Preterm birth has been associated with extremes of body mass index (BMI). The etiology of preterm birth in obese and underweight patients is complex and actively deliberated. Because of the strong association between shortened cervix in the middle trimester and preterm delivery, we sought to investigate the association of BMI and cervical length in the context of preterm delivery.METHODS:
A retrospective cohort study was conducted of 393 women who underwent cervical screening using a transvaginal ultrasound for cervical length measurement (CL) in 2012. We collected cervical length, BMI, fetal weight at delivery, gestational age at delivery and other maternal and fetal characteristics. SPSS was used for data analysis as appropriate; P less than .05 was considered significant.RESULTS:
Women were separated into 3 groups based on BMI: group 1 (BMI less than 20), group 2 (BMI 20 to 29), and group 3 (BMI 30 or greater). The majority of the patients (98.3%) had cervical length screening without a specific indication. Group 3 had the highest risk of preterm birth (11.8% , 16.4% , 30.9% ; P=.002), were older, and were more likely to be delivered for a maternal indication (2% , 3.2% , 4.1% ; non-significant). However, CL (cm) among the 3 groups was similar (3.59 , 3.55 , 3.44  non-significant). The Pearson correlation coefficient between BMI and CL did not show statistical correlation.CONCLUSION:
Despite the increased risk of preterm delivery in obese women, the mechanism of preterm birth does not appear to be preceded by a short cervix in mid-gestation. Other mechanisms warrant further investigation.