Abstract 20040: Risk Assessment in Pregnant Women With Congenital Heart Disease
Introduction: As more women with congenital heart disease (CHD) survive into their childbearing years, clinicians are faced with assessing pregnancy risk. Published risk stratification schemes have shown variable accuracy in predicting adverse cardiac outcomes in pregnancy for women with CHD.
Hypothesis: The modified World Health Organization (WHO) criteria are superior to the CARPREG and ZAHARA risk scores in predicting pregnancy outcomes in CHD.
Methods: This is a retrospective, single center study examining women ≥ 18 years with CHD who delivered between 1998 and 2014. CARPREG and ZAHARA risk scores and modified WHO criteria were applied to each subject. Cardiac events (i.e. at least one of the following: arrhythmia, heart failure or pulmonary edema, thromboembolic complications, transient ischemic attack, stroke, endocarditis, myocardial infarction, cardiac arrest, cardiac death) were assessed up to six months post-partum. Logistic regression analyses were performed and the area under the receiver operating curve (AUC) was calculated.
Results: There were 186 pregnancies (mean maternal age 32 years) and the most common CHD lesions were congenital aortic stenosis (17%), atrial septal defect (17%), and tetralogy of Fallot (12%). There were 31 cardiac events (17%) and no deaths. CARPREG and ZAHARA overestimated risk for the highest risk pregnancies: the estimated vs. observed event rates were 75% vs. 25% (N=4 at risk, p=0.021) for CARPREG and 70% vs. 38% (N=13 at risk, p=0.013) for ZAHARA. Figure 1 demonstrates the AUC for CARPREG (0.61), ZAHARA (0.78), and modified WHO (0.67) for cardiac events.
Conclusions: CARPREG and ZAHARA scores overestimate risk for higher risk pregnancies in women with CHD. In contrast to recent studies, ZAHARA risk score emerged as the model with the best discrimination for predicting adverse cardiac events. Further research is necessary to better refine risk stratification schemes for this complex patient population.
Author Disclosures: L.A. Goldberg: None. L.P. Gleason: None. E.S. Ruckdeschel: None. T. Bhamare: None. D. Drajpuch: None. A. Hirshberg: None. S.L. Partington: None. R. Rogers: None. S.K. Srinivas: None. N. Stokes: None. L. Tobin: None. L.D. Levine: None. Y.Y. Kim: None.
- © 2016 by American Heart Association, Inc.