Detection of acute myocardial injury with low field (0.55T) Cardiovascular Magnetic Resonance Imaging

Background: The diagnosis of acute myocardial injury with CMR relies on the detection of myocardial edema and injury with T2-weighted short-tau inversion recovery (T2w-STIR), T2 mapping and LGE, respectively. Clinical CMRs is almost exclusively performed on 1.5T or 3T platforms. Despite its high diagnostic accuracy, hardware costs and technical requirements limit global availability. Affordable lower field strength (0.55T) platforms have been developed to broaden access to CMR.¹ However, these have lower signal-to-noise ratio, which may impact clinical utility. We sought to determine the diagnostic accuracy of T2w-STIR and T2 mapping at 0.55T compared to 1.5T.

Methods: Ten patients with clinically suspected acute myocardial injury (median [IQR] age 33 [39-47] and peak median [IQR] troponin I 696 [359-1415] ng/L) underwent T2w-STIR and T2 mapping on 0.55T and 1.5T scanners (MAGNETOM FreeMax and MAGNETOM Aera, Siemens Healthineers, Forchheim, Germany) on the same day. Additionally, LGE images were performed on 1.5T. T2 mapping at 0.55T was performed using a research sequence with one additional average acquisition and one less recovery heart beat to enable scanning in a single breath hold. The overall clinical presentation (symptoms, ECG changes, troponin and angiography results) was regarded as the reference standard. All subjects provided written informed consent.





T2 mapping and T2w-STIR image interpretations were evaluated blinded to clinical data. Qualitative analysis was performed using a 4-point Likert scale (1-non-diagnostic; 2-fair; 3-good; 4-excellent).²

Results: The reference range for T2 mapping on 0.55T in 21 healthy volunteers was 56-84ms.





The median (IQR) time from peak troponin to scan was 4 (4-6) days. Acute myocarditis was diagnosed in 7 patients, acute myocardial infarction in 1 patient and 2 patients had normal scans. Image quality for T2 mapping on 0.55T was diagnostic in 100% of the images (50% excellent, 50% good), whilst 90% of the images were diagnostic for T2w-STIR imaging (40% excellent, 30% good, 20% fair, 10% non-diagnostic).





The sensitivity (95% confidence interval [CI]) of T2 mapping for detecting myocardial edema at 0.55T was 89% (57-99%), with indeterminate specificity (no true negatives). The sensitivity (95% CI) of T2w-STIR was 56% (27-81%), specificity 100% (5-100%) with the wide CI reflecting small sample size. The sensitivity and specificity (95% CI) of the T2 mapping sequence for detecting myocardial edema on 1.5T was 78% (45-96%) and 100% (5-100%). Similarly, the sensitivity and specificity (95% CI) of T2w-STIR was 78% (45-96%) and 100% (5-100%).

Conclusion: We have demonstrated the feasibility of T2 mapping and T2w-STIR for detecting myocardial edema on 0.55T. T2 mapping had high sensitivity for detecting myocardial edema. The sensitivity of T2w-STIR was lower than T2 mapping, which has been demonstrated.³ This early proof-of-concept study is a promising step towards the widespread use of 0.55T for detecting acute cardiovascular pathology.

Figure. Acute myocarditis 37-year-old-male presented with fever, chest pain, and raised troponin I of 396ng/L (normal range <13 ng/L). Myocardial edema is seen in the epicardial to mid wall of the basal to mid-inferior and lateral walls. T2 mapping revealed increased T2 values ranging from 85-95ms (normal range <84ms) at 0.55T and 61-74ms (normal range <55ms) at 1.5T. T2 STIR images reveal increased epicardial to mid-myocardial signal intensity in the basal to mid inferior and lateral walls. LGE images reveals enhancement in the corresponding areas, confirming the diagnosis of acute myocarditis. CMR= cardiovascular magnetic resonance; ECG= electrocardiogram; LGE= late gadolinium enhancement; STIR= short-tau inversion recovery