151 - Altered fetal brain development in human fetuses with non-syndromic isolated structural musculoskeletal birth defects

Poster Number: 151
Publication Number: 151.135

Lana Vasung, Boston Children's Hospital, Harvard Medical School, Boston, MA, United States; Esha Ahmad, Boston Children's Hospital, North Beach, MD, United States; Clemente Velasco-Annis, Boston Children's Hospital, Sunderland, MA, United States; edward Yang, Boston Children's Hospital, Jamaica Plain, MA, United States; Ali Gholipour, Boston Children's Hospital, Boston, MA, United States; Henry Feldman, Boston Children's Hospital, Boston, MA, United States; Patricia Ellen Grant, Boston Children's Hospital, Boston, MA, United States

Background: Proprioception (sense of self-movement, force, and body position) is altered in several congenital conditions that fall into the category of non-syndromic isolated musculoskeletal structural birth defects of the body (niMSBDs). However, to what extent are proprioceptive stimuli important for prenatal brain development and whether fetuses with niMSBDs have abnormal brain development remains a critical knowledge gap.

Objective: We aimed to identify whether fetuses with niMSBDs, compared to the controls, have altered brain development before birth.

Design/Methods: From the hospital database, we retrospectively collected in-vivo structural T2-weighted MRI of 69 fetuses (30 controls and 39 cases with niMSBDs including arthrogryposis, skeletal dysplasia, short long bones [ICD-10-CM codes: Q74.3, Q68.8, Q74.9, Q66.8, Q69, Q70-73, Q79.9]) between 17-37 gestational weeks (GW). The study was approved by the institutional review board. Fetal brain MRIs were automatically reconstructed and segmented using state-of-the-art MRI tools ensuring high accuracy and reproducibility. We calculated the volumes (in mm3) of transient fetal compartments (ganglionic eminence, proliferative zones, subplate zone, intermediate zone, and cortical plate), lateral ventricles, and other brain regions (limbic, basal ganglia with the thalamus, cerebellum). Relative hemispheric volumes (% hemisphere) were calculated, log-transformed, and used as a dependent variable. The robust regression method was used to identify and downweight any extreme values. We hypothesized that the dependent variable will be explained by the following factors or covariates: hemisphere (left-right), age (in GW), brain zone/region, and group (control or niMSBDs). We also included the interaction term (region*group) to provide region-specific group di!erences. Sex wasn't taken into consideration due to a large number of missing data. The resulting p-values were adjusted by the Holm procedure to limit the familywise type I error rate to 5%.

Results: Compared to the controls, fetuses with niMSBDs had significantly smaller intermediate zone (-51.2±7.5% SE) and cerebellum (-47.4±8.0% SE) after correction for multiple comparisons.

Conclusion(s): Our results indicate that early development of the cerebellum and intermediate zone might be altered in fetuses with niMSBDs.