in patients with uterine fibroids of childbearing age was 0.971(95%CI:0.951~0.990),the sensitivity was 0.898,and the specificity was 0.936.Hosmer-Lemeshow deviation test showedχ2=2.866(P=0.943).Bootstrap internal verification showed that the MAE of the calibra⁃tion curve of the nomogram model was 0.015.Conclusion The nomogram model constructed based on the risk factors of postoperative re⁃currence in patients with uterine fibroids of childbearing age has good discrimination and calibration,and good predictive ability.It can be used for patients with high risk of recurrence after hysteroscopic uterine fibroids removal,providing reference for the formulation of screen⁃ing and intervention programs.