Strain parameters for predicting the prognosis of non-ischemic dilated cardiomyopathy using cardiovascular magnetic resonance tissue feature tracking.

ABSTRACT:

Background

A considerable number of non-ischemic dilated cardiomyopathy (NDCM) patients had been found to have normalized left ventricular (LV) size and systolic function with tailored medical treatments. Accordingly, we aimed to evaluate if strain parameters assessed by cardiovascular magnetic resonance (CMR) feature tracking (FT) analysis could predict the NDCM recovery.

Methods

79 newly diagnosed NDCM patients who underwent baseline and follow-up CMR scans were enrolled. Recovery was defined as a current normalized LV size and systolic function evaluated by CMR.

Results

Among 79 patients, 21 (27%) were confirmed recovered at a median follow-up of 36 months. Recovered patients presented with faster heart rates (HR) and larger body surface area (BSA) at baseline (P < 0.05). Compared to unrecovered patients, recovered pateints had a higher LV apical radial strain divided by basal radial strain (RS_api/bas) and a lower standard deviation of time to peak radial strain in 16 segments of the LV (SD16-TTPRS). According to a multivariate logistic regression model, RS_api/bas (P = 0.035) and SD16-TTPRS (P = 0.012) resulted as significant predictors for differentiation of recovered from unrecovered patients. The sensitivity and specificity of RS_api/bas and SD16-TTPRS for predicting recovered conditions were 76%, 67%, and 91%, 59%, with the area under the curve of 0.75 and 0.76, respectively. Further, Kaplan Meier survival analysis showed that patients with RS_api/bas ≥ 0.95% and SD16-FTPRS ≤ 111 ms had the highest recovery rate (65%, P = 0.027).

Conclusions

RS_api/bas and CMR SD16-TTPRS may be used as non-invasive parameters for predicting LV recovery in NDCM. This finding may be beneficial for subsequent treatments and prognosis of NDCM patients. Registration number: ChiCTR-POC-17012586.

SUBMITTER: Gao C 

PROVIDER: S-EPMC7958458 | biostudies-literature | 2021 Mar

REPOSITORIES: biostudies-literature