不同年龄段不孕患者控制性超排卵中卵巢低反应的预测

目的研究不同年龄段不孕患者超排卵过程中卵巢低反应的发生情况,探讨不同年龄段患者卵巢低反应的相关因素及比较各预测指标的诊断价值。方法回顾性分析中山大学附属孙逸仙医院生殖医学中心2424个体外受精/卵细胞浆内单精子注射（IVF/ICSI）周期。按年龄分为4组,比较各组卵巢低反应的发生率及妊娠率;logistic回归分析基础FSH、基础LH、基础E2、窦卵泡数（AFC）、卵巢容积和BMI等与卵巢低反应的相关性,并根据回归结果计算联合指标的ROC曲线。结果 18～30岁组不孕患者卵巢低反应率为9.0%（77/852）,基础FSH与AFC联合预测卵巢低反应的ROC曲线下面积为0.726;31～35岁组卵巢低反应率为19.7%（172/871）,基础FSH、基础LH、AFC联合预测卵巢低反应的ROC曲线下面积为0.789;36～40岁组卵巢低反应率34.9%（190/545）,基础FSH、卵巢体积、AFC和年龄联合预测卵巢低反应的ROC曲线下面积为0.831;≥41岁组卵巢低反应率为69.2%（108/156）,AFC预测卵巢低反应的ROC曲线下面积为0.809。结论随着年龄增大,卵巢低反应发生率增加、妊娠率下降;不同年龄段预测卵巢低反应的指标不同,综合多指标的预测价值较单一指标预测价值高,建议综合多个有效指标评估卵巢的反应性。

Prediction of low ovarian response in infertile patients of different ages undergoing controlled ovarian hyperstimulation

Objective To evaluate the proportion of low ovarian response in different ages of infertility patients,and explore the value of relevant factors in predicting low ovarian response in different ages. Methods A total of 2 424 IVF/ICSI cycles in our centre were analyzed restrospectively. According to the ages,the patients were divided into 4 groups:18～30 years old;31～35 years old;36～40 years old; ≥41years old. The proportion of low ovarian response and clinical pregnancy rates were evaluated. Logistic regression analysis was used for prediction of the probability of low response and the relevant parameters (FSH、LH、AFC、ovarian volume、BMI、E2). The value of this trial was checked by analyzing the area under the ROC curve.Results In group of 18～30 years old,the proportion of low ovarian response was 9%,FSH and AFC were relative with low response. In group of 31～35 years old,the proportion of low ovarian response was 19.7%,FSH,LH and AFC were relative with low response; In group of 36～40 years old,the proportion of low ovarian response was 34.9%,and FSH,LH,ovarian volume and AFC were relative with low response; In group of≥41 years old,the proportion of low ovarian response was 69.2%,and only AFC were relative with low response. Conclusions The proportion of low ovarian response increased and the clinical pregnancy rates decreased along with age. The prediction value of relevant parameters in different ages was different. In clinical practice patients of different ages need to be evaluated ovarian response combining with effective parameters.