Prevalence of acute neurological complications and pathological neuroimaging findings in critically ill COVID-19 patients with and without VV-ECMO treatment

Abstract

Acute brain injuries such as intracerebral hemorrhage (ICH) and ischemic stroke have been reported in critically ill COVID-19 patients as well as in patients treated with veno-venous (VV)-ECMO independently of their COVID-19 status. The purpose of this study was to compare critically ill COVID-19 patients with and without VV-ECMO treatment with regard to acute neurological symptoms, pathological neuroimaging findings (PNIF) and long-term deficits. The single center study was conducted in critically ill COVID-19 patients between February 1, 2020 and June 30, 2021. Demographic, clinical and laboratory parameters were extracted from the hospital’s databases. Retrospective imaging modalities included head computed tomography (CT) and magnetic resonance imaging (MRI). Follow-up MRI and neurological examinations were performed on survivors > 6 months after the primary occurrence. Of the 440 patients, 67 patients received VV-ECMO treatment (15%). Sixty-four patients (24 with VV-ECMO) developed acute neurological symptoms (pathological levels of arousal/brain stem function/motor responses) during their ICU stay and underwent neuroimaging with brain CT as the primary modality. Critically ill COVID-19 patients who received VV-ECMO treatment had a significantly lower survival during their hospital stay compared to those without (p < 0.001). Among patients treated with VV-ECMO, 10% showed acute PNIF in one of the imaging modalities during their ICU stay (vs. 4% of patients in the overall COVID-19 ICU cohort). Furthermore, 9% showed primary or secondary ICH of any severity (vs. 3% overall), 6% exhibited severe ICH (vs. 1% overall) and 1.5% were found to have non-hemorrhagic cerebral infarctions (vs. < 1% overall). There was a weak, positive correlation between patients treated with VV-ECMO and the development of acute neurological symptoms. However, the association between the VV-ECMO treatment and acute PNIF was negligible. Two survivors (one with VV-ECMO-treatment/one without) showed innumerable microhemorrhages, predominantly involving the juxtacortical white matter. None of the survivors exhibited diffuse leukoencephalopathy. Every seventh COVID-19 patient developed acute neurological symptoms during their ICU stay, but only every twenty-fifth patient had PNIF which were mostly ICH. VV-ECMO was found to be a weak risk factor for neurological complications (resulting in a higher imaging rate), but not for PNIF. Although logistically complex, repeated neuroimaging should, thus, be considered in all critically ill COVID-19 patients since ICH may have an impact on the treatment decisions and outcomes.