1. IntroductionSevere acute respiratory syndrome (SARS-CoV-2) causing COVID-19 disease has emerged as a pandemic around the globe. Although comorbid conditions such as diabetes, hypertension, chronic lung diseases (particularly COPD), chronic kidney disease and cardiovascular disease have been reported as major risk factors for COVID-19 mortality, the prognostic value of chronic liver disease (in particular liver cirrhosis) is still undefined [1,2]. However, patients with cirrhosis following SARS-CoV-2 infection proved to be at a higher risk for poor outcomes. Registry data showed that overall mortality in patients with cirrhosis following COVID-19 infection ranged from 16% to 42% and poor outcome increased stepwise with the severity of cirrhosis [2,3,4,5]. Furthermore, patients with decompensated cirrhosis following SARS-CoV-2 infection were more susceptible to Intensive Care Unit (ICU) admission, renal replacement therapy and invasive mechanical ventilation compared to those without chronic liver disease (CLD) [6]. Acute-on-chronic liver failure (ACLF), which is a life-threatening syndrome, occurred in 12-50% of decompensating patients in COVID-19 infection [2,4,7]. Despite respiratory failure being the leading cause (71%) in COVID-19, liver-related manifestations (19%) are also common [2]. Nevertheless lung injury and hepatic dysfunction are tied by multiple links such as an altered gut–lung axis, gut dysbiosis, cirrhosis-associated immune dysfunction (CAID) and pulmonary complications due to ascites [8]. Primary vaccination is the most effective tool for preventing severe SARS-CoV-2 infection and is especially recommended for vulnerable individuals including patients with liver cirrhosis [9,10,11]. Notably, patients with cirrhosis were reported to display an accelerated decline in antibody titres compared to healthy individuals [12,13]. In Hungary since 15 January, 2021 five different COVID-19 vaccines have been commonly used: two mRNA-based vaccines (BNT162b2-Pfizer-BioNTech, mRNA-1273-Moderna), two vector vaccines (AZD1222-Astra Zeneca, Gam-COVID-Vac-Sputnik V) and one inactivated vaccine (HB02-Sinopharm) [14]. To date, there are limited data about the efficacy of different SARS-CoV-2 vaccine platforms (mRNA; viral vector; inactivated) in patients with cirrhosis.

Therefore, we aimed to investigate the impact of primary vaccination with different SARS-CoV-2 vaccines on patients with cirrhosis. Furthermore, the aim of our study was to evaluate the effectiveness of mRNA vaccines, as compared to other COVID-19 vaccines, in cirrhosis.

4. Discussion

Patients with liver cirrhosis are known to have a higher risk of in-hospital mortality and deterioration in cirrhosis severity following COVID-19 infection. Recently, multiple types of COVID-19 vaccines have been administered to the entire population including patients with chronic liver disease to prevent severe COVID-19 outcomes. We conducted a multicentre matched cohort study to investigate the efficacy of vaccines against SARS-CoV-2 in patients with liver cirrhosis following COVID-19 infection (GROUP B) compared to patients with liver cirrhosis without COVID-19 infection (GROUP C) and COVID-19 patients without cirrhosis (GROUP A).

In line with international data, our study demonstrated that patients in GROUP B were more frequently associated with adverse outcomes including oxygen support or mechanical ventilation compared to those in GROUP A (Figure 3) [24,25]. Previous studies showed that remdesivir was associated with liver injury in COVID-19 patients [26,27,28]. Moreover, Gao and colleagues reported that COVID-19 patients receiving corticosteroids had an increased risk of drug-induced liver injury (DILI) in contrast to non-user COVID-19 patients [29]. Therefore, in our dataset, patients receiving remdesivir or steroids were mostly patients in GROUP A (Figure 3). Consistently, patients in GROUP B received convalescent COVID-19 plasma (CCP) more commonly compared to those in GROUP A due to contraindications to start remdesivir in patients with five times the upper limit of transaminases. Higher administration rates of CCP in GROUP B might have resulted in improved outcomes, especially when the protein supplementation effect of this treatment is also considered (Figure 3) [30].A study of a COVID-19 cohort of 220,727 US patients reported that hospital mortality rates of COVID-19 patients with cirrhosis and those without cirrhosis were 8.9% and 3.9%, respectively [25]. Regarding cirrhosis cases, the death rates in our study were not significantly different in the two groups: 9.6% (GROUP B) and 11.8% (GROUP C), respectively. Patients with cirrhosis had similarly poor prognosis regardless of COVID-19 infection, which was in accordance with a North American multicentre matched cohort study [7].As for the etiology of cirrhosis, alcohol-related liver cirrhosis was found to be a leading cause and patients with alcohol-use disorder are notably more susceptible to hepatic decompensation following COVID-19 infection, in line with recently published studies [6,31,32].The Clinical Practice Guidelines of the European Association for the Study of the Liver (EASL) emphasized the significance of regular clinical screening to the prompt detection and treatment of the complications of cirrhosis [33]. Surveillance procedures were likely delayed during the pandemic, leading to disease progression and increased occurrence of decompensation events. In our findings, patients with liver cirrhosis who contracted COVID-19 infection more frequently developed hepatic decompensation events with worsening ascites, significantly more severe encephalopathy stages and higher rates of acute variceal haemorrhage. Although viral infections could trigger ACLF, we found lower rates in GROUP B compared to GROUP C considering ACLF is mostly associated with bacterial infections and poor hospital outcome (Figure 4) [34,35]. According to a large registry cohort of 745 COVID-19 patients with chronic liver disease, liver cirrhosis severity classified by Child–Pugh score is a reliable predictor of in-hospital mortality [2]. We also found that disease progression in liver cirrhosis was precipitated by COVID-19 infection (Figure 5). Although there was a stepwise increase in cirrhosis progression following COVID-19 infection, GROUP B did not seem to be associated with higher mortality rates grouped by different CTP cirrhosis stages, in contrast to GROUP C. In accordance with international data, our current results demonstrate that patients with cirrhosis, without regard to COVID-19 status, remained at higher risk of in-hospital mortality [36]. A North American matched cohort of 762 patients reported that COVID-19-related in-hospital mortality was lower in patients with cirrhosis after receiving one or two mRNA vaccines in comparison with unvaccinated patients with liver cirrhosis [10]. Moreover, the two-dose administration of mRNA vaccines was associated with the highest efficacy of preventing a severe clinical course of COVID-19 infection compared to other vaccines [37]. In contrast to international data, our study showed that vaccines failed to prevent COVID-19-related hospitalisations in adults [9,38]. Nevertheless, in our survival analysis, unvaccinated patients, in particular those with liver cirrhosis, had significantly worse chances of survival (Figure 6). Regardless of COVID-19 status and liver cirrhosis, primary vaccination with mRNA vaccines, compared to viral vector or inactivated vaccines, significantly improved the survival rates, in accordance with previously published studies [39,40]. Comparing the effectiveness of primary vaccination with the two mRNA vaccines, Pfizer-BioNTech was found to be more effective in preventing symptomatic COVID-19 infection requiring hospital admission, and significantly decreased the need for oxygen support and mechanical ventilation in patients with cirrhosis (Table 3). Furthermore, primary vaccination with Pfizer-BioNTech in patients with cirrhosis significantly protected against decompensating events associated with hepatic encephalopathy. However, primary vaccination with Moderna in patients with cirrhosis was significantly associated with a poor outcome following COVID-19 infection compared to those vaccinated with Pfizer-BioNTech (Table 3).Apart from mRNA vaccines, the Hungarian vaccination campaign included the AZD1222-AstraZeneca, Gam-COVID-Vac-Sputnik V and HB02-Sinopharm vaccines. In a nationwide, retrospective study investigating 3 740 066 Hungarian citizens, the overall estimated effectiveness of AstraZeneca, Sputnik-V and Sinopharm against COVID-19-related death was lower compared to Pfizer-BioNTech and Moderna [14]. Our study also found evidence that primary vaccination with viral vector and inactivated vaccines was less effective at preventing in-hospital mortality in COVID-19 patients, regardless of liver cirrhosis.

Altogether, our data provide evidence that COVID-19 infection is associated with the progression of liver cirrhosis and the development of acute hepatic decompensation events possibly due to aggravated immune dysfunction. Moreover, our findings demonstrate the effectiveness of primary vaccination with mRNA vaccines in patients with liver cirrhosis, predominantly in those receiving the Pfizer-BioNTech vaccine.

5. Limitations

The limitations of our data are mostly due to the retrospective study design and relatively small sample size. First, the number of enrolled patients with liver cirrhosis and those following COVID-19 infection were limited in our study. The higher mortality in GROUP C compared to GROUP B supports the notion that patients in GROUP C were admitted in a more vulnerable condition owing to advanced disease progression of cirrhosis. Second, the single-dose Ad26.COV2.S-Janssen vaccine was excluded from our investigation. As the cases were recorded over a prolonged time interval, a potential bias is expected as different cohorts were exposed to different contagion conditions. The study period overlapped with different waves of multiple variants of SARS-CoV-2, which may have influenced the effectiveness of the COVID-19 vaccines investigated in our analysis. Third, some vaccines were categorically indicated for elderly patients or patients with comorbidities, hence confounding patient characteristics could lead to perceived variation in the effectiveness of COVID-19 vaccines. Therefore, further prospective studies are needed to evaluate the effectiveness of all approved COVID-19 vaccines immunizing against new, upcoming variants in patients with cirrhosis.

On the other hand, the strengths of our study include its multicentre matched cohort design and its focus on a susceptible population group with a higher risk of COVID-19-related mortality. Although limited data are available about the impact of COVID-19 vaccination regimens on the clinical outcome of patients with liver cirrhosis, our methodology exclusively allows better interpretation to evaluate the impact of different COVID-19 vaccines on hospital outcomes in COVID-19 infection and liver cirrhosis.