Thrombotic and Thromboembolic Complications After Vaccination Against COVID-19: A Systematic Review

Thromboembolic complications after the COVID-19 vaccination have been reported from all over the world. We aimed to identify the thrombotic and thromboembolic complications that can arise after receiving various types of COVID-19 vaccines, their frequency, and distinguishing characteristics. Articles published in Medline/PubMed, Scopus, EMBASE, Google Scholar, EBSCO, Web of Science, the Cochrane Library, the CDC database, the WHO database, ClinicalTrials.gov, and servers like medRxiv.org and bioRxiv.org, as well as the websites of several reporting authorities between December 1, 2019, and July 29, 2021, were searched. Studies were included if they reported any thromboembolic complications post-COVID-19 vaccination and excluded editorials, systematic reviews, meta-analyses, narrative reviews, and commentaries. Two reviewers independently extracted the data and conducted the quality assessment. Thromboembolic events and associated hemorrhagic complications after various types of COVID-19 vaccines, their frequency, and distinguishing characteristics were assessed. The protocol was registered at PROSPERO (ID-CRD42021257862). There were 59 articles, enrolling 202 patients. We also studied data from two nationwide registries and surveillance. The mean age of presentation was 47 ± 15.5 (mean ± SD) years, and 71.1% of the reported cases were females. The majority of events were with the AstraZeneca vaccine and with the first dose. Of these, 74.8% were venous thromboembolic events, 12.7% were arterial thromboembolic events, and the rest were hemorrhagic complications. The most common reported event was cerebral venous sinus thrombosis (65.8%), followed by pulmonary embolism, splanchnic vein thrombosis, deep vein thrombosis, and ischemic and hemorrhagic stroke. The majority had thrombocytopenia, high D-dimer, and anti-PF4 antibodies. The case fatality rate was 26.5%. In our study, 26/59 of the papers were of fair quality. The data from two nationwide registries and surveillance revealed 6347 venous and arterial thromboembolic events post-COVID-19 vaccinations. COVID-19 vaccinations have been linked to thrombotic and thromboembolic complications. However, the benefits far outweigh the risks. Clinicians should be aware of these complications because they may be fatal and because prompt identification and treatment can prevent fatalities.


Introduction And Background
COVID-19 is caused by infection from the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and the first case was reported from Wuhan, Hubei Province, China, in December 2019 [1]. COVID-19 was declared a pandemic by the World Health Organization (WHO) on March 11, 2020 [1]. It is now evident that COVID-19 is associated with prothrombotic complications due to immunothrombosis and endothelial dysfunction. Immunothrombosis is caused by the activation of neutrophils and monocytes and causes dysregulated coagulation cascade activation. This in turn leads to the formation of thrombi in blood vessels [2]. available, and commentaries were excluded. Animal and post-mortem studies, as well as those that could not be translated into English, were removed. The same cases can be reported in different databases; however, we made an effort to prevent duplication by comparing the patient's age, sex, date of admission, hospital name, and clinical, laboratory, and radiological characteristics. Additionally, we emailed the relevant corresponding author for clarification if there were any uncertainties.

Data Extraction and Study Quality Assessment
Two team members (AK and TF) searched and extracted data from various databases. Two reviewers independently reviewed all the articles and selected articles based on inclusion and exclusion criteria. Duplicate studies were excluded from the analysis. Any discrepancies were resolved by discussing them with the third reviewer. Reporting was done according to the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA). Two investigators (AK and TF) independently assessed the risk of bias in the included studies with the National Institutes of Health (NIH) Quality Assessment Tool [13]. The following information was extracted: author's name, publication date, study design, sample size, demographic characteristics, age, gender, comorbidities, type of vaccine received, number of doses received, time of onset of symptoms after vaccination, site of thrombosis and incidence, clinical features, laboratory parameters, imaging features, treatment received, patient outcome, study follow-up, and limitations.

Outcome Measures and Data Synthesis Strategy
We assessed the venous and arterial thromboembolic manifestations reported following different types of vaccination against COVID-19, as well as the vaccine dose, timing of onset of symptoms after vaccination, clinical features, laboratory and imaging results, treatment received, and outcome. Because there were few registries available and to prevent duplication with the reported cases, the descriptive analysis did not include national registries, surveillance, or reporting agencies' data. For the categorical variables, we used simple and relative frequencies and proportions. We employed measures of central tendency (mean or median) and dispersion (standard error and standard deviation) for continuous variables.

Results
Among the 1859 articles identified through database searching and other sources, 59 studies enrolling 202 patients were included in this systematic review ( Figure 1).

FIGURE 1: PRISMA flow chart for study selection
Out of them, one was a retrospective cohort, one was a retrospective survey, 17 were case series ( Table 2), and the rest were case reports. Most of the studies reported were from the United States, the United Kingdom, and Germany. A population-based cohort study and a nested incident-matched case-control study were also included. These two investigations reported a total of 6347 venous and arterial thromboembolic events. We also looked at three independent registries from the United States, the United Kingdom, and India.

Dosage and Vaccination
In 199 of the reported cases, information on the type of vaccine received was available. One hundred fiftythree of them received the AstraZeneca vaccine, 26 received the Pfizer/BioNTech vaccine, 16 received the J&J/Janssen vaccine, and four received the Moderna vaccine. The vaccine dose was documented in 152 patients, with 138 patients experiencing complications after the first dose (most common with the AstraZeneca vaccine) and another 14 experiencing complications after the second dose. Among the patients who developed complications with the second dose, most (11 patients) received the Pfizer/BioNTech vaccine.

Events Reported
The total number of thrombotic, thromboembolic, and hemorrhagic events reported among 202 patients was 306. Of those, 229 (74.8%) were venous thromboembolic events, and 39 (12.7%) were arterial thromboembolic events. Among them, in 174 patients for whom data was available, the mean duration of the onset of symptoms after vaccination was 9.8 ± 5 days. The majority of patients (79.9%) reported their initial symptoms between six and 14 days after vaccination, with the duration ranging from 30 minutes to 37 days. CVST was the most commonly reported form among all thromboembolic events, followed by pulmonary embolism and splanchnic thrombosis. Multiple thromboses were reported in 59 individuals (29.2%).

Nature of the Reported Events
CVST was reported in 133 patients (65.8%), and among them, 33 (24.8%) had associated brain parenchymal haemorrhage secondary to CVST. Also, 12 patients also had coexisting internal jugular vein thrombosis (IJVT). Other common systemic thromboses reported were pulmonary embolism (17.8%), splanchnic vein thrombosis (12.9%), and DVT (7.4%). Twenty patients presented with ischemic stroke, and the most common territory involved was the MCA territory (3/5 of reported cases). Other territories reported were the basilar artery and cerebellar infarcts. Two patients had transient ischemic attacks (TIAs), and one patient had an ICA non-occlusive thrombus. Seven patients had hemorrhagic strokes at admission. One patient had cardiac arrest, and three patients had myocardial infarction, including one case of triple coronary artery thrombosis and one case of right coronary artery thrombosis. Systemic venous system thromboses, such as iliac, inferior vena cava, iliofemoral, basivertebral, and bilateral superior ophthalmic vein thrombosis, were identified in seven cases. Other arterial systemic thromboses were found in 13 patients, with the most common being aortic thrombosis, followed by celiac, splenic, iliac, spinal, and femoral artery thrombosis. In five patients, cutaneous thrombosis associated with skin necrosis, dermal petechiae, and subdermal hematoma was reported. Five individuals had immune thrombocytopenic purpura (ITP), although one of them had a history and a family history of thrombocytopenia. Four patients were diagnosed with TTP. Six patients developed DIC during the course of the illness. Adrenal haemorrhage was reported in five patients. Thromboses and bleeding caused more neurological events (CVST, stroke, and TIAs) than non-neurological events ( Figure 2).

Investigations
The majority of patients (78.6%) had thrombocytopenia, with the lowest platelet count reported being 1 × 10 9 /L. A large number of individuals (87.4%) had elevated D-dimer. Heparin-induced thrombocytopenia (HIT) screen/anti-PF4 antibodies were tested in 150 patients and were positive in 86% of them. Four individuals had a heterozygous Factor V Leiden mutation, two had a prothrombin G20210A mutation, and one had a methylenetetrahydrofolate reductase (MTHFR) mutation. Five individuals tested positive for lupus anticoagulant, and one patient each had low protein S and low antithrombin III levels.

Treatment and Outcome
Treatment data were available for 161 patients. The majority of patients (128; 79.5%) received heparin or non-heparin anticoagulants. Intravenous immunoglobulin (IVIG) was given to 61 patients, steroids to 42 patients, and plasmapheresis (PE) to 11 patients. Intravenous methylprednisolone was the corticosteroid most frequently used, but data were scarce. Rituximab and eculizumab were administered to four and two patients, respectively, and one patient received caplacizumab. Out of the nine patients who improved completely, five received IVIG, one received PE, seven received IVIG or PE with steroids and heparin or other non-heparin anticoagulants, and two received only anticoagulants. In four cases, neurosurgical decompression procedures were performed. In 185 patients, data on their outcomes were available. At the time of publication, 110 patients were still being treated in the hospital, four patients were critical or in ICU care, and 49 patients (26.5%) had died. Twenty-two patients were discharged, and nine of those patients had fully recovered.  (7), cardiac events (3), adrenal haemorrhage (2), hemorrhagic stroke (2), and ischemic stroke (2). Out of the 27 patients who expired and whose treatment data was documented, 14 patients didn't receive any steroids, IVIG, anticoagulants, or immunosuppressants. Although data was lacking, patients with normal D-dimer and platelet counts had better outcomes. Only six patients received IVIG, and eight patients received combination therapy. Five patients received only heparin. Data on the type of vaccine administered among those who died was available in 48 patients, and 41 (85.4%) patients received the AstraZeneca vaccine, three patients each received the Pfizer/BioNTech and J&J/Janssen vaccines, and only one received the Moderna vaccine.

National Registries, Surveillance, and Reporting Agencies' Data
According to a population-based cohort study conducted by Pottegård

Risk of Bias
In our study, 26/59 of the papers were of fair quality (NIH Quality Assessment Tool) [13]. There were 17 good-quality articles and 16 poor-quality ones ( Tables 5-6).

Discussion
In this systematic review, we assessed the thrombotic and thromboembolic complications after receiving the COVID-19 vaccination, which included 59 articles with 202 patients and 306 events. Apart from that, data from two nationwide registries and surveillance revealed 6347 venous and arterial thromboembolic incidents. We also analyzed data from various reporting agencies separately. To the best of our knowledge, this is the most comprehensive review of the subject.
Data on the underlying mechanism of the AstraZeneca vaccine initiating TTS supports a two-step mechanism similar to the pathogenesis of autoimmune heparin-induced thrombocytopenia (HIT). As a result of a pronounced B-cell response triggered by vaccine-induced PF4/adenovirus aggregates and proinflammatory reactions, anti-PF4 antibodies are formed. These are IgG antibodies that activate platelets via low-affinity platelet FcγIIa receptors and the high-titer anti-PF4 antibodies that, in turn, stimulate platelets and neutrophils, causing neutrophils to release NETs (neutrophil extracellular traps). This leads to thrombosis in TTS [36]. The mechanism through which other types of vaccinations cause TTS is unknown.
For a definitive diagnosis of TTS, the American Society of Hematology requires that all five points listed below be met [12] (1) administratration of the COVID vaccine four to 42 days prior to symptom onset; (2) any venous or arterial thrombosis (often cerebral or abdominal); (3) thrombocytopenia (platelet count < 150 × 10 9 /L) (4) positive PF4 HIT ELISA; (5) markedly elevated D-dimer (more than four times the upper limit of normal).
The vast majority of our patients met all five criteria, although a few did not, owing to an insufficient evaluation in many studies, such as anti-PF4 antibodies, D-dimer, etc. The WHO interim guidance case definition of TTS includes the COVID vaccine 30 days before symptom onset [37].
TTS appears to be extremely uncommon. According to CDC and FDA data, the J&J/Janssen vaccine caused 3.1 cases per million doses, and the Moderna vaccine caused 0.006 cases per million doses [9]. The MHRA in the United Kingdom reported an overall incidence of major thromboembolic events with concurrent thrombocytopenia of 8.5 cases per million doses after the AstraZeneca vaccine, but 14.9 cases per million doses after first or unknown doses, 1.9 cases per million doses after second doses, and 0.4 cases per million doses after the Pfizer/BioNTech vaccine [10]. Only two cases of TTS were reported by the CDC and FDA after more than 356 million doses of Moderna vaccine were administered in the United States, and only two cases of major thromboembolic events with concurrent thrombocytopenia were reported by the MHRA after 1.4 million first doses and 0.9 million second doses of Moderna vaccine were administered in the United Kingdom [9,10]. So the highest incidence of TTS is with AstraZeneca, followed by J&J/Janssen, and the least with the Moderna vaccine. The majority of people who received the AstraZeneca vaccine experienced complications after the first dose. The majority of those who developed complications after the second dose of the vaccine received the Pfizer/BioNTech vaccine. Other adenoviral vaccines, such as Gam-COVID-Vac/SputnikV and Ad5-nCOV, have not been linked to thromboembolic complications. It's possible that this is related to the usage of different adenovirus species serotypes or insufficient reporting [38]. As previously stated, thrombotic and thromboembolic complications were more common in our study with the AstraZeneca vaccination. However, it is too early to comment on vaccine type and predisposition to these side effects, as the majority of people received the AstraZeneca vaccine, and thus the events reported were higher with the same vaccine. More research is needed to compare the risk of thromboembolic complications among different types of vaccines.
The most common events reported were venous thromboembolic events, with CVST being the most common, followed by pulmonary embolism and splanchnic thrombosis. Compared to this, the most common venous thromboembolic events recorded in HIT are pulmonary embolism and DVT [39]. Other events reported in TTS were DVT, internal jugular vein thrombosis, ischemic stroke, TIA, hemorrhagic stroke, cardiac arrest, myocardial infarction, thrombosis in the iliac, inferior vena cava, iliofemoral, basivertebral, bilateral superior ophthalmic veins, and arterial thrombosis in the aorta, celiac, splenic, iliac, spinal, femoral, and leg arteries, cutaneous thrombosis associated with skin necrosis and dermal petechiae, subdermal hematoma, ITP, TTP, DIC, and adrenal hemorrhage. TTS, therefore, occurs more frequently in atypical regions, such as the cerebral venous sinuses and splanchnic circulation, and HIT more frequently manifests as DVT and pulmonary embolism. Furthermore, DVT and pulmonary embolism are the most often observed thrombotic complications associated with COVID-19 infection [40].
In individuals with minor symptoms or who do not require hospitalization, the reported incidence of thrombotic complications after COVID-19 was reported to be very low. The reported incidence of venous thromboembolism (VTE) was only 0.09% [41]. In a study by Hill et al., only 3.1% of 2748 patients hospitalized with COVID-19 developed VTE, including DVT (1.5%) and pulmonary embolism (1.3%), while only 0.3% had arterial thromboses [41]. Another meta-analysis found a 14.1% in-hospital prevalence of VTE (pulmonary embolism or DVT), with a greater frequency in ICU patients (22.7%) [42]. Various other studies found the pooled incidence of ischemic stroke in COVID-19 patients to be 1.2%, CVST to be 0.08%, and myocardial infarction to be 0.5% in hospitalized, non-ICU-treated patients [43]. Hence, the prevalence of thrombotic complications, notably VTE, is substantially higher after COVID-19 infection than after COVID-19 vaccination, with a much higher prevalence rate among severe COVID-19 cases. It might be attributed to longer hospital stays among COVID-19 patients who are severely ill.
The majority of patients who developed thromboembolic events were young and female. HIT, which has a similar underlying mechanism, is also more common in females than in males [44]. The majority of the patients had no prior medical history or risk factors for the events.
Most of the patients in our study had the onset of symptoms within 14 days of receiving the vaccine dose, ranging from 30 minutes to 37 days post-vaccination. A complete blood count, peripheral smear, Ddimer, fibrinogen, anti-PF4 heparin antibodies on the enzyme-linked immunosorbent assay (ELISA), and imaging (computed tomography or magnetic resonance imaging with venography or angiography) should be used to rule out thrombosis in individuals suspected of having TTS. In the absence of heparin treatment, anti-PF4 antibodies are highly indicative of TTS. Our patients showed thrombocytopenia, high D-dimer levels, and positive HIT screens or anti-PF4 antibodies in the majority of cases. The platelet count and Ddimer levels were normal in a couple of the cases. Patients with normal D-dimer and platelet counts had better outcomes, despite the dearth of available data.
A systematic review of VIIT and CVST after AstraZeneca and J&J/Janssen vaccines, which included 14 articles and 49 patients, was recently published [45]. The majority of the patients in that study were female, and the PF4 IgG assay and D-dimer were both positive in the majority of the cases. Our research yielded comparable results. Symptoms emerged in the majority of patients one week after the first vaccine dosage (range, 4-19 days) in their study. However, the majority of patients in our research developed symptoms within 14 days of receiving the vaccination dose (a range of 30 minutes to 37 days). It's possible that this is related to the fact that we incorporated more publications.
In patients with suspected TTS, a multidisciplinary approach is suggested, comprising physicians from neurology, neurosurgery, hematology, critical care, internal medicine, radiology, and the emergency department [37]. Heparin is not advised in any of the recommendations. Platelet infusion should be avoided because of its similarities to HIT, except in emergencies where surgery is strongly indicated or severe thrombocytopenia is present. IVIG and non-heparin-based anticoagulants are recommended. The recommended IVIG dose is 1 g/kg per day for two days or 0.4 g/kg per day for five days. Parenteral direct thrombin inhibitors (argatroban or bivalirudin), direct oral anticoagulants (dabigatran, rivaroxaban, apixaban), fondaparinux, and danaparoid are non-heparin anticoagulants that are recommended [12,37,46]. If IVIG treatment is ineffective, the NICE guideline also suggests adding short courses of high-dose corticosteroids (methylprednisolone 1 g for three days or dexamethasone 20 to 40 mg for four days) [46]. Aspirin should be avoided because it can increase the risk of bleeding and is ineffective in preventing TTS [12]. Plasma exchange, eculizumab, and rituximab can be considered as treatment options in patients not responding to IVIG and anticoagulation [12,46]. Only a small number of individuals received these treatments (plasma exchange, eculizumab, and rituximab), and the prognosis was good. Those who received IVIG had a better prognosis as compared to others.
The limitation of our study was that the majority of the research in our review was case reports. The likelihood of thromboembolic complications reported following COVID-19 vaccination may be greater due to the potential underreporting of side effects. The strength of our analysis is that we used data from national registries, monitoring agencies, and reporting agencies, as well as observational studies and case reports, to compile it. The majority of the studies were of fair to good quality.

Conclusions
COVID-19 vaccines are generally safe, and the risk of death from COVID-19 disease far outweighs the risk of TTS from the COVID-19 vaccine. Thrombotic complications are more likely due to COVID-19 infection than post-vaccination, so concern about these events should not deter patients from getting the COVID-19 vaccine. Clinicians should be aware of TTS as it is potentially fatal, and early diagnosis and appropriate treatment can save lives. Vaccine side-effect profiles must be evaluated in future trials.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.