摘要
目的 分析围绝经期异常子宫出血(AUB)患者宫腔镜术后复发的风险因素,基此构建诺莫图模型,并评价其临床预测患者术后复发的价值.方法 前瞻性选取2020年1月至2023年8月濮阳惠民医院收治的200例围绝经期AUB患者作为研究对象,所有患者均接受宫腔镜手术,术后6个月失访5例,根据是否复发分为复发组57例和未复发组138例.比较两组患者的一般资料和实验室指标.采用Lasso-Logistic回归方程筛选影响AUB患者宫腔镜术后复发的因素,采用R语言软件构建诺莫图模型,并采用受试者工作特征(ROC)曲线、校准曲线、决策曲线评价诺莫图模型对AUB患者宫腔镜术后复发的预测效能.结果 复发组患者的体质量指数(BMI)、AUB时间、子宫内膜厚度、术后使用曼月乐占比、慢性疾病史占比、妇科疾病史占比分别为(25.34±1.25)kg/m2、(62.62±5.58)d、(8.92±1.34)mm、15.79%、29.82%、24.56%,明显高于未复发组的(21.65±1.34)kg/m2、(40.41±4.33)d、(6.78±1.01)mm、3.62%、14.49%、7.97%,血清血管通透因子(VEGF)、细胞生长因子(bFGF)分别为(23.55±5.12)pg/mL、(9.91±1.34)pg/mL,明显低于未复发组的(30.34±4.68)pg/mL、(14.40±3.32)pg/mL,差异均有统计学意义(P<0.05);Lasso-Logistic回归方程分析结果显示,BMI、慢性疾病史、术后使用曼月乐及血清VEGF、bFGF均是围绝经期AUB患者宫腔镜术后复发影响因素(P<0.05);根据Lasso-Logistic回归方程分析结果构建术后复发诺莫图模型,模型评价及验证结果显示,该模型预测围绝经期AUB患者宫腔镜术后复发的曲线下面积(AUC)为0.801(95%CI:0.733～0.870),预测术后复发概率与实际概率相吻合,当阈值概率处于15%～100%时,临床净获益率最大.结论 BMI、慢性疾病史、术后使用曼月乐及血清VEGF、bFGF均是围绝经期AUB患者宫腔镜术后复发影响因素,根据上述因素构建诺莫图模型对术后复发具有良好预测价值及临床正向净获益,可为临床预测术后复发提供参考,以指导后续临床决策.
Abstract
Objective To analyze the risk factors for recurrence of abnormal uterine bleeding(AUB)after hys-teroscopy in patients with perimenopause,to construct a nomogram model based on the risk factors,and to evaluate the model's clinical predictive value for postoperative recurrence.Methods A prospective study was conducted on 200 pa-tients with perimenopausal AUB admitted to Puyang Huimin Hospital from January 2020 to August 2023.All patients underwent hysteroscopy surgery.Five patients were lost to follow-up within 6 months after surgery.Patients were divid-ed into a recurrence group(57 cases)and a non-recurrence group(138 cases).The general information and laboratory in-dicators of the two groups of patients were compared.The Lasso-Logistic regression equation was used to screen the fac-tors that affect the recurrence of AUB patients after hysteroscopy.The nomograph model was constructed using the R language software.The receiver operating characteristic(ROC)curve,calibration curve,and decision curve were drawn to evaluate the efficacy of the nomogram model for predicting the recurrence of AUB patients after hysterosco-py.Results The body mass index(BMI),AUB duration,endometrial thickness,proportion of postoperative use of Mi-rena,proportion of chronic disease history,and proportion of gynecological disease history in the recurrence group were(25.34±1.25)kg/m2,(62.62±5.58)d,(8.92±1.34)mm,15.79%,29.82%,and 24.56%,respectively,which were signifi-cantly higher than(21.65±1.34)kg/m2,(40.41±4.33)d,(6.78±1.01)mm,3.62%,14.49%,and 7.97%in the non-recurrent group;the serum vascular endothelial growth factor(VEGF)and basic fibroblast growth factor(bFGF)were(23.55±5.12)pg/mL and(9.91±1.34)pg/mL,respectively,which were significantly lower than(30.34±4.68)pg/mL and(14.40±3.32)pg/mL in the non-recurrent group;the differences were statistically significant(P<0.05).The results of the Las-so-Logistic regression equation analysis showed that BMI,a history of chronic diseases,postoperative use of Mirena,se-rum levels of VEGF and bFGF were all factors that affected the recurrence of perimenopausal AUB patients after hyster-oscopy(P<0.05).Based on the analysis results of the Lasso-Logistic regression equation,a nomogram model for post-operative recurrence was constructed.The model evaluation and validation results showed that the area under the curve(AUC)of the model for predicting postoperative recurrence in perimenopausal area under the curve(AUC),patients af-ter hysteroscopy was 0.801(95%CI:0.733-0.870),and the predicted probability of postoperative recurrence was con-sistent with the actual probability.When the threshold probability was between 15%and 100%,the clinical net benefit rate was maximized.Conclusion BMI,history of chronic diseases,postoperative use of Mirena,serum VEGF,and bFGF are all influencing factors of the recurrence of perimenopausal AUB patients after hysteroscopy.Constructing a nomogram model based on these factors has good predictive value and positive clinical net benefit for postoperative re-currence,which can provide a reference for clinical prediction of postoperative recurrence and guide subsequent clini-cal decisions.
维普
万方数据