In April 2022, the Fourth People's Hospital of Shanghai, Tongji University, was designated as one of the hospitals responsible for treating patients infected with the Omicron variant during the outbreak of the epidemic in Shanghai. In this study, a cohort of 772 patients was included out of a total of 2,645 patients recorded in the medical system between 10 April and 27 June 2022. The inclusion criteria comprised confirmed cases of the SARS-CoV-2 Omicron variant, individuals aged 60 and above, those with TnI test results within 24 h of hospitalization, and those with complete medical information. Exclusion criteria applied to unconfirmed cases of the Omicron variant, individuals under 60 years of age, those without TnI results within 24 h of hospitalization, and those lacking medical information. The study was conducted following the 1975 Declaration of Helsinki (8), was approved by the Ethics Committee of the Fourth People's Hospital of Tongji University (No. 2022097-001), and was published in the Chinese Clinical Trials Registry (CHiCTR2200065440). Written informed consent was waived by the ethics committee of the designated hospital for patients during the SARS-CoV-2 Omicron variant pandemic.

Two researchers retrieved demographic characteristics (age and gender), clinical data (symptoms, comorbidities, treatment methods, complications, and outcomes), and laboratory findings from each patient admitted to the hospital using the hospital's electronic medical records database. Data for all the clinical biomarkers in this study were collected immediately after the admission of the patients.

According to the Fourth Universal Definition of Myocardial Infarction guidelines, myocardial injury is defined as a serum level of the cardiac biomarker TnI above 99% of the upper reference limit (5, 6). In this study, the serum level of TnI was measured in patients admitted to the hospital within the first 24 h. The patients were classified according to the presence or absence of myocardial injury; a level above 0.028 ng/ml is defined as an abnormal serum TnI level according to test specifications.

The clinical course classification and treatment was in accordance with the ninth version of the national COVID-19 guidance (9). To confirm the presence of infection, 2019-nCoV detection kits (Shanghai Bojin Medical Technology Company) were used to detect the cycle threshold (Ct) of the ORF1ab gene (Ct-ORF1ab) and the N gene (Ct-N) of nasopharyngeal swab samples using real-time reverse transcriptase-polymerase chain reaction. Both the Ct-N gene and Ct-ORF1ab greater than 35 were judged as negative. Mild cases are defined as those with mild symptoms such as fever, fatigue, and loss of taste/smell but without radiographic evidence of pneumonia. Moderate cases refer to those with typical symptoms of respiratory infection (e.g., fever, dry cough, fatigue) and radiographic evidence of pneumonia. Severe cases refer to those patients with at least one of the following conditions: breathing problems, low oxygen saturation, low PaO₂/FiO₂ (PaO₂ denotes a partial pressure of oxygen in arterial blood; FiO₂ denotes the fraction of inspired oxygen), or progressive symptoms combined with pulmonary imaging showing an obvious progress of lesions (>50%) within 24–48 h. Patients who meet any of three criteria, namely, respiratory failure, shock, or organ failure requiring admission to an intensive care unit, are considered severe cases. In this study, we combined severe and critical patients into a severe group for statistical analysis.

Other laboratory indicators, including data on white blood cell (WBC, 10⁹/L) count, neutrophil (NEUT, 10⁹/L) count, neutrophil% (NEUT%), lymphocyte (LYM, 10⁹/L) count, lymphocyte% (LYM%), monocyte (MON, 10⁹/L) count, monocyte% (MON%), hemoglobin (g/L), C-reaction protein (CRP, mg/L), procalcitonin (ng/ml), serum amyloid A (SAA, mg/L), interleukin-6 (pg/ml), myoglobin (ng/L), creatinine kinase-myocardial band (CK-MB, ng/L), N-terminal pro-B-type natriuretic peptide (NT-proBNP, pg/ml), total protein (TP, g/L), lactate dehydrogenase (LDH, U/L), aspartate aminotransferase (AST, U/L), alanine aminotransferase (ALT, U/L), glomerular filtration rate (estimated) (eGFR), blood urea nitrogen (BUN), and serum creatinine (sCr), were collected from patients who were admitted to the hospital within the first 24 h.

Descriptive statistics were obtained for all study variables. Statistical analysis results of continuous variables were described as the mean ± standard deviation (SD) if they had a Gaussian distribution, as determined using the Kolmogorov–Smirnov test. Otherwise, they were expressed as the median and interquartile range (IQR). Categorical variables were compared using Pearson's chi-squared test, but in the case of limited data, Fisher's exact test was used. Continuous variables were compared using t-tests or Mann–Whitney U tests, as appropriate. The stratified log-rank statistic was used to compare the probability of patients requiring intensive care. The optimal statistical cutoff values for distinguishing myocardial injury were calculated based on the receiver operating characteristic (ROC) curve. Considering the power analysis, there are 26 concomitant variables in our study; the number of myocardial injury cases required was 260 based on the 10 EPV (10 events per variable) method. Logistic regression was employed to analyze potential risk factors for the occurrence of myocardial injury. All statistical analyses were performed using SPSS 20 or R v4.1.3, and all graphs were drawn using GraphPad. The significance threshold for all analyses was set at P < 0.05.

Out of a total of 2,645 patients recorded in the medical system between 10 April and 27 June 2022, we excluded 828 patients whose SARS-CoV-2 Omicron variant status was not confirmed, 302 patients aged below 60 years, 362 patients who lacked TnI test results within the initial 24 h of hospitalization, and 381 patients with unavailable medical information. Consequently, a cohort of 772 patients was included in this study. Figure 1 depicts the patient recruitment process.

In the enrolled population, the median age was 82 years (range, 60–104 years), with 326 (42.23%) patients being men. Among the cohort, 263 patients (34.07%) had a myocardial injury, while 509 patients (65.93%) did not have a myocardial injury (Table 1). Among these patients, the most common symptoms were cough [458 patients (59.33%)], sputum production [339 patients (43.91%)], and fever [199 patients (25.78%)]. Sore throat, shortness of breath, rhinorrhea, and chest tightness were present in 102 (13.21%), 66 (8.55%), 43 (5.57%), 37 (4.79%), and 19 (4.6%) patients, respectively. Dizziness [21 (2.72%) patients], headache [19 (2.46%) patients], and myalgia [16 (2.07%) patients] were rare. Hypertension [471(61.01%) patients], coronary artery disease [232 (30.05%) patients], cerebrovascular disease, including cerebral hemorrhage and cerebral infarction [218 (28.24%) patients], and diabetes [191 (24.74%) patients] were the most common pre-existing diseases. The proportions of arrhythmia, tumor, chronic kidney disease, chronic obstructive pulmonary disease, and heart failure were 12.31% (n = 95), 11.01% (n = 85), 7.12% (n = 55), 2.33% (n = 18), and 2.07% (n = 16), respectively.

Overall, patients with myocardial injury were older than those without myocardial injury [median (range), 87 (61–104) vs. 78 (60–101), respectively, P < 0.001]. Patients with myocardial injury were stratified by gender and age. The probability of myocardial injury increased with age in both male and female patients (P < 0.001). There was a significant difference in the gender distribution of patients with myocardial injury between 71 and 80 years (P = 0.016); however, there was no significant difference in the gender distribution of patients with myocardial injury among those aged 60–70, 81–90, and >90 years (Figure 2). Patients with myocardial injury were more likely to have shortness of breath and chest tightness than those without myocardial injury. Moreover, pre-existing diseases, including hypertension, coronary heart disease, cerebrovascular disease, arrhythmia, chronic kidney disease, and heart failure, were present more often among those with myocardial injury than those without myocardial injury (all P’s < 0.01) (Table 1).

Among all 772 patients, the median levels of D-dimer, interleukin-6, CRP, and SAA were all elevated, while the median levels of TP were decreased compared with normal values/ranges. The median levels of other pathological indicators, including WBC, NEUT, LYM, MON, and D-dimer, were within the normal range (Table 2). Patients with myocardial injury had higher median values of WBC (10⁹/L), NEUT (10⁹/L), NEUT%, hemoglobin, D-dimer, interleukin-6, procalcitonin, CRP, SAA, myoglobin, CK-MB, NT-proBNP, LDH, ALT, BUN, and sCr compared with those without myocardial injury, but neo-coronavirus Ct-ORF1ab values, Ct-N values, LYM (10⁹/L), LYM%, TP, and eGFR were lower.

During hospitalization, 704 patients of those included in our study (91.19%) were treated with Paxlovid; the proportions of patients treated with heparin for anticoagulation, immunoglobulin, glucocorticoids, transnasal high-flow oxygen therapy, non-invasive ventilation, and invasive mechanical ventilation were 75.91% (n = 586), 5.05% (n = 39), 16.32% (n = 126), 4.53% (n = 35), 11.53% (n = 89), and 4.40% (n = 34), respectively. The proportions of mild, moderate, and severe cases were 43.52% (n = 336), 43.13% (n = 333), and 13.34% (n = 103), respectively. The median duration of hospitalization was 11 days (IQR, 8–15 days); during their hospital stay, 89 patients (23.30%) were transferred to the intensive care unit for treatment (Table 3).

A higher proportion of patients with myocardial injury were given heparin therapy, glucocorticoid therapy, transnasal high-flow oxygen therapy, non-invasive ventilation, and invasive mechanical ventilation compared with those without myocardial injury (Table 2). Individuals with myocardial injury experienced both a prolonged duration of Omicron infection and an extended hospitalization compared with those without myocardial injury.

Patients infected with the Omicron variant were divided into three groups (mild, moderate, and severe) according to the severity of their disease; those with severe disease had higher serum TnI levels than patients with mild or moderate disease (Figure 3A). A higher proportion of those with myocardial injury progressed to severe disease than those without myocardial injury (Table 3). Furthermore, patients with myocardial injury were more likely to be transferred to the intensive care unit for treatment than those without myocardial injury (P < 0.001) (Figure 3B).

An ROC curve analysis was performed to assess the predictive value of routine pathological indicators for myocardial injury, excluding myocardial indicators such as Ct-ORF1ab, Ct-N, WBC, NEUT, NEUT%, LYM, LYM%, hemoglobin, D-dimer, interleukin-6, procalcitonin, CRP, SAA, TP, LDH, ALT, eGFR, BUN, and sCr. The results, depicted in Figure 4, indicated that these indicators showed varying degrees of accuracy in predicting myocardial injury. The sensitivity and specificity of these indicators in predicting myocardial injury were both statistically significant (P < 0.05). The respective cutoff values for these indicators were determined to be within ranges, such as 21.275 for Ct-ORF1ab, 22.055 for Ct-N, 6.115 × 10⁹/L for WBC, 3.615 × 10⁹/L for NEUT, 70.05% for NEUT%, 0.945 × 10⁹/L for LYM, 18.05% for LYM%, 110.5 g/L for hemoglobin, 0.615 mg/L for D-dimer, 22.69 pg/ml for interleukin-6, 0.0435 ng/ml for procalcitonin, 19.15 mg/L for CRP, 21.33 mg/L for SAA, 58.47 g/L for TP, 218.065 U/L for LDH, 32.59 U/L for ALT, 83.5 for eGFR, 6.865 mmol/L for BUN, and 81.3 μmol/L for sCr.

Based on the optimal cutoffs determined by the ROC curve, the values of laboratory biomarkers were converted to categorical variables. As shown in Table 4, indicators with statistically significant differences included age, hypertension, coronary artery disease, cerebrovascular disease, arrhythmia, chronic kidney disease, and heart failure, and categorical variables for indicators were identified in the logistic regression analyses. In Table 4, the univariate logistic regression analyses revealed significant associations. The odds ratios (ORs) and 95% confidence intervals (CIs) were determined for various factors. All of these associations were statistically significant (P < 0.05). In the multivariable logistic regression analyses, several factors were identified as potential risk factors for myocardial injury. These included advanced age, pre-existing coronary heart disease, interleukin-6 levels >22.69 pg/ml, procalcitonin levels >0.0435 ng/ml, D-dimer levels >0.615 mg/L, and sCr levels >81.30 μmol/L. The corresponding ORs and 95% CIs were 1.063 (1.036–1.091), 2.171 (1.350–3.491), 2.101 (1.202–3.673), 3.199 (1.967–5.202), 2.108 (1.273–3.489), and 3.077 (1.841–5.141), respectively.

This study has several limitations. First, it is a retrospective, single-center analysis; larger cohorts of older adults from multiple centers are needed to further explore the characteristics. Second, our analysis focused on myocardial injury within 24 h of admission. The dynamics of TnI levels and the timing of myocardial injury onset and its duration were not explored. Third, this study focused on the early identification of myocardial injury among older adults infected with the Omicron variant; some measures associated with cardiovascular diseases, such as lipid levels in the blood, were not included in the analysis because these values were not measured or recorded in many patients.