Early diagnostic and prognostic value of myocardial strain derived from cardiovascular magnetic resonance in patients with cardiac amyloidosis.

BACKGROUND

Cardiac amyloidosis (CA) is one of the causes of heart failure with preserved ejection fraction. Cardiac magnetic resonance (CMR) with late gadolinium enhancement (LGE) and extracellular volume (ECV) fractions is a preferred method to identify CA. However, the requirement of contrast limits its use in renal deficiency patients. Myocardial strain is a promising method without contrast. We sought to assess the early diagnostic and prognostic value of strain.

METHODS

This retrospective study enrolled 31 patients with systemic amyloidosis (SA) in Peking University First Hospital from January 2014 to January 2019. The patients were categorized into three groups, including 11 CA patients with left ventricular hypertrophy (CA-LVH group), 9 CA patients without LVH (CA-NLVH group), and 11 patients with extracardiac SA (SA group). Strain analysis was performed with CMR images. A least absolute shrinkage and selection operator (LASSO) was used to generate strain score. The receiver operating characteristic (ROC) curve was used to evaluate the early diagnostic efficacy of strain score and other single strain parameter. The primary endpoint was defined as death from all cause or rehospitalization for heart failure. A Cox proportional hazards model was used to assess the index value on the prognosis.

RESULTS

In CA patients, as the left ventricular wall thickens, the global and regional strain decrease significantly. A new strain score (strain score = 0.00893 × mid-septal circumferential peak strain + 0.02285 × apical radial peak strain + 0.1541 × apical circumferential peak strain + 0.33097 × epicardial circumferential average peak strain + 0.42232 × endocardial longitudinal average peak strain) generated using LASSO showed that the area under the ROC curve was 0.909. All the patients with outcome events were in CA groups, four were in CA-LVH group and one in CA-NLVH group. New York Heart Association (NYHA) grade [hazard ratio (HR) =14.29, 95% confidence interval (CI): 2.34-87.34, P<0.01], brain natriuretic peptide (HR =20.05, 95% CI: 2.21-182.36, P=0.008), cardiac injury biomarker (HR =11.59, 95% CI: 1.03-130.36, P=0.047), E/E' (mitral inflow to mitral relaxation velocity ratio) (HR =1.09, 95% CI: 1.00-1.18, P=0.040), end-systolic left ventricular volume (HR =1.04, 95% CI: 1.00-1.18, P=0.039) and LGE volume (HR =1.11, 95% CI: 1.02-1.20, P=0.012) positively correlate with events. Better renal function (HR =0.92, 95% CI: 0.86-0.98, P=0.011) and ejection fraction (HR =0.94, 95% CI: 0.88-0.99, P=0.027) appear to be protective factors. Although with no statistical difference, the strain damage had a tendency to predict poor prognosis, i.e., mid-ventricular circumferential strain with HR of 1.25 (95% CI: 1.0-1.57, P=0.050) and strain score with HR of 1.30 (95% CI: 0.98-1.73, P=0.067).

CONCLUSIONS

Myocardial strain decreased in CA patients. The integrated magnetic resonance imaging (MRI) strain score can serve as a useful tool to identify early myocardial involvement in amyloidosis. The strain damage had a tendency to predict poor prognosis.