ABSTRACT
Background
Smallness for gestational age and reduced head circumference at birth are consistently associated with later neurodevelopmental disorders (NDD), but it is unclear whether similar associations are present at foetal biometry in mid‐gestation.
Objectives
To investigate associations between second‐trimester foetal biometric measurements and later NDD.
Methods
We conducted a population‐based cohort study using data from routine second‐trimester ultrasound scans (gestational weeks 18–21) in Denmark. We included all singleton euploid live births from 2008 to 2015. Foetal biometrics included head circumference, biparietal diameter (BPD), abdominal circumference (AC), femur length, estimated foetal weight (EFW), and the head‐abdomen ratio. NDD comprised cerebral palsy (CP), epilepsy, intellectual disability, and autism spectrum disorders (ASD). Children were followed from birth until diagnosis, death, emigration, or 31 December 2022. Associations were examined using multivariable Cox proportional hazards models, and cumulative incidence through age 10 years was estimated accounting for death as a competing risk. Adjusted mean differences in biometric Z‐scores were examined using multivariable linear regression.
Results
Among 337,028 children, 13,414 (4.0%; 69.7% males) were diagnosed with at least one NDD. Larger BPD was associated with higher hazards of epilepsy, intellectual disability, and ASD. Larger AC and EFW were associated with lower hazard of intellectual disabilities. Individuals with epilepsy, ASD, and intellectual disabilities had slightly larger mean BPD Z‐score at their second–trimester scans, compared to children without NDD. Furthermore, smaller HC was associated with lower hazards and cumulative incidence of epilepsy, whereas shorter FL was associated with higher hazards of CP and epilepsy.
Conclusions
Mid‐gestation head size did not follow the expected risk pattern. These findings propose a rethinking of the relationship between restricted brain growth and neurodevelopmental impairment, suggesting that atypical early brain development may be reflected in accelerated rather than reduced growth, or that restricted growth‐related causes of NDD act later in pregnancy.
Keywords: foetal biometry, foetal growth, foetal weight, neurodevelopmental disorders, prenatal ultrasound
1. Background
The human brain is formed in the first trimester and thereafter undergoes complex development, including rapid neuronal proliferation, migration, and synaptogenesis [1, 2, 3]. These processes are strongly influenced by genetic and epigenetic factors, which interact with placental function and intrauterine exposures to shape foetal growth trajectories and long‐term neurodevelopment [4]. Impaired foetal growth may be present as early as the first trimester [5], with early foetal growth restriction (FGR) being linked to reduced cerebral volume, decreased structural complexity [6, 7], and increased risk of neurodevelopmental disorders (NDD) [8, 9], suggesting permanent alterations in brain structure and function [10, 11, 12].
NDD often originate in utero from a spectrum of causal factors, with external harmful influence, preterm birth, and FGR being central pathways [13]. However, the timing and mechanisms linking impaired foetal growth to harmful brain influence remain unclear. While FGR may directly impair cerebral development, it is also plausible that early prenatal brain injury triggers suboptimal foetal growth or that impaired foetal growth increases foetal brain vulnerability. Moreover, severe FGR can prompt cerebral blood flow redistribution to prioritise the brain above the rest of the body [10], causing asymmetric growth and increased risk of mortality and neurological morbidity [8, 14, 15]. Delayed foetal growth and a small head size have been associated with adverse motor and cognitive functioning in infancy [10, 11], yet most evidence relies on growth measures from late pregnancy or size at birth. While early developmental assessments in infancy and childhood provide some insight into developmental capacities, clinical diagnoses often emerge later in childhood. Therefore, we aimed to investigate associations between mid‐pregnancy foetal brain and body size and later NDD at the population level.
2. Methods
2.1. Study Design
This cohort study included all singleton live births from 2008 to 2015 with complete second‐trimester assessment between gestational age (GA) 18 + 0 (weeks + days) to 21 + 6. GA was based on crown‐rump length measured at the first‐trimester assessment [16]. We excluded children with major congenital malformations or chromosomal anomalies according to the EUROCAT guide [17].
In Denmark, all pregnant women are offered a nationwide prenatal screening programme that includes free first‐ and second‐trimester ultrasound assessments, with a participation rate exceeding 95% [18]. The second‐trimester ultrasound assessment is performed between gestational weeks 18–21 and includes screening for congenital malformations and measurement of foetal biometrics. All assessments are conducted by sonographers or specialists in foetal medicine, certified in accordance with the Fetal Medicine Foundation (FMF) guidelines [19]. Data are recorded in the Danish Fetal Medicine Database [20]. Residents in Denmark are assigned a unique civil registration number at birth or immigration, which we use to link individual information. Data were added from the Danish Medical Birth Register and the Danish Population Education Register, and information on death and emigration was sourced from the Danish Migration and Causes of Death registries.
2.2. Exposure
We included the standard second‐trimester foetal biometrics of head circumference (HC), biparietal diameter (BPD), abdominal circumference (AC), and femoral length (FL), all measured in millimetres (mm). Estimated foetal weight (EFW) was calculated using the Hadlock formula [21], including HC, AC, and FL, while the HC/AC ratio was estimated and used to assess foetal asymmetry [22]. All biometrics were continuous and assumed to be normally distributed after visual inspection. The biometrics were standardised to Z‐scores to account for variation in GA at measurement, using Dutch population norms [23] and FMF reference data [24], which are the standard references in Denmark. Measurements above or below 5 standard deviations (SD) were considered and set to missing, which did not materially affect percentile‐based categorisation.
2.3. Outcomes
We included the outcomes of cerebral palsy (CP), epilepsy, intellectual disability, and autism spectrum disorders (ASD). CP diagnoses were obtained from the Danish Cerebral Palsy Follow‐up Program (CPOP), which validates diagnoses through yearly follow‐ups, finalised around ages 4–5 [25, 26]. Epilepsy, intellectual disability, and ASD diagnoses were identified from the Danish National Patient Register, which includes diagnoses for all individuals discharged or seen in ambulatory clinics at Danish hospitals based on the International Classification of Diseases, 10th revision (ICD‐10) [27]. Epilepsy was defined as at least two separate notations on two different dates (ICD‐10 diagnosis codes G40–G41*). In Denmark, epilepsy is defined in accordance with the International League Against Epilepsy [28]. Intellectual disability (ICD‐10 diagnosis codes F70–F79*, R620) and ASD (ICD‐10 diagnosis codes F84* (not including F842 (Rett syndrome)), F88–F89*) were defined by at least one notation.
CP diagnoses in the Danish CPOP have high validity, with around 91% sensitivity and 94% positive predictive value [26]. Whereas for epilepsy, diagnoses in the Danish National Patient Registry have a positive predictive value of around 81% [29], but to improve this, we used two registry notations to define epilepsy. No studies have validated the diagnoses of ASD and intellectual disability, but these diagnoses do not have similar, clear‐cut diagnostic criteria. However, the quality of diagnoses in the Danish National Patient Registry is considered good, with a high sensitivity due to the widespread registration [30].
2.4. Covariates
Birthweight (grams, continuous); relative birthweight categorised as large (LGA, > 97th centile), appropriate (AGA), or small for GA (SGA < 3rd centile) [24]; GA (days, continuous); preterm birth (< 37 weeks); sex; maternal age (years, continuous); maternal BMI (continuous); maternal educational level at year of birth categorised according to the International Standard Classification of Education as high, middle, or low [31]; hypertension in pregnancy and maternal smoking, defined as any self‐reported smoking during pregnancy. Confounding covariates to adjust for in the multivariable analyses were selected based on their presumed causal pathways, depicted in a directed acyclic graph (DAG; Figure 1) [32]. The DAG was based on previous research and the authors' expert judgement, and minimal sufficient adjustment sets for the total effects were created using the free DAGitty software [33, 34].
FIGURE 1.

Directed acyclic graph of the causal relationship between foetal size in the second trimester of pregnancy and neurodevelopmental disorder.
2.5. Statistical Analysis
Individuals were followed from birth until the first occurrence of an outcome of interest, death, emigration, or the end of the follow‐up period (31 December 2022), whichever came first. The maximum follow‐up time was thus almost 15 years. Cox proportional hazards models were used to estimate cause‐specific hazard ratios (HR) and 95% confidence intervals (CI) for each NDD according to second‐trimester biometrics, modelled both per 1 standard deviation (SD) increase in Z‐score and as binary indicators of measurements below the 10th centile or above the 90th centile. Adjusted models included sex, maternal body mass index (BMI), maternal educational level, and smoking during pregnancy. To account for death as a competing risk, cumulative incidence through age 10 years was estimated using the Aalen‐Johansen estimator. As a secondary analysis, adjusted mean Z‐scores were compared between individuals with and without NDD; these analyses were restricted to individuals who survived and remained in Denmark for at least 7 years to ensure comparable ascertainment. Descriptive differences of the mean Z‐scores, stratified by sex, were also presented in Figure 2.
FIGURE 2.

Mean Z‐scores for foetal biometrics measured in the second trimester (GA 18–21) in children with neurodevelopmental disorders (NND) compared to children without and stratified by sex. Z‐scores are derived from Dutch population norms and the Fetal Medicine Foundation reference data.
Data management and statistical analyses were performed using R (version 4.4.0) [35].
2.6. Missing Data
No variables had more than 3% missing data (Table 1), and the proportions were on par between cases and controls. The four variables in the multivariable models had an overall missingness of 4.9%.
TABLE 1.
Descriptive maternal and child characteristics in the cohort with and without neurodevelopmental disorders (cerebral palsy, epilepsy, intellectual disabilities, and autism spectrum disorder).
| Neurodevelopmental disorder (n = 13,414) | No neurodevelopmental disorder (n = 323,614) | Total cohort (n = 337,028) | |
|---|---|---|---|
| Maternal variables | |||
| Maternal age (years), mean (±SD) | 29.4 (5.2) | 29.9 (4.9) | 29.9 (4.9) |
| BMI (kg/m2), mean (±SD) | 24.8 (5.7) | 23.8 (4.9) | 23.9 (5.0) |
| < 18.5 | 785 (6%) | 20,465 (7%) | 21,250 (6%) |
| 18.5–24.9 | 6804 (52%) | 187,053 (60%) | 193,857 (59%) |
| 25.0–29.9 | 3125 (24%) | 67,862 (22%) | 70,987 (22%) |
| ≥ 29.9 | 2298 (18%) | 38,578 (12%) | 40,876 (13%) |
| Missing | 402 (3%) | 9656 (3%) | 10,058 (3%) |
| Parity | |||
| Nulliparous | 6542 (49%) | 150,918 (47%) | 157,460 (47%) |
| Multiparous | 6711 (51%) | 170,348 (53%) | 177,059 (53%) |
| Missing | 161 (1%) | 2348 (1%) | 2509 (1%) |
| Smoking during pregnancy | |||
| Yes | 1899 (14%) | 30,549 (10%) | 32,448 (10%) |
| No | 11,228 (86%) | 286,558 (90%) | 297,786 (90%) |
| Missing | 287 (2%) | 6507 (2%) | 6794 (2%) |
| Hypertension during pregnancy | |||
| Yes | 912 (7%) | 15,992 (5%) | 16,904 (5%) |
| No | 12,502 (93%) | 307,622 (95%) | 320,124 (95%) |
| Diabetes in pregnancy | |||
| Yes | 688 (5%) | 11,589 (4%) | 12,277 (4%) |
| No | 2726 (95%) | 312,025 (96%) | 314,751 (96%) |
| Maternal educational level | |||
| High | 4923 (37%) | 144,272 (45%) | 149,195 (45%) |
| Middle | 2852 (21%) | 48,527 (15%) | 51,379 (15%) |
| Low | 5527 (42%) | 126,534 (40%) | 132,061 (40%) |
| Missing | 112 (1%) | 4281 (1%) | 4393 (1%) |
| Child variables | |||
| Sex | |||
| Female | 4063 (30%) | 161,208 (50%) | 165,271 (49%) |
| Male | 9351 (70%) | 162,406 (50%) | 171,757 (51%) |
| Gestational age at birth (weeks), median (25th–75th centiles) | 40 (39, 41) | 40 (39, 41) | 40 (39, 41) |
| > 36 + 6 (Term) | 12,443 (93%) | 310,192 (96%) | 322,635 (96%) |
| < 37 + 0 (preterm) | 971 (7%) | 13,422 (4%) | 14,393 (4%) |
| Birthweight (g), mean (±SD) | 3481 (610) | 3523 (531) | 3521 (534) |
| AGA | 11,951 (91%) | 295,293 (93%) | 307,244 (93%) |
| SGA (< 3rd centile) | 485 (4%) | 8872 (3%) | 9357 (3%) |
| LGA (> 97nd centile) | 735 (6%) | 14,297 (4%) | 15,032 (5%) |
| Missing | 243 (2%) | 5152 (2%) | 5395 (2%) |
| Gestational age at 2. trim scan (weeks), mean (SD) | 20 (0.6) | 20 (0.6) | 20 (0.6) |
Note: Hypertension in pregnancy included chronic hypertension, gestational hypertension, preeclampsia, and eclampsia.
Abbreviations: AGA, appropriate for gestational age; BMI, body mass index; g, grams; kg, kilogram; LGA, large for gestational age; m, meter; n, number; SD, standard deviation; SGA, small for gestational age.
2.7. Sensitivity Analyses
We undertook an exploratory sensitivity analysis to exclude those with selected perinatal sentinel events at birth that are potentially linked to NDD, as these conditions may reflect intrapartum complications that could independently influence later neurodevelopment, thereby obscuring the true association between foetal biometrics and NDD. Therefore, we performed similar analyses, excluding all with an event at birth of shoulder dystocia (DO660), uterine rupture (DO71*), umbilical cord prolapse (DO690), or placental abruption (DO45*).
2.8. Ethics Approval
The study was approved by the Danish Data Protection Agency (P‐2023‐209). In Denmark, ethical approval is not required for register‐based studies.
3. Results
We identified 355,785 liveborn singleton children with complete second‐trimester scans from 2008 to 2015. After excluding 18,757 (5.3%) with major congenital malformations or chromosomal anomalies, a total of 337,028 (94.7%) children were included in the final cohort (Figure 3). Among these, 13,414 (4.0%) developed at least one clinically important NDD during the study period, with 2250 (0.7%) diagnosed with epilepsy, 692 (0.2%) with CP, 2668 (0.8%) with intellectual disability, and 9280 (2.8%) with ASD. Some children developed multiple conditions; with the combination of ASD and intellectual disability being the most common in 849 (0.3%) children, followed by epilepsy and ASD (n = 127, 0.04%), epilepsy and intellectual disability (n = 87, 0.03%), CP and intellectual disability (n = 72, 0.02%), CP and epilepsy (n = 69, 0.02%). Very few children had more than two diagnoses.
FIGURE 3.

Flow chart of the study cohort.
Descriptive characteristics of children with and without NDD are presented in Table 1. Mothers of children with NDD were more likely to have lower educational levels, be nulliparous, smoke during pregnancy, and have a high BMI. Moreover, children with NDD were much more frequently male and born preterm.
The overall mean second‐trimester foetal biometric measurements were 171 mm (SD 8.7) for HC, 47.6 mm (SD 2.8) for BPD, 148.1 mm (SD 9.6) for AC, 31.2 mm (SD 2.2) for FL, and 318.0 g (SD 42.9) for EFW. As expected, female foetuses had smaller measurements of HC, BPD, AC, AC/HC ratio, and EFW (Figure 2).
3.1. Association With NDD and Foetal Biometrics Measured in the Second Trimester (GA 18–21)
3.1.1. Epilepsy
Foetuses that later developed epilepsy had slightly larger mean HC Z‐scores and mean BPD Z‐scores at their second‐trimester scans, compared to those without NDD (Table S1). Therefore, the hazard of epilepsy increased per one SD increase in HC measurements, and HC measurements, along with HC/AC‐ratio and BPD, in the lower 10th centile were associated with reduced hazards and cumulative risks of epilepsy (Table 2). Conversely, FL measurements in the lower 10th centiles were linked with increased hazard and cumulative risk of epilepsy.
TABLE 2.
Differences in hazard and cumulative risk for neurodevelopmental disorders (NND) in individuals with abnormal second‐trimester (GA 18–21) foetal biometric measurements.
| Biometric | HR per 1 SD increase in Z‐score | HR of NND in foetuses with measurements < 10th centile | HR of NND in foetuses with measurements > 90th centile | Cumulative incidence (%) of NND at age 10 in foetuses with measurements ≥ 10th versus < 10th centile | Cumulative incidence (%) of NND at age 10 in foetuses with measurements ≤ 90th versus > 90th centile |
|---|---|---|---|---|---|
| Epilepsy (2250 events and 715 death) | |||||
| HC | 1.07 (1.02, 1.12) | 0.74 (0.61, 0.89) | 1.05 (0.91, 1.20) | 0.68 (0.65, 0.71) versus 0.51 (0.41, 0.60) | 0.66 (0.63, 0.69) versus 0.71 (0.61, 0.80) |
| BPD | 1.03 (0.99, 1.08) | 0.89 (0.78, 1.01) | 1.08 (0.92, 1.26) | 0.67 (0.64, 0.70) versus 0.62 (0.54, 0.69) | 0.66 (0.63, 0.69) versus 0.72 (0.61, 0.83) |
| AC | 1.01 (0.96, 1.06) | 0.92 (0.77, 1.09) | 0.88 (0.75, 1.03) | 0.67 (0.64, 0.70) versus 0.60 (0.50, 0.71) | 0.67 (0.64, 0.70) versus 0.59 (0.50, 0.68) |
| HC/AC‐ratio | 1.03 (0.98, 1.07) | 0.82 (0.70, 0.95) | 0.96 (0.83, 1.11) | 0.68 (0.65, 0.71) versus 0.56 (0.47, 0.64) | 0.67 (0.64, 0.70) versus 0.63 (0.54, 0.72) |
| FL | 0.99 (0.94, 1.04) | 1.34 (1.14, 1.58) | 1.12 (0.95, 1.33) | 0.65 (0.62, 0.68) versus 0.90 (0.76, 1.04) | 0.66 (0.63, 0.69) versus 0.74 (0.62, 0.87) |
| EFW | 1.01 (0.97, 1.06) | 1.02 (0.88, 1.18) | 1.01 (0.89, 1.15) | 0.66 (0.63, 0.69) versus 0.66 (0.57, 0.76) | 0.66 (0.63, 0.69) versus 0.68 (0.60, 0.77) |
| Cerebral palsy (692 events and 726 deaths) | |||||
| HC | 1.00 (0.92, 1.09) | 1.09 (0.81, 1.47) | 1.03 (0.81, 1.33) | 0.21 (0.19, 0.22) versus 0.20 (0.14, 0.25) | 0.20 (0.18, 0.22) versus 0.23 (0.18, 0.29) |
| BPD | 1.04 (0.96, 1.12) | 0.94 (0.74, 1.18) | 1.29 (1.00, 1.67) | 0.21 (0.19, 0.23) versus 0.17 (0.14, 0.21) | 0.20 (0.18, 0.21) versus 0.28 (0.21, 0.34) |
| AC | 1.01 (0.92, 1.10) | 0.94 (0.68, 1.29) | 1.11 (0.86, 1.43) | 0.21 (0.19, 0.22) versus 0.18 (0.12, 0.23) | 0.20 (0.18, 0.22) versus 0.24 (0.18, 0.30) |
| HC/AC‐ratio | 1.00 (0.93, 1.08) | 1.17 (0.92, 1.48) | 0.94 (0.72, 1.22) | 0.20 (0.18, 0.22) versus 0.23 (0.18, 0.29) | 0.21 (0.19, 0.22) versus 0.19 (0.14, 0.24) |
| FL | 0.95 (0.87, 1.05) | 1.39 (1.04, 1.85) | 0.87 (0.61, 1.23) | 0.20 (0.18, 0.22) versus 0.27 (0.19, 0.34) | 0.21 (0.19, 0.22) versus 0.18 (0.12, 0.25) |
| EFW | 0.98 (0.91, 1.05) | 1.00 (0.76, 1.31) | 0.95 (0.75, 1.20) | 0.21 (0.19, 0.22) versus 0.19 (0.14, 0.23) | 0.20 (0.19, 0.22) versus 0.21 (0.17, 0.26) |
| Intellectual disabilities (2668 events and 723 deaths) | |||||
| HC | 1.01 (0.97, 1.05) | 1.08 (0.92, 1.27) | 1.05 (0.93, 1.18) | 0.78 (0.75, 0.81) versus 0.69 (0.58, 0.79) | 0.75 (0.72, 0.78) versus 0.97 (0.86, 1.08) |
| BPD | 1.04 (1.00, 1.08) | 0.98 (0.86, 1.10) | 1.17 (1.03, 1.33) | 0.79 (0.76, 0.82) versus 0.67 (0.60, 0.75) | 0.75 (0.72, 0.78) versus 1.08 (0.95, 1.21) |
| AC | 0.95 (0.91, 0.99) | 1.04 (0.89, 1.23) | 0.91 (0.79, 1.04) | 0.77 (0.74, 0.81) versus 0.75 (0.63, 0.86) | 0.77 (0.74, 0.80) versus 0.81 (0.70, 0.92) |
| HC/AC‐ratio | 1.05 (1.01, 1.10) | 0.86 (0.75, 0.99) | 1.00 (0.88, 1.14) | 0.78 (0.75, 0.82) versus 0.68 (0.59, 0.77) | 0.77 (0.74, 0.80) versus 0.81 (0.71, 0.91) |
| FL | 0.96 (0.92, 1.01) | 1.22 (1.04, 1.42) | 0.93 (0.78, 1.09) | 0.76 (0.73, 0.79) versus 0.95 (0.81, 1.10) | 0.77 (0.74, 0.80) versus 0.82 (0.69, 0.95) |
| EFW | 0.95 (0.92, 0.99) | 1.12 (0.98, 1.29) | 0.90 (0.80, 1.01) | 0.77 (0.74, 0.80) versus 0.79 (0.68, 0.89) | 0.76 (0.73, 0.80) versus 0.83 (0.74, 0.92) |
| Autism (9280 events and 744 deaths) | |||||
| HC | 0.99 (0.97, 1.01) | 1.03 (0.94, 1.13) | 1.00 (0.94, 1.06) | 2.44 (2.38, 2.50) versus 1.74 (1.56, 1.92) | 2.31 (2.25, 2.37) versus 3.10 (2.90, 3.31) |
| BPD | 1.02 (1.00, 1.04) | 1.01 (0.94, 1.08) | 1.05 (0.98, 1.12) | 2.45 (2.39, 2.52) versus 1.99 (1.85, 2.13) | 2.32 (2.26, 2.38) versus 3.22 (2.98, 3.45) |
| AC | 1.01 (0.99, 1.04) | 0.92 (0.84, 1.01) | 0.93 (0.87, 1.00) | 2.43 (2.37, 2.49) versus 1.79 (1.61, 1.98) | 2.36 (2.30, 2.42) versus 2.65 (2.45, 2.85) |
| HC/AC‐ratio | 0.99 (0.97, 1.01) | 1.00 (0.94, 1.08) | 0.92 (0.85, 0.98) | 2.40 (2.34, 2.46) versus 2.31 (2.14, 2.49) | 2.40 (2.34, 2.46) versus 2.26 (2.09, 2.43) |
| FL | 1.00 (0.98, 1.03) | 1.02 (0.93, 1.12) | 1.05 (0.97, 1.15) | 2.39 (2.33, 2.44) versus 2.40 (2.16, 2.64) | 2.37 (2.31, 2.43) versus 2.68 (2.44, 2.92) |
| EFW | 1.00 (0.98, 1.02) | 0.98 (0.91, 1.06) | 1.02 (0.96, 1.08) | 2.44 (2.38, 2.50) versus 1.89 (1.72, 2.05) | 2.32 (2.26, 2.37) versus 2.88 (2.71, 3.05) |
Note: All analyses are adjusted for sex, maternal body mass index, maternal educational level, and smoking in pregnancy.
Abbreviations: AC, abdominal circumference; BPD, biparietal diameter; EFW, estimated foetal weight; FL, femur length; GA, gestational age; HC, head circumference; HR, hazard ratio; NND, neurodevelopmental disorders; SD, standard deviation.
3.1.2. Cerebral Palsy
Female foetuses who later developed CP had higher mean Z‐scores for BPD and AC (Figure 2); however, these differences were not observed in the multivariable linear models. Notably, a BPD measurement in the upper 90th centile was associated with higher hazard and cumulative risk of CP (Table 2). Furthermore, FL measurements in the lower 10th centile were also associated with an increased hazard of CP.
3.1.3. Intellectual Disabilities
Foetuses who later developed intellectual disabilities had somewhat larger BPD mean Z‐scores at their second‐trimester scans (Table S1). Thus, the hazard of intellectual disabilities increased per 1 SD increase in BPD measurements, and BPD measurements in the lower 10th centile had reduced cumulative risks at age 10, whereas BPD measurements in the upper 90th centile were associated with higher hazards and cumulative risks (Table 2). In contrast, the hazard of intellectual disabilities decreased per 1 SD increase in AC and EFW.
3.1.4. Autism Spectrum and Related Disorders
ASD later in life was also associated with slightly larger mean Z‐scores of BPD at the second‐trimester scan (Table S1), and the hazard of ASD increased per 1 SD increase in BPD measurements. Lower cumulative incidence at age 10 was observed for HC, BPD, AC, and EFW below the 10th centile, whereas cumulative incidence was generally higher for measurements above the 90th centile. In the adjusted analyses, BPD was positively associated with ASD, whereas AC and HC/AC ratio above the 90th centile showed inverse associations.
3.2. Sensitivity Analyses
The results remained largely unchanged after excluding 5378 (1.6%) children with at least one sentinel perinatal event (Tables S2 and S3).
4. Comment
4.1. Principal Findings
In this population‐based cohort study, we found modest associations between foetal head and body size in mid‐gestation, measured during routine second‐trimester ultrasound scans, and later NDD. The most consistent pattern was that a larger head size was associated with higher hazards of all four NDDs, whereas larger AC and EFW were associated with lower hazards of intellectual disability. For epilepsy, a smaller head size, as judged by HC and BPD in the lower 10th centile, was associated with reduced hazard and cumulative incidence by age 10, while shorter femur length was associated with higher hazard.
4.2. Strengths of the Study
This is the largest study to date investigating associations between foetal ultrasound biometrics and NDD. Thus, a key strength of the study is the substantial sample size, included from a nationwide database with a very high participation rate [18]. Second‐trimester biometrics were measured across an almost entire pregnant population by certified collectors, minimising selection and information bias. Furthermore, children were followed for an extended period, with outcomes obtained from comprehensive nationwide registries. For CP, all diagnoses were validated by specialists at 4–5 years of age, reducing the risk of misclassification [26]. Moreover, given the rarity of NDD, the large cohort was critical to examine such associations robustly.
4.3. Limitations of the Data
A limitation is the lack of specific diagnostic criteria for epilepsy, intellectual disabilities, and autism, as these were based on registry data without access to detailed clinical records, increasing the risk of misclassification bias. Foetal biometrics obtained by ultrasound are subject to inter‐ and intraobserver variability [36]. However, foetal ultrasound examinations in Denmark follow standardised national protocols and are performed by certified sonographers, limiting measurement heterogeneity.
4.4. Interpretation
Other studies have explored associations between foetal biometrics at different stages of pregnancy and adverse neurodevelopment, primarily using neurodevelopmental tests administered in infancy or early childhood as measures of neurological performance rather than focusing on clinical conditions. In addition, the existing literature shows mixed results. Two large studies, including approximately 2200 and 1500 pregnancies, found no associations between foetal biometrics at varying gestational ages and developmental delays at 14 or 24 months of age [37, 38]. Similarly, a study of 211 pregnancies observed no link between HC at 18 weeks of gestation and IQ score at age 9 years [39], whereas a small case–control study found no differences in HC and FL measured at week 18 in foetuses who later developed language impairments [40]. Our research group also found no differences in foetal BPD and crown‐rump length, measured at gestational weeks 11–14, in nearly 400 foetuses who later developed CP [41]. In contrast, a large Dutch study of 3000 low‐risk pregnancies reported decreasing risks of suboptimal neurodevelopmental status with each SD increase in AC, AC‐HC ratio, and EFW at 20 and 30 weeks [42]. The same group linked higher HC and EFW in mid‐pregnancy with better self‐help abilities at 6–18 months [43]. Furthermore, two studies with longer follow‐ups reported that small HC at 14 weeks was associated with adverse cognitive outcomes at 6–8 years [44], while a high EFW in the second and third trimesters was positively linked to better childhood development at ages 1 and 5 [45]. Lastly, a large study of 7000 pregnancies found that higher HC and EFW at 18 weeks correlated with better academic achievement [46].
The associations identified in our study cannot be used to predict NDD early [41]. However, our findings indicate that abnormal foetal head growth may influence neurodevelopment as early as the second trimester of pregnancy. Thus, these results contribute to the etiological understanding of the conditions as we may move closer to the period of origin. Notably, our results indicate that the established link between FGR and NDD is either absent or not detectable at the second‐trimester scan. While previous research identified a U‐shaped relationship between relative birthweight and the risk of CP [47], we did not observe a similar pattern for mid‐gestation EFW.
The timing of neurological impairment is important when understanding NDD. In the case of CP, the etiological belief has shifted from intrapartum causes to antenatal origins [48], a perspective further supported by our findings. The prior linkage between infant head size and negative neurodevelopment has concerned small head size [11, 49] and not, as in our case, large measurements. Small head size at birth is associated with reduced brain volume [50] and correlates with psychomotor retardation and cognitive delay [11, 49]. However, it is also established that brain growth accelerates markedly during the latter half of pregnancy and continues rapidly after birth [3]. A potential explanation for these discrepant findings is that foetal head size in mid‐gestation may reflect growth trajectories rather than absolute brain volume. Studies have linked abnormal early foetal BPD growth to adverse outcomes, with slow growth being associated with a risk of intrauterine growth restrictions [51, 52], and fast growth correlating with preterm birth [53, 54] and macrosomia [51]. Furthermore, early postnatal neuroimaging studies in children with ASD have demonstrated accelerated brain growth during infancy, which may not be apparent from measurements at birth alone [55, 56]. Thus, larger foetal head measurements in mid‐gestation may reflect altered timing of neurodevelopmental processes rather than sustained macrocephaly, potentially reconciling our findings with the established associations between small head size and NDD. Lastly, our somewhat unexpected findings can also reflect abnormal foetal programming during a vulnerable developmental stage; for example, large foetal head size could be an early sign of FGR, with prioritisation of cerebral circulation leading to larger heads and contributing to abnormal neurodevelopment in childhood. However, this hypothesis warrants further investigation in future studies.
Measurements of HC and BPD are correlated, as both reflect head size; however, a larger HC showed weaker associations with the outcomes. This discrepancy may arise because BPD is more reproducible, as its measurement is relatively straightforward compared with HC.
5. Conclusions
This nationwide study of more than 330,000 pregnancies found that foetuses who later developed epilepsy, ASD, or intellectual disabilities had larger BPD measurements at their second‐trimester ultrasound scans at gestational weeks 18–21. This is surprising, since small head size at birth and poor postnatal head growth are robustly associated with neurodevelopmental impairment. While the effect size is small and confounding cannot be excluded, our study suggests we should rethink early restricted brain growth as an unambiguous predictor of neurodevelopmental impairment.
Author Contributions
M.L.L. conceived the study and developed the study design. M.L.L. and O.B.P. obtained and curated the data. M.L.L. conducted the statistical analyses. M.L.L., G.G., L.K., C.H.H., and O.B.P. interpreted the results. M.L.L. drafted the manuscript. All authors contributed to the interpretation of findings, critically revised the manuscript for important intellectual content, and approved the final version for publication.
Funding
This work was supported by Elsass Fonden (23‐B01‐0499), Novo Nordisk Fonden (NNFSA170030576).
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Mean Z‐scores for foetal biometrics measured in the second trimester (GA 18–21) in children with neurodevelopmental disorders (NND) compared to children without.
Table S2: Sensitivity analyses on the cohort without obstetric sentinel events. Mean Z‐scores for foetal biometrics measured in the second trimester (GA 18–21) in children with neurodevelopmental disorders (NND) compared to children without.
Table S3: Sensitivity analyses on the cohort without obstetric sentinel events. Differences in hazard and cumulative risk for neurodevelopmental disorders (NND) in individuals with abnormal second‐trimester (GA 18–21) foetal biometric measurements.
Acknowledgements
This work was supported by the Elsass Foundation (23‐B01‐0499). Olav Bjørn Petersen holds a professorship funded by the Novo Nordisk Foundation (NNFSA170030576). The funding parties did not participate in the study.
Data Availability Statement
The data that support the findings of this study are available with permission from Statistics Denmark: www.dst.dk/en/TilSalg/data‐til‐forskning. These data are available under licence and subject to restrictions.
References
- 1. Dobbing J. and Sands J., “Quantitative Growth and Development of Human Brain,” Archives of Disease in Childhood 48 (1973): 757–767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Sailaja K. and Gopinath G., “Ultrastructure of Developing Substantia Nigra in Humans,” International Journal of Developmental Neuroscience 14 (1996): 761–770. [DOI] [PubMed] [Google Scholar]
- 3. Stiles J. and Jernigan T. L., “The Basics of Brain Development,” Neuropsychology Review 20 (2010): 327–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Durbagula S., Korlimarla A., Ravikumar G., Valiya Parambath S., Kaku S. M., and Visweswariah A. M., “Prenatal Epigenetic Factors Are Predisposing for Neurodevelopmental Disorders—Considering Placenta as a Model,” Birth Defects Research 114 (2022): 1324–1342. [DOI] [PubMed] [Google Scholar]
- 5. Smith G. C. S., “First Trimester Origins of Fetal Growth Impairment,” Seminars in Perinatology 28 (2004): 41–50. [DOI] [PubMed] [Google Scholar]
- 6. Tolsa C. B., Zimine S., Warfield S. K., et al., “Early Alteration of Structural and Functional Brain Development in Premature Infants Born With Intrauterine Growth Restriction,” Pediatric Research 56 (2004): 132–138. [DOI] [PubMed] [Google Scholar]
- 7. Samuelsen G. B., Pakkenberg B., Bogdanović N., et al., “Severe Cell Reduction in the Future Brain Cortex in Human Growth–Restricted Fetuses and Infants,” American Journal of Obstetrics and Gynecology 197 (2007): 56.e1–56.e7. [DOI] [PubMed] [Google Scholar]
- 8. Murray E., Fernandes M., Fazel M., Kennedy S. H., Villar J., and Stein A., “Differential Effect of Intrauterine Growth Restriction on Childhood Neurodevelopment: A Systematic Review,” BJOG: An International Journal of Obstetrics & Gynaecology 122 (2015): 1062–1072. [DOI] [PubMed] [Google Scholar]
- 9. Sacchi C., Marino C., Nosarti C., Vieno A., Visentin S., and Simonelli A., “Association of Intrauterine Growth Restriction and Small for Gestational Age Status With Childhood Cognitive Outcomes,” JAMA Pediatrics 174 (2020): 772–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Baschat A. A., “Neurodevelopment Following Fetal Growth Restriction and Its Relationship With Antepartum Parameters of Placental Dysfunction,” Ultrasound in Obstetrics & Gynecology 37 (2011): 501–514. [DOI] [PubMed] [Google Scholar]
- 11. Miller S. L., Huppi P. S., and Mallard C., “The Consequences of Fetal Growth Restriction on Brain Structure and Neurodevelopmental Outcome,” Journal of Physiology 594 (2016): 807–823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Korkalainen N., Ilvesmäki T., Parkkola R., Perhomaa M., and Mäkikallio K., “Brain Volumes and White Matter Microstructure in 8‐ to 10‐Year‐Old Children Born With Fetal Growth Restriction,” Pediatric Radiology 52 (2022): 2388–2400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Stanley F. J., Blair E., and Alberman E. D., Cerebral Palsies: Epidemiology and Causal Pathways (Cambridge Univesity Press, 2000). [Google Scholar]
- 14. Scherjon S., Briët J., Oosting H., and Kok J., “The Discrepancy Between Maturation of Visual‐Evoked Potentials and Cognitive Outcome at Five Years in Very Preterm Infants With and Without Hemodynamic Signs of Fetal Brain‐Sparing,” Pediatrics 105 (2000): 385–391. [DOI] [PubMed] [Google Scholar]
- 15. Hernandez‐Andrade E., Benavides Serralde J. A., and Cruz‐Martinez R., “Can Anomalies of Fetal Brain Circulation Be Useful in the Management of Growth Restricted Fetuses?,” Prenatal Diagnosis 32 (2012): 103–112. [DOI] [PubMed] [Google Scholar]
- 16. Robinson H. P. and Fleming J. E. E., “A Critical Evaluation of Sonar “Crown‐Rump Lenght” Measurements,” BJOG: An International Journal of Obstetrics and Gynaecology 82 (1975): 702–710. [DOI] [PubMed] [Google Scholar]
- 17. Broe A., Damkier P., Pottegård A., Hallas J., and Bliddal M., “Congenital Malformations in Denmark: Considerations for the Use of Danish Health Care Registries,” Clinical Epidemiology 12 (2020): 1371–1380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Ekelund C. K., Petersen O. B., Jørgensen F. S., et al., “The Danish Fetal Medicine Database: Establishment, Organization and Quality Assessment of the First Trimester Screening Program for Trisomy 21 in Denmark 2008–2012,” Acta Obstetricia et Gynecologica Scandinavica 94 (2015): 577–583. [DOI] [PubMed] [Google Scholar]
- 19. The Fetal Medicine Foundation . Last accessed November 2021, https://fetalmedicine.org/.
- 20. Ekelund C. K., Kopp T. I., Tabor A., and Petersen O. B., “The Danish Fetal Medicine Database,” Clinical Epidemiology 8 (2016): 479–483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Hadlock F. P., Harrist R. B., Sharman R. S., Deter R. L., and Park S. K., “Estimation of Fetal Weight With the Use of Head, Body, and Femur Measurements ‐ A Prospective Study,” American Journal of Obstetrics and Gynecology 151 (1985): 333–337. [DOI] [PubMed] [Google Scholar]
- 22. Campbell S. and Thoms A., “Ultrasound Measurements of the Fetal Head to Abdomen Circumference Ratio in the Assessment of Growth Retardation,” BJOG: An International Journal of Obstetrics & Gynaecology 84 (1977): 165–174. [DOI] [PubMed] [Google Scholar]
- 23. Verburg B. O., Steegers E. A. P., De Ridder M., et al., “New Charts for Ultrasound Dating of Pregnancy and Assessment of Fetal Growth: Longitudinal Data From a Population‐Based Cohort Study,” Ultrasound in Obstetrics & Gynecology 31 (2008): 388–396. [DOI] [PubMed] [Google Scholar]
- 24. Nicolaides K. H., Wright D., Syngelaki A., Wright A., and Akolekar R., “Fetal Medicine Foundation Fetal and Neonatal Population Weight Charts,” Ultrasound in Obstetrics & Gynecology 52 (2018): 44–51. [DOI] [PubMed] [Google Scholar]
- 25. Rasmussen H. M., Nordbye‐Nielsen K., Møller‐Madsen B., et al., “The Danish Cerebral Palsy Follow‐Up Program,” Clinical Epidemiology 8 (2016): 457–460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Larsen M. L., Hoei‐Hansen C. E., and Rackauskaite G., “The Diagnosis of Cerebral Palsy in Two Danish National Registries: A Validation Study,” Scandinavian Journal of Public Health 53 (2024): 1–7. [DOI] [PubMed] [Google Scholar]
- 27. Madsen M., Jørgensen J., Mellemkjær L., Olsen J. H., and Andersen T. F., “The Danish National Hospital Register : A Valuable Source of Data for Modern Health Sciences,” Danish Medical Bulletin 46 (1999): 263–268. [PubMed] [Google Scholar]
- 28. Fisher R. S., Acevedo C., Arzimanoglou A., et al., “ILAE Official Report: A Practical Clinical Definition of Epilepsy,” Epilepsia 55 (2014): 475–482. [DOI] [PubMed] [Google Scholar]
- 29. Christensen J., Vestergaard M., Olsen J., and Sidenius P., “Validation of Epilepsy Diagnoses in the Danish National Hospital Register,” Epilepsy Research 75 (2007): 162–170. [DOI] [PubMed] [Google Scholar]
- 30. Thygesen L. C., Daasnes C., Thaulow I., and Brønnum‐Hansen H., “Introduction to Danish (Nationwide) Registers on Health and Social Issues: Structure, Access, Legislation, and Archiving,” Scandinavian Journal of Public Health 39 (2011): 12–16. [DOI] [PubMed] [Google Scholar]
- 31. Schneider S. L., “The International Standard Classification of Education 2011,” Comparative Social Research 30 (2013): 365–379. [Google Scholar]
- 32. Williams T. C., Bach C. C., Matthiesen N. B., Henriksen T. B., and Gagliardi L., “Directed Acyclic Graphs: A Tool for Causal Studies in Paediatrics,” Pediatric Research 84 (2018): 487–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Larsen M. L., Rackauskaite G., Greisen G., et al., “Declining Prevalence of Cerebral Palsy in Children Born at Term in Denmark,” Developmental Medicine & Child Neurology 64 (2022): 715–722. [DOI] [PubMed] [Google Scholar]
- 34. Textor J., van der Zander B., Gilthorpe M. S., Liśkiewicz M., and Ellison G. T. H., “Robust Causal Inference Using Directed Acyclic Graphs: The R Package ‘Dagitty’,” International Journal of Epidemiology 45 (2016): 1887–1894. [DOI] [PubMed] [Google Scholar]
- 35. R Core Team , R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2020). [Google Scholar]
- 36. Sarris I., Ioannou C., Chamberlain P., et al., “Intra‐ and Interobserver Variability in Fetal Ultrasound Measurements,” Ultrasound in Obstetrics & Gynecology 39 (2012): 266–273. [DOI] [PubMed] [Google Scholar]
- 37. Álamo‐Junquera D., Sunyer J., Iñiguez C., et al., “Prenatal Head Growth and Child Neuropsychological Development at Age 14 Months,” American Journal of Obstetrics and Gynecology 212 (2015): 661.e1–661.e11. [DOI] [PubMed] [Google Scholar]
- 38. Karantha S. C., Upadhyay R. P., Jain A., et al., “Association of Fetal Ultrasound Anthropometric Parameters With Neurodevelopmental Outcomes at 24 Months of Age,” PLoS One 18 (2023): e0296215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Gale C. R., “Critical Periods of Brain Growth and Cognitive Function in Children,” Brain 127 (2004): 321–329. [DOI] [PubMed] [Google Scholar]
- 40. Whitehouse A. J. O., Zubrick S. R., Blair E., Newnham J. P., and Hickey M., “Fetal Head Circumference Growth in Children With Specific Language Impairment,” Archives of Disease in Childhood 97 (2012): 49–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Larsen M. L., Rackauskaite G., Pihl K., et al., “First‐Trimester Biomarkers and the Risk of Cerebral Palsy,” Early Human Development 167 (2022): 105564. [DOI] [PubMed] [Google Scholar]
- 42. van Batenburg‐Eddes T., de Groot L., Steegers E. A. P., et al., “Fetal Programming of Infant Neuromotor Development: The Generation R Study,” Pediatric Research 67 (2010): 132–137. [DOI] [PubMed] [Google Scholar]
- 43. Henrichs J., Schenk J. J., Barendregt C. S., et al., “Fetal Growth From Mid‐ to Late Pregnancy Is Associated With Infant Development: The Generation R Study,” Developmental Medicine & Child Neurology 52 (2010): 644–651. [DOI] [PubMed] [Google Scholar]
- 44. Walker S. P., Thame M. M., Chang S. M., Bennett F., and Forrester T. E., “Association of Growth in Utero With Cognitive Function at Age 6–8 Years,” Early Human Development 83 (2007): 355–360. [DOI] [PubMed] [Google Scholar]
- 45. von Ehrenstein O. S., Mikolajczyk R. T., and Zhang J., “Timing and Trajectories of Fetal Growth Related to Cognitive Development in Childhood,” American Journal of Epidemiology 170 (2009): 1388–1395. [DOI] [PubMed] [Google Scholar]
- 46. Norris T., Johnson W., Petherick E., et al., “Investigating the Relationship Between Fetal Growth and Academic Attainment: Secondary Analysis of the Born in Bradford (BiB) Cohort,” International Journal of Epidemiology 47 (2018): 1475–1484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Jarvis S., Glinianaia S. V., Torrioli M.‐G., et al., “Cerebral Palsy and Intrauterine Growth in Single Births: European Collaborative Study,” Lancet 362 (2003): 1106–1111. [DOI] [PubMed] [Google Scholar]
- 48. Korzeniewski S. J., Slaughter J., Lenski M., Haak P., and Paneth N., “The Complex Aetiology of Cerebral Palsy,” Nature Reviews. Neurology 14 (2018): 528–543. [DOI] [PubMed] [Google Scholar]
- 49. Baschat A. A., “Fetal Growth Restriction ‐ From Observation to Intervention,” Journal of Perinatal Medicine 38 (2010): 239–246. [DOI] [PubMed] [Google Scholar]
- 50. Cooke R. W. I., Lucas A., Yudkin P. L. N., and Pryse‐Davies J., “Head Circumference as an Index of Brain Weight in the Fetus and Newborn,” Early Human Development 1 (1977): 145–149. [DOI] [PubMed] [Google Scholar]
- 51. Kirkegaard I., Henriksen T. B., and Uldbjerg N., “Early Fetal Growth, PAPP‐A and Free Beta‐hCG in Relation to Risk of Delivering a Small‐for‐Gestational Age Infant,” Ultrasound in Obstetrics & Gynecology 37 (2011): 341–347. [DOI] [PubMed] [Google Scholar]
- 52. Simic M., Stephansson O., Petersson G., Cnattingius S., and Wikström A.‐K., “Slow Fetal Growth Between First and Early Second Trimester Ultrasound Scans and Risk of Small for Gestational Age (SGA) Birth,” PLoS One 12 (2017): e0184853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Kirkegaard I., Uldbjerg N., Petersen O. B., Tørring N., and Henriksen T. B., “PAPP‐A, Free β‐hCG, and Early Fetal Growth Identify Two Pathways Leading to Preterm Delivery,” Prenatal Diagnosis 30 (2010): 956–963. [DOI] [PubMed] [Google Scholar]
- 54. Elenis E., Wikström A.‐K., and Simic M., “Accelerated Fetal Growth in Early Pregnancy and Risk of Preterm Birth: A Prospective Cohort Study,” BMC Pregnancy and Childbirth 20 (2020): 764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Sacco R., Gabriele S., and Persico A. M., “Head Circumference and Brain Size in Autism Spectrum Disorder: A Systematic Review and Meta‐Analysis,” Psychiatry Research: Neuroimaging 234 (2015): 239–251. [DOI] [PubMed] [Google Scholar]
- 56. Courchesne E., Carper R., and Akshoomoff N., “Evidence of Brain Overgrowth in the First Year of Life in Autism,” Jama 290 (2003): 337–344. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Mean Z‐scores for foetal biometrics measured in the second trimester (GA 18–21) in children with neurodevelopmental disorders (NND) compared to children without.
Table S2: Sensitivity analyses on the cohort without obstetric sentinel events. Mean Z‐scores for foetal biometrics measured in the second trimester (GA 18–21) in children with neurodevelopmental disorders (NND) compared to children without.
Table S3: Sensitivity analyses on the cohort without obstetric sentinel events. Differences in hazard and cumulative risk for neurodevelopmental disorders (NND) in individuals with abnormal second‐trimester (GA 18–21) foetal biometric measurements.
Data Availability Statement
The data that support the findings of this study are available with permission from Statistics Denmark: www.dst.dk/en/TilSalg/data‐til‐forskning. These data are available under licence and subject to restrictions.
