Martinez-Portilla Raigam Jafet, Hawkins-Villarreal Ameth, Alvarez-Ponce Pamela, Chinolla-Arellano Zarela Lizbeth, Moreno-Espinosa Ana Lisbeth, Sandoval-Mejia Ana Lilia, Moreno-Uribe Nallely
Fetal Medicine Unit, Obstetrics Department, Hospital "La Raza" Medical Center, Autonomous University of Mexico, Mexico City,
Fetal i+D Fetal Medicine Research Center, BCNatal - Barcelona Center for Maternal-Fetal and Neonatal Medicine (Hospital Clínic and Hospital Sant Joan de Deu), Institut Clínic de Ginecologia, Obstetricia i Neonatologia, Institut d'Investigacions Biomèdiques August Pi i Sunyer, Universitat de Barcelona, Barcelona,
Fetal Diagn Ther. 2019;45(3):168-175. doi: 10.1159/000488080. Epub 2018 Apr 10.
To assess the added value of maternal serum levels of IL-6 in women with preterm-prelabor rupture of membranes (PPROM) as a non-invasive test for the prediction of histological chorioamnionitis (HCA).
This was a prospective cohort study of pregnant women between 20 + 0 and 36 + 6 weeks of gestation with a confirmed diagnosis of PPROM. Logistic regression models were created for the prediction of HCA and compared by assessing the improvement in their Naegelkerke R2 as a measure of goodness of fit. Predictive performance of all models was assessed by receiver operating characteristics curve (ROC) analysis and compared by the DeLong method.
From 47 women with PPROM, 31 (66%) developed HCA. Maternal serum IL-6 ≥19.5 pg/dL was the best cut-off point for the prediction of HCA (OR = 15; 95% CI: 3.6-61; p < 0.01). A model comprising maternal characteristics and IL-6 ≥19.5 pg/dL showed an area under the curve of 0.85 (95% CI: 0.74-0.95), significantly improving the previous models of IL-6 ≥19.5 pg/dL (R2: 23.3 vs. 34.1%; p = 0.01) or maternal characteristics (R2: 8.4 vs. 34.1%; p < 0.01).
A model comprising maternal serum levels of IL-6 plus maternal characteristics proves to be a good non-invasive predictor of HCA.
评估胎膜早破（PPROM）孕妇血清白细胞介素-6（IL-6）水平作为预测组织学绒毛膜羊膜炎（HCA）的非侵入性检测方法的附加价值。
这是一项对妊娠20 + 0至36 + 6周、确诊为PPROM的孕妇进行的前瞻性队列研究。建立逻辑回归模型以预测HCA，并通过评估其奈格尔克R2的改善情况作为拟合优度的衡量指标进行比较。所有模型的预测性能通过受试者工作特征曲线（ROC）分析进行评估，并采用德龙方法进行比较。
47例PPROM孕妇中，31例（66%）发生了HCA。孕妇血清IL-6≥19.5 pg/dL是预测HCA的最佳切点（OR = 15；95% CI：3.6 - 61；p < 0.01）。一个包含孕妇特征和IL-6≥19.5 pg/dL的模型显示曲线下面积为0.85（95% CI：0.74 - 0.95），显著改善了之前仅IL-6≥19.5 pg/dL的模型（R2：23.3%对34.1%；p = 0.01）或仅孕妇特征的模型（R2：8.4%对34.1%；p < 0.01）。
一个包含孕妇血清IL-6水平和孕妇特征的模型被证明是HCA的良好非侵入性预测指标。
