Acute Hemorrhagic Encephalitis Related to COVID-19

Discussion

One of the biggest challenges facing physicians and health care workers addressing the COVID-19 pandemic is the wide variability in presenting features, which range from completely asymptomatic individuals to severe acute respiratory distress requiring mechanical ventilation, unconventional treatments, and extensive life-saving measures.³ Features attributable to CNS involvement by COVID-19 are rare but were described in an early population-based study from Wuhan.

We present a suspected case of acute hemorrhagic encephalitis in a previously healthy 48-year-old COVID-19–positive patient. Acute hemorrhagic encephalitis is believed to be a postinfectious process that is a severe variant of acute disseminated encephalomyelitis and has been described in other viral respiratory infections such as H1N1 Influenza. Another similar entity is acute necrotizing hemorrhagic encephalopathy, although lack of thalamic involvement in our case makes this less likely.^7,–,9

The first case of acute hemorrhagic encephalitis associated with COVID-19 was reported in Detroit, Michigan, in March 2020.⁶ In that case, thalamic involvement suggested that the underlying diagnosis was acute hemorrhagic necrotizing encephalopathy. The patients are demographically similar; however, the cases differ in location and extent of intracranial features (table). It is suspected that the large bilateral intraparenchymal hemorrhage with intraventricular extension was related to the presence of full anticoagulation. Heparin infusion was initiated because the patient's elevated d-dimer; at that time, anticoagulation was deemed the appropriate treatment.

Imaging was completed to determine the underlying etiology of the bilateral parietal-occipital intraparenchymal hemorrhage. An MRI and a CT angiogram (timed to optimize assessment of arterial and venous phases) demonstrated patency of arterial and venous structures. There was no evidence of vasculopathy, aneurysm (including peripheral mycotic lesions), or vascular malformation. The pattern of edema and hemorrhage was inconsistent with watershed infarct. Posterior reversible encephalopathy syndrome (PRES) was considered unlikely due to the extensive hemorrhage and progressive nature of the findings despite supportive care. It is noted that the patient was briefly treated with tocilizumab, for which there is 1 case report of associated PRES.¹⁰ No other precipitating factors for PRES were identified. Small vessel vasculitis was not suspected based on the available imaging. A digital subtraction angiogram would have definitively ruled out small vessel vasculitis; however, given the patient's diagnosis and unstable condition, this procedure could not be performed.

Having ruled out other differential diagnoses and with the absence of virus in the CSF, the presumed diagnosis of exclusion was one of acute hemorrhagic encephalitis. There are many proposed mechanisms by which SARS-CoV-2 virus can cause nervous system injury, including neural expression of angiotensin-converting enzyme 2 receptors, retrograde neuronal transport, vascular dissemination, systemic or focal hypoxic injury, or as a consequence of cytokine storm.^5,11,12 Acute hemorrhagic encephalitis may be related to an autoimmune process with an inflammatory response leading to perivascular demyelination and associated edema, hemorrhage, and necrosis.¹³ Support for this comes from the finding that CSF for viral PCR is often negative, as it was for this patient, which suggests that the process is not due to direct viral infiltration of brain parenchyma.^13,14 Similarly, the presence of an elevated IgG index with elevated CSF IgG ratio⁸ and a neutrophil predominant CSF leukocyte count⁷ is in keeping with an inflammatory process rather than a direct viral effect. However, an inflammatory process could only be confirmed with biopsy, which the patient did not undergo. Furthermore, it is also possible that the CSF neutrophilia was influenced by the presence of intraventricular hemorrhage. Although treatment strategies directed toward mitigating the inflammatory response have been used in acute hemorrhagic encephalitis, such as IV immunoglobulin, plasma exchange, or corticosteroids, the outcomes in the few reported adult cases suggest that prognosis is poor.¹⁴

The extensive hemorrhage was most likely related to the use of systemic anticoagulation. There is ongoing debate regarding the appropriate management of patients with COVID-19 who display features consistent with a prothrombotic state, with or without evidence of tissue thrombosis. The present case highlights the need for ongoing neurologic surveillance and the clear risk of neurologic complications in patients with COVID-19. This is particularly true when anticoagulation is deemed the appropriate treatment strategy and even more so when clinical assessments are limited because of the use of sedation and paralytics to assist in ventilation.

We present a case of a previously healthy 48-year-old woman with COVID-19 who developed acute hemorrhagic encephalitis, a complication which has been rarely described to this point. The underlying mechanism for developing such complication is believed to be autoimmune or paraviral in nature and warrants further investigation. At present, it appears that the initial neurologic presentation of COVID-19–related hemorrhagic encephalitis is altered level of consciousness, which may prompt further neurologic examination and imaging to exclude this feature. The observed intraparenchymal hemorrhage was most likely exacerbated by systemic anticoagulation and highlights the potential for neurologic complications in these severely ill patients with COVID-19. These findings may inform risk assessment and medical management in these patients, particularly as it pertains to the use of systemic anticoagulation and to the importance of ongoing neurologic assessment.