This page undergoes regular review and was last comprehensively reviewed on January 18, 2023. Some sections may reflect more recent updates.
Overview
In addition to common adverse effects such as injection site pain, fatigue and headache, some more serious safety signals have been identified in relation to COVID-19 vaccination. Here, we review key evidence and clinical recommendations related to these rarer presentations. For more information about common side effects, refer to our mRNA and viral vector COVID-19 vaccine pages.
Existing information about the safety of mRNA bivalent boosters is very limited and comprises information largely collected from clinical trials testing a bivalent booster with a BA.1 rather than a BA.4/BA.5 antigen. However, preliminary data suggests that safety outcomes with bivalent booster vaccines are mild and similar to safety outcomes with monovalent boosters.
Safety signals under investigation
Recently, CDC’s Vaccine Safety Datalink, a real-time vaccine safety surveillance system, identified data that met the statistical criteria to prompt additional investigation of ischemic stroke in adults 65 years of age and older who received the Pfizer-BioNTech COVID-19 bivalent booster vaccine. This statistical safety signal, which identified a potential for slightly increased risk of ischemic stroke, is still being investigated by VSD. This preliminary signal was not identified with the bivalent Moderna COVID-19 booster vaccine, nor was it identified in analyses using the Centers for Medicare and Medicaid Services database, the Veterans Affairs database, the Vaccine Adverse Event Reporting System or Pfizer-BioNTech’s global safety database. It was also not identified in any other age groups except adults 65 years old and older. No other country so far has identified a similar safety signal.
At this time, no change in vaccination practice is recommended.
Myocarditis
Myopericarditis (inflammation of the cardiac lining and/or muscle) is a known, rare complication of vaccination that was best recognized in the pre-COVID-19 era following immunization against smallpox (Halsell, June 2003; Su, January 2021). Mechanisms underlying vaccine-associated myopericarditis are uncertain but favored to represent nonallergic, immune-mediated effects in the setting of post-vaccination inflammatory milieu including possible antibody cross-recognition of cardiac tissue antigens (“molecular mimicry”) (Bozkurt, July 2021).
Although not noted during the landmark clinical trials of COVID-19 vaccines, myopericarditis temporally related to mRNA vaccination (Pfizer BioNTech 162b2 or Moderna mRNA-1273) has become better documented in the post-authorization phase. Specifically, there is an association between COVID-19 mRNA vaccination and higher relative rates of myopericarditis, particularly in younger males after the second vaccine dose. Absolute case numbers, however, are very low and predominately characterized by a short, self-limited syndrome. It is important to assess these rare adverse effects in the context of both the serious morbidity associated with COVID-19 itself and significant benefits afforded by COVID-19 vaccination.
Clinical Presentation
A typical clinical presentation of vaccine-associated myopericarditis consists of chest pain and/or dyspnea beginning 1-5 days after administration of a vaccine dose, particularly the second of the two-dose mRNA series. Mild fever has been reported in several cases (Marshall, June 2021; Mevorach, October 2021), though molecular testing for acute SARS-CoV-2 infection is uniformly negative.
This syndrome is most frequently observed in younger adult males (see Incidence and Epidemiology). Several early series described a total of 20 males (85% <30 years of age) with similar symptomatology predominately after Pfizer vaccination who presented with elevated inflammatory markers (e.g., C-reactive protein) and moderately elevated troponin levels (Marshall, June 2021; Rosner, June 2021; Abu Mouch, June 2021). A single larger series of 23 male U.S. military personnel (median age 25) described a very similar syndrome occurring 1-4 days following a second dose of mRNA vaccination; two thirds had received the Moderna vaccine (Montgomery, June 2021).
Cardiac evaluation of myopericarditis cases has most commonly identified nonspecific electrocardiographic abnormalities, particularly S-T segment or T wave changes. Cardiac function may be normal by echocardiography, especially in younger patients and those with isolated pericarditis, though there have been several reports of mild to moderately depressed left ventricular ejection fraction (Montgomery, June 2021) (see Outcomes). Cardiac MRI, if performed, frequently shows evidence of myopericarditis such as patchy late gadolinium enhancement. Management is conservative, with therapy primarily consisting of nonsteroidal anti-inflammatory medications, with administration of intravenous immune globulin and/or corticosteroids for select cases.
Outcomes
The characteristic clinical course of vaccine-associated myopericarditis includes brief hospitalization and significant symptom improvement occurring within a week of onset, without need for higher-level care or readmission. The largest published myopericarditis series utilized claims data and was adjudicated by clinical experts in Israel, identifying 136 definite or probable cases temporally related to vaccination among >5 million Pfizer vaccinees (Mevorach, October 2021). The majority of cases (95%) were considered mild based on quick resolution of symptoms and lab abnormalities, with short hospital stays. Similar findings were described in a U.S. health system study of 2.4 million vaccinees, where 15 total cases of myopericarditis were diagnosed (all in men <32 years of age), of whom 100% required hospital stays ≤3 days.
The Mevorach et al. national series did identify seven total cases of severe cardiac dysfunction temporally related to vaccination, including a single case of fatal myocarditis in a 22-year-old male among 358,511 vaccinees <25 years of age. Two additional severe presentations of myocarditis have been described in middle-aged adults 1-2 weeks after mRNA vaccination, characterized by depression of left ventricular function and shock, with one death (Verma, September 2021). Although myocarditis was diagnosed by biopsy in these latter two cases, causal relation to vaccination was not definitively established.
Epidemiology & Incidence Estimates
Population and health system-level studies have confirmed a consistent signal of increased relative incidence of myocarditis after vaccination, approximately 2-to-3-fold higher than unvaccinated control populations (Barda, September 2021; Mevorach, October 2021; Simone, October 2021). Absolute observed case numbers, however, remain very low, amounting to 5-30 excess myocarditis cases per million vaccine doses in the general population, with zero expected attributable deaths given mild clinical course. Importantly, the risk of myocarditis following COVID-19 vaccination is reduced in all age groups compared to the risk of myocarditis following SARS-CoV-2 infection (Block, April 2022).
Much benefit/risk focus has centered on adolescent and young adult males given higher relative incidence of vaccine-associated myopericarditis in this group. Adjudicated CDC review of VAERS data (passive reporting) detected the highest case rates in males ages 16-17, though this still accounted for <75 cases per million vaccinees (i.e., ~1/14,000 vaccinees) (Rosenblum slides, August 2021 [PDF]). A recent VAERS and v-safe analysis reported in a CDC MMWR indicated risk of myocarditis in adolescent males (aged 12-15 or 16-17) was less common after a Pfizer-BioNTech booster dose compared to the second dose of the Pfizer-BioNTech primary series (Hause, March 2022). An Israeli series found about 2-fold higher incidence of myopericarditis (~1/6,636) among boys ages 16-19 years than in the U.S. series after two-dose Pfizer vaccination (Mevorach, October 2021). In both analyses, myocarditis incidence was 10-fold lower in female versus male adolescents, with observed case rates of ≤1/100,000 in young female vaccinees.
Of emerging interest is potential higher incidence of myopericarditis among recipients of the Moderna versus the Pfizer vaccine. For example, in unadjusted analysis, a Canadian health ministry detected a 6-fold higher rate among male Moderna vaccinees ages 18-24 after the second vaccination (47 cases, rate ~1/3,500) as compared with Pfizer vaccinees (26 cases, ~1/23,000) (Public Health Ontario, December 2021 [PDF]). Several Nordic countries have considered restricting Moderna vaccination in younger males (Paterlini, October 2021), though other European countries continue to strongly endorse the benefits of mRNA vaccination in young adults (Haaf, October 2021).
However, recent evidence from v-safe does not indicate a substantial difference in risk of myocarditis between Moderna and Pfizer-BioNTech recipients (Shimabukuro, July 2022 [PDF]). A pattern of risk of myocarditis or pericarditis with the Johnson & Johnson/Janssen vaccine has not been identified, according to CDC. Additionally, there is very limited population-level data for the risk of myocarditis after Novavax (NVX-CoV2373) vaccine, but clinical trial evidence suggests a possible slight increased risk of myocarditis in vaccine recipients compared to placebo recipients. Ongoing surveillance and review of trials data in the U.S. continues.
Risk versus Benefit of Vaccination in Younger Populations
Although there is a clear association between mRNA vaccination and the uncommon complication of myopericarditis, the benefits of vaccination in younger populations are significant. For example, in CDC’s analysis, for every million mRNA vaccine doses in males ages 12-19 years, ~40 cases of myocarditis would be expected (with zero deaths), whereas these vaccines would prevent 11,000 cases of COVID-19, including 650 hospitalizations, 138 ICU admissions and six deaths (Gargano, July 2021). Additional national-level analyses in the U.S. support the protective benefits of vaccination, indicating that even in the context of the Delta variant, hospitalization rates remain 10-fold lower among vaccinated versus unvaccinated children and adolescents (Delahoy, September 2021).
In addition to framing vaccine-associated myopericarditis risk versus that of a control, unvaccinated population, it is important to consider cardiac complications due to COVID-19 itself that may be prevented through vaccination. CDC review of hospital claims data noted a >40% increase in myocarditis diagnoses in 2020 versus 2019 (pre-pandemic), with cases peaking along with COVID-19 waves. Overall myocarditis incidence was 0.15% in patients with COVID-19 (i.e., 1/667 infections) in contrast to <0.01% in those without COVID-19 (i.e., <1/10,000 persons). COVID-19 was associated with an overall 16-fold higher relative risk of myocarditis, including >30-fold higher risk for children <16 years of age (Boehmer, September 2021). The described incidence of myocarditis associated with COVID-19 in children is 10-to-15-fold higher than the reported case rates of myopericarditis after mRNA vaccination in adolescents (see Epidemiology & Incidence Estimates).
Vaccine-Induced Thrombotic Thrombocytopenia
In post-authorization studies, both the Oxford-AstraZeneca and Johnson & Johnson/Janssen COVID-19 vaccines have been demonstrated to be safe. One safety signal identified in post-authorization surveillance studies of both these viral-vector vaccines is an association with rare blood clotting events with low platelet counts — thrombosis with thrombocytopenia syndrome (TTS) and vaccine-induced thrombotic thrombocytopenia (VITT) — which were not observed in the Phase 3 trials. The American Society of Hematology maintains an up-to-date resource on TTS/VITT that includes a case definition and recommendations regarding diagnostic evaluation and acute management.
Clinical Presentation
VITT is a syndrome characterized by venous or arterial thrombosis associated with thrombocytopenia and detectable anti-platelet factor 4 antibodies that occurs within 3 weeks after receipt of either the Oxford-AstraZeneca or Johnson & Johnson/Janssen COVID-19 vaccine. The American Society of Hematology has developed a case definition with recommendations for diagnostic evaluation and management of patients with suspected VITT.
Epidemiology & Incidence Estimates
In post-authorization surveillance of the Johnson & Johnson/Janssen COVID-19 vaccine, a small number (N=6) of rare thrombotic events — cerebral venous sinus thrombosis (CVST) — associated with thrombocytopenia were identified among vaccine recipients based on data reported to VAERS. This led to a brief pause in the use of this vaccine on April 13, 2021, and a review of safety data by CDC’s Advisory Committee on Immunization Practices on April 23, 2021. In this review, 15 total cases of thrombosis with thrombocytopenia syndrome were identified, including 12 cases of CVST. All the TTS cases occurred among women, and 13 of 15 were in women aged 18-49 years old. The median age of the case patients was 37 years, and the median interval from vaccination to symptom onset was 8 days (range, 6-15 days) (MacNeil, April 2021). There was one case of CVST with thrombocytopenia in a male during the Phase 3 trial of Ad26.COV2.S (Sadoff, April 2021).
The Oxford-AstraZeneca COVID-19 vaccine has been associated with both arterial and venous thrombotic events. In an analysis of data reported to EudraVigilance (a drug safety reporting system for the European Union), the European Medicines Agency identified 169 cases of CSVT and 53 cases of splanchnic vein thrombosis following receipt of the Oxford-AstraZeneca COVID-19 vaccine (out of more than 34 million doses administered). These thrombotic events occurred concurrently with thrombocytopenia, usually within 2 weeks of receipt of the vaccine, and mostly among women under age 60. Subsequently, three independent case series from Norway (Schultz, April 2021), Germany/Austria (Greinacher, April 2021) and the United Kingdom (Scully, April 2021) described a total of 39 patients (27 women, 12 men) with venous thromboses associated with thrombocytopenia that occurred within 4 weeks (range, 5-24 days) after vaccination with the Oxford-AstraZeneca COVID-19 vaccine. Given the resemblance of this syndrome to heparin-induced thrombocytopenia, in two of these studies the investigators tested case patients for anti-platelet factor 4 antibodies and found elevated titers in the majority of patients, leading them to propose a new syndrome called VITT.
Recommendations
The risk of VITT is still lower after vaccination with Oxford-AstraZeneca or Johnson & Johnson/Janssen COVID-19 vaccines, compared to the risk of blood clot with COVID-19 infection. However, as a result of rare reports of VITT, CDC’s Advisory Committee on Immunization Practices preferentially recommends an mRNA vaccine over the Johnson & Johnson/Janssen vaccine, given the high availability of mRNA vaccines in the U.S. This recommendation is similar to a preferential recommendation made by Canada’s National Advisory Committee on Immunization to receive an mRNA vaccine over the Oxford-AstraZeneca vaccine, where mRNA vaccines are available. (The Oxford-AstraZeneca vaccine has not yet been approved for use in the U.S.)
Neurological Events
In the routine process of vaccine safety assessment, rare neurological outcomes are often investigated among potential outcomes (even in cases where vaccine clinical trials did not identify concerns for these outcomes). Such rare neurological outcomes may include Guillain-Barré syndrome and facial paralysis (Bell’s palsy). Neither GBS nor Bell’s palsy was identified as a consistent or statistically significant outcome associated with vaccination in clinical trials for mRNA vaccines or the Johnson & Johnson/Janssen vaccine (Baden, February 2021) or the Novavax recombinant subunit vaccine (Dunkle, December 2021; Heath, June 2022). This finding is not unexpected, as these outcomes are very rare and may require population-level studies to assess risk post-vaccination.
In a large-scale study conducted using data from the Vaccine Safety Datalink, GBS was not more common in mRNA vaccine recipients than background rates of GBS (unadjusted incidence rate: 1.3 per 100,000 person-years [95% CI: 0.7-2.4 events per 100,000 person-years]). However, GBS was more common than background rates in Johnson & Johnson/Janssen vaccine recipients (unadjusted incidence rate: 32.4 per 100,000 person-years [95% CI: 14.8-61.5 events per 100,000 person-years]) (Hanson, April 2022; ACIP slides, July 2021 [PDF]).
In a large-scale study conducted using data from the Vaccine Safety Datalink, Bell’s palsy was not more common than background rates after receipt of COVID-19 vaccines (Klein, September 2021), consistent with another large-scale observational study conducted in Israel (Shasha, January 2022) and a case-control study also conducted in Israel (Shemer, June 2021).
There have been reports of sudden sensorineural hearing loss, including tinnitus, after COVID-19 vaccination. An investigation of SSNHL outcomes using VAERS identified no relationship between COVID-19 vaccines (Moderna, Pfizer-BioNTech and Johnson & Johnson/Janssen) and SSNHL (Formeister, February 2022). Another investigation of SSNHL outcomes using a large-scale dataset in Israel found a potential relationship between Pfizer-BioNTech vaccine receipt and SSNHL (Yanir, February 2022) but noted that the effect size identified was small (standardized incidence ratios of SSNHL: 1.35 [95% CI: 1.09-1.65]) and that the benefits of Pfizer-BioNTech vaccine still outweighed the risks. Importantly, neither study identified strong demographic or other risk criteria associated with SSNHL outcomes, or considered the severity of such outcomes. WHO recently reported slightly higher than expected numbers of cases of hearing loss and tinnitus identified after COVID-19 vaccination in Vigibase (a drug safety monitoring database). Most of the events reported in Vigibase were classified as nonserious. WHO has noted that SSNHL outcomes should be monitored further.