Myocardial injury as a predictor of mortality and adverse outcomes in COVID-19

Background: The American College of Cardiology suggested physicians should only measure troponin and brain natriuretic peptide (BNP) if myocardial infarction or heart failure were suspected in people with COVID-19. We aimed to evaluate the use of biomarkers on admission to hospital and the impact on mortality and morbidity. Methods: Consecutive patients presenting with COVID-19(reverse transcription PCR positive) between Feb27-May20 2020 were included in this retrospective, observational, single-center study. Clinical information was collected on admission and during hospitalization by physicians and later analysed by specialist cardiology registrars. 1675 patients were PCR +ve with 1036 having a high sensitivity troponin T(hsTropT) on admission. 371(35.8%) patients were hs TropT negative(<15ng/L) and 664(64.1%) had evidence of myocardial injury on admission(hsTropT ≥15ng/L). Subsequently demographic details were compared, as well as primary outcomes of death, ICU admission and COVID severity. Secondary outcomes were ARDS, myocardial infarction (MI);comparison with other biomarkers: NT-proBNP, d-dimer, CRP,LDH and ferritin. Results: Demographic data revealed no significant increase in proportions of Black, Asian or ethnic minorities in the myocardial injury group, however, patients were older(74.9±13.5 v 54.7±13.7yrs;p <0.001) and had significantly more co-morbidities such as diabetes(37 v 13%), hypertension(34 v 29%), ischemic heart disease(16 v 2%), other cardiac conditions(59 v 5%), malignancy(11 v 1%), COPD(9 v 4%), CKD stage ≤3 (40 v 3%) (p <0.01). Mortality was significantly higher in the myocardial injury group, 302(45.5%) v 29(7.8%) p <0.001, as were secondary outcomes of critical COVID (47 v 19%;p<0.001), ARDS (20 v 4%;p<0.001), Type 1 MI (1.6 v 0.01%;p<0.01) and Type 2 MI (44 v 26%;p<0.001). Interestingly, ICU admission (19 v 23%;p=0.09), pulmonary embolism (11 v 6%;p=0.22), stroke (1.1 v 0.5%;p=0.05) did not reach significance. Analysis of bio-markers on admission (Fig 1.) demonstrated hs Trop T (AUC 0.75 CI 0.69-0.81) and NT-pro BNP (AUC 0.75 CI 0.69-0.81) had more sensitvity 83%;85% and specificty 52%;58%, respectively at predicting death than d-dimer, CRP, LDH and ferritin. Conclusion: Early detection of elevated hsTropT and NT-proBNP predicts mortality and morbidity in patient with COVID-19. Routine measurement of cardiac biomarkers should be considered in patients with COVID-19 at the time of hospital admission in order to optimise risk stratification and guide monitoring. (Figure Presented).