Ultrasonic fetal weight estimation in multiple and singleton pregnancy

Thirty-five multiple and 168 singleton fetal weight estimations were made in an Australian population using the tables of Warsof et al., which have been computed from a North American population. The estimated fetal weights were compared to the actual birth weights within 96 h of estimation. Correlation coefficients of the actual birth weight with the estimated fetal weight for the total group, singleton and multiple pregnancies respectively, were 0.967, 0.969 and 0.933. The percentages of estimated fetal weights falling within 10% of the actual birth weight were for the three groups, 79.3%, 77.9% and 85.7% respectively. Individual weight sub-groups, less than 1500 g, 1500-2500 g, 2500-4000 g and greater than 4000 g, showed correlation coefficients of 0.84, 0.838, 0.839 and 0.759 respectively, which compare favourably to published studies of North American populations using this method. It is concluded that ultrasonic fetal weight estimation using the North American computerized tables of Warsof et al. is appropriate for an Australian population and is valid for both singleton and multiple pregnancies.