基于Logistic回归分析建立产后压力性尿失禁的盆底超声评分模型及其诊断价值

Pelvic floor ultrasonic scoring model for postpartum stress urinary incontinence based on Logistic regression analysis and its diagnostic value

目的基于Logistic回归分析建立产后压力性尿失禁(SUI)的盆底超声评分模型,并探讨其诊断价值。方法选取我院236例产妇,根据产妇产后3个月内是否出现SUI将其分为SUI组38例和非SUI组198例。应用盆底超声获取膀胱颈移动度(BND)、尿道旋转角(URA)、静息及Valsalva状态下膀胱尿道后角(PUVA)和肛提肌裂孔面积(LHA)。采用多因素Logistic回归分析产后SUI的影响因素。根据危险因素的回归系数构建盆底超声评分模型;绘制受试者工作特征(ROC)曲线分析该评分模型对产后SUI的诊断价值。结果SUI组BND、URA及Valsalva状态下LHA和PUVA均明显高于非SUI组,差异均有统计学意义(均P<0.05)。多因素Logistic回归分析显示,BND、URA及Valsalva状态下LHA和PUVA是产后SUI的影响因素(OR=4.412、4.176、2.379、2.841,均P<0.05)。根据多因素Logistic回归分析建立盆底超声评分模型,模型总分0~6分。该评分模型诊断产后SUI的ROC曲线下面积为0.903(95%可信区间:0.780~0.953),以4分为截断值的敏感性和特异性分别为87.7%和78.8%;BND诊断产后SUI的ROC曲线下面积为0.843(95%可信区间:0.725~0.903),以12.2 mm为截断值的敏感性和特异性分别为85.6%和79.3%;二者曲线下面积比较差异有统计学意义(Z=2.437,P=0.016)。结论BND、URA、Valsalva状态下LHA及PUVA是产后SUI的影响因素,基于上述因素构建的盆底超声评分模型对产后SUI有一定诊断价值。

Objective To establish a pelvic floor ultrasonic scoring model for postpartum stress urinary incontinence(SUI)based on Logistic regression analysis,and to explore its clinical diagnostic value.Methods A total of 236 parturients in our hospital were selected and divided into SUI group(38 cases)and non-SUI group(198 cases)according to whether SUI occurred within 3 months after delivery.Pelvic floor ultrasound was used to obtain bladder neck descent(BND),urethral rotation angle(URA),posterior urethrovesial angle(PUVA)and levator hiatus area(LHA)in resting state and Valsalva state.Multivariate Logistic regression analysis was used to determine the influencing factors of postpartum SUI.According to the regression coefficients of risk factors,a pelvic floor ultrasonic scoring model was established,and receiver operating characteristic(ROC)curve was drawn to analyze the diagnostic value of the the scoring model for postpartum SUI.Results The BND,URA,and LHA,PUVA in Valsalva state of SUI group were significantly higher than those of non-SUI group,the differences were statistically significant(all P<0.05).Multivariate Logistic regression analysis showed that BND,URA,and LHA,PUVA in Valsalva state were the influencing factors of postpartum SUI(OR=4.412,4.176,2.379,2.841,all P<0.05).According to the results of multivariate Logistic regression analysis,a pelvic floor ultrasonic scoring model was established,with a total score of 0~6.The area under the ROC curve of the scoring model for diagnosing postpartum SUI was 0.903(95%CI:0.780~0.953),the sensitivity and specificity were 87.7%and 78.8%when the cut-off value was 4 points.The area under ROC curve of BND for diagnosing postpartum SUI was 0.843(95%CI:0.725~0.903),the sensitivity and specificity were 85.6%and 79.3%when the cut-off value was 12.2 mm.The area under the curve between the two was statistically significant(Z=2.437,P=0.016).Conclusion BND,URA,and LHA,PUVA in Valsalva state are the influencing factors of postpartum SUI,and the pelvic floor ultrasonic scoring model based on the above factors has certain diagnostic value for postpartum SUI.