Obstetrics & Gynecology Email Alerts
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Obstetrics & Gynecology 2000;96:157-161
© 2000 by The American College of Obstetricians and Gynecologists
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by SONG, T.-B.
Right arrow Articles by KIM, E.-K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by SONG, T.-B.
Right arrow Articles by KIM, E.-K.

ORIGINAL RESEARCH

Fetal Weight Prediction by Thigh Volume Measurement With Three-Dimensional Ultrasonography

TAE-BOK SONG, MD, THOMAS R. MOORE, MD, JI-YOUNG LEE, MD, YOON-HA KIM, MD and EUN-KYUNG KIM, MD

From the Department of Obstetrics and Gynecology, Chonnam National University Medical School and Research Institute of Medical Sciences, Chonnam National University, the Department of Diagnostic Radiology, Chosun University Medical School, Kwangju, Korea; and the Division of Perinatal Medicine, Department of Reproductive Medicine, University of California, San Diego, School of Medicine, La Jolla, California.

Address reprint requests to: Tae-Bok Song, MD Chonnam National University Medical School Department of Obstetrics and Gynecology 8 Hakdong, Dongku Kwangju, 501-190 Korea E-mail: tbsong{at}chonnam.chonnam.ac.kr


   Abstract
Top
Abstract
Materials and Methods
 Results
 Discussion
 References
 
Objective: To evaluate the usefulness and accuracy of a simple method of predicting fetal weight by measuring fetal thigh volume with three-dimensional ultrasonography.

Methods: In 84 pregnant women, fetuses without structural or chromosomal anomalies were studied prospectively and cross-sectionally. Biparietal diameter (BPD), abdominal circumference (AC), and femur length (FL) were measured by two-dimensional ultrasound. Fetal thigh volume was measured by three-dimensional ultrasound, using three cross-sectional images of femur, from proximal, middle, and distal parts of femur diaphysis. Infants were delivered within 48 hours after ultrasound examinations.

Results: Modified thigh volume measurements using three cross-sectional images of femur by three-dimensional ultrasound were correlated strongly with birth weight (R2 = 0.921, P < .001). Using linear and polynomial regression, we calculated a new best-fit formula: Birth weight (g) = 165.32 + 28.78 x modified thigh volume (mL). The mean and standard deviation of the residual were 121.8 and 110.4, respectively, in three-dimensional formulas, which were significantly smaller than those of two-dimensional formulas.

Conclusion: Thigh volume measurement using three cross-sectional images of femur by three-dimensional ultrasound was simple, and there was better accuracy with this method than with two-dimensional ultrasound methods for predicting fetal weight during the third trimester of pregnancy.

Initial attempts to estimate fetal weight by ultrasound consisted of individual fetal measurements such as biparietal diameter (BPD)1 or abdominal circumference (AC).2 Subsequent reports showed that accuracy of estimated fetal weight is improved when multiple fetal measurements are used. The simplest methods that give reasonably accurate results are based on two measurements, AC and BPD3,4 or AC and femur length (FL).5–7

With the advent of three-dimensional ultrasound, some researchers found it useful for fetal weight estimation by using limb circumferences,8 upper arm volumes,9 and thigh volumes.10 Fetal limb volume is related to fetal growth and nutrition.11 The accuracy of three-dimensional ultrasound in volumetry has been validated in many organ systems, in vitro and in vivo.12–15 Hence, thigh volume assessed by three-dimensional ultrasound should effectively predict birth weight. In other studies, fetal upper arm or thigh volume assessed by three-dimensional ultrasound achieved satisfactory results in birth weight prediction.9,10 The only drawback was the long time needed to measure volumes. In this study, we assessed fetal thigh volume by a simple method. We determined the usefulness and accuracy of that method of assessing fetal thigh volume by three-dimensional ultrasonography for predicting birth weight.


   Materials and Methods
Top
Abstract
 Materials and Methods
Results
 Discussion
 References
 
Between August and November 1998, two- and three-dimensional ultrasonographic examinations were done consecutively on 84 pregnant women with singleton fetuses by one obstetrician. This study was prospective and cross-sectional, so each case was included for only one examination. There were 18 cases with maternal disease such as hypertension or diabetes (Table 1Go). There was a case with polyhydramnios and four cases with oligohydramnios, defined as amniotic fluid index (AFI) under 5 cm. Small for gestational age (SGA) infants were delivered in ten cases, and large for gestational age (LGA) infants in 17 cases. All were delivered within 48 hours after ultrasound examinations at the Chonnam University Hospital, a tertiary medical center in southwestern Korea, and all subjects were Korean. In our routine, fetuses scheduled for induction or elective cesarean delivery have prenatal ultrasound examinations to estimate their weights.


View this table:
[in this window]
[in a new window]
  		Table 1. Demographic Data for the 84 Subjects
 
Complete sets of fetal biometric variables (BPD, AC, and FL) were assessed with a two-dimensional ultrasonographic scanner with a 3.5-MHz transabdominal probe according to methods previously described.3,5 Biparietal diameter was measured on an axial plane at the level of the thalamic nuclei with a clear midline echo and clearly discernible lateral ventricle. Abdominal circumference was evaluated when the umbilical vein and fetal stomach could be identified on the same plane. Femur length was measured at the ossified shaft on a long-axis plane. Those indices were then calculated to predict an estimated birth weight with the standard formulas (Table 2Go).3,5


View this table:
[in this window]
[in a new window]
  		Table 2. Formulas Used for Estimating Fetal Weight by Two-Dimensional Ultrasound
 
Thigh volume was assessed with a three-dimensional ultrasound scanner (Combison 530; Kretztechnik, Zipf, Austria) with a 5.0-MHz transabdominal sector transducer. We chose the best saggital plane of the femur to take a volume sample. Data acquisition lasted 4 seconds. After that period, we could manipulate the three different planes, x, y, and z. The scanned volume of the whole thigh was stored in the built-in computer or on compact disks for further analysis.

The formats of thigh volume measurement by three-dimensional ultrasonography are shown in Figures 1Go and 2Go. We used the traditional plane for measuring the femur length on the first screen (Figure 1AGo and Figure 2Go, upper left panel) and rotated the plane to put the femur accurately in a horizontal position. Then we fixed the plane as an anchor and moved the cursor (XM) along the femur. The transaxial plane (PM) was shown on the second screen (Figure 1BGo and Figure 2Go, upper right panel), and the dot cursor was marked along the outline of the thigh. We measured areas at distal, middle, and proximal diaphysis. The cursor location of XD and XP (Figure 1AGo) was about 4 mm inside the end of femur diaphyses because the outline of thigh was shown more clearly there than that of the end of diaphysis. The shaded area in Figure 1BGo, which corresponds to the upper right panel of Figure 2Go, was measured. The thigh volume was calculated automatically from three cross-sectional images. Each measurement of simple thigh volume took about 2 minutes.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 1. A) Plane of femur length. Volume of region between arrows (from to XD XP) is volume assessed for thigh. B) Corresponding area transaxial plane (PM) to cursor line XM is shown. Shaded area = of corresponding plane, PM. From XD to XP shaded area of PM was measured at three cross-sections (PD, PM, PP), and volume data were integrated automatically by a built-in computer in the three-dimensional ultrasonography unit.

 


View larger version (76K):
[in this window]
[in a new window]
  		Figure 2. Images of three-dimensional ultrasonographic thigh volume acquisition show traditional plane of femur length on upper left panel. After the first screen was fixed, we moved the cursor along the femur at three points (distal, middle, and proximal diaphysis), and the corresponding transaxial plane was shown on upper right panel. The dot cursor was marked along the outline of the thigh on upper right panel and integrated volume was shown on the screen. The third plane, which is perpendicular to the above two screens, was shown simultaneously in lower left panel. Cursor movement in scanning box was displayed in lower right panel.

 
With the linear and polynomial regression methods, we calculated a new best-fit formula on thigh volume versus birth weight. The accuracy of the formulas was examined by the mean of residual (absolute value of predicted birth weight minus actual birth weight) and the standard deviation (SD) of residual.

Intraobserver error was determined by measuring thigh volume on the same subject 30 times by the same physician (JYL). To validate the new formula, we prospectively evaluated 102 cases that met the criteria. The mean and SD of the residual of the new formula and the traditional formula were also compared.

We used SPSS for Windows Base 10.0 statistical package (SPSS Inc., Chicago, IL) to analyze the data. Regression analysis with coefficient of determination (R2) was used to determine the relationship between independent and dependent variables. The paired t test was used to compare the mean and SD between different formulas. Intraobserver error was examined by one-sample t test. P < .05 was considered statistically significant.


   Results
Top
Abstract
 Materials and Methods
 Results
Discussion
 References
 
Two-dimensional ultrasonographic growth measures and three-dimensional ultrasonographic acquisitions for thigh volume could be made without difficulty in all 84 subjects. Population characteristics are provided in Table 1Go. No fetuses had congenital anomalies. The mean value (±SD) of actual birth weight was 3070.1 ± 586.7 g (range 1460–4060 g), the mean gestational age was 37.7 ± 2.2 weeks (range 34–41 weeks), and the mean thigh volume was 100.9 ± 19.5 mL (range 64.5–133.5 mL).

Intraobserver error, expressed as intraobserver coefficient of variation, was 3.12% (n = 30, mean 110.30 ± 3.44 mL) and considered not significant. Three-dimensional ultrasound volumetry was highly reproducible for the assessment of fetal thigh volume. All measurements were done by one author, so there was no interobserver error.

The relationship between birth weight and thigh volume was highly statistically significant (R2 = 921, n = 84, P < .001). Although polynomial equations using thigh volume as the independent variable and birth weight as the dependent variable were calculated from the first to the third order, the simple linear regression equation had the best fit for thigh volume versus birth weight. From that regression analysis, we derived a new best-fit formula: Birth weight (g) = 165.32 + 28.78 x thigh volume (mL).

As given in Table 3Go, the mean and SD of residual of the new formula and the previous two formulas for predicting birth weights in Table 2Go were compared by paired t test. The new formula using thigh volume measured by three-dimensional ultrasonography had the lowest mean and SD of residual. The new formula was statistically superior to the Shepard and Hadlock formulas (P < .001 each).


View this table:
[in this window]
[in a new window]
  		Table 3. Comparison of Accuracy of Two- and Three-Dimensional Formulas for Estimating Birth Weight
 
Another 102 fetuses were measured by two- and three-dimensional ultrasonography to prospectively evaluate the accuracy of our new formula for predicting fetal weight. The mean value ± SD of actual birth weights was 3170.9 ± 590.3 g (range 1530–4300 g), the mean gestational age was 38.2 ± 2.0 weeks (range 32–41 weeks), and mean thigh volume was 105.5 ± 22.8 mL (range 42.0–149.8 mL). Fetal weights estimated with the new equation showed no difference compared with actual birth weight (paired t test, t = -0.355, n = 102, P = .724). As shown in Table 4Go, the comparison between our three-dimensional equation and the other traditional two-dimensional formulas showed that the new formula was more accurate than the other formulas and had the lowest mean and SD of residual. The new formula had better accuracy than the Shepard and Hadlock formulas (P = .011 and .002, respectively) for predicting fetal weight during the third trimester of pregnancy.


View this table:
[in this window]
[in a new window]
  		Table 4. Prospective Comparison of Accuracy of Two- and Three-Dimensional Formulas for Estimating Birth Weight
 

   Discussion
Top
Abstract
 Materials and Methods
 Results
 Discussion
References
 
Fetal weight estimates, like fetal age estimates, are important in perinatal treatment decisions. Initially, fetal BPD was used to estimate fetal mass, but because the head comprises only 20% of total fetal volume, it is not surprising that this method was so inaccurate it was of no clinical value. Multiple fetal variables such as BPD or head circumference, AC, and FL also have been used to predict fetal weight by two-dimensional ultrasonography, with improved accuracy.1–5 It is relatively convenient to predict fetal weight with two-dimensional formulas, but many conditions can decrease their accuracy, such as inaccurate section plane of AC because of uterine contraction, fetal breathing, or abdominal compression from other parts of the body, as with oligohydramnios and deeply engaged fetal head during labor. Under those conditions, errors in estimating fetal weight will be greater with two-dimensional ultrasound equations using BPD, AC, and FL.

Many in vitro and in vivo studies validated the accuracy of three-dimensional ultrasound in volumetry. Estimation of fetal weight using three-dimensional ultrasound has been studied,8–10 and Jeanty et al11 reported that fetal limb volume can be a new measurement to assess fetal growth and nutrition.

Liang et al9 reported that upper arm volume assessed with three-dimensional ultrasound significantly improved prediction of fetal weight compared with the traditional two-dimensional formulas. Chang et al10 reported that fetal thigh volumetry assessed by three-dimensional ultrasound predicted birth weight more accurately than two-dimensional ultrasound formulas. However, those three-dimensional assessments of volume were relatively meticulous and time-consuming. According to Chang et al,10 the measurements of thigh volume were cross-sectional images at 3-mm intervals, and a complete assessment of thigh volume usually took 10–15 minutes.

To shorten the data collection time, we used a simple method. Fetal thigh volume was measured by using three cross-sectional images at proximal, middle, and distal diaphysis of the femur instead of 3-mm intervals as used by Chang et al.10 Our method was easier and less time-consuming than that of Chang et al.10 As shown in Table 3Go, the three-dimensional method using only three cross-sectional images of the femur had better accuracy than two-dimensional methods.

The 84 cases in our study were classified into appropriate for gestational age, LGA, and SGA groups to determine differences among them (Table 5Go). Fetal weight prediction by the three-dimensional ultrasonographic method had the lowest values for mean and SD of residual in all three groups. The paired t test showed statistically significant P values in all except the SGA group. Further studies with more subjects are necessary because the number of cases in the SGA group was small (n = 10).


View this table:
[in this window]
[in a new window]
  		Table 5. Comparison of Accuracy of Two- and Three-Dimensional Formulas According to Size for Gestational Age
 
The weakness of our study was the problem of compared population. Fetal weight prediction formulas for two-dimensional ultrasound were constructed by using different populations. Our formula using three-dimensional ultrasound was derived from the same population; therefore, the comparison is unfair, placing the two-dimensional method at a disadvantage because two-dimensional formulas were derived from different populations.


   Footnotes
 
PII S0029-7844(00)00884-X

Received October 21, 1999. Received in revised form February 24, 2000. Accepted March 16, 2000.


   References
Top
Abstract
 Materials and Methods
 Results
 Discussion
 References
 
1. Ianniruberto A, Gibbons JM. Predicting fetal weight by ultrasound B-scan cephalometry: An improved technique with disappointing results. Obstet Gynecol 1971;37:689–94.[Abstract/Free Full Text]

2. Campbell S, Wilkin D. Ultrasonic measurement of the fetal abdominal circumference in estimation of fetal weight. Br J Obstet Gynaecol 1975;82:689–97.[Medline]

3. Warsof SL, Gohari P, Berkowitz RL, Hobbins JC. The estimation of fetal weight by computer-assisted analysis. Am J Obstet Gynecol 1982;142:47–54.[Medline]

4. Shepard MJ, Richards VA, Berkowitz RL, Warsof SL. An evaluation of two equations for predicting fetal weight by ultrasound. Am J Obstet Gynecol 1982;142:47–54.

5. Hadlock FP, Harrist RB, Carpenter RJ, Deter RL, Oark SK. Sonographic estimation of fetal weight. The value of femur length in addition to head and abdomen measurements. Radiology 1984; 150:535–40.[Abstract/Free Full Text]

6. Warsof SL, Wolf P, Coulehan J, Queenam TJ. Comparison of fetal weight estimation formulas with and without head measurements. Obstet Gynecol 1986;67:569–73.[Abstract/Free Full Text]

7. Campbell WA, Vintzileos AM, Neckles S, Weinbaum PJ, Nochimson DJ. Use of the femur length to estimate fetal weight in premature infants: Preliminary results. J Ultrasound Med 1985;4: 583–90.[Abstract]

8. Favre R, Bader AM, Nisand G. Prospective study on fetal weight estimation using limb circumferences obtained by three-dimensional ultrasound. Ultrasound Obstet Gynecol 1995;6:140–4.[Medline]

9. Liang RI, Chang FM, Yao BL, Chang CH, Yu CH, Ko HC. Predicting birth weight by fetal upper arm volume using three-dimensional ultrasound. Am J Obstet Gynecol 1997;177:632–8.[Medline]

10. Chang FM, Liang RI, Ko HC, Yao BL, Chang CH, Yu CH. Three-dimensional ultrasound-assessed fetal thigh volumetry in predicting birth weight. Obstet Gynecol 1997;90:331–9.[Abstract]

11. Jeanty P, Romero R, Hobbins JC. Fetal limb volume: A new parameter to assess fetal growth and nutrition. J Ultrasound Med 1985;4:273–82.[Abstract]

12. Kyei-Mensah A, Zaidi J, Pittrof R, Shaker A, Campbell S, Tan SL. Transvaginal three-dimensional ultrasound: Accuracy of follicular volume measurements. Fertil Steril 1996;65:371–6.[Medline]

13. Riccabona M, Nelson TR, Pretorius DH, Davidson TE. In vivo three-dimensional sonographic measurement of organ volume validation in urinary bladder. J Ultrasound Med 1996;15:627–32.[Abstract]

14. Nosir TFM, Fioretti PM, Vletter WB, Boersma E, Salustri A, Postma JT, et al. Accurate measurement of left ventricular ejection fraction by three-dimensional echocardiography: A comparison with radionucleotide angiography. Circulation 1996;94:460–6.[Abstract/Free Full Text]

15. King DL, King DL Jr, Shao MYC. Evaluation of in vitro measurement accuracy of a three-dimensional ultrasound scanner. J Ultrasound Med 1991;10:77–82.[Abstract]




This article has been cited by other articles:


Home page
 J Ultrasound MedHome page
L. F. Goncalves, W. Lee, J. Espinoza, and R. Romero
Three- and 4-Dimensional Ultrasound in Obstetric Practice: Does It Help?
J. Ultrasound Med., December 1, 2005; 24(12): 1599 - 1624.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by SONG, T.-B.
Right arrow Articles by KIM, E.-K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by SONG, T.-B.
Right arrow Articles by KIM, E.-K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS