ACCURACY OF CLINICAL VERSUS SONOGRAPHIC ESTIMATION OF FOETAL WEIGHT AT TERM AT UNIVERSITY OF ILORIN TEACHING HOSPITAL, ILORIN, NIGERIA

Abstract

Background:
Accurate estimation of foetal weight in late pregnancy and labour is extremely useful in the
management of labour and delivery. It helps obstetricians in making decisions about appropriate
mode of delivery, trial of labour, planned vaginal breech delivery, instrumental vaginal delivery
and trial of labour after caesarean delivery all with the aim of reducing the morbidity and
mortality associated with birth trauma and perinatal asphyxia. Clinical formulae and sonographic
estimation are the two common methods of foetal weight estimation. However the superiority of
sonographic estimation over the clinical methods has been a source of debate. Also there are
concerns about the influence of some biological variables such as maternal BMI and actual birth
weight categories on the accuracy of both methods of foetal weight estimation.
Aim:
The aim of this study is to determine and compare the accuracy of clinical estimation of foetal
weight using Dare’s formula with that of sonographic estimation using hadlock 3 formula. Also
the influence of maternal BMI and actual birth weight categories on the accuracies of these
methods was determined.
Study design/Setting:
A prospective comparative study conducted at the University of Ilorin Teaching Hospital.
Methods:
Three hundred consenting and eligible pregnant women planned for either vaginal delivery or
caesarean section were recruited for the study using systematic random sampling. Data collection
sheet was used to obtain information on socio-demographic characteristics, clinical parameters
and sonographic parameters. The patient’s height and weight were measured and the BMI was
calculated. In-utero clinical estimation of foetal weight was carried out using a flexible tape
measure calibrated in centimetre to measure the symphysio-fundal height and abdominal girth at
the level of the umbilicus and the product of the two was calculate as the foetal weight in
grammes. The patients also had sonographic estimation of foetal weight using Hadlock 3 formula
in the foetal assessment unit. After delivery, the newborn babies were weighed within 30 minutes
of delivery.
Outcomes measured include estimated foetal weight error, absolute estimated foetal weight error,
absolute percentage error, and absolute percentage error less than 10% (i.e. an estimated foetal
weight that is within ±10% of the actual birth weight) for both clinical and sonographic methods
of estimation. Comparison was made using the chi-square test and student’s t-test, where
necessary. Subgroup analysis based on Preset BMI categories and ABW categories was done to
assess the influence of BMI and ABW on the accuracy of both clinical and sonographic
estimation of foetal weight. A p-value of < 0.05 was taken as statistically significant.
Results:
The mean age of patients included was 30.54±3.92 (range 18-40) years. The mean parity was
1.61±1.12 (range 0-4). The mean gestational age at delivery was 39.26±1.34 (range 37-41.9)
weeks. The mean maternal BMI was 30.41±5.4 (range 20.8-48.3) kg/m2. Of the 300 participants,
13.7% have normal BMI while 41% were overweight and 45.3% were obese. The mean
birthweight in the study population was 3206.35±455.935 (range 2000-4750) g. Majority of the
participants delivered normal weight babies accounting for 88.7% while 3.3% and 8.0%
delivered low birthweight and macrosomic babies respectively.
Clinical method estimated foetal weight with error margin within 10% of actual birthweight in
53.3% of the participants which is significantly lower than that of sonographic method with error
margin within 10% of actual birthweight in 68% of the participants (p value=0.0003272). The
mean error of clinically estimated foetal weight was 406.14±464.670g while that of ultrasound
estimated foetal weight was 125.13±307.788g.
The result showed that BMI significantly affected the error in estimation of foetal weight by
clinical method (p value=0.000) but the influence of BMI on error in ultrasound estimation of
foetal weight was not statistically significant (p value=0.339). Accuracy of ultrasound estimated
foetal weight is significantly higher than that of clinical method in the obese patients with error
margin within 10% of actual birthweight in 61.5% and 29.16% of the participants respectively (p
value= 0.00000234). In the morbidly obese category, clinical estimation tends to overestimate
foetal weight with error margin of 1194.58±627.634g while ultrasound tends to underestimate
foetal weight with error margin of -27.84±357.300g.
Actual birth weight category influence the percentage errors of foetal weight estimation of both
clinical and ultrasound estimation of foetal weight (p values=0.032 and 0.000) respectively.
Accuracy of both clinical and ultrasound estimation of foetal weight were similar at the extremes
of birth weight. However in the macrosomic foetal weight category, while clinical method
overestimated foetal weight, ultrasound method underestimated foetal weight with mean error of
352.38±581.134g and -376.75±256.415g respectively.
Conclusion:
Ultrasound estimation of foetal weight is more accurate and superior to clinical estimation
generally. While ultrasound estimation of foetal weight is better than clinical method in the
obese category, clinical method is as good as ultrasound in other weight categories.

ACCURACY OF CLINICAL VERSUS SONOGRAPHIC ESTIMATION OF FOETAL WEIGHT AT TERM AT UNIVERSITY OF ILORIN TEACHING HOSPITAL, ILORIN, NIGERIA