影响剖宫产瘢痕妊娠术中大出血发生的高危因素分析及风险预测模型构建

Analysis of high risk factors affecting the occurrence of massive hemorrhage during cesarean scar pregnancy and the construction of risk prediction model

目的探讨影响剖宫产术后子宫瘢痕妊娠(cesarean scar pregnancy,CSP)术中大出血发生的高危因素,并构建风险预测模型。方法回顾性分析青岛市中心医院2008年10月至2017年1月收治的277例CSP患者的临床资料,根据治疗方案的不同,将患者分为建模组(170例)和校模组(107例),分别构建风险模型,并利用受试者工作特征曲线(receiver operating characteristic curve,ROC)检验两者的预测效能。根据最终的治疗结局,将建模组中行"子宫动脉栓塞术(uterine arterial embolization,UAE)+甲氨蝶呤"的73例患者定义为高危组,行"UAE+清宫术"的97例患者为低危组。将校模组中行"腹腔镜下双侧子宫动脉结扎术+病灶清除术"的30例患者定义为高危组,行"腹腔镜监视下清宫术"的77例患者为低危组。结果 (1)建模组中低危组与高危组患者在年龄、距前次手术时间、停经时间、治疗前β-人绒毛膜促性腺激素(beta human chorionic gonadotropin,β-h CG)、病灶最大径线、病灶前壁肌层厚度、早孕期阴道流血、超声分型、胎血管搏动等方面比较,差异有统计学意义(P <0. 05)。进一步行Logistic回归分析,发现年龄、停经时间、超声分型、病灶最大径线是独立性危险因素。(2)校模组中低危组与高危组患者在年龄、停经时间、病灶最大径线、超声分型、胎血管搏动等方面,差异有统计学意义(P <0. 05)。进一步行Logistic回归分析,发现年龄、停经时间、超声分型、病灶最大径线是独立性危险因素。(3)建模组的ROC曲线下面积为0. 793,95%CI为0. 765～0. 942,截断值P=0. 441。校模组的ROC曲线下面积为0. 793,95%CI为0. 632～0. 921,截断值P=0. 496。两种模型的AUC差异无统计学意义(P> 0. 05)。结论发病年龄小、停经时间长、包块型CSP、病灶径线大是CSP患者发生大出血的主要危险因素,临床医师对存在上述危险因素的患者应提高警惕。应用预测模型,当患者风险概率>44. 1%时,治疗方案应首先考虑UAE联合化疗或腹腔镜下双侧子宫动脉结扎术,病灶清除术。

Objective To explore the risk factors affecting the occurrence of massive hemorrhage in cesarean scar pregnancy （CSP） patients and to construct a risk prediction model. Methods Retrospectively analyzed the clinical data of 277 patients with CSP admitted to Qingdao Center Hospital from October 2008 to January 2017. According to the different treatment options, patients were divided into modeling group （ 170 cases） and school module group （ 107 cases）. Risk models were constructed separately and the predictive efficacy of the two was tested using the receiver operating characteristic curve （ROC）. Results ① In the low - risk group and high - risk group in the modeling group, the age, the time of the previous operation, the menstrual time, the treatment of human choroidal gonadotropin （ β - hCG）, the maximum diameter of the lesion, thickness of anterior wall of lesion, vaginal bleeding in early pregnancy, ultrasound classification, fetal pulsation, etc ; the differences were statistically significant （ P 〈 0. 05 ）. Further logistic regression analysis showed that age, menstrual time, ultrasound classification, and lesion maximum diameter were independent risk factors. ② There were significant differences in age, menstrual time, maximum lesion diameter, ultrasound classification, and fetal pulsation between the low - risk group and the high - risk group in school module group （ P 〈 0. 05 ）. Further logistic regression analysis showed that age, menstrual time, ultrasound classification, and lesion maximum diameter were independent risk factors. ③ The area under the ROC curve of the modeling group was 0. 793, 95 % CI was 0. 765 - 0. 942, and the cut - off value was P = 0. 441. The area under the ROC curve of the school module was 0. 793, 95 % CI was 0. 632 - 0. 921, and the cutoff value was P = 0. 496. There was no significant difference in AUC between the two models （ P 〉 0. 05 ）. Conclusion The age of onset is small, the menopause time is long, CSP, and the diameter of the lesion is the main risk factor for the occurrence of major bleeding in patients with cesarean scar pregnancy. Patients with these risk factors should be vigilant. Using the predictive model, when the patient＇s risk probability is 〉 44. 1%, the treatment plan should first consider uterine arterial embolization combined with chemotherapy or laparoscopic bilateral uterine artery ligation.