Written by Preethi Ramchand, MD; Sara Stern-Nezer, MD, MPH
Original Article: Scutelnic, A., Krzywicka, K., Mbroh, J et al. Management of Cerebral Venous Thrombosis Due to Adenoviral COVID-19 Vaccination. Ann Neurol (2022); 92: 562-573. https://doi.org/10.1002/ana.26431
Background
One of the documented adverse effects of adenovirus-based severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccinations is cerebral venous thrombosis (CVT). The postulated pathophysiological mechanism is an immune-mediated consumptive thrombocytopenia, also known as vaccine-induced immune thrombocytopenia (VITT), similar in mechanism to heparin-induced thrombocytopenia (HIT). Because of this similarity, treatment guidelines published on March 28, 2021, as defined by the International Society of Thrombosis and Haemostasis (ISTH), departed from standard CVT recommendations to those more similar to the management of HIT avoiding heparin due to hypothesized cross-reactivity of platelet-activating antibodies and focusing instead on: immune-modulating therapies, avoidance of heparin-type anticoagulants, and avoidance of platelet transfusions. This study aims to establish whether guideline changes have translated into changes in clinical practice and to determine whether compliance with recommendations led to reduced mortality from VITT-induced CVT (VITT-CVT).
Methods
Patient data was collected from a prospective international CVT registry. Patients included in the present study developed CVT, confirmed by neuroimaging or autopsy, within 28 days of SARS-CoV-2 vaccination and were selected according to pre-specified criteria by the British Society for Haematology for possible, probable, and definitive VIIT-CVT. Exclusion criteria included CVT after 28 days of vaccination, administration of non-adenovirus SARS-CoV-2 vaccine, and those with low suspicion of VITT Treatment considered compliant with 2021 ISTH VIIT-CVT. Recommendations included all three of the following criteria:
- Administration of immune-modulating treatment (intravenous immunoglobulins (IVIG) or plasma exchange)
- Treatment with non-heparin anticoagulants (or no anticoagulants if profound thrombocytopenia)
- Avoidance of platelet transfusion needed for pre-operatively
Descriptive statistics were used to assess temporal trends in management, treatment adherence, and treatments and outcomes for patients treated with various regimens. Nonparametric statistics were used to compare the following parameters among patients pre- and post-dissemination of modified VITT-CVT ISTH recommendations: baseline variables, adherence to recommendations, treatment regimens, and mortality. Binary logistic regression was used to calculate odds ratios for mortality per type of treatment regimen, with adjustment for age, presence of coma (GCS< 9), presence of ICH at onset, and baseline platelet count.
Results
In total, 99 patients met inclusion criteria for definite, probable or possible VITT. The median age in this group was 47, of which 75% were female. 92% received the ChAdOx1 nCov-19 vaccine (Vaxzevria, AstraZeneca/Oxford) and 8% received the Ad26.COV2.S vaccine (Janssen/Johnson & Johnson). Mortality decreased over time from March to May (52% v. 29%, p = 0.034), without significant difference in mortality between treatment-compliant and non-compliant groups (32% vs. 52%, adjusted OR = 0.43, 95% CI = 0.16-1.19). Of the 3 management criteria listed above, only immunomodulation was associated with improved mortality; anti-coagulation use, and platelet transfusion management variations were not significantly associated with mortality after adjustment for age, coma, ICH and initial platelet count.
Effects of immunomodulation: 65% of the study population received immune therapy, of which 94% was IVIG, 5% was IVIG and plasma exchange, 38% received adjuvant steroid treatment, and 11% received steroid monotherapy. 86% of the study population received systemic anticoagulation, of which 15% was heparin-based, 59% was non-heparin anticoagulation, and 26% was a mixture of both. However, patients treated with immunomodulation had a lower mortality rate (29% vs. 70%, adj OR = 0.19, 95% CI 0.06-0.58).
Effects of platelet transfusion: 27% of all patients received a platelet transfusion, of which 56% were for preoperative optimization. Although there was no significant difference between baseline platelet count among those who received and did not receive transfusion, patients who were transfused were more likely to have intracranial hemorrhage (81% vs 68%), worse baseline neurological function, and underwent more decompressive hemicraniectomies.
Temporal variation in outcomes: Cases diagnosed earlier in the cohort had a higher incidence of ICH at onset (p = 0.015). 44% of the 99 patients in the study were treated per the new recommendations, and there was significant increase in compliance over time with regards to both the composite of all three key components (13% to 46% to 76% from March, April, and May onwards, respectively; p < 0.001) as well as to each of the three components individually.
Commentary
The primary findings of this study are that:
1) Adherence to VITT-CVT recommendations increased over time
2) CVT mortality decreased over time
3) Despite the above, there was no overall difference in mortality among those who were treated according to the new 3-prong recommendations versus those who were not.
That being said, those treated with immune therapy did have a lower incidence of death compared to those who were not, potentially speaking to the underlying immune-driven pathophysiology of this specific disease.
Increased adherence to new treatment recommendations did not seem to drive the finding that mortality decreased over time. The authors speculate that this could be due to decreased severity of CVT over time (citing the lower incidence of intracranial hemorrhage at time of onset over time) as well as improved physician awareness of the disease, but mortality remains higher than CVT unrelated to vaccination.
This study offers a granular look at patient-level data and includes patients across multiple hospitals and countries with a prospective design. However, despite these strengths, the small sample size may underestimate mortality of things such as platelet transfusion or the effects of guideline adherence. Additionally, the confounders and complexities posed by heterogenous clinical practices may introduce bias, however it is challenging to study rare diseases.
Although there was an increase in adherence and decrease in mortality over time, this was not found to be a causal relationship, speaking to the complexity of the disease and the need for ongoing research in this area, but favoring immunomodulatory therapy and a different approach to vaccine-associated CVT than our standard approach.
#LiteratureWatch#NEWSReview