RESUMO
AIM: To develop and validate a graphical method for ultrasonic foetal weight estimation with predictive qualities comparable to state-of-the-art computational methods. METHOD: Study data were obtained from 3839 consecutive singleton pregnancies with ultrasound examination within 7 days of delivery. We translated the well established Hadlock 2-parameter formula based on abdomen circumference and femur length into a nomogram [corrected]. We compared the measured foetal weight with estimations obtained from this new nomogram and three other methods. RESULTS: Measured by the foetal weight percentage error the new nomogram underestimates the foetal weight on average by 2.5 % (9.86) (mean [SD]) with uniform results over the complete birth weight spectrum. By the same measurement the Hadlock 4-parameter formula underestimates the foetal weight by 1.2 % (9.3) whereas the Hansmann nomogram overestimates the foetal weight by 14.5 % (16.1) The difference between the estimated foetal weight percentage error obtained from the new nomogram and from the Hadlock 2-parameter formula is 0.06 % (0.27) and hence clinically irrelevant. CONCLUSION: The new nomogram has predictive qualities comparable to state-of-the-art computational methods and is thus not only as reliable but also easy to use in situations when computers are not available. It can be recommended for foetal weight estimation over the whole spectrum of birth weight.
