论文部分内容阅读
目的:通过孕妇血清学指标建立不良妊娠结局的预测模型。方法:采用回顾性调查的方法,对2008年～2010年深圳地区唐氏综合征筛查神经管缺陷高风险118例孕妇不良妊娠结局的血清学指标进行逐步判别,建立相应的判别模型。结果:118例中,随访确诊61例不良妊娠。逐步判别分析将孕妇年龄及甲胎蛋白水平纳入判别模型后判别效果最好;回顾性误判分析显示有17例发生误判,总误判率为14.4%,总准确率为85.6%;而交叉核实法显示有18例发生误判,总误判率为15.3%,总准确率为84.7%。判别模型的灵敏度为83.3%,特异性为90.0%。结论:利用孕妇血清学指标建立判别模型,有利于提高高风险孕妇不良妊娠结局预测的准确性,有助于指导临床诊断。
Objective: To establish the prediction model of adverse pregnancy outcome by pregnant women serological indicators. Methods: A retrospective survey was conducted to determine the serological markers of 118 pregnancies with adverse pregnancy outcomes of Down’s syndrome screening for screening of neural tube defects in Shenzhen from 2008 to 2010. The corresponding discriminant models were established. Results: In 118 cases, 61 cases of adverse pregnancy were diagnosed at follow-up. The discriminant analysis showed that 17 cases were misjudged, the total false positive rate was 14.4% and the total accuracy was 85.6% Verification showed that 18 cases were miscarriage of justice, the total false positive rate was 15.3%, with a total accuracy of 84.7%. The discriminative model has a sensitivity of 83.3% and a specificity of 90.0%. Conclusion: Establishing the discriminant model by using the serological markers of pregnant women will help to improve the accuracy of prediction of adverse pregnancy outcomes in high-risk pregnant women and help to guide the clinical diagnosis.