Pathik D. Wadhwa’s research while affiliated with University of California System and other places

Importance: Maternal diabetes (MD) and maternal obesity (MO) have been robustly established to confer health risks in offspring. Additionally, mounting evidence suggests that these fetal programming effects vary by sex, but whether these factors independently or interactively influence infant brain development remains unclear. Objectives: To characterize interactions between MD, MO, and sex on offspring subcortical brain volumes. Design, setting and participants: This was a cross-sectional study of 1,966 infants from six international cohorts. Exposures: MD and MO Main outcomes and measures: MRI-based subcortical brain volumes (thalamus, amygdala, hippocampus, pallidum, putamen, caudate) were segmented and mixed effects models were used to examine associations, controlling for age at scan, prematurity, birthweight, maternal education, and intracranial volume. Backward elimination regression was used to identify the best fitting model (3-way interaction, 2-way interaction, no interaction) for each region and false discovery rate (FDR) corrections were applied. Results: Of 1,966 infants, 46% were female (N=909), 9% were exposed to MD (N=172), and 21% were exposed to MO (N=386). MRI scans were performed at (mean±SD) 25.9±18.8 days of age. There was a significant interaction between MD, MO and sex in the thalamus (standardized β=−0.32, 95%CI −0.54 to −0.11, FDR corrected P=0.014). In female infants, MD (standardized β=-0.10, 95%CI −0.02 to −0.003, P=0.04) and MO (standardized β =−0.09, 95%CI −0.14 to −0.03, P=0.003) were independently and negatively associated with thalamic volume. In males, a significant interaction between MD and MO was observed (standardized β =−0.20, 95%CI −0.34 to −0.06, P=0.005), with post hoc analysis showing that males with combined exposure to MD and MO had lower thalamic volume compared to those with one or neither exposure (all Ps<0.05). In the hippocampus, an interaction between MO and infant sex was identified (standardized β =0.15, 95%CI 0.05 to 0.26, FDR corrected P=0.015), whereby MO (independent of MD) was associated with lower offspring hippocampal volume in females only (standardized β =−0.12, 95%CI −0.2 to −0.05, P=0.002). Conclusion and relevance: Our results suggest independent, interactive associations of intrauterine exposure to MD and MO with infant subcortical brain volumes, varying by sex. This has implications for future metabolic disorders, among other health risks.

Importance Short sleep duration during pregnancy and the perimenopausal period has been associated with adverse cardiometabolic outcomes. However, it remains unclear how sleep duration changes after delivery and whether such changes are associated with the cardiometabolic health of birthing people. Objective To investigate whether persistently short sleep during pregnancy and after delivery is associated with incident hypertension and metabolic syndrome. Design, Setting, and Participants This secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be Heart Health Study (NuMoM2b-HHS), an ongoing prospective cohort study, was conducted between September 5, 2023, and March 1, 2024, in 8 US academic medical centers. Participants were aged 18 years or older at NuMoM2b enrollment; recruited during their first pregnancy between October 1, 2010, and September 30, 2013; and followed up for a mean (SD) of 3.1 (0.9) years after delivery. Exposures Self-reported short sleep duration (<7 hours) during pregnancy and 2 to 7 years after delivery was defined as persistent short sleep. Main Outcomes and Measures Incident hypertension and metabolic syndrome (MetS) at follow-up. Regression models were used to estimate relative risks of incident hypertension and MetS by sleep duration pattern. Hypertension analyses excluded participants with hypertension at baseline, and MetS analyses excluded participants with MetS at baseline. Multivariable models included a priori covariates of baseline age and time from delivery to follow-up. Incident hypertension analyses included an additional covariate of body mass index at baseline. Results Among 3922 participants (mean [SD] age, 27.3 [5.4] years; 598 Hispanic [15.2%], 485 non-Hispanic Black [12.4%], and 2542 non-Hispanic White [64.8%]), 565 individuals (14.4%) experienced persistent short sleep. Non-Hispanic Black (adjusted odds ratio [aOR], 2.17; 95% CI, 1.59-2.97) and unmarried (aOR, 1.68, 95% CI, 1.29-2.19) participants were significantly more likely to experience persistent short sleep compared with non-Hispanic White and married participants, respectively. Persistent short sleep was associated with higher odds of incident MetS (aOR, 1.60; 95% CI, 1.21-2.11) but not incident hypertension (aOR, 0.91; 95% CI, 0.69-1.19). Conclusions and Relevance In this study, short sleep duration that persisted from pregnancy to 2 to 7 years after delivery was associated with a greater risk for adverse cardiometabolic outcomes. Future studies should explore whether sleep-targeted interventions during and after pregnancy are associated with improved cardiometabolic health outcomes, particularly among populations at increased risk.

Background: A major challenge in epidemiology is knowing when an exposure effect is large enough to be clinically important, in particular how to interpret a difference in mean outcome in unexposed/exposed groups. Where it can be calculated, the proportion/percentage beyond a suitable cut-point is useful in defining individuals at high risk to give a more meaningful outcome. In this simulation study we compute differences in outcome means and proportions that arise from hypothetical small effects in vulnerable sub-populations. Methods: Data from over 28,000 mother/child pairs belonging to the Environmental influences on Child Health Outcomes Program were used to examine the impact of hypothetical environmental exposures on mean birthweight, and low birthweight (LBW) (birthweight < 2500g). We computed mean birthweight in unexposed/exposed groups by sociodemographic categories (maternal education, health insurance, race, ethnicity) using a range of hypothetical exposure effect sizes. We compared the difference in mean birthweight and the percentage LBW, calculated using a distributional approach. Results: When the hypothetical mean exposure effect was fixed (at 50, 125, 167 or 250g), the absolute difference in % LBW (risk difference) was not constant but varied by socioeconomic categories. The risk differences were greater in sub-populations with the highest baseline percentages LBW: ranging from 3.1-5.3 percentage points for exposure effect of 125g. Similar patterns were seen for other mean exposure sizes simulated. Conclusions: Vulnerable sub-populations with greater baseline percentages at high risk fare worse when exposed to a small insult compared to the general population. This illustrates another facet of health disparity in vulnerable individuals.

Background: The Modified Checklist for Autism in Toddlers (M-CHAT) is a common pediatric screening tool with mixed accuracy findings. Prior evidence supports M-CHAT screening for developmental concerns, especially in toddlers born preterm. This study examined M-CHAT accuracy in a large, nationwide sample. Methods: 3393 participants from the Environmental influences on Child Health Outcomes (ECHO) program were included. Harmonized M-CHAT (M-CHAT-H) results were compared with parent-reported autism diagnosis and autism-related characteristics to assess accuracy for term and preterm children, together and separately. Generalized estimating equations, clustering for ECHO cohort and controlling for demographic covariates, were used to examine associations between developmental and behavioral characteristics with M-CHAT-H accuracy. Results: Sensitivity of the M-CHAT-H ranged from 36 to 60%; specificity ranged from 88 to 99%. Positive M-CHAT-H was associated with more developmental delays and behavior problems. Children with severe motor delays and more autism-related problems were more likely to have a false-negative M-CHAT-H. Children with fewer behavior problems and fewer autism-related concerns were more likely to have a false-positive screen. Conclusion: The M-CHAT-H accurately detects children at low risk for autism and children at increased risk with moderate accuracy. These findings support use of the M-CHAT-H in assessing autism risk and developmental and behavioral concerns in children. Impact: Previous literature regarding accuracy of the Modified Checklist for Autism in Toddlers (M-CHAT) is mixed but this study provides evidence that the M-CHAT performs well in detecting children at low risk for autism and consistently detects children with developmental delays and behavioral problems. The M-CHAT moderately detects children at increased risk for autism and remains a useful screening tool. This study examines M-CHAT accuracy in a large-scale, nationwide sample, examining associations between screening accuracy and developmental outcomes. These findings impact pediatric screening for autism, supporting continued use of the M-CHAT while further elucidating the factors associated with inaccurate screens.

Background The mediobasal hypothalamus (MBH) is a key brain area for regulation of energy balance. Previous neuroimaging studies suggest that T2‐based signal properties indicative of cellular inflammatory response (gliosis) are present in adults and children with obesity, and predicts greater adiposity gain in children at risk of obesity. Objectives/Methods The current study aimed to extend this concept to the early life period by considering if, in full‐term healthy neonates (up to n = 35), MRI evidence of MBH gliosis is associated with changes in early life (neonatal to six months) body fat percentage measured by DXA. Results In this initial study, neonatal T2 signal in the MBH was positively associated with six‐month changes in body fat percentage. Conclusion This finding supports the notion that underlying processes in the MBH may play a role in early life growth and, by extension, childhood obesity risk.

Robust segmentation is critical for deriving quantitative measures from large-scale, multi-center, and longitudinal medical scans. Manually annotating medical scans, however , is expensive and labor-intensive and may not always be available in every domain. Unsupervised domain adaptation (UDA) is a well-studied technique that alleviates this label-scarcity problem by leveraging available labels from another domain. In this study, we introduce Masked Au-toencoding and Pseudo-Labeling Segmentation (MAPSeg), a unified UDA framework with great versatility and superior performance for heterogeneous and volumetric medical image segmentation. To the best of our knowledge, this is the first study that systematically reviews and develops a framework to tackle four different domain shifts in medical image segmentation. More importantly, MAPSeg is the first framework that can be applied to centralized, federated, and test-time UDA while maintaining comparable performance. We compare MAPSeg with previous state-of-the-art methods on a private infant brain MRI dataset and a public cardiac CT-MRI dataset, and MAPSeg outperforms others by a large margin (10.5 Dice improvement on the private MRI dataset and 5.7 on the public CT-MRI dataset). MAPSeg poses great practical value and can be applied to real-world problems. Our code and pretrained model will be available later.