论文部分内容阅读
目的:探索行辅助生殖技术卵泡期长效长方案超促排卵治疗的患者获取最佳卵巢反应的相关指标，并建立卵巢反应性的诺曼预测模型。方法:本研究为回顾性队列研究，分析2018年7月1日至2019年7月30日期间于福建省妇幼保健院生殖医学中心接受辅助生殖助孕治疗的1289例患者临床资料，按照获卵数分为卵巢低反应组（获卵数≤5枚）164例，卵巢正常反应组（5枚18枚）234例。通过logistic回归筛选影响卵巢反应性的独立因素，即为模型入组变量，并依据模型中的回归系数建立诺曼预测模型。结果:三组患者的年龄、抗苗勒管激素（anti-Müllerian hormone，AMH）水平、基础窦卵泡计数（antral follicle count，AFC）差异均有统计学意义［（32.43±3.99）岁，（31.48±3.89）岁，（29.91±3.73）岁；（2.53±1.90）μg/L，（3.79±2.20）μg/L，（5.94±3.12）μg/L；10.24±3.10，14.50±3.29，19.81±3.44；均n P0.05）。三组患者行促排卵的促性腺激素（gonadotropin，Gn）起始剂量差异有统计学意义［（182.62±53.96）U、（166.79±48.20）U、（159.13±43.92）U，n P0.05）。多因素逐步回归分析显示，女方年龄［0.93（0.90~0.96），n P=0.007］、AFC［1.07（1.03~1.09），n P=0.001］、AMH［1.29（1.20~1.39），n P=0.001］、基础卵泡刺激素［0.79（0.73~0.86），n P=0.001］、黄体生成素［1.11（1.06~1.23），n P=0.010］、Gn起始剂量［1.00（1.00~1.01），n P=0.003］、Gn使用总量［1.00（0.99~1.00），n P=0.001］，是否为子宫内膜异位症［0.63（0.47~0.86），n P=0.001］和多囊卵巢综合征［0.30（0.22~0.91），n P=0.030］是超促排卵过程中发生卵巢不同反应的独立因素。根据上述因素构建卵巢反应性的预测模型，预测卵巢最佳反应状态的准确性为95%。用2019年8月1日至2019年10月30日期间该中心的306例同类患者数据对上述模型进行验证，共279例患者的预测卵巢反应（获卵数）与实际相符，符合度为91.2%。模型的一致性指数是0.71。n 结论:筛选出行卵泡期长效长方案超促排卵中影响卵巢反应性的相关因素，成功建立的诺曼模型能够有效、直观、可视化地预测超促排卵中的卵巢反应性。“,”Objective:To explore the relative factors for best ovarian response in patients undergoing assisted reproductive technology with follicular phase long-acting long protocol, and to establish a Nomogram prediction model of ovarian response.Methods:This retrospective cohort study analyzed the clinical data of 1289 patients who received assisted reproductive treatment in the Center for Reproductive Medicine of Fujian Maternity and Child Health Hospital from July 1, 2018 to July 30, 2019. According to the number of oocytes retrieved, there were 164 cases in the low ovarian response group (≤5 oocytes retrieved), 891 cases in the normal ovarian response group (the number of retrieved oocytes was >5, and ≤18), and 234 cases in the high ovarian response group (>18 oocytes retrieved). Independent factors affecting ovarian reactivity were screened by logistic regression, which were the model entry variables, and a Nomogram prediction model was established based on the regression coefficients in the model.Results:There were statistically significant differences in age, anti-Müllerian hormone (AMH) level and antral follicle count (AFC) among the three groups [32.43±3.99, 31.48±3.89, 29.91±3.73; (2.53±1.90) μg/L, (3.79±2.20) μg/L, (5.94±3.12) μg/L; 10.24±3.10, 14.50±3.29, 19.81±3.44; all n P0.05). The initial dosage of gonadotropin (Gn) used for ovarian hyperstimulation among the three groups was statistically different [(182.62±53.96) U, (166.79±48.20) U, (159.13±43.92) U,n P0.05). Multifactorial stepwise aggression analysis showed that female age [0.93(0.90-0.96),n P=0.007], AFC [1.07(1.03-1.09), n P=0.001], AMH [1.29(1.20-1.39), n P=0.001], basal follicle-stimulating hormone [0.79(0.73-0.86), n P=0.001], luteinizing hormone value [1.11(1.06-1.23), n P=0.010], initial dosage of Gn used [1.00(1.00-1.01), n P=0.003], total dosage of Gn usd [1.00(0.99-1.00), n P=0.001] and the presence or absence of diagnosis of endometriosis [0.63(0.47-0.86), n P=0.001] and polycystic ovary syndrome [0.30(0.22-0.91), n P=0.030] were independent factors for the occurrence of different ovarian responses during ovarian hyperstimulation. The prediction model of ovarian reactivity was constructed based on the above factors, and the accuracy of predicting the optimal ovarian response state was 95%. The above model was verified with 306 patients\' data from August 1, 2019 to October 30, 2019 in this center, and the predicted ovarian response (number of oocytes obtained) of a total of 279 patients was consistent with the actual situation, with a coincidence degree of 91.2%. The consistency index of the model was 0.71.n Conclusion:We screened out the relevant factors affecting ovarian response in patients undergoing assisted reproductive technology with follicular phase long-acting long protocol, and established a Nomogram prediction model of ovarian response, which could effectively, intuitively and visually predict ovarian reactivity in hyperstimulation.