Medium-term outcomes of myocarditis and pericarditis following BNT162b2 vaccination among adolescents in Hong Kong

ABSTRACT In this study, we examined the clinical and electrophysiological outcomes of adolescents in Hong Kong who developed myocarditis or pericarditis following BNT162b2 vaccination for COVID-19, and followed-up for 60–180 days after their initial diagnosis. Clinical assessments included electrocardiogram (ECG) and echocardiogram at the initial admission and follow-up were compared. Treadmill testing was also performed in some cases. Between 14 June 2021 and 16 February 2022, 53 subjects were approached to participate in this follow-up study, of which 28 patients were followed up for >60 days with a median follow-up period of 100 days (range, 61–178 days) and were included in this study. On admission, 23 patients had ECG abnormalities but no high-grade atrioventricular block. Six patients had echocardiogram abnormalities, including reduced contractility, small rim pericardial effusions, and hyperechoic ventricular walls. All patients achieved complete recovery on follow-up. After discharge, 10 patients (35.7%) reported symptoms, including occasional chest pain, shortness of breath, reduced exercise tolerance, and recurrent vasovagal near-syncope. At follow-up, assessments, including ECGs, were almost all normal. Among the three patients with possible ECG abnormalities, all their echocardiograms or treadmill testings were normal. Sixteen patients (57.1%) underwent treadmill testing at a median of 117 days post-admission, which were also normal. However, at follow-up, there was a significant mean bodyweight increase of 1.81 kg (95%CI 0.47-3.1 kg, p = 0.01), possibly due to exercise restriction. In conclusion, most adolescents experiencing myocarditis and pericarditis following BNT162b2 vaccination achieved complete recovery. Some patients developed non-specific persistent symptoms, and bodyweight changes shall be monitored.


Introduction
The BNT162b2 vaccine was demonstrated to be largely safe and effective against COVID-19 in clinical trials. [1,2] However, following the widespread use of the BNT162b2 vaccine worldwide, numerous reports of myocarditis and pericarditis, particularly in male adolescents receiving their second dose 21 days after the first. [3][4][5] In Hong Kong, our pharmacovigilance population-based cohort study showed that all patients had mild symptoms during their acute admission and did not require inotropic, ventilatory, and circulatory support, [6] which largely agreed with the most recent evidence on the mild clinical presentation during the hospital course and short-term outcomes in adolescents and children. [7][8][9][10][11][12][13] The United States Centers for Disease Control and Prevention (CDC) have also surveyed COVID-19 patients and their clinicians to understand the long-term health effects of myocarditis following mRNA vaccination. [14] However, data on the medium-term clinical outcomes at follow-up is lacking. Here, we describe the medium-term (60-180 days) clinical and noninvasive electrophysiological outcomes of Hong Kong adolescents diagnosed with myocarditis following BNT162b2 vaccination.

Methods
This was an observational follow-up study of the population-based vaccine safety monitoring program initiated by the Hong Kong regulatory authority. [6] Following BNT162b2 vaccination, all children diagnosed with myocarditis or pericarditis according to the definitions for myocarditis and pericarditis of the Cardiovascular Injury-Coalition for Epidemic Preparedness Innovations (CEPI) and the Brighton Working Group were followed up. [15] These patients were referred from all regional hospitals to a centralized clinic set up at the Hong Kong Children's Hospital for follow-up and counseling. Visits to the clinic were arranged within 12 weeks of the patient's initial admission. A pediatric cardiologist (ST) and two pediatric immunologists (GC and MK) provided the follow-up assessments.
The inclusion criteria for participation in this study were: 1) adolescents between 12 and 18 years of age at the time of diagnosis, 2) follow-up was between 60 and 180 days from the initial diagnosis, and 3) written consent was provided. The exclusion criteria were: 1) > 18 or <12 years of age at the time of diagnosis, 2) received COVID-19 vaccines other than the BNT162b2 vaccine, 3) had an alternative cause such as viral infection, 4) follow-up was <60 or >180 days from the initial diagnosis, and 5) did not provide consent for the study.
Serial ECGs, serial echocardiograms, and cardiac magnetic resonance imaging (cMRI) performed during acute admission and follow-up at the referring hospitals were reviewed by a single investigator (ST). All ECGs were performed at a speed of 25 mm/second and 10 mm/mV gain. Reviewed parameters included heart rate, QRS axis, PR interval, QRS duration, QT and QTc intervals, ST/T wave abnormalities, including T wave inversion (excluding aVR and V1-V2) and biphasic or flat T waves, and QRS voltages. Treadmill testing conducted around three months after discharge was only available at some centers due to the limited resources during the COVID-19 pandemic. All subjects were advised to avoid strenuous exercise for three months after discharge or until treadmill testing was normal.
Clinical symptoms and bodyweight changes between acute admission and follow-up were recorded and reviewed. One sample T-test was used to determine any significant changes in the bodyweight, and a p-value of <0.05 is considered statistically significant. Assessments included serial ECGs, serial echocardiograms, cMRI, and treadmill testing.
This study was approved by the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (Reference: UW 21-548), the Hospital Authority Central Institutional Review Board (CIRB-2021-003-4), and the Department of Health Ethics Committee (LM21/ 2021). Written consent and assent were obtained from the parents and participating patients, respectively.

Results
Between 14 June 2021 and 16 February 2022, 53 adolescents who had received the BNT162b2 vaccination were diagnosed with myocarditis or pericarditis according to the CEPI and Brighton Working Group Criteria. Among these patients, 28 fulfilled the eligibility criteria and consented to participate in this study. Table 1 gives a summary of the patient characteristics, which showed 17 (60.7%) had myocarditis and 11 (39.3%) had perimyocarditis, of which 17 (60.7%) were definite cases, seven (25.0%) were probable cases, and four (14.3%) were possible cases. The median follow-up period was 100 days (range from 61 to 178 days). Other potential viral infections, including SARS-CoV-2 infection, were excluded as possible causes of myocarditis and pericarditis. Details of the investigations have been published elsewhere. [6] At follow-up, 18 (64.3%) patients were asymptomatic and 10 (35.7%) had non-specific symptoms. Five patients complained of occasional chest pains or discomfort (patients 5, 11, 14, 20, and 27), two had occasional shortness of breath (patients 8 and 17), one (patient 25) had reduced exercise tolerance, and one (patient 19) had recurrent vasovagal near-syncope. All patients showed normal ECGs, echocardiograms, and treadmill ECGs that did not suggest progression to chronic heart failure. One patient (patient 16) had persistent sinus tachycardia and incidentally had goiter with hyperthyroidism unrelated to the myocarditis, which was treated with carbimazole and propranolol. At follow-up, there was a significant mean bodyweight gain of 1.81 kg (95%CI 0.47-3.1 kg, p = 0.01). Overall, 18 subjects (64.3%) gained between 0.5 and 8.8 kg bodyweight, whereas five subjects (17.9%) lost between 0.2 and 7 kg. Detailed physical examination and biochemical analysis, including thyroid function tests, for subjects with weight loss, were performed and did not reveal any significant pathologies.
Repeat ECGs in 27 patients performed at their referring hospitals were retrieved and reviewed. Analysis of the ECGs at the initial admission showed t- hat 10 patients (35.7%) had sinus tachycardia (>100 bpm). No arrhythmias were observed, except for two cases of rare isolated premature ventricular contractions in one patient with definite myocarditis and one with definite perimyocarditis. Transient 1stdegree atrioventricular block was noted in one patient with perimyocarditis, but no patients developed highgrade AV block. An elevated ST segment (defined as ≥2 mm) was observed in 18 patients. A mild reduction in QRS voltage was noted in three patients with myocarditis and two with perimyocarditis. All ECGs performed during the last follow-up were mostly normal, although one patient had possible left ventricular hypertrophy and ST elevation in V3 and V4, one had ST elevation, and one had short PR intervals.
All echocardiograms reported in the electronic medical records were retrieved for review. Analysis of the echocardiograms on admission showed normal cardiac function in 16 (94.1%) out of 17 patients with myocarditis and in nine (81.8%) out of 11 patients with perimyocarditis. One patient with myocarditis and two with perimyocarditis had borderline left ventricular (LV) functions. Three patients with perimyocarditis had a small amount of pericardial effusion. One patient with perimyocarditis had incidental findings of a small coronary fistula. Of the three patients with borderline LV dysfunction (one with myocarditis and two with perimyocarditis), two had repeat echocardiograms within a week of the resolution of LV function. One patient with perimyocarditis did not receive a repeat echocardiogram at admission, although the repeat echocardiogram 7 months later was normal.
Except for one patient with possible myocarditis, cMRIs were performed on 27 patients on admission. Analysis of cMRI of patients with myocarditis (n = 16) showed one (6.23%) had borderline LV function, one (6.3%) had borderline RV function, and one (6.3%) had a borderline biventricular function. Nine patients (56.3%) also exhibited elevated extracellular volume, T2 mapping, and early and late gadolinium enhancement. Analysis of cMRI of patients with perimyocarditis (n = 11) showed three (27.3%) had borderline biventricular functions and one (9.1%) had borderline LV function. Six patients (54.5%) also had features of myocarditis, and eight (72.7%) had features of pericarditis (subepicardial or epicardial late gadolinium enhancement and pericardial effusion). Among those patients with borderline LV function, one with myocarditis and one with perimyocarditis also had echocardiograms showing borderline LV systolic function.
Due to the restrictions and limited resources during the ongoing COVID-19 outbreaks, fewer scheduled treadmill tests were available for patients. Only 16 (57.1%) out of 28 patients (7 in the myocarditis group and 9 in the perimyocarditis group) underwent treadmill testing at a median of 117 days post-admission. The results from the 16 treadmill tests showed that the patients had normal heart rate and blood pressure response, with no inducible ischemic changes or arrhythmias. However, one patient with probable myocarditis exhibited a T wave inversion in the inferior leads and anterolateral leads during the treadmill test performed two months post-admission. This patient developed perimyocarditis one month after the second dose of the BNT162b2 vaccine.

Discussion
This is the first study to investigate the medium-term follow-up of adolescents with myocarditis and pericarditis following BNT162b2 vaccinations by an in-depth examination of clinical and cardiac outcomes. Most patients achieved complete recovery, as seen by normal serial ECGs, serial echocardiograms, and treadmill ECGs at follow-up. Only a small percentage of patients experienced chronic symptoms, including intermittent non-exertional chest pain and shortness of breath, but no specific treatment was offered as no cardiovascular pathologies could be identified. It is unlikely that these symptoms were psychosomatic, as these subjects did not complain of emotional distress after discharge or have psychiatric illnesses or mood disorders. Although echocardiograms showed most had a normal cardiac function, the use of echocardiograms alone can miss the presence of borderline ventricular function. Long-term follow-up in these patients is needed to monitor their progress.
Exercise restriction for 3-6 months is standard practice for managing patients with myocarditis and pericarditis. [16] Given that our patients only had mild symptoms, we recommended they avoid strenuous exercise for only three months, as prolonged exercise restriction may not be beneficial. [6] An unintended consequence of the exercise restriction has possibly led to many of these patients having increased bodyweight after this short period. Therefore, we recommend that patients should be given dietary and activity guidance to avoid bodyweight fluctuations, which is associated with higher allcause mortality, cardiovascular disease morbidity, and hypertension, as demonstrated in a systematic review and meta-analysis. [17] To ensure physical fitness for the resumption of physical activities such as competitive sports, a series of cardiac assessments, including ECGs, echocardiograms, and treadmill tests, should be performed around 12 weeks after discharge.
Although the latest CDC interim recommendations suggest that subjects with a history of myocarditis or pericarditis after a dose of an mRNA vaccine should generally avoid receiving any further COVID-19 vaccines, [18] a recent population-based study conducted in Hong Kong comparing adults who had received the BNT162b2 and CoronaVac vaccines showed that the latter group had a significantly lower risk of myocarditis and pericarditis. [3] Our local study also showed that both BNT162b2 and CoronaVac vaccines are able to confer robust T cell immunity. [19] As the Hong Kong Government has approved the use of Cor-onaVac in children older than three years, we recommend that patients in Hong Kong who develop myocarditis and pericarditis following the BNT162b2 vaccine should receive the CoronaVac as their booster dose.
Findings from this study should be interpreted with the following caveats. First, the echocardiograms and treadmill tests were performed in different centers due to limited resources. Nevertheless, pediatric cardiologists with similar years of experience working in close collaboration performed and reported these investigations. We also acknowledge that follow-up echocardiograms and treadmill testing were unavailable for some patients as these follow-up investigations were also performed in the referring hospitals and are limited by resource availability. However, we expect the follow-up echocardiograms to be largely normal as many of these cases had a normal baseline echocardiogram even during their acute admissions. Second, due to limited resources, follow-up cMRI was not arranged as part of the follow-up protocol. However, as most subjects had mild symptoms at the initial assessments, they were expected to have normal cMRI findings at the follow-up. Third, we only recruited adolescent patients in this study. Hence, our results may not apply to adults. Last but not least, as our study sample was from the Chinese population, we cannot exclude the possibility of ethnic differences in the extent or rate of recovery in other populations. Further studies with larger sample sizes and in different populations are warranted.

Conclusion
Adolescents with myocarditis and perimyocarditis following BNT162b2 Vaccination were recommended exercise restriction for three months, with most achieving complete recovery. The gain in bodyweight during recovery was likely due to the exercise restriction. Thus appropriate activity and dietary education should be provided at discharge. Shortening the exercise restriction period to less than three months could be considered in patients with pericarditis without myocardial involvement or if abnormal ECG findings are rapidly resolved. Long-term follow-up will be warranted for these individuals regarding their cardiovascular risk outcome. Information and advice on future vaccinations should also be incorporated into the follow-up.