Serum levels of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans in a sample of Lebanese pregnant women: The role of dietary, anthropometric, and environmental factors

Utilizing machine learning to classify persistent organic pollutants in the serum of pregnant women: a predictive modeling approach

Polychlorinated biphenyls (PCBs), organochlorine pesticides (OCPs), polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs), and per- and poly-fluoroalkyl substances (PFAS) are persistent organic pollutants (POPs) that remain detrimental to critical subpopulations, namely pregnant women. Required tests for biomonitoring are quite expensive. Moreover, statistical models aiming to discover the relationships between pollutants levels and human characteristics have their limitations. Therefore, the objective of this study is to use machine learning predictive models to further examine the pollutants' predictors, while comparing them. Levels of 33 congeners were measured in the serum of 269 pregnant women, from whom data was collected regarding sociodemographic, dietary, environmental, and anthropometric characteristics. Several machine learning algorithms were compared using "Python" for each pollutant: support vector machine (SVM), random forest, XGBoost, and neural networks. The aforementioned characteristics were included in the model as features. Prediction, accuracy, precision, recall, F1-score, area under the ROC curve (AUC), sensitivity, and specificity were retrieved to compare the models between them and among pollutants. The highest performing model for all pollutants was Random Forest. Results showed a moderate to acceptable performance and discriminative power among all POPs, with OCPs' model performing slightly better than all other models. Top related features for each model were also presented using SHAP analysis, detailing the predictors' negative or positive impact on the model. In conclusion, developing such a tool is of major importance in a context of limited financial and research resources. Nevertheless, machine learning models should always be interpreted with caution by exploring all evaluation metrics.

Dioxins and furans maternal transfer: A study of breast milk and cord serum levels among Lebanese mothers and associations with newborn anthropometric measurements

Dioxins and furans (PCDD/Fs) are anthropogenic pollutants that persist in the environment for long years, bioaccumulating in food & contaminating humans. In pregnancy, they can transfer through the placenta and reach the fetus, which negatively affects fetal growth. They can also reach newborns through breastfeeding. In this study, we focused on this critical subpopulation and identified the presence of PCDD/Fs among pregnant women in breast milk (n = 41) and cord serum (n = 49); we assessed the correlation between different matrices, evaluated the predictors and associations with newborn anthropometric measurements. Over 70.7 % of PCDD/Fs were detected in breast milk and 46.9-55.1 % in cord serum. Cord/maternal serum and breast milk to maternal serum ratios were > 1 with a significant positive Spearman correlation (0.669-0.729). Breast milk & maternal serum PCDD/Fs were associated inversely with age and positively with red meat intake. Cord serum PCDD/Fs were inversely associated with pre-pregnancy weight loss and passive smoking. Parity and gestational weight gain showed positive associations with Z-scores at birth. Z-score differences showed negative and positive associations with passive smoking and pre-pregnancy BMI respectively.

Maternal Serum, Cord and Human Milk Levels of Per- and Polyfluoroalkyl Substances (PFAS), Association with Predictors and Effect on Newborn Anthropometry

BACKGROUND

The understanding of per- and polyfluoroalkyl substances (PFAS) health effects is rapidly advancing among critical populations. Therefore, the objective of this study was to assess PFAS serum levels among Lebanese pregnant women, cord serum and human milk levels, their determinants, and effects on newborn anthropometry.

METHODS

We measured concentrations of six PFAS (PFHpA, PFOA, PFHxS, PFOS, PFNA and PFDA) using liquid chromatography MS/MS for 419 participants, of which 269 had sociodemographic, anthropometric, environmental and dietary information.

RESULTS

The percentage of detection for PFHpA, PFOA, PFHxS and PFOS was 36.3-37.7%. PFOA and PFOS levels (95th percentile) were higher than HBM-I and HBM-II values. While PFAS were not detected in cord serum, five compounds were detected in human milk. Multivariate regression showed that fish/shellfish consumption, vicinity to illegal incineration and higher educational level were associated with an almost twice higher risk of elevated PFHpA, PFOA, PFHxS and PFOS serum levels. Higher PFAS levels in human milk were observed with higher eggs and dairy products consumption, in addition to tap water (preliminary findings). Higher PFHpA was significantly associated with lower newborn weight-for-length Z-score at birth.

CONCLUSIONS

Findings establish the need for further studies, and urgent action to reduce exposure among subgroups with higher PFAS levels.