Ultrasonography-based Fetal Weight Estimation: Finding an Appropriate Model for an Indian Population

Area of Science:

• Obstetrics and Gynecology
• Medical Imaging
• Perinatology

Background:

• Limited data exists on the accuracy of fetal weight estimation models in the Indian population.
• Commonly used ultrasonography parameters include abdominal circumference (AC), biparietal diameter (BPD), femur length (FL), and head circumference (HC).

Purpose of the Study:

• To systematically evaluate existing fetal weight estimation models for their appropriateness in the Indian population.
• To identify the most accurate models for fetal weight estimation in Indian pregnant women.

Main Methods:

• Retrospective analysis of 149 pregnant women from Bengaluru, India (gestational age ≥ 34 weeks, delivery within 7 days of scan).
• Fetal weight was estimated using 34 different models, excluding those for specific conditions like IUGR or macrosomia.
• Models were ranked based on mean percentage error (MPE) and random error.

Main Results:

• Only 12 out of 34 models had MPE within ±10%; only seven had random error <10%.
• Western population-based models tended to overestimate fetal weight.
• Woo's (AC-BPD) model was ranked highest, followed by Jordaan (AC), Combs (AC-HC-FL), Hadlock (AC-HC), and Hadlock-3 (AC-HC-FL). Models using AC and AC-BPD showed significantly lower MPE.

Conclusions:

• Existing fetal weight estimation models demonstrate higher errors in the Indian population compared to their native populations.
• Direct application of current models without consideration may lead to inaccuracies.
• There is a critical need for developing fetal weight estimation models specifically for the Indian population.