Purpose: To determine the accuracy of ultrasound in fetal weight estimation and to evaluate maternal and/or fetal factors that could interfere in the result.
Methods: This was a transverse prospective study, involving 106 patients, with 212 fetal weight evaluations, by two observers, within 24 h to delivery. The following parameters were measured: biparietal diameter, head circumference, abdominal circumference, and femoral length. Fetal weight was estimated using the Hadlock formula and the results were compared to birth weight. The maternal factors examined were: weight, BMI, and skin to uterus distance measured by ultrasound, and the fetal factors were: presentation, position, placental localization and thickness, fetal weight, and amniotic fluid index (AFI).
Results: There was good correlation between estimated fetal weight and birth weight (R=0.97). In 79.2% and in 92.4% of cases the estimated fetal weight was within 10% and 15% of birth weight, respectively. The only maternal factor that presented a positive correlation with percent error in the estimate of fetal weight was the skin to uterus distance (R ≥ 0.56). Fetal weight showed negative correlation with percent error (R>-0.36; p<0.001), with a significant tendency to overestimate fetal weight in the group of very low weight - <1000 g (p<0.05). The AFI showed a low negative correlation with percent error (R=-0.21; p<0.001) with no difference between AFI groups (p=0.516).
Conclusion: Ultrasound presented good accuracy in the estimation of fetal weight. The error of weight estimate was directly proportional to the skin to uterus distance and inversely proportional to fetal weight. AFI did not interfere significantly in the ultrasound prediction of fetal weight.