Sexual dimorphism of the fetal brain biometry: an MRI-based study

Fetal growth assessment is a key component of prenatal care. Sex-specific fetal brain nomograms on ultrasound are available and are clinically used. In recent years, the use of fetal MRI has been increasing; however, there are no sex-specific fetal CNS nomograms on MRI. The study aimed to assess the differences in fetal brain biometry and growth trajectories and to create population-based standards of the fetal brain on MRI. In this cross-sectional study, brain structures of singleton fetuses with normal brain MRI scans were analyzed: biparietal diameter, occipitofrontal diameter, trans-cerebellar diameter, and the corpus callosum were measured and converted into centiles. Sex-specific nomograms were created. A total of 3848 MRI scans were performed in one tertiary medical center between 2011 and 2019; of them, 598 fetuses met the inclusion criteria, 300 males and 298 females between 28- and 37-weeks’ gestation. Males had significantly larger occipitofrontal diameter than females (median 75%, IQR 54–88%; median 61%, IQR 40–77%) and biparietal diameter (median 63%, IQR 42–82%; median 50%, IQR 25–73%), respectively (p < 0.001). The cerebellum had the greatest growth rate, with a 1.5-fold increase in diameter between 28 and 37 weeks’ gestation, with no measurement difference between the sexes (p = 0.239). No significant difference was found in the corpus callosum (p = 0.074). Measuring both sexes on the same nomograms may result in over-estimation of male fetuses and under-estimation of females. We provide fetal sex-specific nomograms on two-dimensional MRI.

T2-weighted magnetic resonance imaging at 32 gestational week, showing measurement of structures in the fetal brain: occipitofrontal diameter (A) and length of the corpus callosum (B) at midline sagittal plan; biparietal diameter (C) and trans-cerebellar diameter (D) at coronal plan

Introduction
Fetal growth assessment is a key component of prenatal care, with several ultrasound and MRI-based reference ranges available [1][2][3].
Sex-specific nomograms for the pediatric population exist and are routinely used [4,5]. However, only recently (2019), Galjaard et al. [6] published sex-specific longitudinal fetal brain growth curves, based on the measurements of over 27,000 fetuses on ultrasound, allowing integration with neonatal and pediatric WHO sex-specific head growth curves [5]. They showed that the head circumference (HC) and biparietal diameter (BPD) are considerably larger in males compared to females from 20 weeks of gestation onwards (p < 0.001) [6]. In recent years, the use of MRI in the assessment of the fetal brain has been investigated and data on the fetal central nervous system (CNS) is accumulating [7][8][9]. However, there are no fetal sex-specific nomograms for MRI. This study aimed to assess the differences in the fetal CNS biometry and the growth trajectories among males and females on MRI.

Materials and methods
This is a cross-sectional study, approved by the institutional review board (IRB) of our medical center. MRI scans were obtained between 2011 and 2019 at one tertiary referral medical center. The study population included singleton fetuses between 28 and 37 gestational weeks. Pregnancy dating was based on ultrasound measurement of crown-rump length in the first trimester. Only examination with normal brain MRI were included. Examinations of fetuses with mild isolated extra-cranial anomalies or maternal cytomegalovirus infection with no intracranial anomalies were also included, as acceptable in similar studies [1,[10][11][12]. Fetuses with central nervous system anomalies were excluded. The MRI quality, MRI two-dimensional measurements, maternal age during gestation, fetal sex, and presentation were recorded. The two-dimensional measurements included fetal brain biparietal diameter (BPD), occipitofrontal diameter (OFD), trans-cerebellar diameter (TCD), and the corpus callosum (CC). All measurements were converted into centiles according to Tilea et al. (Fig. 1) [1]. For each structure measured, only satisfactory images in terms of quality and alignment were selected to be measured. Fetal brain MRI was performed using a 1.5 T system (Optima 1.5 T; GE Healthcare Ultrasound, Milwaukee, WI, USA). Single-shot fast spin-echo T2-weighted sequences in orthogonal planes were performed using the following parameters: section thickness 3-4 mm; no gap; flexible coil (eight-channel cardiac coil); matrix 320 × 224; echo time (TE) 90 ms; and repetition time (TR) 1298 ms. The field of view was determined by the size of the fetal head: 24 cm for the smaller fetuses and 30 cm for the larger fetuses. T1 fast-spoiled gradientecho sequences were performed only in the axial plane with a larger field of view (400 mm), of section thickness 4 mm, gap 0.5 mm, TR 160 ms, and TE 2.3 ms [13].

Statistical analysis
Categorical variables were reported as frequency and percentage. Variables were evaluated for normal distribution. Variables with normal distribution were presented as mean and standard deviation. Variables with abnormal distribution were presented as a median and interquartile range. Independent-sample t test and Mann-Whitney test were used to compare the brain measurements of male and female fetuses in the whole cohort and each week. The mean values in female and male fetuses, as well as 95% CI of the mean, were presented using a plot. Generalized estimating equations model was used to study the association between brain measurements and gestational age. The centiles were estimated by the Generalized Additive Models for Location, Scale, and Shape model. Normal distribution, smoothing splines and Rigby and Stasinopoulos algorithm were applied. All statistical tests were two sided and p < 0.05 was considered statistically significant. IBM SPSS statistics version 25 (IBM corp. Armnok, New-York, USA, 2017) and R (version 3.6.1, R Foundation for Statistical Computing, Vienna, Austria, 2019) was used for all statistical analysis. The "R" package "gamlss" (Rigby RA and Stasinopoulos DM. gamlss: Generalized Additive Models for Location Scale and Shape) was used to calculate the centiles.

Results
A total of 3848 fetal brain MRI scans were performed in our medical center between 2011 and 2019, of which 598 fetuses were included in the study, 300 males and 298 females between 28-and 37-weeks' gestation (Supplement A, B). The mean gestational age was 33.2 weeks (SD 2.26). Mean maternal age was 31.76 years (SD 4.87). Male prevalence ranged from 43% to 58% each week (Fig. 1). The cerebellum revealed the greatest growth rate, with a 1.5-fold increase (146.2%) in diameter between 28 and 37 weeks' gestation, with a relative 3-5% growth rate per week. Occipitofrontal diameter and the biparietal diameter showed 126.25% and 133.44% increase, respectively. The corpus callosum demonstrated the lowest growth rate with only 121% increase. Differences in MRI values in centiles of the biparietal diameter, occipitofrontal diameter, trans-cerebellar diameter, and the corpus callous between males (n = 300) and females (n = 298) are presented in Supplement C.
Sex-specific nomograms were constructed for each of these structures in males and females and are shown in Supplement D and Table 1.

Discussion
The differences among sexes from early fetal period to adulthood are well documented in the literature. Our study further delineates these differences and is in accordance with previous publications [6,[14][15][16][17][18][19]. We have found that the biparietal diameter and the occipitofrontal diameter were significantly larger in male fetuses on two-dimensional MRI (p < 0.001). Melamed et al. [20], have also shown that the most distinct difference among males and females was the sonographic bone-bone biparietal diameter, with a male-tofemale ratio of 1.021, observed as soon as the early second trimester [20]. In a study previously published by our group, we have also shown a potential difference between the sexes in fetal brain dimensions, persisting after birth [21].
The corpus callosum measurements were similar between males and females (p = 0.074). Corpus callosum sexual dimorphism has been a matter of debate in the literature. A study by Achiron et al. showed that female fetuses had wider corpus callosum than males on ultrasound between 16 and 36 weeks' gestation [14]. Their findings regarding the CC length, however, were insignificant statistically, as the mean length of the CC was 27.46 ± 8.5 mm (95% CI 25.78 ± 29.13) in males and 26.93 ± 8.4 (95% CI 25.27 ± 28.59) mm in females. Differences in the adult corpus callosum measurements have also been reported, implicating the ongoing development of the corpus callosum beyond the fetal period [22,23]. In other studies, however, no sex differences were found in the fiber composition of the adult corpus callosum.
The trans-cerebellar diameter growth pattern and measurements were similar among males and females (p = 0.239), as previously shown on ultrasound [24]. The cerebellum also showed the greatest growth rate, with a 1.5-fold increase in diameter between 28 and 37 weeks' gestation.
Recently (2021), a study examining the volumetric growth of intracranial structures in healthy fetuses also showed that the cerebellum had the greatest weekly agerelated change of 19.4% [19]. Hatab et al. showed that the cerebellar volume correlates with gestational age, making it a possible marker for gestational age on ultrasound, however, they did not investigate the influence of fetal sex [25]. Similar findings in our study and the homogeneity among sexes suggest that trans-cerebellar diameter can be a possible marker for fetal gestational age on two-dimensional MRI as well.
The underlying reason for the differences in the growth trajectories between males and females is unclear, and it there is a volumetric asymmetry in the male brain in favor of the right hemisphere, while the fetal female is likely to have two hemispheres of the same size or a left hemisphere that is slightly larger than its right counterpart, supporting the hypothesis that testosterone in utero has different effects on the fetal brain structures. Although the genetic basis for these differences is still unclear, some genetic evidence has also been published regarding the differences existing between males and females, even from the early fetal period [26,27]. Franklin et al. study in mice showed that the androgenregulated, sexually dimorphic Rbm48 gene expression might present a molecular mechanism by which perinatal androgens control the development of sexual dimorphism in cortical and hippocampal structure and function.
Our study has several limitations. This study focused on gestational ages 28-37; therefore, its conclusions are limited to the third trimester. Several other studies have suggested that there is no significant difference in fetal growth in the early pregnancy, with an acceleration in the latter half of gestation, making the focus on the third trimester valuable.
Another limitation is that the study population consisted of cases referred for MRI due to a suspected condition, which may be a source of selection bias. Nevertheless, we have only included cases with normal fetal brain scans. Moreover, fetuses with extra-cranial congenital malformations comprised only a minor part of the study cohort (11.5%), and these malformations were not expected to influence fetal brain development. Other population-based studies have also included infants with congenital malformations [28][29][30]. Another limitation is its cross-sectional design; while it is feasible to conduct a longitudinal study of fetal growth on ultrasound, MRI is not available for multiple recurrent assessments.
Our findings regarding the alternation in fetal sex-related growth trajectories expand the understanding of fetal development. Measuring both sexes on the same nomograms may result in over-estimation of male fetuses and under-estimation of females. As the clinical use of two-dimensional MRI is constantly increasing, there is a need for such nomograms on MRI as well. This is the first MRI-based study focusing on the differences between males and females in fetal brain biometry. It is encompassed a large cohort of 598 fetuses. While BPD and OFD ultrasound measurements are of the fetal calvaria, measurements on two-dimensional MRI are of the fetal brain parenchyma, adding valuable information.

Conclusions
This study shows evidence for sexual dimorphism on twodimensional MRI, as occipitofrontal and biparietal diameters were larger in males. No significant differences were found between males and females in the corpus callosum or the trans-cerebellar diameter. Measuring both sexes on the same nomograms may result in over-estimation of male fetuses and under-estimation of females. We provide sexspecific two-dimensional MRI biometric nomograms of the fetal brain in the third trimester that are available for clinical use.