RESUMO
UNLABELLED: Ultrasound fetal weight estimation (EFW) has become a routine practice in obstetrics with a major impact on obstetrical management. OBJECTIVES: To evaluate the accuracy of sonographic fetal weight estimations performed by residents in obstetrics at the delivery day and to search parameters influencing this exam. METHODS: A prospective study performed in 299 patients at 37 weeks of gestation or more, during their delivery day. Ultrasound fetal weight estimations were performed by residents and compared to birth weights (BW). The estimated fetal weight (EFW) was calculated using the formula of Hadlock: Log10 EPF=1,335+0.0316 BIP+0,0457 PA+0,1623 LF-0,0034 PA LF. RESULTS: Mean absolute difference between EFW and BW was 200g [100-450]. Mean absolute percentage error was 5.71 % [2.7-11.37]. The proportion of the EFW within 10 % of BW was 69.6 %. There was a good correlation between EFW and BW (R=0.79). Obesity and fetal macrosomia had negatively influenced ultrasound performance. For the diagnosis of macrosomia, ultrasound has a sensibility of 38 % and a specificity of 99.59 %. CONCLUSION: Ultrasound fetal weight estimations performed by residents in the delivery room were as accurate as those performed by experimented sonographers. Thus, the predictive performance remains limited with a low sensitivity in the diagnosis of macrosomia.