Aust NZ J Obstet Gynaecol 1992; 32: 3: 196
Clinical Versus Ultrasound Estimation of Fetal Weight S. Raman’, R. Urauhart* and M. Yusof’ Department of Obstetrics and Gynaecology, University Hospital, Kuala Lumpur, Wst Malaysia EDITORIAL COMMENT: In clinical medicine there is always the tendency to grant authenticity to an opinion expressed in a typewritten report concerning a pathology or radiology investigation and to override the clinician’s opinion which in truth is often more accurate. Obstetricians are well aware of the fallibility of their clinical judgement and this applies to the outcome of trial of labour, and assessment of fetal weight; thus clinical opinion tends to be unjustifiably underrated. In this series from West Malaysia the clinicians were more than equal to the technology of ultrasonography. However, this is not the end of the story since the importance of estimation of fetal size is when the baby is small and of questionable viability (isit over 800 g and worth a Caesarean section for afetal indication, or less than 600 g and not?), or large enough to warrant Caesarean section because of the risk of severe morbidity from shoulder dystocia or uterine rupture if the patient has had a previous Caesarean or has a breech presentation. Although there are occasions when fetal weight in excess of 4,000 g would indicate Caesarean section (breechpresentation), usually a trial of labour is indicated when the infant is large and pments cephalically because the vast majority are delivered uneventfully. It is more important for the clinician to judge when to do a Caesarean section for arrest of the big head in the midpelvis rather than to perform elective abdominal delivery. In the editor’s experience ultrasonography is often very useful and accurate in assessment of fetal weight at gestations below 30 weeks;fetal weight is most likely to be underestimated by the clinician when the fetus is premature and growth retarded - the head is often deeply engaged and associated oligohydramnios causes the clinician to undercall fetal size. How many Caesarean sections would have been performed in retrospect, in patients with premature rupture of the membranes f breech presentation, had it been known that the birth-weight exceeded 800 g? In the case of cephalopelvic disproportion there is too much emphasis on fetal weight rather than the size and shape of the baby’s skull - it is the head and/or shoulders that obstruct labour, not total body weight!
Summary: A prospective study was carried out on 50 patients who had their fetal weight estimated by 3 clinicians of different seniority and compared to the ultrasound estimated fetal weights using 3 different formulas. All the patients delivered within 24 hours of their clinical and ultrasound estimates. A wide range of birth-weights (1,800-4,500g) was estimated among the 3 different races (Malay, Chinese and Indians). The results showed that there was no significant difference in birth-weight estimation amongst the 3 clinicians as well as between the 3 ultrasound formulas used. There was however significant difference between these 2 groups when compared with the actual birthweight with clinical estimation being superior to ultrasound estimation in our population. This level of significance did not extend beyond 4,000 g fetal weight (actual) thus making both clinical and ultrasound estimation of fetal weight equally accurate after 4,000 g. This has important implications for developing countries where there is a lack of technologically advanced ultrasound machines capable of doing sophisticated functions like fetal weight estimations but has experienced clinicians who could perform this function equally well if not better.
1. Associate Professor. 2. Lecturer. 3. Medical Officer.
Address for correspondence: Dr. S. Raman, Department of Obstetrics and Gynaecology, University Hospital, 59100 Kuala Lumpur, West Malaysia.
s. RAMANET AL
Estimation of fetal weight forms an integral part of the obstetric management of the high risk pregnancy. It is particularly important in breech presentation and cases of previous Caesarean section where vaginal delivery is contemplated. Other situations would include suspected macrosomia, intrauterine growth retardation and preterm deliveries. Obstetricians have always prided themselves on their ability to accurately predict fetal weight, but there are wide variations between individuals. This was highlighted by Ong and Sen (1) who showed that the more senior doctor did not necessarily estimate with the greatest accuracy. More recently ultrasound has come to play a major role in obstetrical care including estimation of fetal weight. Several charts have been devised for this purpose using various mathematical computations of multiple fetal measurements. It has been reported that ultrasound estimation is an accurate method of estimating birth-weight (2-4). The aim of this study was to prospectively compare clinical estimation of fetal weight by 3 doctors of varying grades of experience with ultrasound examination using 3 different formulas.
MATERIALS AND METHODS A prospective study was carried out on 50 women who were either to undergo elective Caesarean section or who were in early labour with the fetal head unengaged and considered suitable for accurate ultrasound assessment. The 3 authors independently estimated the fetal weight based on careful clinical palpation. The patients subsequently underwent ultrasound examination with a SAL 77A Toshiba real time ultrasound scanner using a 3.5 MegaHz probe. All the ultrasound measurements were made by one of us (SR). Two measurements each of the biparietal diameter (BPD), (total of 4 measurements), femur length (FL), fetal trunk area (FTA) and the abdominal circumference (AC) were taken and the mean of the 2 measurements calculated. The BPD was measured from the outer to the inner edge of the fetal skull except for the Osaka University Chart where outer to outer measurements were used. The AC and FTA were measured at the level of the bifurcation of the portal sinus and the stomach. The FTA was measured using the ellipse method with a performed calculation obtained from the machine. The AC was calculated by measuring the anteroposterior and transverse abdominal diameters, and multiplying the sum of these diameters by 1.57. The FL was measured with the femur horizontal and measuring the distance from the greater trochanter to the lateral condyle. The fetal weight was calculated for the various charts using the formulas described below. This was not done for Campbell’s chart as it is based on abdominal circumference alone, and was found not to be accurate in our population by an earlier study. Shepard et a1 (5) Log 10 birthweight=1.7492 + 0.166 (BPD) + 0.046 (AC) - 2.646 (AC x BPD)/1000
197
Osaka University Chart (6) Estimated fetal birth-weight =1.25647 x BPD’ + 3.50665 x FTA x FL + 6.30994 Hadlock et a1 (7) Log 10 birth-weight=1.3598 + 0.051 (AC) + 0.1844 (FL) - 0.0037 (AC x FL) The 3 different racial groups (Malay, Chinese and Indian) that make up our population were studied. All patients delivered within 24 hours of the clinical and ultrasound estimates being made. The actual birthweight at delivery was measured and noted (Range birth-weight 1,800-4,500 g). The data were analyzed using a linear regression bivariate analysis approach where, of the 2 variables, the dependent variable is the clinical assessment of birth-weight prior to birth while the independent variable is the actual birth-weight. A test of homogenicity of regression was applied among the groups to see whether the group could be represented by a single regression line. The regression lines were compared a priori among the 3 estimating clinicians by measuring the coefficient of the regression and the sum of squares due to the use of separate vectors. This indicates whether there is significant departure from parallelism of the regression lines, indicating significant differences between the estimating clinicians. The same technique was used for the ultrasound regression coefficients where the Osaka, Shepard and Hadlock charts were compared (8,9).
RESULTS There were no significant differences between the 3 doctors estimating fetal weight using clinical skills nor were there any significant differences between the 3 ultrasound charts used to assess the weight of the fetus in utero. The sample was therefore collapsed to give 150 variables in the 2 separate groups (clinical estimation versus ultrasound estimations) compared. The mean birth-weight estimated by the clinicians, by ultrasound and the actual birth-weight at delivery are shown in table 1. The standard deviations for both the clinical and ultrasound estimates are narrow. This suggests accuracy of estimation in the sample. There was however a significant difference between the mean ultrasound estimate of fetal weight, the clinical estimate and actual birth-weight (p
Clinical Versus Ultrasound Estimation of Fetal Weight S. Raman’, R. Urauhart* and M. Yusof’ Department of Obstetrics and Gynaecology, University Hospital, Kuala Lumpur, Wst Malaysia EDITORIAL COMMENT: In clinical medicine there is always the tendency to grant authenticity to an opinion expressed in a typewritten report concerning a pathology or radiology investigation and to override the clinician’s opinion which in truth is often more accurate. Obstetricians are well aware of the fallibility of their clinical judgement and this applies to the outcome of trial of labour, and assessment of fetal weight; thus clinical opinion tends to be unjustifiably underrated. In this series from West Malaysia the clinicians were more than equal to the technology of ultrasonography. However, this is not the end of the story since the importance of estimation of fetal size is when the baby is small and of questionable viability (isit over 800 g and worth a Caesarean section for afetal indication, or less than 600 g and not?), or large enough to warrant Caesarean section because of the risk of severe morbidity from shoulder dystocia or uterine rupture if the patient has had a previous Caesarean or has a breech presentation. Although there are occasions when fetal weight in excess of 4,000 g would indicate Caesarean section (breechpresentation), usually a trial of labour is indicated when the infant is large and pments cephalically because the vast majority are delivered uneventfully. It is more important for the clinician to judge when to do a Caesarean section for arrest of the big head in the midpelvis rather than to perform elective abdominal delivery. In the editor’s experience ultrasonography is often very useful and accurate in assessment of fetal weight at gestations below 30 weeks;fetal weight is most likely to be underestimated by the clinician when the fetus is premature and growth retarded - the head is often deeply engaged and associated oligohydramnios causes the clinician to undercall fetal size. How many Caesarean sections would have been performed in retrospect, in patients with premature rupture of the membranes f breech presentation, had it been known that the birth-weight exceeded 800 g? In the case of cephalopelvic disproportion there is too much emphasis on fetal weight rather than the size and shape of the baby’s skull - it is the head and/or shoulders that obstruct labour, not total body weight!
Summary: A prospective study was carried out on 50 patients who had their fetal weight estimated by 3 clinicians of different seniority and compared to the ultrasound estimated fetal weights using 3 different formulas. All the patients delivered within 24 hours of their clinical and ultrasound estimates. A wide range of birth-weights (1,800-4,500g) was estimated among the 3 different races (Malay, Chinese and Indians). The results showed that there was no significant difference in birth-weight estimation amongst the 3 clinicians as well as between the 3 ultrasound formulas used. There was however significant difference between these 2 groups when compared with the actual birthweight with clinical estimation being superior to ultrasound estimation in our population. This level of significance did not extend beyond 4,000 g fetal weight (actual) thus making both clinical and ultrasound estimation of fetal weight equally accurate after 4,000 g. This has important implications for developing countries where there is a lack of technologically advanced ultrasound machines capable of doing sophisticated functions like fetal weight estimations but has experienced clinicians who could perform this function equally well if not better.
1. Associate Professor. 2. Lecturer. 3. Medical Officer.
Address for correspondence: Dr. S. Raman, Department of Obstetrics and Gynaecology, University Hospital, 59100 Kuala Lumpur, West Malaysia.
s. RAMANET AL
Estimation of fetal weight forms an integral part of the obstetric management of the high risk pregnancy. It is particularly important in breech presentation and cases of previous Caesarean section where vaginal delivery is contemplated. Other situations would include suspected macrosomia, intrauterine growth retardation and preterm deliveries. Obstetricians have always prided themselves on their ability to accurately predict fetal weight, but there are wide variations between individuals. This was highlighted by Ong and Sen (1) who showed that the more senior doctor did not necessarily estimate with the greatest accuracy. More recently ultrasound has come to play a major role in obstetrical care including estimation of fetal weight. Several charts have been devised for this purpose using various mathematical computations of multiple fetal measurements. It has been reported that ultrasound estimation is an accurate method of estimating birth-weight (2-4). The aim of this study was to prospectively compare clinical estimation of fetal weight by 3 doctors of varying grades of experience with ultrasound examination using 3 different formulas.
MATERIALS AND METHODS A prospective study was carried out on 50 women who were either to undergo elective Caesarean section or who were in early labour with the fetal head unengaged and considered suitable for accurate ultrasound assessment. The 3 authors independently estimated the fetal weight based on careful clinical palpation. The patients subsequently underwent ultrasound examination with a SAL 77A Toshiba real time ultrasound scanner using a 3.5 MegaHz probe. All the ultrasound measurements were made by one of us (SR). Two measurements each of the biparietal diameter (BPD), (total of 4 measurements), femur length (FL), fetal trunk area (FTA) and the abdominal circumference (AC) were taken and the mean of the 2 measurements calculated. The BPD was measured from the outer to the inner edge of the fetal skull except for the Osaka University Chart where outer to outer measurements were used. The AC and FTA were measured at the level of the bifurcation of the portal sinus and the stomach. The FTA was measured using the ellipse method with a performed calculation obtained from the machine. The AC was calculated by measuring the anteroposterior and transverse abdominal diameters, and multiplying the sum of these diameters by 1.57. The FL was measured with the femur horizontal and measuring the distance from the greater trochanter to the lateral condyle. The fetal weight was calculated for the various charts using the formulas described below. This was not done for Campbell’s chart as it is based on abdominal circumference alone, and was found not to be accurate in our population by an earlier study. Shepard et a1 (5) Log 10 birthweight=1.7492 + 0.166 (BPD) + 0.046 (AC) - 2.646 (AC x BPD)/1000
197
Osaka University Chart (6) Estimated fetal birth-weight =1.25647 x BPD’ + 3.50665 x FTA x FL + 6.30994 Hadlock et a1 (7) Log 10 birth-weight=1.3598 + 0.051 (AC) + 0.1844 (FL) - 0.0037 (AC x FL) The 3 different racial groups (Malay, Chinese and Indian) that make up our population were studied. All patients delivered within 24 hours of the clinical and ultrasound estimates being made. The actual birthweight at delivery was measured and noted (Range birth-weight 1,800-4,500 g). The data were analyzed using a linear regression bivariate analysis approach where, of the 2 variables, the dependent variable is the clinical assessment of birth-weight prior to birth while the independent variable is the actual birth-weight. A test of homogenicity of regression was applied among the groups to see whether the group could be represented by a single regression line. The regression lines were compared a priori among the 3 estimating clinicians by measuring the coefficient of the regression and the sum of squares due to the use of separate vectors. This indicates whether there is significant departure from parallelism of the regression lines, indicating significant differences between the estimating clinicians. The same technique was used for the ultrasound regression coefficients where the Osaka, Shepard and Hadlock charts were compared (8,9).
RESULTS There were no significant differences between the 3 doctors estimating fetal weight using clinical skills nor were there any significant differences between the 3 ultrasound charts used to assess the weight of the fetus in utero. The sample was therefore collapsed to give 150 variables in the 2 separate groups (clinical estimation versus ultrasound estimations) compared. The mean birth-weight estimated by the clinicians, by ultrasound and the actual birth-weight at delivery are shown in table 1. The standard deviations for both the clinical and ultrasound estimates are narrow. This suggests accuracy of estimation in the sample. There was however a significant difference between the mean ultrasound estimate of fetal weight, the clinical estimate and actual birth-weight (p
