Authors:

M Montenbruck¹ , S Kelle² , S Esch³ , F Andre⁴ , G Korosoglou⁵ , H Steen¹ , ¹Marienhospital, Cardiology / Cardiac imaging - Hamburg - Germany , ²Deutsches Herzzentrum Berlin - Berlin - Germany , ³Marien hospital - Hamburg - Germany , ⁴University Hospital of Heidelberg - Heidelberg - Germany , ⁵Academic Teaching Hospital Weinheim, Department of internal medicine / Cardiology and vascular medicine - Hamburg - Germany ,

Citation:

Background:

Global longitudinal strain has become an alternative to ejection fraction in identifying reduced cardiac function in the left (LV) or right (RV) ventricles.  However, these global metrics are not able to characterize patients in which the heart compensates for regional dysfunction.  More sensitive metrics are needed to detect subclinical regional dysfunction before cardiac remodeling results in changes in ejection fraction (EF) and global longitudinal strain (GLS).  Fast-SENC intramyocardial strain (fSENC) is a unique cardiac magnetic resonance imaging (CMR) modality that measures intramyocardial contraction in 1 heartbeat per image plane. This prospective registry compares segmental fSENC to global metrics GLS and LVEF based on modified ACC/AHA Heart Failure Stage that categorized differing levels of structural heart disease for Stage B and C.

Methods:

A single center, prospective registry of MRI scans acquired with a 1.5T scanner were evaluated for conventional CMR diagnostics including biventricular EF, volumes and mass.  In addition, fSENC scans were acquired and processed with the MyoStrain software to quantify intramyocardial LV & RV strain.  Three short axis scans (basal, midventricular, & apical) were used to calculate strain in 16 LV & 6 RV longitudinal segments while three long axis scans (2-, 3- & 4-chamber) were used to calculate 21 LV & 5 RV circumferential segments.  All metrics were compared based on ACC/AHA Heart Failure Stage determined by full CMR exam.

Results:

A total of 977 scans in 779 patients were included in the study; this population included 210 myocarditis, 46 dilated cardiomyopathy, and 30 ischemic cardiomyopathy cases.  Patients had an average (± stdev) age of 55 (17) yrs and BMI of 26 (5) kg/m2; 48% had arterial hypertension, 12% diabetes mellitus, 33% valve disease, 24% cancer, 7% atrial fibrillation, 13% pulmonary disease, 5% left bundle branch block, 35% hypercholesterolemia, and 24% coronary artery disease. Figure 1 shows the relationship between segmental strain, calculated as the percent of normal LV segments (longitudinal & circumferential) based on intramyocardial fSENC < -17%, versus LVEF based on ACC/AHA Heart Failure stage.

Conclusion:

Segmental fSENC detects subclinical LV dysfunction well before changes in EF or GLS.  Incorporating both longitudinal and circumferential components into segmental fSENC metrics provides an alternative metric that shows consistent changes in heart failure progression irrespective of risk factors or underlying cardiac disease.