SUBMITTED:

August 2021

Accepted:

March 2022

G.W.C.M. Beelen (1), T.D. Kooiman (2), M. Coppens (3), S. Slot (2) Departments of 1 Emergency Medicine and 2 Intensive Care Northwest Clinics Alkmaar, Alkmaar, the Netherlands 3 Department of Vascular Medicine, Amsterdam Cardiovascular Sciences, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands

Correspondence:

G.W.C.M. Beelen - gwcm.beelen@nwz.nl
Clinical problem solving

Vaccine-induced thrombotic thrombocytopenia with saddle pulmonary embolism, a therapeutic challenge

Keywords:

Case
A 63-year-old woman was brought to the emergency department because of chest pain, severe dyspnoea and syncope. She had been experiencing abdominal discomfort and vomiting since her vaccination with ChAdOx1 nCoV-19 (AstraZeneca) 12 days earlier. Her past medical history included a cholecystectomy and a deep venous thrombosis of the lower extremity in 2010 without known predisposing factors. Upon arrival of the ambulance, the patient was responsive. However, once in the ambulance respiratory failure developed followed by cardiac arrest directly after arrival at the emergency department. Cardiopulmonary resuscitation (CPR) and advanced life support were started without any delay and the patient was intubated. After six CPR blocks (2 minutes each) with pulseless electrical activity, return of spontaneous circulation (ROSC) occurred with a blood pressure of 72/44 mmHg and a heart rate of 106/min without the use of additional inotropes or vasopressors. Mottling of the extremities and congestion of the upper jugular veins were noted. Bedside echocardiography showed right ventricular dilation and the ECG showed a T-wave inversion pattern in the anterior leads as well as the McGinn-White sign (figure 1), as an indication of acute right heart strain. With these findings, massive pulmonary embolism was considered the most likely diagnosis. This was confirmed by CT pulmonary angiogram, which showed a saddle pulmonary embolism with central obstruction (figure 2). The right ventricle was dilated, with a right ventricle/left ventricle ratio >1. Based on these findings, a loading dose of 10 mg alteplase was administered. COVID-19 was ruled out based on a negative SARS-CoV-2 RT-PCR on a nasopharyngeal swab and the absence of specific abnormalities on the CT scan. Laboratory testing (table 1) showed severe thrombocytopenia (11 x 109/l), with a prolonged prothrombin time and prolonged activated partial thromboplastin time (aPTT), indicating a disseminated intravascular coagulation (DIC)-like state. Other abnormalities included mildly increased inflammatory parameters, elevated liver enzymes and acute kidney injury.

Risk factors that might have contributed to the development of the pulmonary embolism included recent immobilisation and possibly a genetic predisposition, given the history of deep venous thrombosis. However, because our patient presented with acute thrombosis and thrombocytopenia within 30 days of receiving her SARS-CoV-2 vaccination, we also had a high suspicion of vaccine-induced thrombotic thrombocytopenia (VITT). This was later confirmed with a positive HIT-ELISA test and a heparin-induced platelet activation assay (HIPAA).

Clinical problems
The combination of severe thrombocytopenia and prolonged clotting times with the cardiac arrest due to obstructive shock caused by a saddle pulmonary embolism led to several therapeutic challenges in our patient. Firstly, there was an indication for reperfusion treatment with fibrinolytic therapy given the cardiac arrest and severe hypotension following ROSC. A recombinant tissue plasminogen activator (rt-PA, alteplase) is the most commonly used fibrinolytic agent in this setting. Although alteplase does not directly affect the platelet factor 4 (PF4) mediated pathway of VITT, it does lead to thrombus dissolution by converting plasminogen into plasmin. Importantly, bleeding diathesis is a relative contraindication for fibrinolysis.[1,2] Although our patient did not show any signs of bleeding at the time, the severe thrombocytopenia and DIC-like state increased her bleeding risk significantly. Indeed, bleeding complications have been reported in multiple patients with VITT throughout the literature.[3,4] Given the relatively stable situation following ROSC, we therefore deemed it necessary to confirm the diagnosis using a CT scan before administering fibrinolytic therapy.

Figure 1 NJCC V30-5.l

This led to the administration of a loading dose of 10 mg alteplase after 37 minutes following ROSC. Moreover, we decided to re-evaluate the situation before continuing with the rest of the alteplase infusion. Unfortunately, literature regarding systemic thrombolysis in patients with thrombocytopenia proved to be scarce.[5-7] Recommendations are solely based on expert opinion and include the use of endovascular management. However, the latter is not commonly used in our centre and the patient developed progressive signs of shock, requiring quick intervention. Upon arrival in the ICU, bedside point-of-care ultrasound (POCUS) showed an unchanged, dilated right ventricle with an obliterating left ventricle, consistent with persistent obstructive shock. Therefore, we decided to continue the alteplase infusion according to protocol with another 90 mg administered in two hours. No major bleeding occurred during and after treatment.

In addition to the above, we started treatment according to the current Dutch guidelines for the diagnosis and treatment of VITT, in consultation with an expert in vascular medicine.[4] This included administration of intravenous immunoglobulin (IVIG) 1g/kg per day for two consecutive days, and non-heparin anticoagulants. The choice of the non-heparin anticoagulant posed a second clinical challenge. Dutch guidelines suggest adjusting treatment to the patient’s clinical status and the anticipated need to stop anticoagulation, based on the risk of bleeding or the need for invasive procedures. Based on the presumed high risk of bleeding in our patient – due to the severe thrombocytopenia and recent fibrinolytic therapy – we chose intravenous argatroban. This selective thrombin inhibitor has a short half-life of 52 minutes and is also used in adult patients with heparin-induced thrombocytopenia and thrombosis (HIT).[8] The recommended initial dosage is 2 μg/kg/min for adult patients without hepatic impairment, or 0.5 μg/kg/min for critically ill patients. The dosage is then adjusted to a steady-state aPTT of 1.5-3.0 times the initial baseline value. For our patient, this would have resulted in a target value of 75-150 sec. Since her baseline aPTT was prolonged due to the DIC-like state and recent fibrinolytic therapy, we aimed for a narrower range of 75-100 sec.[7] In addition, we regularly measured fibrinogen levels,
aiming to keep them >1.0 g/l as recommended in the Dutch guidelines. The first fibrinogen level was measured after the loading dose of alteplase, showing a decreased value of 0.8 g/l (normal 2.0-4.0 g/l) (table 2). A few hours later, the aPTT became immeasurable due to very low fibrinogen levels (0.3 g/l), making it impossible to target dosing of argatroban.

Table 2 NJCC V30-5.l
Figure 2 NJCC V30-5.l

Despite administration of a total of 8 g fibrinogen, the fibrinogen level remained <1.0 g/l and the aPTT was still more than 100 seconds. Therefore, after careful consideration, the decision was made to switch to fondaparinux.

In addition to the above, targeted temperature management was started for 24 hours, with a target temperature of 36 °C in light of the patient’s coagulopathy. Within 24 hours after admission, the obstructive shock gradually resolved. The platelet count normalised, as did the renal function. No bleeding complications occurred. However, after cessation of the sedation, the patient developed a post-anoxic status epilepticus. This was thought to be due to the prolonged shock state, since CPR had been started without delay. In the next few days, she was treated with four antiepileptic drugs and three sedatives. At day seven of admission, distributive shock occurred concurrent with a rise in inflammatory parameters. No specific site of infection could be identified, but ceftriaxone, vancomycin and hydrocortisone were started based on a positive sputum culture (Klebsiella pneumoniae) and the possibility of a catheter-related infection. A component of obstructive shock was excluded with POCUS at day eight, which showed no signs of elevated right ventricular pressures. Although her haemodynamic status improved over the next few days, the status epilepticus persisted. At day 11 of admission, the prognosis was deemed infaust. The treatment was discontinued and the patient died 12 days after her initial presentation at the emergency department.

Discussion
VITT is a relatively new and rare vaccine-associated adverse event, characterised by thrombocytopenia and thrombosis occurring 4-28 days after vaccination with the ChAdOx1-S/nCoV-19 (AstraZeneca) or Ad26.COV2.S (Janssen) SARS-CoV-2 vaccines. Thrombosis is often seen at unusual sites, including the cerebral venous sinus and splanchnic veins. In addition to venous thrombi, arterial thromboembolisms, including aortoiliac thrombosis and stroke, have been described.[9,10] VITT closely resembles HIT, which is characterised by antibodies against the multimolecular complexes of heparin and platelet factor 4 (PF4, a protein stored in platelet α granules and released during platelet activation). These antibodies induce the release of procoagulant platelet microparticles including PF4 itself, thus creating a cascade of further platelet activation and thrombin generation. This cycle causes widespread thrombosis and a consumptive coagulopathy with thrombocytopenia, hypofibrinogenaemia and elevated D-dimer concentrations. In VITT, antibodies against PF4 are found in the absence of heparin as a trigger. It is currently unclear whether these are autoantibodies induced by the strong inflammatory stimulus of vaccination or antibodies induced by the vaccine that cross-react with PF4 and platelets. HIT-ELISA tests have the most appropriate sensitivity for anti-PF4 antibodies in VITT, but positive results must be confirmed by a functional HIPAA.[11] Its incidence has been estimated at 0.73 per 100,000 persons receiving a first dose of the ChAdOx1-S/nCoV-19 vaccine.[12] A higher risk of VITT is seen in patients with younger age and female sex.[11] With her age of 63, our patient belongs to the 15% of patients aged >60 years at the time of diagnosis of VITT. In the Netherlands, 40 cases of thrombosis with thrombocytopenia after vaccination with a COVID-19 vaccine (AstraZeneca and Janssen) were reported to the Netherlands Pharmacovigilance Centre Lareb until 1 August 2021. VITT was confirmed in 22 of these cases based on a positive HIT-ELISA and/or HIPAA or based on clinically severe thrombosis. Of these patients, 18 had received the AstraZeneca vaccine. Of the 40 patients with thrombocytopenia and thrombosis, six died, three of whom were presumed to have developed VITT following the AstraZeneca vaccine, according to Lareb.[13] In this article we have described one of these cases. Given the massive scale on which vaccinations are still taking place worldwide, it seems likely that some of our colleagues will encounter the same therapeutic challenges as we did.

The first and most difficult problem we encountered was the combination of pulmonary embolism and obstructive shock with severe thrombocytopenia. Although there are several published guidelines for the treatment of venous thromboembolism, recommendations on the management of pulmonary embolism in patients with severe thrombocytopenia are lacking. In this particular case we treated our patient with a full dose of alteplase despite the severe thrombocytopenia, given the lack of readily-available alternative options in our hospital. Our treatment was successful: the patient became haemodynamically stable and did not develop any bleeding. Alternatives to systemic thrombolysis include embolectomy by aspiration, catheter-directed fragmentation with local fibrinolytic therapy and surgical embolectomy. The last two procedures have been successfully performed in a few patients with HIT or thrombocytopenia due to another cause,[5,14,15] and current guidelines list embolectomy and catheter-directed treatment as reasonable alternatives in patients with contraindications to systemic thrombolysis if expertise and resources are available on-site.[1] Currently, several centres in the Netherlands provide these treatment options. In our case, we chose not to transfer the patient to an intervention centre, since this would have caused a significant delay in reperfusion treatment.

Table 2 NJCC V30-5.l

Secondly, the decision regarding anticoagulation therapy following alteplase infusion was challenging. Due to the high estimated bleeding risk in this particular case, we started with argatroban given its short half-life. However, argatroban dosing was complicated by immeasurable aPTT values due to severe hypofibrinogenaemia, which might have been caused by both a consumptive coagulopathy due to VITT and the fibrinolytic therapy. In the setting of hypofibrinogenaemia following alteplase treatment, fibrinogen replacement is not generally recommended, although this type of hypofibrinogenaemia has been associated with a higher bleeding risk in multiple patient populations. However, Dutch VITT guidelines do recommend considering fibrinogen replacement below a threshold of 1.0 g/l in patients with VITT and thrombosis. In our case, and after consultation with a vascular medicine specialist in a tertiary centre, we decided to comply with the VITT guidelines given the need for a reliable parameter to guide argatroban dosing. For future patients presenting with VITT and an indication for fibrinolytic therapy, we recommend measuring fibrinogen levels and aPTT values before the start of this treatment. In case of immeasurable aPTT values, early and aggressive administration of fibrinogen should complement argatroban therapy, for example by using a suppletion threshold of 1.5 g/l, in order to prevent levels from dropping below 1 g/l. When using an alternative agent, other characteristics should be carefully weighted. In this specific example, our second choice was fondaparinux. This agent does not require follow-up monitoring of clotting times. However, it has a longer half-life, which makes its use less attractive in case of a high risk of bleeding. Other alternatives include danaparoid or direct oral anticoagulants (DOACs). Of note, the absorption of DOACs is unpredictable in periods of shock. Given the multitude of options, consultation with an expert in vascular medicine is warranted. Moreover, setting up a pulmonary rapid response team might aid in therapeutic decision-making, and this is also recommended by the 2019 ESC guidelines.

Conclusion
VITT is a relatively new and rare side effect of the AstraZeneca and Janssen SARS-CoV2 vaccines. Although it has a low incidence, consequences for the individual patient can be severe. In our patient, who presented with a cardiac arrest and obstructive shock due to pulmonary embolism, treatment with alteplase was successful and did not cause any bleeding despite severe baseline thrombocytopenia. However, subsequent aPTT values were immeasurable due to hypofibrinogenaemia, hampering adequate argatroban dosing. For future patients presenting with VITT and an indication for fibrinolytic therapy, we recommend measuring fibrinogen levels before the start of this treatment. When using argatroban as a subsequent anticoagulant therapy, early and aggressive replacement of fibrinogen may be necessary. Alternatively, other anticoagulants may be considered, taking other drug characteristics into account (e.g. absorption, half-life, availability of a specific reversal agent). Furthermore, endovascular treatment of pulmonary embolism might form an alternative if resources are available on site. Consultation with an expert in vascular medicine is warranted and formation of a pulmonary rapid response team is encouraged.

Disclosures
All authors declare no conflicts of interest. No funding or financial support was received.

References:

  1. Konstantinides SV, Meyer G, Bueno H, et al. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European respiratory society (ERS). Eur Heart J. 2020;41:543-603.
  2. Ucar EY. Update on thrombolytic therapy in acute pulmonary thromboembolism. Eurasian J Med. 2019;51:185-9.
  3. Pottegård A, Lund LC, Karlstad Ø, et al. Arterial events, venous thromboembolism, thrombocytopenia, and bleeding after vaccination with Oxford-AstraZeneca ChAdOx1-S in Denmark and Norway: Population based cohort study. BMJ. 2021 May 5;373:n1114.
  4. FMS Leidraad_trombopenie_met-zonder_trombose_vaccinatie. https://demedischspecialist.nl/sites/default/files/leidraad_trombopenie_met-zonder_ trombose_vaccinatie.
  5. Muñoz Tovar RA, Alvarez Perdomo LC, Rojas Molina SM, Salazar SJ. Submassive Pulmonary Thromboembolism in a Patient with Thrombocytopenia: Therapeutic Challenge. Case Rep Crit Care. 2019; doi: 10.1155/2019/1919401.
  6. Zalpour A, Hanzelka K, Patlan JT, Rozner MA, Yusuf SW. Saddle pulmonary embolism in a cancer patient with thrombocytopenia: A treatment dilemma. Cardiol Res Pract. 2011. doi: 10.4061/2011/835750.
  7. Cohen JI, Cooper MR, Greenberg CS. Streptokinase Therapy of Pulmonary Emboli With Heparin-Associated Thrombocytopenia. Arch Intern Med [Internet]. 1985;145:1725-6.
  8. Grouzi E. Update on argatroban for the prophylaxis and treatment of heparin-induced thrombocytopenia type II. J Blood Med. 2014;5:131-41.
  9. Scully M, Singh D, Lown R, et al. Pathologic Antibodies to Platelet Factor 4 after ChAdOx1 nCoV-19 Vaccination. New Engl J Med. 2021;384:2202-11.
  10. Greinacher A, Thiele T, Warkentin TE, Weisser K, Kyrle PA, Eichinger S. Thrombotic Thrombocytopenia after ChAdOx1 nCov-19 Vaccination. New Engl J Med. 2021;384:2092-101.
  11. Klok F. ‘Vaccine-induced immune thrombotic thrombocytopenia.’ Lancet Haematol. 2022;9:e73-80.
  12. Chan BT, Mpa M, Bobos P, et al. Meta-Analysis of Risk of Vaccine-Induced Immune Thrombotic Thrombocytopenia Following ChAdOx1-S Recombinant Vaccine. Available from: https://doi.org/10.1101/2021.05.04.21256613.
  13. https://www.lareb.nl/pages/update-van-bijwerkingen.
  14. Badreldin HA, Rimsans J, Connors JM, Wiviott SD. Use of systemic bivalirudin with catheter-directed thrombolysis in a patient with heparin-induced thrombocytopenia: A case report. Cath Cardiovasc Interv. 2018;91:E17-20.
  15. Taniguchi S, Fukuda I, Minakawa M, Watanabe KI, Daitoku K, Suzuki Y. Emergency pulmonary embolectomy during the second trimester of pregnancy: Report of a case. Surg Today. 2008;38:59-61.