Emergency department presentations for chest complaints after mRNA COVID‐19 vaccinations in children and adolescents

Abstract

Objective

To investigate characteristics and management of children presenting with chest complaints to a tertiary paediatric ED post‐mRNA COVID‐19 vaccine.

Methods

This was a retrospective medical record review with data linkage to the Australian Immunisation Register. The study setting was the Royal Children's Hospital, Melbourne, Australia. Children <18 years who had a troponin blood test performed in hospital within 14 days of receiving mRNA COVID‐19 vaccination were included. Elevated troponin and myocarditis or pericarditis as per Brighton criteria was the primary outcome. Vaccination status, length of stay, investigations and clinical management were secondary outcomes.

Results

Six hundred and ten patients had a troponin test in 13 months. After exclusion of trauma‐related tests (n = 31), known cardiac patients (n = 75) and others (n = 145), 359 troponins were obtained due to chest complaints and related symptoms, with 283 troponins assessed to be mRNA vaccination‐related. There was a temporal peak in presentations with a 30‐fold monthly increase in troponin post‐commencement of mRNA COVID‐19 vaccines. In those with chest complaints following mRNA vaccination, mean age was 14 years and 50.4% were female. Fourteen out of 283 (5%) vaccine‐related troponins were abnormal with 14 patients assessed to have vaccine‐associated myocarditis. No patients had pericarditis.

Conclusions

There was a large number of possible mRNA COVID‐19 vaccine‐related chest complaints presenting to the ED. Few patients had abnormal troponins or myocarditis.

ED presentations for mRNA COVID‐19 vaccination‐related chest complaints, as measured by the conduct of troponin tests in children <18 years of age, increased 30‐fold compared to prior to roll out of the vaccine. Of 283 vaccine‐related chest complaints, only 14 (5%) were found to have Brighton criteria‐defined myocarditis and none had pericarditis. Myocarditis outcomes were benign. Although mRNA COVID‐19 vaccinations (Comirnaty BNT162b2 COVID‐19 [Pfizer‐BioNTech] and Spikevax mRNA‐1273 [Moderna]) led to increased emergency presentations, the incidence of confirmed inflammatory heart conditions was low.

Key findings.

• ED presentations for mRNA COVID‐19 vaccination‐related chest complaints as measured by the conduct of troponin tests in children <18 years of age increased 30‐fold compared to before vaccine roll out.

• Of 283 vaccine‐related chest complaints, only 14 (5%) were found to have Brighton criteria defined myocarditis and none had pericarditis. Myocarditis outcomes were benign.

• While mRNA COVID‐19 vaccinations [Comirnaty BNT162b2 COVID‐19 (Pfizer‐BioNTech) and Spikevax mRNA‐1273 (Moderna)] led to increased emergency presentations, the incidence of confirmed inflammatory heart conditions was low.

Introduction

The COVID‐19 pandemic caused most morbidity and mortality in older patients and vaccines initially targeted the older population groups. Successively younger age groups were included in national vaccine programmes once the vaccine became more widely available and older patient groups had achieved sufficient vaccine coverage. Before the introduction of mRNA COVID‐19 vaccination in children and adolescents, information had become available about a link between the vaccines and myocarditis and pericarditis in adults. ¹ , ² , ³ It had also become apparent that younger adults were at increased risk, with the highest incidence reported among males 16–24 years of age following a second vaccination dose. ¹ , ²

In September 2021, vaccination for all persons aged 12–16 years was made available in the state of Victoria in Australia using Comirnaty BNT162b2 COVID‐19 (Pfizer‐BioNTech) and Spikevax mRNA‐1273 (Moderna). ⁴ A high two‐dose coverage of mRNA COVID‐19 vaccination was obtained in the paediatric age group (12–15 years 91% and 16–17 years 96.5%) ⁵ as Melbourne was the city that had the longest period of lockdowns and school closures in Australia. ⁶

Following a rapid increase in ED presentations after introduction of mRNA COVID‐19 vaccinations, the Australian Technical Advisory Group on Immunisation (ATAGI) in collaboration with the Paediatric Research in Emergency Departments International Collaborative and the Australasian College for Emergency Medicine, released clinical guidance for review of chest pain in September 2021. This recommended all patients who developed either cardiac or non‐specific symptoms without an alternate cause within 14 days of receiving an mRNA vaccine, to present to healthcare for assessment of possible vaccine‐related myocarditis or pericarditis (Fig. S1). ⁸

Based on this steep rise in ED presentations for chest complaints after the introduction of the mRNA COVID‐19 vaccinations in adolescents, we set out to investigate chest complaints possibly related to COVID‐19 vaccination and subsequent diagnosis of myocarditis or pericarditis using Brighton Collaboration criteria. ⁹ We did this by assessing all patients with troponin tests obtained within 14 days of mRNA COVID‐19 vaccination based on linkage with the Australian Immunisation Registry (AIR).

Methods

Study design

We conducted a single‐site study using retrospective electronic medical record data of children who presented to the ED at the Royal Children's Hospital (RCH) in Melbourne, Australia. Vaccination data were also obtained through linkage of records with the AIR. RCH is a large tertiary children's hospital with an annual ED census of 100 000. The study was approved as a quality assurance project by the RCH Research Office (QA/80737/RCHM‐2021). The development of the project was undertaken without patient or family involvement.

Study population

We assessed all patients aged <18 years who had a troponin test conducted at RCH between 1 January 2021 and 31 January 2022. The troponin test was a high sensitivity Troponin I test. Our inclusion criteria were patients aged under 18 years who presented to ED with chest complaints and related symptoms within 14 days of a COVID‐19 vaccination and had a troponin test taken.

Although all patients' medical records were reviewed, patients were excluded from further analysis if the troponin test was following trauma, for a known cardiac disorder, or medical notes or indications for troponin were missing. In addition, we excluded patients who had been diagnosed with COVID‐19, PIMS‐TS and other specified alternative diagnoses (sepsis, oncological diagnosis) and those who had not received a COVID‐19 vaccination within the previous 14 days.

Study process

Data were obtained from the electronic medical records in two ways. Data from electronic medical records were extracted by the Centre for Health Analytics at RCH and imported into the REDCap ⁷ study database using a standardised case report form. This data defined the cohort and included clinical and demographic characteristics, investigations completed, admission status and length of stay data. Second, further clinical data, such as indication for troponin test, symptoms, examination findings, investigation results, treatment and follow up, were extracted manually through individual record reviews conducted by an emergency doctor (MP) and nurse practitioner (SF). In addition, data linkage with the AIR database was undertaken to obtain vaccination data including the vaccine administered, dose number and date.

Patients with an elevated troponin test who had received mRNA COVID‐19 vaccination within 14 days prior to symptoms occurring were defined as having possible, probable or definite myocarditis or pericarditis using Brighton Collaboration criteria (Tables [Link], [Link]). ⁹ These patients had a further review of their medical records by a paediatric cardiologist (BJ).

Statistical analysis

Data were analysed using Stata IC 16.0 (StataCorp, College Station, TX, USA). We summarised data for continuous variables as medians with interquartile ranges (IQRs) or numbers and proportions, as appropriate. Key proportions were compared using chi‐squared tests and continuous variables were compared using rank‐sum tests.

Outcomes and definitions

The primary outcome was to determine the incidence of myocarditis and pericarditis using Brighton Collaboration criteria ⁹ with symptoms developing within 14 days of mRNA COVID‐19 vaccination.

The secondary outcomes were to measure the burden of care in relation to patients presenting to the ED who had a troponin test taken for vaccine‐related symptoms which included symptoms that are commonly associated with myocarditis such as chest pain, shortness of breath, palpitations, dizziness, syncope, fever, abdominal pain, vomiting, cough, oedema and poor feeding.

We assessed risk factors, presenting symptoms, timing of symptoms related to vaccination, abnormal investigations such as ECGs, chest X‐rays, echocardiograms, cardiac MRI and cardiac biopsies as well as interventions in patients with an abnormal troponin test.

An abnormal troponin test using the VITROS XT 7600 platform (Ortho Clinical Diagnostics, Raritan, NJ, USA) was defined as higher than 11 ng/L.

ECG changes such as benign early repolarisation, sinus arrhythmia, incomplete right bundle branch block and non‐specific changes were classified as normal ECGs. ECGs were classified as abnormal if any abnormality was recorded in the ECG report provided by a senior cardiologist, or by a senior ED clinician as recorded in the patients' note.

Results

In the 13‐month period between 1 January 2021 and 31 January 2022, we identified 610 patients who had a troponin blood test in hospital. After exclusion of presentations based on the indication for troponin testing, including trauma‐related (n = 31), known cardiac patients (n = 75), patients with PIMS‐TS (n = 32), COVID‐19 infection (n = 53), other causes (sepsis, oncological disease) (n = 21) and missing documentation or indication for troponin (n = 38), 359 troponins were identified as being obtained due to chest complaints (Fig. 1). Of these, 283 troponins were categorised to be vaccine‐related as the patient's symptoms developed within 14 days of mRNA vaccine administration.

Figure 1.

Patient flow chart. COVID 19, coronavirus; PIMS‐TS, Paediatric Inflammatory Multisystem Temporally associated with SAS‐COV‐2 (severe acute respiratory syndrome coronavirus 2).

Figure 2 shows a low rate of troponin tests and ECGs performed in hospital until August 2021 with a steep rise in tests from September 2021 and a peak in October 2021 following rollout of the vaccine programme in Victoria for ages 12–16 years.

Figure 2.

Frequencies of overall and vaccine‐related troponins ordered between January 2021 and January 2022. Troponin: normal (); abnormal (). Vaccine‐related: normal (); abnormal ().

There were equal numbers of males and females presenting to the ED with chest complaints. Chest pain or discomfort was the most common presenting symptom (n = 252, 89%), followed by shortness of breath (n = 75, 26.5%) and palpitations (n = 65, 23%). Few patients had pre‐existing medical conditions (Table 1). The median number of days from vaccine administration to symptom onset for all patients was 2.0 days (IQR 1.0–7.0). Most patients (n = 257, 90.8%) had received Comirnaty BNT162b2 COVID‐19 (Pfizer‐BioNTech) and presented within 2 days of receiving the vaccine. More patients presented following their second dose of vaccination (n = 164, 57.9%) compared to their first (n = 119, 42%) (Table 2).

TABLE 1.

Characteristics of all patients who had troponin test performed in hospital and vaccine‐related symptoms

	All troponin		Troponin normal		Troponin abnormal
	n = 283		n = 269		n = 14
	n	(%)	n	(%)	n	(%)
Sex
Male	140	(49.5)	128	(47.6)	12	(85.7)
Female	143	(50.5)	141	(52.4)	2	(14.3)
Age, mean (SD)	14.2	(1.8)	14.2	(1.9)	14.5	(1.2)
Age, median (IQR)	14.4	(13.1–15.6)	14.4	(13.1–15.7)	14.6	(14.1–14.9)
Age
5–11 years	18	(6.4)	18	(6.7)	0	(0.0)
12–16 years	265	(93.6)	251	(93.3)	14	(100.0)
Aboriginal/Torres Strait Islander	6	(2.1)	5	(1.9)	1	(7.1)
Pre‐existing medical condition
Heart disease	0	(0.0)	0	(0.0)	0	(0.0)
Chronic renal disease	1	(0.4)	0	(0.0)	1	(7.1)
Immunosuppressed	3	(1.1)	3	(1.1)	0	(0.0)
Connective tissue disease	0	(0.0)	0	(0.0)	0	(0.0)
Diabetes mellitus	3	(1.1)	2	(0.7)	1	(7.1)
Other predisposing factors	55	(19.4)	52	(19.3)	3	(21.4)
Symptoms
Chest pain/discomfort	252	(89.0)	240	(89.2)	12	(85.7)
Palpitations	65	(23.0)	63	(23.4)	2	(14.3)
Dizziness/syncope	35	(12.4)	34	(12.6)	1	(7.1)
Shortness of breath	75	(26.5)	72	(26.8)	3	(21.4)
Fever	28	(9.9)	26	(9.7)	2	(14.3)
Oedema	0	(0.0)	0	(0.0)	0	(0.0)
Cough	11	(3.9)	10	(3.7)	1	(7.1)
Abdominal pain	13	(4.6)	13	(4.8)	0	(0.0)
Vomiting	13	(4.6)	13	(4.8)	0	(0.0)
Poor feeding	1	(0.4)	1	(0.4)	0	(0.0)
Other†	90	(31.8)	87	(32.3)	3	(21.4)
mRNA vaccine
Pfizer	257	(90.8)	246	(91.4)	11	(78.6)
Moderna	26	(9.2)	23	(8.6)	3	(21.4)

^†Other symptoms included headache, myalgias, sore throat, anorexia and nausea. IQR, interquartile range; SD, standard deviation.

TABLE 2.

Length of stay, investigations ordered in the ED, and time from vaccination for patients presenting with mRNA COVID vaccine‐related cardiac symptoms

	All troponin			Troponin normal			Troponin abnormal
	n = 283			n = 269			n = 14			P
Length of stay in hospital (h), median (IQR)	283	3.2	(2.5–4.2)	269	3.1	(2.5–4.0)	14	36.1	(14.8–54.5)	<0.001
Length of stay in ED (h), median (IQR)	283	3.1	(2.4–4.0)	269	3.0	(2.4–3.9)	14	5.5	(4.1–9.8)	<0.001
Investigations in the ED
Blood test		79	(27.9)		73	(27.1)		6	(42.9)	0.225
ECG		276	(97.5)		265	(98.5)		11	(78.6)	0.003
Chest X‐ray		32	(11.3)		28	(10.4)		4	(28.6)	0.060
ECHO		1	(0.4)		0	(0.0)		2	(14.3)	0.050
Represented to ED within 14 days		22	(7.8)		18	(6.7)		4	(28.6)	0.017
Onset of symptoms post mRNA COVID vaccine
Median no. of days (IQR)
1st	119	2.0	(1.0–7.0)	117	2.0	(1.0–7.0)	2	4.5	(3.0–6.0)	0.481
2nd	164	2.0	(1.0–6.0)	152	1.0	(1.0–6.0)	12	2.0	(2.0–3.0)	0.634
3rd	0
Pfizer median no. of days (IQR)
1st	107	2.0	(1.0–7.0)	105	2.0	(1.0–7.0)	2	4.5	(3.0–6.0)	0.465
2nd	150	2.0	(1.0–6.0)	141	2.0	(1.0–6.0)	9	3.0	(2.0–3.0)	0.618
3rd	0
Moderna median no. of days (IQR)
1st	12	3.0	(0.0–14.5)	12	3.0	(0.0–14.5)	0			–
2nd	14	1.0	(0.0–2.0)	11	1.0	(0.0–3.0)	3	2.0	(1.0–2.0)	0.604
3rd	0

ECG, electrocardiogram; ECHO, echocardiogram; h, hour; IQR, interquartile range.

Nearly all patients who had a troponin test performed in the ED also had an ECG performed (n = 276, 97.5%), 27.9% had further blood tests performed (n = 79) and 11.3% had received a chest X‐ray (n = 32). Of all the patients discharged from hospital with vaccine‐related chest complaints, 7.8% (n = 22) represented to the ED within 14 days with ongoing symptoms.

Of the 283 patients who were judged as having a vaccine‐related presentation, 14 (4.9%) patients had an abnormal troponin blood test either at presentation or during admission. Abnormal troponin tests occurred predominately in males (n = 12, 85.7%). Patients with a normal troponin test stayed in the ED for an average of 3 h, or 5.5 h if the troponin test was abnormal. The 14 patients with abnormal troponin tests were assessed to have vaccine‐associated myocarditis (Table 3) with seven patients classified as ‘definite’ and seven as ‘probable’ myocarditis as per the Brighton criteria. ⁹ No patients were diagnosed with vaccine‐associated pericarditis (Fig. 1). All patients with a definite or probable diagnosis of myocarditis were reported to Surveillance of Adverse Events Following Vaccination In the Community, ¹⁰ the Victorian Vaccine Service and the Australian Therapeutic Goods Administration. ¹¹

TABLE 3.

Investigations and management of patients diagnosed with mRNA COVID vaccine‐related myocarditis

	Myocarditis		Definite		Probable
	n = 14		n = 7		n = 7
ECG, n (%)
Normal	9	(64.2)	4	(57.1)	5	(71.4)
Abnormal	5	(35.7)	3	(42.8)	2	(28.5)
Chest X‐ray, n (%)
Normal	4	(28.5)	2	(28.5)	2	(28.5)
Abnormal	0	(0.0)	0	(0.0)	0	(0.0)
ECHO, n (%)
Normal	6	(42.8)	4	(57.1)	2	(28.5)
Abnormal	4	(28.5)	3	(42.8)	1	(14.2)
MRI, n (%) *includes outpatient MRI
Normal	4	(28.5)	0	(0.0)	4	(57.1)
Abnormal	7	(50.0)	7	(100.0)	0	(0.00)
Cardiac biopsy, n (%)
Normal	0
Abnormal	0
Hospital stay, n (%)
Admitted	13	(92.9)	7	(100.0)	6	(85.7)
Not admitted	1	(7.1)	0	(0.0)	1	(14.3)
Length of stay, n, median (IQR)	40.7	(25.0–54.5)	52.8	(47.0–61.3)	25.0	(5.6–40.7)
Management, n (%)
Respiratory support	0
Cardiac support	0
ICU admission	0
Medication+	1	(7.1)
No specific management	13	(92.9)
Complications at discharge, n (%)
Heart failure	0
Arrhythmias	0
Ongoing chest pain, n (%)	1	(7.1)
Exercise limitations	10	(71.4)
Follow up, n (%)
Vaccinologist	9	(64.2)
Cardiology	13	(92.9)

ECG, electrocardiogram; ECHO, echocardiogram; ICU, intensive care unit; IQR, interquartile range; MRI, magnetic resonance imaging; NG, nasogastric tube.

One patient who had pericarditis in the weeks preceding vaccine administration presented with symptoms within the 14‐day window of mRNA vaccination. This patient with abnormal troponin was judged to not be vaccine‐related.

All patients with myocarditis had an ECG and 35.7% (n = 5) had abnormal findings. Most common abnormalities were a mild ST elevation (28.5%) and short PR interval (7%). Ten (71.4%) patients underwent an echocardiogram, four (28.5%) had abnormal findings; the most common abnormalities were trivial/mild pericardial effusion (21.4%) and normal to low systolic function (7.1%). Eleven of the 14 patients (78.5%) diagnosed with vaccine‐associated myocarditis underwent cardiac MRI either as an inpatient or outpatient, and of those, 63.6% (n = 7) had abnormal findings. Most common abnormalities seen on MRI were late gadolinium enhancement (36.3%), myocardial oedema (36.3%) and epicardial fibrosis (9%). Thirteen patients (92.9%) were admitted to hospital; of these, one (7.1%) received a short course of non‐steroidal anti‐inflammatory medication, whereas the remaining 12 patients required no symptomatic treatment. The patients with myocarditis admitted to hospital stayed an average of 36.1 h. On discharge, one patient (7.1%) had ongoing symptoms of chest pain, and 10 patients (71.4%) were advised to avoid exercise until cardiology follow up. Follow up with cardiology and vaccinologists occurred in 13 and nine patients, respectively. None of the patients received ongoing medication, with information available between 1 month and 18 months after the index visit.

Discussion

In the present study, we linked data from electronic medical records at a tertiary children's hospital and national immunisation registry data. We found a spike in presentations for chest complaints to the ED during the state‐wide rollout of mRNA COVID‐19 vaccines for adolescents. During this period, there was a 30‐fold increase in number of troponin tests performed at its peak compared to baseline. We assessed only 5% of vaccine‐related chest complaints to have vaccine‐associated myocarditis with seven patients classified as ‘definite’ and seven as ‘probable’ myocarditis using internationally accepted criteria. ⁹

This surge in troponin and ECG tests being performed is likely reflective of a large proportion of the adolescent population in Victoria being vaccinated in a short time period. This rapid vaccination was prompted by guidance from ATAGI and the availability of multiple state government‐run vaccine hubs providing vaccine delivery ⁴ in the context of prolonged lockdowns in the preceding months. This led to a significant increase in presentations to the ED and increased demands on the hospital workforce. A collaboration between ATAGI and other relevant national bodies led to guidance for emergency staff that was available before the spike (Fig. S1) and appears to have been followed at the study hospital with most patients. Patients with a normal troponin test and normal ECG were able to be safely discharged within a median of 3 h with minimal additional investigations.

By linking tertiary children's hospital and immunisation registry data, we were able to pinpoint the timing of the vaccination and the relationship with the ED visit. The timing of the presentations was consistent with prior reports of vaccine‐related chest complaints and myocarditis and pericarditis in particular. ¹ , ¹² , ¹³

It is of note that the emergency presentations for vaccine‐related chest complaints were equally split between males and females, but abnormal tests and myocarditis had a male predominance, consistent with previous reports. ¹⁴ , ¹⁵ Chest pain or discomfort (89%) was the most common presenting symptom, followed by shortness of breath (26.5%) and palpitations (23%), with symptoms broadly similar between those with and without myocarditis. Ng et al. ¹⁶ described similar symptoms in patients presenting with chest pain post‐vaccination with or without myocarditis.

We found patients with vaccine‐related chest complaints most commonly presented within 2 days of their second vaccination (57.9% presented on Day 2 after the second vaccination). In contrast to our findings, Ng et al. reported a peak in symptoms occurring 8 days following mRNA COVID‐19 vaccination. ¹⁶ Of the patients with abnormal troponin tests, 85.7% presented 2 days (IQR 2.0–3.0 days) after the second vaccine dose. These data are consistent with previous reports of cardiac inflammation. ¹⁷ , ¹⁸ Similar to a systematic review by Pillay et al., ¹⁸ we found no difference in the incidence of vaccine‐related chest pain based on pre‐existing conditions or risk factors.

Our research adds further evidence that vaccine‐associated myocarditis is a usually mild and self‐limiting illness. Of the 14 patients identified as having vaccine‐associated myocarditis, 13 patients were admitted to hospital, but only one patient (7.1%) required treatment (NSAIDs). No patients required ICU admission or vasoactive agents.

Limitations

Although the present study provided a unique link between hospital and immunisation registry data, there were several limitations. The study was single‐centre and retrospective. Although we followed guidance for high‐quality data extraction, clinical information extracted was dependent upon accurate documentation by the treating clinicians. ¹⁹ To minimise interpretation errors for patients with abnormal troponin or vaccine‐associated myocarditis or pericarditis, a further chart review was performed by a cardiologist (BJ) to validate patient classification and exclusion. In addition, based on no patient being diagnosed with vaccine‐related pericarditis, a second review of all abnormal ECGs during the study period was performed to ensure no patient with pericarditis had been missed. No further patients with pericarditis were identified following this.

As a result of the identification of the patient cohort by troponin test ordered, it is possible that the present study has missed patients who had troponin tests obtained in the community only and were referred to hospital for further assessment, or patients who presented with chest complaints but did not have a troponin test taken. There was a low number of children under 12 years of age who received mRNA COVID‐19 vaccinations based on ATAGI advice at the time (roll out January 2022 for 5–11 year olds) to extrapolate the findings to this younger cohort. It will also be important to monitor new vaccines being developed with the mRNA platform, such as influenza vaccines, for potential AEs of special interest such as myocarditis and pericarditis. Although this may be an AE specific to COVID‐19 vaccines, monitoring for new vaccines will help support vaccine confidence in the future.

Conclusion

The present study indicates that a large spike of mRNA vaccine‐related emergency presentations for chest complaints occurred during a state‐wide rollout to adolescents. However, vaccine‐associated myocarditis was a rare, benign and self‐limiting illness.

Supporting information

Table S1. Brighton Collaboration criteria for myocarditis.

Table S2. Brighton Collaboration criteria for pericarditis.

Figure S1. mRNA chest pain guideline ‐ algorithm.

Data availability statement

Author elects to not share data.