ADULTS WITH REPAIRED COARCTATION: REFINING THE NEED FOR MRI IMAGING

Abstract

Guidelines for the management of adult congenital heart disease recommend evaluation of patients with repaired coarctation (rCoA) by MRI or CT at least once in adulthood and repeated at intervals of 5 years or less. The incidence of pathology identified by this approach is uncertain. In an era of rationalization of healthcare costs, it is desirable to identify a low-risk subset of patients that may not require advanced tomographic imaging. Our objective was to determine the yield of MRI in detecting recoarctation or aneurysm in rCoA. We hypothesized that clinical assessment could identify patients at high risk for recoarctation or aneurysm, and conversely that prediction rules could be developed to identify those at low risk of complications. We performed a retrospective study in adults post surgical repair of aortic coarctation who had undergone MRI imaging. Recoarctation was defined as a ratio of the minimal aortic diameter/aortic diameter at the diaphragm of <0.6. Aneurysm was defined as a ratio of the widest diameter of the proximal descending aorta/aortic diameter at the diaphragm of >1.5. An optimal repair was defined as no imaging evidence of recoarctation or aneurysm. 74 consecutive patients with rCoA underwent MRI, mean age 32 years (range 18-69), 60% male. Recoarctation or aneurysm was identified in 17 patients (23%). Clinical predictors of recoarctation or aneurysm were upper-lower blood pressure gradient (ULG)>20 mmHg (OR 9.5, 95% CI 2.1- 43.7, p=0.004) and a history of reintervention after coarctation repair (OR 4.7, 1.4- 15.6, p=0.01). A prediction rule combining clinical factors (repair before age 5 without need for reintervention, no Dacron patch, no ULG and no hypertension) had a positive predictive value of 95% (95% CI 77%- 100%) for identifying patients with an optimal repair. In our cohort, 22 patients satisfied the prediction rule, representing 30% of the sample. Routine MRI imaging after repair of aortic coarctation is of moderate yield with 23% of patients having recoarctation or aneurysm. The strongest predictors of these complications are ULG>20 mmHg and a history of reintervention. Imaging should routinely be performed in such patients. Our clinical risk predictor (repair before age 5 without need for intervention, absence of Dacron patch, absence of ULG and absence of hypertension) has excellent predictive value for identifying an optimal repair. This tool has the potential to identify a subset of patients who are at low risk of complications and may not require imaging.