Vaccine 42 (2024) 2200–2211

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

COVID-19 vaccines and adverse events of special interest: A multinational

Global Vaccine Data Network (GVDN) cohort study of 99 million
vaccinated individuals
K. Faksova a, *, D. Walsh b, c, Y. Jiang b, c, J. Griffin c, A. Phillips d, A. Gentile e, J.C. Kwong f, g, h,
K. Macartney d, i, M. Naus j, n, Z. Grange k, S. Escolano l, G. Sepulveda m, A. Shetty m, A. Pillsbury d,
C. Sullivan k, Z. Naveed j, n, N.Z. Janjua j, n, N. Giglio e, J. Perälä o, S. Nasreen f, p, x, H. Gidding d, i,
P. Hovi q, T. Vo r, F. Cui s, L. Deng d, L. Cullen k, M. Artama r, H. Lu b, c, H.J. Clothier c, m, K. Batty t,
J. Paynter u, H. Petousis-Harris c, u, J. Buttery c, m, v, S. Black c, u, A. Hviid a, w
a
Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark
b
Department of Statistics, University of Auckland, New Zealand
c
Global Vaccine Data Network, Global Coordinating Centre, Auckland, New Zealand
d
National Centre for Immunisation Research and Surveillance, Westmead, New South Wales, Australia
e
Department of Epidemiology, Ricardo Gutierrez Children Hospital, Buenos Aires University, Argentina
f
ICES, Toronto, Ontario, Canada
g
Public Health Ontario, Toronto, Ontario, Canada
h
Department of Family and Community Medicine, Temerty Faculty of Medicine and the Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario,
Canada
i
The University of Sydney, Australia
j
British Columbia Centre for Disease Control, Vancouver, British Columbia, Canada
k
Public Health Scotland, Glasgow, Scotland, United Kingdom
l
Université Paris-Saclay, UVSQ, Inserm, CESP, High Dimensional Biostatistics for Drug Safety and Genomics, Villejuif, France
m
Murdoch Children’s Research Institute, Parkville, Victoria, Australia
n
School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Canada
o
Department of Health Security, Finnish Institute for Health and Welfare, Helsinki, Finland
p
Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
q
Department of Public Welfare, Finnish Institute for Health and Welfare, Helsinki, Finland
r
Faculty of Social Sciences, Tampere University, Finland
s
School of Public Health, Peking University, China
t
Auckland UniServices Limited at University of Auckland, New Zealand
u
School of Population Health, University of Auckland, New Zealand
v
University of Melbourne, Parkville, Victoria, Australia
w
Pharmacovigilance Research Center, Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen,
Denmark
x
School of Public Health, SUNY Downstate Health Sciences University, Brooklyn, NY, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Background: The Global COVID Vaccine Safety (GCoVS) Project, established in 2021 under the multinational
Vaccine safety surveillance Global Vaccine Data Network™ (GVDN®), facilitates comprehensive assessment of vaccine safety. This study
Pharmacovigilance aimed to evaluate the risk of adverse events of special interest (AESI) following COVID-19 vaccination from 10
Adverse events following immunization
sites across eight countries.
Adverse events of special interest
COVID-19
Methods: Using a common protocol, this observational cohort study compared observed with expected rates of 13
Observed vs. expected analysis selected AESI across neurological, haematological, and cardiac outcomes. Expected rates were obtained by
participating sites using pre-COVID-19 vaccination healthcare data stratified by age and sex. Observed rates were
reported from the same healthcare datasets since COVID-19 vaccination program rollout. AESI occurring up to
42 days following vaccination with mRNA (BNT162b2 and mRNA-1273) and adenovirus-vector (ChAdOx1)

* Corresponding author at: Department of Epidemiology Research, Statens Serum Institut, Artillerivej 5, Copenhagen S 2300, Denmark.
E-mail address: [email protected] (K. Faksova).

https://doi.org/10.1016/j.vaccine.2024.01.100
Received 29 January 2024; Accepted 30 January 2024
Available online 12 February 2024
0264-410X/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
K. Faksova et al. Vaccine 42 (2024) 2200–2211

vaccines were included in the primary analysis. Risks were assessed using observed versus expected (OE) ratios
with 95 % confidence intervals. Prioritised potential safety signals were those with lower bound of the 95 %
confidence interval (LBCI) greater than 1.5.
Results: Participants included 99,068,901 vaccinated individuals. In total, 183,559,462 doses of BNT162b2,
36,178,442 doses of mRNA-1273, and 23,093,399 doses of ChAdOx1 were administered across participating sites
in the study period. Risk periods following homologous vaccination schedules contributed 23,168,335 person-
years of follow-up. OE ratios with LBCI > 1.5 were observed for Guillain-Barré syndrome (2.49, 95 % CI:
2.15, 2.87) and cerebral venous sinus thrombosis (3.23, 95 % CI: 2.51, 4.09) following the first dose of ChAdOx1
vaccine. Acute disseminated encephalomyelitis showed an OE ratio of 3.78 (95 % CI: 1.52, 7.78) following the
first dose of mRNA-1273 vaccine. The OE ratios for myocarditis and pericarditis following BNT162b2, mRNA-
1273, and ChAdOx1 were significantly increased with LBCIs > 1.5.
Conclusion: This multi-country analysis confirmed pre-established safety signals for myocarditis, pericarditis,
Guillain-Barré syndrome, and cerebral venous sinus thrombosis. Other potential safety signals that require
further investigation were identified.

1. Introduction 2.2. Data source and study population

Since declaration of the COVID-19 pandemic by the World Health The GCoVS Project compiled electronic healthcare data on AESI
Organization (WHO) on March 11, 2020 [1] more than 13.5 billion doses related to COVID-19 vaccines from participants across multiple sites
of COVID-19 vaccines have been administered worldwide [2]. As of within the GVDN network, including Argentina, Australia – New South
November 2023, at least 70.5 % of the world’s population had received at Wales, Australia – Victoria, Canada – British Columbia, Canada –
least one dose of a COVID-19 vaccine [2]. This unparalleled scenario un­ Ontario, Denmark, Finland, France, New Zealand, and Scotland [10].
derscores the pressing need for comprehensive vaccine safety monitoring The healthcare data comprised of either individual- or population-level
as very rare adverse events associated with COVID-19 vaccines may only data, depending on the availability in the study sites (Supplementary
come to light after administration to millions of individuals. Table 1).
In anticipation of this unprecedented global rollout of COVID-19 Immunization registers containing individual-level vaccination data
vaccines, the Safety Platform for Emergency vACcines (SPEAC) initia­ were utilized by the majority of study sites. These registers covered the
tive formulated a list of potential COVID-19 vaccine adverse events of same population and geographic region as the data sets used to calculate
special interest (AESI) in 2020 [3]. AESI selection was based on their background rates. We also examined population-level data on vaccina­
pre-established associations with immunization, specific vaccine plat­ tion uptake using regularly updated dashboards from the study sites. If
forms or adjuvants, or viral replication during wild-type disease; theo­ the number of individuals vaccinated in specific age and gender groups
retical concerns related to immunopathogenesis; or supporting evidence was available, we converted those numbers into person-years based on
from animal models using candidate vaccine platforms [3]. the post-vaccination risk period. Unlike the registers with individual-
One flexible approach for assessing AESI is the comparison of level data, the age and sex strata used in this approach might not have
observed AESI rates following the introduction of a vaccine program matched the strata used in the background rates calculations.
with the expected (or background) rates based on historical periods pre- Participants were individuals vaccinated with COVID-19 vaccines in
vaccine roll out [4,5]. Such comparisons can be executed rapidly and the populations represented by the sites. To the extent possible, stan­
can play a key role in early detection of potential vaccine safety signals dardized methods were applied across sites. Patient types included
or when regulatory and public health agencies need rapid assessment of hospital inpatients (Australia – New South Wales, France, New Zealand,
an emerging safety signal [4,6]. Observed versus (vs.) expected (OE) Scotland), and combinations of inpatient and outpatient emergency
analysis was integral in identifying thrombosis with thrombocytopenia department patients (Argentina, Australia – Victoria, Canada, Denmark,
syndrome (TTS) as a safety signal, prompting the suspension of use of Finland). In countries without clearly defined patient types, hospital
the ChAdOx1 (AstraZeneca COVID-19 vaccine) on March 11, 2021, in contact duration was used as a proxy for patient types. As an example, a
Denmark and Norway [7,8]. contact duration of five hours or longer was used as a proxy for in­
These evaluations are not only valuable early-on in large-scale vac­ patients in Denmark. Site-specific characteristics of data sources and
cine deployment, but also as the vaccination program matures, espe­ data are presented in Supplementary Table 1.
cially if they can be conducted in a multi-country context. We conducted
a global cohort study following the Observed vs. Expected Analyses of
2.3. Study period and follow-up
COVID-19 Adverse Events of Special Interest Study Protocol [9] with
data from 10 sites across eight countries participating in the unique
The study periods varied across countries, commencing on the date
Global COVID Vaccine Safety (GCoVS) Project [10] of the Global Vac­
of the site-specific COVID-19 vaccination program rollout, and
cine Data Network™ (GVDN®) [11]. The GCoVS Project, initiated in
concluding at the end of data availability (Table 1). In general, the study
2021, is a Centers for Disease Control and Prevention (CDC) funded
periods spanned from December 2020 until August 2023. The shortest
global collaboration of investigators and data sources from multiple
study period observed occurred in Australia – New South Wales,
nations for the purpose of COVID-19 vaccine safety monitoring.
including 11 months from February 2021 to December 2021. Argentina
had the longest study period, from December 2020 to August 2023,
2. Methods
encompassing a total of 32 months.
The risk intervals used after each dose were 0–7 days, 8–21 days,
2.1. Study design
22–42 days, and 0–42 days. For each vaccination dose, day 0 was
denoted the day of vaccine receipt. For this manuscript, we present re­
This retrospective observational study was designed to estimate the
sults for the risk interval of 0–42 days only. More data are presented on
OE ratios of selected AESIs after COVID-19 vaccination in a multi-
the GVDN dashboard with all latest updates from participating sites
country population cohort.
[12]. Outcome events that occurred outside the study period were not
included. A 365-day washout period for outcome events was used to
define incident outcomes. Outcome events were considered incident if

2201
K. Faksova et al. Vaccine 42 (2024) 2200–2211

there was no record of the same outcome event during the preceding Table 2
365-day washout period. An individual may have contributed several Total number of vaccinations by brand.
outcome events on the condition they were separated in time by at least Vaccine platform Vaccine brand Total doses
the washout period of 365 days.
Inactivated Covilo or SARS-CoV-2 Vaccine (Vero Cell) 134,550
[Sinopharm (Beijing)]
Covaxin [Bharat Biotech] 1,660
2.4. Study variables and outcomes CoronaVac or Sinovac [Sinovac Biotech] 31,598
Inactivated (Vero cell) [Sinopharm (Wuhan)] 623
2.4.1. Adverse events of special interest (AESI) Nucleic acid- Comirnaty or Riltozinameran or Pfizer/ 3,516,963
based BioNTech COVID-19 Vaccine Bivalent [Pfizer/
Thirteen conditions representing AESI of specific relevance to the
BioNTech]
current landscape of real-world vaccine pharmacovigilance were Comirnaty or Tozinameran [Pfizer/BioNTech 183,677,660
selected from the list compiled by the Brighton Collaboration SPEAC or Fosun-BioNTech]
Project [3] and in response to the safety signals of thrombosis with Comirnaty or Tozinameran Paediatric [Pfizer/ 2,439,086
thrombocytopenia syndrome [7,8] (Supplementary Table 2). The con­ BioNTech or Fosun-BioNTech]
Spikevax bivalent Original/Omicron 2,750,476
ditions chosen matched the AESI for which background rates were [Moderna]
recently generated by GVDN sites [13]. AESI were identified using Elasomeran or Spikevax or TAK-919 Half Dose 400,395
harmonized International Classification of Diseases 10th Revision (ICD- [Moderna or Takeda]
10) codes. Neurological conditions selected included Guillain-Barré Elasomeran or Spikevax or TAK-919 36,222,514
[Moderna or Takeda]
syndrome (GBS), transverse myelitis (TM), facial (Bell’s) palsy, acute
Protein-based MVC-COV1901 [Medigen] 16
disseminated encephalomyelitis (ADEM), and convulsions (generalized Covovax or Nuvaxoid [Novavax or Serum 66,856
seizures (GS) and febrile seizures (FS)) as potential safety signals have Institute of India]
been identified for some of these conditions [14–16]. Hematologic Non-replicating Convidecia or Convidence [CanSino] 3,938
conditions included cerebral venous sinus thrombosis (CVST), viral vector Covishield or Vaxzevria [AstraZeneca or 23,094,620
Serum Institute of India]
splanchnic vein thrombosis (SVT) and pulmonary embolism (PE); the
Sputnik Light or Gam-COVID-Vac [Gamaleya 26
unusual site thromboses (CVST and SVT) were selected as markers of Research Institute]
potential TTS that could be accurately identified using diagnostic codes Sputnik V [Gamaleya Research Institute] 84,460
[17,18]. Thrombocytopenia and immune thrombocytopenia (ITP) were Janssen [Janssen/Johnson & Johnson] 1,137,505
also included due to their association with TTS and reports of ITP as an
independent safety signal [7,19,20]. Myocarditis and pericarditis were
protein-based Novavax (n = 66,856) vaccine, and the adenovirus-vector
included as cardiovascular conditions and the OE ratios were evaluated
Janssen/Johnson & Johnson (n = 1,137,505) and Gamaleya Research
separately for each condition [21–23].
Institute/Sputnik (n = 84,460) vaccines. The total number of doses of
each vaccine brand administered are outlined in Table 2. Exposure to
2.4.2. COVID-19 vaccines
COVID-19 vaccine by platform/type, brand, and dose data were avail­
As of November 2023, multiple vaccines against COVID-19 were in
able at the individual level to determine the number of observed cases by
use by the GCoVS sites representing multiple platform types such as
vaccine type/brand and dose profile and within the 0–42 days post-
inactivated, nucleic acid-based (mRNA), protein-based, and non-
vaccination risk interval.
replicating viral vector platforms (Table 2). For this manuscript, we
focused on three vaccines that recorded the highest number of doses
administered, Pfizer/BioNTech BNT162b2, Moderna mRNA-1273, and 2.5. Statistical analysis
Oxford/Astra Zeneca/Serum Institute of India ChAdOx1 vaccines. The
cumulative number of doses of other vaccines administered (n) across 2.5.1. Calculation of observed vs. expected ratios for each site
study sites were relatively low, with exceptions for the inactivated For each site, we calculated the observed number of events for each
Sinopharm (n = 134,550) and Sinovac (n = 31,598) vaccines, the AESI in the risk interval after introduction of COVID-19 vaccination. To

Table 1
Population summary by site. (Only Pfizer/BioNTech BNT162b2, Moderna mRNA-1273, and Oxford/Astra Zeneca/Serum Institute of India ChAdOx1 vaccines and
doses 1–4 included).

Vaccines: Pfizer/BioNTech (BNT162b2), Moderna (mRNA-1273), and Oxford/Astra Zeneca/Serum Institute of India (ChAdOx1).

2202
K. Faksova et al. Vaccine 42 (2024) 2200–2211

calculate the expected number of cases, we used pre-COVID-19 vacci­ period was 10,000 or greater. The combined numbers of events and the
nation background rates data from 2015 to 2019 (2019–2020 for OE ratio was calculated with 95 % CIs derived using the exact Poisson
Denmark) collected in the GCoVS Background Rates of AESI Following distribution. No event (i.e., zero) observed for a vaccine brand and dose
COVID-19 vaccination study [13]. The observed follow-up period in profile was reported separately without CI.
person-years for a given vaccination profile and post-vaccination period
was stratified according to age group and sex. Each of the age-sex 2.5.3. Sensitivity analysis
stratified person-years were multiplied by the corresponding age-sex Firstly, we conducted site-specific sensitivity analyses to further
stratified background rate. This resulted in the expected number of explore potential associations of the most significant safety signals
cases in each stratum, which were then summed to give the total number identified in the main analysis. The observed rates reported by sites were
of expected cases during the observed follow-up period. considered in the analysis based on the following constraints. For each
The aggregated OE ratios by last dose were calculated by dividing the vaccine brand and dose profile, and post-vaccination period combina­
observed number of cases by the expected number of cases in the post- tion, the OE ratios and 95 % CI were suppressed if fewer than five events
vaccination period, 95 % confidence intervals (CI) were derived using were observed. Secondly, we conducted supplemental analysis including
the exact Poisson distribution. We also calculated OE ratios for homol­ other vaccines and doses administered across sites. The person-years
ogous schedules for BNT162b2, mRNA-1273, and ChAdOx1 vaccines up threshold for reporting was lowered from 10,000 to 1,000 person-
to four doses. Both the aggregated OE ratios and those specific to ho­ years compared to the main aggregated OE ratios analysis, allowing
mologous schedules are presented. for broader scope of vaccines to be analysed.
We considered an OE ratio a potential safety signal of concern where
the lower bound of the 95 % CI (LBCI) was greater than one and reached
2.6. Ethical approval
statistical significance [5]. However, we prioritised potential safety
signals of concern for further evaluation where the LBCI was greater
Approval from the relevant Human Research Ethics Committees was
than 1.5, due to increased statistical evidence and the higher likelihood
either acquired or an exemption obtained for all participating sites
of being a true signal, based on expert opinion from the CDC and GVDN
(Supplementary Table 3).
collaborators.
3. Results
2.5.2. Combining results across sites
The results were aggregated across sites by summing the observed
The total vaccinated population across all sites comprised
number of events for each AESI and the age-sex stratified person-years
99,068,901 individuals. Most vaccine recipients were in the 20–39 and
for a given vaccination profile and post-vaccination period. For each
40–59-year age groups (Table 1). In total, 183,559,462 doses of
AESI, individual vaccine profiles were reported if the cumulative
BNT162b2, 36,178,442 doses of mRNA-1273, and 23,093,399 doses of
amount of follow up (in person-years) in the 0–42 days post-vaccination
ChAdOx1 were administered across all the sites in the study periods. The

Table 3
Aggregated OE Ratios by last dose, neurological conditions, period 0–42 days.

AESI: GBS = Guillain-Barré syndrome, TRM = Transverse myelitis, BP = Facial (Bell’s) palsy, ADEM = Acute disseminated encephalomyelitis, FSZ = Febrile seizures,
GSZ = Generalised seizures.
Vaccines: Pfizer/BioNTech (BNT162b2), Moderna (mRNA-1273), and Oxford/Astra Zeneca/Serum Institute of India (ChAdOx1).

2203
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Table 4
Aggregated OE Ratios by last dose, haematologic conditions, period 0–42 days.

AESI: THR = Thrombocytopenia, ITP = Idiopathic thrombocytopenia, PEM = Pulmonary embolism, CVST = Cerebral venous sinus thrombosis, SVT = Splanchnic vein
thrombosis.
Vaccines: Pfizer/BioNTech (BNT162b2), Moderna (mRNA-1273), and Oxford/Astra Zeneca/Serum Institute of India (ChAdOx1).

highest numbers of doses were administered in France (120,758,419), of mRNA-1273 (1.36, 95 % CI: 1.02, 1.77 and 1.44, 95 % CI: 1.04, 1.95,
followed by Canada – Ontario (32,159,817) and Australia – Victoria respectively), and for generalised seizures following a first mRNA-1273
(15,617,627). In total, 23,168,335 person-years contributed to the OE dose (1.15, 95 % CI: 1.10, 1.20) and a fourth BNT162b2 dose (1.09, 95 %
ratios for the AESI following homologous schedules. The population CI: 1.04, 1.14). No increased OE ratios were identified following a third
summary is presented in Table 1, and more detailed information on the dose of any vaccine. The results are concordant with the OE ratios of
other administered vaccines are presented in Supplementary Table 4. In homologous schedules; however, an increased OE ratio for generalized
the results sections below, we provide both aggregated OE ratios seizures following a homologous schedule of four doses of mRNA-1273
(Tables 3–5) and detailed OE ratios for homologous schedules (1.33; 95 % CI: 1.07, 1.63) was identified (Fig. 1). These outcomes did
(Figs. 1–3), including the number of events and person-years. Overall, not meet the threshold for a prioritised safety signal following
95.8 % and 86.6 % of vaccinations were included in the aggregated and vaccination.
the homologous schedules analysis, respectively (Supplementary Table
5). The primary results from the individual sites as well as additional risk 3.2. Hematologic conditions
periods and meta-analyses for each AESI are available in the interactive
GVDN Observed vs Expected (OE) Dashboard [12]. The OE ratio of CVST was 3.23 (95 % CI: 2.51–4.09) within 42 days
after a first dose of ChAdOx1, fulfilling the threshold of a prioritised
3.1. Neurological conditions safety signal (Table 4). In total, 21 events were expected, while 69
events were observed (Fig. 2).
There was a statistically significant increase in GBS cases within 42 Increased OE ratios were also identified for thrombocytopenia after a
days after a first ChAdOx1 dose (OE ratio = 2.49; 95 % CI: 2.15, 2.87), first dose of ChAdOx1 (1.07; 95 % CI: 1.03, 1.12), BNT162b2 (1.11; 95
indicating a prioritised safety signal (Table 3). Seventy-six GBS events % CI: 1.08, 1.14), and mRNA-1273 (1.33; 95 % CI 1.25, 1.42), as well as
were expected, and 190 events were observed (Fig. 1). The OE ratio for after a third dose of ChAdOx1 (1.95; 95 % CI: 1.29, 2.84). Immune
ADEM within 42 days after a first mRNA-1273 dose also fulfilled the thrombocytopenia also demonstrated increased OE ratios after a first
significance threshold of a prioritised safety signal (3.78; 95 % CI: 1.52, dose of ChAdOx1 (1.40; 95 % CI: 1.24, 1.58) and BNT162b2 (1.08; 95 %
7.78), with two expected events compared with seven observed events CI: 1.01, 1.16). Pulmonary embolism OE ratios were increased following
(Fig. 1). first doses of ChAdOx1 (1.20; 95 % CI: 1.16, 1.24), BNT162b2 (1.29; 95
Statistically significant differences were also found for transverse % CI: 1.26, 1.32), and mRNA-1273 (1.33, 95 % CI: 1.26, 1.40), as well as
myelitis (OE ratio = 1.91; 95 % CI: 1.22, 2.84) and ADEM (OE ratio = after a third dose of ChAdOx1 (1.88; 95 % CI: 1.32, 2.58). The OE ratio
2.23; 95 % CI: 1.15, 3.90) after a first ChAdOx1 dose. Bell’s palsy had an of CVST was 1.49 (95 % CI: 1.26, 1.75) after a first dose and 1.25 (95 %
increased OE ratio after a first dose of BNT162b2 (1.05; 95 % CI: 1.00, CI: 1.06, 1.46) after a second dose of BNT162b2. An increased OE ratio
1.11) and mRNA-1273 (1.25; 95 % CI: 1.11, 1.39). There were also for SVT was found after a first dose of BNT162b2 (1.25; 95 % CI: 1.17,
increased OE ratios for febrile seizures following a first and second dose 1.34) and mRNA-1273 (1.23; 95 % CI: 1.03, 1.47); a second dose of

2204
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Table 5
Aggregated OE Ratios by last dose, cardiovascular conditions, period 0–42 days.

mRNA-1273 (1.17; 95 % CI: 1.01, 1.36); and a fourth dose of BNT162b2 homologous schedules (Fig. 3), except for the OE ratio of 1.23 (95 % CI:
(1.30, 95 % CI: 1.06, 1.59) and mRNA-1273 (1.53, 95 % CI: 1.05, 2.16). 0.45–2.69) after receipt of the fourth mRNA-1273 dose, which did not
These outcomes did not meet the threshold for a prioritised safety signal meet the threshold for a safety signal. The homologous OE ratio
following vaccination. following a third dose of ChAdOx1 was not reported as only a small
number of third doses of ChAdOx1 were given across study sites
(Table1).
3.3. Cardiovascular conditions
3.4. Sensitivity analysis
Increased OE ratios fulfilling the threshold of prioritised safety sig­
nals for myocarditis were consistently identified following a first, second
Secondary analyses were conducted to further explore GBS, ADEM,
and third dose of mRNA vaccines (BNT162b2 and mRNA-1273)
CVST, myocarditis, and pericarditis at the site-specific level. We report
(Table 4). The highest OE ratio was observed following a first and sec­
the aggregated OE ratios by last dose and site in the period 0–42 days
ond dose of mRNA-1273 (3.48; 95 % CI: 3.00, 4.01 and 6.10; 95 % CI:
after vaccination in Supplementary Tables 6–10. It was not possible to
5.52, 6.72, respectively). The OE ratio following a third dose of mRNA-
report results for all sites and study outcomes due to insufficient person-
1273 was 2.01 (95 % CI: 1.60, 2.49). The numbers of events for up to
years or less than five events observed by site privacy criteria. The
four doses of homologous schedules are shown in Fig. 3. The OE ratios of
majority of identified safety signals following specific vaccine brand and
homologous schedules align with the aggregated OE ratios. The ho­
dose combinations from the main analysis were, however, confirmed by
mologous OE for myocarditis following four doses of mRNA-1273 vac­
individual sites where data were available. The supplementary analysis
cine could not be estimated due to a lack of observed events.
with person-years threshold of 1,000 and including other vaccines and
Similarly, the OE ratio for pericarditis fulfilled the threshold of a
doses administered within the GVDN sites, showed an increased OE ratio
prioritised safety signal following a first and fourth dose of mRNA-1273,
for some outcomes, e.g. for generalized seizures following a first dose of
with OE ratios of 1.74 (95 % CI: 1.54, 1.97) and 2.64 (95 % CI: 2.05,
Gamaleya Research Institute/Sputnik vaccine (5.50, 95 % CI: 2.74, 9.84)
3.35) respectively. An increased ratio of 6.91 (95 % CI: 3.45, 12.36),
(Supplementary Tables 11–13).
fulfilling the threshold of a prioritised safety signal, was also observed
following a third dose of ChAdOx1. The aggregated OE ratios for peri­
carditis were increased following all doses of all the three vaccines
presented (Table 4). The results are very similar to the ratios of

2205
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Fig. 1. Number of events and OE ratios (with 95 % confidence interval) for homologous schedules by dose 1–4, neurological conditions. AESI: GBS = Guillain-Barré
syndrome, TRM = Transverse myelitis, BP = Facial (Bell’s) palsy, ADEM = Acute disseminated encephalomyelitis, FSZ = Febrile seizures, GSZ = Generalised sei­
zures. Vaccines: AZD = Oxford/Astra Zeneca/Serum Institute of India ChAdOx1, BNT = Pfizer/BioNTech (BNT162b2), MOD = Moderna (mRNA-1273).

4. Discussion explained by a smaller sample size and different outcome measures.

Overall, this evidence supports our findings of a GBS safety signal
This multi-country cohort study was conducted in the unique setting following ChAdOx1 vaccination. Although rare, this association was
of the GVDN. To date, the number of such large systematically coordi­ acknowledged by the WHO, the European Medicines Agency (EMA), and
nated studies across diverse geographical locations and populations is Therapeutic Goods Administration (TGA) of Australia, resulting in GBS
limited. However, several studies have previously assessed the risks of being listed as a rare side effect following exposure to ChAdOx1
the identified safety signals following COVID-19 vaccination, primarily [15,29,30].
in single site settings. We investigated the association between COVID- The identified increased risk of CVST following ChAdOx1 vaccina­
19 vaccination and 13 AESIs comprising neurological, haematological, tion in this study is corroborated by multiple studies. An increased OE
and cardiovascular conditions across 10 sites in eight countries ratio was observed in a nationwide cohort study from Denmark and
including Europe, North America, South America, and Oceania. In this Norway, with increased rates of venous thromboembolic events,
study including more than 99 million people vaccinated against SARS- including CVST with an excess rate of 2.5 events per 100,000 vaccina­
CoV-2, the risk up to 42 days after vaccination was generally similar tions following ChAdOx1 [7]. Based on a variety of methodologies, other
to the background risk for the majority of outcomes; however, a few studies have also reported increased incidence of CVST after vaccination
potential safety signals were identified. We observed potential safety [31,32]. Ultimately, this rare but concerning safety signal led to the
signals for GBS and CVST after the first dose of ChAdOx1 based on more withdrawal of the ChAdOx1 vaccine from COVID-19 vaccine programs
than 12 million doses administered. or implementation of age-based restrictions in multiple countries [8].
Overall, studies of the vector-based vaccines such as the ChAdOx1, It is crucial to acknowledge the significance threshold of prioritised
have observed a higher incidence of GBS after vaccination compared safety signals applied in this study (LBCI > 1.5). This threshold was
with the background incidence; whereas, most studies of the mRNA selected based on expert opinion within the GVDN and at CDC, to focus
vaccines, such as BNT162b2 and mRNA-1273, have not observed in­ on those outcomes most likely to be true signals. Some observed events,
creases of GBS [14,15,24–27]. Atzenhoffer et al. [24] reported an although not fulfilling this threshold, may still hold clinical importance
elevated OE ratio > 2.0 for adenovirus-vectored COVID-19 vaccines, and require further investigation. For instance, ITP with an OE ratio >
across countries contributing to VigiBase, an international database of 1.0 and LBCI of 1.2 following vaccination with ChAdOx1 aligns with
adverse drug events and Patone et al. [27] reported 38 excess cases of findings reported in the literature as a potential signal. This concurrence
GBS per 10 million exposed in the 1–28 days risk period following is highlighted in a study conducted in Victoria, Australia, which
vaccination with ChAdOx1 in England. The authors did not observe an observed a substantially higher than expected rate of ITP following
increased risk in those who received BNT162b2. In contrast, a study by ChAdOx1 vaccination [33].
Li et al. [28] showed no increased risk of GBS for ChAdOx1, while only Moreover, we observed significantly higher risks of myocarditis
SARS-CoV-2 infection was associated with a higher risk. The discrep­ following the first, second and third doses of BNT162b2 and mRNA-
ancy, compared with the results of Patone et al. [27], could however be 1273 as well as pericarditis after the first and fourth dose of mRNA-

2206
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Fig. 2. Number of events and OE ratios (with 95 % confidence interval) for homologous schedules by dose 1–4, hematologic conditions. AESI: THR = Thrombo­
cytopenia, ITP = Idiopathic thrombocytopenia, PEM = Pulmonary embolism, CVST = Cerebral venous sinus thrombosis, SVT = Splanchnic vein thrombosis. Vac­
cines: AZD = Oxford/Astra Zeneca/Serum Institute of India (ChAdOx1), BNT = Pfizer/BioNTech (BNT162b2), MOD = Moderna (mRNA-1273).

1273, and third dose of ChAdOx1, in the 0–42 days risk period. The events based on 1,035,871 person-years and 10.5 million doses
elevated rates of pericarditis following ChAdOx1 vaccination identified administered; however, the number of cases of this rare event were small
in this study rely on a limited number of observed counts in the meta- and the confidence interval wide, so results should be interpreted with
analysis. The wide confidence interval underscores the substantial un­ caution and confirmed in future studies. Although some case reports
certainty of characterizing pericarditis as a safety signal following have suggested a possible association between COVID-19 vaccination
ChAdOx1 vaccination. However, our study confirms findings of previ­ and ADEM, there was no consistent pattern in terms of vaccine or timing
ously identified rare cases of myocarditis and pericarditis following first following vaccination, and larger epidemiological studies have not
and second doses of mRNA vaccines [21–23,34]. A large cohort study of confirmed any potential association [41–44]. Moreover, case reports
23.1 million residents across four Nordic countries revealed an increased may report on coincidental events and do not establish association nor
risk of myocarditis among young males aged 16–24 years, based on 4–7 indicate causality, thus larger observational studies are warranted to
excess events in 28 days per 100,000 vaccinees after a second dose of further investigate our finding. To address this, a follow-up study is
BNT162b2, and between 9 and 28 per 100,000 vaccinees after a second currently being undertaken within the GVDN, focusing on a de­
dose of mRNA-1273 [22]. Similarly, studies from British Columbia, mographic not included in our analysis. Based on reports of rare ADEM
Canada reported cases of myocarditis to be higher among those cases to the European Database of Suspected Adverse Drug Reaction,
receiving a second dose compared with a third dose, and for those who EMA assessed the potential association of ADEM following vaccination
received a second dose of the mRNA-1273 vaccine compared with the with ChAdOx1 [45]. Frontera et al. [46] concluded that chances of
BNT162b2 vaccine [35,36]. Patone et al. [37] estimated extra myocar­ having a neurological event following acute SARS-CoV-2 infection were
ditis events to be between one and 10 per million persons in the month up to 617-fold higher than following COVID vaccination, suggesting that
following vaccination, which was substantially lower than the 40 extra the benefits of vaccination substantially outweigh the risks. A safety
events per million persons observed following SARS-CoV-2 infection signal for generalized seizures was identified following Gamaleya
period. A systematic review by Alami et al. [38] concluded that mRNA Research Institute/Sputnik vaccination, however the number of vacci­
vaccinated individuals were twice as likely to develop myocarditis/ nations was relatively low compared with other vaccines in this study.
pericarditis compared with unvaccinated individuals, with a rate ratio of Further studies are warranted to explore this potential safety signal.
2.05 (95 % CI 1.49–2.82). Given the evidence, WHO issued updated Conducting a cohort analysis in the unique multi-country context of
guidance regarding these safety signals and mRNA COVID-19 vaccina­ the GVDN leverages a vast and diverse data pool. Aggregating data from
tion, and EMA provided updates to the Product Information for multiple countries on more than 99 million vaccine recipients has
BNT162b2 and mRNA-1273 vaccines [21,23]. TGA as well as the CDC significantly increased the sample size and the statistical power
continue to monitor and review data on myocarditis and pericarditis compared with many previous safety studies. This enhances the ability
following COVID-19 vaccination [39,40]. to detect safety signals, especially for extremely rare adverse events, as
Another potential safety signal was identified for ADEM after the first the larger sample size provides greater precision in estimating observed
dose of mRNA-1273 vaccine, with five more observed than expected rates.

2207
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Fig. 3. Number of events and OE ratios (with 95 % confidence interval) for homologous schedules by dose 1–4, cardiovascular conditions. AESI: MYO = Myocarditis,
PER = Pericarditis. Vaccines: Vaccines: AZD = Oxford/Astra Zeneca/Serum Institute of India (ChAdOx1), BNT = Pfizer/BioNTech (BNT162b2), MOD = Moderna
(mRNA-1273).

Results based on data across Europe, North and South America and detecting associations with lower confidence intervals below 1.5 that
Oceania offer stronger external validity, enabling findings to be more maintain statistical significance. The safety signals identified in this
generalizable to a broader range of populations and healthcare settings study should be evaluated in the context of their rarity, severity, and
participating in the global COVID-19 vaccination programme. More­ clinical relevance. Moreover, overall risk–benefit evaluations of vacci­
over, multi-country analyses facilitate comparisons between countries nation should take the risk associated with infection into account, as
with varying vaccination strategies, population demographics, and multiple studies demonstrated higher risk of developing the events
healthcare systems, yielding insights into how these factors may influ­ under study, such as GBS, myocarditis, or ADEM, following SARS-CoV-2
ence vaccine safety profiles. Data used in our analysis were drawn from infection than vaccination. Finally, the use of ICD-10 codes is subject to
multiple databases, including healthcare databases, national immuni­ considerations about specificity and sensitivity, and application may
zation registries, and vaccination dashboards, allowing the identifica­ vary by country.
tion of potential safety signals from various sources.
The results from our study should, however, be interpreted consid­ 5. Conclusion
ering multiple limitations. Our analyses inherently involve heteroge­
neity in data collection, quality, and reporting standards across Observed vs. expected analyses in a multi-country context of the
countries. These differences in healthcare infrastructure and surveil­ GVDN and the GCoVS Project offers a larger and more diverse dataset,
lance systems can introduce bias and affect the comparability of results. enhanced generalizability, and improved statistical power over single
The participating sites across the eight countries implemented varied site or regional studies. It also presents challenges related to data het­
vaccination strategies, including vaccine types, dosing schedules, and erogeneity, population confounding factors, and variations in vaccina­
prioritization of vaccine recipients. Moreover, the multi-country ana­ tion strategies and reporting systems. The involvement of researchers
lyses are susceptible to population confounding factors, such as differ­ and data sources from diverse regions of the world promotes inclusivity,
ences in pre-existing health conditions, genetic factors, ethnic profiles, reduces potential biases, and fosters collaboration in the pursuit of a
and behavioural patterns, which was not possible to adjust for in our shared public health goal. While our study confirmed previously iden­
analysis. We consider our approach suitable for application in large tified rare safety signals following COVID-19 vaccination and contrib­
datasets representing average populations. However, age- and sex- uted evidence on several other important outcomes, further
specific historic background rates that are not adjusted for factors like investigation is warranted to confirm associations and assess clinical
prior disease may not provide a suitable comparison, for example, in the significance. This could be addressed by conducting association studies
early stages of a vaccination campaigns where people with co- specific to individual outcomes by applying methodologies such as the
morbidities were vaccinated prior to other population groups. self-controlled case series (SCCS) to validate the associations [6].
Potential underreporting across countries may have led to an un­
derestimation of the significance of potential safety signals. It is
important to recognize the potential for false negatives, especially when

2208
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Disclaimer Artama: Conceptualization, Investigation, Methodology, Validation,

Writing – review & editing. H. Lu: Data curation, Formal analysis,
All analyses, inferences drawn, opinions, conclusions, and state­ Software, Validation. H.J. Clothier: Conceptualization, Methodology,
ments are those of the authors and do not necessarily represent the Validation, Writing – review & editing, Data curation, Formal analysis,
official views of, nor an endorsement by, CDC/HHS, or the U.S. Gov­ Project administration. K. Batty: Conceptualization, Methodology,
ernment. For more information, please visit cdc.gov. Project administration, Validation, Writing – review & editing. J.
Parts of this material are based on data and/or information compiled Paynter: Conceptualization, Methodology, Supervision, Data curation,
and provided by the Canadian Institute for Health Information and the Formal analysis, Writing – review & editing. H. Petousis-Harris:
Ontario Ministry of Health. The analyses, conclusions, opinions, and Conceptualization, Funding acquisition, Investigation, Methodology,
statements expressed herein are solely those of the authors and do not Resources, Supervision, Writing – review & editing, Project adminis­
reflect those of the funding or data sources; no endorsement is intended tration. J. Buttery: Conceptualization, Funding acquisition, Investiga­
or should be inferred. Parts of this material are based data and/or in­ tion, Methodology, Resources, Supervision, Validation, Writing – review
formation provided by the British Columbia Ministry of Health. All in­ & editing, Project administration. S. Black: Conceptualization, Funding
ferences, opinions, and conclusions drawn in this manuscript are those acquisition, Investigation, Methodology, Project administration, Re­
of the authors, and do not reflect the opinions or policies of the Data sources, Supervision, Validation, Writing – review & editing. A. Hviid:
Steward(s). Conceptualization, Investigation, Methodology, Supervision, Valida­
tion, Writing – review & editing.
Funding statement
Declaration of competing interest
The GCoVS project is supported by the Centers for Disease Control
and Prevention (CDC) of the U.S. Department of Health and Human The authors declare the following financial interests/personal re­
Services (HHS) as part of a financial assistance award totalling US lationships which may be considered as potential competing interests:
$10,108,491 with 100 % per cent funded by CDC/HHS. Jeffrey C. Kwong reports financial support was provided by Centers for
The Ontario site contributing to this study was supported by Public Disease Control and Prevention. Naveed Z. Janjua reports financial
Health Ontario and by the ICES, which is funded by an annual grant support was provided by Centers for Disease Control and Prevention.
from the Ontario Ministry of Health. JCK is supported by a Clinician- Anders Hviid reports financial support was provided by Global Vaccine
Scientist Award from the University of Toronto Department of Family Data Network. Helen Petousis-Harris reports financial support was
and Community Medicine. provided by New Zealand Ministry of Health. Steven Black reports a
relationship with GSK that includes: consulting or advisory. Jeffrey C.
CRediT authorship contribution statement Kwong reports a relationship with Canadian Institutes of Health
Research that includes: funding grants. Jeffrey C. Kwong reports a
K. Faksova: Visualization, Writing – original draft, Writing – review relationship with Public Health Agency of Canada that includes: funding
& editing. D. Walsh: Data curation, Formal analysis, Investigation, grants. Naveed Z. Janjua reports a relationship with AbbVie Inc that
Methodology, Software, Validation, Conceptualization, Writing – review includes: consulting or advisory and speaking and lecture fees. Naveed
& editing, Visualization. Y. Jiang: Conceptualization, Data curation, Z. Janjua reports a relationship with Gilead Sciences Inc that includes:
Formal analysis, Investigation, Methodology, Software, Supervision, speaking and lecture fees. Anders Hviid reports a relationship with In­
Validation, Visualization, Writing – review & editing. J. Griffin: dependent Research Fund Denmark that includes: funding grants.
Conceptualization, Writing – review & editing, Methodology. A. Phil­ Anders Hviid reports a relationship with Lundbeck Foundation that in­
lips: Conceptualization, Methodology, Writing – review & editing, cludes: funding grants. Anders Hviid reports a relationship with Novo
Investigation, Validation. A. Gentile: Data curation, Investigation, Su­ Nordisk Foundation that includes: funding grants. Anders Hviid reports
pervision, Validation. J.C. Kwong: Data curation, Supervision, Valida­ a relationship with VAC4EU that includes: consulting or advisory.
tion, Writing – review & editing, Investigation, Methodology. K. Finnish Institute for Health and Welfare (THL) conducts Public-Private
Macartney: Data curation, Supervision, Validation, Writing – review & Partnership with vaccine manufacturers and has received research
editing, Investigation, Methodology. M. Naus: Data curation, Supervi­ funding from Sanofi Inc. Petteri Hovi has been an investigator in these
sion, Validation, Investigation, Methodology. Z. Grange: Data curation, studies, but has received no personal remuneration. Helen Petousis-
Supervision, Validation, Conceptualization, Investigation, Methodol­ Harris has served on expert advisory boards and had speaking engage­
ogy. S. Escolano: Data curation, Supervision, Validation, Investigation, ments for Pfizer and GSK. She has also received research funding from
Methodology, Writing – review & editing. G. Sepulveda: Data curation, GSK. She has not received any personal honoraria. If there are other
Formal analysis, Software, Validation. A. Shetty: Data curation, Vali­ authors, they declare that they have no known competing financial in­
dation, Investigation, Methodology. A. Pillsbury: Data curation, Vali­ terests or personal relationships that could have appeared to influence
dation, Investigation, Methodology, Writing – review & editing. C. the work reported in this paper.
Sullivan: Data curation, Validation, Investigation, Methodology,
Writing – review & editing. Z. Naveed: Data curation, Validation, Data availability
Investigation, Methodology, Writing – review & editing. N.Z. Janjua:
Data curation, Writing – review & editing. N. Giglio: Data curation, The authors do not have permission to share data.
Investigation, Methodology, Validation. J. Perälä: Data curation,
Investigation, Methodology, Validation. S. Nasreen: Conceptualization, Acknowledgements
Data curation, Validation, Writing – review & editing. H. Gidding:
Conceptualization, Validation, Writing – review & editing, Investiga­ The Observed vs Expected Analyses Of COVID-19 Vaccine Adverse
tion, Methodology. P. Hovi: Conceptualization, Validation, Writing – Events of Special Interest Study Protocol was developed by the Observed
review & editing, Investigation, Methodology. T. Vo: Conceptualization, vs. Expected Work Group led by Anders Hviida,b. Members of the Work
Validation, Formal analysis, Investigation, Methodology, Writing – re­ Group were Nelson Aguirre Duartec, Miia Artamad, Karin Battye, Steven
view & editing. F. Cui: Conceptualization, Investigation, Methodology, Blackc,f, Hannah Chisholmc, Hazel Clothierg,h,i, Fuqiang Cuij, Lucy
Validation. L. Deng: Conceptualization, Investigation, Methodology, Dengk, Lucy Cullenl, Heather Giddingk,m,n, Petteri Hovio, Yannan Jiangc,
Validation, Writing – review & editing. L. Cullen: Conceptualization, Janine Paynterc, Helen Petousis-Harrisc, Anastasia Phillipsk, John
Investigation, Methodology, Validation, Writing – review & editing. M. Sluyterc, Thuan Vod,o, and Daniel Walshc, Eric Weintraubp.

2209
K. Faksova et al. Vaccine 42 (2024) 2200–2211

Work group affiliations cited 2023 December 13]. Available from: www.globalvaccinedatanetwork.org/Da
ta-Dashboards.
a. Statens Serum Institut, Denmark; b. University of Copenhagen,
[13] Phillips A, Jiang Y, Walsh D, Andrews N, Artama M, Clothier H, et al. Background
Denmark; c. University of Auckland, New Zealand; d. Tampere University, rates of adverse events of special interest for COVID-19 vaccines: a multinational
Finland; e. Auckland UniServices Limited at University of Auckland, New Global Vaccine Data Network (GVDN) analysis. Vaccine 2023;41:6227–38. https://
Zealand; f. University of Cincinnati and Children’s Hospital, USA; g. doi.org/10.1016/J.VACCINE.2023.08.079.
[14] Hanson KE, Goddard K, Lewis N, Fireman B, Myers TR, Bakshi N, et al. Incidence of
Victorian Department of Health, Australia; h. Murdoch Childrens Research Guillain-Barré syndrome after COVID-19 vaccination in the vaccine safety datalink.
Institute, Australia; i. Centre for Health Analytics, Melbourne Children’s JAMA Netw Open 2022;5:E228879. https://doi.org/10.1001/
Campus, Australia; j. Peking University, Beijing, China; k. National Centre JAMANETWORKOPEN.2022.8879.
[15] World Health Organization. Statement of the WHO Global Advisory Committee on
for Immunisation Research and Surveillance of Vaccine Preventable Dis­ Vaccine Safety (GACVS) COVID-19 subcommittee on reports of Guillain-Barré
eases, Australia; l. Public Health Scotland, Glasgow, Scotland, United Syndrome (GBS) following adenovirus vector COVID-19 vaccines n.d. https
Kingdom; m. Kolling Institute, Northern Sydney Local Health District, ://www.who.int/news/item/26-07-2021-statement-of-the-who-gacvs-covid-19-s
ubcommittee-on-gbs (accessed October 23, 2023).
Australia; n. University of Sydney Northern Clinical School, Australia; o. [16] Keizer RJ, Huitema ADR, Schellens JHM, Beijnen JH. Clinical pharmacokinetics of
Finnish Institute for Health and Welfare, Finland; p. Vaccine Safety therapeutic monoclonal antibodies. Clin Pharmacokinet 2010;49:493–507.
Datalink, Centers for Disease Control and Prevention. https://doi.org/10.2165/11531280-000000000-00000.
[17] World Health Organization. Global Advisory Committee on Vaccine Safety
The following individuals contributed as GVDN site leads: Anders (GACVS) review of latest evidence of rare adverse blood coagulation events with
Hviid (Denmark); Angela Gentile (Argentina); Sylvie Escolano (France); AstraZeneca COVID-19 Vaccine (Vaxzevria and Covishield). Accessed October 23,
Eero Poukka (Finland); Jeffrey C. Kwong (Ontario, Canada); Kristine 2023. https://www.who.int/news/item/16-04-2021-global-advisory-
committee-on-vaccine-safety-(gacvs)-review-of-latest-evidence-of-rare-
Macartney (New South Wales, Australia); Jim Buttery (Victoria,
adverse-blood-coagulation-events-with
Australia); Monika Naus (British Columbia, Canada); Zoe Grange -astrazeneca-covid-19-vaccine-(vaxzevria-and-covishield).
(Scotland); and Helen Petousis-Harris (New Zealand). [18] Burn E, Li X, Kostka K, Stewart HM, Reich C, Seager S, et al. Background rates of
The following individuals contributed as GVDN site investigators: five thrombosis with thrombocytopenia syndromes of special interest for COVID-19
vaccine safety surveillance: incidence between 2017 and 2019 and patient profiles
Gonzalo Sepulveda and Aishwarya Shetty (Victoria, Australia); Alexis from 38.6 million people in six European countries. Pharmacoepidemiol Drug Saf
Pillsbury (New South Wales, Australia); Christopher Sullivan (Scotland); 2022;31:495–510. https://doi.org/10.1002/PDS.5419.
Naveed Zaeema (British Columbia, Canada); Norberto Giglio [19] Greinacher A, Thiele T, Warkentin TE, Weisser K, Kyrle PA, Eichinger S.
Thrombotic thrombocytopenia after ChAdOx1 nCov-19 vaccination. N Engl J Med
(Argentina); Jori Perälä (Finland); Sharifa Nasreen (Ontario, Canada); 2021;384:2092–101. https://doi.org/10.1056/NEJMOA2104840/SUPPL_FILE/
Han Lu (New Zealand). NEJMOA2104840_DISCLOSURES.PDF.
[20] Hviid A, Hansen JV, Thiesson EM, Wohlfahrt J. Association of AZD1222 and
BNT162b2 COVID-19 vaccination with thromboembolic and thrombocytopenic
Appendix A. Supplementary material events in frontline personnel: a retrospective cohort study. Ann Intern Med 2022;
175:541–6. https://doi.org/10.7326/M21-2452.
Supplementary data to this article can be found online at https://doi. [21] World Health Organization. COVID-19 subcommittee of the WHO Global Advisory
Committee on Vaccine Safety (GACVS): updated guidance regarding myocarditis
org/10.1016/j.vaccine.2024.01.100. and pericarditis reported with COVID-19 mRNA vaccines. Accessed October 23,
2023. https://www.who.int/news/item/09-07-2021-gacvs-guidance-myocarditis
References -pericarditis-covid-19-mrna-vaccines.
[22] Karlstad Ø, Hovi P, Husby A, Härkänen T, Selmer RM, Pihlström N, et al. SARS-
CoV-2 vaccination and myocarditis in a Nordic cohort study of 23 million residents.
[1] World Health Organization. WHO Director-General’s opening remarks at the media
JAMA Cardiol 2022;7:600–12. https://doi.org/10.1001/
briefing on COVID-19 - 11 March 2020 2020. https://www.who.int/director-gen
JAMACARDIO.2022.0583.
eral/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefin
[23] Gargano JW, Wallace M, Hadler SC, Langley G, Su JR, Oster ME, et al. Use of
g-on-covid-19—11-march-2020 (accessed October 13, 2023).
mRNA COVID-19 vaccine after reports of myocarditis among vaccine recipients:
[2] Our World in Data. Coronavirus (COVID-19) Vaccinations - Our World in Data
update from the advisory committee on immunization practices — United States,
2023. https://ourworldindata.org/covid-vaccinations (accessed October 13, 2023).
June 2021. MMWR Morb Mortal Wkly Rep 2021;70:977–82. 10.15585/MMWR.
[3] Safety Platform for Emerging Vaccines S. D2.3 Priority list of adverse events of
MM7027E2.
special interest: COVID-19. Brighton Collaboration; 2020. Available from:
[24] Atzenhoffer M, Auffret M, Pegat A, Masmoudi K, Khouri C, Bertin B, et al. Guillain-
https://zenodo.org/records/6656179#.Y-0yxuyZOnN.
Barré syndrome associated with COVID-19 vaccines: a perspective from
[4] van der Boom MDX, van Eekeren R, van Hunsel FPAM. Observed-over-Expected
spontaneous report data. Clin Drug Investig 2022;42:581. https://doi.org/
analysis as additional method for pharmacovigilance signal detection in large-
10.1007/S40261-022-01164-4.
scaled spontaneous adverse event reporting. Pharmacoepidemiol Drug Saf 2023;
[25] Otero-Losada M, Petrovsky N, Alami A, Crispo JA, Mattison D, Capani F, et al.
32:783–94. https://doi.org/10.1002/pds.5610.
Disproportionality analysis of adverse neurological and psychiatric reactions with
[5] Mahaux O, Bauchau V, Van Holle L. Pharmacoepidemiological considerations in
the ChAdOx1 (Oxford-AstraZeneca) and BNT162b2 (Pfizer-BioNTech) COVID-19
observed-to-expected analyses for vaccines. Pharmacoepidemiol Drug Saf 2016;25:
vaccines in the United Kingdom. Expert Opin Drug Saf 2023;22:343–9. https://doi.
215. https://doi.org/10.1002/PDS.3918.
org/10.1080/14740338.2022.2120607.
[6] Li R, Stewart B, Weintraub E. Evaluating efficiency and statistical power of self-
[26] Walker JL, Schultze A, Tazare J, Tamborska A, Singh B, Donegan K, et al. Safety of
controlled case series and self-controlled risk interval designs in vaccine safety.
COVID-19 vaccination and acute neurological events: a self-controlled case series
J Biopharm Stat 2016;26:686–93. https://doi.org/10.1080/
in England using the OpenSAFELY platform. Vaccine 2022;40:4479–87. https://
10543406.2015.1052819.
doi.org/10.1016/J.VACCINE.2022.06.010.
[7] Pottegård A, Lund LC, Karlstad Ø, Dahl J, Andersen M, Hallas J, et al. Arterial
[27] Patone M, Handunnetthi L, Saatci D, Pan J, Katikireddi SV, Razvi S, et al.
events, venous thromboembolism, thrombocytopenia, and bleeding after
Neurological complications after first dose of COVID-19 vaccines and SARS-CoV-2
vaccination with Oxford-AstraZeneca ChAdOx1-S in Denmark and Norway:
infection. Nat Med 2021;27:2144. https://doi.org/10.1038/S41591-021-01556-7.
population based cohort study. BMJ 2021;373:n1114. https://doi.org/10.1136/
[28] Li X, Raventós B, Roel E, Pistillo A, Martinez-Hernandez E, Delmestri A, et al.
bmj.n1114.
Association between covid-19 vaccination, SARS-CoV-2 infection, and risk of
[8] European Medicines Agency. Signal assessment report on embolic and thrombotic
immune mediated neurological events: population based cohort and self-controlled
events (SMQ) with COVID-19 Vaccine (ChAdOx1-S [recombinant])-Vaxzevria
case series analysis. BMJ 2022:376. https://doi.org/10.1136/BMJ-2021-068373.
(previously COVID-19 Vaccine AstraZeneca) (Other viral vaccines). 2021.
[29] European Medicines Agency. Vaxzevria (previously COVID-19 Vaccine
[9] Global Vaccine Data Network. Observed vs. expected analyses of COVID-19 vaccine
AstraZeneca) | European Medicines Agency. Accessed October 26, 2023.
adverse events of special interest study protocol. Version 1.4. Auckland (NZ):
https://www.ema.europa.eu/en/medicines/human/EPAR/vaxzevria#product-
Global Vaccine Data Network; 2022. Available from: https://www.globalvaccineda
information-section.
tanetwork.org/sites/globalvaccinedatanetwork.org/files/2024-01/GVDN-observe
[30] COVID-19 vaccine weekly safety report - 09-12-2021 | Therapeutic Goods
d-vs.-expected-analyses-of-COVID-19-vaccine-AESI-V1.4.pdf.
Administration (TGA). Accessed December 5, 2023. https://www.tga.gov.
[10] Global Vaccine Data Network. Global COVID Vaccine Safety (GCoVS). Auckland
au/news/covid-19-vaccine-safety-reports/covid-19-vacc
(NZ): Global Vaccine Data Network; 2022 [accessed 2023 October 13]. Available
ine-weekly-safety-report-09-12-2021#section-1298.
from: https://www.globalvaccinedatanetwork.org/global-covid-vaccine-safe
[31] Schulz JB, Berlit P, Diener HC, Gerloff C, Greinacher A, Klein C, et al. COVID-19
ty-gcovs.
vaccine-associated cerebral venous thrombosis in Germany. Ann Neurol 2021;90:
[11] Global Vaccine Data Network. Auckland (NZ): Global Vaccine Data Network; 2023
627–39. https://doi.org/10.1002/ANA.26172.
[accessed 2023 October 13]. Available from: https://www.globalvaccinedatanet
[32] Andrews NJ, Stowe J, Ramsay ME, Miller E. Risk of venous thrombotic events and
work.org/.
thrombocytopenia in sequential time periods after ChAdOx1 and BNT162b2
[12] Global Vaccine Data Network. GVDN: Observed vs expected (OE) dashboard.
Auckland (NZ): Global Vaccine Data Network; 2023 [updated 2023 December 8;

2210
K. Faksova et al. Vaccine 42 (2024) 2200–2211

COVID-19 vaccines: a national cohort study in England. Lancet Reg Heal Eur 2022; 19-vaccine-safety-report-13-07-2023#myocarditis-and-pericarditis-after-covid19-
13. doi: 10.1016/J.LANEPE.2021.100260. vaccination (accessed December 5, 2023).
[33] Gordon SF, Clothier HJ, Morgan H, Buttery JP, Phuong LK, Monagle P, et al. [40] Myocarditis and Pericarditis After mRNA COVID-19 Vaccination | CDC n.d. https
Immune thrombocytopenia following immunisation with Vaxzevria ChadOx1-S ://www.cdc.gov/coronavirus/2019-ncov/vaccines/safety/myocarditis.html
(AstraZeneca) vaccine, Victoria. Australia Vaccine 2021;39:7052–7. https://doi. (accessed December 5, 2023).
org/10.1016/J.VACCINE.2021.10.030. [41] Brock K, Reyes SC, Conner C, Gillson N, Weiss M, Elfituri O, et al. Acute
[34] Buchan SA, Seo CY, Johnson C, Alley S, Kwong JC, Nasreen S, et al. Epidemiology disseminated encephalomyelitis (ADEM)-like illness in a pediatric patient
of myocarditis and pericarditis following mRNA vaccination by vaccine product, following COVID-19 vaccination. BJR | Case Rep 2023;9. https://doi.org/10.1259/
schedule, and interdose interval among adolescents and adults in Ontario, Canada. BJRCR.20220097.
JAMA Netw Open 2022;5:E2218505. https://doi.org/10.1001/ [42] Nimkar SV, Yelne P, Gaidhane SA, Kumar S, Acharya S, Gemnani RR. Fatal acute
JAMANETWORKOPEN.2022.18505. disseminated encephalomyelitis post-COVID-19 vaccination: a rare case report.
[35] Naveed Z, Li J, Spencer M, Wilton J, Naus M, García HAV, et al. Observed versus Cureus 2022;14. https://doi.org/10.7759/CUREUS.31810.
expected rates of myocarditis after SARS-CoV-2 vaccination: a population-based [43] Permezel F, Borojevic B, Lau S, de Boer HH. Acute disseminated encephalomyelitis
cohort study. CMAJ 2022;194:1529–36. https://doi.org/10.1503/CMAJ.220676. (ADEM) following recent Oxford/AstraZeneca COVID-19 vaccination. Forensic Sci
[36] Naveed Z, Li J, Wilton J, Spencer M, Naus M, Velásquez García HA, et al. Med Pathol 2022;18:74–9. https://doi.org/10.1007/S12024-021-00440-7.
Comparative risk of myocarditis/pericarditis following second doses of BNT162b2 [44] Khan S, Khan S, Waqar Z, Mobeen H, Khan N, Hassan M. Post Moderna Vaccine
and mRNA-1273 coronavirus vaccines. J Am Coll Cardiol 2022;80:1900–8. https:// associated acute disseminated encephalomyelitis (ADEM) - a case report. Pak J
doi.org/10.1016/J.JACC.2022.08.799. Neurol Sci 2022;17. 10.56310/PJNS.V17I01.186.
[37] Patone M, Mei XW, Handunnetthi L, Dixon S, Zaccardi F, Shankar-Hari M, et al. [45] European Medicines Agency. COVID-19 Vaccine Safety Update.; 2021. https://www
Risks of myocarditis, pericarditis, and cardiac arrhythmias associated with COVID- .ema.europa.eu/en/documents/covid-19-vaccine-safety-update/covid-19-vacc
19 vaccination or SARS-CoV-2 infection. Nat Med 2021;28:410–22. https://doi. ine-safety-update-vaxzevria-previously-covid-19-vaccine-astrazeneca-18-june-
org/10.1038/s41591-021-01630-0. 2021_en.pdf.
[38] Alami A, Krewski D, Farhat N, Mattison D, Wilson K, Gravel CA, et al. Risk of [46] Frontera JA, Tamborska AA, Doheim MF, Garcia-Azorin D, Gezegen H, Guekht A,
myocarditis and pericarditis in mRNA COVID-19-vaccinated and unvaccinated et al. Neurological events reported after COVID-19 vaccines: an analysis of vaccine
populations: a systematic review and meta-analysis. BMJ Open 2023;13:e065687. adverse event reporting system. Ann Neurol 2022;91:756–71. https://doi.org/
https://doi.org/10.1136/BMJOPEN-2022-065687. 10.1002/ANA.26339.
[39] COVID-19 vaccine safety report - 13-07-2023 | Therapeutic Goods Administration
(TGA) n.d. https://www.tga.gov.au/news/covid-19-vaccine-safety-reports/covid-

2211