key: cord-1029535-pskjnb1q
authors: Baldini, Tommaso; Asioli, Gian Maria; Romoli, Michele; Carvalho Dias, Mariana; Schulte, Eva C.; Hauer, Larissa; Aguiar De Sousa, Diana; Sellner, Johann; Zini, Andrea
title: Cerebral venous thrombosis and severe acute respiratory syndrome coronavirus‐2 infection: A systematic review and meta‐analysis
date: 2021-02-02
journal: Eur J Neurol
DOI: 10.1111/ene.14727
sha: a64bf5999511171978ce617d988bf937d99ae01d
doc_id: 1029535
cord_uid: pskjnb1q

BACKGROUND AND PURPOSE: Severe acute respiratory syndrome coronavirus‐2 (SARS‐CoV‐2) infection predisposes patients to arterial and venous thrombosis. This study aimed to systematically review the available evidence in the literature for cerebral venous thrombosis (CVT) in association with coronavirus disease‐2019 (COVID‐19). METHODS: We searched MEDLINE, Embase, and Cochrane Central Register of Controlled Trials databases to identify cases of COVID‐19–associated CVT. The search period spanned 1 January 2020 to 1 December 2020, and the review protocol (PROSPERO‐CRD42020214327) followed Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines. Identified studies were evaluated for bias using the Newcastle‐Ottawa scale. A proportion meta‐analysis was performed to estimate the frequency of CVT among hospitalized COVID‐19 patients. RESULTS: We identified 57 cases from 28 reports. Study quality was mostly classified as low. CVT symptoms developed after respiratory disease in 90%, and the mean interval was 13 days. CVT involved multiple sites in 67% of individuals, the deep venous system was affected in 37%, and parenchymal hemorrhage was found in 42%. Predisposing factors for CVT beyond SARS‐CoV‐2 infection were present in 31%. In‐hospital mortality was 40%. Using data from 34,331 patients, the estimated frequency of CVT among patients hospitalized for SARS‐CoV‐2 infection was 0.08% (95% confidence interval [CI]: 0.01–0.5). In an inpatient setting, CVT accounted for 4.2% of cerebrovascular disorders in individuals with COVID‐19 (cohort of 406 patients, 95% CI: 1.47–11.39). CONCLUSIONS: Cerebral venous thrombosis in the context of SARS‐CoV‐2 infection is a rare, although there seems to be an increased relative risk. High suspicion is necessary, because the diagnosis of this potentially life‐threatening condition in COVID‐19 patients can be challenging. Evidence is still scarce on the pathophysiology and potential prevention of COVID‐19–associated CVT.

The outbreak of the novel severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in December 2019 in Wuhan, China, has rapidly evolved into a pandemic. The clinical features of SARS-CoV-2 infection (coronavirus disease-19 ) and its prognosis are manifold. They range from asymptomatic infection to severe viral pneumonia with respiratory failure and high fatality rates [1] . Importantly, angiotensin converting enzyme-2, the primary receptor utilized by SARS-CoV-2 for cell entry, is not only expressed in the lungs but also in the central nervous system and vascular endothelial cells [2] . There is emerging evidence for neurological complications of SARS-CoV-2 infection [3, 4] . Both neuroinvasive disease and parainfectious complications [3] as well as an increased risk of stroke and thrombotic complications, have been described in patients with SARS-CoV-2 infection [5] [6] [7] [8] . Likewise, pulmonary embolism was identified as a major cause of sudden death in these patients [9] . Several factors have been hypothesized to contribute to this observation. They include immobility, reduced effectivity of thromboprophylaxis, prothrombotic events caused by cytokine storm, and either tropism of SARS-CoV-2 to endothelial cells or the ability of SARS-CoV-2 to damage endothelial cells [5, 6, 8, 10] .

Despite the attention given to cerebrovascular thrombotic events, few reports have addressed the risk of cerebral venous sinus thrombosis (CVT) in patients with SARS-CoV-2 infection. Given the higher risk of thrombosis among patients with SARS-CoV-2 infection, it is not unforeseen that the number of reports on CVT in the context of COVID-19 is increasing in the literature [11] [12] [13] . CVT is a rarer form of venous thrombosis, which predominantly affects younger individuals. As CVT is a potentially life-threatening cause of stroke that may be preventable in the context of COVID-19, it is important to describe the clinical, radiological, and paraclinical features, management, and prognosis of the condition. In this study, we aimed to summarize the current knowledge on the frequency and disease characteristics of SARS-CoV-2-related CVT on the basis of a systematic review of the literature and meta-analysis.

A systematic review was carried out to collect all cases reporting Reference lists and cited articles were also reviewed to increase the identification of relevant studies.

No limitations were imposed on study type: case reports, case series, observational and interventional studies, as well as randomized controlled trials were considered. We restricted studies to those published in English and excluded studies based on animal models and preclinical settings. Reviews, editorials, and letters were discarded unless they provided original data. Biases were assessed with the Newcastle-Ottawa scale, which included ratings of selection bias, assessment bias, comparability issues, causality, and reporting bias as previously performed [14, 15] .

The following data were extracted from studies identified via the search strategy outlined above: study design, number of patients Qualitative and quantitative synthesis was performed to assess occurrence rate of CVT among people hospitalized with SARS-CoV-2 infection. Moreover, we studied the spectrum of known risk factors for CVT in these patients. We also evaluated clinical and radiological features, management (including anticoagulation and anticonvulsive drugs), and short-term prognosis. Regarding the latter, we examined the clinical status as defined by each study, including the corresponding modified Rankin Scale score at the last available follow-up, as well as mortality after CVT. We reported lack of data whenever needed.

Summary statistics were calculated, and descriptive statistics were presented as mean and standard deviation for continuous variables, and counts and percentage for categorical variables.

We used t tests and χ 2 tests as appropriate. We performed statistical analyses pooling data extracted from selected studies.

To calculate the proportion of CVT in hospitalized patients with SARS-CoV-2 infection, we excluded case reports and case series, as no denominator was available, and pooled estimates only from large cohort studies including consecutive hospitalized patients with SARS-CoV-2 infection. Heterogeneity was assessed by means of Cochrane's Q test and I 2 statistics [16] . Meta-analysis of proportions was performed by using a DerSimonian-Laird random-effects model due to consistent differences in design and assessment across studies. Reported probability values were twosided, with significance set at p < 0.05. Visual inspection of funnel plots was used to assess reporting bias. Sensitivity analysis through a leave-one-out paradigm was planned. Statistical analysis was performed with R v.3.3.1, and the "meta" package for meta-analysis of proportions.

The systematic search yielded 90 articles, of which 62 underwent full-text assessment ( Figure 1 ). Thirty-four were excluded because they were commentaries, letters, or reviews (Supplementary Material, Table 1 for excluded studies and reason).

Overall, 28 reports of CVT in patients with SARS-CoV-2 infection were included in the qualitative synthesis (Table 1 ) .

Four were retrospective observational studies describing cerebrovascular diseases in a cohort of SARS-CoV-2-positive patients [19, 22, 38, 39] , two were prospective observational studies with consecutive enrolment [23, 36] , one was a retrospective multicenter case series [40] ; the remaining articles were retrospective case series or case reports. The largest study (n = 17,799) investigated all stroke subtypes, including CVT, ischemic, and hemorrhagic stroke, among patients hospitalized with SARS-CoV-2 infection across multiple hospitals in Europe, Asia, America, and

Bias assessment revealed low quality for almost all studies, with only two case series [28, 40] with moderate quality (Table S1 ). Quality issues were mainly related to selection and reporting bias, as most of the cases were single reports. Moreover, studies addressing CVT occurrence in larger samples often did not report specific methods of assessment, causality, outcome, or follow-up. Limitations in study quality were also related to uncertain exposure to SARS-CoV-2 infection in some studies, because diagnosis could be based on clinical and radiological features or serology in the presence of a negative F I G U R E 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart. 

nasopharyngeal swab polymerase chain reaction (PCR) test but high clinical suspicion (Table S2) [18, 20] .

Overall, 57 CVT cases were collected from 28 reports (Table 1) [ . Mean age was 53.5 years, with 11 cases presenting CVT at age <50 years and balanced gender distribution (50% female).

SARS-CoV-2 infection was ascertained through nasopharyngeal swab PCR in 50 out of 53 cases (92.1%; Table 1 ). Clinical and radiological criteria were used to establish diagnosis in three studies [18] [19] [20] , one of which used such criteria to diagnose COVID-19 even in light of a negative nasopharyngeal swab PCR test in 17% of the total cohort [19] . All patients received COVID-19-specific treatment, which reflected local standard operating procedures and included hydroxychloroquine, lopinavir, ritonavir, and antibiotics in suspected cases of superimposed bacterial pneumonia.

Six studies reported the occurrence of CVT in consecutively enrolled cohorts of patients with SARS-CoV-2 infection [19, 22, 23, 36, 38, 39] . Occurrence rates varied across the studies: 0.001% among all those diagnosed with COVID-19 in Singapore [36, 39] , 0.02% to 1% in multicenter cohorts of hospitalized patients with COVID-19 (n = 17,799) [22, 23, 38] , and 0.06% among hospitalized patients with SARS-CoV-2 infection referred for neurological assessment [19] .

In pooling data from studies reporting events in hospitalized SARS-CoV-2 patients (n = 34,331) [19, 22, 23, 38] , an estimated proportion of 0.08% of cases had CVT (95% confidence interval [CI]: 0.01-0.50, p heterogeneity = 0.007; Figure 2a ). Random-effects modeling was justified by heterogeneity attributable to large samples, few events, and interstudy bias ( Figure S1 , Table 3 ). This would translate into an estimate of approximately 0.8 cases per 1,000 hospitalized patients with SARS-CoV-2. In pooling data from studies reporting numbers of cerebrovascular events among hospitalized patients (n = 406), CVT was reported in 4.19% of those cases (95% CI: 1.47-11.39, p heterogeneity = 0.02; Figure 2b) , with leave-oneout sensitivity analyses yielding rates ranging from 2.8% to 5.7% (Table S3) .

In four cases, CVT diagnosis preceded by a few days the onset of COVID-19-related systemic symptoms, whereas in all other reports, CVT signs, symptoms, and diagnosis followed the onset of COVID-19 (36/40, 90%) ( Table 1 ). The interval between onset of COVID-19 respiratory F I G U R E 2 Forest plot for proportion estimates of patients having CVT among those hospitalized with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection (a) and among those hospitalized with SARS-CoV-2 infection and reported to have a cerebrovascular event (b). CI, confidence interval; CVT, cerebral venous thrombosis. symptoms and CVT manifestation had a wide variability, ranging from the very same day of COVID-19 onset to 47 days after COVID-19 started.

Risk factors for CVT were reported inconsistently. Overall, details of known risk factors for CVT independent of SARS-CoV-2 infection were reported in 11 cases (30.6%): five women [27, 28, 40] were taking oral contraceptives; one individual had a previous diagnosis of polycythemia vera for which aspirin was taken as primary prevention [17] , two individuals had solid tumors (one with breast cancer using hormone therapy and one with B-cell lymphoma [30, 31] ), one individual was a 3-year old child with concomitant tuberculous meningitis [37] and another individual had concomitant traumatic occipital skull fracture [36] . No patient had preexisting thrombophilia or a history of previous CVT or deep venous thrombosis (Table 1) .

All patients had neurological signs or symptoms due to CVT. Among reports with detailed clinical features, an isolated headache pattern was reported in only one case [31] , whereas all other cases presented with encephalopathy, focal signs, or seizures. Altered mental status was common (60.5%), whereas focal neurological signs varied according to CVT location and affected brain region, ranging from hemiparesis to aphasia. Epileptic seizures were reported in 10 cases (27.8%), and were focal (n = 3) [24, 27, 35] , generalized tonic-clonic (n = 2) [25, 33] or refractory status epilepticus (Table 2 ) [29] .

Computed tomography angiography was the most frequent imaging technique used for diagnostic assessment (30/43, 69.8%; Table 3 ). Only one patient underwent digital subtraction angiography, showing signs consistent with extensive hemispheric venous congestion [28] . Involvement of multiple venous vessels was more frequent than thrombosis of a single vessel (29 vs. 14/43). The transverse sinus was most frequently affected (65%), followed by the sigmoid sinus (47%), the superior sagittal sinus (44%), and the straight sinus (21%). The deep venous system was involved in 37% of cases, whereas thrombosis in cortical veins was detected in 21% of cases.

Hemorrhagic lesions were reported in 42% of cases (Table 3) .

Regarding coagulation, fibrinogen was abnormal in 54.5% of cases (mean fibrinogen = 490.8 ± 112.9 mg/dl), whereas D-dimer levels were above threshold in all but two cases [39, 44] (mean = 7812 ± 15,062 ng/ml) ( Table S4 ). Cerebrospinal fluid testing for SARS-CoV-2 by PCR was negative in all cases where it was available (n = 5/5) [21, 28, 29, 32, 33] . No data were available on opening pressure.

Nine patients were treated with anticonvulsive medication, in one instance as prophylactic treatment (Table 4 ) [17] . Anticoagulants were administered to 37 patients (95%); one pediatric patient was treated with antiplatelets [37] and one patient received endovascular treatment (mechanical thrombectomy and local thrombolysis) [28] . Follow-up imaging was reported in only four cases, at variable time points ranging from 1 to 4 weeks after admission, and showed partial [18, 44] or no recanalization [17, 26] .

In-hospital mortality was high, as 14/35 patients died (40%). One of them had recurrent contralateral CVT and associated hemorrhage (with persistent left transverse sinus thrombosis) after 2 weeks from the initial CVT [26] . Among them, six had nonhemorrhagic lesions and seven had parenchymal hemorrhage [20, 30, 39, 42] . Parenchymal hemorrhage tended to be more frequent in those not surviving CVT (60% vs. 40%, p = 0.1). Full or partial recovery was reported in 21 cases, nine of which had a full recovery at the last available follow-up [27, [33] [34] [35] [36] 39, 44] .

This systematic review including 57 patients disclosed that CVT in the context of SARS-CoV-2 infection is a rare but life-threatening complication, which was predominantly seen in patients with mild to moderate COVID-19 disease. In detail, we determined a frequency of 0.08% among patients hospitalized for SARS-CoV-2 infection. In addition, CVT represented 4.2% of all cerebrovascular events among patients hospitalized for COVID-19. These results support a potential higher occurrence rate of CVT in SARS-CoV-2 patients, given an expected rate of only 5 to 20 per million per year in the general population [12, 45] .

Conditions associated with CVT can be classified as predisposing (e.g., genetic prothrombotic diseases, antiphospholipid syndrome, cancer) or precipitant (oral contraceptives, infections, drugs with prothrombotic action) [46] . In 90% of cases of our cohort, neurological adequately sized control group [3, 47] .

Considering that the pandemic continues, it is necessary to raise awareness for CVT as a potential complication of SARS-CoV-2 infection. This should be particularly emphasized to curtail missed or delayed diagnosis of a potentially treatable condition such as CVT, which requires specific imaging workup [13] . The diagnosis is complicated by mild and subtle clinical presentations that may be seen as common residual symptoms of COVID-19 infection, with isolated headache pattern potentially underrecognized, as it is underreported in this review. Thus, a low threshold for diagnostic consideration for CVT and subsequent intracerebral vessel imaging (e.g., computed tomography venography) should be maintained in the acute and subacute phase of COVID-19 in case of headache, encephalopathy, mental status changes, focal neurological signs, or seizures [11] . Elevated D-dimer and fibrinogen levels can raise suspicion of CVT but are also commonly observed during the acute phase of systemic SARS-CoV-2 infection [49] . Computed tomography venography may be preferred over magnetic resonance imaging given the substantially shorter scan timing and broader availability, which is critical in times of limited hospital resources and risk for spread of infection to hospital personnel [50] .

We found a high rate of thrombosis of the cerebral deep venous system (37%) and involvement of multiple sinuses (67%). Despite the limitations due to the quality of the reports, the involvement of deep veins seems more frequent than usual, with the International Study on Cerebral Vein and Dural Sinus Thrombosis reporting rates of deep venous system involvement of as low as 11% [51] . In conclusion, our systematic review raises awareness for CVT in the context of SARS-CoV-2 infection. Prospective studies and analysis of registries are warranted to confirm our findings, to identify further peculiar features of CVT in people infected with SARS-CoV-2 and the characteristics of post-COVID-19 CVT, and to provide potential insights into the ascertainment and treatment of the underlying thrombophilic state.

The authors have no conflicts of interest to declare. Conceptualization (supporting); data curation (supporting); formal analysis (supporting); funding acquisition (supporting); investigation (supporting); methodology (supporting); project administration (supporting); resources (supporting); software (supporting); supervision (supporting); validation (supporting); visualization (supporting); writing-original draft (supporting); writing-review and editing (equal). Andrea Zini:

Conceptualization (lead); data curation (equal); formal analysis (equal); funding acquisition (equal); investigation (equal); methodology (equal); project administration (equal); resources (equal); software (equal); supervision (equal); validation (equal); visualization (equal); writing-original draft (lead); writing-review and editing (lead).

Data collected for this systematic review will be available upon request from the corresponding author (M.R.). 

Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study

Of mice and men: COVID-19 challenges translational neuroscience

A systematic review of neurological manifestations of SARS-CoV-2 infection: the devil is hidden in the details

The international European Academy of Neurology survey on neurological symptoms in patients with COVID-19 infection

Pulmonary embolism in patients with COVID-19

Neurologic manifestations of hospitalized patients with coronavirus disease

Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UKwide surveillance study

COVID-19 related stroke in young individuals

Registry of arterial and venous thromboembolic complications in patients with COVID-19

The need to manage the risk of thromboembolism in COVID-19 patients

European Stroke Organization guideline for the diagnosis and treatment of cerebral venous thrombosis -Endorsed by the European Academy of Neurology

Cerebral venous thrombosis

Diagnosis and management of cerebral venous thrombosis: a statement for healthcare professionals from the American Heart Association/ American Stroke Association

Mothership versus drip-and-ship model for mechanical thrombectomy in acute stroke: a systematic review and meta-analysis for clinical and radiological outcomes

Bridging versus direct mechanical thrombectomy in acute ischemic stroke: a subgroup pooled meta-analysis for time of intervention, eligibility, and study design

Quantifying heterogeneity in a meta-analysis

Thrombosis and bleeding as presentation of COVID-19 infection with polycythemia vera. A case report

Cerebral venous sinus thrombosis associated with COVID-19

Neurologic manifestations in 1760 COVID-19 patients admitted to Papa Giovanni XXIII Hospital

Cerebral venous thrombosis in patients with severe COVID-19 infection in intensive care

Cerebral deep venous thrombosis and COVID-19: case report

Cerebrovascular events and outcomes in hospitalized patients with COVID-19: The SVIN COVID-19 Multinational Registry

Risk of stroke in hospitalized SARS-CoV-2 infected patients: a multinational study

Extensive cerebral venous sinus thrombosis: a potential complication in a patient with COVID-19 disease

Cerebral venous thrombosis: a typical presentation of COVID-19 in the young

COVID-19-associated acute cerebral venous thrombosis: clinical, CT, MRI and EEG features

COVID-19 as triggering co-factor for cortical cerebral venous thrombosis?

Cerebral venous thrombosis associated with COVID-19

Atypical deep cerebral vein thrombosis with hemorrhagic venous infarction in a patient positive for COVID-19

Concomitant brain arterial and venous thrombosis in a COVID-19 patient

Cerebral venous thrombosis associated with COVID-19 infection: Causality or coincidence?

Venous cerebral thrombosis in COVID-19 patient

First case of Covid-19 presented with cerebral venous thrombosis: a rare and dreaded case

Cerebral venous sinus thrombosis as a presentation of COVID-19

Cerebral venous thrombosis and SARS-CoV-2 infection

Neurology of COVID-19 in Singapore

Child with tuberculous meningitis and COVID-19 coinfection complicated by extensive cerebral sinus venous thrombosis

Venous thromboembolism in non-critically ill patients with COVID-19 infection

Cerebral venous thrombosis in patients with COVID-19 infection: a case series and systematic review

Cerebral venous sinus thrombosis associated with SARS-CoV-2; a multinational case series

COVID-19 related cerebrovascular thromboembolic complications in three young patients

Arterial and venous strokes in the setting of COVID-19

High prevalence of deep venous thrombosis in non-severe COVID-19 patients hospitalized for a neurovascular disease

Cerebral venous thrombosis in COVID-19-associated coagulopathy: a case report

Trends in incidence and epidemiologic characteristics of cerebral venous thrombosis in the United States

Cerebral venous thrombosis: an update

Defining causality in COVID-19 and neurological disorders

Neurological associations of COVID-19

Coagulation parameters: an efficient measure for predicting the prognosis and clinical management of patients with COVID-19

The stroke mothership model survived during COVID-19 era: an observational single-center

Prognosis of cerebral vein and dural sinus thrombosis: results of the international study on cerebral vein and dural sinus thrombosis (ISCVT)

Endothelial cell infection and endotheliitis in COVID-19

National Institutes of Health N, COVID-19 Treatment Guidelines Panel N. COVID-19 Treatment Guidelines Panel. Coronavirus Disease 2019 (COVID-19) Treatment Guidelines