2021 年 85 巻 6 号 p. 944-947
Background: Several studies have reported elevated troponin levels in coronavirus disease 2019 (COVID-19) patients, so we investigated myocardial damage by measuring high-sensitivity troponin T (hsTnT) levels and analyzed the relationship with comorbidities.
Methods and Results: Of 209 patients who recently recovered from COVID-19, 65% had an elevated hsTnT level that was higher than levels in patients with acute phase infection despite most patients (79%) having a mild illness. The hsTnT levels correlated with disease severity, sex, comorbidities, and ACEi and ARB use.
Conclusions: Myocardial damage occurs in the recovery phase of COVID-19, and its evaluation, regardless of patient age, should be considered.
The novel coronavirus disease 2019 (COVID-19) primarily affects the lungs by causing interstitial pneumonitis and severe acute respiratory distress syndrome; it also affects multiple organs, including the cardiovascular system.1 Several reports suggested that elevated troponin in COVID-19 patients was a marker of worsening cardiovascular disease and increased mortality.2,3
Despite its aging population, COVID-19-related fatalities in Japan are lowest among the developed world. Factors such as cultural habits, universal masking, genetics, and relative immunity due to the mandatory Bacillus Calmette-Guérin (BCG) tuberculosis vaccination are probably related to the decreased mortality.4–6 However, the roles of these factors remain unclear. To the best of our knowledge, there are no studies reporting sequelae of COVID-19, particularly myocardial damage.
This study aimed to evaluate the presence of myocardial damage by measuring its serological biomarkers and performing echocardiography, and to describe the relationship with comorbidities in Japanese patients who had recently recovered from COVID-19.
We included patients who were enrolled in the COVIPLA registry, to be studied for convalescent plasma therapy between April and September 2020 at the National Center for Global Health and Medicine (NCGM). Eligible patients were aged 20–69 years, with weight >45 kg in men and >40 kg in women, and had recovered from COVID-19 more than 3 weeks prior. Their diagnosis of COVID-19 was confirmed using either the antigen test or polymerase chain reaction (PCR), and the date of onset was confirmed based on the medical certificate or discharge summary. Blood samples were obtained for complete cell count, and echocardiography was performed to evaluate ejection fraction (EF) by trained echocardiographers who were blinded to the patients’ clinical information. Later, the high-sensitivity troponin T (hsTnT) level was measured using a commercially standardized sample kit (Roche Diagnostics, Tokyo, Japan) and serum samples stored at −112°F obtained from the COVIPLA registry. Because our study focused on detecting cardiac damage caused by COVID-19 and was not specifically evaluating cases of COVID-19 causing acute myocardial infarction, we did not perform electrocardiogram. The local laboratory cutoff for detectable hsTnT was >0.003 ng/mL and values above the 99th percentile (0.013 pg/mL) were recorded as a significant increase.7 All examinations in the COVID-19 recovery period were performed at NCGM. Patients were provided with the details of the study and were given the choice of opting out at any time.
Statistical AnalysisCategorical data are presented as number (percentage) and continuous variables as mean (standard deviation, SD) and median (interquartile range [IQR]).
Comparison between patients’ groups was conducted using a t-test for normally distributed parameters and the Wilcoxon rank-sum test for non-normally distributed data. Fisher’s exact test and chi-square test were used for proportions. A two-sided P-value <0.05 was considered statistically significant. All statistical analyses were carried out using R (Version 4.0.3, The R Foundation, Vienna, Austria).
Of 209 patients, 106 (51%) were male, and the mean (SD) age was 45 (9) years. The median (IQR) time interval between COVID-19 diagnosis and hsTnT evaluation was 56 (36–96) days. A total of 165 patients (79%) did not require oxygen, 38 (18%) required oxygen inhalation, 4 (2%) required intubation, and 2 patients (1%) required extracorporeal membranous oxygenation (ECMO). All patients had normal left ventricular EF (mean: 64.9%, range: 62.2–68.7%).
Elevation of hsTnT During COVID-19 RecoveryFor comparison, we divided the patients into 2 groups: hsTnT-negative (<0.003 ng/mL) group having 74 patients (35.7%), and positive (≥0.003 ng/mL) group having 135 patients (64.6%). Their characteristics are summarized in the Table. In the hsTnT-positive group, hsTnT was significantly elevated (>0.013 ng/mL) in 2 patients (1%), whose values were 0.015 ng/mL and 0.031 ng/mL. The patient with a hsTnT level of 0.015 ng/mL was a 50-year-old man with no underlying medical conditions; he did not require oxygen administration and had demonstrated improvement without medication, including steroids. He underwent blood sampling on the 79th day of illness, and his EF was 57.6%. In contrast, the patient with a hsTnT level of 0.031 ng/mL was a 53-year-old man on either angiotensin-converting enzyme inhibitor (ACEi) or angiotensin II receptor blocker (ARB) for hypertension, whose condition improved with azithromycin and oxygen administration. He underwent blood sampling on the 35th day and his EF was 57.6%.
| COVID-19 (n=209) |
hsTnT >0.003 ng/mL (n=135) |
hsTnT ≤0.003 ng/mL (n=74) |
P valuea | |
|---|---|---|---|---|
| Patient characteristics | ||||
| Age, mean (SD) years | 45 (19) | 50 (42.5–56) | 36 (31–45) | <0.001 |
| Male sex, n (%) | 106 (51) | 93 (69) | 13 (18) | <0.001 |
| Days since onset (SD), days | 56 (62) | 54 (35–97.5) | 56.5 (32.3–89.8) | 0.507 |
| Comorbidities and medication | ||||
| Hypertension, n (%) | 33 (16) | 30 (22) | 3 (4) | <0.001 |
| Diabetes, n (%) | 16 (8) | 16 (12) | 0 (0) | 0.002 |
| Dyslipidemia, n (%) | 26 (12) | 24 (18) | 2 (3) | 0.002 |
| History of smoking, n (%) | 47 (23) | 32 (24) | 15 (20) | 0.608 |
| COPD, n (%) | 1 (1) | 1 (1) | 0 (0) | 1 |
| OMI, n (%) | 0 (0) | 0 (0) | 0 (0) | 1 |
| CABG, n (%) | 0 (0) | 0 (0) | 0 (0) | 1 |
| Arrhythmia, n (%) | 4 (2) | 4 (3) | 0 (0) | 0.300 |
| Use of ACEi/ARB, n (%) | 24 (12) | 23 (17) | 1 (1) | <0.001 |
| Use of β-blocker, n (%) | 1 (1) | 1 (1) | 0 (0) | 1 |
| Severity of COVID-19 | ||||
| Non-severe; no oxygen n (%) | 165 (79) | 96 (71) | 69 (93) | <0.001 |
| Severe; inhalation of oxygen, intubation, and ECMO n (%) |
44 (21) | 39 (28) | 4 (5) | <0.001 |
| Laboratory data | ||||
| White blood cells, mean (SD), μL | 5,310 (1,725) | 5,310 (1,630) | 5,330 (1,875) | 0.652 |
| Hemoglobin, mean (SD), mg/dL | 13.7 (2.2) | 14.1 (2.1) | 13.1 (1.5) | <0.001 |
| Platelets, mean (SD), 104/μL | 24.6 (7.1) | 23.5 (7) | 26.0 (5.9) | 0.009 |
| Echocardiography | ||||
| Ejection fraction, (SD) % | 64.9 (6.5) | 65.4 (6.5) | 63.6 (6.5) | 0.069 |
aP-value for 2-group comparison (hsTnT >0.003 vs. hsTnT ≤0.003) using t-test for normally distributed parameters and Wilcoxon rank-sum test for non-normally distributed data. Fischer exact and chi-square tests were used for proportions. ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CABG, coronary artery bypass grafting; COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 2019; ECMO, extracorporeal membrane oxygenation; hsTnT, high-sensitivity troponin T; OMI, old myocardial infarction; SD, standard deviation.
There were significant differences in disease severity (mild [no oxygen], and severe [requiring oxygen, intubation or ECMO] disease), sex, age, comorbidities (hypertension, diabetes, and dyslipidemia), use of ACEi/ARB, hemoglobin (all P<0.001), and platelet count (P=0.009). Correlation between hsTnT level and time since diagnosis revealed that the hsTnT values showed a linear regression over time irrespective of disease severity. There was no significant correlation with a history of smoking (Figure).
Correlation of hsTnT levels with disease severity (A), time from symptom onset to sampling (B) and correlation of hsTnT levels with history of smoking, sex, comorbidities, age, and use of ACEi/ARB (C–I). The number of patients in each group (of the respective panels) are as follows: No vs. Yes=103 vs. 47 (C), Male vs. Female=106 vs. 103 (D), No vs. Yes=176 vs. 33 (E), No vs. Yes=183 vs. 26 (F), No vs. Yes=193 vs. 16 (G), <60 vs. ≥60 187 vs. 22 (H), No vs. Yes=181 vs. 24 (I). ACEi, angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; ECMO, extracorporeal membranous oxygenation; hsTnT, high-sensitivity troponin T.
Cardiovascular involvement irrespective of preexisting conditions was suspected in 65% of the study patients because of an elevated hsTnT level, despite most of them (79%) having had a mild case of disease. In addition, the EF of all patients was within the normal range. In 2 patients with a high hsTnT level, no notable characteristics were observed because disease severity had been mild to moderate, and they had no history of cardiac disease. ACEi or ARB use might trigger an elevation in hsTnT level. A German study of COVID-19 patients in the recovery phase reported an association between hsTnT levels elevated slightly above the detection limit, and myocardial damage on magnetic resonance imaging.8 Their results revealed that 71% of the recovered patients had positive hsTnT levels, which is similar to our results. Surprisingly, the rate of myocardial damage was similar between the Japanese and German patients despite the cumulative deaths per 100,000 population being significantly different (68.0 in Germany vs. 4.5 in Japan).9
Our study showed a linear regression of the hsTnT values over time irrespective of disease severity in both groups, and that the hsTnT levels were relatively higher in the patients requiring oxygen. The study performed in Germany showed similar results, although no significant correlation with duration between a positive test for COVID-19 and hsTnT testing was found.8 Interestingly, although these results show that myocardial damage is more likely to occur in the recovery phase, the clinical significance of detecting a slight elevation in hsTnT remains unclear. Possible reported causes of myocardial injury by COVID-19 are as follows: myocarditis, stress cardiomyopathy, microangiopathy, pulmonary embolism, acute respiratory distress syndrome, and systemic inflammatory response syndrome.1,2,10–13 Additionally, pneumonia can trigger an elevation in troponin level by hypoxemia, increased sympathetic activity, increased inflammatory activity within coronary atherosclerotic plaques, and endothelial dysfunction.14 Because the pulmonary sequelae of COVID-19, such as dyspnea, decreased respiratory function, and residual ground-grass opacity, persist in some patients even after 3 months of onset, unimproved lung damage might indirectly cause myocardial damage.15 Therefore, further research on the acute phase of COVID-19 is necessary.
There is no study evaluating the correlation of hsTnT level with the comorbidities of patients in the recovery phase. However, correlation in the acute phase has been reported. A study performed in China reported that cardiac damage was more likely in male patients with hypertension.16 A meta-analysis of 16 studies revealed that the incidence of cardiac injury was 24.4% in hospitalized COVID-19 patients. In subgroup analyses, factors such as older age, hypertension, and chronic obstructive respiratory disease were associated with an increased risk of developing a cardiac injury.17
To the best of our knowledge, this is the first study in which Japanese patients in the recovery phase after COVID-19 underwent evaluation for myocardial damage. Despite the younger age of the patients on average, the high rate of troponin positivity was a remarkable finding suggesting that myocardial damage may be present even in mild cases.
Study LimitationsThis was a single-center retrospective study and the findings cannot be generalized in pediatric (≤20 years) and elderly (≥70 years) patients. We did not assess cardiac function before and during the acute phase of COVID-19, so elevated hsTnT levels may not be solely due to COVID-19 and the relationship between an elevated hsTnT level and pre-existing myocardial damage was not evaluated. Some patients with negative hsTnT during the recovery phase might have been positive during the acute phase. The majority of the patients included in our study had mild disease. We did not evaluate the status or severity of the comorbidities.
Troponin T was positive in most of the Japanese COVID-19 patients studied during the recovery phase, and it correlated with disease severity, sex, comorbidities, and ACEi and ARB use. Myocardial damage may be overlooked, so its evaluation in both the acute and chronic phases should be considered regardless of patient age.
We thank the staff of the COVIPLA Registry at the National Center for Global Health and Medicine for their support.
None.
This work was supported by the Health, Labor, and Welfare Policy Research Grants, Research on Emerging and Reemerging Infectious Diseases and Immunization (grant no. 20HA1006).
The study was approved by the Ethics Committee of Center Hospital of the National Center for Global Health and Medicine (approval no. NCGM-G-003559-00).
