Myocardial strain analysis by feature tracking cardiac magnetic resonance to identify subclinical cardiac dysfunction in patients with MINOCA

Figures

• 

  Fig. 1. Flow chart of participant selection and diagnosis of MINOCA. (A) Selection process of enrolled cases of the study. (B) Clinical diagnosis procedures of MINOCA and exclusion cases

• 

  Fig. 2. Multi-parametric images of a typical MINOCA case. Imaging from a 48-year-old female patient who presented with sudden chest tightness for 8 h and chest pain for 2 h. The value of peak cTnI was 1.79 ng/ml. (A–B) Coronary angiography indicated no stenosis in the coronary artery. (C) T2-STIR imaging demonstrated a hyperintense area of edematous myocardium in the lateral wall (arrow). (D–E) LGE short-axis and 4-chamber image demonstrated delayed enhancement in sub-endocardial myocardium of the lateral wall (arrows). (F–H) Voxel-derived FT of myocardium at end-diastole. (I) Area of infarcted areas (yellow). The proportion of the infarct area is 7.13%. (J) Area of edema area (blue). The proportion of the edema area is 10.05%. (K) GRS curve demonstrated a GRS of 35.8%. (L) GCS curve demonstrated a GCS of -16.3%. (M) GLS curve demonstrated a GLS of -14.9%

• 

  Fig. 3. MINOCA patients with normal routine MRI. Imaging from a 32-year-old male patient who presented with sudden chest pain for 5 h. The value of peak cTnI was 0.68 ng/ml. (A) The ECG at the time of admission indicated partial ST elevation. (B) One day after admission, ECG indicated inversion of T wave in lead II, III and AVF, indicating new inferior infarction. (C–D) Coronary angiography indicated no stenosis in the coronary artery. E T2-STIR images demonstrated no edema in LV myocardium. (F–G) LGE short-axis and 4-chamber image demonstrated no delayed enhancement in LV myocardium. (H) GRS curve demonstrated a GRS of 36.2%. (L) GCS curve demonstrated a GCS of -18.2%. (M) GLS curve demonstrated a GLS of -16.6%

• 

  Fig. 4. GLS and GLSR Boxplots in Patient Subgroups. (A) GLS in the three groups. The GLS in MINOCA patients with CMR abnormal was the lowest; the GLS in MINOCA patients with CMR normal was higher but still lower than controls. P < 0.05 for the comparison between any two groups. (B) GLSR in the three groups. The GLSR in MINOCA patients with CMR abnormal was the lowest; the GLSR in MINOCA patients with CMR normal was higher, but still lower than controls. P < 0.05 for comparison between any two groups (Negative sign in GLS and GLSR means myocardial shortening. A greater negative value indicates higher GLS and GLSR.)

• 

  Fig. 5. Scatter Plot of GLS with the Proportions of the Infarct and Edema Size. (A) The Y-axis indicates the values of GLS, and the X-axis indicates the proportion of infarct myocardial size to the total LV myocardial mass. GLS showed a positive linear correlation with the infarcted myocardium size. (B) The Y-axis indicates the values of GLS, and the X-axis indicates the proportion of edema myocardial size to the total LV myocardial mass. GLS showed a positive linear correlation with the edema myocardium size. (C) The Y-axis indicates the values of GLS, and the X-axis indicates the transmural extent grade of myocardial Infarction for each MINOCA patient. GLS showed a positive rank correlation with transmural extent

• 

  Fig. 6. ROC Curve of GLS Diagnosis of MINOCA. The area under the ROC curve was 0.968, P = 0.000; the maximum Youden’s index = 0.8 and the optimal cut-off point for GLS was <-16.9%