LAUSR

Increase in native T1 mapping values provides insight into presence of disease and its evolution. Despite the surge of evidence, the immediate clinical application of these techniques is complicated by several variants of T1 mapping sequences, whose diagnostic bioequivalence is not known. We undertook an intra-individual comparison of two native T1 measurements based on modified Look-Locker (MOLLI) schemes, MOLLI 3(2)3(2)5 (flip angle, FA 50°, T1-FFM) and MOLLI 5(3)3 (FA 35°, T1 Long) in 30 patients with myocarditis proven by left ventricular endomyocardial biopsy. We examined their agreement with high-sensitive troponin T (hs-TnT) levels, as the reference standard for myocyte injury. All patients underwent a routine cardiovascular magnetic resonance (CMR) scan using a 3-Tesla clinical scanner. Native T1 values were estimated using a septal region of interest (ROI) in a single mid ventricular short axis (SAX) slice. Areas of late gadolinium enhancement (LGE) were excluded from ROIs. Venous blood was sampled at the time of the CMR study and hs-TnT analyzed using standardized commercially available test kits. A p value <0.05 was considered statistically significant. Native T1 by T1-FFM was moderately associated with MOLLI 5(3)3 (FA 35°) (Pearsons r=0.877, p<0.0001), however the Bland Altman analysis showed a poor agreement with a fixed bias of 46ms (p 0.001) and evidence of proportional bias (p 0.008). There was a significant correlation between native T1-FFM and hs-TnT (rho=0,500, p 0.018), whereas T1 Long showed no association (rho=0,374, p 0,086). Although native T1 by T1 Long showed a moderate correlation with the native T1 FFM in patients with biopsy proven myocarditis, only T1 FFM had a significant correlation with hsTnT suggesting a greater sensitivity for myocardial injury.