RESUMO
BACKGROUND: The prevalence of metabolic syndrome is rapidly increasing in the United States. We hypothesized that prediction models using data obtained during pregnancy can accurately predict the future development of metabolic syndrome. OBJECTIVE: This study aimed to develop machine learning models to predict the development of metabolic syndrome using factors ascertained in nulliparous pregnant individuals. STUDY DESIGN: This was a secondary analysis of a prospective cohort study (Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be Heart Health Study [nuMoM2b-HHS]). Data were collected from October 2010 to October 2020, and analyzed from July 2023 to October 2023. Participants had in-person visits 2 to 7 years after their first delivery. The primary outcome was metabolic syndrome, defined by the National Cholesterol Education Program Adult Treatment Panel III criteria, which was measured within 2 to 7 years after delivery. A total of 127 variables that were obtained during pregnancy were evaluated. The data set was randomly split into a training set (70%) and a test set (30%). We developed a random forest model and a lasso regression model using variables obtained during pregnancy. We compared the area under the receiver operating characteristic curve for both models. Using the model with the better area under the receiver operating characteristic curve, we developed models that included fewer variables based on SHAP (SHapley Additive exPlanations) values and compared them with the original model. The final model chosen would have fewer variables and noninferior areas under the receiver operating characteristic curve. RESULTS: A total of 4225 individuals met the inclusion criteria; the mean (standard deviation) age was 27.0 (5.6) years. Of these, 754 (17.8%) developed metabolic syndrome. The area under the receiver operating characteristic curve of the random forest model was 0.878 (95% confidence interval, 0.846-0.909), which was higher than the 0.850 of the lasso model (95% confidence interval, 0.811-0.888; P<.001). Therefore, random forest models using fewer variables were developed. The random forest model with the top 3 variables (high-density lipoprotein, insulin, and high-sensitivity C-reactive protein) was chosen as the final model because it had the area under the receiver operating characteristic curve of 0.867 (95% confidence interval, 0.839-0.895), which was not inferior to the original model (P=.08). The area under the receiver operating characteristic curve of the final model in the test set was 0.847 (95% confidence interval, 0.821-0.873). An online application of the final model was developed (https://kawakita.shinyapps.io/metabolic/). CONCLUSION: We developed a model that can accurately predict the development of metabolic syndrome in 2 to 7 years after delivery.
RESUMO
OBJECTIVE: To determine whether adverse pregnancy outcomes are associated with a higher predicted 30-year risk of atherosclerotic cardiovascular disease (CVD; ie, coronary artery disease or stroke). METHODS: This was a secondary analysis of the prospective Nulliparous Pregnancy Outcomes Study-Monitoring Mothers-to-Be Heart Health Study longitudinal cohort. The exposures were adverse pregnancy outcomes during the first pregnancy (ie, gestational diabetes mellitus [GDM], hypertensive disorder of pregnancy, preterm birth, and small- and large-for-gestational-age [SGA, LGA] birth weight) modeled individually and secondarily as the cumulative number of adverse pregnancy outcomes (ie, none, one, two or more). The outcome was the 30-year risk of atherosclerotic CVD predicted with the Framingham Risk Score assessed at 2-7 years after delivery. Risk was measured both continuously in increments of 1% and categorically, with high predicted risk defined as a predicted risk of atherosclerotic CVD of 10% or more. Linear regression and modified Poisson models were adjusted for baseline covariates. RESULTS: Among 4,273 individuals who were assessed at a median of 3.1 years after delivery (interquartile range 2.5-3.7), the median predicted 30-year atherosclerotic CVD risk was 2.2% (interquartile range 1.4-3.4), and 1.8% had high predicted risk. Individuals with GDM (least mean square 5.93 vs 4.19, adjusted ß=1.45, 95% CI, 1.14-1.75), hypertensive disorder of pregnancy (4.95 vs 4.22, adjusted ß=0.49, 95% CI, 0.31-0.68), and preterm birth (4.81 vs 4.27, adjusted ß=0.47, 95% CI, 0.24-0.70) were more likely to have a higher absolute risk of atherosclerotic CVD. Similarly, individuals with GDM (8.7% vs 1.4%, adjusted risk ratio [RR] 2.02, 95% CI, 1.14-3.59), hypertensive disorder of pregnancy (4.4% vs 1.4%, adjusted RR 1.91, 95% CI, 1.17-3.13), and preterm birth (5.0% vs 1.5%, adjusted RR 2.26, 95% CI, 1.30-3.93) were more likely to have a high predicted risk of atherosclerotic CVD. A greater number of adverse pregnancy outcomes within the first birth was associated with progressively greater risks, including per 1% atherosclerotic CVD risk (one adverse pregnancy outcome: 4.86 vs 4.09, adjusted ß=0.59, 95% CI, 0.43-0.75; two or more adverse pregnancy outcomes: 5.51 vs 4.09, adjusted ß=1.16, 95% CI, 0.82-1.50), and a high predicted risk of atherosclerotic CVD (one adverse pregnancy outcome: 3.8% vs 1.0%, adjusted RR 2.33, 95% CI, 1.40-3.88; two or more adverse pregnancy outcomes: 8.7 vs 1.0%, RR 3.43, 95% CI, 1.74-6.74). Small and large for gestational age were not consistently associated with a higher atherosclerotic CVD risk. CONCLUSION: Individuals who experienced adverse pregnancy outcomes in their first birth were more likely to have a higher predicted 30-year risk of CVD measured at 2-7 years after delivery. The magnitude of risk was higher with a greater number of adverse pregnancy outcomes experienced.
Assuntos
Resultado da Gravidez , Humanos , Feminino , Gravidez , Adulto , Resultado da Gravidez/epidemiologia , Estudos Prospectivos , Nascimento Prematuro/epidemiologia , Doenças Cardiovasculares/epidemiologia , Doenças Cardiovasculares/etiologia , Estudos Longitudinais , Hipertensão Induzida pela Gravidez/epidemiologia , Diabetes Gestacional/epidemiologia , Fatores de Risco , Recém-Nascido , Medição de RiscoRESUMO
Importance: There is no consensus regarding the best method for prediction of hypertensive disorders of pregnancy (HDP), including gestational hypertension and preeclampsia. Objective: To determine predictive ability in early pregnancy of large-scale proteomics for prediction of HDP. Design, Setting, and Participants: This was a nested case-control study, conducted in 2022 to 2023, using clinical data and plasma samples collected between 2010 and 2013 during the first trimester, with follow-up until pregnancy outcome. This multicenter observational study took place at 8 academic medical centers in the US. Nulliparous individuals during first-trimester clinical visits were included. Participants with HDP were selected as cases; controls were selected from those who delivered at or after 37 weeks without any HDP, preterm birth, or small-for-gestational-age infant. Age, self-reported race and ethnicity, body mass index, diabetes, health insurance, and fetal sex were available covariates. Exposures: Proteomics using an aptamer-based assay that included 6481 unique human proteins was performed on stored plasma. Covariates were used in predictive models. Main Outcomes and Measures: Prediction models were developed using the elastic net, and analyses were performed on a randomly partitioned training dataset comprising 80% of study participants, with the remaining 20% used as an independent testing dataset. Primary measure of predictive performance was area under the receiver operating characteristic curve (AUC). Results: This study included 753 HDP cases and 1097 controls with a mean (SD) age of 26.9 (5.5) years. Maternal race and ethnicity were 51 Asian (2.8%), 275 non-Hispanic Black (14.9%), 275 Hispanic (14.9%), 1161 non-Hispanic White (62.8% ), and 88 recorded as other (4.8%), which included those who did not identify according to these designations. The elastic net model, allowing for forced inclusion of prespecified covariates, was used to adjust protein-based models for clinical and demographic variables. Under this approach, no proteins were selected to augment the clinical and demographic covariates. The predictive performance of the resulting model was modest, with a training set AUC of 0.64 (95% CI, 0.61-0.67) and a test set AUC of 0.62 (95% CI, 0.56-0.68). Further adjustment for study site yielded only minimal changes in AUCs. Conclusions and Relevance: In this case-control study with detailed clinical data and stored plasma samples available in the first trimester, an aptamer-based proteomics panel did not meaningfully add to predictive utility over and above clinical and demographic factors that are routinely available.