ABSTRACT
Objective To compare the application of Goldman , CARPREG and ZAHARA in predicting cardiac outcomes in pregnant women with congenital heart disease. Methods A retrospective analysis of 144 pregnant women with CHD , enrolled in The Third Affiliated Hospital of Guangzhou Medical University between Jan 1992 and June 2015 was conducted. The mortality and the incidence of cardiac events were counted by three risk estimation methods. The sensitivity and calibration of three systems were compared by using Hosmer-Lemeshow goodness of-fit analysis and Aera Under Curve (AUC). Results In terms of mortality rates, the sensitivity of CARPREG was higher than that of Goldman (AUC:0.874 and 0.889 respectively;P = 0.568 >0.05);. In terms of the incidence of cardiac events, the sensitivity of ZAHARA was higher than that of Goldman (AUC:0.912 and 0.773 respectively;P = 0.063 > 0.05); Conclusions All 3 risk estimation methods have good performance in sensitivity of predicting the incidence of cardiac events and mortality in pregnant women with CHD. The ZAHARA has a better calibration in predicting the incidence of cardiac events , and we recommend ZAHARA as the first choice of evaluation system for this kind of patients.