Objective
We tested the hypothesis whether small-for-gestational-age (SGA) fetuses have different brain stem and cerebellar morphometry when compared with appropriate-for-gestational-age (AGA) fetuses and whether the differences in these structures were associated with their neonatal neurobehavior.
Study Design
Magnetic resonance imaging was performed on 51 SGA fetuses and 47 AGA fetuses at 37 weeks’ gestation. Pontine width, medullar width, vermian width and height, cerebellar primary fissure’s depth, and cerebellar volume were measured and corrected by biparietal diameter and cerebellar volume by total intracranial volume. Ratios were compared between cases and control subjects. The association between morphometric differences and neurobehavioral outcome in SGAs was tested.
Results
Brainstem and cerebellar ratios were significantly larger in SGA fetuses: pontine width, SGA 0.143 ± 0.01 vs AGA 0.135 ± 0.01 ( P < .01); medullar width, SGA 0.088 ± 0.01 vs AGA 0.083 ± 0.01 ( P = .03); vermian width, SGA 0.181 ± 0.03 vs AGA 0.162 ± 0.02 ( P < .01); vermian height, SGA 0.235 ± 0.03 vs AGA 0.222 ± 0.01 ( P < .01); cerebellar volume, SGA 0.042 ± 0.01 vs AGA 0.038 ± 0.00 ( P = .04); with deeper cerebellar primary fissure in SGAs, SGA 0.041 ± 0.01 vs AGA 0.035 ± 0.01 ( P = .01). Medullar, cerebellar biometries, and volumetry were significantly associated with different Neonatal Behavioral Assessment Scale cluster scores in SGA infants.
Conclusion
Brain stem and cerebellar morphometric measurements are significantly different in term SGA fetuses, which are associated significantly with their neurobehavioral outcome. This finding supports the existence of brain microstructural changes in SGA fetuses and lays the basis for potential image biomarkers to detect fetuses who are at risk.
Intrauterine growth restriction (IUGR) is defined as the underachievement of the genetic growth potential. Placental insufficiency is considered one of the main causes of this condition, by inducing an undernourished and hypoxic environment in which the growth-restricted fetus develops. Exposure of the developing fetal brain to adverse factors can have lifelong consequences. An increasing amount of evidence has shown how severely affected IUGR preterm infants present micro and macro structural brain differences, which are associated to a higher risk for long-term neurologic morbidity. Signs of brain reorganization are not exclusive of severe small for gestational age (SGA), which is present independently of prematurity in close-to-term SGA fetuses. Differences in brain metabolism, microstructure, and cortical sulcation pattern have been reported in this condition at term. Also, neurobehavioral and neurocognitive difficulties are present in this subgroup from the neonatal period all the way to late infancy.
The relevance of an impaired neurodevelopment in SGA fetuses should be stressed because of its high prevalence, which affects 10% of all deliveries at term. One of the current challenges in this field is to improve the characterization of changes that are induced by IUGR in the fetal brain and ideally to find potential brain biomarkers to identify cases at highest risk. Brain stem and cerebellum are crucial structures in brain function. The first is considered a crossway for motor and sensory systems, and the latter has a pivotal role in motor control and other tasks such as memory, attention, or language. Thus, they are involved in critical functions that have been described to be altered in children with IUGR. Additionally, these structures can be easily visualized in a midsagittal view in standard neurosonography, which facilitates a potential transference of these measurements to fetal brain ultrasound scans. The existence of changes in brain stem and cerebellum in SGA fetuses and their potential relationship with neurobehavior has not been yet assessed.
In this study, we performed brain stem and cerebellum morphometric analysis in 51 SGA fetuses and in 47 appropriate-for-gestational-age term fetuses and analyzed the association of these measurements with the neurobehavioral outcome in the SGA group.
Material and Methods
Study cohort
The protocol for this study was approved by the institutional ethics committee (Institutional Review Board 2008/4422), and all participants gave written informed consent. A prospective cohort of 51 SGA fetuses was included. SGA was defined by an estimated fetal weight and confirmed birthweight <10th percentile according to local standards and normal umbilical artery pulsatility index (<95th percentile). A control group of 47 consecutive pregnancies with appropriate-for-gestational-age fetuses with a confirmed birthweight >10th percentile was selected among normal pregnancies and observed at our institution after agreeing to participate in the study. Cases with congenital malformations, chromosomal abnormalities, perinatal infections, chronic maternal disease, and noncephalic presentations were considered to be not eligible for this study.
Clinical and ultrasound data
Gestational age was corrected from fetal crown-rump length in the first trimester. Prenatal Doppler ultrasound examinations were performed with a 6-2–MHz curved-array transducer (General Electric Voluson E8; GE Medical Systems, Zipf, Austria) within 1 week from magnetic resonance imaging (MRI) scan.
Fetal MRI acquisition
All cases were scanned at 37 weeks of gestation on a clinical 3.0 Tesla MR system, in a scanner (TIM TRIO 3.0 T; Siemens, Munich, Germany) without sedation. A body coil with 8 elements was wrapped around the mother’s abdomen. Routine fetal imaging required between 15-30 minutes. Fetal neuroimaging consisted on single-shot, fast spin echo T2 weighted sequences (TR 990 ms, TE 137 ms, slice thickness 3.5 mm, no gap, field of view 260 mm, voxel size 1.4 × 1.4 × 3.5 mm, matrix 192 × 192, flip angle 180 degrees, acquisition time 24 seconds) that were acquired in the 3 orthogonal planes. If the quality of the images was suboptimal, sequences were repeated. Structural MRI images were reviewed for the presence of anatomic abnormalities by an experienced neuroradiologist who was blinded to group membership.
Fetal imaging post processing and delineation
Offline analyses of brain biometric and volumetric measurements were performed with the semiautomatic ImageJ software (version 1.46r; Image J, U.S. National Institutes of Health Bethesda, Maryland, MD) and ITK-SNAP software (University of Pennsylvania, Pennsylvania, PA) respectively.
Biometric measurements
All biometric parameters were corrected by biparietal diameter and therefore expressed as ratios ( Figure 1 ). Biparietal diameter was measured in an axial plane at the level of the thalami. Delineation was performed perpendicularly to the interhemispheric fissure from the inner table of the skull on 1 side to the outer table on the contralateral side at the widest part of the skull.
The rest of the biometries were performed in a midsagittal plane: pontine width (distance between the anterior and posterior boundaries of the pons in its widest section, including the tegmentum, perpendicularly to its rostrocaudal axis); medullar width (distance between the anterior and posterior portions of the medulla, just inferiorly to the pons); vermian width or anterioposterior length (maximal distance between the most anterior part of the central lobule and the most posterior part of the tuber ); vermian height or craniocaudal length (maximal distance between the most cranial part of the culmen and the most caudal part of the uvula ); cerebellar primary fissure’s depth (measured from its deepest point to the cerebellar surface, on the fissure that divides the anterior from the posterior lobe of the cerebellum).
Volumetric measurements
Cerebellar volume (CBV) was corrected by total intracranial brain volume (TICV) and thus expressed as a ratio ( Figure 2 ). CBV and TICV were obtained by measure of the area in each slice that contained cerebellar tissue for CBV and extra- and intraventricular cerebrospinal fluid, cerebral, cerebellar, and brain stem parenchymal tissue for TICV, as reported elsewhere. All areas that were obtained from this multiplanar analysis were added and multiplied by 3.5 mm of slice thickness to obtain volumetry based on Cavalieri’s principle considering no gap interval between them.
Interobserver variability
To establish biometric and volumetric measurement reproducibility, a total of 15 and 10 cases, respectively, were assessed completely independently by 2 experienced examiners who were blinded to group membership. Interclass correlation coefficients for total agreement were calculated to define the reliability of the measurements.
Neonatal Behavioral Assessment Scale test
Postnatal follow up was offered to all SGA patients. The Neonatal Behavioral Assessment Scale (NBAS) test was performed in 44 SGA patients prospectively at 43 ± 2.9 weeks’ gestation by 1 of 2 observers who were accredited by The Brazelton Institute (Harvard Medical School, Boston, MA) and who were blinded to the perinatal outcomes. This test is an acceptable method to assess both cortical and subcortical functions from 3 days to 6 weeks after delivery. This test evaluates 35 items that are rated on a 1 to 9 scale, where 9 is the best performance for some areas and for others this is represented by the central score of 5. Items are grouped into 6 clusters and include habituation (habituation to light, rattle, bell, and tactile stimulation of the foot), motor (general tone, elicited activity, spontaneous activity, and motor maturity), social-interactive (responses to visual, animate, and inanimate auditory stimuli and alertness), organization of state (irritability, state lability, maximal excitation, and reaction time), and regulation of state (self-quieting and hand-to-mouth responses). The social-interactive cluster was subscored for visual and auditory stimuli. In addition, as reported recently by the authors of the NBAS test, an aggregation of individual items (alertness, quality of the alert responsiveness, and cost of attention) was used to evaluate the capacity of the newborn infant’s attention. Neonates were assessed in the afternoon between feedings in a small, semidark quiet room with a temperature between 22° and 27°C in the presence of >1 parents. Scores from each NBAS cluster were converted into percentiles according to normal curve references for our population and classified as normal if they were ≥5th percentile; an abnormal result was a score <5th percentile.
Statistical analysis of biometric, volumetric, and clinical data
Student t test for independent samples and Pearson’s χ 2 or Fisher exact tests were used to compare quantitative and qualitative data, respectively. The Student t test for independent samples and a multivariate analysis of covariance (MANCOVA) were conducted to detect differences in biometric and volumetric measurements in both areas of interest; adjustment was made for maternal smoking, neonatal sex, and gestational age at MRI. The association between measurements and NBAS test cluster scores in SGA cases was assessed by MANCOVA test that considered the same covariates. Also, this association was tested with the performance of an ordinal regression in which the number of NBAS test clusters <5th percentile were considered to be the dependent variable and brain stem and cerebellar measurements to be the independent variables. The same adjustment covariates were considered in this test. SPSS software (version 17.0; SPSS, Chicago, IL) was used for statistical analyses.
Results
Clinical characteristics in the study population
All 98 fetuses from our sample underwent fetal MRI at 37 weeks’ gestation. Maternal characteristics and time of MRI scans did not differ between cases and control subjects ( Table 1 ). As expected, SGA fetuses were delivered earlier with higher rates of labor induction and cesarean section ( Table 2 ).
| Variable | Fetus | P value a | |
|---|---|---|---|
| Small for gestational age (n = 51) | Appropriate for gestational age (n = 47) | ||
| Maternal age, y b | 30.55 ± 5.59 | 32.21 ± 4.6 | .11 |
| Maternal smoking, % | 29.4 | 8.5 | .01 |
| Maternal body mass index, kg/m 2 b | 21.85 ± 3.29 | 23.21 ± 3.58 | .05 |
| White ethnicity, % | 72.3 | 73.5 | .91 |
| Primiparity, % | 76.5 | 66.0 | .25 |
| Low socioeconomic status, c % | 26.7 | 18.5 | .41 |
| Gestational age at magnetic resonance imaging, wk b | 37.31 ± 0.90 | 37.52 ± 0.83 | .23 |
a Student t test for independent samples, Pearson’s χ 2 , or Fisher exact test as appropriate
b Results are expressed as mean ± SD
c Routine occupations, long-term unemployment, or never worked (United Kingdom National Statistics Socio-economic Classification).
| Variable | Fetus | P value a | |
|---|---|---|---|
| Small for gestational age (n = 51) | Appropriate for gestational age (n = 47) | ||
| Gestational at birth, wk b | 38.64 ± 1.03 | 40.01 ± 1.02 | < .01 |
| Birthweight, g b | 2438 ± 282.13 | 3454 ± 305.9 | < .01 |
| Birthweight percentile b | 3.73 ± 6.64 | 54.76 ± 24.44 | < .01 |
| Neonatal sex: male, % | 53.2 | 52.9 | .98 |
| Labor induction, % | 74.5 | 17.8 | < .01 |
| Emergency cesarean delivery, % | 23.4 | 6.4 | .02 |
| Cesarean delivery, % | 37.3 | 13.0 | .01 |
| Neonatal acidosis, c % | 12.0 | 7.3 | .45 |
| Apgar score <7 at 5 minutes, % | 0 | 0 | |
a Student t test for independent samples, Pearson’s χ 2 , or Fisher exact test as appropriate
b Results are expressed as mean ± SD
Biometric and volumetric analysis
Brain stem, cerebellar ratios, and CBV/TICV were significantly larger in cases than control subjects ( Table 3 ). Also, primary fissure depth was significantly more pronounced in SGA fetuses after adjustment by biparietal diameter.
| Variable | Fetus a | P value b | P value c | |
|---|---|---|---|---|
| Small for gestational age (n = 51) | Appropriate for gestational age (n = 47) | |||
| Biparietal diameter, mm | 91.82 ± 4.28 | 100.13 ± 3.51 | < .01 | < .01 |
| Pontine width/biperietal diameter, mm | 0.143 ± 0.01 | 0.135 ± 0.01 | < .01 | < .01 |
| Medullar width/biperietal diameter, mm | 0.088 ± 0.01 | 0.083 ± 0.01 | .01 | .03 |
| Vermian width/biperietal diameter, mm | 0.181 ± 0.03 | 0.162 ± 0.02 | < .01 | < .01 |
| Vermian height/biperietal diameter, mm | 0.235 ± 0.03 | 0.222 ± 0.01 | < .01 | < .01 |
| Primary fissure depth/ biperietal diameter, mm | 0.041 ± 0.01 | 0.035± 0.01 | .01 | .01 |
| Cerebellar volume/total intracranial volume, mm | 0.042 ± 0.01 | 0.038 ± 0.00 | .09 | .04 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
