Maternal congenital cardiac disease: evaluating prognostic scores and the location of the lesion

We sought to describe both maternal and neonatal outcomes in women with congenital heart disease delivering at a single academic medical center and compare three prognostic scoring systems - mWHO, ZAHARA, CARPREG, to determine the best predictor of cardiac event in pregnancy. We conducted a retrospective cohort study of pregnancies occurring in women with congenital heart disease who received prenatal care and delivered at Vanderbilt University Medical Center between 2009 - 2016. We evaluated the association of each cardiac risk system with postpartum cardiac event and calculated the sensitivity, specificity, PPV and NPV of each system. We also sought to compare the frequency of postpartum cardiac event and other maternal and neonatal morbidity markers with the type of predominant cardiac lesion (left/complex versus right). Student’s t-test and chi-square tests were used for analysis. A total of 81 pregnancies were included. Most patients were Caucasian (56, 69.1%), utilized public insurance (55, 67.9%) and had surgically repaired lesions (61, 75%). Most cardiac lesions were mWHO I/II (63, 77.8%). Seven patients had a postpartum cardiac event. Women with a mWHO ≥ 3 were more likely to have a postpartum cardiac event (RR 4.67, p = 0.03). Specificity was greatest with the ZAHARA system, 95%. Sensitivity was greatest for the CARPREG, 71%. PPV was low for all three scoring systems (mWHO 22%, CARPREG 16%, ZAHARA 33%). NPV was similar among the three systems (mWHO 95%, CARPREG 96%, ZAHARA 93%). Postpartum ICU admission (57% vs 42.9%) and overall cardiac events (57 vs 42.9%) occurred more frequently in women with right sided cardiac lesions (p = 0.40). Preterm birth (68.5% vs 31.5%, p = 0.16) and IUGR (68.5% vs 31.5%, p = 0.13) were more common in women with right sided cardiac lesions. A mWHO ≥ 3 increases risk for significant postpartum cardiac event. The mWHO, CARPREG and ZAHARA systems demonstrate the ability to rule out the possibility of a postpartum cardiac event with higher specificity and NPV. There is no evidence of association of location of cardiac lesion with maternal or neonatal morbidity; however, results are limited due to sample size. Additional research is needed to further evaluate the maternal and obstetrical risks in pregnancy in women with congenital heart disease.