In-Hospital Mortality in Hemorrhagic Myocardial Infarction
Keyur Vora, Ankur Kalra, Chirag Shah, Kinjal Bhatt, Andreas Kumar, Tejas Pandya, Vishal Poptani, Shing Fai Chan, Dhirendra Pratap Singh, Nithya Jambunathan, Ramesh Subramanian, Khalid Youssef, Saravanan Kanakasabai, Robert Finney, Ankit A. Desai, Rolf P. Kreutz, Richard J. Kovacs, Subha V. Raman, Deepak L. Bhatt, Rohan Dharmakumar
Abstract
BACKGROUND: Advances in acute ST-elevation myocardial infarction (STEMI) care have substantially decreased in-hospital mortality; however, in absolute terms, in-hospital mortality still remains high. Reperfusion injury, particularly intramyocardial hemorrhage following primary percutaneous coronary intervention (PCI), is a major predictor of adverse cardiovascular outcomes in the long term, but whether it contributes to in-hospital mortality is not known. METHODS: We performed a multicenter study to investigate the use of post-PCI high-sensitivity cardiac troponin I (hs-cTn-I) as a diagnostic tool to identify hemorrhagic myocardial infarction (MI) by determining hourly hs-cTn-I thresholds (every hour up to 12 hours, and at 16, 20, 24, and 48 hours post-PCI). We then investigated the relationship between patients classified as having hemorrhagic MI based on post-PCI hs-cTn-I cutoff values and in-hospital mortality using STEMI registries containing information about 6180 patients across seven hospitals in a single large health system in the United States. RESULTS: We enrolled 154 patients in a discovery cohort and 53 patients in a validation cohort. Hemorrhagic MI was diagnosed by cardiac magnetic resonance imaging. Post-PCI hs-cTn-I cutoff values for the determination of hemorrhagic MI were time dependent, with a sensitivity greater than 0.91, a specificity greater than 0.86, and an area under the curve (AUC) greater than 0.92 over the first 10 hours post-PCI, decreasing to a sensitivity greater than>0.84, a specificity greater than 0.80, and an AUC greater than 0.84 thereafter. The STEMI registry analysis demonstrated that patients classified as having hemorrhagic MI based on hs-cTn-I cutoff values had a 2.81-fold greater risk for in-hospital mortality than those classified as having had nonhemorrhagic MI (adjusted odds ratio, 2.81; 95% confidence interval, 2.17 to 3.64). CONCLUSIONS: Post-PCI troponin kinetics may have the potential to diagnose hemorrhagic MI, which was associated with in-hospital mortality. (Funded by the National Institutes of Health National Heart, Lung, and Blood Institute (grant numbers HL133407, HL136578, and HL147133) and others; ClinicalTrials.gov ID, NCT05872308).