Background T1 and T2 cardiac magnetic resonance (CMR) mapping methods have shown promise for infarct characterisation in patients with acute ST-elevation myocardial infarction (STEMI). Non-ST elevation MI is typically a sub-acute problem with infarct characteristics that are less readily defined. We prospectively studied the diagnostic accuracy of two novel (T1, T2 mapping) and one established (T2 STIR) CMR methods for imaging the ischaemic area-at-risk (AAR) in patients with a recent NSTEMI.

Methods NSTEMI patients underwent contrast-enhanced CMR at 3.0 Tesla (T) after percutaneous coronary intervention. The presence/extent of infarction was assessed with late gadolinium enhancement imaging (Gadovist, 0.1 mmol/kg). The infarct-related territory (IRA) was identified independently using a combination of angiographic, ECG and clinical findings. AAR was assessed with T1, T2 and T2 STIR methods by 2 observers who were blind to all of the clinical data. Comparisons were made between CMR and clinical findings.

Results Seventy-three NSTEMI patients (mean age 57 ± 10 yrs, 78% male) underwent 3.0 T MRI. The mean infarct size was 5.5 ± 7.2% of left ventricular (LV) volume. The AAR T1 and T2 times were 1323 ± 68 ms and 57 ± 5 ms, respectively. The extent of AAR (% of LV volume) estimated with T1 (15.8 ± 10.6%) and T2 maps (16.0 ± 11.8%) was similar (p = 0.838), and moderately well correlated (r = 0.82, p < 0.001). The 95% limits of agreement for mean AAR estimated with T1 versus T2 maps were -13% and 13%.

Mean AAR estimated with T2 STIR (7.8 ± 11.6%) was lower than that estimated with T1 (p < 0.001) or T2 maps (p < 0.001). There were moderate correlations between AAR estimated with T1 maps vs. T2 STIR (r = 0.54, p < 0.001), and AAR estimated with T2 maps vs. T2 STIR (r = 0.46, p < 0.001). The 95% limits of agreement for mean myocardial AAR estimated with T1 maps vs. T2 STIR were -28% and 12%, and for T2 maps vs. T2 STIR -32% and 16%.

The IRA was correctly identified in 52 patients (71%) when with T1 CMR, 56 (77%) with T2 CMR, and 32 (44%) with T2 STIR CMR. The diagnostic accuracies of T1 and T2 CMR for identification of the IRA were similar (p = 0.125) whereas T1 CMR and T2 CMR had higher diagnostic accuracy vs. T2 STIR (both p < 0.001). Inter-observer agreement for IRA assignment was moderately high when analysed with T1 (κ = 0.790, p < 0.001) and T2 (κ = 0.794, p < 0.001) maps, but low with T2 STIR (κ = 0.555, p < 0.001).

Conclusion T1 and T2 maps have higher diagnostic accuracy than T2 STIR maps, implying superior clinical utility with T1 and T2 CMR for infarct characterisation in NSTEMI patients.