Cardiovascular safety of COVID-19 vaccines in real-world studies: a systematic review and meta-analysis

ABSTRACT

Objectives

To evaluate the cardiovascular safety of COVID-19 vaccines in the real world.

Methods

Studies reported on any COVID-19 vaccine-related cardiovascular events in the population aged ≥12 years between 1 January 2020 and 15 June 2022 were included.

Results

A total of 42 studies were included in this meta-analysis. Myocarditis risk was mainly seen after the second (risk ratio [RR], 2.09; 95% confidence interval [CI]: 1.59–2.58) and third (RR, 2.02; 95% CI: 1.04–2.91) dose. A total of 5 vaccines were analyzed, among which mRNA-1273 (RR, 3.13; 95% CI: 2.11–4.14) and BNT162b2 (RR, 1.57; 95% CI: 1.30–1.85) vaccines were associated with myocarditis risk. No significant increase in risk of myocardial infarction (RR, 0.96) or arrhythmia (RR, 0.98) events was observed following vaccination. The risk of cardiovascular events (myocarditis, RR, 8.53; myocardial infarction, RR, 2.59; arrhythmia, RR, 4.47) after SARS-CoV-2 infection was much higher than after vaccination.

Conclusions

The risk of myocarditis was observed after COVID-19 vaccination, but it was much lower than that following the SARS-CoV-2 infection. No significant increased risk of myocardial infarction or arrhythmia was found after COVID-19 vaccination.

KEYWORDS:

• COVID-19 vaccine
• myocarditis
• myocardial infarction
• arrhythmia
• SARS-CoV-2

1. Introduction

The World Health Organization reported that, as of 23 June 2022, there have been a total of more than 539 million confirmed cases of COVID-19 including 6,324,112 cumulative deaths worldwide [1]. An effective and well-tolerated vaccine is an effective measure to reduce the infection rate and to curb the coronavirus pandemic.

Several large studies have evaluated that clinical trials of the COVID-19 vaccines have demonstrated a good safety [2] and efficacy [3]. COVID-19 vaccines, including the mRNA vaccine, adenovirus vector vaccine, inactivated vaccine, and the protein subunit vaccine, are available in different dosage and inter-dose intervals.

Although some vaccines have passed the safety test of phase III clinical trial, rare adverse events can only be reported in the real-world study after marketing. With the gradual reporting of COVID-19 vaccine-related cardiovascular events [2,4–7], the potential adverse event of vaccines raised hesitancy to vaccinate [8]. These may affect vaccination rates and may lead to a longer duration of the COVID-19 pandemic. Moreover, COVID-19 infection increases the risk of cardiovascular events. In addition, the recent emergence of more infectious and immune evasive mutants has made the management of outbreaks more difficult. Therefore, a realistic and objective assessment of vaccine safety has become a major concern for the international community.

Randomized controlled trials are usually conducted under harsh and ideal conditions, with a high degree of consistency in population selection and study settings, which has some limitations in assessing vaccine safety [9]. However, in the case of larger vaccinations in a population, the factors that need to be considered (e.g. individual willingness to vaccinate and group heterogeneity) are more complex than in randomized controlled trials. Real-world studies provide a better indication of the cardiovascular safety of the vaccine in vaccinated populations. Several studies have recently reported real-world cardiovascular safety assessments of the COVID-19 vaccine, but the results remain controversial [2,10–12]. Studies by Barda [2] and Karlstad [9] concluded that the mRNA vaccine was associated with an elevated risk of myocarditis. However, Ip S [12] found little evidence to suggest a higher incidence of these events after the second vaccination. Whiteley [11] found the protective effect of mRNA vaccine on myocardial infarction. Therefore, meta-analyses of real-world studies remain necessary. In this study, we analyzed the risk of cardiovascular events of COVID-19 vaccine and compared vaccine with SARS-CoV-2 to evaluate the cardiovascular safety of COVID-19 vaccine in the real world.

2. Methods

2.1. Search strategy

This study was registered in PROSPERO (CRD 42022344375). A systematic search was done according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement [13] for the conduct of meta-analyses of real-world studies. Searches were conducted on PubMed, EMBASE, and Web of Science databases, without language restrictions from 1 January 2020 to 15 June 2022. Gray literature from Google Scholar and preprint reports were added to the searches of electronic databases. The terms of Medical Subject Headings and relevant text words consisted the following: ‘COVID-19 vaccine,’ ‘SARS-CoV-2 vaccine,’ ‘myocarditis, myopericarditis, cardiac inflammations,’ ‘myocardial infarction,’ ‘arrhythmia, cardiac arrhythmia’ ‘cardiovascular events.’

2.2. Inclusion criteria and exclusion criteria

All real-world studies were included in the systematic review and meta-analysis. Intervention studies (such as randomized controlled studies), animal studies, comments, review articles, case reports, or studies without available data were excluded from the meta-analysis.

We conducted descriptive analyses of the available data following dose one, dose two, and dose three of CoronaVac, ChAdOx1-S, Ad26.COV2.S, mRNA-1273, and BNT162b2 vaccines among individuals aged at least 12 years. The association between cardiovascular events and COVID-19 vaccination was evaluated in this population. Outcomes were myocarditis, myocardial infarction, and arrhythmia. The Tenth Revision of International Classification of Diseases code was used for myocarditis I40, myocardial infarction I21, I22, and arrhythmia I44, I45, I46, I47, I48, I49.

2.3. Data extraction and quality evaluation

For each included study, data extraction and quality evaluation were performed independently by two authors (Y.F.C and E.W.H) using a standardized electronic form. Any disagreements were settled by consultation. Data sought were the first author, year of publication, country/region, type of study, study period, data sources, outcomes, observation window (days), dose number, vaccine type, population characteristics, sample size (total persons and total doses), events. Intra-study risk of bias was evaluated using the Joanna Briggs Institute checklist [14,15] for prevalence studies and the Newcastle-Ottawa Scale [16] for cohort studies.

2.4. Data analysis

The incidence rate and risk ratio (RR) were used to estimate effect sizes. We applied the inverse variance method to calculate Ln RR and SE of the values of the natural logarithm of each study RR. SARS-CoV-2-infected cardiovascular risks were compared with those related to vaccination. The heterogeneity between effect values across studies was estimated by I² statistic. Fixed-effects models were used if I² ≤ 50%. Random-effects models were used if I² > 50%, representing significant heterogeneity. Random-effects meta-regression analyses were used to investigate the associations of population size, vaccine sample size, geographical location, and follow-up duration with the observed incidence rate and RR. Sample sizes for both population and doses were divided into three groups of less than 1 million, 1–10 million, or greater than 10 million. Subgroup analyses were performed for sex, age (<40 years or ≥40 years), dose number, vaccine type, geographical location (North America, Europe, or Asia), and follow-up duration (<21 days or ≥21 days) to assess the incidence rate and risk ratio of cardiovascular events. Evidence of publication bias was examined by funnel plot, Begg’s test. The robustness and reliability of the meta-analysis results were evaluated by sensitivity analyses. Statistical analyses were performed with Stata version 17.0. For all analyses, a P value of less than 0.05 was deemed significant.

3. Results

3.1. Search results and characteristics of included studies

The literature search screened 1296 studies of which 102 were reviewed in full text (Figure 1). Fifty-nine articles were excluded: case report, review (n = 22), absence of outcomes of interest (n = 18), lacked sample data (n = 13), specific population (n = 6). Finally, a total of 42 studies were included in the meta-analyses, of which reported 16,978 cases of myocarditis [2,5,6,10,12,17–48], 28,451 cases of myocardial infarction [2,7,11,19,20,22,24,40,44,46,49–51] and 96,269 cases of cardiac arrhythmia [2,19,21,22,24,40,41] (Table S1). The overall risk of bias in these studies was rated as moderate-to-high quality (Tables S2 and S3). The population size ranged from 8553 to 557,868,504. Funnel plots and Begg’s test identified no strong evidence of publication bias. P values are all >0.05 (Figure S1). Sensitivity analyses showed no significant change in the results of the incidence and risk ratio of cardiovascular events after excluding each study one by one, indicating the robustness of the meta-analysis (Figure S2).

Figure 1. Article Identification flow chart following the PRISMA guidelines.

3.2. Association of myocarditis and COVID-19 vaccine

A total of 1,604,254,833 people who received 2,575,129,450 doses of the vaccine were eligible for inclusion in the studies. Of those with myocarditis, the median time from the last vaccination to symptom onset was 3.2 (0.6–5.8) days after any doses, and 2 (1–4) days after the second dose. The overall incidence rate of myocarditis after COVID-19 vaccination was 14.80 (12.96–16.65) events per million persons, 8.84 (7.77–9.91) events per million doses. The meta-regression analyses identified no association between population size (P = 0.057), vaccine sample size (P = 0.277), geographical location (P = 0.448), or follow-up duration (P = 0.970) and myocarditis incidence rate. Subgroup analyses indicated that the incidence of myocarditis was more frequent in males than in females (18.14 [14.47–21.81] vs. 4.86 [3.69–6.03] events per million people, P < 0.001), and in aged 12–39 years than in 40 years or older (17.75 [11.14–24.35] vs. 5.20 [1.53–8.86] events per million people, P < 0.001). The incidence rate per 1000,000 doses was 5.51 (2.78–8.24) for first dose, 13.66 (10.39–16.9) for second dose and 5.92 (1.77–10.06) for third dose (Figure 2). Six of the included studies reported incidence rate (11.55[5.59–17.52] events per million persons) of myocarditis in unvaccinated population. Two of the included studies reported the incidence rate (107.88[61.80–153.96] events per million persons) of myocarditis following SARS-CoV-2 infection (Table S5).

Figure 2. Incidence rate for myocarditis events after COVID–19 vaccination.

Of the included studies, 17 reported vaccination and myocarditis risk. Figure 3 shows the risk ratios in the vaccination analyses for myocarditis events. In the total population, the risk of myocarditis events was higher in the vaccinated than in the unvaccinated, with a risk ratio of 1.39 (95% CI, 1.13 − 1.65). The meta-regression analyses identified no association between population size, geographical location (P = 0.466), or follow-up duration (P = 0.955) and myocarditis risk ratio (P = 0.082). Subgroup analyses by dose and vaccine type found higher risks of myocarditis following second vaccination (RR, 2.09; 95% CI: 1.59–2.58), third vaccination (RR, 2.02; 95% CI: 1.04–2.91) compared without vaccination. Furthermore, there were increased risks of myocarditis following a second dose of mRNA-1273 (RR, 7.27; 95%CI: 5.33–9.21), of BNT162b2 (RR, 1.55; 95%CI: 1.36–1.74) (Table S4). There was no evidence of myocarditis risk following CoronaVac or ChAdOx1-S vaccination. In the vaccinated population, males were more likely to have a high risk of myocarditis than females (RR, 3.44; 95% CI: 2.61–4.54); the risk of myocarditis was higher in older as compared to younger (<40 vs. ≥40 years: RR, 2.20; 95% CI: 1.06–4.55). Among the unvaccinated, the risk ratio was higher in males (male vs. female: RR, 1.71; 95% CI: 1.49–1.93), while lower in younger (<40 vs. ≥40 years: RR, 0.72; 95% CI: 0.61–0.83). In addition, studies from North America (RR, 1.56; 95%CI: 1.04–2.07) and with follow-up period of more than 21 days (RR, 1.32; 95%CI: 1.05–1.58) also showed COVID-19 vaccine-related myocarditis risk. Four of the included studies reported a risk of myocarditis following the SARS-CoV-2 infection. The combination revealed that infection substantially increased the risk of myocarditis (RR, 8.53; 95% CI: 2.15–14.91).

Figure 3. Risk ratio for myocarditis events after COVID–19 vaccination.

3.3. Association of myocardial infarction and COVID-19 vaccine

A total of 610,438,891 people who received 670,082,343 doses of vaccine were included in the myocardial infarction event analyses. The overall incidence rate of myocardial infarction after COVID-19 vaccination was 1.73 (1.63–1.82) events per 10,000 persons, 1.62 (1.46–1.77) events per 10,000 doses. The meta-regression analyses identified no association between population size (P = 0.643), vaccine sample size (P = 0.822), geographical location (P = 0.873), or follow-up duration (P = 0.814) and myocardial infarction incidence rate. Subgroup analyses by dose found that the myocardial infarction incidence rate after the first vaccination and the second vaccination were 1.81 and 1.10 events per million doses, respectively (Table 1). The incidence rate associated with SARS-CoV-2 infection (9.81 events per 10,000 persons) was much higher than that associated with the vaccine (P < 0.001) (Table S5).

Table 1. Incidence rate for myocardial infarction events following COVID–19 vaccination.

	Studies, No.	Events, No.	Total, No.	Incidence rate (95% confidence interval), per 10,000	Heterogeneity, I^2
Total vaccinations
Persons	11	26,698	610,438,891	1.73(1.63–1.82)	100.0%
Doses	10	28,451	670,082,343	1.62(1.46–1.77)	100.0%
Sex
Male	2	1160	10,371,966	1.03(0.78–1.28)	71.8%
Female	2	394	10,722,873	0.37(0.33–0.40)	71.8%
Age
< 40 years	2	100	10,792,995	0.09(0.07–0.11)	0.0%
≥ 40 years	5	2733	21,573,293	1.54(0.09–2.98)	99.4%
Dose No.
First	5	9160	45,715,167	1.81(0.06–3.55)	100.0%
Second	3	1244	18,207,826	1.10(0.02–2.22)	99.3%
Vaccine type
CoronaVac	1	8764	70,423,565	0.38(0.35–0.41)	100.0%
ChAdOx1-S	3	7956	22,352,515	2.12(−1.64–5.87)	100.0%
Ad26.COV2.S	1	128	5,891,211	0.22(0.18–0.26)	0.0%
mRNA-1273	2	548	131,842,583	0.98(0.91–2.87)	97.4%
BNT162b2	7	10,275	647,096,265	0.96(0.90–1.03)	99.9%
Geographical location
North America	5	1827	299,816,175	1.04(0.31–1.77)	99.4%
Europe	3	23,258	54,215,447	3.93(2.85–5.01)	99.7%
Asia	3	1613	256,380,571	0.47(−0.12–1.06)	99.9%
Follow-up duration
< 21 days	5	2491	297,540,362	1.54(−0.45–3.53)	99.7%
≥ 21 days	6	2226	262,556,384	1.82(0.53–3.10)	100.0%

No association between the vaccine and myocardial infarction risk was observed following either the first dose (RR, 0.97; 95% CI: 0.86–1.09) or the second dose (RR, 1.06; 95% CI: 0.99–1.14). Myocardial infarction risk was not associated with the ChAdOx1-S (RR, 0.91; 95% CI: 0.76–1.06) or BNT162b2 vaccines (RR, 0.95; 95% CI: 0.85–1.05). In the vaccinated population, males were more likely to have a high risk of myocardial infarction than female (RR, 2.61; 95% CI: 1.65–4.14); the risk of myocardial infarction was higher in older as compared to younger (≥40 vs. <40 years: RR, 11.73; 95% CI: 4.83–28.48) (Figure 4). Among the unvaccinated, the risk was higher in males (male vs. female: RR, 2.86; 95% CI: 1.54–4.18) and in those older than 40 (≥40 vs. <40 years: RR, 5.23; 95% CI: 4.98–5.49) for myocardial infarction. Two of the included studies reported the risk of myocardial infarction following SARS-CoV-2 infection. The combination revealed that infection was associated with a substantially increased risk of myocardial infarction (RR, 2.59; 95% CI: 2.38–2.79).

Figure 4. Risk ratio for myocardial infarction events following COVID-19 vaccination.

3.4. Association of cardiac arrhythmia and COVID-19 vaccine

A total of 59,951,135 people and 372,839,175 doses of vaccine were included in the cardiac arrhythmia study. The overall incidence rate of cardiac arrhythmia after COVID-19 vaccination was 9.62 events per 10,000 persons and 5.09 events per 10,000 doses. Subgroup analyses by vaccination dose found that the cardiac arrhythmia incidence rates after the first vaccination and the second vaccination were 6.07 and 5.04 events per million doses, respectively (Table 2).

Table 2. Incidence rate for cardiac arrhythmia events following COVID–19 vaccination.

	Studies, No.	Events, No.	Total, No.	Incidence rate (95% confidence interval), per 10,000	Heterogeneity, I^2
Total vaccinations
Persons	4	88,726	59,951,135	9.62(−4.31–23.55)	100.0%
Doses	5	96,269	372,839,175	5.09(3.17–7.01)	100.0%
Sex
Male	3	46,401	27,886,355	9.85(−8.36–28.06)	100.0%
Female	3	41,965	31,787,342	7.76(−6.53–22.06)	100.0%
Age
< 40 years	3	5009	22,452,974	2.14(−0.88–5.16)	100.0%
≥ 40 years	3	83,383	44,570,456	11.69(−10.91–34.29)	100.0%
Dose No.
First	4	43,848	59,193,521	6.07(−1.09–13.23)	100.0%
Second	3	44,643	47,214,867	5.04(−5.34–15.43)	100.0%
Vaccine type
CoronaVac	1	485	17,030,243	0.28(0.26–0.31)	0.0%
ChAdOx1-S	2	50,215	91,159,837	4.17(−0.06–8.39)	100.0%
Ad26.COV2.S	1	144	5,891,211	0.24(0.20–0.28)	0.0%
mRNA-1273	2	375	1,513,526	6.87(−3.76–17.49)	99.2%
BNT162b2	5	45,047	257,165,581	5.32(3.08–7.57)	100.0%
Geographical location
North America	1	343	277,438	12.36(11.06–13.67)	0.0%
Europe	1	86,754	38,615,491	22.47(22.32–22.62)	0.0%
Asia	2	1629	21,058,206	1.82(−0.42–4.06)	99.3%
Follow-up duration
< 21 days	1	343	277,438	12.36(11.06–13.67)	100.0%
≥ 21 days	3	88,383	59,673,697	8.70(−7.44–24.85)	100.0%

No association was observed of cardiac arrhythmia and the vaccine in total population (RR, 0.98; 95% CI: 0.84–1.12), including first vaccination (RR, 0.97; 95% CI: 0.91–1.02) and second vaccination (RR, 0.97; 95% CI: 0.94–1.01). No association was found with the ChAdOx1-S, mRNA-1273, or BNT162b2 vaccines, whether in Europe (RR, 0.93; 95% CI: 0.91–0.96) or Asia (RR, 1.00; 95% CI: 0.80–1.20). Among the vaccinated population, the risk ratio of cardiac arrhythmia events was higher in older individuals (≥40 vs. <40 years: RR, 3.83; 95% CI: 1.82–8.07), similar results in unvaccinated individuals (≥40 vs. <40 years: RR, 2.27; 95% CI: 1.78–2.89) (Figure 5). Three of the included studies reported the risk of cardiac arrhythmia following SARS-CoV-2 infection. The combination revealed that infection substantially increased the risk of cardiac arrhythmia (RR, 4.47; 95% CI: 3.32–5.63).

Figure 5. Risk ratio for cardiac arrhythmia events following COVID-19 vaccination.

4. Discussion

In this systematic review and meta-analysis, we examined the association between COVID-19 vaccines and cardiovascular events to evaluate the cardiovascular safety profile of COVID-19 vaccine. The study results show an increased risk of myocarditis after vaccination, but it is a rare adverse event with a rate of 14.80 events per million people. In the general population, the increased risk of myocarditis after vaccination was mostly in young men and observed following the second and third doses. Our results do not indicate statistically significant associations between vaccine and myocardial infarction or cardiac arrhythmia events. The incidence and risk of COVID-19-associated cardiovascular events were greatly higher than that of COVID-19 vaccine-related cardiovascular events.

This study reported the risk of myocarditis following COVID-19 vaccination (RR,1.39). The incidence of myocarditis was 8.84 (7.77–9.91) events per million doses. A meta-analysis [52] indicated that the overall incidence of myopericarditis following COVID-19 vaccination was 18.2 (10.9–30.3) per million doses. However, verification of the causal relationship between myocarditis and the vaccines is not possible. Rare cases of myocarditis have also been reported after vaccination for other diseases (such as smallpox vaccine and influenza vaccine). In addition, the incidence of myopericarditis was significantly higher following smallpox vaccinations (132.1 [81.3–214.6] per million doses) compared with COVID-19 vaccination [52]. Vaccine-induced myocarditis is suspected to be the result of an autoimmune phenomenon [53]. The potential mechanisms might be mRNA immune reactivity, antibodies to SARS-CoV-2 spike glycoproteins cross-reacting with myocardial contractile proteins, and hormonal differences [54]. In addition, Li et al. [55] suggest that COVID-19 mRNA vaccine-associated myopericarditis in the murine model may be due to inadvertent intravenous administration or rapid return from the lymphatic circulation resulting in elevated systemic levels of mRNA-vaccine lipid-nanoparticles.

The findings from the present systematic review show that, the CoronaVac vaccine had the lowest incidence of myocarditis, while the Ad26.COV2.S vaccine had the lowest incidence and risk of myocardial infarction and cardiac arrhythmias. Multiple studies have shown that, inactivated vaccines had the highest safety profile and the lowest incidence of myocarditis and myocardial infarction events [56–60]. Safety comparisons between mRNA vaccines and adenovirus vector vaccines have been controversial. A head-to-head comparison [33] of BNT162b2 and mRNA-1273 vaccines indicated that myocarditis risk was higher after mRNA-1273 vaccination than after BNT162b2 vaccination. Both vaccines were associated with an excess risk of myocarditis [2]. Meanwhile, another head-to-head comparison [61] for men aged 18–25 years do not indicate a statistically significant difference in myocarditis risk between recipients of mRNA-1273 and BNT162b2. This study demonstrated an increased risk of myocarditis after mRNA-1273 vaccination than after BNT162b2 vaccination. This is possibly due to the heterogeneity of the population: individuals receiving ChAdOx1 or BNT162b2 vaccines are on average older than those receiving mRNA-1273 vaccine; the interval between the first and second doses is different in BNT162b2 vaccine (3 weeks) and mRNA-1273 (4 weeks). In our meta-analysis, the incidence of myocarditis was 14.80 events per million people following at least one dose of vaccination. Data from a large health-care organization of Israeli [48] indicated the estimated incidence was 2.13 events per 100,000 people. These rates are much lower than the incidence rate described for SARS-CoV-2-infected myocarditis.

Our analysis concluded that the time from the last vaccination to the onset of myocarditis symptoms was 3.2 days after any dose, and 2 days after the second dose. Most cases were considered mild or moderate in severity, with a short duration [48,62]. A systematic evaluation by Pillay [63] concluded that most myocarditis occurs within the first week after vaccination, usually on day 3–4. The overall survival rate of patients with COVID-19 mRNA vaccine-associated myocarditis was greater than 99% [54]. Rosenblum [44] found that from vaccine adverse event reporting system reports, more than half of the patients showed mild adverse reactions and a short duration of symptoms.

Among the vaccinated population, the incidence of myocarditis, myocardial infarction, and arrhythmia events differ in age and sex. These events all occur more commonly in males compared with females. The risk of myocardial infarction and arrhythmia was largely confined to those older than 40 years. For myocarditis, the risk was higher in people aged under 40 years, and vaccine increases the risk in this group. The results are consistent with the characteristics of myocarditis in most studies [41,64], while the explanation is still unclear. The histological changes included an inflammatory infiltrate admixed with macrophages, T-cells, eosinophils, and B cells [65]. A few studies have concluded that the mechanism might be the inhibition effects of testosterone on anti-inflammatory cells and the immune response to Th1 type cells in males, and the inhibitory effects of estrogen on pro-inflammatory T cells in females [65].

We found no evidence of an increased risk of any myocarditis, myocardial infarction or arrhythmia after the first administration. People who received the vaccine had the highest incidence and risk ratio for myocarditis after the second vaccination, followed by the third. Most studies [10,23,61] have shown a higher risk of myocarditis after the second dose of mRNA vaccine. For example, a cohort study from the United States [61] concluded that among young men, the pooled incidence rate was highest after the second dose for BNT162b2 and mRNA-1273. However, only a few studies [40,41] reported statistically no significant risk for myocarditis and the second-dose vaccine, especially CoronaVac and ChAdOx1-S vaccine. The meta-analysis shows a higher incidence of myocardial infarction after the first than the second dose. There could be the following reasons: one, patients with myocardial infarction after the first dose did not receive the second dose; another, patients who received the first dose were older than those who received the second dose. Increased age is a risk factor for myocardial infarction [66]. Most of the studies included in our review did not report on cardiovascular outcomes after receiving the third dose. Therefore, further studies are needed to determine the outcomes of the third dose.

In this study, no significant increased risk of myocardial infarction or arrhythmia was found after any vaccination. There was a greater risk of myocarditis (RR, 8.53), myocardial infarction (RR, 2.59), cardiac arrhythmia (RR, 4.47) following SARS-CoV-2 infection, and that virus significantly increases the risk of many other serious adverse events [2]. Even among the young male population, COVID-19-associated myocarditis were 6 times higher than that of COVID-19 vaccine-related myocarditis [67].

We conducted random-effects models for meta-analysis and implemented multiple subgroup analyses to assess potential sources of heterogeneity. The heterogeneity might be due to the diversity of study populations, follow-up, vaccine components, and diagnostic criteria. Comparisons between incidence rates from different sources may yield systematic errors. In addition, some of the studied outcomes also had population-level heterogeneity that followed age patterns. The incidence of myocardial infarction and cardiac arrhythmia increased with age. In this analysis, the heterogeneity did not change significantly after our subgroup analysis by sex, age, vaccine type, geographical location, and follow-up duration. However, some outcome events are derived from a small number of studies and should be carefully interpreted.

Limitations of this study. First, studies on cardiovascular events of the vaccine were included in this meta-analysis, while studies on other adverse events (such as immune thrombosis, thrombocytopenia, stroke) of COVID-19 vaccines were not included. Second, the significant heterogeneity observed across studies could not be explored nor explained by subgroup analyses or meta-regression. Third, most studies did not report outcomes in people under 12 years of age who were vaccinated. Fourth, studies should be carried out to understand the long-term prognosis of patients with COVID-19 vaccine-related myocarditis, myocardial infarction, and arrhythmia, although most cases recover from the acute course. Fifth, most enrolled studies were conducted in 2021, when the delta variant was predominant, while the incidences of myocarditis may be different in people with different variants. Continued surveillance of vaccine-related myocarditis risk on the existing vaccines against emerging variations is essential. Sixth, due to different age groups in different studies, which restrict to perform a more detailed subgroup analysis of elders, adults, and adolescents.

In conclusion, the present study suggests that myocarditis following COVID-19 vaccination is rare, and the highest risk was in men aged under 40 years after the second dose of mRNA-1273 or BNT162b2. No significant risk of myocardial infarction or arrhythmia was found after any vaccination. The risk of cardiovascular events was much higher after the SARS-CoV-2 infection than after COVID-19 vaccination. Additional studies are needed to investigate long-term safety, not only on the studied endpoints but also on other COVID-19 vaccination outcomes.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or material discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

Y.F.C., E.W.H. and B.L. conceived the study. Y.F.C and E.W.H. collected the data. Y.F.C. analyzed and interpreted the data, and wrote the first draft of the paper. All authors edited and approved the final manuscript.

Ethical approval

Patients or public were not involved in setting the research question or the outcome measures, nor were they involved in the design or conduct of the study. No participants were asked to advise on interpretation or writing up of the manuscript. There are no plans to involve patients in the dissemination of the study findings.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14760584.2023.2150169

Additional information

Funding

This paper was not funded.