Abstract
Objective
Estimated fetal weight (EFW) is essential for managing pregnancy and delivery. Currently, two primary methods are used for EFW: Clinical and sonographic estimation. Both might be influenced by maternal habitus. Obesity has become a global epidemic; however, its effect on EFW and the accuracy of the two methods has yet to be determined. The aim of the present study was to describe the accuracy of clinical EFW (cEFW) and sonographic EFW (sEFW) as a function of maternal body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) by comparing them to actual birth weight.
Methods
This was a retrospective cohort study at a single tertiary center (2014–2020), including term (37–42 weeks) singleton deliveries with documented BMI, cEFW, and sEFW performed within 10 days of delivery, and known birth weight. At our institution, cEFW is routinely performed at admission by trained obstetricians as part of a standardized Leopold‐based assessment. Participants were stratified into BMI groups: <25, 25–29.9, 30–34.9, and ≥35. The primary outcome was estimation accuracy across BMI strata, expressed as absolute error (in grams [g]) and as deviations greater than 10% and 20%. Secondary outcomes included diagnostic performance for large‐for‐gestational‐age (LGA) neonates, assessed using receiver operating characteristic (ROC) curves and the area under the curve (AUC). Multivariable logistic regression identified factors associated with an error rate exceeding 20%.
Results
Among 19 397 deliveries, 2182 women qualified for analysis. Participants were distributed across BMI groups as follows: <25 (n = 408), 25–29.9 (n = 809), 30–34.9 (n = 634), and ≥35 (n = 331). Mean birth weight rose progressively with BMI, from 3045 ± 519 g in the lowest BMI group to 3557 ± 453 g in the ≥35 group (P < 0.01). Across BMI ≥25, cEFW showed larger absolute error than sEFW: BMI 25–29.9 (222 ± 184 vs. 202 ± 156 g), 30–34.9 (238 ± 187 vs. 204 ± 154 g), and ≥35 (254 ± 207 vs. 211 ± 161 g) (all P < 0.01). Rates of sEFW deviation >20% were low and stable across BMI (0.6%–1.7%; P = 0.42), whereas cEFW >20% increased with BMI (2.0%–6.0%; P < 0.01). In adjusted models, cEFW had higher odds of >20% error that escalated with BMI (adjusted odds ratio [aOR] 3.34, 5.87, 9.17 for BMI 25–29.9, 30–34.9, ≥35; all P < 0.01). In adjusted models, maternal BMI was not associated with >20% error in sEFW. For LGA detection, sEFW outperformed cEFW across strata, with the most significant gap at BMI ≥35 (AUC 0.873 vs. 0.791).
Conclusion
Maternal obesity significantly reduces the accuracy of clinical EFW, whereas sonographic EFW maintains stable performance across BMI strata. Sonographic assessment should therefore be prioritized for fetal weight estimation in women with elevated BMI, particularly in the late third trimester, when weight‐based delivery decisions are made.
Keywords: body mass index, fetal weight estimation, large‐for‐gestational‐age, obesity
1. INTRODUCTION
Obesity has become a global public health crisis, with rising rates among women of reproductive age. 1 Maternal obesity is associated with a range of adverse pregnancy outcomes, including altered fetal growth, gestational diabetes, hypertensive disorders, and stillbirth. 2 , 3 Obese pregnant individuals are at increased risk for delivering macrosomic infants, 4 which elevates the likelihood of labor induction failure, shoulder dystocia, and both elective and emergency cesarean deliveries. 5 These comorbidities often necessitate changes in clinical decision making regarding the timing and mode of delivery, thereby influencing perinatal outcomes. 6
Accurate estimated fetal weight (EFW) is essential for guiding prenatal care and intrapartum management. EFW informs individualized clinical decisions about the mode and timing of delivery, enabling risk stratification for complications such as fetal growth restriction (FGR), large for gestational age (LGA), or macrosomia. 7 Multiple methods are available for estimating fetal weight, including clinical assessment, maternal anthropometric measurements, and sonographic imaging. 8 The accuracy of these methods is critically important, as it directly impacts obstetric planning and outcomes. 9
The Leopold maneuvers, a series of four abdominal palpation techniques traditionally used to assess fetal position and presentation, are commonly employed by experienced providers to estimate fetal weight through clinical examination. 10 Although widely used, the accuracy of clinical EFW (cEFW) varies with provider experience and patient characteristics, such as maternal habitus. 11 Sonographic EFW (sEFW), by contrast, is based on standardized biometric measurements such as biparietal diameter, head circumference, abdominal circumference, and femur length, which are incorporated into formulas like Hadlock's to estimate fetal weight. 12
While sEFW is considered more objective, its accuracy may still be influenced by maternal obesity due to compromised image quality. 13 Conversely, cEFW may integrate maternal habitus into the assessment, but it is more dependent on the provider's skill, potentially leading to greater variability and reduced accuracy. 14 Previous studies comparing clinical and sonographic methods have yielded mixed results, 15 many of which involved small cohorts, and the limited evidence regarding the optimal EFW method among obese patients. Some have reported comparable accuracy, while others indicate that sEFW outperforms clinical methods in pregnancies complicated by FGR or macrosomia. 16 , 17
Therefore, this study aimed to evaluate the accuracy of clinical versus sonographically estimated fetal weight across different body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) categories.
2. MATERIALS AND METHODS
2.1. Study population
This was a retrospective cohort study conducted at a single tertiary medical center, utilizing institutional electronic medical records from January 2014 to February 2020.
The study period ended in 2020 because this was the most recent date for which complete and validated electronic medical record data (including both clinical and sonographic EFW) were available. Ending the study period before the COVID‐19 pandemic minimized changes in obstetric care patterns that could influence EFW practice and documentation.
Inclusion criteria comprised women with singleton pregnancies who delivered at term (37–42 weeks' gestation), with documented maternal BMI, clinical and sonographic EFW recorded within 10 days of delivery and known neonatal birth weight. The 10‐day interval between EFW assessment and delivery was chosen to ensure clinical relevance, as fetal weight changes rapidly near term, while preserving an adequate sample size for analysis.
Exclusion criteria included preterm (<37 weeks) and post‐term (>42 weeks) deliveries, multifetal gestations, and cases with incomplete or missing relevant data. To ensure methodological consistency and reduce interoperator and interequipment variability, we restricted the analysis to cases in which sEFW was performed in‐house, using standardized protocols.
2.2. Clinical and sonographic EFW
Late third‐trimester sEFW is not obtained routinely at our center. Instead, ultrasound is ordered based on clinical indications, including suspected fetal growth restriction or macrosomia, maternal comorbidities (e.g., pregestational or gestational diabetes, hypertensive disorders), obesity, or discordance between cEFW and gestational age. Consequently, our cohort represents a clinically selected population rather than the entire cohort of term pregnancies.
Sonographic EFW was calculated using standard fetal biometric parameters according to the Hadlock IV formula, which incorporates abdominal circumference (AC), femur length (FL), biparietal diameter (BPD), and head circumference (HC). 18 Ultrasound measurements were obtained via transabdominal scanning by an experienced obstetrician or a certified ultrasound technician at our tertiary ultrasound unit. Clinical EFW was performed using the Leopold‐Pavlik abdominal palpation technique. 19 Clinical EFW was performed in routine practice by obstetric residents, fellows, and attending obstetricians. Trainees performed examinations under attending supervision as part of their formal residency. All clinicians performing cEFW receive structured teaching and supervised training in Leopold maneuvers as part of the residency/fellowship curriculum; no additional study‐specific training was provided.
In our institution, cEFW is typically performed as part of the physical examination before the sonographic assessment; thus, clinicians performing cEFW were not aware of the sEFW result at the time of palpation. Assessments were performed at admission to labor. In cases where the initial cEFW appeared clinically discordant, either notably low or notably high, an attending obstetrician was called to repeat or confirm the assessment, reflecting standard practice in our unit.
2.3. Definitions
Fetuses and newborns large‐for‐gestational‐age (LGA) and small‐for‐gestational‐age (SGA) were defined as having an estimated fetal weight or an actual birth weight above the 90th or below the 10th percentile, respectively, according to gestation‐specific growth curves widely used for the Israeli population. 20 For the primary analyses, maternal BMI was categorized according to established guidelines into four groups: <25 (normal), 25–29.9 (overweight), 30–34.9 (obesity class I), and ≥35 (obesity class II–III). 21 Obesity classes II and III were grouped due to a relatively small sample size in class III, allowing for more stable statistical analysis.
2.4. Outcomes
The primary outcome was the accuracy of clinical and sonographic EFW methods in predicting actual neonatal birth weight across maternal BMI categories. The secondary outcome was the diagnostic accuracy of both clinical and sonographic EFW in identifying LGA and SGA fetuses across maternal BMI categories.
2.5. Data collection and parameters
Data were obtained from the hospital's electronic medical record system and included various maternal, obstetric, and neonatal parameters. Maternal characteristics included maternal age at delivery, pre‐pregnancy BMI, BMI at the time of EFW, gestational weight gain (GWG), gravidity, parity, chronic hypertension, and pregestational diabetes. Obstetric features included the presence of hypertensive disorders during pregnancy (gestational hypertension and pre‐eclampsia), amniotic fluid volume abnormalities (polyhydramnios and oligohydramnios), gestational diabetes mellitus (GDM) screening values using the 50 g glucose challenge test (GCT) and 100 g oral glucose challenge test (OGTT), gestational age at delivery, and mode of delivery (vaginal, vacuum‐assisted, or cesarean), cEFW, Ultrasound parameters included sEFW and fetal biometry measurements (BPD, HC, AC, FL), along with placental location. Neonatal parameters included gender and birth weight.
2.6. Ethics approval and consent to participate
All methods were carried out in accordance with relevant guidelines and regulations. This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of “The Rabin Medical Center institutional review board” (no.: 0130‐20‐RMC) at Rabin Medical Center. Informed consent was waived by “The Rabin Medical Center institutional review board” due to the retrospective design. The data were anonymized before use.
2.7. Statistical analysis
Data analysis was performed using IBM SPSS Statistics for Windows (version 27.0, IBM Corporation, Armonk, New York, USA) and R software (version 3.6.0, R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were assessed for normality using the Shapiro–Wilk test and visual inspection of histograms and Q‐Q plots. Normally distributed data are presented as mean ± standard deviation (SD) and compared using one‐way analysis of variance (ANOVA) across BMI categories (<25, 25–30, 30–35, and ≥35).
Categorical variables are shown as counts and percentages and compared using the χ 2 test or Fisher exact test when expected cell counts were less than 5. Spearman's rank correlation coefficients were calculated to evaluate the association between birth weight and sonographic or clinical EFW within each BMI group.
The predictive accuracy of sonographic and clinical EFW in identifying LGA fetuses was assessed separately within each BMI group using receiver operating characteristic (ROC) curve analysis. The area under the curve (AUC) was determined with 95% confidence intervals (CI), and AUCs were compared descriptively across BMI groups and between sonographic and clinical assessments.
The error in EFW was measured by the absolute difference between estimated and actual birth weights in grams (mean absolute error [MAE]). Percentage deviations exceeding 10% and 20% were analyzed based on BMI category. Logistic regression models were created to identify factors independently associated with a deviation of more than 20% in cEFW and sEFW. Variables with P less than 0.05 in univariate analysis and those deemed clinically relevant (such as BMI category, GDM, LGA, SGA) were included in the multivariable models. Adjusted odds ratios (aOR) and 95% confidence intervals (CIs) are reported.
All statistical tests were two‐tailed, and P less than 0.05 was regarded as statistically significant.
3. RESULTS
During the study period, 19 397 singleton deliveries with complete in‐house documentation took place at our institution. Of these, 1495 deliveries were excluded because they were pre‐ or post‐term. An additional 7305 cases were excluded due to missing maternal BMI data, and 8415 cases were excluded because clinical or sonographically EFW assessments were done more than 10 days before delivery.
A total of 2182 singleton live births (11% of 19 397 deliveries) met the inclusion criteria (Figure 1). Participants were stratified into four BMI groups: <25 (n = 408), 25–29.9 (n = 809), 30–34.9 (n = 634), and ≥35 (n = 331). The mean maternal BMI was 29.7 ± 5.2, and the mean gestational age at delivery was 39.0 ± 1.3 weeks. Birth weight increased across BMI categories, ranging from 3045 ± 519 g in the <25 group to 3557 ± 453 g in the ≥35 group (P < 0.01). Additional baseline characteristics, including pre‐pregnancy BMI and mode of delivery, are presented in Table 1.
FIGURE 1.
Flow diagram of study population selection and distribution across maternal body mass index (BMI) categories.
TABLE 1.
Maternal and obstetrical characteristics of the study group stratified according to maternal BMI.
| N | BMI | Overall | P value | |||
|---|---|---|---|---|---|---|
| <25 | 25–29.9 | 30–34.9 | ≥35 | |||
| 408 (19%) | 809 (37%) | 634 (29%) | 331 (15%) | 2182 | ||
| Maternal age (years) | 30.77 ± 4.8 | 31.82 ± 4.81 | 32.07 ± 4.91 | 32.19 ± 5.18 | 31.75 ± 4.92 | <0.01 |
| Pre‐pregnancy BMI | 22.9 ± 1.62 | 27.54 ± 1.4 | 32.14 ± 1.38 | 38.54 ± 3.18 | 29.67 ± 5.22 | |
| Gestational weight gain (kg) | 9.76 ± 6.82 | 12.60 ± 8.0 | 14.06 ± 9.88 | 14.68 ± 16.38 | 12.81 ± 10.20 | <0.01 |
| Neonatal birth weight (g) | 3045 ± 519 | 3379 ± 524 | 3506 ± 488 | 3557 ± 453 | 3380 ± 532 | <0.01 |
| Mode of delivery | <0.01 | |||||
| Normal vaginal delivery (%) | 74.3 | 74.5 | 69.4 | 60.1 | 70.9 | |
| Vacuum extraction (%) | 11.8 | 8.9 | 8 | 9.4 | 9.3 | |
| Cesarean section (%) | 13.7 | 16.3 | 22.6 | 29.9 | 19.7 | |
| Forceps (%) | 0.2 | 0.4 | 0 | 0.3 | 0.2 | |
| Macrosomia | 19 (4.7%) | 100 (12.4%) | 97 (15.3%) | 52 (15.7%) | 268 (12.3%) | <0.01 |
| Low birth weight a | 66 (16.2%) | 41 (5.1%) | 21 (3.3%) | 5 (1.5%) | 133 (6.1%) | <0.01 |
| Pregestational DM | 49 (12.0%) | 128 (14.0%) | 119 (18.1%) | 75 (22.1%) | 371 (17.0%) | <0.01 |
| Type 1 DM | 7 (1.7%) | 15 (1.9%) | 4 (0.6%) | 2 (0.6%) | 28 (1.3%) | |
| Type 2 DM | 42 (10.3%) | 113 (14.0%) | 115 (18.1%) | 73 (22.1%) | 343 (15.7%) | |
| Chronic hypertension | 48 (11.8%) | 93 (11.5%) | 92 (14.5%) | 59 (17.8%) | 292 (13.4%) | 0.021 |
| Gestational DM | 59 (14.5%) | 174 (21.5%) | 178 (28.1%) | 115 (35.7%) | 526 (24.1%) | <0.01 |
| Gestational Hypertension | 2 (0.5%) | 12 (1.5%) | 12 (1.9%) | 13 (3.9%) | 39 (1.8%) | <0.01 |
| Pre‐eclampsia | 3 (0.7%) | 2 (0.2%) | 6 (0.9%) | 5 (1.5%) | 16 (0.7%) | 0.125 |
| Large‐for‐gestational age b | 14 (3.4%) | 81 (10.0%) | 81 (12.8%) | 44 (13.3%) | 220 (10.1%) | <0.01 |
| Small‐for‐gestational age c | 101 (24.8%) | 74 (9.2%) | 35 (5.5%) | 9 (2.7%) | 219 (10.1%) | <0.01 |
Note: BMI, calculated as weight in kilograms divided by the square of height in meters. Data presented as n (%) for dichotomous variables and mean ± SD for continuous variables.
Abbreviations: BMI, body mass index; DM, diabetes mellitus; SD, standard deviation.
Actual birth weight below 2500 g.
Actual birth weight above the 90th percentile.
Actual birth weight below the 10th percentile.
The MAE between the estimated and actual fetal weights was significantly larger in the clinical than in the sonographic EFW when maternal BMI exceeded 25 (222 vs. 202 g for BMI 25–29.9, 238 vs. 205 g for BMI 30–34.9, and 254 vs. 212 g for BMI above 35; P < 0.01) (Table 2, Figure 2).
TABLE 2.
Mean absolute error between actual birth weight and clinical and sonographic EFW, stratified according to maternal BMI.
| BMI Groups | ||||||||
|---|---|---|---|---|---|---|---|---|
| <25 | 25–29.9 | 30–34.9 | ≥35 | |||||
| Sonographic | Clinical | Sonographic | Clinical | Sonographic | Clinical | Sonographic | Clinical | |
| Mean ± SD (g) | 193 ± 152 | 203 ± 161 | 202 ± 156 | 222 ± 184 | 204 ± 154 | 238 ± 187 | 211 ± 161 | 254 ± 207 |
Note: BMI, calculated as weight in kilograms divided by the square of height in meters.
Abbreviations: BMI, body mass index; EFW, estimated fetal weight; SD, standard deviation.
FIGURE 2.
The mean absolute error (MAE) between the estimated and actual fetal weights is significantly larger in the clinical than in the sonographic estimated fetal weight (EFW) when maternal body mass index (BMI) exceeds 25.
The analysis of birth weight percentage deviation by maternal BMI shows that BMI was not associated with sEFW estimation errors. Deviations represent both overestimation and underestimation relative to actual birth weight.
For deviations greater than 10%, the rates were relatively consistent across BMI categories, ranging from 18.4% to 23.0% (P = 0.36). For deviations greater than 20%, the rates remained low and stable at 0.6%–1.7% (P = 0.42). In contrast, clinical estimates displayed a different pattern. While the >10% deviation rates were similar across BMI categories (21.5%–26.5%, P = 0.20), the frequency of more substantial errors (>20% deviation) increased progressively with higher BMI, from 2.0% in women with a BMI <25 to 6.0% in women with a BMI ≥35. This association was statistically significant (P < 0.01) (Table 3).
TABLE 3.
Birth weight percentage deviation stratified according to maternal BMI.
| Overall | BMI | |||||
|---|---|---|---|---|---|---|
| <25 | 25–29.9 | 30–34.9 | ≥35 | P value | ||
| sEFW deviation (%) | ||||||
| >10% | 19.90% | 23.0% | 19.3% | 19.6% | 18.4% | 0.36 |
| >20% | 1.20% | 1.7% | 1.4% | 0.6% | 1.2% | 0.42 |
| cEFW deviation (%) | ||||||
| >10% | 23.60% | 26.5% | 21.5% | 23.5% | 25.7% | 0.2 |
| >20% | 3.80% | 2.0% | 3.3% | 4.6% | 6.0% | <0.01 |
Note: BMI, calculated as weight in kilograms divided by the square of height in meters.
Abbreviations: BMI, body mass index; cEFW, clinical estimated fetal weight; sEFW, sonographic estimated fetal weight.
A multivariate logistic regression model was used to evaluate factors associated with significant EFW errors (greater than 20%), with actual birth weight as the reference (Table 4). The analysis demonstrated that cEFW was significantly more likely than sEFW to exceed the 20% error threshold, and this discrepancy increased proportionally with maternal BMI. Specifically, among women with a BMI of 25–29.9, the aOR for a clinical estimation error greater than 20% was 3.34 (95% CI: 1.45–7.70); for a BMI of 30–34.9, aOR = 5.87 (95% CI: 2.5–13.8); and for a BMI greater than 35, aOR = 9.17 (95% CI: 3.64–23.11), all with P values <0.01.
TABLE 4.
Logistic regression for EFW deviation exceeding 20%.
| Clinical EFW | Sonographic EFW | |||
|---|---|---|---|---|
| aOR | P value | aOR | P value | |
| LGA | 0.80 (0.31–2.05) | 0.64 | 0 | 1 |
| SGA | 13.76 (8.00–23.66) | <0.01 | 3.07 (1.20–7.89) | 0.02 |
| GDM | 1.78 (1.06–2.99) | 0.03 | 1.69 (0.71–4.05) | 0.24 |
| BMI group | ||||
| <25 | Reference | <0.01 | Reference | — |
| 25–29.9 | 3.34 (1.45–7.70) | 1.02 (0.38–2.73) | 0.98 | |
| 30–34.9 | 5.87 (2.50–13.80) | 0.50 (0.14–1.80) | 0.28 | |
| ≥35 | 9.17 (3.64–23.11) | 0.98 (0.26–3.60) | 0.97 | |
Note: BMI, calculated as weight in kilograms divided by the square of height in meters. Variables with P < 0.05 in univariate analysis and those considered clinically relevant (e.g., BMI category, GDM, LGA, SGA) are included in the multivariable models. aOR and 95% CIs are reported.
Abbreviations: aOR, adjusted odds ratio; BMI, body mass index; CI, confidence interval; EFW, estimated fetal weight; GDM, gestational diabetes mellitus; LGA, large‐for‐gestational‐age, SGA, small‐for‐gestational‐age.
Furthermore, among the subgroup of SGA fetuses, cEFW was significantly more likely to differ by over 20% when compared to sEFW, with an aOR of 13.76 (95% CI: 8.00–23.66).
In contrast, sEFW performance was not significantly influenced by maternal BMI. Across all BMI categories, the adjusted odds ratios remained close to unity and did not achieve statistical significance (e.g., BMI 25–29.9: aOR = 1.02, P = 0.98; BMI 30–34.9: aOR = 0.5, P = 0.28; BMI ≥35: aOR = 0.98, P = 0.97).
The diagnostic accuracy of both sonographic and clinically estimated fetal weight for identifying LGA was evaluated using ROC analysis across maternal BMI categories. For sEFW, the AUC was highest in the first group (BMI <25), at 0.934. AUC values declined with increasing BMI: 0.862 for BMI 25–29.9, 0.856 for BMI 30–34.9, and 0.873 for BMI ≥35. For cEFW, a similar pattern of declining AUC was observed as BMI increased. The AUC was 0.925 for BMI <25, 0.854 for BMI 25–25.9, and 0.854 for BMI 30–34.9. However, in women with BMI >35, the clinical EFW showed a notable decline in diagnostic performance, with an AUC of 0.791. When comparing the two methods, sEFW consistently outperformed cEFW across all BMI categories, with the greatest difference observed in women with BMI >35 (AUC 0.873 vs. 0.791) (see Figure 3 for full AUC values).
FIGURE 3.
Receiver operating characteristic (ROC) curves for sonographic and clinical large for gestational age (LGA) prediction as a factor of body mass index (BMI). Sonographic estimated fetal weight (EFW) consistently outperformed clinical EFW in predicting LGA fetuses in all BMI categories, with the most significant discrepancy observed in women with a BMI greater than 35 (area under the curve [AUC] 0.873 vs. 0.791).
4. DISCUSSION
4.1. Principal results
In this retrospective cohort study, we evaluated the accuracy of cEFW versus sEFW across BMI strata. Our findings indicate that sEFW consistently outperformed cEFW in predicting both actual birth weight and the presence of LGA neonates, particularly among women with elevated BMI.
As maternal BMI increased, cEFW showed greater inaccuracy, as evidenced by higher mean absolute error, diminished correlation with neonatal birth weight, and reduced discriminatory performance for LGA detection.
4.2. Compared to what we know
The influence of maternal habitus on the accuracy of estimated fetal weight has been a subject of ongoing investigation, with prior studies reporting mixed results. Daykan et al. reported greater accuracy of sEFW than clinical estimation in tall, term‐pregnant women, but only in 329 women. 22 Sgayer et al. conducted a prospective analysis examining whether maternal pre‐delivery BMI influenced the accuracy of sEFW. They found no significant association between maternal BMI and the precision of the ultrasound‐based weight estimates. 23 Similarly, Melamed et al. evaluated the performance of both methods among obese women. They concluded that while both estimations showed reduced accuracy in higher BMI groups, the difference between the two methods did not reach clinical significance. 24 These findings suggest that under certain conditions, clinical estimation may remain a viable alternative when sonography is unavailable. However, these studies are relatively small.
Conversely, other studies have highlighted the superiority of sEFW, especially in patients with increased adiposity. Farrell et al. reported that, in a cohort of term pregnancies, sonographic estimates were significantly more accurate than clinical estimates among women with BMI >30, with notably lower mean percentage error and standard deviation. 25 Similarly, Deter et al. demonstrated that ultrasound‐based methods provided significantly better fetal weight predictions across all BMI ranges, and that the discrepancy between methods increased proportionally with maternal BMI. 26 These studies support our findings by confirming that sEFW is less affected by maternal body composition and tactile limitations than clinical palpation.
Our study also adds to the literature on the relative accuracy of EFW methods in SGA and LGA fetuses. A survey by Ben‐Haroush et al. found that sonographic estimation significantly outperformed clinical palpation in identifying LGA fetuses, with clinical estimates often underestimating birth weight in macrosomic neonates. 27 Chauhan et al. reported that cEFW was significantly less accurate than sonographic methods in detecting both SGA and LGA infants, particularly in nulliparous women. 28 These findings align with our own, which show that cEFW is more prone to extreme deviations at both ends of the birth weight spectrum.
Specifically, multiple studies have reported inferior clinical performance in SGA fetuses. Gifty et al. conducted a prospective blinded study comparing sEFW with clinical estimation at term. They found that ultrasound had slightly higher sensitivity for detecting low birth weight (<2.5 kg), whereas clinical estimates frequently overestimated fetal weight, particularly among less‐experienced examiners. 29 Dude et al. evaluated the performance of sonographic EFW for identifying SGA infants across BMI categories. They found universally poor sensitivity (≤50%) and a low positive predictive value, particularly in obese patients. Clinical estimates fared even worse in individuals with high BMI and in cases with atypical amniotic fluid volumes, reinforcing the challenges of accurately detecting growth restriction under these conditions. 30 The increased error observed in SGA fetuses may be explained by altered fetal body composition, reduced soft‐tissue mass, and limitations of both palpation and biometric formulas in accurately measuring smaller fetal dimensions. These factors may contribute to systematic under‐ or overestimation in both clinical and sonographic methods.
In terms of diagnostic performance for detecting LGA fetuses, sEFW again proved superior. Among women with a BMI greater than 35, the AUC for sEFW was 0.873, compared with 0.791 for cEFW, indicating a significant difference in discriminatory ability.
As previously mentioned, the MAE between the estimated and actual fetal weights was significantly larger in the clinical than in the sonographic EFW when maternal BMI exceeded 25. Although statistically significant, the absolute differences in MAE (20–40 g) are modest and may have limited clinical significance at the individual level.
4.3. Clinical and research implications
These findings have significant clinical implications that extend beyond maternal obesity. Accurate fetal weight estimation is crucial for several critical decisions in obstetric care, including timing and mode of delivery, as well as neonatal resuscitation preparedness. The demonstrated superiority of sEFW in high‐BMI populations supports the prioritization of ultrasound‐based assessment, particularly in settings where the consequences of fetal over‐ or underestimation, such as missed macrosomia or undetected growth restriction, can lead to adverse maternal or neonatal outcomes.
Maternal GWG should be interpreted in the context of pre‐pregnancy BMI. According to the Institute of Medicine (IOM) recommendations, the advised total GWG is 11.5–16.0 kg for women with normal BMI, 7.0–11.5 kg for overweight women, and 5.0–9.0 kg for women with obesity. 31 Excess GWG has been associated with fetal overgrowth, higher birth weight, and adverse obstetric outcomes. 32 In our cohort, GWG patterns among overweight and obese women are notable in light of these recommendations and parallel the increase in neonatal birth weight across BMI strata. This may contribute to the higher prevalence of LGA and to the reduced accuracy of clinical fetal weight estimation at higher BMI.
These findings may inform counseling regarding shoulder dystocia risk and mode of delivery, particularly in women with obesity and suspected LGA. In such patients, reliance on cEFW alone, given its greater inaccuracy at higher BMI, may underestimate fetal size and thereby shoulder dystocia risk. Since sEFW is not routinely done at term, incorporating sEFW into late‐pregnancy assessment could support more informed shared decision making about induction timing, trial of labor versus planned cesarean, and intrapartum preparedness for shoulder dystocia. 33
Our data suggest that women with elevated BMI and those with additional risk factors for abnormal fetal growth (e.g., diabetes, hypertensive disorders) may particularly benefit from third‐trimester sEFW close to delivery. However, our retrospective design and indication‐based cohort do not allow us to define optimal universal screening criteria, and prospective studies are needed to determine which patients should routinely undergo late‐pregnancy ultrasound.
In settings where a formal late‐third‐trimester ultrasound is not readily available, bedside (point‐of‐care) ultrasound may provide an alternative sonographic EFW to support counseling, provided that operators are appropriately trained and aware of measurement uncertainty.
Because diabetes is clinically relevant to fetal growth, diabetes variables were included in multivariable models of >20% error, and we discuss the potential role of diabetes in late‐pregnancy ultrasound selection and EFW accuracy. A dedicated BMI‐stratified diabetic subgroup analysis was not performed due to limited power within strata and incomplete indication metadata.
The strengths of the present study include its large sample size, stratification by standardized BMI categories, and the use of both continuous (MAE, correlation coefficients) and categorical (ROC/AUC, >10%–20% deviation) accuracy measures. Performing clinical and sonographic EFW within a narrow time window before delivery enhances comparability.
Limitations of this study include its retrospective design and the potential for interobserver variability in both clinical and sonographic fetal weight estimation. Although all assessments were performed by trained providers, the accuracy of ultrasound imaging and abdominal palpation may still vary according to maternal habitus, and residual confounding cannot be entirely excluded. Provider training level was not uniformly captured in the dataset, precluding stratified analysis by operator experience.
In addition, BMI was used as a global proxy for adiposity, and we lacked standardized data on fat distribution (e.g., central vs. peripheral), which may influence both palpation and ultrasound image quality.
Only 11% of eligible women had both cEFW and sEFW within 10 days of delivery, reflecting real‐world practice where late third‐trimester ultrasound is performed selectively rather than routinely. This selection pattern may limit generalizability; however, the observed decline in clinical palpation accuracy with rising BMI was consistent and biologically plausible.
Finally, restricting the analysis to in‐house sonographic examinations enhanced methodological consistency but may limit the generalizability of our findings to other clinical settings.
5. CONCLUSION
Maternal BMI significantly influences the accuracy of estimated fetal weight, with cEFW becoming progressively less reliable as BMI increases. sEFW demonstrates superior and stable performance across all BMI groups and should be considered the preferred method for estimated fetal weight, particularly in overweight and obese patients. These findings reinforce the need for objective, imaging‐based evaluation of fetal size in prenatal care and delivery planning, especially as maternal obesity continues to rise globally.
AUTHOR CONTRIBUTIONS
Amir Krencel, Bar Narkis, Rotem Yedidia Moser, Or Bercovich, Natav Hendin, Asaf Romano, Eran Hadar, Yinon Gilboa, and Ohad Houri each made substantial contributions to the conception and design of the work and/or the acquisition, analysis, or interpretation of data. All authors contributed to drafting the manuscript and/or revising it critically for important intellectual content. All authors gave final approval of the version to be published and agree to be accountable for all aspects of the work, ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
FUNDING INFORMATION
No financial support was received for this study.
CONFLICT OF INTEREST STATEMENT
The authors report no conflict of interest.
Krencel A, Narkis B, Yedidia Moser R, et al. Influence of maternal body mass index on the accuracy of clinical and sonographic fetal weight estimation. Int J Gynecol Obstet. 2026;175:263‐272. doi: 10.1002/ijgo.71012
The paper or any part of it has not been presented or accepted for an upcoming presentation at a scientific meeting.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
REFERENCES
- 1. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980‐2013: a systematic analysis for the global burden of disease study 2013. Lancet. 2014;384(9945):766‐781. doi: 10.1016/S0140-6736(14)60460-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Abenhaim HA, Kinch RA, Morin L, Benjamin A, Usher R. Effect of prepregnancy body mass index categories on obstetrical and neonatal outcomes. Arch Gynecol Obstet. 2007;275(1):39‐43. doi: 10.1007/s00404-006-0219-y [DOI] [PubMed] [Google Scholar]
- 3. Catalano PM, Ehrenberg HM. The short‐ and long‐term implications of maternal obesity on the mother and her offspring. BJOG. 2006;113(10):1126‐1133. doi: 10.1111/j.1471-0528.2006.00989.x [DOI] [PubMed] [Google Scholar]
- 4. Henriksen T. The macrosomic fetus: a challenge in current obstetrics. Acta Obstet Gynecol Scand. 2008;87(2):134‐145. doi: 10.1080/00016340801899289 [DOI] [PubMed] [Google Scholar]
- 5. Weiss JL, Malone FD, Emig D, et al. Obesity, obstetric complications and cesarean delivery rate‐‐a population‐based screening study. Am J Obstet Gynecol. 2004;190(4):1091‐1097. doi: 10.1016/j.ajog.2003.09.058 [DOI] [PubMed] [Google Scholar]
- 6. Kominiarek MA, Chauhan SP. Obesity before, during, and after pregnancy: a review and comparison of five National Guidelines. Am J Perinatol. 2016;33(5):433‐441. doi: 10.1055/s-0035-1567856 [DOI] [PubMed] [Google Scholar]
- 7. Dudley NJ. A systematic review of the ultrasound estimation of fetal weight. Ultrasound Obstet Gynecol. 2005;25(1):80‐89. doi: 10.1002/uog.1751 [DOI] [PubMed] [Google Scholar]
- 8. Balsyte D, Schäffer L, Burkhardt T, Wisser J, Kurmanavicius J. Sonographic prediction of macrosomia cannot be improved by combination with pregnancy‐specific characteristics. Ultrasound Obstet Gynecol. 2009;33(4):453‐458. doi: 10.1002/uog.6282 [DOI] [PubMed] [Google Scholar]
- 9. Melamed N, Yogev Y, Meizner I, Mashiach R, Ben‐Haroush A. Sonographic prediction of fetal macrosomia: the consequences of false diagnosis. J Ultrasound Med. 2010;29(2):225‐230. doi: 10.7863/jum.2010.29.2.225 [DOI] [PubMed] [Google Scholar]
- 10. pryer O, Husslein H, Concin N, et al. Fetal weight estimation at term—ultrasound versus clinical examination with Leopold's manoeuvres: a prospective blinded observational study. BMC Pregnancy Childbirth. 2019;19(1):122. doi: 10.1186/s12884-019-2251-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Chauhan SP, Hendrix NW, Magann EF, Morrison JC, Kenney SP, Devoe LD. Limitations of clinical and sonographic estimates of birth weight: experience with 1034 parturients. Obstet Gynecol. 1998;91(1):72‐77. doi: 10.1016/s0029-7844(97)00590-5 [DOI] [PubMed] [Google Scholar]
- 12. Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Estimation of fetal weight with the use of head, body, and femur measurements‐‐a prospective study. Am J Obstet Gynecol. 1985;151(3):333‐337. doi: 10.1016/0002-9378(85)90298-4 [DOI] [PubMed] [Google Scholar]
- 13. Scioscia M, Vimercati A, Ceci O, Vicino M, Selvaggi LE. Estimation of birth weight by two‐dimensional ultrasonography: a critical appraisal of its accuracy. Obstet Gynecol. 2008;111(1):57‐65. doi: 10.1097/01.AOG.0000296656.81143.e6 [DOI] [PubMed] [Google Scholar]
- 14. Hendrix NW, Grady CS, Chauhan SP. Clinical vs. sonographic estimate of birth weight in term parturients. A randomized clinical trial. J Reprod Med. 2000;45(4):317‐322. [PubMed] [Google Scholar]
- 15. Peregrine E, O'Brien P, Jauniaux E. Clinical and ultrasound estimation of birth weight prior to induction of labor at term. Ultrasound Obstet Gynecol. 2007;29(3):304‐309. doi: 10.1002/uog.3949 [DOI] [PubMed] [Google Scholar]
- 16. Ghosal S, Gleason JL, Grantz KL, Chen Z. Maternal BMI and diagnostic accuracy of using estimated fetal growth to predict abnormal birthweight: Results from NICHD fetal growth studies. Diagnostics (Basel). 2025;15(11):1398. doi: 10.3390/diagnostics15111398 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Gaudet L, Ferraro ZM, Wen SW, Walker M. Maternal obesity and occurrence of fetal macrosomia: a systematic review and meta‐analysis. Biomed Res Int. 2014;2014:640291. doi: 10.1155/2014/640291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Hadlock FP, Harrist RB, Martinez‐Poyer J. In utero analysis of fetal growth: a sonographic weight standard. Radiology. 1991;181(1):129‐133. doi: 10.1148/radiology.181.1.1887021 [DOI] [PubMed] [Google Scholar]
- 19. Sherman DJ, Arieli S, Tovbin J, Siegel G, Caspi E, Bukovsky I. A comparison of clinical and ultrasonic estimation of fetal weight. Obstet Gynecol. 1998;91(2):212‐217. doi: 10.1016/s0029-7844(97)00654-6 [DOI] [PubMed] [Google Scholar]
- 20. Dollberg S, Seidman DS, Armon Y, Stevenson DK, Gale R. Standardized birthweight curves for Israeli newborns. Isr J Med Sci. 1996;32(3–4):128‐132. [Google Scholar]
- 21. Obesity in pregnancy: ACOG practice bulletin, number 230. Obstet Gynecol. 2021;137(6):e128‐e144. doi: 10.1097/AOG.0000000000004395 [DOI] [PubMed] [Google Scholar]
- 22. Daykan Y, Shavit M, Yagur Y, et al. Fetal weight estimation in tall women: is ultrasound more accurate than clinical assessment? A prospective trial. Arch Gynecol Obstet. 2022;305(3):567‐572. doi: 10.1007/s00404-021-06177-6 [DOI] [PubMed] [Google Scholar]
- 23. Sgayer I, Awwad S, Aiob A, Mikhail SM, Lowenstein L, Odeh M. Pre‐delivery BMI and the accuracy of fetal weight estimation in very preterm infants. J Obstet Gynaecol Can. 2024;46(10):102643. doi: 10.1016/j.jogc.2024.102643 [DOI] [PubMed] [Google Scholar]
- 24. Melamed N, Yogev Y, Meizner I, Mashiach R, Bardin R, Ben‐Haroush A. Sonographic fetal weight estimation: which model should be used? J Ultrasound Med. 2009;28(5):617‐629. doi: 10.7863/jum.2009.28.5.617 [DOI] [PubMed] [Google Scholar]
- 25. Farrell T, Holmes R, Stone P. The effect of body mass index on three methods of fetal weight estimation. BJOG. 2002;109(6):651‐657. doi: 10.1111/j.1471-0528.2002.01249.x [DOI] [PubMed] [Google Scholar]
- 26. Deter RL, Lee W, Sangi‐Haghpeykar H, Tarca AL, Yeo L, Romero R. Individualized fetal growth assessment: critical evaluation of key concepts in the specification of third trimester size trajectories. J Matern Fetal Neonatal Med. 2014;27(6):543‐551. doi: 10.3109/14767058.2013.833904 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Ben‐Haroush A, Yogev Y, Bar J, et al. Accuracy of sonographically estimated fetal weight in 840 women with different pregnancy complications prior to induction of labor. Ultrasound Obstet Gynecol. 2004;23(2):172‐176. doi: 10.1002/uog.940 [DOI] [PubMed] [Google Scholar]
- 28. Chauhan SP, West DJ, Scardo JA, Boyd JM, Joiner J, Hendrix NW. Antepartum detection of macrosomic fetus: clinical versus sonographic, including soft‐tissue measurements. Obstet Gynecol. 2000;95(5):639‐642. doi: 10.1016/s0029-7844(99)00606-7 [DOI] [PubMed] [Google Scholar]
- 29. Gifty D, Cherian NE, Raj SLA, Yenuberi HG, Rathore S, Mathews JE. A prospective blinded diagnostic accuracy study to assess the accuracy of fetal weight estimation at term by clinical method using Leopold's maneuvers and ultrasonography in predicting the actual birth weight. J Family Med Prim Care. 2024;13(10):4499‐4504. doi: 10.4103/jfmpc.jfmpc_221_24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Dude AM, Davis B, Delaney K, Yee LM. Identifying fetal growth disorders using ultrasound in obese nulliparous women. J Matern Fetal Neonatal Med. 2021;34(11):1768‐1773. doi: 10.1080/14767058.2019.1648420 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Rasmussen KM, Yaktine AL, Institute of Medicine (US) and National Research Council (US) Committee to Reexamine IOM Pregnancy Weight Guidelines . Weight Gain during Pregnancy: Reexamining the Guidelines. National Academies Press (US); 2009. doi: 10.17226/12584 [DOI] [PubMed] [Google Scholar]
- 32. American College of Obstetricians and Gynecologists . ACOG Committee opinion no. 548: weight gain during pregnancy. Obstet Gynecol. 2013;121(1):210‐212. doi: 10.1097/01.aog.0000425668.87506.4c [DOI] [PubMed] [Google Scholar]
- 33. Duewel AM, Doehmen J, Dittkrist L, et al. Antenatal risk score for prediction of shoulder dystocia with focus on fetal ultrasound data. Am J Obstet Gynecol. 2022;227(5):753.e1‐753.e8. doi: 10.1016/j.ajog.2022.06.008 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.