The relationship between myocardial and hepatic T1 relaxometry on cardiovascular magnetic resonance imaging and clinical outcomes late after tetralogy of Fallot repair.

backgroundCardiovascular magnetic resonance imaging (CMR) can estimate diffuse myocardial fibrosis using native T1 mapping and extracellular volume fraction (ECV) quantification.

objectivesThis study aimed to explore associations between, and prognostic value of, myocardial and hepatic parametric imaging measurements after tetralogy of Fallot repair (rTOF).

methodsAdults with rTOF were prospectively enrolled along with healthy controls. Patients were categorized by residual hemodynamic load (pulmonary stenosis and/or pulmonary regurgitation). Available clinical data were recorded at study entry. Myocardial and hepatic native T1 were measured and ECV was calculated from pre/post-contrast T1 maps. The primary outcome was a composite of death, resuscitated sudden death, sustained ventricular tachycardia (VT), VT requiring invasive therapy, or heart failure requiring hospital admission. Multivariable models, adjusted for age and sex, examined the association between CMR measures and adverse cardiovascular events.

resultsIn total, n=181 patients (56% male; median age 33 years [IQR 26-47]) and 20 healthy age and sex-matched controls were studied. Hepatic and myocardial ECV measures correlated modestly (r=0.31, p<0.001). Hepatic native T1 was higher in rTOF patients as compared with controls (572±102 versus 529±12 msec, p=0.006) with statistically significant differences between hemodynamic load type (p=0.044). Right ventricular (RV) native T1 was increased in rTOF as compared with controls (1078±43 versus 997±43 msec, p<0.001) with a trend towards significant differences when stratified by hemodynamic load (p=0.091). Pathologic late gadolinium enhancement (LGE) was seen in the RV in 46% (n=83) and in the left ventricle (LV) in 37% (n=66). At median follow-up 2.05 years (IQR 0.75-2.81), the composite primary outcome occurred in 9% (n=17). Independent predictors of adverse events included RV end-systolic volume (HR 1.43, [95% CI 1.09-1.99], p=0.009), RV native T1 (HR 1.15, [95% CI 1.07-1.25], p<0.001), and LV LGE (HR 3.18, [95% CI 1.07-10.23], p=0.038); hepatic parametric imaging was not associated with outcome.

conclusionIn rTOF, myocardial and hepatic parametric imaging measures were modestly correlated and were increased as compared with controls. Diffuse myocardial fibrosis in the RV (native T1) and replacement fibrosis in the LV (LGE) were independently associated with major adverse cardiovascular outcomes, although hepatic relaxometry measures were not. These findings warrant further validation in larger populations with longer follow-up.