Acute Cerebrovascular Diseases
A stroke incidence of 1–2% in hospitalized COVID-19 patients has been established by large prospective studies and an observational stroke registry involving 17 healthcare networks across four countries.[34,43–47] In one registry, more than one-third of COVID-19 patients with acute ischemic stroke and over half with intracerebral hemorrhage died during their hospitalization, which was higher in comparison to the median in-hospital, nonstroke COVID-19 mortality rate of 12%.
Acute Ischemic Stroke
Initial reports from hospitalized COVID-19 patients in Wuhan, China found COVID-related stroke patients were more likely to be older, present with severe infection, and have cardiovascular risk factors, such as hypertension, diabetes, and a history of prior stroke. However, other data have reported the occurrence of stroke in patients younger than 50 years of age and without vascular risk factors.[46,47] Patients with COVID-19 and stroke have significantly higher rates of respiratory failure, thromboembolic events, and in-hospital mortality than contemporary non-COVID stroke patients and historical stroke controls.[48,49] Patients with COVID-19 and ischemic stroke were also less likely to receive thrombolytics, mechanical thrombectomy, or have a favorable discharge disposition.
There are several proposed etiologies for COVID-related ischemic stroke including hypercoagulable state, cardioembolism related to COVID-19 cardiomyopathy, and direct viral-induced endotheliopathy.[44,46,48–50] Despite these proposed mechanisms, the majority of studies report predominantly cryptogenic infarction and embolic strokes of unknown source (ESUS).[44,48] Some have theorized that the hypercoagulable state associated with COVID-19 may contribute to higher stroke risk. The COVID-19 related prothrombotic state has been associated with systemic endothelial cell dysfunction, abnormal flow dynamics, and platelet activation with thrombin generation and fibrin formation.[48,51] Elevated pro-inflammatory markers including D-dimer, fibrinogen and cytokines have been well documented in COVID-19 patients and, in some studies, correlate with a two-fold increased risk of thrombotic events and a higher odds of critical illness, acute kidney injury and ischemic stroke.[46,52–55]
Empiric therapeutic anticoagulation protocols for patients with elevated D-dimer levels had been suggested to reduce thromboembolic complications; however, on-going clinical trials have been paused for futility and safety concerns. Although prophylactic dosing of anticoagulation remains indicated in most hospitalized patients, therapeutic anticoagulation in the absence of a clear indication, such as a venous or arterial thromboembolic event, is not recommended at this time.[57–59]
Other causes of ischemic stroke, such as viral-induced CNS vasculitis, have been postulated. Postmortem histological analysis of COVID-19 patients has revealed lymphocytic endotheliitis within the endothelial cells of multiple organs, including lung, heart, kidney, small intestine, and liver.[20,57] Evidence of endothelial abnormalities have been noted in the brain, but co-localization of SARS-CoV-2 virions with this pathological injury has not been reported.[20,60]
The prevalence of radiographic confirmed primary intracerebral hemorrhage (ICH) during COVID-19 infection is less common than ischemic stroke, with reported prevalence rates between 0.2 and 0.4%.[36,47,61] COVID-related ICHs are more likely to be intraparenchymal with lobar location, have multiple foci, and be multicompartmental. Only rare accounts of nonaneurysmal subarachnoid hemorrhage have been reported. COVID-related ICH has been significantly associated with coagulopathy secondary to therapeutic anticoagulation.[61,62] 67–90% of ICH cases were fully anticoagulated at the time of diagnosis, most often with intravenous unfractionated heparin.[34,47,58]
Some of these hemorrhages may also be related to bleeding into the hypoxic-ischemic brain in the context of anticoagulation, microhemorrhages (possibly related to critical illness or hypoxia) and hyper-inflammation.[34,35] Indeed, pathologic studies of COVID-19 patients with multisystem organ failure have revealed diffuse petechial hemorrhages at the gray–white junction, as well as fresh hemorrhages across multiple brain territories including the frontal lobe, thalamus, diencephalon, pons, and corpus callosum. All patients in this series had evidence of diffuse intravascular thrombosis and half had disease consistent with endothelial inflammation. There was no immunohistochemistry done to determine if SARS-CoV-2 co-localizes with this injury and the underlying mechanism remains unclear, though hypoxic-ischemic injury, critical illness and hyperinflammation remain possible culprits. Another autopsy study of 18 brains using magnetic resonance (MR) microscopy with histopathological correlation found that brain hyperintensities are seen on MR microscopy correlated with thinning of the basal lamina, while hypointensities represented areas of microhemorrhage and fibrinogen leakage into the parenchyma. However, no evidence of SARS-CoV-2 viral invasion of the endothelium was found. Instead, perivascular T cells adjacent to the endothelial cells were postulated to have contributed to vascular injury.
Cerebral Sinus Thrombosis
A systematic review identified 14 reported cases of cerebral venous sinus thrombosis (CVT) within a median of 7 days from the onset of COVID-19 symptoms to diagnosis. Initial imaging of CVT-related stroke revealed the involvement of the transverse sinuses (75%), sigmoid sinuses (50%), or deep venous sinuses (33%) at presentation. With the exception of one patient, the majority were managed with therapeutic anticoagulation with a mortality rate of 45.5%. To date, the incidence of COVID-related CVT remains much lower than that of acute ischemic and hemorrhagic stroke.
Curr Opin Infect Dis. 2021;34(3):217-227. © 2021 Lippincott Williams & Wilkins