多囊卵巢综合征患者冻融周期活产率的影响因素及诺莫模型的建立

Influencing factors of live birth rate in patients with polycystic ovary syndrome after frozen embryo transfer and establishment of nomogram model

目的分析影响多囊卵巢综合征(PCOS)患者冻融周期活产率的因素并构建诺莫预测模型。方法纳入取卵后冻融周期移植的129例PCOS患者为训练集,通过单因素及多因素分析筛选活产率的影响因素,运用R软件构建诺莫模型,然后经过非同期61例患者构成的验证集外部验证。结果训练集和验证集的活产率分别为57.56%和42.62%,有无优质胚胎(P=0.002)、移植胚胎数目(P=0.021)、身体质量指数(BMI)(P=0.024)及是否血脂异常(P=0.029)是预测活产率的独立影响因素,由此构建的诺莫模型ROC曲线下面积(AUC)在训练集和验证集分别为0.765(95%CI为0.632~0.859)和0.713(95%CI为0.639~0.782),校准曲线反映模型亦较好的校准能力。结论有无优质胚胎、移植胚胎数目、BMI及是否血脂异常是影响PCOS患者首次冻融周期妊娠结局的影响因素,该模型可以可视化、量化地体现患者的预后,为临床决策及医患沟通提供直观的信息。

Objective To analyze the factors affecting the live birth rate of polycystic ovary syndrome(PCOS)patients after frozen embryo transfer,and establish a nomogram prediction model.Methods One hundred and twenty-nine patients with PCOS after frozen embryo transfer(FET)were included.The influencing factors of live birth rate were screened by univariate and multivariate analysis.The nomogram model was constructed by using R software,and then external verification was conducted by the validation set composed of 61 patients from different periods.Results The live birth rate in women on training cohort and validation cohort was 57.56 and 42.62%,respectively.Four variates significantly associated with live birth rate were high quality of embryo(P=0.002),number of transplanted embryos(P=0.021),body mass index(P=0.024)and serum lipid level(P=0.029).This predictive model built on above variates showed good discrimination and calibration abilities,with an area under the curve(AUC)of 0.765(95%CI 0.632~0.859)for the training cohort and 0.713(95%CI 0.639~0.782)for the validation cohort.Conclusion Our evidence-based nomogram presents graphically risk factors and prognostic models for fist FET outcomes in women with PCOS,which can offer useful guidance to clinicians and patients for individual therapy.