European journal of obstetrics, gynecology, and reproductive biology · 2023
First-trimester prediction of preterm prelabour rupture of membranes incorporating cervical length measurement.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Rode Line, Wulff Camilla B, Ekelund Charlotte K, Hoseth Eva, Petersen Olav B, Tabor Ann, El-Achi Vanessa, Hyett Jon A, McLennan Andrew C
Аннотация
OBJECTIVES: To examine early pregnancy risk factors for preterm prelabour rupture of membranes (PPROM) and develop a predictive model.
STUDY DESIGN: Retrospective analysis of a cohort of mixed-risk singleton pregnancies screened in the first and second trimesters in three Danish tertiary fetal medicine centres, including a cervical length measurement at 11-14 weeks, at 19-21 weeks and at 23-24 weeks of gestation. Univariable and multivariable logistic regression analyses were employed to identify predictive maternal characteristics, biochemical and sonographic factors. Receiver operating characteristic (ROC) curve analysis was used to determine predictors for the most accurate model.
RESULTS: Of 3477 screened women, 77 (2.2%) had PPROM. Maternal factors predictive of PPROM in univariable analysis were nulliparity (OR 2.0 (95% CI 1.2-3.3)), PAPP-A < 0.5 MoM (OR 2.6 (1.1-6.2)), previous preterm birth (OR 4.2 (1.9-8.9)), previous cervical conization (OR 3.6 (2.0-6.4)) and cervical length ≤ 25 mm on transvaginal imaging (first-trimester OR 15.9 (4.3-59.3)). These factors all remained statistically significant in a multivariable adjusted model with an AUC of 0.72 in the most discriminatory first-trimester model. The detection rate using this model would be approximately 30% at a false-positive rate of 10%. Potential predictors such as bleeding in early pregnancy and pre-existing diabetes mellitus affected very few cases and could not be formally assessed.
CONCLUSIONS: Several maternal characteristics, placental biochemical and sonographic features are predictive of PPROM with moderate discrimination. Larger numbers are required to validate this algorithm and additional biomarkers, not currently used for first-trimester screening, may improve model performance.
STUDY DESIGN: Retrospective analysis of a cohort of mixed-risk singleton pregnancies screened in the first and second trimesters in three Danish tertiary fetal medicine centres, including a cervical length measurement at 11-14 weeks, at 19-21 weeks and at 23-24 weeks of gestation. Univariable and multivariable logistic regression analyses were employed to identify predictive maternal characteristics, biochemical and sonographic factors. Receiver operating characteristic (ROC) curve analysis was used to determine predictors for the most accurate model.
RESULTS: Of 3477 screened women, 77 (2.2%) had PPROM. Maternal factors predictive of PPROM in univariable analysis were nulliparity (OR 2.0 (95% CI 1.2-3.3)), PAPP-A < 0.5 MoM (OR 2.6 (1.1-6.2)), previous preterm birth (OR 4.2 (1.9-8.9)), previous cervical conization (OR 3.6 (2.0-6.4)) and cervical length ≤ 25 mm on transvaginal imaging (first-trimester OR 15.9 (4.3-59.3)). These factors all remained statistically significant in a multivariable adjusted model with an AUC of 0.72 in the most discriminatory first-trimester model. The detection rate using this model would be approximately 30% at a false-positive rate of 10%. Potential predictors such as bleeding in early pregnancy and pre-existing diabetes mellitus affected very few cases and could not be formally assessed.
CONCLUSIONS: Several maternal characteristics, placental biochemical and sonographic features are predictive of PPROM with moderate discrimination. Larger numbers are required to validate this algorithm and additional biomarkers, not currently used for first-trimester screening, may improve model performance.