Sonographic fetal weight estimation –
Upcoming SlideShare
Loading in...5
×
 

Sonographic fetal weight estimation –

on

  • 214 views

 

Statistics

Views

Total Views
214
Views on SlideShare
214
Embed Views
0

Actions

Likes
0
Downloads
1
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Sonographic fetal weight estimation – Sonographic fetal weight estimation – Document Transcript

  • ORIGINAL ARTICLE Sonographic fetal weight estimation – is there more to it than just fetal measurements? Oshri Barel1,2 , Ron Maymon1,2*, Zvi Vaknin1,2 , Josef Tovbin1,2 and Noam Smorgick1,2 1 Department of Obstetrics and Gynecology, Assaf Harofeh Medical Center, Zerifin, Israel 2 Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel *Correspondence to: Ron Maymon. E-mail: maymonrb@bezeqint.net ABSTRACT Objectives The primary aim of this study was to evaluate the effects of different maternal, fetal, and examiner related factors on the accuracy of sonographic fetal weight estimation (SFWE). Methods A retrospective cohort study analyzing 9064 SFWEs performed within 1 week prior to delivery, including singleton pregnancies with a gestational age of 37 to 42 weeks, was recorded at one medical center from January 2004 to September 2011. Predicted birth weights were calculated according to models by Sabbagha et al., Hadlock et al., and Combs et al. and were compared with the actual birth weight. Effects of different factors on SFWE accuracy were assessed. The systematic error, random error, and mean absolute percentage error were used as measures of accuracy. Results High maternal weight, height, body mass index, multiparity, older maternal age, diabetes, and fetal male sex were associated with underestimation of SFWE (P < 0.05). Fetal presentation and the sonographer’s experience influenced SFWE differently using various models. The amniotic fluid index did have a significant effect on SFWE. Overall, more than 90% of the systematic errors were unaccounted for in the factors we assessed. Conclusions Many maternal and fetal factors significantly influence the SFWE; nevertheless, most errors are probably due to inherent problems in SFWE formulas. © 2013 John Wiley & Sons, Ltd. Funding sources: None Conflicts of interest: None declared INTRODUCTION Ultrasound estimation of the fetal weight is one of the most common ways to assess the growth of a fetus in utero to evaluate an ongoing pregnancy or to prepare for delivery. Information regarding intrauterine growth restriction or excess growth (macrosomia) may influence the pregnancy follow-up and the timing and mode of delivery. Additionally, knowledge of the fetal weight is an important factor affecting fetal mortality.1 However, although numerous methods were developed to compute the sonographic fetal weight estimation (SFWE) from fetal measurements, a high random error of more than 7% characterizes most of them, undermining the accuracy of the SFWE and possibly affecting clinical decisions regarding pregnancy follow-up and delivery.2 In addition to the inherent random errors of these methods, various clinical and technical factors may affect the accuracy of the SFWE. These factors may or may not include maternal factors such as body mass index (BMI);3–5 pregnancy factors such as fetal sex, multiple pregnancy, and amniotic fluid volume;3,6,7 and technical factors related to the experience and fatigue of the ultrasonographer.3,8 Models for prediction of fetal weight using maternal characteristics with or without combination with sonographic fetal measurements have also been developed.9,10 Indeed, previous studies have found conflicting results regarding those various clinical and technical factors and were performed on a relatively small sample of patients. Thus, the aim of the current study was to determine the effect of clinical, sonographic, and technical factors on the accuracy of SFWE in a large retrospective cohort. MATERIALS AND METHODS This retrospective cohort study assessed sonographic and obstetric data of deliveries in Assaf Harofe Medical Center between January 2004 and September 2011. The study cohort comprised of parturient women who referred to our gynecologic and obstetrical ultrasound unit for SFWE within 1 week prior to delivery. Most women were referred for routine ultrasound exam, because it is customary in our department to perform such evaluation to each parturient reporting for any reason during weekday mornings, if such estimation was not performed in the previous 2 weeks. Inclusion criteria were a live-birth singleton pregnancy, birth weight (BW) over 1500 g, Prenatal Diagnosis 2014, 34, 50–55 © 2013 John Wiley & Sons, Ltd. DOI: 10.1002/pd.4250
  • and gestational age between 37 and 42 completed weeks. Exclusion criteria were detection of a fetal abnormality or a major malformation, active labor at the time of SFWE, or ruptured membranes. The computerized database used in our department was searched to obtain the sonographic fetal measurements taken within 1 week before delivery. Sonographic fetal measurements, including biparietal diameter, head circumference, abdominal circumference, and femur diaphysis length, were performed according to formal standards.11–13 Amniotic fluid index (AFI) was measured and recorded in the standard four-quadrant assessment technique.14 Oligohydramnios was defined as AFI ≤ 5 cm and polyhydramnios as AFI > 24 cm. Subsequently, the expected BW was recalculated by using the models by Sabbagha et al.15 (designed for appropriate for gestational age fetuses) and Combs et al.,16 which proved to be the most accurate in our population in a previous study,17 and also by using the popular model by Hadlock et al.18 (utilizing abdominal circumference, femur diaphysis length, head circumference, and biparietal diameter). Those calculated expected BW were compared with the actual BW, also obtained from the departmental computerized database. The SFWEs were performed in our obstetrics ultrasound unit by ultrasound technicians and by physicians trained in obstetrics and gynecology. Some physicians received additional training in obstetrical ultrasound and were defined in this study as ultrasound specialists. Additional demographics, clinical data, and sonographic data were extracted from the patient’s computerized medical records; these were taken at the time of admission to delivery (within 1 week following the ultrasound examination) and included maternal age, maternal height and weight, obstetrical history, gestational age at delivery, mode of delivery, fetal presentation, and fetal sex. The gestational age was determined according to the last menstrual period and the first trimester ultrasound where available, in patients with no first trimester ultrasound, the gestational age was corroborated with the second trimester ultrasound. The gestational age was corrected when there was a disparity of >6 days between the last menstrual period and the dating according to the first trimester ultrasound and >10 days between the last menstrual period and the dating according to the second trimester ultrasound. We did not record ethnicity or race because the rate of intermarriage between individuals of widely different geographic and ethnic origins is currently high in Israel. The study was approved by the local Institutional Review Board. Statistical analysis Data were collected on a standard spreadsheet (Microsoft Excel 2010). Statistical analysis was performed using SPSS software (Version 15, Chicago, IL, USA) by the Tel Aviv University statistical laboratory; P-values of <0.05 were considered statistically significant. Fetal ultrasound measurements were used in the calculations of the formulas for the models analyzed. Descriptive parameters are expressed as mean ± standard deviation. Frequencies are presented as percentages. The analysis was performed in several ways: percentage error was calculated by subtracting the actual BW from the calculated BW and then dividing the difference by the actual BW and multiplying by 100. The mean percentage error (MPE), expressing the systematic error, was calculated from the percentage error. Absolute percentage error and mean absolute percentage error (MAPE) were calculated the same way by using the absolute value of the difference between the estimated BW and the actual BW. Random error, which is the standard deviation of MPE, was also calculated. Percentage errors were compared using the Student’s t-test, the Pearson’s correlation test, and the analysis of variance test in reference to maternal age, parity, weight, height, BMI, diabetes status, gestational age, time from the ultrasound examination to delivery, fetal gender, fetal presentation, and the amount of amniotic fluid. Levene’s test for equality of variance was used to compare random errors. Multivariate stepwise linear regression was also performed in order to evaluate the influence of different variables on SFWE results. RESULTS Included in this study were 9064 SFWE estimations performed during the week prior to delivery (mean time to delivery, 1.6 ± 1.8 days). Total delivery rate in our institute within that period of time consisted of 74 970 births; 67 149 of them between 37 and 42 weeks gestational age. The mean maternal age of our subjects was 30.2 ± 5.0 years (range, 17–48 years), and the median parity was 2 (range, 1–13). Gestational diabetes mellitus (GDM) type A1 was recorded in 536 (5.9%) women, whereas 265 (2.9%) had insulin-dependent gestational or pregestational diabetes mellitus. The mean newborn weight at delivery was 3322 ± 467 g (range, 1680–5420 g), and the mean gestational age at delivery was 39.3 ± 1.2 weeks (range, 37–42 weeks). A cephalic presentation was recorded in 8689 (95.8%) fetuses, a breech presentation in 348 (3.8%) fetuses, and other presentations in 27 (0.3%) fetuses. Other maternal characteristics are described in Table 1. Maternal and gestational characteristics were evaluated in correlation with BW. Increasing maternal weight, height, BMI, parity status, and advanced gestational age were all associated with higher BW (P < 0.001). The SFWE were evaluated with the model by Sabbagha et al.15 (which uses gestational age in addition to sonographic fetal measurements) and with the models by Combs et al.16 and Hadlock et al.16,18 (which use only fetal measurements) and were compared with the actual BW. The systematic errors Table 1 The maternal characteristics of 9064 cases of sonographic fetal weight estimation included in the study Maternal characteristics Result Maternal age (years) 30.2 ± 5 (17–48) Parity 2.1 ± 1.3 (1–13) Maternal weight (kg) 78 ± 14.1 (42–150) Maternal height (m) 1.63 ± 0.06 (1.38–1.83) Maternal body mass index (kg/m 2 ) 29.1 ± 4.8 (16.4–49.7) Data are expressed as mean ± standard deviation (range). Factors affecting SFWE 51 Prenatal Diagnosis 2014, 34, 50–55 © 2013 John Wiley & Sons, Ltd.
  • for the total population were 0.7%, 3.8%, and 5.8% using the models by Sabbagha et al.,15 Combs et al.,16 and Hadlock et al.17 respectively. Groups were analyzed according to different variables in order to investigate the effect of each factor on the accuracy of fetal weight estimation. The factors we assessed were maternal (weight, height, age, parity, diabetes status, and BMI), fetal (gender, presentation, AFI, and actual BW), and the training and experience of the performer of weight estimation; the results are listed in Table 2. Multivariate stepwise linear regression demonstrated an effect of maternal height, BMI, age, maternal diabetes, gestational age, parity status, and fetal gender as significant in affecting the MPE of SFWE using the methods by Sabbagha et al.15 and Combs et al.16 (P < 0.001). All of these factors except maternal age and BMI were found to significantly affect the weight estimation using the method by Hadlock et al.18 The summary of the influence of each factor on the results of fetal weight estimation is presented in Table 3. Nevertheless, although many of the factors we assessed were found to be significant, the coefficient of determination (R2 ) was 0.042 for Hadlock et al.,18 0.044 for Combs et al.,16 and 0.097 for Sabbagha et al.;15 meaning that only 4.2% to 9.7% of the difference in systematic errors could be attributed to those variables, and more than 90% was caused by other factors. Regarding maternal characteristics, maternal weight and height were found to influence the SFWE, with increasing maternal weight and height causing an underestimation of the SFWE using the models by Sabbagha et al.15 and by Combs et al.15,16 (P < 0.001 for both weight and height).The BMI also influenced the SFWE in the same way (P < 0.001), although the actual difference was less than 1% (Table 2). Maternal diabetes was associated with an underestimation of fetal weight, with an estimation of 2.6% to 2.7% less than the actual BW in GDM-A1 and insulin-dependent diabetes, accordingly, using the model by Sabbagha et al.15 (P < 0.05). The model by Combs et al.16 was found to be more accurate for women with diabetes than for the rest of the population [3.6% overestimation for women without diabetes vs 2.2% overestimation for GDM-A1 and 2.7% for insulin-dependent diabetes (P < 0.05)]. Maternal age was also found to be an independent variable affecting SFWE using the models by Sabbagha et al.15 and by Combs et al.,16 with older age associated with an underestimation of fetal weight (P< 0.001). The model by Hadlock et al.18 was related with an overestimation of fetal weight in all of the groups evaluated. Analysis of the effects of these variables on the results of SFWE by Hadlock et al.18 demonstrated a small although significant (P 0.001) improvement in accuracy with higher maternal weight, height, gestational age, and parity status; nevertheless, the results were still less accurate in our population than the results with the models by Sabbagha et al.15 and by Combs et al.16 Regarding fetal characteristics, fetal presentation was not found to significantly affect the systematic error of fetal weight estimation using both models by Sabbagha et al.15 and by Combs et al.16 MAPE, which is another measure of expressing the overall accuracy, was slightly higher for breech and other non-cephalic presentations (8% and 8.4%, respectively) compared with cephalic presentations (6.9%), (P< 0.05). Breech presentation was associated with a slight improvement in accuracy using the model by Hadlock et al.18 (P< 0.05). Conversely, fetal gender was found to be a significant factor in SFWE using all three models. The model by Sabbagha et al.15 produced an underestimation of 1.6% in male fetuses weight estimation while being very accurate for female fetuses, with only a 0.4% overestimation (P< 0.05). The model by Combs et al.,16 on the other hand, tended to overestimate all SFWE by 2.8% in male fetuses and by 4.4% in female fetuses (P< 0.05). The model by Hadlock et al.18 was associated with an overestimation of 2.8% in male fetuses and 4.4% in female fetuses (P< 0.05). The AFI also influenced the SFWE. Oligohydramnios (AFI< 5cm) was associated with an overestimation, and polyhydramnios (AFI > 24 cm) was associated with an underestimation of SFWE in respect to the normal amount of amniotic fluid using the models by Sabbagha et al.15 and by Combs et al.16 The model by Hadlock et al.18 was associated with an overestimation of 5.9% in cases of oligohydramnios, an overestimation of 3.4% in cases with normal AFI and a lower overestimation of 2.9% in cases of polyhydramnios (P < 0.05). We then sought to evaluate the effect of sonographers’ experience on the accuracy of SFWE. Sonographers with experience of at least 2 years in our ultrasound unit were found to have MAPE closer to the actual BW using the models by Combs et al.16 (P < 0.05) and Hadlock et al.18 However, there was a significant difference in systematic errors in the SFWE between those with more than 2 years and those with less than 2 years of experience only with the model by Hadlock et al.18 The SFWE of ultrasound technicians were also compared with those of physicians with or without specialized ultrasound training. The SFWE results obtained by physicians were systematically lower than those obtained by technicians. Using the model by Combs et al.,16 physicians with ultrasound specialty had very accurate results with lower systematic errors than other physicians, and ultrasound technicians had the highest systematic errors (P < 0.05). The model by Hadlock et al.18 also demonstrated better accuracy of fetal weight estimation by US specialists than by physicians and better accuracy of SFWE performed by physicians than by US technicians. On the other hand, using the model by Sabbagha et al.,15 technicians were found to be most accurate, and ultrasound specialists were the least accurate (P < 0.001). Although this difference was found statistically significant, the difference in accuracy between physicians and technicians was no more than 2.6% to 1.5% using the methods by Combs et al.16 and Sabbagha et al.,15 respectively. DISCUSSION In this retrospective cohort study, we tested more than 9000 fetuses and investigated the effects of different maternal, fetal, and examiner variables on the accuracy of SFWE. We found that many factors did affect the SFWE significantly, although this effect was small, and its clinical significance is questionable. In particular, increasing maternal height and weight, advanced gestational age, maternal diabetes, and parity were associated with an underestimation of fetal weight using the models by Combs et al.16 and Sabbagha et al.15 and a small improvement in weight estimation using the model by Hadlock et al.18 These O. Barel et al.52 Prenatal Diagnosis 2014, 34, 50–55 © 2013 John Wiley & Sons, Ltd.
  • Table2Factorsaffectingsonographicfetalweightestimation Factor Systematicerror (MPE)by Sabbagha(%) Systematic error(MPE) byCombs(%) Systematic error(MPE) byHadlock(%) Randomerrorby Combs Randomerrorby Sabbagha Random errorby Hadlock MAPEby Sabbagha(%) MAPEby Combs(%) MAPEby Hadlock(%) Fetal presentation Cephalic(N=8689)À0.63.56.6*7.98.28.16.36.9*8.3 Breech(N=348)À1.54.25.1*8.29.18.26.78.0*7.7 Other(N=27)À1.05.58.3*7.79.07.46.18.4*9.1 Amnioticfluid index Oligohydramnios (N=381) 2.2*5.9*8.6*8.18.38.46.6*8.2*9.9* Normal(N=8344)À0.7*3.4*6.4*7.78.08.16.1*6.8*8.2* Polyhydramnios (N=339) À2.3*2.9*6.8*7.88.08.16.5*6.7*8.5* FetalgenderMale(N=4661)À1.6*2.8*5.8*7.67.97.96.26.5*7.8* Female(N=44403)0.4*4.4*7.2*7.98.28.36.27.2*8.8* Maternal diabetes Nodiabetes (N=8263) À0.5*3.6*6.6*7.98.28.36.16.88.1 GDM-A1(N=536)À2.6*2.2*5.7*7.98.77.96.66.78.2 Insulin-treateddiabetes (N=265) À2.7*2.7*5.9*8.59.78.26.57.18.4 ExaminerTechnician(N=8362)À0.9*3.8*6.8*12.5*13.1*8.56.36.98.5 Physician(N=301)À1.8*2.4*4.8*8.3*8.7*8.26.97.28.2 USspecialist(N=401)À2.4*1.3*4.4*8.39.07.86.97.07.8 Examiner’s experience Lessthan2years (N=1131) À0.33.96.9*8.28.68.56.47.3*8.7* 2yearsormore (N=7933) À0.73.56.5*7.78.08.06.16.8*8.3* Meanpercentageerror(MPE)representsthesystematicerrorofeachmodel.MPEiscalculatedasthemeandifferencebetweenfetalweightestimationandactualbirthweight(BW)dividedbytheactualBWandexpressedinpercent.Meanabsolute percentageerror(MAPE)iscalculatedasthemeanabsolutevalueofthedifferencebetweenthefetalweightestimationandactualBWdividedbytheactualBWandexpressedinpercent.ThestandarddeviationofMPErepresentstherandomerror. RandomerrorswerecomparedusingLevene’stestforequalityofvarianceandwerenotfoundtobesignificantlydifferent(P>0.05) *RepresentsP<0.05. Factors affecting SFWE 53 Prenatal Diagnosis 2014, 34, 50–55 © 2013 John Wiley & Sons, Ltd.
  • factors were also associated with greater BW. This finding may be explained by regression toward the mean, which is an inherent mathematical property of SFWE models on the basis of nonlinear regression analysis. We can therefore assume that, because all of these factors are associated with larger newborn weights, the results of SFWE are lower than expected because of regression toward the mean. Previous studies have found different effects of fetal sex on SFWE. Some studies have not described a clear association,2,19 whereas others have not reported an association and even generated fetal sex specific models for SFWE.20 In the current study, we have evaluated over 9000 SFWE and have found a correlation between male gender and an underestimation of fetal weight using the model by Sabbagha et al.15 On the other hand, this model is very suitable for female fetuses with a systematic error of only 0.4% (P < 0.001). The same tendency was also found using the model by Combs and Hadlock but with higher systematic errors (Table 2). There is conflicting evidence regarding the influence of sonographer’s experience on SFWE. Predanic et al.21 investi- gated the learning curve in estimating fetal weight; there were significant improvements in accuracy amongst residents in training up to 24months, at which time, the best performance was achieved. Conversely, Ben-Aroya et al.8 claimed that neither experience nor fatigue influenced the accuracy of fetal weight estimation performed by residents. We found a significant effect of sonographer’s experience on the systematic errors only using the model by Hadlock et al.,18 although a slight impact was also found in overall accuracy (expressed as MAPE) in favor of more experienced sonographers with over 24months experience (P< 0.05) when using the model by Combs et al.16 and Hadlock et al.16,18 When we compared physicians and technicians, the systematic error was lower for technicians using the model by Sabbagha et al.15 with a systematic error of less than 1% (P< 0.05). There are a few possible explanations for this phenomenon. First, those pregnancies evaluated by physicians may have included more complex cases such as intrauterine growth restriction and macrosomia. This diversity of cases could increase the margin of error. Second, physicians possibly use clinical judgment when assessing fetal weight, which may slightly influence their fetal measurements. The physicians’ SFWE was closer to the actual BW when using the models by Combs et al.16and Hadlock et al.16,18 but not when using the model by Sabbagha et al.,15 which is less dependent on the fetal measurements because gestational age is also a part of the equation. Random errors were also higher for SFWE performed by technicians in comparison with those performed by physicians and ultrasound specialists Fetal presentation also seemed to affect the accuracy of SFWE in previous studies by Dammer et al.22 (who investigated 244 fetuses) and by Melamed et al.23 (who investigated 165 cases). We evaluated this hypothesis in 348 cases and found a significant difference using the model by Hadlock et al.18 We could not demonstrate a significant impact of breech presentation on the systematic error using the model by Combs et al.16 or by Sabbagha et al.15,16 Amniotic fluid index also had an effect on SFWE with a tendency for overestimation of fetal weight in cases of oligohydramnion and underestimation in cases of polyhydramnios (P< 0.001). This finding is in contrast with previous studies,3,6,24 which found no influence of AFI on SFWE. A possible explanation for this finding might be that polyhydramnios was associated with higher BW, whereas oligohydramnios was associated with lower BW in our population (P < 0.05), and the SFWE tended to regress toward the mean thereby causing this effect. Our study presents several limitations. This is a retrospective cohort study, and the data are derived from a facility- based rather than a population-based registry. This may undermine the possibility to generalize our conclusions. One major weakness in this study is that although many of the variables we studied significantly affect the fetal weight estimation error, the actual combined contribution to the MPE was less than 10%. This indicates that the Table 3 Effect of different factors on systematic error using the model by Sabbagha et al.15 Factor Effect on systematic error by Sabbagha Coefficient of determination (R 2 ) (total effect on systematic error accuracy expressed in %) Maternal age Underestimation* 0.1 Maternal weight Underestimation* 2 Maternal body mass index Underestimation* 1 Maternal height Underestimation* 3 Parity Underestimation* 0.7 Maternal diabetes Underestimation* 0.2 Gestational age Underestimation* 2.3 Fetal male gender Underestimation* 0.7 Fetal presentation No significant effect Oligohydramnios Overestimation* 4 Polyhydramnios Underestimation* 4 Sonographers’ experience No significant effect Ultrasound technicians versus physicians No significant effect *Represents P < 0.001. O. Barel et al.54 Prenatal Diagnosis 2014, 34, 50–55 © 2013 John Wiley & Sons, Ltd.
  • factors we assessed are only on the tip of the iceberg and that most causes for SFWE errors are still unaccounted for. Random errors are still the major causes for the inherent errors in SFWE, and the factors we evaluated did not significantly influence these errors. In conclusion, many maternal, fetal, and examiner related factors significantly influence the SFWE. Knowledge of the influence of these factors on the SFWE may help the clinician to understand whether the fetal weight estimation performed tends for overestimation or underestimation of the actual BW, possibly, allowing for improved management of pregnancy and delivery. Nevertheless, even after adjusting for these factors, fetal weight estimation will only improve by up to 10% of systematic errors. Further research has to be performed in order to find more accurate fetal weight estimation formulas or other factors that might be accountable for systematic and random errors. WHAT’S ALREADY KNOWN ABOUT THIS TOPIC? • Most of the studies so far found conflicting evidence regarding the effect of maternal, fetal, and examiner related factors on the accuracy of sonographic fetal weight estimation. WHAT DOES THIS STUDY ADD? • This study evaluated over 9000 cases and found a significant effect of several factors on the accuracy of sonographic fetal weight estimation. Nevertheless, even after adjusting for these factors, fetal weight estimation will only improve by up to 10%. REFERENCES 1. Barker DJP. Long-term outcome of retarded fetal growth. In Clinical Obstetrics and Gynecology, Divon MY (ed.). Philadelphia, PA: Lippincott–Raven, 1997;853–63. 2. Dudley NJ. A systematic review of the ultrasound estimation of fetal weight. Ultrasound Obstet Gynecol 2005;25:80–9. 3. Heer IM, Kumper C, Vogtle N, et al. Analysis of factors influencing the ultrasonic fetal weight estimation. Fetal Diagn Ther 2008;23:204–10. 4. Blann DW, Prien SD. Estimation of fetal weight before and after amniotomy in the laboring gravid woman. Am J Obstet Gynecol 2000;182:1117–20. 5. Farrell T, Holmes R, Stone P. The effect of body mass index on three methods of fetal weight estimation. BJOG 2002;109:651–7. 6. Benacerraf BR, Gelman R, Frigoletto FD Jr. Sonographically estimated fetal weights: accuracy and limitation. Am J Obstet Gynecol 1988;159:1118–21. 7. Townsend RR, Filly RA, Callen PW, Laros RK. Factors affecting prenatal sonographic estimation of weight in extremely low birthweight infants. J Ultrasound Med 1988;7:183–7. 8. Ben-Aroya Z, Segal D, Hadar A, et al. Effect of OB/GYN residents’ fatigue and training level on the accuracy of fetal weight estimation. Fetal Diagn Ther 2002;17:177–81. 9. Nahum GG, Stanislaw H, Huffaker BJ. Accurate prediction of term birth weight from prospectively measurable maternal characteristics. J Reprod Med 1999;44:705–12. 10. Mazouni C, Rouzier R, Ledu R, et al. Development and internal validation of a nomogram to predict macrosomia. Ultrasound Obstet Gynecol 2007;29:544–9. 11. Chitty LS, Altman DG, Henderson A, Campbell S. Charts of fetal size, 2: head measurements. Br J Obstet Gynaecol 1994;101:35–43. 12. Chitty LS, Altman DG, Henderson A, Campbell S. Charts of fetal size, 3: abdominal measurements. Br J Obstet Gynaecol 1994;101:125–31. 13. Chitty LS, Altman DG, Henderson A, Campbell S. Charts of fetal size, 4: femur length. Br J Obstet Gynaecol 1994;101:132–5. 14. Rutherford SE, Phelan JP, Smith CV, Jacobs N. The four-quadrant assessment of amniotic fluid volume: an adjunct to antepartum fetal heart rate testing. Obstet Gynecol 1987;70:353–6. 15. Sabbagha RE, Minogue J, Tamura RK, Hungerford SA. Estimation of birth weight by use of ultrasonographic formulas targeted to large-, appropriate-, and small-for-gestational-age fetuses. Am J Obstet Gynecol 1989;160:854–60. 16. Combs CA, Jaekle RK, Rosenn B, et al. Sonographic estimation of fetal weight based on a model of fetal volume. Obstet Gynecol 1993;82:365–70. 17. Barel O, Vaknin Z, Tovbin J, et al. Assessment of accuracy for multiple sonographic fetal weight estimation formulas: a 10-year experience from a single center. J Ultrasound Med 2013;32:815–23. 18. Hadlock FP, Harrist RB, Sharman RS, et al. Estimation of fetal weight with the use of head, body, and femur measurements – a prospective study. Am J Obstet Gynecol 1985;151:333–7. 19. Ott WJ, Doyle S, Flamm S. Accurate ultrasonic estimation of fetal weight. Am J Perinatol 1985;2:178–82. 20. Melamed N, Yogev Y, Ben-Haroush A, et al. Does use of a sex-specific model improve the accuracy of sonographic weight estimation? Ultrasound Obstet Gynecol 2012;39:549–57. 21. Predanic M, Cho A, Ingrid F, Pellettieri J. Ultrasonographic estimation of fetal weight: acquiring accuracy in residency. J Ultrasound Med 2002;21:495–500. 22. Dammer U, Goecke TW, Voigt F, et al. Sonographic weight estimation in fetuses with breech presentation. Arch Gynecol Obstet 2013;287:851–8. 23. Melamed N, Ben-Haroush A, Meizner I, et al. Accuracy of sonographic fetal weight estimation: a matter of presentation. Ultrasound Obstet Gynecol 2011;38:418–24. 24. Meyer WJ, Font GE, Gauthier DW, et al. Effect of amniotic fluid volume on ultrasonic fetal weight estimation. J Ultrasound Med 1995;14:193–7. Factors affecting SFWE 55 Prenatal Diagnosis 2014, 34, 50–55 © 2013 John Wiley & Sons, Ltd.