Large Vessel Occlusions By Free Floating Thrombi in Strokes During the COVID-19 pandemic- A Single Center Observational Study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.344655
Source of Support: None, Conflict of Interest: None
Keywords: Anticoagulation, COVID-19, free-floating thrombus, ischemic stroke, large vessel occlusion
Coronavirus disease (COVID-19), the pandemic declared by the World Health Organization on March 11, 2020 that has shaken the world with its widespread medical, social, and economic ramifications, is still poorly understood. While its prominent symptoms have been respiratory, with influenza-like illness, dyspnea, respiratory failure, and severe hypoxemia, extra-pulmonary manifestations such as diarrhea, renal failure, gastroenteritis, myocarditis, myalgias, and low exercise tolerance have also come to the fore. Reported neurological manifestations initially included ageusia, anosmia, acute hemorrhagic necrotizing encephalopathy, meningoencephalitis, central and peripheral demyelination, and seizures.,
However, subsequently increased incidence of strokes (5%–6%) have been reported, more so in moderate to severe COVID.,,,, In tandem, increased incidence of pulmonary embolism, deep vein thrombosis, and renal arterial thrombi have also been noted., This has led to the modification of treatment strategies for patients with moderate and severe disease across the world. Although there is profuse literature on stroke in COVID-19, no clear consensus regarding treatment and prevention has emerged. Here, we share the experience at our center with strokes during this period with special reference to a particular subset and describe its characteristic clinico-radiological features, treatment, and outcomes.
This study was conducted at a tertiary care center in Western Rajasthan, a desert state in India. It was an ambispective study (retrospective till June 2020 and prospective thereafter), approved by the Institutional Ethics Committee (AIIMS/IEC/2020-21/3038). Data were collected for all stroke patients admitted under the department of Neurology from January to December 2020. Details pertaining to demographic, clinical characteristics, disease progression, risk factors, laboratory evaluation, neuroimaging features, treatment, and outcomes were collected. Patients were categorized as ischemic stroke, hemorrhagic stroke, and venous strokes based on initial imaging [computed tomography (CT) or magnetic resonance imaging (MRI brain)]. Patients were subclassified into strokes with onset before the official outbreak of COVID-19 in India and those thereafter. This was done with a view to analyze the change in stroke profile across the pandemic.
Radiological evaluations such as angiography, venography, and perfusion studies were done as per stroke type. For ischemic strokes, non-contrast CT head and CT angiography (CTA) was done to define the core of infarct and cerebral blood flow at baseline and follow-up. CTA was performed to include the aortic arch upwards. CT perfusion studies were added for patients who presented within 24 h of ictus to assess perfusion deficits. The radiological findings of all patients were carefully reviewed by the neuroradiologists (S.T. and J.P.).
All relevant hematological and biochemical investigations for stroke were done with evaluation for thrombophilic states wherever possible or as indicated. Screening with electrocardiography, transthoracic echocardiography, and arterial Doppler of limbs was done routinely, and Holter monitoring as clinically indicated. Treatment was accorded as per the standard protocol of stroke management at the institute, which is as per National Stroke guidelines and recommendations during the COVID pandemic. The best possible medical, surgical and endovascular treatment was offered depending on stroke duration, type, and clinical situation.
The first case of COVID-19 in India was reported on January 30, 2020 and a country-wide and state-wide lockdown was rigorously implemented from March 25, 2020. Our institution is located in Jodhpur, a city in the state of Rajasthan, India having a catchment area including southern and western parts of this western border state and neighboring states of Gujarat, Madhya Pradesh, Haryana, and Maharashtra. It is a COVID-dedicated hospital but also continues to admit non-COVID patients. The first case of COVID-19 was reported in Jodhpur on March 21, 2020. The total number of stroke patients admitted under the Neurology services from January to December 2020 was 238, of which there were 56 (38 ischemic, 15 hemorrhagic, and three venous) strokes before lockdown and 182 (115 ischemic, 49 hemorrhagic, and 18 venous) during the COVID pandemic period spanning 9 months (April–December 2020).
Among all ischemic strokes, 25 and 86 patients had large vessel occlusion (LVO), nine and 18 patients had small vessel occlusion (SVO), whereas four and one patient had cardioembolic stroke in the pre-COVID lockdown period and COVID pandemic period, respectively. The in-hospital mortality was 5.4% in the pre-lockdown period and 11.5% in the COVID pandemic period.
Out of the 153 ischemic stroke patients, 21 patients (13.7%) had free-floating thrombus (FFT) in the vessels, 18 of whom were admitted between May and December 2020. An FFT was defined on CTA as a filling defect within the vessel lumen completely surrounded by contrast on at least two contiguous axial source images, producing the classical “donut sign.”
The detailed clinical and risk factor characteristics of these patients with FFT are elucidated in [Table 1]. There were 14 male and four female patients with a mean age of 51.6 (29–76 years). Seven patients had stroke in young, that is, were ≤45 years. Mean NIHSS at stroke onset was 15.2 (range: 3–25), and mean mRS was 4.5 (range: 3–5). Seven patients (38.9%) had been respiratory features of COVID-19, of which five had stroke in the second week, 4–10 days after testing positive for COVID-19. Traditional vascular risk factors for stroke were absent in seven patients and four had hypertension alone.
Hematological and biochemical profile suggesting heightened inflammatory state (raised ESR, hs-CRP, IL-6, ferritin) was noted in 14 patients (77.8%). Increased D-dimer levels were seen in ten patients, hyperhomocysteinemia in ten patients, MTHFR mutations in four patients, antiphospholipid antibody syndrome in one patient, while none had evidence of vasculitis. Eight (44.4%) tested positive for COVID-19 by nasopharyngeal swab RT-PCR, while seven were negative and three were not tested (as presented earlier in the pandemic when governmental indications for testing included influenza-like illness or high-risk contacts). Among the remaining without PCR-proven COVID-19, nine (50%) had surrogate markers of heightened inflammatory state and/or increased D-dimer, one of which had additional radiological features of COVID-19-specific lung involvement [Table 2].
Radiological features of infarct location and size as well as vessel and thrombus characteristics are given in [Table 2]. Out of all patients with LVO, 18.9% had FFT from January to December 2020. Of these, 20.9% and 12% had FFT in the COVID pandemic period versus the pre-COVID period, respectively [Figure 1]. Of stroke with FFT during COVID pandemic period, six had ASPECTS ≥7 on initial CT while ten had ASPECTS ≤6. Seven patients had multiple embolic infarcts in one or more arterial territories. On CTA, LVO was noted in 13 and five patients in anterior and posterior circulation, respectively. Multiple sites for FFT were seen in four patients while eight patients had thrombi in proximal vessels, that is, aortic arch, subclavian, brachiocephalic, or proximal common carotid artery [Figure 2]. One patient additionally had limb ischemia with gangrene of digits [Figure 3]. The presence of multiple arterial territorial infarcts and proximal arterial thrombi was a peculiar radiological feature in LVO with FFT.
Six patients in this group presented within the window period and were thrombolysed while one underwent mechanical thrombectomy (MT). All patients were given anticoagulation either immediately or 24 h after intravenous thrombolysis (IVT). Unfractionated heparin (UFH) or low molecular weight heparin (LMWH) was given and shifted to oral anticoagulants after 7–10 days. Five patients (27.7%) with FFT had a massive volume of infarcts involving complete MCA, ICA, or bilateral vertebrobasilar territories and rapidly deteriorated with impending herniation, three of whom were surgically treated with decompression craniectomy. All five patients in this subgroup could not be salvaged despite the best medical and surgical therapy and died [Figure 1] and [Figure 3]. Average mRS at discharge was 3.1 (range: 1–6), and mean mRS for the survivors (13 patients) at 3 months was 1.77 (range: 0–4).
This observational single-center study from western India has brought to fore a specific stroke causation during the COVID pandemic comprising FFT in the large vessels supplying cerebral vasculature. A total of 9.9% of all stroke types (18 patients) and 15.6% of ischemic strokes during the COVID pandemic period had FFT. Among ischemic strokes, those due to LVO was the predominant subtype (74.8%) reporting to the emergency room (ER) for neurological management. This could be due to inherent referral bias and patient preference bias. Patients with SVO may not have severe enough strokes and families either may have refused admission or did not seek to attend the ER due to treatment hesitancy for fear of contracting COVID-19 infection in the hospital. Other studies on stroke patterns during the pandemic have also shown a high incidence of LVO., Interestingly, more than one-fifth of LVO had FFT during the pandemic period, thereby emphasizing a higher predilection for it.
An FFT consists of friable thrombi that are either completely detached or partially attached to underlying vessel with one end free. The FFT tends to move in an arc-like pendulous manner with each cardiac cycle while remaining attached and is prone to break-away fragments. It may be formed due to underlying ulcerated atherosclerotic plaques and other vessel wall pathology such as dissection, trauma, or other cause for endothelial discontinuity.,, In addition, thrombophilic states such as malignancy, pregnancy, antiphospholipid syndrome, prothrombotic predisposition due to genetic mutations, and sepsis may contribute. Rarely, FFT may occur with a non-aneurysmal uninvolved vessel wall or without known etiology.
FFT is a potentially life-threatening situation that can predispose to ischemic strokes as well as peripheral and pulmonary embolism. Previous reports suggest an increased frequency of FFT in the aorta where it can manifest in 73% with embolic events such as distal embolism and limb ischemia. FFT in cerebrovascular circulation is invariably symptomatic with neurological manifestations.,, However, FFT has rarely been reported in cerebral ischemic stroke with an estimated incidence of 0.05%–1.45%.,, On the contrary, our study showed higher frequency during the COVID pandemic with intraluminal thrombi in large extracranial as well as intracranial vessels of anterior and posterior circulation and/or immediate proximal vasculature from aortic arch upwards (44.4%) posing a high risk. Interestingly, this and multiple thrombi (22.2%) predisposed to recurrent strokes in some patients even while on optimum antiplatelets, and transient ischemic attacks (TIAs) or minor strokes with stuttering course preceding the acute event in 50% [Table 1], thereby emphasizing need for early detection.
Analysis of risk factors showed a single risk factor in four patients while seven had none. In the pre-COVID era, advanced age and multiple cardiovascular risk factors predisposed to FFT, especially carotid plaque instability and underlying vessel wall pathology.,,, In addition, the heterogeneous group of hematological disorders presenting as cerebral ischemic events needs to be ruled out, especially in young strokes. In the present pandemic, a proinflammatory environment, endothelial dysfunction, and hypercoagulable state caused by COVID-19 may be deemed responsible for thrombus formation.,, Increased thrombotic events such as pulmonary thromboembolism, limb ischemia, DVT, and strokes (31%) have been noted during the pandemic and established in autopsy studies.,,,, However, the exact pathophysiology and complexity of the procoagulant state related to COVID-19 are yet to be clearly elucidated. A dangerous combination of hyper-inflammatory state with associated tissue hypoxia, prolonged immobilization, and diffuse intravascular coagulation play an instrumental role in some way. SARS-CoV-2 infection is known to trigger a cascade of inflammation with the release of mediators such as IL-6, hs-CRP, and surrogate markers such as ferritin. An increased D-dimer level has also been noted accompanying the cytokine storm in COVID-19.
Mohamud et al. suggested the interplay between increased thrombogenicity and plaque inflammation ultimately leading to plaque-rupture with thrombosis. However, the severity of stroke in COVID-19 may not be directly associated with carotid atherothrombosis as seen in 50% of our patients, where even mild atherosclerotic disease served as site for acute thrombus formation. Most reports with FFT in COVID patients were in patients without atherosclerosis.,, Additionally, these thrombotic episodes occurred despite receiving anticoagulant therapy in some cases.
While 44.4% of FFT were found to be COVID-PCR positive, 38.9% had active COVID infection and an additional 50% had surrogate inflammatory markers supporting suspected COVID. An unprecedented high number of patients with FFT seen at our center (20.9% of all LVO) during the ongoing pandemic raises various questions. In addition, the true COVID status of those testing PCR-negative in our subset is open to debate as a negative RT-PCR nasopharyngeal swab test with a detection rate of 65% is insufficient to rule out COVID-19. Thus, although a direct cause and effect relationship with COVID-19 infection and FFT as cause of stroke cannot be established in the present study, high frequency of a particular event repeating itself over a short time period definitely raises concerns. This may eventually be clearer as the pandemic evolves. However, this may suggest the possibility of thrombogenic state induced by symptomatic as well as asymptomatic SARS-CoV-2 infection during the ongoing pandemic.
To date, consequences of FFT in COVID-19 have been sparingly reported. While COVID-related aortic FFT presenting with limb ischemia and pulmonary symptoms have been reported, only three patients were detected during CTA screen for cause of stroke without FFT in the cerebral vasculature.,, The present study is the first report of documented FFT in cerebral vasculature in strokes noted during the COVID-19 pandemic.
The management of FFT-related stroke in COVID-19 needs to be aggressive and comprises medical, surgical, endovascular, or a combination of modalities. While recently published guidelines for stroke management in COVID-19 suggest otherwise, all patients with FFT reported here underwent anticoagulation irrespective of initial IVT/MT given. Anticoagulation was started immediately or after 24 h if thrombolysed. Published literature has shown that anticoagulation with or without antiplatelets led to complete dissolution of FFT in 86% when treated for 2–24 weeks without significant difference in outcome compared with the surgical arm. Though no randomized trials are there to support either therapeutic modality yet, early medical therapy may suffice in most cases without the need for emergent surgical or endovascular therapy.
Compared to the high mortality of up to 38% in COVID strokes, all patients improved remarkably from mean mRS at admission of 4.5 to mean mRS at discharge of 3.1., Only 27.8% died, with no age predilection (29–71 years), majority with no or single risk factor but having complete territorial malignant infarcts with brain edema not responding even to surgical decompression. Of the survivors, 69.2% had mRS of ≤2 at 3-month follow-up. The authors' experience with anticoagulation in stroke with this subset of ischemic stroke with FFT has been very encouraging. A rapid improvement in physical deficits in a short period of time and near-complete resolution of floating thrombus component augers well with a good outcome and further points to inflammatory origin and transient nature of thrombus.
The presence of FFT in the cerebral vasculature has been frequently observed in ischemic strokes during the COVID-19 pandemic. Active search for FFT in COVID positive or suspect patients, especially in young patients, without significant risk factors, multiple territorial infarcts with/without significant influenza-like illness, as well as assessment for factors contributing to thrombophilia need to be explored. Use of anticoagulation is encouraging, with good stroke outcomes as early as 1–2 months. With the renewed and steeper spike in COVID-19 cases, especially with new variant mutations, the resurgence of this finding needs to be actively explored early in the course of illness to reduce morbidity and mortality.
The authors would like to thank the ICMR- National Centre for Disease Informatics and Research, Bengaluru for funding the ongoing stroke registry at our center, a subset of which was studied in the present paper. We would also like to thank the Administration of All India Institute of Medical Sciences, Jodhpur, nursing staff, and healthcare workers who provided their support, as well as patients who were assessed in the course of the study.
Financial support and sponsorship
ICMR- National Centre for Disease Informatics and Research, Bengaluru, India.
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2]