Hanna Åmark1, Magnus Westgren2, Martina Persson1,3. 1. Department of Clinical Science and Education, Unit of Obstetrics and Gynecology, Karolinska Institute, Södersjukhuset, Stockholm, Sweden. 2. Department of Clinical Sciences, Intervention & Technology, Karolinska Institutet, Stockholm, Sweden. 3. Department of Medicine, Clinical Epidemiology Unit, Karolinska University Hospital, Stockholm, Sweden.
Abstract
INTRODUCTION: Infants born large for gestational age (LGA) have increased risks of adverse perinatal outcomes. Maternal obesity, defined as body mass index (BMI) ≥30 kg/m2 , is one of the most prevalent risk factors for LGA and the proportion of pregnancies complicated by obesity is increasing. Early identification of women with BMI ≥30 kg/m2 at increased risk of giving birth to an LGA infant may open possibilities for prevention, aiming at decreasing the incidence of LGA. MATERIAL AND METHODS: A population-based cohort study using information from the first-trimester screening database, which was cross-linked with the Swedish Medical Birth Register. The database included 139 277 full-term singletons without fetal anomalies born between 2006 and 2015 to mothers without prepregnancy diabetes. Of these, 9.1% (n = 12 704) were infants of mothers with BMI ≥30 kg/m2 . For all women with BMI ≥30 kg/m2 , a prediction model for LGA to be used in early pregnancy was constructed based on information on biochemical markers and maternal characteristics. A similar model, as well as a prepregnancy prediction model, were constructed for parous women with BMI ≥30 kg/m2 . In parous women, data from the previous pregnancy were also used. Receiver operating characteristic curve and area under curve (AUC) were calculated. RESULTS: The predictive models for LGA in parous women with BMI ≥30 kg/m2 prepregnancy and in early pregnancy had AUCs of 0.80 (95% CI 0.78-0.82) and 0.81 (95% CI 0.79-0.82), respectively. For all women with BMI ≥30 kg/m2 , the prediction of LGA in early pregnancy had an AUC of 0.66 (95% CI 0.64-0.67). CONCLUSIONS: Performance of the prepregnancy and early pregnancy prediction models for LGA in parous women with BMI ≥30 kg/m2 was good. The predictive capacity was largely driven by previous child's birthweight. First-trimester measurements of fetal size did not improve the predictive capacity in parous women. Predictions of LGA in all women with BMI ≥30 kg/m2 in early pregnancy, without taking previous child's birthweight into account, remain difficult.
INTRODUCTION:Infants born large for gestational age (LGA) have increased risks of adverse perinatal outcomes. Maternal obesity, defined as body mass index (BMI) ≥30 kg/m2 , is one of the most prevalent risk factors for LGA and the proportion of pregnancies complicated by obesity is increasing. Early identification of women with BMI ≥30 kg/m2 at increased risk of giving birth to an LGA infant may open possibilities for prevention, aiming at decreasing the incidence of LGA. MATERIAL AND METHODS: A population-based cohort study using information from the first-trimester screening database, which was cross-linked with the Swedish Medical Birth Register. The database included 139 277 full-term singletons without fetal anomalies born between 2006 and 2015 to mothers without prepregnancy diabetes. Of these, 9.1% (n = 12 704) were infants of mothers with BMI ≥30 kg/m2 . For all women with BMI ≥30 kg/m2 , a prediction model for LGA to be used in early pregnancy was constructed based on information on biochemical markers and maternal characteristics. A similar model, as well as a prepregnancy prediction model, were constructed for parous women with BMI ≥30 kg/m2 . In parous women, data from the previous pregnancy were also used. Receiver operating characteristic curve and area under curve (AUC) were calculated. RESULTS: The predictive models for LGA in parous women with BMI ≥30 kg/m2 prepregnancy and in early pregnancy had AUCs of 0.80 (95% CI 0.78-0.82) and 0.81 (95% CI 0.79-0.82), respectively. For all women with BMI ≥30 kg/m2 , the prediction of LGA in early pregnancy had an AUC of 0.66 (95% CI 0.64-0.67). CONCLUSIONS: Performance of the prepregnancy and early pregnancy prediction models for LGA in parous women with BMI ≥30 kg/m2 was good. The predictive capacity was largely driven by previous child's birthweight. First-trimester measurements of fetal size did not improve the predictive capacity in parous women. Predictions of LGA in all women with BMI ≥30 kg/m2 in early pregnancy, without taking previous child's birthweight into account, remain difficult.