Article Text

Original research
Risk stratification and prognosis prediction using cardiac biomarkers in COVID-19: a single-centre retrospective cohort study
  1. Madoka Sano1,
  2. Toshiaki Toyota1,
  3. Takeshi Morimoto2,
  4. Yu Noguchi3,
  5. Ryo Shigeno4,
  6. Ryosuke Murai1,
  7. Taiji Okada1,
  8. Yasuhiro Sasaki1,
  9. Tomohiko Taniguchi1,
  10. Kitae Kim1,
  11. Atsushi Kobori1,
  12. Natsuhiko Ehara1,
  13. Makoto Kinoshita1,
  14. Asako Doi5,
  15. Keisuke Tomii6,
  16. Yasuki Kihara7,
  17. Yutaka Furukawa1
  1. 1 Department of Cardiovascular Medicine, Kobe City Medical Center General Hospital, Kobe, Japan
  2. 2 Center for Clinical Research and Innovation, Kobe City Medical Center General Hospital, Kobe, Japan
  3. 3 Department of Cardiovascular Medicine, Tenri Hospital, Tenri, Japan
  4. 4 Department of Cardiovascular Medicine, The Sakakibara Heart Institute of Okayama, Okayama, Japan
  5. 5 Department of Infectious disease, Kobe City Medical Center General Hospital, Kobe, Japan
  6. 6 Department of Respiratory Medicine, Kobe City Medical Center General Hospital, Kobe, Japan
  7. 7 Kobe City Medical Center General Hospital, Kobe, Japan
  1. Correspondence to Dr Toshiaki Toyota; totoyota{at}kuhp.kyoto-u.ac.jp

Abstract

Objective There is a need for a robust tool to stratify the patient’s risk with COVID-19. We assessed the prognostic values of cardiac biomarkers for COVID-19 patients.

Methods This is a single-centre retrospective cohort study. Consecutive laboratory-confirmed COVID-19 patients admitted to the Kobe City Medical Center General Hospital from July 2020 to September 2021 were included. We obtained cardiac biomarker values from electronic health records and institutional blood banks. We stratified patients with cardiac biomarkers as high-sensitive troponin I (hsTnI), N-terminal pro-B-type natriuretic peptide (NT-proBNP), creatine kinase (CK) and CK myocardial band (CK-MB), using the clinically relevant thresholds. Prespecified primary outcome measure was all-cause death.

Results A total of 917 patients were included. hsTnI, NT-proBNP, CK and CK-MB were associated with the significantly higher cumulative 30-day incidence of all-cause death (hsTnI: <5.0 ng/L group; 4.3%, 5.0 ng/L–99%ile upper reference limit (URL) group; 8.8% and ≥99% ile URL group; 25.2%, p<0.001. NT-proBNP: <125 pg/mL group; 5.3%, 125–900 pg/mL group; 10.5% and ≥900 pg/mL group; 31.9%, p<0.001. CK: <upper normal limit (UNL) group; 10.6%, UNL to 3 times of UNL group; 16.4% and ≥3 times of UNL group; 23.5%, p<0.001. CK-MB: <UNL group; 7.8%, UNL to 3 times of UNL group; 20.4% and ≥3 times of UNL group; 38.9%, p<0.001). The adjusted risk for all-cause death remained significant for each threshold of cardiac biomarkers.

Conclusions Elevation of cardiac biomarkers was associated with poor prognosis of COVID-19 patients.

  • COVID-19
  • REGISTRIES
  • Cardiology
  • Risk Factors

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • This study included relatively large real-world databases on COVID-19 that systematically evaluated the effects of cardiac biomarkers on clinical outcomes.

  • This study evaluated the clinical outcomes in patients with COVID-19, including cardiovascular complications, with a median follow-up period of 31 days; thus, further studies evaluating long-term prognosis are needed.

  • The current study population included patients with relatively severe COVID-19 in in-hospital settings.

  • The severity and prognosis of COVID-19 in our study might differ from those of the current COVID-19, caused by recent prevalent genetic variants of SARS-CoV-2 and penetration of vaccination for SARS-CoV-2.

  • Few patients were vaccinated for SARS-CoV-2 in the study population, thus the results can be adoptable for populations who have not vaccinated yet.

Introduction

The global pandemic of COVID-19 caused by SARS-CoV-2 has affected hundreds of millions of people since December 2019 and raised problems not only clinical but social problems, including the disposition of COVID-19 patients requiring advanced medical management, taking balance into the capacity of healthcare providers. Although the most common manifestation of COVID-19 is respiratory symptoms, including pneumonia and acute respiratory distress syndrome (ARDS), myocardial injury detected by the presence of elevated cardiac troponin levels is frequently seen in severe COVID-19 patients, and it has been reported to be associated with poor prognosis and might be helpful to stratify patients with COVID-19.1–6 However, the data regarding the incidences of myocardial injury and cardiovascular comorbidities or analysis on risk stratification in COVID-19 patients are limited. This study aims to assess the incidence of cardiovascular complications in patients with COVID-19 from the real-world data and to explore the prognostic values of cardiac biomarkers for COVID-19 patients.

Methods

Study design and population

The real-world evaluation of clinical outcomes of COVID-19 focused on Myocardial Injury study is a retrospective cohort study aiming at evaluating the real-world clinical outcomes of COVID-19 focused on myocardial injury. The study was conducted at the Kobe City Medical Center General Hospital in Japan, one of the core hospitals in COVID-19 treatment. The study enrolled consecutive COVID-19 patients admitted to the hospital from July 2020 to September 2021, diagnosed with COVID-19 by positive PCR of nasal or pharyngeal swab specimens. Baseline data on demographic information, medical history, vital signs, disease severity, laboratory tests including cardiac biomarker values and clinical outcomes were collected from electronic health records. We classified the severity of COVID-19 as follows; mild: patients without evidence of viral pneumonia and hypoxia; moderate: patients with clinical signs or chest imaging of pneumonia but without hypoxia; severe: patients with hypoxia with SpO2<94% on room air, a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen < 300 mm Hg, a respiratory rate >30 breaths/min or lung infiltrates >50%; and critical: patients with respiratory failure, septic shock or multiorgan dysfunction, according to National Institutes of Health classification.7 We collected high-sensitive troponin I (hsTnI), N-terminal pro-B-type natriuretic peptide (NT-proBNP),creatine kinase (CK) and CK-myocardial band (CK-MB) in the first available blood test within 24 hours from admission as cardiac biomarkers of interest. If the cardiac biomarkers had not been measured, we additionally measured them obtained at the same time course using an institutional serum blood bank, which enrols consecutive COVID-19 patients from July 2020. We measured them in priority order of hsTnI, NT-proBNP, CK-MB and CK, because the amount of blood sample was limited and all the cardiac biomarkers could not be measured. They were measured using standardised analysers: hsTnI (reference values; 0–28 ng/L, 99th%tile of male; 34.2 ng/L, 99th%tile of female; 15.6 ng/L); Architect i2000 SR (Abbott, Illinois, USA), NT-proBNP (reference values; 0–125 pg/mL); Cobas8000/e801 (Roche diagnostics, Meylan, France), CK (reference values of male; 60–270 U/L, reference values of female; 40–150 U/L); LABOSPECT 008α (Hitachi-hightech, Tokyo, Japan) and CK-MB (reference values; 0–2.2 ng/mL); Architect i2000 SR (Abbott Laboratories). Patients with biomarkers were stratified according to the values of hsTnI, NT-proBNP, CK or CK-MB, using the clinically relevant thresholds (hsTnI: 5.0 ng/L and 99th percentile of the upper reference limit (99%ile URL), NT-proBNP: 125 pg/mL and 900 pg/mL, and CK/CK-MB: the upper normal limit (UNL) and three times of UNL ((3UNL)). The 99%ile URL of hsTnI was 34.2 ng/L for males and 15.6 ng/L for females.8 9 Patients were treated according to the treatment guidelines at that time. In brief, patients with hypoxia were treated with dexamethasone, and remdesivir was added if the onset was within 14 days. The addition of tocilizumab or baricitinib was considered for patients with rapidly progressive respiratory failure.7 10 The primary outcome measure was all-cause death. We defined advanced respiratory support as a composite of endotracheal intubation use or high-flow nasal cannula oxygen therapy (HFNC). Myocardial infarction was adjudicated according to the academic research consortium definition.11 Stroke was defined as ischaemic or haemorrhagic stroke with neurological symptoms lasting >24 hours. The definitions of secondary endpoints are described in online supplemental tables S1–S4. We included heart failure, Takotsubo syndrome, myocarditis, pericarditis, venous thromboembolism (VTE) and tachyarrhythmia as clinical outcome measures of interest. Major adverse cerebrocardiovascular event was defined as a composite of cardiovascular death, myocardial infarction, stroke, heart failure, Takotsubo syndrome, myocarditis, pericarditis or VTE. Due to the study’s retrospective nature, an opt-out approach was used for participant involvement. Written informed consent was obtained separately from the patients when they participated in the serum blood bank registry on admission. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Supplemental material

Statistical analysis

Categorical variables were expressed as numbers and percentages and compared with the χ2 test. Continuous variables were expressed as median with IQR or mean and SD and were compared using the Wilcoxon rank-sum test or Student’s t-test based on their distribution. The cumulative incidences were estimated using the Kaplan-Meier method, and differences among each group were assessed using a log-rank test. Cox proportional hazards model was used to estimate the effect of variables for the primary outcome measure. The multivariable Cox proportional hazards assumptions were conducted for the primary outcome measure, using following covariates. The risk-adjusting covariates included age of ≥75 years, body mass index of ≥25, history of cardiovascular disease, diabetes, hypertension, chronic kidney disease, stroke, chronic obstructive pulmonary disease, chronic liver disease, malignancy, immunosuppressive status, pregnancy, vaccination and severity on admission of >moderate, systolic blood pressure of <100 mm Hg on admission, white cell count ≥9x109/L and CRP≥5.0 mg/dL (online supplemental table S5). All p values were two sided, and p values <0.05 were considered significant. Missing values were not imputed and were handled as missing value. All statistical analyses were performed using JMP software (V.17.0, SAS institute).

Patient and public involvement

This study did not include patients in developing research questions, designing or recruiting participants. The result will be presented to the patients by publishing the academic article.

Results

Patient characteristics

In the study period, 917 patients with COVID-19 were included (figure 1). The patient characteristics including medical history, comorbidities and disease severity are also shown in online supplemental table S5. Notably, approximately 14% of patients had a history of cardiovascular disease, and only 2% had been vaccinated against SARS-CoV-2. More than half of the patients were classified as having severe COVID-19 on admission. Patients with elevated cardiac biomarkers were older and had more cardiovascular disease and cardiac risk factors.

Figure 1

Study population. *If the cardiac biomarkers were not taken, we additionally measured them using an institutional blood bank, using the first available blood test within 24 hours from admission. CK, creatine kinase; CK-MB, creatine kinase myocardial band; HsTnI, high-sensitive troponin I; NT-pro BNP, N-terminal pro-B-type natriuretic peptide; N, number; 99%ile URL, 99th percentile of the upper reference limit; UNL, upper normal limit.

Cardiac biomarkers and clinical outcomes

At admission, the median hsTnI and NT-proBNP levels were 8.3 (IQR: 0–24) ng/L and 155 (IQR: 45–1983) pg/mL, respectively (figure 2). Among the 544 patients with hsTnI values, 365 (67.0%) patients had elevated hsTnI of ≥5.0 ng/L, and 134 patients (24.6%) had TnI of ≥99% ile URL (table 1 and figure 1). Among 546 patients with NT-proBNP values, 295 patients (54.0%) had elevated NT-pro-BNP of ≥125 pg/mL, and 93 patients (17.0%) had NT-proBNP of ≥900 pg/mL (table 1 and figure 1). At admission, the median CK and CK-MB levels were 102 U/L (IQR: 54–220 U/L) and 0.5 ng/mL (IQR: 0.2–1.1 ng/mL), respectively (figure 2). Among the 887 patients with CK values, 230 patients (26%) had elevated CK of ≥UNL and 59 patients (6.7%) had CK of ≥3 UNL. In addition, among 484 patients with CK-MB values, 57 patients (11.7%) had elevated CK-MB of ≥UNL and 18 patients (3.7%) had CK-MB of ≥3 UNL (table 1).

Figure 2

Distribution of biomarker values for patients with each value. (A) High-sensitive troponin I, (B) NT-pro BNP, (C): CK, (D) CK-MB. CK, creatine kinase; CK-MB, creatine kinase myocardial band; HsTnI, high-sensitive troponin I; NT-pro BNP, N-terminal pro-B-type natriuretic peptide.

Table 1

Clinical outcomes at 30 days: stratified by cardiac biomarkers

The median follow-up period was 31 (IQR: 11–90) days. The median hospital stay was 16 (IQR: 6–19) days. In the study population, the cumulative incidence of all-cause death at 30 days after admission was 12.7% (table 1). The incidences of individual cardiovascular events were not high; myocardial infarction: 0.5%, heart failure: 1.4%, Takotsubo syndrome: 0.6%, VTE: 3.0% (pulmonary embolism: 1.6% and deep vein thrombosis: 2.4%), tachyarrhythmia: 5.8%, acute pericarditis: 0.1% and stroke: 1.1% (table 1 and online supplemental table S6).

Higher levels of hsTnI, NT-proBNP, CK or CK-MB were associated with the significantly higher cumulative 30-day incidence of all-cause death relative to the lower levels of each biomarker (hsTnI: <5.0 ng/L group; 4.3%, 5.0 ng/L–99%ile URL group; 8.8% and ≥99% ile URL group; 25.2%, p<0.001. NT-proBNP: <125 pg/mL group; 5.3%, 125–900 pg/mL group; 10.5% and ≥900 pg/mL group; 31.9%, p<0.001. CK: <UNL group; 10.6%, UNL to 3 UNL group; 16.4% and ≥3 UNL group; 23.5%, p<0.001. CK-MB: <UNL group; 7.8%, UNL to 3UNL group; 20.4% and ≥3 UNL group; 38.9%, p<0.001) (table 1 and figure 3). The adjusted risk for all-cause death remained significant for higher thresholds of hsTnI and NT-proBNP (hsTnI≥99% ile URL: HR 1.98, 95% CI 1.11 to 3.54, p=0.02, and NT-proBNP≥900 pg/mL: HR 3.60, 95% CI 1.86 to 6.98, p<0.001). Lower thresholds of hsTnI and NT-proBNP remained similar trends, with the numerical increase of risk for all-cause death (hsTnI≥5.0 ng/L: HR 1.52, 95% CI 0.68 to 3.42, p=0.31, and NT-proBNP≥125 pg/mL: HR 1.77, 95% CI 0.85 to 3.67, p=0.12). In addition, CK and CK-MB demonstrated a significantly higher risk for all-cause death at each threshold after adjustment (CK≥UNL: HR 1.86, 95% CI 1.22 to 2.84, p=0.005, CK≥3UNL: HR 2.04, 95% CI 1.11 to 3.76, p=0.03, CK-MB≥UNL: HR 2.99, 95% CI 1.52 to 5.85, p=0.002 and CK-MB≥3 UNL, HR 5.17, 95% CI 1.85 to 14.5, p=0.004). Almost all deaths occurred within 60 days of admission, and the mortality was slightly increased at 90 days. Elevation of cardiac biomarkers was also associated with higher mortality at 60 and 90 days.

Figure 3

Cumulative incidences for all-cause death. (A) High-sensitive troponin I, (B) NT-pro BNP, (C) CK, (D) CK-MB. CK, creatine kinase; CK-MB, creatine kinase myocardial band; NT-pro BNP, N-terminal pro-B-type natriuretic peptide.

The cumulative 30-day incidence of advanced respiratory support elevated rapidly after admission, and the 30-day incidence was stratified by using the lower thresholds of each biomarker; however, those were similar using the higher cut-off values (hsTnI: <5.0 ng/L group; 25.4%, 5.0 ng/L–99%ile URL group; 44.9% and ≥99% ile URL group; 45.7%, p<0.001. NT-proBNP: <125 pg/mL group; 30.2%, 125–900 pg/mL group; 48.8% and ≥900 pg/mL group; 43.8%, p<0.001. CK: <UNL group; 37.0%, UNL to 3 UNL group; 50.7% and ≥3 UNL group; 54.1%, p<0.001. CK-MB: <UNL group; 35.5%, UNL to 3UNL group; 54.2% and ≥3 UNL group; 56.3%, p<0.001) (table 1 and figure 4).

Figure 4

Cumulative incidences for advanced respiratory support. (A) High-sensitive troponin I, (B) NT-pro BNP, (C) CK, (D) CK-MB. CK, creatine kinase; CK-MB, creatine kinase myocardial band; NT-pro BNP, N-terminal pro-B-type natriuretic peptide.

Discussion

The major findings of the analysis were as follows: (1) Each cardiac biomarker, namely hsTnI and NT-proBNP, could well stratify the mortality risk of COVID-19 patients requiring admission; (2) Even in the severely conditioned patients with a high incidence of cardiac injury, the incidences of cardiovascular complications were rare in the Japanese real-world practice.

The highly infectious nature of the SARS-CoV-2 coerces us to manage unexpected increases of patients with COVID-19, and the patients who need admission sometimes exceed the healthcare provider’s limited capacity. Current practical guidelines recommend assessing the patient’s risk by using risk factor burden as cardiovascular disease, respiratory disease or immunocompromised status. In addition, recent reports demonstrated values of cardiac biomarkers for the stratification of COVID-19 risk. A better understanding of the clinical characteristics and early risk stratification might be helpful in COVID-19 patients to decide the appropriate disposition of the infected patients, especially during the pandemic status. In the present study, we investigated the impact of cardiac biomarkers measured at the time of hospital admission on the prognoses of COVID-19 patients, including relatively large real-world databases.

Although the current study population included many severely ill-conditioned COVID-19 patients, there were few patients with cardiovascular complications such as myocardial infarction, myocarditis, pericarditis and VTEs. There was a concern for thrombotic complications during the first pandemic of COVID-19. From the preomicron variants era analysis, the overall case fatality rate of COVID-19 was reported to be 2.3%, and the mortality reached 10.5% in patients with underlying cardiovascular disease.12 Sparse evidence exists for the incidence of cardiovascular events in COVID-19; a single-centre analysis from Italy reports the cardiovascular incidence of 15.7% (pulmonary embolism; 9.4%, arrhythmias; 3.3%, myocardial infarction; 2.3% and myocarditis; 0.8%), and the population included many patients with cardiac biomarker elevation.13 An international registry among 3011 hospitalised COVID-19 patients reported cardiac incidence of 11.6% (acute coronary syndrome; 0.5%, type II cardiac ischaemia; 0.8%, heart failure; 1.8% and arrhythmia/conduction disorders; 8.6%) and pulmonary embolism; 6.6%.14 On the other hand, a neural network study using Polish National Registry of Invasive Cardiology Procedures reported that the COVID-19 infection had not affected periprocedural death in acute coronary syndrome and concluded that the evaluation of classic clinical risk factors is also important during the COVID-19 pandemic.15

The difference in the reported incidences of cardiovascular complications might have come from several reasons. First, there are many differences in patients included in each analysis; enrolled country, population, practice pattern and predominant variants of COVID-19 during the study periods. Second, we adopted the criteria for analysis of cardiovascular events along with the cardiovascular clinical studies; thus, the adjudication process might differ. Third, we adopted antithrombotic therapy for most patients in need in line with the treatment recommendations for COVID-19 and achieved a low incidence of thrombotic events. Although our data demonstrated relatively low impact on cardiovascular complications due to COVID-19, however, we need to watch for the incidence of cardiovascular complications; in recent populations with newer variants or long-term effects after COVID-19 infection.

Contrary to the low incidence of cardiovascular complications, we frequently observed COVID-19 patients with elevated cardiac biomarkers: hsTnI≥99% ile in 24.6%, NT-proBNP≥125 pg/mL in 35.7%, CK>UNL in 25.9% and CK-MB>UNL in 11.8%. Former reports indicated similar rates of COVID-19 patients with elevated biomarker values: troponin, 6.5%–36%, BNP or NT-proBNP, 13%–49%, CK, 12%–46% and CKMB, 5.1%–11%.4–6 16 17 Consistent with the current study, patients with elevated cardiac biomarkers had a significantly higher incidence of mortality than those without cardiac biomarker elevation. Current analysis revealed the high ability for risk stratification for mortality in COVID-19 patients. In addition, limited studies evaluated cardiovascular biomarker potential for the risk of advanced respiratory support.18 In our analysis, only the low-cut-off values (TnI of 5.0 ng/L and NT-proBNP of 125 pg/mL) stratified the risk well. We may be able to use the cardiac biomarkers for the risk stratification of non-cardiac events. However, we need to know that the ability of higher cut-off values for the stratification of the advanced respiratory support risk was not so high. A recent study using machine learning model has highlighted the importance of conventional risk factors as age, kidney function and albumin for predicting COVID-19 prognosis.19 Combination of multiple risk factors and cardiac biomarkers might be useful for more accurate prediction of COVID-19 outcomes.

The mechanisms underlying cardiac involvement in COVID-19 would be multiple and unclear; however, most biomarker elevations seem unrelated to clinically evident cardiovascular events. Troponin and CK-MB are released into circulation when myocardial necrosis occurs and thus are established as serum-specific cardiac markers of myocardial injury.11 Increased levels of them in patients with COVID-19 can be attributed to several factors, including direct viral invasion of the myocardium via binding to ACE2 receptors, inflammatory myocarditis in the context of cytokine storm and endothelial dysfunction, or myocardial ischaemia/infarction as a consequence of oxygen supply-demand imbalance, microvascular and macrovascular thrombosis, stress-induced cardiomyopathy or acute coronary syndrome from acute inflammation-triggered destabilisation of atheromas.20 NT-proBNP is a sensitive indicator of cardiac haemodynamic stress and plays a central role in diagnosing and managing heart failure.9 SARS-CoV-2 causes myocardial injury and microvascular thrombosis, which lead to ischaemic or inflammatory cardiac dysfunction, ventricular wall stress and right heart overload secondary to the pulmonary consequences of the disease including hypoxic vasoconstriction, pulmonary hypertension, ARDS and pulmonary embolism.5 CK-MB is another myocardial injury marker, and CK is widely used as an alternative to myocardial injury markers, which may reflect the elevation of CK-MB. In addition, some viral tissue invasions may cause muscle injury. CK-MB or CK elevation stratified the event risk well but was not as specific as the other biomarkers. We found 12%–36% cardiac biomarker elevation using the conventional threshold; however, the elevation of cardiac biomarkers does not necessarily reflect cardiovascular complications such as myocardial infarction or heart failure. We also adopted 5.0 ng/L of hsTnI, which is familiar for the cardiologist for excluding acute coronary syndrome, as a low threshold for the stratification. The value demonstrated a favourable ability for stratification; however, there might be a more potent threshold for risk stratification of COVID-19 patients. The current study population predominantly included preomicron variant infection, few-penetration rates of anti-SARS-CoV2 vaccinations and before established antiviral treatment strategy. We need to assess the adaptability of the results in the more recent population.

Although how to use these cardiac biomarkers has not been well defined, measuring them at the time of admission would help identify high-risk COVID-19 patients. Accurate prediction and targeted management for high-risk patients in earlier stages of the disease will contribute to a further reduction in mortality. Troponins may reflect myocardial injury, and BNPs may reflect cardiac functional burden; thus, a combination of these biomarkers might predict a patient’s risk of COVID-19 well. Furthermore, each biomarker has a mechanism in the COVID-19 infection; however, all biomarkers demonstrated an acceptable ability for risk stratification in our analysis. The consistent results for risk stratification may allow physicians to use CK or CK-MB for risk stratification in daily clinical practice, especially in cases without hsTnI or NT-proBNP.

Study limitations

Several limitations of this study should be acknowledged. First, a single-centre retrospective study design precluded any definitive conclusions because of the retrospective study nature. Despite careful statistical adjustments for potential confounding factors and additional measurement of cardiac biomarkers using blood banks, the inherent limitations of the observational study design impede the ability to draw definitive conclusions owing to selection bias in the performance of blood tests and residual samples, as well as the presence of confounders that were not accounted for. The Kobe City Medical Center General Hospital is one of the tertiary hospitals to which many severely ill patients tend to be referred for inpatient management. Thus, the generalisability of the results to other populations should be evaluated carefully. Second, serial data of cardiac biomarkers were not obtained throughout each patient’s hospital stay, although dynamic changes and peak values of each cardiac biomarker during admission may add additive value in prognostication. In addition, most biomarker values were obtained from the serum blood bank. We obtained the cardiac biomarker value as much as possible in the priority order of hsTnI, NT-proBNP, CK-MB and CK, without attending physician’s discretion. We could not assess patients without available blood samples, however, current analysis could evaluate not only patients suspected cardiovascular complications, but also patients with blood samples. Third, the disease severity and prognosis of COVID-19 in our study might differ from the current COVID-19, caused by recent prevalent genetic variants of SARS-CoV-2. Fourth, the median follow-up period was 31 days in the current analysis, and many patients were not followed up after discharge, which makes it difficult to assess the long-term clinical outcomes. However, the Kaplan-Meier curve suggests that most clinical events occur within 60 days, with a slight increase from 60 to 90 days. This seems to reflect the acute phase of the COVID-19 illness, but longer-term data are necessary to understand the long-term effects. In addition, few patients were vaccinated for SARS-CoV-2 in the current population. Even after the penetration of vaccination for SARS-CoV-2, not a few populations have not vaccinated yet, thus the results can be adoptable for those patients.

Conclusions

Elevation of hsTnI, NT-proBNP, CK or CK-MB at the time of admission was associated with a poor prognosis in the current relatively severely ill COVID-19 patients. Measurement of cardiac biomarkers can be an attractive option for risk stratification and deciding appropriate management in patients with COVID-19.

Supplemental material

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information.

Ethics statements

Patient consent for publication

Ethics approval

The study protocol was approved by the Institutional Review Board of Kobe City Medical Center General Hospital (approval number: zn211101).

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors All authors met four criteria for authorship: contributions to the conception or design of the work, drafting the work or reviewing it critically, final approval of the version to be published and agreement to be accountable for all aspects of the work. TTo is a guarantor of the work. Namely, TTo and TM mainly designed the work; MS, TTo, TM, YN, RS, RM, TO, YS, TTa, KK, AK, NE, MK, AD, KT, YK and YF drafted and reviewed the work. All authors approved the manuscript and agreed to be accountable for all aspects of the work.

  • Funding The authors have no financial conflicts of interest to disclose concerning the presentation. This study was financially supported by JSPS KAKENHI, grant number 22K16124.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.