Diffusion tensor imaging in patients with suspected cardiomyopathies (RF_TH_201)

Diffusion Tensor Imaging in cardiovascular magnetic resonance (DTI-CMR) is a novel non-contrast imaging technique that can non-invasively assess myocardial architecture. This study investigates apparent diffusion coefficient (ADC) and fractional anisotropy (FA) derived from DTI-CMR in patients with cardiomyopathies and compares them with healthy volunteers.

Methods:

A motion-compensated, cardiac-gated, single-shot echo-planar imaging-based diffusion sequence was performed in patients with clinical suspicion of various cardiomyopathies and normal volunteers on a 1.5T scanner (Sola, Siemens) and 3T scanner (Vida, Siemens Healthineers). Short-axis views were acquired using a b-value of 350 s/mm2, 3 and 6 diffusion -encoding directions (3DD and 6DD), echo-time of 64 ms, repetition time of 100 ms, and GRAPPA acceleration factor of 2. The 6DD acquisition provides ADC and FA measurements as compared to 3DD, which provides only ADC measurements. Left ventricular myocardial contours were drawn manually on basal and mid short axis slices using a custom software developed in MATLAB to generate ADC and FA maps. Mean values were compared using two-sample t tests (p< 0.05). Multiple disease group analysis was performed using one-way analysis of variance (ANOVA).

Results:

We imaged 72 patients and 19 healthy controls, after excluding images for poor quality, a total of 55 patients and 19 controls were included for analysis. 42 patients and 19 controls were imaged on a 1.5T (Sola, Siemens) and 13 patients were imaged on a 3T (Vida, Siemens) scanner. Comparison of baseline characteristics, clinical variables and imaging variables between patients and healthy controls are shown in Table 1. The ADC values are significantly higher in the patient group compared to control group in both 6DD (1.71 vs. 1.45, p< 0.001) and 3DD (1.55 vs. 1.31, p=0.035). No significant difference in FA values between the two groups is observed (0.45 vs. 0.47, p=0.390). In patients, when comparing ADC values obtained from 6DD, those with diagnosis of active myocarditis (N=8) demonstrated the highest values as compared to HCM (N=6), other cardiomyopathies (N=20), no cardiomyopathy (N=5), and normal controls (N=17) (p < 0.001). No significant difference in FA values between these groups (Fig. 1). In regression analysis, there are no significant correlations in ADC values obtained from 6DD with age, sex, body mass index, race, LVM, LVMI, or LVEF as variables (p=0.08-0.94). ADC values obtained from 6DD may correlate with native T1 (r² = 0.20, p=0.020), T2 (r² = 0.28, p=0.005) and ECV (r² = 0.44, p< 0.001) mapping values, and the correlations are better than ADC values obtained from 3DD (Fig. 2).

Conclusion:

DTI-CMR imaging has the potential to detect pathological processes due to changes in myocardial microarchitecture. In our study, ADC values can be significantly elevated in various types of cardiomyopathies, especially active myocarditis, and correlate with tissue mapping results weakly/moderately. Future studies are needed to validate the use of this technique in characterizing different disease states.