Native T1-mapping using cardiovascular magnetic resonance detects myocardium at risk during the first week following myocardial infarction in a swine model and in patients - comparison to contrast-enhanced cine steady-state free precession.

BACKGROUND

Myocardium at risk (MaR) can be evaluated by cardiovascular magnetic resonance (CMR) imaging using contrast-enhanced steady state free precession (CE-SSFP) in patients after ST-elevation myocardial infarction (STEMI). However, CE-SSFP utilizes gadolinium contrast, which is contraindicated in patients with severe renal insufficiency. Native T1-mapping is a non-contrast CMR method which has been shown feasible in assessing MaR, enabling patients with gadolinium contrast contraindications to be examined. However, native T1-mapping data have been presented in the sub-acute phase suggesting to also depict infarct size (IS), as assessed by late gadolinium enhancement (LGE). Therefore, it is unclear whether native T1-mapping depicts MaR or IS during the first week after reperfusion. We hypothesized that native T1-mapping agrees with MaR as assessed by CE-SSFP and overestimates IS as assessed by LGE in an experimental pig model and in patients during the first week after STEMI.

METHODS

A retrospective analysis was performed using CMR images from an infarct/reperfusion experimental pig model. CMR imaging was performed at 2 h, 24 h and 7 days after reperfusion in a serially imaged group (n = 7) and at 4 days in a single-timepoint imaged group (n = 4). Also, STEMI patients with a single vessel LAD occlusion (n = 11) were CMR imaged between 3 to 7 days after reperfusion. Native T1-mapping MOLLI, CE-SSFP and LGE were acquired for each scan in both animals and patients. In animals, images with an additional T1-mapping sequence, SASHA, were acquired. Enhanced areas on T1-maps, CE-SSFP and LGE images were quantified and compared.

RESULTS

In pigs, native T1-mapping MOLLI agreed with CE-SSFP in the single-timepoint- and serially imaged groups (bias: 0.3 ± 6.6% (mean ± 2SD), and 0.9 ± 18%), respectively. Native T1-mapping SASHA also agreed with CE-SSFP in the serially imaged group (bias: -0.1 ± 18%). However, MOLLI overestimated IS by LGE in pigs in the serially- and single-timepoint imaged groups (bias: 21 ± 26%, and 18 ± 17%), respectively. Similar results were seen in patients (MOLLI vs. CE-SSFP: 0.8 ± 7.5%, and MOLLI vs. LGE: 31 ± 22%).

CONCLUSION

Our findings suggest that native T1-mapping agrees with CE-SSFP during the first week after myocardial infarction when evaluating MaR. Also, native T1-mapping overestimates the LGE hyperintense area, indicating that native T1-mapping does not primarily depict infarct size.