A comparison of intergrowth-21st method to hadlock formula in Estimating fetal weight in low-risk pregnancies

Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Dr Sikolia Wanyonyi

Second Supervisor/Advisor

Prof Marleen Temmerman

Third Supervisor/Advisor

Prof Sudhir Vinayak


Obstetrics and Gynaecology (East Africa)


Introduction; Several ultrasound methods have been used to estimate fetal weight but the Hadlock formula has been found to be the most reliable and is currently preferred by many clinicians. Hadlock method has been used and extensively researched for decades worldwide. However, in 2017 the Intergrowth-21st group developed a new formula to estimate fetal weight using abdominal circumference and head circumference. This was meant to develop standards that could address differences across different populations. We therefore undertook to compare the accuracy of Hadlock and Intergrowth-21st methods of estimating fetal weight in a low-risk population, to determine if the observed difference could significantly alter our interpretation of fetal growth.

Objectives; the primary objective was to compare the Intergrowth-21st and Hadlock formulae in estimating fetal weight in low-risk pregnancies and determine the factors that affect their prediction of fetal weight.

Methods; A secondary analysis of data from the Routine Third Trimester Ultrasound (ROTTUS) trial was undertaken. Two hundred and eighty-four women who were scanned between 36weeks and 37weeks and 0 days were included. The estimated fetal weight using Hadlock and Intergrowth 21-st formulas were converted to percentiles. The accuracy of the two formulas was compared using: 1) Receiver operating characteristics curve analysis with respective area under the curve, sensitivity and specificity of each formula (using paired t tests or Wilcoxon signed rank test) 2) The mean of the percentage error was used as a marker for the systematic error expressed as a percentage of birth weight 3) The proportion of fetuses with an estimated fetal weight within ±15% of birth weight was used as a marker of acceptable margin of error (using Chi-square test or Fisher's exact test). A logistic regression model was constructed for the two ultrasound formulae based on estimated fetal weight within ±15% of birth weight as the dependent variable, against parameters on bivariate analysis and a multivariate analysis was conducted on these factors to control for confounders.

Results; There were 284 total study participants. Majority were in the age group of 30-34 years. Only one third of the participants had a normal BMI with majority being overweight or obese. The proportion of fetuses within ±15% of the birth weight was 78.2% and 62.7% for Hadlock and Intergrowth-21st methods respectively (p<0.0001). The area under the curve for Hadlock and Intergrowth-21st methods was 0.8209 (CI 0.74347-0.89838) and0.872 (CI 0.80865-0.93544) respectively, when predicting large for gestational age fetuses (p<0.0393).The mean absolute percentage error for Hadlock and Intergrowth-21st methods was -7.2 and -10.1 (p<0.0001) respectively. The accuracy of Intergrowth -21st method when the ultrasound-to-delivery interval was <8days and>21days, was 88.9% and 48.9% respectively (p<0.0001).

Conclusion; Hadlock formula was more accurate than Intergrowth-21st method in estimating fetal weight in low-risk pregnancies. Intergrowth performed better in predicting fetal weight in large for gestation fetuses. However, both Hadlock and Intergrowth-21st methods demonstrated acceptable overall accuracy. There was no justification to replace Hadlock with Intergrowth-21st method. Either formula can be used based on the clinician’s discretion.

This document is available in the relevant AKU library