An electrocardiography score predicts heart failure hospitalization or death beyond that of cardiovascular magnetic resonance imaging.

The electrocardiogram (ECG) and cardiovascular magnetic resonance imaging (CMR) provide powerful prognostic information. The aim was to determine their relative prognostic value. Patients (n = 783) undergoing CMR and 12-lead ECG with a QRS duration < 120 ms were included. Prognosis scores for one-year event-free survival from hospitalization for heart failure or death were derived using continuous ECG or CMR measures, and multivariable logistic regression, and compared. Patients (median [interquartile range] age 55 [43-64] years, 44% female) had 155 events during 5.7 [4.4-6.6] years. The ECG prognosis score included (1) frontal plane QRS-T angle, and (2) heart rate corrected QT duration (QTc) (log-rank 55). The CMR prognosis score included (1) global longitudinal strain, and (2) extracellular volume fraction (log-rank 85). The combination of positive scores for both ECG and CMR yielded the highest prognostic value (log-rank 105). Multivariable analysis showed an association with outcomes for both the ECG prognosis score (log-rank 8.4, hazard ratio [95% confidence interval] 1.29 [1.09-1.54]) and the CMR prognosis score (log-rank 47, hazard ratio 1.90 [1.58-2.28]). An ECG prognosis score predicted outcomes independently of CMR. Combining the results of ECG and CMR using both prognosis scores improved the overall prognostic performance.

Automated In-Line Artificial Intelligence Measured Global Longitudinal Shortening and Mitral Annular Plane Systolic Excursion: Reproducibility and Prognostic Significance.

Background Global longitudinal shortening (GL-Shortening) and the mitral annular plane systolic excursion (MAPSE) are known markers in heart failure patients, but measurement may be subjective and less frequently reported because of the lack of automated analysis. Therefore, a validated, automated artificial intelligence (AI) solution can be of strong clinical interest. Methods and Results The model was implemented on cardiac magnetic resonance scanners with automated in-line processing. Reproducibility was evaluated in a scan-rescan data set (n=160 patients). The prognostic association with adverse events (death or hospitalization for heart failure) was evaluated in a large patient cohort (n=1572) and compared with feature tracking global longitudinal strain measured manually by experts. Automated processing took ≈1.1 seconds for a typical case. On the scan-rescan data set, the model exceeded the precision of human expert (coefficient of variation 7.2% versus 11.1% for GL-Shortening, P=0.0024; 6.5% versus 9.1% for MAPSE, P=0.0124). The minimal detectable change at 90% power was 2.53 percentage points for GL-Shortening and 1.84 mm for MAPSE. AI GL-Shortening correlated well with manual global longitudinal strain (R2=0.85). AI MAPSE had the strongest association with outcomes (χ2, 255; hazard ratio [HR], 2.5 [95% CI, 2.2-2.8]), compared with AI GL-Shortening (χ2, 197; HR, 2.1 [95% CI,1.9-2.4]), manual global longitudinal strain (χ2, 192; HR, 2.1 [95% CI, 1.9-2.3]), and left ventricular ejection fraction (χ2, 147; HR, 1.8 [95% CI, 1.6-1.9]), with P<0.001 for all. Conclusions Automated in-line AI-measured MAPSE and GL-Shortening can deliver immediate and highly reproducible results during cardiac magnetic resonance scanning. These results have strong associations with adverse outcomes that exceed those of global longitudinal strain and left ventricular ejection fraction.

Extracellular Volume and Global Longitudinal Strain Both Associate With Outcomes But Correlate Minimally.

OBJECTIVES: This study examined how extracellular volume (ECV) and global longitudinal strain (GLS) relate to each other and to outcomes. BACKGROUND: Among myriad changes occurring in diseased myocardium, left ventricular imaging metrics of either the interstitium (e.g., ECV) or contractile function (e.g., GLS) may consistently associate with adverse outcomes yet correlate minimally with each other. This scenario suggests that ECV and GLS potentially represent distinct domains of cardiac vulnerability. METHODS: The study included 1,578 patients referred for cardiovascular magnetic resonance (CMR) without amyloidosis, and it quantified how ECV associated with GLS in linear regression models. ECV and GLS were then compared in their associations with incident outcomes (death and hospitalization for heart failure). RESULTS: ECV and GLS correlated minimally (R2 = 0.04). Over a median follow-up of 5.6 years, 339 patients experienced adverse events (149 hospitalizations for heart failure, 253 deaths, and 63 with both). GLS (univariable hazard ratio: 2.07 per 5% increment; 95% CI: 1.86 to 2.29) and ECV (univariable hazard ratio: 1.66 per 4% increment; 95% CI: 1.51 to 1.82) were principal variables associating with outcomes in univariable and multivariable Cox regression models. Similar results were observed in several clinically important subgroups. In the whole cohort, ECV added prognostic value beyond GLS in univariable and multivariable Cox regression models. CONCLUSIONS: GLS and ECV may represent principal but distinct domains of cardiac vulnerability, perhaps reflecting their distinct cellular origins. Whether combining ECV and GLS may advance pathophysiological understanding for a given patient, optimize risk stratification, and foster personalized medicine by targeted therapeutics requires further investigation.

Extracellular Volume Associates With Outcomes More Strongly Than Native or Post-Contrast Myocardial T1.

OBJECTIVES: Because risk stratification data represents a key domain of biomarker validation, we compared associations between outcomes and various cardiovascular magnetic resonance (CMR) metrics quantifying myocardial fibrosis (MF) in noninfarcted myocardium: extracellular volume fraction (ECV), native T1, post-contrast T1, and partition coefficient. BACKGROUND: MF associates with vulnerability to adverse events (e.g., mortality and hospitalization for heart failure [HHF]), but investigators still debate its optimal measurement; most histological validation data show strongest ECV correlations with MF. METHODS: We enrolled 1,714 consecutive patients without amyloidosis or hypertrophic cardiomyopathy from a single CMR referral center serving an integrated healthcare network. We measured T1 (MOdified Look-Locker Inversion recovery [MOLLI]) in nonenhanced myocardium, averaged from 2 short-axis slices (basal and mid) before and 15 to 20 min after a gadolinium contrast bolus. We compared chi-square test values from CMR MF measures in univariable and multivariable Cox regression models. We assessed "dose-response" relationships in Kaplan-Meier curves using log-rank statistics for quartile strata. We also computed net reclassification improvement (NRI) and integrated discrimination improvement (IDI for Cox models with ECV vs. native T1). RESULTS: Over a median of 5.6 years, 374 events occurred after CMR (162 HHF events and 279 deaths, 67 with both). ECV yielded the best separation of Kaplan-Meier curves and the highest log-rank statistics. In univariable and multivariable models, ECV associated most strongly with outcomes, demonstrating the highest chi-square test values. Native T1 or post-contrast T1 did not associate with outcomes in the multivariable model. ECV provided added prognostic value to models with native T1, for example, in multivariable models IDI = 0.0037 (95% confidence interval [CI]: 0.0009 to 0.0071), p = 0.02; NRI = 0.151 (95% CI: 0.022 to 0.292), p = 0.04. CONCLUSIONS: Analogous to histological previously published validation data, ECV myocardial fibrosis measures exhibited more robust associations with outcomes than other surrogate CMR MF measures. Superior risk stratification by ECV supports claims that ECV optimally measures MF in noninfarcted myocardium.