Let’s begin by characterizing the global burden of viral hepatitis and viral hepatitis–related liver disease. It is important to realize that viral hepatitis is responsible for the majority—nearly 80%—of hepatocellular carcinoma cases,¹ and viral hepatitis claims more than 2 lives per minute by liver cancer or liver failure.

The World Health Organization has responded with ambitious targets to reduce viral hepatitis infections and deaths significantly by 2030,² but those targets were set before SARS-CoV-2 complicated the picture. As of December 2021, we have lost more than 5 million people worldwide to the COVID-19 pandemic and more than 800,000 in the United States alone.³ The number of COVID-19 cases is increasing, and the United States is clearly in “first place” for number of cases and deaths. This is not just that we do more testing and have the greatest transparency—it has to with interacting factors that influence rates of vaccination, social distancing, and mask wearing. But the issue under consideration here is how we can respond and adapt to pressures from both viral diseases and continue to make progress towards eliminating viral hepatitis across the globe.

How has the COVID-19 pandemic affected viral hepatitis elimination thus far? We will have that data someday—but it will take years to realize the full impact, since it takes years for the effects of chronic inflammation and fibrosis to take their toll. Furthermore, we do not know how long pandemic-related healthcare delays will last. In the meantime, Blach and colleagues⁴ have used modeling to predict what we might see in more than one half of the world’s countries if, for example, hepatitis C elimination programs around the world experienced a 1-year delay in delivery of services because of the COVID-19 pandemic.

Not surprising, they estimate that a 1-year delay in programming would result in many fewer new diagnoses and treatment starts, which would then affect downstream rates of viral transmission, chronic infections, and mortality by 2030.

The modeling performed here is conservative in that it assumed harm behaviors that increase viral hepatitis risk remained stable at prepandemic rates, but we know that in many populations, harm behaviors have increased under the stress of the pandemic. We look forward to more detail on hepatitis B and liver disease, in general, as modeling becomes more advanced. 

We looked at what the model projected for missed treatment starts and excess incident hepatitis C virus (HCV) cases globally, but how is this increased burden of disease distributed? What we see from their stratification by World Bank income level is that the highest proportion of missed treatments and excess HCV cases occurs in lower-income countries.⁴

Some low- and middle-income countries have ready access to inexpensive, generic antiviral medications for hepatitis C treatment. The availability of these inexpensive generic drugs has likely contributed to the marked increase in treatment starts seen in Georgia and Egypt in recent years, for example. But delays in diagnosis would dramatically decrease treatment starts in such countries, because treatment start rates have been high.

By contrast, some US states still restrict access to treatment for people who inject drugs, so the model does not predict as many missed treatment starts in the United States where treatment starts are relatively low.

Under this model, only high-income countries are projected to achieve the WHO target of reducing liver-related deaths by 65% by 2030. No countries are projected to reduce new hepatitis C infections by 90% by 2030.

Egypt and Georgia are unusual in low- and middle-income countries, but they do serve as a model for other countries in the world for how they could bring hepatitis C care forward. 

We have looked at how COVID-19–related delays may affect HCV diagnosis and treatment, but how about people who have CLD? How does CLD affect their risk of COVID-19? In this important study by Wang and colleagues,⁵ the authors analyzed nationwide electronic health record data in the United States to ascertain the odds of getting COVID-19 among patients with vs without CLD. The data here pertains to persons with a recent healthcare encounter for CLD, that is, within the past 1 year.

They found that the risk of COVID-19 was higher in patients with CLD, stratified by alcohol-induced liver disease, hepatitis B or C infection, and metabolic-associated fatty liver disease, with odds adjusted for age, sex, and race. In addition, Black patients were at increased risk of contracting COVID-19, which is consistent with other studies of COVID-19 risk and suggests factors such as access to healthcare and social adversity.

Here we see COVID-19 mortality among patients with CLD in the 2 racial groups predominant in this large, retrospective analysis.⁵ Black patients experienced significantly greater mortality from COVID-19 alone and CLD alone. But when looking at the smaller subgroup of patients who had both a recent encounter for CLD and COVID-19, the difference in mortality rates was not statistically significant.

Let’s take a deeper dive into CLD and COVID-19 risk and discuss COVID-19 outcomes by liver disease stage. CLD and cirrhosis are associated with immune dysregulation, and our patients with cirrhosis are really considered immune suppressed. There is a concern that this could worsen COVID-19 outcomes as well as liver disease outcomes.

This study by Marjot and colleagues⁶ took a retrospective look at individuals with COVID-19 and CLD, stratified by Child-Pugh cirrhosis stage, from 2 international registries. Major contributing countries included the United Kingdom, United States, China, Spain, Singapore, Egypt, Mexico, and Iran.

Data from these patients were compared with data from patients with COVID-19 but without CLD from a UK hospital network. As you can see, most of the liver disease in this cohort was induced by nonalcoholic fatty liver disease/metabolic associated fatty liver disease, and 25% was attributable to viral hepatitis. In this analysis, liver disease was not associated with hospitalization rate, but once patients with liver disease were hospitalized with COVID-19, there was a very large gradient between Child-Pugh A and Child-Pugh C scores for serious COVID-19 outcomes. In fact, mortality was 4 times greater among patients with cirrhosis than among those without cirrhosis.

This mirror bar chart demonstrates the 4-fold increase in mortality among the group with cirrhosis.6 Although the group without cirrhosis has the expected age-related gradient for COVID-19 mortality, the mortality in the group with cirrhosis was more evenly distributed among the age groups, consistent with the idea of immune dysregulation associated with cirrhosis. Besides cirrhosis, factors that were significantly associated with mortality by univariate analysis were age, cardiovascular disease, and baseline serum creatinine level.

This is very important Sankey diagram follows patients’ clinical courses through COVID-19. On the left is CLD without cirrhosis, showing a high survival rate (92%).⁶ On the right is the group with CLD with cirrhosis with a high death rate of 32%, including mostly patients who were ventilated and patients who were admitted to the intensive care unit. The overall survival rate in patients with cirrhosis has decreased to 68%.

In my experience, the 2 most common noninvasive tests for assessing liver fibrosis are the FIB-4 score and elastography, most commonly FibroScan. In this study, Li and colleagues⁷ compared several serum biomarkers/calculations for predictive value for mortality from COVID-19.

Increasing FIB-4 score (shown in the red box) was found to predict COVID-19 mortality. In fact, after adjusting for sex, BMI, hypertension, diabetes, and history of CLD, each 1-unit increment in FIB-4 predicted a 79% increased risk of death (odds ratio: 1.79; 95% CI: 1.36-2.35; P <.001). The mean FIB-4 in patients who lived was significantly lower than in patients who died (1.91 vs 3.98; P <.001). FIB-4 scores peaked then normalized in the group that survived COVID-19 but failed to normalize in the group that died. The other indices that many people use—aspartate aminotransferase (AST)–to–platelet ratio index and liver enzyme levels—did not appear to be associated with COVID-19 mortality.

What makes these results interesting is the suggestion that FIB-4 could be a noninvasive, simple risk assessment tool in patients with COVID-19, independent of underlying comorbidities including liver diseases.

We talked about COVID-19 outcomes in patients with diseased livers, but what about patients who are liver transplant recipients?

This retrospective analysis of liver transplantation and COVID-19 was published in 2021.⁸ It included patients from an international transplant registry and a control group with COVID-19 without liver transplant and with similar comorbidities.

The risk of intensive care unit admission and invasive ventilation was higher among liver transplant recipients, but liver transplant did not significantly increase the risk of death in patients with COVID-19.

Similarly, the US multicenter COLD consortium of 112 liver transplant recipients also showed that risk of death was not higher in these patients.⁹ Authors in both studies hypothesize that the immunosuppressive drugs used in liver transplant recipients may mitigate the excess inflammation associated with worse outcomes in COVID-19. It is also worth noting that the liver transplant centers that typically manage these liver transplant recipients are very sophisticated with access to the best and latest clinical tools. Regardless of the mechanism, these results are encouraging for our patients who have received liver transplants.

The data on the opposing sides of this mirror bar chart underscore that the 2 groups have similar case fatality rates from COVID-19.⁸ The variables independently associated with increased risk among liver transplant recipients were age, baseline serum creatinine, and nonliver cancers, with no association was seen with time since transplantation.

In this analysis from Spain,¹⁰ COVID-19 outcomes were assessed among liver transplant recipients who developed COVID-19, with a median follow-up interval of 23 days. The primary outcome was severe COVID-19 (defined as requiring mechanical ventilation, intensive care unit residence, and/or death). The mean age of the participants was approximately 65 years, 71% were male, and the median time since transplant was almost 9 years.

Not surprising, the authors found a 2-fold increase in rate of acquiring COVID-19 among liver transplant recipients, all but 2 of whom were receiving immunosuppressants. As we saw in other retrospective studies, receiving mycophenolate (especially at doses >1000 mg/day, which is quite typical for transplant patients) was an independent predictor of severe COVID-19. There was no association with receiving cyclosporine, tacrolimus, or everolimus and severe COVID-19, and no benefit was seen with complete withdrawal of baseline immunosuppression.

The authors proposed this modified approach to liver transplant care, in which the mycophenolate is either dose reduced, removed, or replaced with an alternative agent when a patient is at high risk of worse outcomes from COVID-19.

We know that COVID-19 is commonly associated with liver involvement that can range the clinical spectrum from mild, asymptomatic elevations in liver enzymes to hepatic decompensation. So let’s review the postulated mechanisms of SARS-CoV-2–induced liver injury.

We know that SARS-CoV-2 infects lung epithelial cells via the ACE2 receptors. This results in immune cell recruitment and activation that, in turn, leads to cytokine production and local inflammation. Mostly, COVID-19 causes mild aminotransferase elevations and hypoalbuminemia.¹¹

In some cases, however, the cytokines reach levels that result in widespread, systemic inflammatory effects. Worse clinical outcomes of COVID-19 occur when massive cytokine elevations occur early and do not decline during the disease course, triggering multiorgan damage, including liver decompensation and failure.

Systemic inflammation is thought to be the major mechanism of liver injury, although detection of viral RNA outside of lung tissue and the presence of ACE2 receptors in liver cells suggest that SARS-CoV-2 may exert some direct cytopathy.¹¹ Hypoxia and hepatic congestion due to the need for mechanical ventilation and vasopressor support may also cause or aggravate acute liver injury, with or without cholestasis, and drugs used to treat COVID-19 are also an important cause of drug-induced liver injury in patients hospitalized with COVID-19. 

How can we summarize COVID-19–related liver injury? First, it is important to note that we do not know the true incidence of elevated liver biomarkers because many patients with COVID-19 are asymptomatic. But we do know that hospitalized patients with COVID-19 without CLD often present with AST elevation, and AST elevation is more common in patients with severe disease vs those with mild COVID-19 (45% vs 15%, respectively).¹¹

Next, among patients with CLD, the highest COVID-19 risk, as we have seen before, is found among patients with metabolic associated fatty liver disease.

Finally, among patients hospitalized with COVID-19 who already have cirrhosis, the risk of decompensation is approximately 20%, and when these patients have a flare of their liver disease, they are at risk for acute-on-chronic liver failure.¹¹

As previously mentioned, liver injury in hospitalized patients may be as result of systemic inflammation or it can come from the COVID-19 treatments themselves. Mechanical ventilation and vasopressors, for example, alter hemodynamics and may result in acute liver injury with or without cholestasis. Drug-induced liver injury is potentially an important factor in COVID-19–related liver injury. The mechanisms of drug-induced liver injury could be direct drug toxicity or even hepatitis B flares when immune-suppressing agents are used to treat severe COVID-19.

The AASLD has published best practice advice for hepatologists during the COVID-19 pandemic.¹² The AASLD consensus statements suggest strategies for evaluating patients with COVID-19 and elevated liver enzymes, which ranges in incidence from 14% to 83%.

Recall that the upper limit of normal for liver enzymes levels is 25 IU/L for women and 35 IU/L for men. Most cases of COVID-19 with elevated liver enzymes involve AST and alanine aminotransferase (ALT) levels that are 1-2 times the upper limit of normal, with normal to modestly increased total bilirubin levels; these changes are typically transient and require only supportive care. Cases of COVID-19 with acute liver injury (AST or ALT >5 times the upper limit of normal) are rare in patients without CLD, and uncommon in children.

How does AASLD suggest we approach the COVID-19 patient with elevated liver enzymes? First, do a full workup, meaning consider other causes of elevated liver enzymes. In viral hepatitis circles, we call this “knowing your ABCs”—consider hepatitis A, B, and C infections. Second, take a full history of patient use of prescription and over-the-counter medications, herbal remedies, and supplements.

If liver tests are improving, continue to monitor; if they are worsening, look for other causes, such as muscle injury (particularly if AST > ALT). Consider a creatine kinase workup or aldolase test, look for cytokine release, obtain a C-reactive protein level and erythrocyte sedimentation rate, and so on.

If imaging is clinically indicated to rule out obstruction or thrombosis, ultrasound is best—not necessarily MRI or CT. In my clinic, every patient with persistently elevated liver enzymes would get an ultrasound, at a minimum. Liver biopsy may be warranted in certain situations, but we do not have a lot of evidence for its role in patients with COVID-19–induced liver injury.

The AASLD has also published guidance for management of CLD during the COVID-19 era.¹² We need to do as much as possible at the community level to avoid hospital exposures and overburdening hospitals: optimizing telemedicine and reimbursement for telemedicine must be emphasized. Screening for COVID-19 symptoms and taking temperatures prior to entering clinic locations are also important.

We need to continue treatment for patients’ underlying viral liver disease. If a patient with COVID-19 needs to start treatment for hepatitis B, it is okay to do so, especially if there is a clinical suspicion for hepatitis B virus flare or when initiating immunosuppressive therapy for COVID-19.

New treatment starts for hepatitis C infection or primary biliary cholangitis are not routinely warranted until recovery from COVID-19, so it can reasonably be delayed.

Surveillance for hepatocellular carcinoma is very important and should be continued as long as the patient does not have active COVID-19; we need to use community-based radiology services when possible to avoid patient exposures in large medical centers. Be sure to discuss risks vs benefits with patients so that they are aware of the importance of ongoing surveillance even in the context of the COVID-19 risk. Liver cancer treatment should go forward whenever possible to avoid delays that might result in progression to incurable disease or removal from the transplant waitlist.

Experts in hepatology from the Asia-Pacific and European regions have also put out expert guidance on managing patients with liver disease in the COVID-19 era; I will not mention those specifically here as there is good consensus among the expert societies and their recommendations are similar.

In the United States, the FDA has approved a third mRNA dose for certain individuals with immunocompromise.¹³ A third dose (at least 4 weeks after the second dose) of the Pfizer/BioNTech mRNA vaccine can be given to patients at least 12 years of age, and a third dose of the Moderna vaccine can be given to persons at least 18 years of age.

What constitutes immunocompromise in this setting? The definition of immunocompromised for these purposes includes active or recent treatment for solid tumor and hematologic malignancies; receipt of solid organ or recent hematopoietic stem cell transplants; severe primary immunodeficiency; advanced or untreated HIV infection; active treatment with high-dose corticosteroids, alkylating agents, antimetabolites, tumor necrosis factor blockers, and other biologic agents that are immunosuppressive or immunomodulatory; and chronic medical conditions such as asplenia and chronic renal disease that may be associated with varying degrees of immune deficit.

Keep in mind that vaccine responses will wane over time in our immunocompromised patients with liver disease, just as in the general population, so on the figure you can see the authorization for boost doses.¹⁴ 

What we know so far is that the vaccines are safe in liver transplant recipients,¹⁵ and a third mRNA dose may be considered because they may have a suboptimal vaccine response.^16,17

It is also worth mentioning the expert guidance from the AASLD¹⁸ on COVID-19 vaccination in patients with liver disease: Do not delay COVID-19 vaccines or non–COVID-19 vaccines and remember that they can be given concomitantly. Do not withhold immunosuppression in an attempt to increase vaccine efficacy. Proceed with vaccination in liver transplant candidates, finishing up the vaccine series by resuming at least 4 weeks after transplant, if interrupted.

To put this in context, we do have unanswered questions about COVID-19 vaccines in our patients. For example, what is the level of vaccine efficacy in solid organ transplant recipients? How does the duration of immunity in immunocompromised patients compare with immunocompetent hosts, for both B-cell–derived and T-cell–derived immunity? Can we safely continue but modify immunosuppression to enhance vaccine responsiveness in liver transplant recipients? Is there an optimal mixing-and-matching strategy for these vaccines, and are there special considerations for timing and dosing in patients who have previously had COVID-19? As more people get vaccinated, we will look for data on the frequency of elevated liver enzymes, acute cellular rejection, or autoimmunity after vaccination.