Secondary Logo

Journal Logo

Brief Report

Disseminated Multifocal Intracerebral Bleeding Events in Three Coronavirus Disease 2019 Patients on Extracorporeal Membrane Oxygenation As Rescue Therapy

Bihlmaier, Karl MD, PhD1; Coras, Roland MD2; Willam, Carsten MD1; Grampp, Steffen MD1; Jabari, Samir MD2; Eichhorn, Philip MD3; Haller, Florian MD3; Kuramatsu, Joji MD4; Schwab, Stefan MD4; Castellanos, Ixchel MD5; Birkholz, Torsten MD5; Schüttler, Jürgen MD5; Altmeppen, Jürgen MD6; Schiffer, Mario MD1; Herbst, Larissa MD1

Author Information
doi: 10.1097/CCE.0000000000000218


At the University Hospital Erlangen, 33 patients with severe acute respiratory distress syndrome (ARDS) due to coronavirus disease 2019 (COVID-19) were treated on different ICU wards, of which six received venovenous extracorporeal membrane oxygenation (ECMO) as rescue therapy. Three venovenous ECMO patients (medium age 56.6 yr) developed sudden anisocoria and absent pupillary reflexes and died shortly after. Cranial CT scans showed fatal severe multifocal intracerebral bleedings in two of the three patients localized in the white matter of both the cerebrum and cerebellum (Fig. 1A). This extensive multifocality, partly exceeding 50 simultaneous bleeding sites, has not been described to date. Even though severe multifocal intracerebral parenchymal hemorrhage of the gray matter is occasionally found in patients undergoing ECMO therapy (1,2), the bleeding pattern of the patients presented here appeared unusual and therefore suggestive for a potential association with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection or endothelialitis. A literature review on cerebral hemorrhage in COVID-19 patients revealed recent retrospective analyses and case reports describing occasional bleeding events (3–7). Here, parenchymal hemorrhage is found merely sporadic but diffuse petechial and in different lobes of the cerebrum or cerebellum. Hemorrhage etiology has been discussed but remains elusive (8,9).

Figure 1.:
Radiological and histopathological findings. A, Cranial CT scan illustrating a simultaneous disseminated multifocal intracerebral bleeding pattern in all lobes of the cerebrum and cerebellum. B, Representative macroscopic slide in coronal frontal plane shows brain congestion with small bleeds (black arrow) as well as larger subcortical parenchymal hemorrhages (red arrow). Note flattening of the brain surface assign for edema. C, Histological examination confirms multiple petechial hemorrhages confined to the subcortical white matter (WM) (black arrows). D, Representative coronal slide in a patient with multiple WM congestion/microhemorrhages (black arrows). Histological examination shows edema confined to the WM (E) with congestion (F) as well as small hemorrhages (G). Dotted line in (C): Gray matter/WM junction. NCX = neocortex/cortical ribbon.

A medical history of arterial hypertension and diabetes was known in two patients; one of them used acetylsalicylic acid as antiplatelet therapy. None of the three had a history of vasculopathy or cerebral comorbidities. All patients presented signs of a hyperinflammatory state with elevated C-reactive protein levels (peak values 310–430 mg/L) and high as well as continuously rising ferritin levels (2,200–7,100 ng/mL). Nota bene raising ferritin levels are a negative predictor for the course of the COVID-19. Both patients presenting fatal hemorrhage did not show acute kidney injury as potential risk factors for cerebrovascular events.

Autopsy in two patients with fatal brain hemorrhage confirmed generalized brain edema with signs of transtentorial herniation of medial parts of the temporal lobe and transforaminal herniation of the cerebellar tonsils (Fig. 1B). Both cases showed secondary Duret hemorrhages of the pons and multiple supratentorial and infratentorial subcortical fresh hemorrhages (Fig. 1C). These were predominantly confined to the subcortical white matter. The prominent brain congestion and the multiple fresh hemorrhagic lesions could be confirmed on histological level at subcortical site with prominent edema of the adjacent cortical ribbon areas. Autopsy in the third patient showed regular macroscopic findings without signs for brain edema or herniation. Interestingly, on macroscopic coronal slides and in microscopy areas with congested subcortical vessels and perivascular hemorrhagic extravasates were found (Fig. 1D). Histology revealed white matter edema, most likely of vasogenic etiology with prominent congestion as well as multifocal microhemorrhages (Fig. 1E–G).

There were no signs of an inflammatory process in any of the three cases. All three patients showed cerebral microangiopathy (small vessel disease) in microscopy of various degrees with hyaline mural thickening of small arteries/arterioles but also affected venules and capillaries. We suspect that the third patient represents the same pathomechanism caught at an earlier stage before macroscopic hemorrhages became evident.

Due to the atypical bleeding patterns and since brain infections of COVID-19 patients are frequently described (9–11), we hypothesized that the multifocal intracerebral hemorrhages could be attributed to a COVID-19–associated vascular pathology, for example, a cerebral endothelialitis. However, after extensive histological analysis including immunohistochemical stainings, we could not confirm any signs of leukocyte infiltrates or any other inflammatory processes, neither of the meninges, the brain parenchyma, nor of the brain vasculature.

  • 1) Since other tissues investigated (e.g., kidney) could also not confirm signs of vascular inflammation, we cannot explain simultaneous multifocal intracerebral bleedings by vascular infiltration and inflammation. Correspondingly, in contrast to positive lung tissue, polymerase chain reaction analyses of multiple cerebral tissue samples, as well as liquor samples, remained negative for SARS-CoV-2 virus RNA. However, the small vessel angiopathy could have potentially contributed to an increased vascular vulnerability and intracerebral blood leakage, in particular, under conditions of increased venous pressure due to venovenous ECMO blood flow and systemic anticoagulation. Venovenous ECMO systems at our hospital are regularly cannulated femoro-femoral in a classic double cannulation configuration. Nevertheless, whether microangiopathy and positive hemosiderin deposits can be attributed to previous SARS-CoV-2 infection remains elusive.
  • 2) Two of three patients received IV immunoglobulins due to hypogammaglobulinemia. The administration of immunoglobulins is clinically associated with hemorrhage, like retroperitoneal bleedings. Since dosage was low in both patients and brain hemorrhage has previously not been reported in the context of immunoglobulin administration, an association of immunoglobulins with simultaneous multifocal intracerebral bleedings is rather unlikely.
  • 3) Thromboembolic events such as deep vein thrombosis or pulmonary embolism are frequently observed in SARS-CoV-2 infection, which led to the recommendation of systemic anticoagulation. In addition to the hemorrhages, we found circumscribed thrombotic cerebrovascular occlusions in our histological examinations.

Since coagulation disorders are associated with intracerebral hemorrhages in ECMO patients, these microinfarctions could have presumably led to secondary disseminated multifocal intracerebral bleedings. Previously, cerebral microhemorrhages were reported in ECMO patients (12–15), which interestingly have been comparable to cerebral injuries seen in high altitude brain edema (16).

At our center, approximately 100 ECMO therapies are carried out per year, with 10–15% general cumulative frequency of intracranial hemorrhages. In the subgroup of six patients suffering from COVID-19–associated ARDS undergoing ECMO support, three presented the novel hemorrhage pattern described.

Anticoagulation targets at our center are platelets above 50,000/µL, quick greater than 50%, antithrombin greater than 70%, and activated partial thromboplastin time 60–80 seconds. Data on target monitoring are shown in Supplemental Figure 1 ( We hypothesize that multifocal intracerebral coagulation events in COVID-19 patients can eventually contribute to the reported disseminated intracerebral bleeding events, and this potential side effect has to be considered under ECMO therapy and anticoagulation strategies in COVID-19 patients.


1. Witherick J, Redler K, Wakerley BR. Extracorporeal membrane oxygenation-related brain haemorrhages. Pract Neurol. 2016; 16:160–161
2. Fletcher-Sandersjöö A, Thelin EP, Bartek J Jr, et al. Incidence, outcome, and predictors of intracranial hemorrhage in adult patients on extracorporeal membrane oxygenation: A systematic and narrative review. Front Neurol. 2018; 9:548
3. Radmanesh A, Derman A, Lui YW, et al. COVID-19-associated diffuse leukoencephalopathy and microhemorrhages. Radiology. 2020. May 21. [online ahead of print]
4. Mao L, Jin H, Wang M, et al. Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China. JAMA Neurol. 2020; 77:1–9
5. Giorgianni A, Vinacci G, Agosti E, et al. Neuroradiological features in COVID-19 patients: First evidence in a complex scenario. J Neuroradiol. 2020. May 15. [online ahead of print]
6. Li J, Long X, Zhu C, et al. A case of COVID-19 pneumonia with cerebral hemorrhage. Thromb Res. 2020; 193:22–24
7. von Weyhern CH, Kaufmann I, Neff F, et al. Early evidence of pronounced brain involvement in fatal COVID-19 outcomes. Lancet. 2020; 395:e109
8. Md Noh MSF. COVID-19 and cerebral hemorrhage: Proposed mechanisms. J Neuroradiol. 2020. Jun 1. [online ahead of print]
9. Aghagoli G, Gallo Marin B, Katchur NJ, et al. Neurological involvement in COVID-19 and potential mechanisms: A review. Neurocrit Care. 2020; 13:1–10
10. Li YC, Bai WZ, Hashikawa T, et al. The neuroinvasive potential of SARS-CoV2 may play a role in the respiratory failure of COVID-19 patients. J Med Virol. 2020; 92:552–555
11. Baig AM. Neurological manifestations in COVID-19 caused by SARS-CoV-2. CNS Neurosci Ther. 2020; 26:499–501
12. Xu Z, Shi L, Wang Y, et al. Pathological findings of COVID-19 associated with acute respiratory distress syndrome. Lancet Respir Med. 2020; 8:420–422
13. Liebeskind DS, Sanossian N, Sapo ML, et al. Cerebral microbleeds after use of extracorporeal membrane oxygenation in children. J Neuroimaging. 2013; 23:75–78
14. Luyt CE, Bréchot N, Demondion P, et al. Brain injury during venovenous extracorporeal membrane oxygenation. Intensive Care Med. 2016; 42:897–907
15. Arachchillage DRJ, Passariello M, Laffan M, et al. Intracranial hemorrhage and early mortality in patients receiving extracorporeal membrane oxygenation for severe respiratory failure. Semin Thromb Hemost. 2018; 44:276–286
16. Riech S, Kallenberg K, Moerer O, et al. The pattern of brain microhemorrhages after severe lung failure resembles the one seen in high-altitude cerebral edema. Crit Care Med. 2015; 43:e386–e389

acute respiratory distress syndrome; cerebral microhemorrhage; coronavirus disease 2019; extracorporeal membrane oxygenation; intracerebral bleeding

Supplemental Digital Content

Copyright © 2020 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of the Society of Critical Care Medicine.