COVID-19 Resources

Vaccine-induced Immune Thrombotic Thrombocytopenia

(Version 1.9; last updated May 9, 2022)

Input from James B. Bussel, MD; Jean M. Connors, MD; Douglas B. Cines, MD; Cynthia E. Dunbar, MD; Laura C. Michaelis, MD; Lisa Baumann Kreuziger, MD; Agnes Y. Y. Lee, MD, MSc; and Ingrid Pabinger-Fasching, MD.

Key Takeaways
Vaccine-induced Immune Thrombotic Thrombocytopenia (VITT)

  • Definitive Diagnosis (must meet all five criteria):
  1. COVID vaccine 4 to 42 days prior to symptom onset# 
  2. Any venous or arterial thrombosis (often cerebral or abdominal)
  3. Thrombocytopenia (platelet count < 150 x 109/L)*
  4. Positive PF4 “HIT” (heparin-induced thrombocytopenia) ELISA
  5. Markedly elevated D-dimer (> 4 times upper limit of normal) 
     
  • Incidence is extremely low. Risk of death and serious outcomes of COVID-19, including thrombosis, far outweigh risk of VITT possibly associated with highly efficacious vaccines.
  • Urgent medical evaluation for VITT is indicated if any of the following develop 4 to 42 days after vaccination:
  • Severe headache
  • Visual changes
  • Abdominal pain
  • Nausea and vomiting
  • Back pain
  • Shortness of breath
  • Leg pain or swelling
  • Petechiae, easy bruising, or bleeding
  • If VITT is suspected, perform immediate CBC with platelet count and imaging for thrombosis based on symptoms.
  • If thrombocytopenia or thrombosis are present, recommend urgent consultation from hematologist with expertise in hemostasis. Avoid use of heparin until VITT has been ruled out or until an alternative other plausible diagnosis has been made.
  • Initial work-up (a normal platelet count is less concerning for VITT*):
  • CBC with platelet count and peripheral smear
  • Imaging for thrombosis based on signs/symptoms
  • PF4-ELISA (HIT assay); draw blood prior to any therapies
  • Fibrinogen and D-dimer
  • Initiate therapy with intravenous immunoglobin and nonheparin anticoagulation pending PF4 ELISA results if:
  • Confirmed thrombosis AND at least one of the following
    • Low platelets* OR
    • Markedly elevated D-dimer OR
    • Both
       

If PF4 ELISA returns negative and there is no thrombocytopenia, VITTis ruled out; treat as standard venous thromboembolism.

OR

  • If thrombocytopenia and very high D-Dimer in absence of known thrombosis, particularly in the presence of severe headache, check PF4 ELISA, and consider treatment (see Q2)
  • If thrombocytopenia but no thrombosis and negative PF4 ELISA: likely ITP (see Q4)
  • Avoid platelet transfusions unless other treatments have been initiated AND life-threatening bleeding or imminent surgery  
  • Consider referral to tertiary care center if VITT is confirmed.

Knowledge about VITT continues to evolve, and updates will be made as new data become available.

#To date, VITT appears far more likely following AstraZeneca/Johnson and Johnson adenoviral vaccines than Moderna/Pfizer mRNA vaccines..

*A patient who presents with thrombosis and a normal platelet count post-vaccination might be in an early stage of VITT. Continued assessment for development of thrombocytopenia/VITT is required. Use of a non-heparin anticoagulant may be indicated if patient is 4 to 42 days post–vaccine with all other facets of the syndrome pending PF4 ELISA testing and additional CBCs.

On April 13, 2021, the U.S. Food and Drug Administration (FDA) and the Centers for Disease Control and Prevention (CDC) suggested pausing administration of the AD26.COV2.S Johnson & Johnson (JJ) vaccine to allow investigation of several cases of severe thrombosis with thrombocytopenia occurring post-vaccination. This announcement came on the heels of the initial reports of similar events in individuals receiving the CHaDOx1 nCov-19 AstraZeneca (AZ) vaccine outside the United States. Comprehensive clinical and laboratory characteristics of VITT have been reported in retrospective series and a large prospective cohort. This syndrome has been termed “vaccine-induced prothrombotic immune thrombocytopenia (VIPIT)” or “vaccine-induced immune thrombotic thrombocytopenia (VITT),” and “thrombosis with thrombocytopenia syndrome (TTS)” in communications from the CDC and FDA. The risk appears far lower for the two mRNA vaccines.

Following the initial April 2021 FDA/CDC review of all sources of event reporting in the United States following JJ vaccine administration, and a detailed risk/benefit analysis, the CDC Advisory Committee on Immunization Practices recommended that the JJ vaccine remain available in the United States for persons aged 18 years and older. An updated CDC review in December 2021, based on better case ascertainment following description of the syndrome, has now estimated the incidence to be as high as 1 in 100,000. The most recent information from the United Kingdom suggests an incidence for the AstraZeneca vaccine of 20.3 per million doses in those aged 18 to 49 years compared to 10.9 per million doses in those aged 50 years and older. Earlier clinical recognition and awareness of VITT has reduced mortality from as high as 50 percent in initial case series to 5 percent in a recent analysis from Australia. On May 5, 2022 the FDA modified the JJ vaccine EAU to limit administration only to those older than 18 years who are unable to receive any other vaccine due to anaphylaxis to mRNA vaccine components or unwillingness or inability to receive any other vaccine, grounded in concerns based on a rate of greater than three cases per million doses, and persistent high mortality in those developing VITT.

This FAQ is designed to provide an overview of considerations around the diagnosis and treatment of VITT. Specialists with expertise in thrombosis should be consulted as soon as VITT is diagnosed or under consideration, and many institutions will devise algorithms based on local availability of laboratory testing and drug availability.

What is VITT? 

A syndrome characterized by venous or arterial thrombosis, particularly at unusual sites including cerebral sinus venous thrombosis (CSVT)/splanchnic thrombosis; mild to severe thrombocytopenia; and positive PF4-heparin ELISA (“HIT” ELISA) was first described following the ChAdOx1 nCov-19 AZ vaccine, utilized extensively in the United Kingdom, Europe, and Canada, but not yet available in the United States. The largest series of 220 definite and 50 probable cases from the UK showed a median age of 48 years, and no gender bias.  Few patients had other known risk factors for thrombosis. Many of the patients were critically ill by the time thrombosis and thrombocytopenia were discovered, and up to one-third of the initially reported patients died.  The presence of severe thrombocytopenia (30 × 109/L) and/or intracranial hemorrhage was associated with the highest mortality.

VITT has also been reported in patients following the Ad26.COV2.S JJ vaccine. The clinical syndrome and time post-vaccination are very similar to cases following the AZ vaccine. Both vaccines utilize recombinant adenoviral vectors (chimpanzee for AZ and human for JJ) encoding the SARS-CoV-2 spike protein immunogen. In June 2021, a report of a VITT-like syndrome in single patient receiving the Moderna mRNA vaccine was published. The UK regulatory agency has reported 15 and two cases of major thrombosis with concurrent thrombocytopenia with Pfizer and Moderna vaccines, respectively; whether these are cases of VITT has not been confirmed.

The striking clinical similarities of VITT to heparin-induced thrombocytopenia (HIT) and the uniformly positive PF4-heparin ELISAs in these index cases led investigators to identify circulating PF4-reactive antibodies able to activate platelets in the absence of heparin. Intravenous immunoglobulin (IVIG) or a monoclonal antibody blocking the Fc receptor were able to prevent platelet activation by these antibodies in vitro. These clinical and laboratory features are similar to rare cases of HIT-like autoimmune thrombosis with thrombocytopenia, previously described following surgery, certain medications or infections in patients not receiving heparin. 

What clinical presentation should trigger consideration of VITT, and what is an appropriate initial work-up?

Mild-to-moderate constitutional symptoms such as fever, fatigue, headache, or muscle aches are common in the first 24 to 48 hours following vaccination and are not suggestive of VITT. Patients with severe, recurrent, or persistent symptoms from 4 to 42 days following COVID-19 vaccination, including intense headache, abdominal pain, back pain, nausea and vomiting, vision changes, change in mental status, shortness of breath, leg pain and swelling, and/or bleeding/petechiae, should be evaluated urgently by a medical provider, and consideration given to underlying VITT. The peak time period for initial symptoms is days 6 to 14. 

Diagnostic criteria of VITT vary between reports and based on geography. The five criteria for “Definitive” VITT (1. COVID vaccine 4 to 42 days prior to symptom onset; 2. Any venous or arterial thrombosis; 3. Thrombocytopenia [platelet count < 150 × 109/L]; 4. Positive PF4 “HIT” [heparin-induced thrombocytopenia] ELISA; 5. Markedly elevated D-dimer [> 4000 FEU or equivalent]) may not be met by every patient on presentation.  As further data are gathered on patients with possible VITT, “grey zone” probable or possible VITT should be considered in patients with any concerning symptoms in the post-vaccine time period. There are no clear answers, and as data on patients with earlier VITT diagnosis become available, recommendations may change. Examples of such situations include:

  • Patients followed by hematologists/oncologists who present post-vaccination with other potential reasons for thrombocytopenia and thrombosis. The PF4 ELISA can assist with diagnosis of VITT in these patients if thrombocytopenia is worsened from baseline.
  • Patients may present with a typical lower extremity venous thromboembolism (VTE) following vaccination in the presence of mild thrombocytopenia or a single low normal value. At present, avoidance of heparins in patients presenting with VTE in the post-vaccine window is reasonable while awaiting PF4 ELISA and following the platelet count. 
  • Patients who have thrombocytopenia and markedly elevated D- without other clear cause, even in the absence of documented thrombosis or suggestive symptoms, might benefit from initiating treatment while waiting for PF4 ELISA results. In particular, IVIG might be beneficial in those with platelet counts below 20 × 109/L as this reflects a consumptive coagulopathy. A recent report documented rapid improvement in laboratory parameters in such a patient treated with both IVIG and a non-heparin anticoagulant, without progression to thrombosis.
  • Patients with typical thrombosis, thrombocytopenia and very high D-dimers in the correct time window following vaccination, but a negative initial PF4 ELISA. This situation arose in 2.7 percent of the largest series and can be started on treatment for VITT (see below) while awaiting repeat PF4 ELISA and specialized platelet activation testing.

Initial work-up (note: draw blood prior to any therapeutic interventions such as IVIG, given potential interference with diagnostic assays):

  1. CBC with platelet count and peripheral smear. Median initial platelet count in largest series was 47 x 109/L with a wide range. Smear is needed to rule out pseudothrombocytopenia from platelet clumping.
  2. Imaging for thrombosis based on symptoms, focused on detection of cerebral sinus venous thrombosis (CSVT) with CT or MRI venogram, splanchnic thrombosis, pulmonary emboli, and/or DVT.
  3. D-dimer: the majority of VITT patients have markedly elevated values, over four times upper limit of normal.
  4. Fibrinogen: some VITT patients are reported to have low values
  5. PF4-heparin ELISA: almost 100 percent of cases reported had positive assays, with optical density greater than 2.0 to 3.0 in the majority. Non-ELISA rapid immunoassays for HIT are not sensitive or specific for VITT and should not be used.

Patients with worrisome symptoms and/or positive imaging in addition to low platelet counts and high D-dimer can be considered to have VITT and should be started on treatment (see below) while awaiting PF4 ELISA results. Whether the degree of PF4 ELISA positivity correlates with risk of VITT is unknown. Patients with low fibrinogen and extremely high D-dimer, suggesting disseminated intravascular coagulation, fall within the VITT syndrome. Microangiopathy with red cell fragmentation and hemolysis has not been a common feature of reported cases; however, at least one case with both VITT and TTP/HUS features has been reported. Thus, review of a blood smear and attention to signs of intravascular hemolysis would be appropriate.

Patients with isolated thrombocytopenia and continued absence of thrombosis may have post-vaccine ITP and not VITT (see below), as confirmed by a negative PF4 ELISA.

How should VITT be treated? 

This is a newly described syndrome, and all recommendations are based on extrapolation from HIT and to non–heparin-dependent autoimmune thrombotic thrombocytopenias, analysis of the clinical features in reported cases, and predictions based on laboratory investigations of pathophysiology. Several groups have published detailed position papers on VITT that include expert consensus recommendations and algorithms. Note that prospective clinical treatment studies do not exist, and that the response rate of these interventions has not been established.

The risk of heparin administration remains unclear. Many patients presenting prior to the initial reports of VITT received heparin anticoagulation, with poor outcomes; however, these patients were clinically very advanced before VITT was recognized. The recent large UK series reported 20 percent mortality in those receiving heparin and 16 percent in those not receiving heparin. While the syndrome resembles HIT, current understanding of VITT pathophysiology suggests that VITT antibodies bind the same site on PF4 as heparin, which can explain the inhibition of platelet activation by heparin seen in some patients’ laboratory testing with VITT.  Given the availability of alternative anticoagulation regimens and lack of consensus regarding the safety of heparin in VITT, we currently suggest avoiding heparin or low molecular weight heparin (LMWH) anticoagulation in these patients.

In patients presenting with thrombocytopenia, documented or suspected thrombosis, and a positive or pending PF4 ELISA 4-42 days post-vaccination, we recommend rapid initiation of treatment, analogous to treatment of severe HIT, including:

  1. IVIG 1 g/kg daily for two days. Note that patients with new clots arising post-IVIG have been reported; thus, patients should be carefully monitored even after treatment has begun, and initiation of anticoagulation coincident with IVIG is recommended.
  2. Non-heparin anticoagulation, chosen based on the clinical status and organ function of the patient:
    1. Parenteral direct thrombin inhibitors (argatroban or bivalrudin), OR
    2. Direct oral anticoagulants without lead-in heparin phase, OR
    3. Fondaparinux, OR 
    4. Danaparoid
  3. Low fibrinogen or bleeding are associated with VITT and should not absolutely preclude anticoagulation, particularly if platelets are 20 x 109 or rising following IVIG initiation. Concurrent replacement of fibrinogen in patients with bleeding and/or very low values should be considered. 
  4. Based on similarities to HIT, avoid platelet transfusions. However, risk/benefit assessment in individual patients with serious bleeding and/or need for surgical intervention may favor platelet transfusion following initiation of IVIG, non-heparin anti-coagulation, and fibrinogen replacement (if deficient).
  5. Corticosteroids have been administered along with IVIG in some cases, with no consensus or data to date on their role.
  6. Avoid aspirin as either treatment or prophylaxis for VITT. Aspirin is not efficacious in preventing HIT antibodies from activating platelets and could increase the risk of bleeding.
  7. Additional therapies: Plasma exchange has been utilized in patients with severe disease. In the UK series, a survival of 90 percent was reported in patients treated with plasma exchange. It can be considered if a patient shows continued thrombosis despite IVIG and non-heparin anticoagulation. The large extravascular volume of distribution of IgG antibodies, causative in both HIT and VITT, prevents rapid or complete removal via plasma exchange, and the concurrent bleeding complications in VITT may make catheter placement and prolonged apheresis challenging. Complement inhibition with eculizumab has been utilized in several patients progressing after IVIG and anticoagulation, with evidence for improvement.

At this time, the duration of risk of thrombosis in patients with VITT is not known. Pending more data, those with documented thrombosis should receive a minimum of three months anticoagulation, as for any provoked VTE.

What if a patient presents with thrombocytopenia or bleeding post-vaccination?

Over 100 cases of new-onset acute immune thrombocytopenia purpura (ITP), with at least one fatal, have been diagnosed in the same timeframe following vaccination as VITT (median, 8 days). These occurrences have been noted following AZ and JJ as well as Moderna and Pfizer vaccines. The platelet count at presentation is often 10 x 109/L, somewhat lower than in VITT (median, 20 x 109/L), and thromboses have not been associated with these cases, although very few cases had a PF4 ELISA checked. Most presented with bleeding. Estimates to date suggest that post–COVID vaccine ITP is rare (1 in 100,000 to 1 in 1,000,000) and may be related to vaccination or represent a coincidental event. Most patients respond to the combination of IVIG, and/or steroids, with platelet transfusions if bleeding. Thrombopoietin agents and possibly a single dose of vincristine may be useful if there is not an immediate (2-4 days or cessation of bleeding) response to IVIG and/or steroids. Avoidance of rituximab is important because of slow onset of action (weeks), negation of recent vaccine-induced immunity, and inability to vaccinate again for more than six months. For more details regarding treatment see the ITP FAQ.

It is important to rule out VITT with a negative PF4 ELISA assay in patients presenting with thrombocytopenia four to 42 days post-vaccination, even in the absence of symptoms suggestive of thrombosis. While awaiting PF4 ELISA results, IVIG could be administered to patients with profound thrombocytopenia and bleeding, given the indication for this medication in the treatment of both ITP and VITT. 

Patients with pre-existing ITP or other causes of thrombocytopenia may have transient further lowering of platelet count following vaccination.  Whether a PF4 ELISA should be sent in such patients in the absence of signs of thrombosis is unclear, but until more information is available, it would seem prudent to send a screening ELISA in those with a clear significant decrease in platelet count occurring during the relevant timeframe.

Should the risk of VITT impact vaccination approaches in hematology patients?

Currently, the AZ and JJ vaccines remain available in many countries around the world and remain critical components in fighting the COVID-19 pandemic. In most settings, the risks of COVID-19 disease far outweigh the very low risk of VITT following AZ or JJ vaccines. However, in settings and countries with availability of mRNA vaccines, the lower risk of VITT and higher efficacy against breakthrough infections (and especially hospitalizations and severe disease for mRNA as compared to adenoviral vaccines) has resulted in both governmental recommendations and policies favoring use of mRNA vaccines, particularly in the need for booster shots to improve protection against newer variants such as Delta and Omicron. Studies have demonstrated efficacy and safety of heterologous regimens (i.e., administration of a different vaccine for a second shot or booster, thus the vast majority of JJ recipients in the United States have received mRNA rather than JJ boosters. On December 16, 2021, the CDC endorsed the vaccine advisory panel’s recommendation that mRNA vaccines be preferred over the JJ vaccine for both initial and booster injections.

Given difficulties in diagnosing VITT in the presence of ongoing thrombocytopenia and lower seroconversion rates for adenoviral vaccines compared to mRNA vaccines in some groups of immunocompromised individuals, many hematology patients will have even stronger indications for use of mRNA versus adenoviral vaccines.

Resources