key: cord-1034736-qk76szs0
authors: Vrsalovic, Mislav; Presecki, Ana Vrsalovic
title: Cardiac injury and mortality in COVID-19: a reappraisal
date: 2020-06-30
journal: J Infect
DOI: 10.1016/j.jinf.2020.06.068
sha: 4ecc4dfcf34c21e1b8850eea97137419bc7caeb5
doc_id: 1034736
cord_uid: qk76szs0

nan

ng/mL, and >1 ng/mL), and the troponin cut-off (0.1 ng/mL) was approximately 5-6 times higher compared to other studies included in the meta-analysis. 3 In our meta-analysis, cardiac injury was defined as serum troponin levels above the 99th percentile upper reference limit.

Kollias et al. pooled the hazard ratios reported in the study, but the method of pooling the data to "study-unique" HR was not explained in the letter. Meta-analysis is used to derive a consolidated estimate of individual studies, and inclusion of post hoc pooled estimates (instead of raw data analysis) may result in an increased heterogeneity. Moreover, in the study of Petrilli et al. a competing risk model for the mortality or hospice outcome was performed.

Hospital discharge was considered to be a competing risk, since mortality data were limited after that point unless the patient was readmitted into the hospital system. As such, the inclusion of this study in the meta-analysis (i.e. meta-analysis investigated the impact of cardiac injury in terms of mortality) can be questioned.

patients. 4 The aforementioned cohort study included both non-severe and severe COVID-19 patients. Consequently, epidemiological, demographic and clinical characteristics of hospitalized patients with severe and non-severe disease were compared. However, the multivariable Cox regression analysis that identified factors associated with death was performed in only 269 severe patients (median age 65 years, interquartile range 54-72 years), and the results cannot be applied to the entire cohort, as was stated in the letter.

Based on the aforementioned methodological issues we performed an additional sub-analysis that included the two largest studies from Wuhan, China, which included 940 patients with COVID-19 (Table 1) , accounting for the fact that both studies analysed hospitalized cases with severe and critical illness. 4,5 A total of 149 (16%) patients died during in-hospital stay, and positive cardiac troponin was present in 200 (21%) patients.

Meta-analysis of those studies that reported adjusted HR (using the generic inverse variance method and fixed effects model), showed a significant association between elevated troponin values and mortality (HR = 3.08; 95% CI 1.95 -4.87) (Fig. 1) , and no significant heterogeneity between studies was detected (Cochran Q = 0.44, p = 0.50).

In conclusion, troponin positivity is common and has an independent prognostic role in hospitalised severe COVID-19 patients. Consequently cardiac injury may serve as an additional risk stratification tool in daily clinical setting. The meta-analysis was conducted using the generic inverse variance method, and pooled HR was reported with 95% confidence interval (CI). There was no significant heterogeneity observed across studies (Cochran Q = 0.44, p = 0.50). 

Cardiac injury and prognosis in covid-19: methodological considerations and updated meta-analysis

Cardiac troponins predict mortality in patients with COVID-19: A meta-analysis of adjusted risk estimates

Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study

Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan

Characteristics and clinical significance of myocardial injury in patients with severe coronavirus disease

The authors declare no conflict of interest.Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.