A Systematic Review of Reported Cases of Immune Thrombocytopenia after COVID-19 Vaccination

With the recent outbreak of the COVID-19 pandemic and emergency use authorization of anti-SARS-CoV-2 vaccines, reports of post-vaccine immune thrombocytopenia (ITP) have gained attention. With this systematic review, we aim to analyze the clinical characteristics, therapeutic strategies, and outcomes of patients presenting with ITP after receiving COVID-19 vaccination. Medline, Embase, and Ebsco databases were systematically explored from inception until 1 June 2022. Case reports and case series investigating the association between the anti-SARS-CoV-2 vaccine and ITP were included. We found a total of 66 patients. The mean age of presentation was 63 years with a female preponderance (60.6%). Sixteen patients had pre-existing ITP. The mean time from vaccine administration to symptom onset was 8.4 days. More ITP events were triggered by mRNA vaccines (BNT162b2 (n = 29) > mRNA-1273 (n = 13)) than with adenoviral vaccines (ChAdOx1-S AstraZeneca (n = 15) > Ad26.COV2-S (n = 9)). Most of the patients were treated with steroids or IVIG, or both. The overall outcome was promising, with no reported deaths. Our review attempts to increase awareness among physicians while evaluating patients presenting with thrombocytopenia after receiving the vaccine. In our solicited opinion, the rarity of these events and excellent outcomes for patients should not change views regarding the benefits provided by immunization.


Introduction
Immune thrombocytopenia (ITP) is an autoimmune disorder characterized by low platelet counts (<100 × 10 3 /µL) unexplainable by an alternative etiology [1]. ITP carries an annual incidence of about 3 cases per 100,000 adults, with a predilection for the female gender in the younger population [2]. Although patients may be asymptomatic upon presentation, typical clinical features include mucocutaneous bleeding, such as petechiae; purpura; ecchymoses; and sometimes, hemorrhage, with intracranial being the most serious [1,3]. While primary ITP is idiopathic in origin, secondary ITP can be caused by other autoimmune disorders, cancer, infection, or medications and accounts for less than a fourth of total ITP cases [4]. Amongst drugs, almost half of the cases are attributed to vaccines, with the measles-mumps-rubella (MMR) vaccine being the most common culprit [5]. With the recent outbreak of the COVID-19 pandemic and emergency use authorization of anti-SARS-CoV2 vaccines, reports of post-vaccine thrombocytopenia have gained attention [6]. Vaccine-induced immune thrombotic thrombocytopenia (VITT) has now been increasingly recognized, predominantly after the administration of adenovirus-vector-based vaccines [7]. It is marked by the formation of widespread thrombi and positive platelet factor 4 antibodies, a lab parameter classically seen in patients with heparin-induced thrombocytopenia. Furthermore, reports of post-vaccine thrombotic thrombocytopenic purpura, defined by induced thrombocytopenia. Furthermore, reports of post-vaccine thrombotic thrombocytopenic purpura, defined by low ADAMTS-13 activity and microangiopathic hemolytic anemia, have also emerged [8]. Contrary to these two entities, vaccine-related ITP involves isolated thrombocytopenia and has a relatively favorable prognosis. That said, patients with ITP still carry a higher thromboembolism risk and increased mortality compared to the general population [2]. Therefore, it becomes imperative to understand its epidemiology in relation to the administration of COVID-19 vaccines. With this review, we aim to analyze the clinical characteristics, presenting features, laboratory parameters, therapeutic strategies, and outcomes of patients presenting with ITP after receiving COVID-19 vaccination.

Search Strategy and Selection of Studies
Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, Medline, Embase, and Ebsco databases were systematically explored from inception until 1 June 2022, with the following keywords: "Purpura, Thrombocytopenic, Idiopathic"(Mesh) AND "COVID- 19 Vaccines"(Mesh) OR "2019-nCoV Vaccine mRNA-1273"(Mesh) OR "BNT162 Vaccine"(Mesh) OR "ChAdOx1 nCoV-19"(Mesh) OR "Ad26COVS1"(Mesh). The search was accomplished by two independent authors (PS and NG). Papers hence identified underwent screening at the title and abstract level. The following inclusion criteria were used: 1. case reports and case series investigating the association between the anti-SARS-CoV-2 vaccine and ITP; 2. presence of isolated thrombocytopenia in the absence of thrombosis; 3. ITP is diagnosed after ruling out other causes of thrombocytopenia. Duplicate articles, articles in languages other than English, and studies reporting other causes of thrombocytopenia were excluded. Cohort studies, data from surveillance systems, and review articles were excluded as detailed information on demographics, treatment, and outcome was required for each patient to synthesize the results of this analysis. After reviewing the full text of the eligible articles and overcoming disagreements through discussion, a total of 43 case reports (66 patients) were included in this study ( Figure 1).

Data Extraction
We extracted the following data: 1. author name and year of publication; 2. gender of the patient and age at presentation; 3. comorbidities; 4. presenting features; 5. platelet Vaccines 2022, 10, 1444 3 of 16 counts at presentation, or nadir if admission counts were not reported; 6. type and dose of the vaccine received; 7. time to symptom onset or presentation post-vaccine, whichever came early; 8. therapies received; 9. outcomes; and 10. whether the second dose, if applicable, was administered or not.

Results
We found a total of 66 patients with ITP following COVID-19 vaccination, as listed in Table 1. The median age of presentation was 52 years (range: 19-86 years) with a female preponderance (60.6%, n = 40). Twenty-four patients had a pre-existing autoimmune disease (17 had pre-existing ITP), one was nine weeks pregnant, and one was receiving immunotherapy (durvalumab) for refractory lung adenocarcinoma. One patient with chronic ITP had a history of flare-up post-Shingrix vaccine. On presentation, two patients had concurrent active Hepatitis C and HIV viral infection, one had autoimmune hemolytic anemia (Evans syndrome), and one patient had weakly positive platelet factor 4 antibodies. Most of the patients (85%) presented with spontaneous mucocutaneous bleeding (gums > nose) or petechiae. Two patients presented with hemoptysis, and none with life-threatening intracranial hemorrhage. The mean time from vaccine administration to symptom onset was 8.4 days, with 73% of patients (n = 48) presenting after the first dose and 27% of patients (n = 18) after the second dose. More ITP events were triggered by mRNA vaccines (BNT162b2 (n = 29) > mRNA-1273 (n = 13)) than with adenoviral vaccines (ChAdOx1-S AstraZeneca (n = 15) > Ad26.COV2-S Johnson & Johnson (n = 9)). On laboratory workup, two patients had positive SS-A antibodies, one had positive GPIb IgG, two had positive lupus anticoagulant, and three had positive GPIIb/IIIa antibodies. A total of 71% of patients (n = 47) had thrombocytopenia of ≤10 × 10 3 /µL. Most patients were treated with steroids or IVIG, or both. Escalation of therapy with rituximab and thrombopoietin receptor agonists (TPO-RA) (eltrombopag or romiplostim) was needed in 22 patients (four patients had pre-existing ITP while 18 were newly diagnosed), out of which, in addition, two received vinca alkaloids, two received aminocaproic acid, one received danazol, one received Rho IgG, and one received fostamatinib. Seven patients, all with platelet counts of >30 × 10 3 /µL, were not treated. The overall outcome was promising, with no reported deaths. Ten patients had a relapse, either during hospitalization or post-discharge. One patient had an emergency room visit due to iatrogenic thrombocytosis from treatment (platelet transfusion, IVIG, steroids, TPO-RA, and vincristine during hospitalization). Four patients who developed ITP after the first dose received the second dose of mRNA-based vaccines with no further relapse. A comparison of new cases of ITP versus relapse post-vaccination is illustrated in Table 2.

Discussion
As of July 2022, over 500 million doses of the COVID-19 vaccine have been delivered across the United States [50]. While there have been around 300,000 reports of adverse outcomes following mRNA vaccination, more than 90% of those were non-serious [51]. Some major adverse events, such as myopericarditis, Guillain-Barre syndrome, and coagulopathy, including ITP, have prompted the need for closer surveillance in the peri-vaccination period [51]. The concept of vaccine-related ITP is not new and has been documented in relation to various other vaccines, such as MMR, influenza, hepatitis B, polio, pneumococcal vaccines, etc. [52]. Most studies reported the occurrence of thrombocytopenia within six weeks of receiving the vaccine and more than 90% of these cases were self-limiting, with only a few progressing to chronic thrombocytopenia [53,54]. With the advent of COVID-19 vaccines, it has become challenging to monitor such cases owing to the emergent need to countermeasure the pandemic, ushering in expedited manufacturing of the vaccines.
Diagnosis of ITP is one of exclusion, involving a thorough history taking to rule out other causes of thrombocytopenia and screening for secondary etiologies of ITP, along with the demonstration of mere thrombocytopenia on peripheral smear without any other hematologic abnormalities [3]. Antiplatelet antibodies are positive in less than two-thirds of patients with ITP, with poor specificity, and are therefore not recommended for diagnosis [55]. In all the cases described above, the authors came to the diagnosis of ITP based on the temporal sequence of events and the absence of any other inciting factors for thrombocytopenia. Despite the question of causality, it is worthwhile to underscore the possible pathophysiological mechanisms by which vaccine-associated ITP might occur (Figure 2). Molecular mimicry, epitope spreading, polyclonal activation, superantigen, and bystander activation are some suggested mechanisms. Analogous to natural infection-causing autoimmunity, both vaccines and their adjuvants carry the structural potential to generate and enhance self-reactivity, respectively [56]. This dysregulated immune response can also lead to the formation of immune complexes, which can additionally perpetuate platelet damage. Furthermore, antisense oligonucleotides, a constituent of the mRNA vaccines, have an inherent ability to cause thrombocytopenia, albeit with a need for a much higher dose than is delivered by a single injection [57]. Possible mechanisms proposed for this effect are platelet consumption through binding of receptors, formation of antibodies on repeated exposure, or an electrostatic platelet-binding effect similar to heparin [57]. Alternatively, a subclinical ITP can manifest as full-blown ITP post-vaccination [6]. Lastly, de novo ITP remains a distinct possibility, particularly in patients developing symptoms some days after vaccination, and is further affirmed by a positive response to traditional ITP-directed therapies, suggesting immune-mediated platelet destruction [6]. The potential mechanism involves vaccine-mediated polyclonal B-and T-cell activation causing both peripheral and bone marrow platelet destruction.
Consistent with natural ITP demographics in patients younger than 65, women were more likely to have vaccine-related ITP, in our review. ITP has long been associated with autoimmune diseases such as systemic lupus erythematosus, thyroid, and inflammatory bowel diseases [58]. A total of 13% of patients in our review had an established prior diagnosis of autoimmune disease, with thyroid disorders being the most common. It is noteworthy that 4.5% (n = 3) of patients with no history of autoimmune disorders had evidence of positive autoantibodies on laboratory workup. Given the short follow-up period, the inference of whether these patients had an undiagnosed autoimmune disease, or if antibodies were elevated as a consequence of ITP, is challenging to make. Tarawneh et al. [22] reported the resolution of SSA-antibodies on follow-up, thereby favoring the latter hypothesis. Two-thirds of the post-vaccine ITP cases were seen after mRNA vaccines (64%), and whether this is due to the upregulation of toll-like receptors by mRNA vaccines leading to further immune activation is still unknown [59]. A formal diagnosis of ITP was present in 24% of patients before presentation, highlighting the possible risk of relapse post-vaccination. The mean time to presentation was around eight days, which is consistent with other epidemiological studies [14,60]. A total of 86% of patients presented with thrombocytopenia of less than or equal to 30 × 10 3 /µL, the commonly accepted threshold of ITP patients presenting with major bleeding [61]. It is essential to highlight that there might have been other cases of subclinical ITP that have gone undetected, given the absence of symptoms. Consistent with natural ITP demographics in patients younger than 65, women were more likely to have vaccine-related ITP, in our review. ITP has long been associated with autoimmune diseases such as systemic lupus erythematosus, thyroid, and inflammatory bowel diseases [58]. A total of 13% of patients in our review had an established prior diagnosis of autoimmune disease, with thyroid disorders being the most common. It is noteworthy that 4.5% (n = 3) of patients with no history of autoimmune disorders had evidence of positive autoantibodies on laboratory workup. Given the short follow-up period, the inference of whether these patients had an undiagnosed autoimmune disease, or if antibodies were elevated as a consequence of ITP, is challenging to make. Tarawneh et al. [22] reported the resolution of SSA-antibodies on follow-up, thereby favoring the latter hypothesis. Two-thirds of the post-vaccine ITP cases were seen after mRNA vaccines (64%), and whether this is due to the upregulation of toll-like receptors by mRNA vaccines leading to further immune activation is still unknown [59]. A formal diagnosis of ITP was present in 24% of patients before presentation, highlighting the possible risk of relapse post-vaccination. The mean time to presentation was around eight days, which is consistent with other epidemiological studies [14,60]. A total of 86% of patients presented with thrombocytopenia of less than or equal to 30 × 10 3 /μL, the commonly accepted threshold of ITP patients presenting with major bleeding [61]. It is essential to highlight that there might have been other cases of subclinical ITP that have gone undetected, given the absence of symptoms.
Standard treatment guidelines for ITP favor a short course of steroids as the first-line therapy, with the addition of intravenous immunoglobulin (IVIG) to rapidly increase counts [62]. In our review, most patients showed an improvement in their platelet count Standard treatment guidelines for ITP favor a short course of steroids as the firstline therapy, with the addition of intravenous immunoglobulin (IVIG) to rapidly increase counts [62]. In our review, most patients showed an improvement in their platelet count and bleeding manifestations with the use of steroids and/or IVIG. Second-line agents were employed in around 33% of patients, which is higher than the use of such agents in a report of over 200,000 ITP patients, wherein only 5% of patients were on such therapeutic modalities [63]. Some patients (n = 7) with counts of >30 × 10 3 /µL were not treated per the 2019 recommended guidelines [62]. Except for one reported death from intracranial bleed due to suspected vaccine-related ITP [64], our review showed an excellent short-term prognosis in terms of zero fatalities and safe discharge if hospitalized, for all 66 patients. While data on long-term outcomes of such patients is lacking, prior studies with other vaccine types have shown patients to have a better prognosis than for viral-associated ITP, which is more likely to progress to a chronic state (28% versus 10% following vaccination) [65]. More prospective data are needed to guide patients on the long-term prognosis and chronicity (if any) of anti-SARS-CoV-2 vaccine-related ITP.
Conflicting evidence exists on whether the aforementioned cases were caused by vaccines or were a mere coincidence. As per some reports, the background incidence rates of ITP remained similar in the pre-and post-vaccination periods [6,66]. On the contrary, the administration of the AstraZeneca vaccine in Scotland, Australia, and England showed a higher-than-expected ITP rate [60,67,68]. More prospective data are needed to adjudicate this relationship accurately and identify predictive biomarkers. Whether the next dose should be advised in such patients and whether of the same vaccine type are other potential areas that need exploring. While only four patients in our review received the next dose, reports of safely immunizing chronic ITP patients, in some cases with an alternate vaccine, have been reassuring [67]. Monitoring of platelet counts in the peri-vaccination period, along with sharing the knowledge to present to the hospital emergently upon the first bleeding symptom, are a few of the potential steps that can be taken while administering the next dose.
The main limitations of our study are the absence of confirmatory tests or a standard definition for anti-SARS-CoV-2 vaccine-associated ITP. Even with a decent sample size, comments on the type of association cannot be made as the included studies were case reports. Furthermore, there exists a likelihood of bias in reporting subclinical cases with an emphasis on reporting instances of severe thrombocytopenia. Lastly, there is always a potential of missing germane articles with any review, despite employing a robust search strategy.

Conclusions
Although vaccine-related ITP is rarely a cause of death, it significantly hampers patients' quality of life owing to the fatigue and adverse effects of therapeutic interventions, making it a critical pathology to understand in the context of accelerated global vaccination efforts. Our review attempts to make physicians conscious of ruling out this entity while evaluating patients presenting with thrombocytopenia after receiving the anti-SARS-CoV-2 vaccine. In our solicited opinion, the rarity of these events and excellent outcomes for patients should not change views regarding the benefits provided by immunization to combat this global crisis. Instead, it should raise awareness about the need for an in-depth anamnesis before vaccination.