Background Whilst international guidelines recognise hypertensive disorders of pregnancy (HDP) as a major risk factor for premature cardiovascular disease (CVD) in women, there is a paucity of recommendations for how to identify those with such high risk. We aimed to determine the predictors of CVD in women with a history of HDP.

Methods Eligible women were identified from the Aberdeen Study of Cardiovascular Health in Women (ASCHW) and European Prospective Investigation into Cancer (EPIC)-Norfolk prospective cohorts and they were followed for incident CVD events through record linkage using ICD 9/10 in both cohorts. Of 4,186 women with a history of HDP identified, 3,468 attended clinic assessment. Missing data were handled using multiple imputation. We examined the relationship between HDP and incident cardiovascular events by carrying out three separate univariate and multivariable logistic regression analyses: lifestyle questionnaire variable analysis, analysis including examination variables and plasma cardiovascular biomarker analysis. The final model consisted of statistically significant predictors (p-value<0.05) derived from the three analyses. Validity of the model was assessed by discrimination (c-statistic) and calibration (Hosmer-and-Lemeshow test).

Results Selected predictors for CVD in women with HDP in the final model were age over 49 years, no university education, high BMI, total cholesterol, triglyceride and plasma fibrinogen; usage of aspirin and lipid lowering medications; hypertension, family history of heart disease, repeated HDP exposure, and the cohort population. All predictors in the final model were statistically significant except total cholesterol levels. The risk factors which conferred the greatest odds ratios of CVD (≥2-fold odds) were: age beyond 49 years (1.99; 95% CI 1.57–2.54), hypertension (2.84; 95% CI 2.36–3.41), aspirin users (2.18; 95% CI 1.54–3.08), having morbidly obese BMI (2.16; 95% CI 1.54–3.04) and the EPIC-Norfolk population (12.57; 95% CI 9.85–16.04). Median AUC was 0.82 and calibration ranged from <0.001 to 0.003 in imputed datasets.

Conclusion We have identified significant predictors of CVD in women with a history of HDP. This suggests that women with a history of HDP should be followed up from the age of 49 years. Biomarkers such as triglyceride and fibrinogen should be monitored, particularly if women have hypertension, high BMI, a family history of CVD or have had repeated exposure to HDP. Further external validation work is recommended to confirm the clinical utility of the proposed predictors of CVD.