The impact of SARS-Cov2-induced acute respiratory syndrome on the liver is reinforced by an interplay between direct viral cytotoxicity on hepatocytes and cholangiocytes, ICU-associated reduced systemic and arterio-hepatic blood flow, medico-toxic side effects, and a severe general inflammatory reaction. The dysregulation of the splanchnic blood flow and oxygen supply is aggravated by an alteration of the hepatic vascular bed. Steatosis and unspecific mild necro-inflammatory infiltration are commonly reported findings [10, 11]. COVID-19 infection can primarily affect the liver parenchyma. Higher elevations in AST and ALT compared to γGT and AP were observed in COVID-19-associated hepatopathy, with only 12% of patients suffering from cholestasis [3, 4]. However, severe hepatocellular necrosis appears to be a secondary effect of disrupted hepatic blood supply due to thrombosis, infarct, or DIC, which may be directly enhanced by virus-induced vasculopathy, or indirectly in cases with a severe disease course.

Nevertheless, the biliary tract can also be damaged. COVID-19-associated SSC is increasingly described in the literature [5,6,7, 12,13,14]. Cholangiocytes and liver sinusoidal endothelial cells (LSECs) express the angiotensin-converting enzyme 2 (ACE2) receptor at a high density on their surface, which is a viral target [9, 15]. The biliary system can therefore become the focus of hepatopathy, and biliary diseases with clinico-serological similarities to primary sclerosing cholangitis (PSC) can occur.

In contrast to the majority of previous case reports, which describe only mild bile duct injury, both our patients have SSC on ERCP imaging, mirrored by extremely high levels of γGT and AP. Neither autoantibodies signalled an autoimmune liver disease nor could parameters for a virus-induced hepatitis be detected on broad serological follow-up analyses.

Our first patient demonstrated a pattern of biliary fibrosis associated with ischemic cholangiopathy of the large and peripheral bile ducts, as well as severe parenchymal bilirubinostasis and cholestasis. Aggregates of neutrophilic granulocytes in a large bile infarct could indicate a former source of biliary septicaemia. This case resembles the so-called ‘secondary sclerosing cholangitis in critically ill patients’ (SSC-CIP) in a progredient state following complex long-term treatment in the ICU.

The liver pathology in the second patient may have been affected by their complex medical history prior to COVID-19 infection. The liver explant showed a striking picture of a biliary cirrhosis with porto-portal septal fibrosis and a massive neoductular reaction, as well as the characteristic features of SSC on ERCP imaging. Bile clots and portal fibrosclerosis with loss of interlobular ducts were findings in this patient which characterise sclerosing cholangitis in critically ill patients (SSC-CIP).

Remarkably, vascular changes of small arteries, portal, and centrilobular veins, as well as adjacent sinusoids with obliterating thrombi in different stages of organisation, were observed in both patients, as reported by other authors [5,6,7, 12,13,14]. Direct endothelial damage and locally altered coagulation during the course of infection have to be considered cofactors for COVID-19-induced liver damage in general and particularly in the development of COVID-19-associated SSC. This consideration should be made in addition to the general impairment of the blood circulation which occurs during complicated ICU treatment. In the presented cases of late-stage liver disease following COVID-19 infection months prior, no morphological or serological signs of an underlying necro-inflammatory process were detected.

In conclusion, COVID-19-associated SSC is a rare, severe, and often fatal disease which is different to primary inflammatory bile duct disorders such as PSC and primary biliary cholangitis (PBC), but with analogies to SSC-CIP. Both presented cases were characterised by prolonged respiratory insufficiency treated by ECMO; both were complicated by septic shock or bacterial pneumonia, which contribute to insufficient oxygenation of the bile duct supporting blood flow; and the course of one case was aggravated by a pre-existing severe medical burden before COVID-19 infection. Additionally, a COVID-19-specific alteration of endothelial cells mediated by the ACE-receptor as an anchor protein for viral entrance, as well as local pro-coagulative conditions followed by vasculopathy and thrombosis, may enhance the liver damage in general and specifically the vulnerability of the bile duct epithelium.