Background—Late gadolinium-enhanced cardiovascular magnetic resonance imaging overestimates infarct size and underestimates recovery of dysfunctional segments acutely post ST-segment–elevation myocardial infarction. We assessed whether cardiovascular magnetic resonance imaging–derived segmental myocardial strain and markers of myocardial injury could improve the accuracy of late gadolinium-enhancement in predicting functional recovery after ST-segment–elevation myocardial infarction. Methods and Results—A total of 164 ST-segment–elevation myocardial infarction patients underwent acute (median 3 days) and follow-up (median 9.4 months) cardiovascular magnetic resonance imaging. Wall-motion scoring, feature tracking–derived circumferential strain (Ecc), segmental area of late gadolinium-enhancement (SEE), microvascular obstruction, intramyocardial hemorrhage, and salvage index (MSI) were assessed in 2624 segments. We used logistic regression analysis to identify markers that predict segmental recovery. At acute CMR 32% of segments were dysfunctional, and at follow-up CMR 19% were dysfunctional. Segmental function at acute imaging and odds ratio (OR) for functional recovery decreased with increasing SEE, although 33% of dysfunctional segments with SEE 76% to 100% improved. SEE was a strong predictor of functional improvement and normalization (area under the curve [AUC], 0.840 [95% confidence interval {CI}, 0.814–0.867]; OR, 0.97 [95% CI, 0.97–0.98] per +1% SEE for improvement and AUC, 0.887 [95% CI, 0.865–0.909]; OR, 0.95 [95% CI, 0.94–0.96] per +1% SEE for normalization). Its predictive accuracy for improvement, as assessed by areas under the receiver operator curves, was similar to that of MSI (AUC, 0.840 [95% CI, 0.809–0.872]; OR, 1.03 [95% CI, 1.02–1.03] per +1% MSI for improvement and AUC, 0.862 [0.832–0.891]; OR, 1.04 [95% CI, 1.03–1.04] per +1% SEE for normalization) and Ecc (AUC, 0.834 [95% CI, 0.807–0.862]; OR, 1.05 [95% CI, 1.03–1.07] per +1% MSI for improvement and AUC, 0.844 [95% CI, 0.818–0.871]; OR, 1.07 [95% CI, 1.05–1.10] per +1% SEE for normalization), and for normalization was greater than the other predictors. MSI and Ecc remained as significant after adjustment for SEE but provided no significant increase in predictive accuracy for improvement and normalization compared with SEE alone. MSI had similar predictive accuracy to SEE for functional recovery but was not assessable in 25% of patients. Microvascular obstruction provided no incremental predictive accuracy above SEE. Conclusions—This multicenter study confirms that SEE is a strong predictor of functional improvement post ST-segment–elevation myocardial infarction, but recovery occurs in a substantial proportion of dysfunctional segments with SEE >75%. Feature tracking–derived Ecc and MSI provide minimal incremental benefit to SEE in predicting segmental recovery. Clinical Trial Registration—URL: http://www.isrctn.com. Unique identifier: ISRCTN70913605.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)