RESUMO
OBJECTIVE: The gestation-adjusted projection method (GAP method) uses third trimester ultrasound fetal weight to predict birthweight. Our study sought to assess if the accuracy of the GAP method in morbidly obese women depended on (1) ultrasound timing or (2) extreme elevations in maternal body mass index (BMI). STUDY DESIGN: We conducted a diagnostic accuracy study from 2007 to 2012 of all singleton pregnancies with BMI >40 kg/m(2) at the time of delivery that had fetal growth assessment between 30+0 and 35+0 weeks (EARLY) and greater than 35+0 weeks (LATE). Next, a 'reference' BMI group (30-35) was compared with subcategories of women with BMI ≥40. Paired t tests, χ(2) tests and analysis of variance were used to determine significance. RESULTS: A total of 235 pregnancies for the first objective and 430 for the second objective were included. The mean absolute percent error was comparable between the 2 periods (EARLY 7.9 ± 6.7%, LATE 7.4 ± 5.6%; P = .33) and across BMI categories (30-35: 7.6 ± 6.3%, 40-50: 7.2 ± 5.7%, >50: 7.8 ± 6.0%; P = .79). The sensitivity and specificity of the GAP method to predict macrosomia during the EARLY time period was 46% (95% confidence interval 28-66) and 97% (94-99) and in the 40-50 BMI subcategory was 72% (47-90) and 96% (90-99), respectively. CONCLUSION: Prediction of birthweight using the GAP method in obese women does not appear to be influenced by timing of ultrasound or extreme BMI. In a population where clinical estimated fetal weight is difficult, the GAP method may aid in delivery planning.