If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Fetal growth standards: the NICHD fetal growth study approach in context with INTERGROWTH-21st and the World Health Organization Multicentre Growth Reference Study
Epidemiology Branch, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD
Division of Intramural Population Health Research, Office of the Director, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD
Division of Intramural Population Health Research, and Biostatistics and Bioinformatics Branch, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MDBiostatistics Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD
Division of Intramural Population Health Research, Office of the Director, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD
Three recently completed longitudinal cohort studies have developed intrauterine fetal growth charts, one in the United States and two international. This expert review compares and contrasts the Eunice Kennedy Shriver National Institute of Child Health and Human Development Fetal Growth Studies, INTERGROWTH-21st and World Health Organization Multicentre Growth Reference Study conclusions in light of differences in aims, sampling frames, and analytical approaches. An area of controversy is whether a single growth reference is representative of growth, regardless of ethnic or country origin. The INTERGROWTH and World Health Organization Fetal studies used a similar approach as the World Health Organization Multicentre Growth Reference Study for infants and children, the aim of which was to create a single international reference for the best physiological growth for children aged 0–5 years. INTERGROWTH made the same assumption (ie, that there would be no differences internationally among countries or racial/ethnic groups in fetal growth when conditions were optimal). INTERGROWTH found differences in crown-rump length and head circumference among countries but interpreted the differences as not meaningful and presented a pooled standard. The World Health Organization Multicentre Growth Reference Study was designed to create a pooled reference, although they evaluated for and presented country differences, along with discussion of the implications. The Eunice Kennedy Shriver National Institute of Child Health and Human Development Study was designed to assess whether racial/ethnic-specific fetal growth standards were needed, in recognition of the fact that fetal size is commonly estimated from dimensions (head circumference, abdominal circumference, and femur length) in which there are known differences in children and adults of differing racial/ethnic groups. A pooled standard would be derived if no racial/ethnic differences were found. Highly statistically significant racial/ethnic differences in fetal growth were found resulting in the publication of racial/ethnic-specific derived standards. Despite all 3 studies including low-risk status women, the percentiles for fetal dimensions and estimated fetal weight varied among the studies. Specifically, at 39 weeks, the 50th percentile for estimated fetal weight was 3502 g for whites, 3330 g for Hispanics, 3263 g for Asians, and 3256 for blacks in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Study, compared with 3186 g for INTERGROWTH and 3403 g for World Health Organization Multicentre Growth Reference Study. When applying these standards to a clinical population, it is important to be aware that different percentages of small- and large-for-gestational-age fetuses will be identified. Also, it may be necessary to use more restrictive cut points, such as the 2.5th or 97.5th, for small-for-gestational-age or large-for-gestational-age fetuses, respectively. Ideally, a comparison of diagnostic accuracy, or misclassification rates, of small-for-gestational-age and large-for-gestational-age fetuses in relation to morbidity and mortality using different criteria is necessary to make recommendations and remains an important data gap. Identification of the appropriate percentile cutoffs in relation to neonatal morbidity and mortality is needed in local populations, depending on which fetal growth chart is used. On a final point, assessment of fetal growth with a one-time measurement remains standard clinical practice, despite recognition that a single measurement can indicate only size. Ultimately, it is knowledge about fetal growth in addition to other factors and clinical judgment that should trigger intervention.
Fetal growth is monitored in pregnancies to ensure fetal well-being and to intervene in the context of maternal or fetal pathology, yet there are many challenges in distinguishing normal from abnormal growth.
Traditionally, cross-sectional fetal biometrics and estimated fetal weight (EFW), calculated using a formula with various combinations of fetal measurements, such as the head circumference, abdominal circumference, and femur length, are compared with reference size-for-gestational-age curves to generate a percentile, with a range of 10th to 90th percentiles often considered appropriate for gestational age.
The choice in reference will therefore affect the percentage of fetuses that are identified as small or large for gestational age (SGA or LGA; often defined as <10th or ≥90th percentiles, respectively). Regarding EFW, intrauterine estimates of fetal weight by ultrasound are highly (r = 0.80–0.91) correlated with actual birthweight, although they can differ by ≥100 g and are more inaccurate at the extremes of EFW, <2000 g and >4000 g.
In theory then, birthweight references, whereby weight is measured directly as opposed to estimated, might seem preferable to assess fetal growth. However, birthweight-for-gestational-age reference percentiles inaccurately describe the preterm growth of fetuses who go on to deliver at term because infants who deliver preterm are more likely to be growth restricted.
Therefore, intrauterine references, despite the drawbacks of estimating fetal weight from ultrasound measurements, tend to be preferred to birthweight references for clinical antepartum monitoring.
that is commonly used in the United States included 392 white women from a single center in Texas where each fetus contributed a single ultrasound, limiting the ability to determine fetal growth prospectively. Until recently, longitudinal ultrasound-based references were based on relatively small studies comprising mostly white women, although larger studies existed outside the United States.
In light of critical data gaps about optimal fetal growth to aid clinical management of pregnant women, 3 longitudinal cohort studies were undertaken and provide new insights about contemporary fetal growth and how best to assess fetal growth: one in the United States, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
However, each has slightly different research aims that have an impact on the interpretation of the findings. We compare and contrast these 3 studies to aid in the application and clinical interpretation.
Approaches and assumptions
One of the main areas of dispute in the area of fetal and child growth is whether a single growth reference is representative of growth, regardless of ethnic or country origin. The INTERGROWTH
Specifically, the WHO MGRS was predicated on the notion that infants and children of well-off parents and whose feeding met the WHO breast-feeding criteria represent optimal growth in size, and the WHO Fetal study was designed as a subsequent study to extend the WHO MGRS to the fetal period.
The INTERGROWTH and WHO Fetal studies on fetal growth started with the same assumption, that there would be no differences internationally in fetal growth when conditions were optimal. For the WHO MGRS, differences in length (0–2 years) and height (2–5 years) were evaluated among 6 countries: Brazil (South America), Ghana (Africa), India (Asia), Norway (Europe), Oman (Middle East), and the United States (North America). Without formal hypothesis testing, these differences were interpreted as small enough to not be meaningful, so the final decision was to create 1 child growth standard.
WHO Multicentre Growth Reference Study Group Assessment of differences in linear growth among populations in the WHO Multicentre Growth Reference Study.
Additional measurements included head circumference, mid–upper-arm circumference, triceps, and subscapular skinfolds, but differences across countries in these dimensions were not tested. INTERGROWTH evaluated for differences in crown-rump length (CRL), head circumference (HC), and newborn length among countries, concluding that the differences were small enough, before pooling.
The likeness of fetal growth and newborn size across non-isolated populations in the INTERGROWTH-21st Project: the Fetal Growth Longitudinal Study and Newborn Cross-Sectional Study.
so the assumption that the small fetal differences do not reflect differences that will persist and be meaningful in infancy and childhood may need to reconsidered.
A key determinant of INTERGROWTH’s decision to pool across sites was whether the standardized site difference at different gestational ages was from a somewhat arbitrary range of –0.5 to 0.5 SD units.
The likeness of fetal growth and newborn size across non-isolated populations in the INTERGROWTH-21st Project: the Fetal Growth Longitudinal Study and Newborn Cross-Sectional Study.
Inappropriate interpretation of centiles could have resulted from pooling of sites given the allowed wide range of standardized site difference. To show this potential, our group previously calculated the probability of being below the lower limit of the standard for a particular site when the standard was constructed using data pooled across different sites for these values.
The probability of being less than the fifth centile varied according to the range of standardized site difference. Specifically, when the standardized site difference was 0.5, then 3.4% of fetuses (targeted centile–pooled centile = 5.0–1.6%) would have been misclassified as not extreme and 7.6% (targeted centile–pooled centile = 12. 6–5. 0%) of fetuses would have been misclassified as extreme.
INTERGROWTH reported the magnitude of within- and between-site variation, and some of the variances reported might be highly statistically significant. Furthermore, INTERGROWTH evaluated only CRL and HC. CRL is known to not vary as much, and HC also has less variation, as demonstrated in the results below. WHO Fetal was designed to create a pooled reference, although they evaluated for and showed country differences along with discussion of the implications.
Unlike the 2 international studies, the NICHD study was designed to assess whether racial/ethnic fetal growth standards were needed, in recognition of the fact that because fetal size is commonly estimated from dimensions, particularly postcranial dimensions (ie, abdominal circumference [AC] and femur length [FL]), in which there are known differences in children and adults of differing racial/ethnic groups, separate standards might be necessary to capture optimal growth and more precisely estimate fetal weight.
Highly statistically significant racial/ethnic differences in fetal growth were found, and by order of detection were as follows: humerus and femur lengths (beginning as early as 10 weeks), abdominal circumference (16 weeks), head circumference (21 weeks), and biparietal diameter (27 weeks) with racial/ethnic differences continuing throughout gestation, so racial/ethnic-specific curves were derived.
One of the continuing debates in designing and conducting an ultrasound or any physical growth study is how to select study subjects, and the terminology of reference vs standards. An ultrasound reference is a sample of pregnancies from a population and by definition contains high-risk pregnancies at risk for fetal growth restriction or overgrowth, including preexisting conditions and pregnancy complications.
An ultrasound standard includes fetuses at low risk for growth disturbances, with the goal of describing how all fetuses should grow, as opposed to traditional reference charts that describe how some have grown at a given place and time. However, distinguishing the normal from abnormal fetal growth remains a challenge, and the term standard in regard to fetal growth is controversial.
It is also important to note that because standards are variance restricted, their percentiles and interpretation are not the same as previous references. For example, the fifth percentile of a reference is not equivalent to the fifth percentile of a standard, in which fetuses are at lower risk for growth aberrations. It may be necessary to use more restrictive cut points, such as the 2.5th or 97.5th percentiles, for SGA or LGA, respectively.
All 3 studies selected healthy women who were positioned for optimal fetal growth and had a known last menstrual period, although the specific inclusion and exclusion criteria varied. The cohort profiles of main differences for the 3 studies are presented in Figure 1. One of the main differences was the racial/ethnic and country variation in women recruited by the 3 studies. The NICHD Study was conducted at 12 US sites (New York [2], New Jersey, Delaware, Rhode Island, Massachusetts, South Carolina, Alabama, Illinois, and California [3]), INTERGROWTH was completed in 8 countries (Brazil, China, India, Italy, Kenya, Oman, the United Kingdom, and the United States), and the WHO Fetal study in 10 countries (Argentina, Brazil, Democratic Republic of the Congo, Denmark, Egypt, France, Germany, India, Norway, and Thailand).
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
AC, abdominal length; EFW, estimated fetal weight; FL, fetal length; HC, head circumference; INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Another main difference was the exclusion of pregnancy complications and fetal factors such as congenital anomalies and stillbirth from the NICHD Study and INTERGROWTH, given the intention of creating standards; the NICHD defined additional a priori exclusion criteria or preterm delivery <37 weeks’ gestation and karyotype abnormalities, neither of which was excluded from INTERGROWTH. WHO Fetal did not exclude pregnancies with complications, with the rationale that they wanted their study to be more of a reference.
The 3 studies used different statistical analytic approaches to model the fetal growth trajectories and calculate the corresponding percentiles. Both INTERGROWTH and NICHD assumed a parametric distribution of the fetal growth trajectories, under a linear mixed model, in which the methods used to create a smoothed mean trajectory differed slightly. After log transformation, the fetal growth data can be reasonably modeled by the linear mixed-effects models, assuming normally distributed random effects and error terms.
WHO Fetal used quantile regression to estimate percentiles directly and made somewhat fewer restrictive assumptions. Despite different model assumptions and smoothing techniques, the approaches are flexible to capture the smooth fetal growth trajectories so that they should yield similar results when applied to the same fetal growth data set. In other words, any differences in results are unlikely due to the different methods used, although they have not been rigorously compared.
The analyses also adjusted for different covariates, which are not able to be summarized here because of the many analyses. For example, the WHO Fetal primary analyses did not adjust for country but secondary analyses adjusted for county and full interaction between country and gestational age.
Another slight difference among studies was the gestational ages that were included in the models to calculate EFW percentiles. Given the clinical uncertainty associated with EFW before 15 weeks, NICHD originally did not intend to report EFW based on the actual measurements taken by sonographers between 10 and 14 weeks but to extrapolate data from the measurements taken at ≥15 weeks, but subsequently they reran the analysis using the actual measurements for EFW at 10–14 weeks and not the extrapolated data because other fetal parameters were presented for that gestational period.
Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO Fetal studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
Study findings
EFW comparison among the 3 studies is presented in Figure 2 and Table 1 (no statistical testing was performed). EFW was plotted for the published estimated third, 50th, and 97th percentiles for INTERGROWTH and NICHD; WHO Fetal published the 2.5th and 97.5th percentiles to approximate ±2 SD. Despite all 3 studies including women with a low-risk status, the percentiles for fetal biometrics and EFW varied among the studies.
Figure 2Estimated fetal weight comparison among the 3 studies
Distribution of the EFW by race/ethnicity and gestation, NICHD Fetal Growth Study–Singletons, INTERGROWTH-21st, and WHO Fetal for 24–40 weeks of gestation (A) and 36–40 weeks of gestation (B). Estimated third, 50th, and 97th percentiles for fetal weight by study; ∗note that values are the 2.5th and 97.5th percentiles for the WHO Fetal study. Also, NICHD and WHO Fetal calculated the EFW from HC, AC, and FL using the Hadlock 1985 formula,
AC, abdominal length; EFW, estimated fetal weight; FL, fetal length; HC, head circumference; INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days)
Note that NICHD and WHO Fetal calculated EFW from HC, AC, and FL using the Hadlock 1985 formula,26 while INTERGROWTH created a new formula12 based on only HC and AC.
NICHD white
NICHD Hispanic
NICHD Asian
NICHD black
INTERGROWTH
WHO Fetal
24
674
651
640
647
668
665
25
787
758
745
751
756
778
26
912
876
862
866
856
902
27
1050
1007
990
994
969
1039
28
1202
1151
1132
1134
1097
1189
29
1369
1311
1287
1289
1239
1350
30
1552
1486
1456
1459
1396
1523
31
1749
1676
1637
1642
1568
1707
32
1960
1879
1830
1837
1755
1901
33
2180
2090
2031
2040
1954
2103
34
2408
2307
2238
2247
2162
2312
35
2637
2521
2448
2452
2378
2527
36
2864
2731
2656
2654
2594
2745
37
3086
2935
2862
2854
2806
2966
38
3299
3134
3065
3054
3006
3186
39
3502
3330
3263
3256
3186
3403
40
3693
3525
3455
3466
3338
3617
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
b Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days)
c Note that NICHD and WHO Fetal calculated EFW from HC, AC, and FL using the Hadlock 1985 formula,
Starting at 26 weeks of gestation and continuing through 40 weeks of gestation, the 50th percentile EFW for INTERGROWTH was smaller than the 50th percentile EFW for WHO Fetal and all racial/ethnic groups in NICHD. The 50th percentile EFW for WHO Fetal was between the NICHD EFW for white and Hispanic women. Specifically, at 39 weeks, the 50th percentile for EFW was 3502 g for white, 3330 g for Hispanic, 3263 g for Asian, and 3256 g for black in the NICHD Study, compared with 3186 g for the INTERGROWTH and 3403 g for the WHO Fetal.
WHO Fetal found country-specific differences similar to NICHD findings for race/ethnicity. Quantile regression with country as a covariate, and interaction terms with gestational age, demonstrated statistically significant variation in fetal growth among countries. For example, at term the fifth percentile for Norway was 3200 g, while it was 2700 g for Egypt and 2800 g using the pooled data from all countries, differences that were also apparent in birthweight. While acknowledging these differences, WHO Fetal chose to present a pooled standard with the rationale that the primary purpose of the study was to develop fetal standards to complement the WHO MGRS
Because differences in EFW are difficult to interpret in light of the different EFW formulas, we also compared fetal biometry among the 3 studies presented in Figure 3, Figure 4, Figure 5 and Table 2, Table 3, Table 4 The differences in AC among studies paralleled that of EFW.
Figure 3Head circumference comparison among the 3 studies
Distribution of head circumference, NICHD Fetal Growth Study–Singletons, INTERGROWTH-21st, and WHO Fetal. Estimated fifth, 50th, and 95th percentiles for head circumference are by race/ethnicity and gestation and study.
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Distribution of head circumference, NICHD Fetal Growth Study–Singletons, INTERGROWTH-21st, and WHO Fetal. Estimated fifth, 50th, and 95th percentiles for abdominal circumference are by race/ethnicity and gestation and study.
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Distribution of head circumference, NICHD Fetal Growth Study–Singletons, INTERGROWTH-21st, and WHO Fetal. Estimated fifth, 50th, and 95th percentiles are for femur length by race/ethnicity and gestation and study.
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
Head circumference (mm) 50th percentiles
NICHD white
NICHD Hispanic
NICHD Asian
NICHD black
INTERGROWTH
WHO Fetal
24
219
218
217
218
219
222
25
231
229
228
229
230
233
26
242
240
239
239
241
244
27
253
250
250
250
251
254
28
263
260
260
259
260
264
29
273
270
269
269
270
273
30
282
279
278
278
278
281
31
290
287
287
286
287
289
32
299
295
295
294
294
296
33
306
303
302
301
302
303
34
313
309
308
307
308
309
35
318
315
314
312
314
315
36
324
320
320
317
319
321
37
328
324
324
321
324
326
38
332
327
328
324
328
332
39
335
330
331
328
331
337
40
338
333
333
331
334
342
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
b Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO Fetal studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
Abdominal circumference (mm) 50th percentiles
NICHD white
NICHD Hispanic
NICHD Asian
NICHD black
INTERGROWTH
WHO Fetal
24
198
195
194
191
191
197
25
210
206
205
202
202
208
26
221
216
215
212
212
219
27
231
227
226
222
223
230
28
242
238
236
232
233
240
29
253
248
247
243
244
250
30
264
259
257
254
254
260
31
275
270
268
265
264
269
32
287
282
279
275
274
279
33
298
293
289
286
284
288
34
308
303
299
296
294
298
35
319
313
309
306
303
307
36
329
323
318
315
313
317
37
338
332
327
324
322
328
38
347
340
336
333
332
338
39
355
348
345
342
341
350
40
361
356
353
351
350
363
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
b Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO Fetal studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
Femur length (mm) 50th percentiles
NICHD white
NICHD Hispanic
NICHD Asian
NICHD black
INTERGROWTH
WHO Fetal
24
43
42
42
43
42
43
25
45
45
44
46
44
46
26
48
47
47
48
47
48
27
50
49
49
50
49
50
28
52
52
51
53
51
52
29
54
54
53
55
53
54
30
56
56
56
57
56
56
31
58
58
58
59
58
59
32
60
60
60
61
60
61
33
62
62
61
63
61
63
34
64
64
63
65
63
65
35
66
66
65
67
65
67
36
68
67
67
68
66
69
37
69
69
68
70
68
70
38
71
70
70
71
69
72
39
72
71
71
73
71
73
40
73
73
72
74
72
73
INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
b Results were reported for the exact day (eg, 16.0 weeks) for the NICHD and WHO studies, while INTERGROWTH results were reported for completed weeks (eg, 16 weeks = 16 weeks 0 days to 16 weeks 6 days).
To directly compare the 2 EFW formulas, we used the NICHD data to calculate EFW using the Hadlock 1985 formula and INTERGROWTH formula (Figure 6). The INTERGROWTH EFW formula performed very close to the NICHD Asian racial/ethnic group but differed from the 3 other racial/ethnic groups. Additional application studies are needed in different populations to assess whether the INTERGROWTH EFW formula outperforms the Hadlock or other EFW formulas in identifying fetuses with other signs of compromise.
Figure 6Comparison of estimated fetal weight formulas
from HC, AC, and FL (add reference) and INTERGROWTH formula based on HC and AC. EFW was calculated using the Hadlock formula: Log10 weight = 1.326 – 0.00326 AC × FL + 0.0107 HC + 0.0438 AC + 0.158 FL and INTERGROWTH-21st formula: Log(EFW) = 5.084820 – 54.06633 × (AC/100)3 – 95.80076 × (AC/100)3 × log(AC/100) + 3.136370 × (HC/100), and plotted across gestation.
AC, abdominal length; EFW, estimated fetal weight; FL, fetal length; HC, head circumference; INTERGROWTH, INTERGROWTH-21st; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; WHO Fetal, World Health Organization Multicentre Growth Reference Study.
Grantz. Fetal growth charts. Am J Obstet Gynecol 2018.
that is commonly used in clinical practice in the United States (Figure 7 and Table 5). Interestingly, the NICHD white 50th percentile EFW was higher than the Hadlock reference, which also was in white women, but the other 3 NICHD racial/ethnic groups had EFW 50th percentiles lower than Hadlock. The population for Hadlock was limited to predominantly middle-class, white women without a history of maternal diseases associated with abnormal fetal growth and no congenital anomalies, so perhaps some of the differences between the white women could be explained by the NICHD cohort having even more restrictions and a healthier cohort, thereby including fetuses growing under more optimal conditions. Alternatively, the NICHD cohort included overweight women (body mass index [BMI] 25.0–29.9 kg/m2) and only nonsmokers, and while body mass index was not reported by Hadlock et al,
Centers for Disease Control and Prevention. Trends in current cigarette smoking among high school students and adults, United States, 1965–2014, Atlanta (GA): Centers for Disease Control and Prevention; 2016.
Nonetheless, perhaps the most important finding is that if the Hadlock 10th percentile is used to identify SGA in clinical practice, a larger percentage of fetuses from black, Hispanic, and Asian women would be labeled as SGA, while fewer fetuses from white women would be, compared with the NICHD standard.
Figure 7Estimated fetal weight comparison between NICHD and Hadlock 1991
These findings are similar to the NICHD analysis, which found the percentage of fetuses classified as being below the fifth percentile for EFW when using the white standard and was substantially higher for black, Hispanic, and Asian fetuses, except for black fetuses less than 18 weeks’ gestation. For example, at 35 weeks’ gestation, 15%, 12%, and 14% of black, Hispanic, and Asian fetuses, respectively, would have been classified as below the fifth percentile based on the white standard.
Therefore, an additional 10% (15% minus 5%), 7% (12% minus 5%) and 9% (14% minus 5%) of black, Hispanic, and Asian fetuses, respectively, would be classified as extreme. Findings were also similar when a pooled standard was used. These findings are also similar to the potential for misclassification using the pooled standard in INTERGROWTH as previously demonstrated.
Implications
Despite having extensive inclusion and exclusion criteria aimed at enrolling healthy women with uncomplicated pregnancies, allowing for optimal fetal growth, none of the 3 studies observed consistent standards for population subgroups. INTERGROWTH observed country-of-origin level differences in maternal height and weight as did WHO Fetal, while the NICHD observed both racial/ethnic differences in maternal size and fetal growth.
Collectively, these data argue for racial/ethnic fetal growth standards. This argument is supported by the fact that size and body proportion differences are known to occur across different races/ethnicities and countries for children and adults.
Mean stature for adult populations varies, and the ratio of sitting height to height, as a measure of body proportion, has also been shown to differ across 4 geographic areas including Australia/New Zealand/Papua New Guinea, Africa, Europe, and Asia.
Differences in body composition have also been found among different Asian ethnic groups, in which for the same BMI, Asian Indians had the highest percentage body fat compared with Malaysians and then Chinese, all of which were higher than whites.
This implies that the differences in fetal growth between the 2 international studies, INTERGROWTH and WHO Fetal, may be, to a large extent, an artifact of the international case mix (ie, maternal characteristics in the countries selected for the sample) and that, again, racial/ethnic-specific standards may improve the precision of fetal growth assessment.
In contrast to the assumption that all fetuses grow the same, INTERGROWTH found in their study of neonatal anthropometry for healthy, low-risk, term deliveries, which included 4321 neonates from the fetal growth cohort plus 20,486 newborns from a cross-sectional cohort, wide variation in birthweight and HC among the countries, consistent with differences in maternal size.
International standards for newborn weight, length, and head circumference by gestational age and sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project.
For example, birthweight ranged (mean [SD]) from 2.9 (0.4) kg in India to 3.5 (0.5) kg in the United Kingdom and HC from 33.1 (1.1) cm to 34.5 (1.3) cm in India and the United Kingdom, respectively.
These differences in body size and proportion have been hypothesized to be explained by both environmental and genetic factors.
Using twin studies, heritability of birth size has been estimated to be up to 40%, with an intergenerational study finding that fetal genetic factors explained 31% of normal variation in birthweight and birth length, and maternal genetic factors explained 22% and 19% of normal variation in these measures, respectively.
Genetic and environmental influences on birth weight, birth length, head circumference, and gestational age by use of population-based parent-offspring data.
Genetic and environmental influences on birth weight, birth length, head circumference, and gestational age by use of population-based parent-offspring data.
It has been hypothesized that the interaction of environmental influences and genetic factors on fetal growth display developmental plasticity that explain phenotypic differences in fetal growth and birthweight, and under this paradigm, a universal fetal growth standard is elusive.
In light of the many complexities underlying racial/ethnic definitions and the fact that it likely is not feasible to repeat these studies in individual populations across the world, it is important to understand how these study findings are applicable to individual populations. Ideally, a comparison of diagnostic accuracy, or misclassification rates, of SGA and LGA in relation to morbidity and mortality using different criteria is necessary to make recommendations and remains an important data gap. Identification for the appropriate percentiles is needed in local populations, depending on which standard is used.
Conclusion
Three recently completed longitudinal observational cohort studies, NICHD, INTERGROWTH and WHO Fetal,
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
have developed intrauterine fetal growth charts. The percentile cutpoints for SGA and LGA varied among the studies. For example, starting at 28 weeks of gestation and onward, the third percentile for INTERGROWTH was below all individual races/ethnicities in the NICHD study and the pooled sample in WHO Fetal study.
When applying these standards to pregnant women under clinical management, it is important to be aware that different percentages of SGA and LGA will be identified and that the percentiles are not interpreted in the same manner as a reference. It might be helpful to use all tools on the belt, in which a simple application could be created to calculate the percentiles (or SD scores) for comparison among the 3 charts.
Also, the assessment of fetal growth with a 1-time measurement (ie, EFW below the 10th percentile at a given gestational age) remains standard clinical practice, despite recognition that a single measurement can indicate only size. At least 2 measurements separated in time are needed to estimate a trajectory, and perhaps one of the greater contributions of these prospective studies will be the ability to estimate fetal growth velocity. Ultimately, it is knowledge about fetal growth in addition to other factors (signs of placental dysfunction or maternal complications) and clinical judgment that should trigger intervention.
The authors alone are responsible for the views expressed in this manuscript, which does not necessarily represent the decisions or the stated policy of the NICHD. The authors acknowledge Alaina Bever and Nicole Gerlanc for their assistance with the tables and figures. The NICHD Fetal Growth standard charts
Correction: The World Health Organization fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight.
The likeness of fetal growth and newborn size across non-isolated populations in the INTERGROWTH-21st Project: the Fetal Growth Longitudinal Study and Newborn Cross-Sectional Study.
Centers for Disease Control and Prevention. Trends in current cigarette smoking among high school students and adults, United States, 1965–2014, Atlanta (GA): Centers for Disease Control and Prevention; 2016.
International standards for newborn weight, length, and head circumference by gestational age and sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project.
Genetic and environmental influences on birth weight, birth length, head circumference, and gestational age by use of population-based parent-offspring data.
This study was supported by the intramural program at the Eunice Kennedy Shriver National Institute of Child Health and Human Development and included American Recovery and Reinvestment Act of 2009 funding (contract numbers HHSN275200800013C, HHSN275200800002I, HHSN27500006, HHSN275200800003IC, HHSN275200800014C, HHSN275200800012C, HHSN275200800028C, and HHSN275201000009C).
32441-9&id=gr1.jpg){kind=link}
32441-9&id=gr2.jpg){kind=link}
32441-9&id=gr3.jpg){kind=link}
32441-9&id=gr4.jpg){kind=link}
32441-9&id=gr5.jpg){kind=link}
32441-9&id=gr6.jpg){kind=link}
32441-9&id=gr7.jpg){kind=link}