Value of characteristics of magnetic resonance imaging in predicting the increta type of pernicious placenta previa by using the sequential multivariate logistic regression analysis
Lu Fangfei1, Li Ming1, Sun Yi1, Hu Anning1, Wang Kun1, Chen Wenqian1, Zhu Xiaoqian1, Yan Xian1, Zhang Bing1, Chen Fei2
1Department of Radiology, the Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing 210008, China; 2Department of Radiology, the Yancheng School of Clinical Medicine of Nanjing Medical University (Yancheng Third People's Hospital), Yancheng 224005, China
摘要目的 探讨MRI征象预测凶险性前置胎盘(PPP)植入类型的价值。方法 回顾性研究。纳入南京大学医学院附属鼓楼医院67例PPP伴胎盘植入患者的临床及MRI资料。患者年龄21~48(31.85±5.12)岁,孕周28~38(32.37±2.47)周。根据不同类型PPP伴胎盘植入的临床及病理诊断标准,将患者分为粘连组(25例)、植入组(29例)和穿透组(13例)。对3组患者的临床资料及MRI征象采用Kruskal-Wallis H检验进行单因素分析,对可能相关因素进行有序多元logistic回归多因素分析,并进一步分析独立危险因素对PPP植入类型的预测价值。结果 单因素分析结果显示,子宫肌层中断/肌层内见胎盘信号、子宫局限性隆突/胎盘向外膨出、胎盘内异常血管、胎盘信号不均匀、子宫下部膨大且宫颈内口胎盘不规则以及膀胱子宫间隙低信号带不连续,为不同植入类型的可能相关因素(P值均＜0.1)。其中,膀胱子宫间隙低信号带不连续[(X1,比值比(OR)20.25)]、子宫肌层中断/肌层内见胎盘信号(X2, OR 13.87)、子宫局限性隆突/胎盘向外膨出(X3, OR 6.16)]和子宫下部膨大且宫颈内口胎盘不规则(X4, OR 4.55),为不同类型胎盘植入的独立危险因素(P值均＜0.05)。建立预测胎盘植入类型的回归模型为:logit(P)=C+3.01X1 +2.63X2+1.82X3+1.52X4(预测“植入”和“穿透”的常量C分别为1.08和5.54),该模型预测3种植入类型的准确率为71.64%,将粘连组与其他类型组相区分的准确率为85.07%。结论 MRI征象对PPP伴胎盘植入类型的诊断具有较高的预测价值,对临床术前评估胎盘植入类型选择不同治疗方案具有一定的指导意义。
Abstract：Objective This study aims to explore the value of image characteristics of magnetic resonance (MR) in predicting the increta type of pernicious placenta previa (PPP).Methods The clinical and the MR data of 67 patients with PPP and placenta increta in the Affiliated Drum Tower Hospital of Nanjing University Medical School were retrospectively analyzed, aged from 21 years to 48 years (31.85 years±5.12 years) and gestational age ranging from 28 weeks to 38 weeks (32.37 weeks±2.47 weeks). According to the clinical and the pathological diagnostic criteria of different types of PPP with placenta implantation, patients were divided into accreta (25 cases), increta (29 cases), and percreta (13 cases) groups. The clinical data and the MR image characteristics of the three groups of patients were analyzed using the single-factor analysis through the Kruskal-Wallis H test to screen the possible correlation factors of different increta types. The sequential multivariate logistic regression was used for the multivariate analysis of possible correlation factors (MR image characteristics), and the independent risk factors of different increta types were screened. The value of independent risk factors (MR image characteristics) in predicting the increta type of PPP was further analyzed.Results The single-factor analysis showed that the myometrium interrupt/placenta signal present in the myometrium, localized protuberance of uterus/placenta bulges outward, abnormal placental vessels, inhomogeneous placental signal, enlarged lower part of the uterus with irregular endocervical placenta, and discontinuous low signal bands in the cysto-uterine space were the possible related factors of different increta types (all P values＜0.1). Among these factors, discontinuous low-signal bands in the cysto-uterine space (X1, OR=20.25), myometrium interrupt/placenta signal present in the myometrium (X2, OR=13.87), localized protuberance of uterus/placenta bulges outward (X3, OR=6.16), and enlarged lower part of the uterus with irregular endocervical placenta (X4, OR=4.55) were the independent risk factors of different increta types (all P values＜0.05). The regression model logit(P)=1.08/5.54+3.01X1+2.63X2+1.82X3+1.52X4 was established to predict the type of placenta increta. The accuracy of this model in predicting the three types of implantation was 71.64%, and the accuracy in differentiating the group accreta with other groups was 85.07%.Conclusions MR image characteristics have a high value in predicting the increta type of PPP, showing certain guiding significance for the clinical preoperative evaluation of placenta increta types and selection of different treatment plans.
陆芳菲, 李茗, 孙艺, 胡安宁, 王坤, 陈文倩, 朱小倩, 闫娴, 张冰, 陈飞. 有序多元logistic回归分析MRI征象预测凶险性前置胎盘植入类型的价值[J]. 中华解剖与临床杂志, 2021, 26(3): 259-264.
Lu Fangfei, Li Ming, Sun Yi, Hu Anning, Wang Kun, Chen Wenqian, Zhu Xiaoqian, Yan Xian, Zhang Bing, Chen Fei. Value of characteristics of magnetic resonance imaging in predicting the increta type of pernicious placenta previa by using the sequential multivariate logistic regression analysis. Chinese Journal of Anatomy and Clinics, 2021, 26(3): 259-264.
Yu L, Hu KJ, Yang HX. A retrospective analysis on the pernicious placenta previa from 2008 to 2014[J]. Chin J Obstet Gynecol, 2016, 51(3): 169-173. DOI:10.3760/cma.j.issn.0529-567x.2016.03.002.
[2]
Wang J, Shi X, Li Y, et al. Prophylactic intraoperative uterine or internal iliac artery embolization in planned cesarean for pernicious placenta previa in the third trimester of pregnancy: an observational study (STROBE compliant)[J]. Medicine (Baltimore), 2019, 98(44): e17767. DOI:10.1097/MD.0000000000017767.
[3]
Wang Y, Hu C, Pan N, et al. Prophylactic uterine artery embolization in second-trimester pregnancy termination with complete placenta previa[J]. J Int Med Res, 2019,47(1):345-352. DOI:10.1177/0300060518801455.
Hu L, Liu Y, Li S. Value of MRI in the diagnosis of pernicious placenta previa in combination with placenta accreta[J]. Radiol Practice, 2019, 34(1): 45-49. DOI:10.13609/j.cnki.1000-0313.2019.01.009.
Zhang YX. Application of descending branch uterine artery ligation in treatment of placenta previa postpartum hemorrhage during caesarean section[J]. Chin J Anat Clin, 2015, 20(4): 356-358. DOI:10.3760/cma.j.issn.2095-7041.2015.04.019.
[6]
He Q, Li YL, Zhu MJ, et al. Prophylactic abdominal aortic balloon occlusion in patients with pernicious placenta previa during cesarean section: a systematic review and meta-analysis from randomized controlled trials[J]. Arch Gynecol Obstet, 2019, 300(5): 1131-1145. DOI:10.1007/s00404-019-05297-4.
Wang GB. Magnetic resonance imaging in the evaluation of pernicious placenta previa[J]. Journal of Shandong University (Health Science), 2016, 54(9): 7-9. DOI:10.6040/j.issn.1671-7554.0.2016.546.
[8]
Maurea S, Romeo V, Mainenti PP, et al. Diagnostic accuracy of magnetic resonance imaging in assessing placental adhesion disorder in patients with placenta previa: correlation with histological findings[J]. Eur J Radiol, 2018,106:77-84. DOI:10.1016/j.ejrad.2018.07.014.
Guo GM, Cao MR, Liu XP, et al. Regression analysis of MRI features in implantable pernicious placenta previa[J]. Journal of Clinical Radiology, 2018, 37(8): 1325-1328. DOI:10.13437/j.cnki.jcr.2018.08.021.
[10]
Wei Y, Luo J, Luo D. Comparison of efficacy between internal iliac artery and abdominal aorta balloon occlusions in pernicious placenta previa patients with placenta accrete[J]. Gynecol Obstet Invest, 2019, 84(4): 343-349. DOI:10.1159/000494493.
Zhang FJ, Cao MR, Liu BG, et al. Diagnostic value of MRI in invasively pernicious placenta previa[J]. J Pract Radiolo, 2015, 31(5): 797-801. DOI:10.3969/j.issn.1002-1671.2015.05.023.
[12]
Dai MJ, Jin GX, Lin JH, et al. Pre-cesarean prophylactic balloon placement in the internal iliac artery to prevent postpartum hemorrhage among women with pernicious placenta previa[J]. Int J Gynaecol Obstet, 2018,142(3):315-320. DOI:10.1002/ijgo.12559.
[13]
Li HT, Luo S, Trasande L, et al. Geographic variations and temporal trends in cesarean delivery rates in China, 2008-2014[J]. JAMA, 2017, 317(1): 69-76. DOI:10.1001/jama.2016.18663.
[14]
D'Antonio F, Iacovella C, Palacios-Jaraquemada J, et al. Prenatal identification of invasive placentation using magnetic resonance imaging: systematic review and meta-analysis[J]. Ultrasound Obstet Gynecol, 2014, 44(1): 8-16. DOI:10.1002/uog.13327.
2021, Vol. 26 
