How Does Underlying Cirrhosis Affect COVID-19?

A 51-year-old man presents to the hospital in Brazil, with shortness of breath, cough, runny nose, fever of 38.1ºC, and diarrhea -- all symptoms that started 3 days previously.

He is a smoker, has high blood pressure, and is an ex-alcoholic, he states (noting that his last drink was about 6 months earlier).

He was admitted, and physical examination finds him to be slightly lethargic, tachypneic, pale, and feverish, but with no evidence of jaundice or cyanosis.

Pulmonary auscultation detects fine crackles in the base of his lungs, but there is no evidence of respiratory distress, with 90% oxygen saturation. The patient's abdomen is distended but not tense, and there is collateral circulation and a small volume of ascites.

His medical history also includes admission to a tertiary hospital in September 2019 due to an increase in abdominal volume. Physical examination at that time noted evidence of chronic liver disease, ascites, and grade I encephalopathy.

He was diagnosed at that time with alcoholic liver cirrhosis, classified using the Child-Turcotte-Pugh (CTP) score, with 10 points in class C, and with a score of 11 on the Model for End-Stage Liver Disease Sodium (MELD-Na).

In the time since that diagnosis, the patient had regular follow-up hepatology appointments to manage his chronic liver disease and resulting decompensations.

During 2019, in an effort to manage the patient's ascites, clinicians started him on diuretic therapy with spironolactone and furosemide. Follow-up included upper digestive system endoscopies, elastic ligation, and beta-blocker therapy to control esophageal varices.

He then developed spontaneous bacterial peritonitis (SBP), which resolved with treatment, and he was started on prophylactic treatment with norfloxacin (Noroxin).

Laboratory test findings upon his admission to the hospital include the following:

• Anemia: Hb 9.2 g/dL (reference value [RV] >12 g/dL)

• Leukopenia: 2,900/mm³ (RV 4,000-12,000/mm³)

• Lymphopenia: 20.2% (RV 23-33%)

• Thrombocytopenia: 123,000/mm³ (RV 140,000-440,000/mm³)

• Albuminemia: 2.4 g/dL (RV 3.5-5 g/dL)

• Alanine aminotransferase (ALT): 71 U/L (RV 17-59 U/L)

• Aspartate aminotransferase (AST): 45 U/L (RV <50 U/L)

• High levels of alkaline phosphatase: 231 U/L (RV <126 U/L)

Prior to his admission, laboratory control tests reveal that the patient already has anemia (Hb 10.7 g/dL) and decreased serum albumin (3.0 g/dL), but the other tests are within normal limits. Clinicians perform paracentesis to analyze ascitic fluid and find it negative for SBP.

Two days after admission, the patient is tested for SARS-CoV-2 and has a CT scan of his chest, which shows an increased number of lymph nodes in the upper and lower paratracheal mediastinal chains as well as in the pre-vascular area.

Multiple ground-glass opacities are evident on the lung CT that are mainly perihilar, especially in the middle and upper third of the lung fields; there is also evidence of thickening of the inter- and intralobular septa, which forms a mosaic pavement pattern, and a slight laminar effusion on the right with an extension for the oblique fissure.

Clinicians diagnose the patient with COVID-19, and start him on auxiliary oxygen 5L/min nasal catheter, which improves his oxygen desaturation. With the diagnosis confirmed, he is started on dexamethasone 6 mg/day.

Three days after the patient's admission, clinicians suspect he has developed a secondary bacterial pulmonary infection. They therefore start him on cefepime and azithromycin and temporarily discontinue the SBP prophylaxis.

After 72 hours, his flu-like symptoms and diarrhea improve, but he remains feverish and short of breath. Suspecting a secondary nosocomial infection, clinicians increase his oxygen flow to 7 L/min.

The respiratory symptoms, however, worsen, and when a culture of tracheal aspirate confirms the presence of Klebsiella pneumoniae sensitive to carbapenems, clinicians switch the antibiotic treatment to meropenem (Merrem).

Dexamethasone is continued for 10 days; 2 days after the antibiotic change and with the aid of respiratory physiotherapy, the patient's clinical status begins to improve. Auxiliary oxygen is gradually withdrawn, and he is discharged from the hospital after 14 days, with all lab tests normal.

Discussion

Clinicians reporting their hospital's first case of SARS-CoV-2 pneumonia in a patient with previous liver cirrhosis stress that the presence of a previous comorbidity of course means there is increased risk of morbidity and mortality.

In this case, the case authors note, the patient's presentation with anemia, leukopenia with lymphopenia, and thrombocytopenia reflects findings of an analysis of COVID-related admission exams, which showed relatively high rates of lymphocytopenia (83.2%), thrombocytopenia (36.2%), and leukopenia (33.7%), which in those patients was likely due to chronic liver disease as well as the infection, the researchers noted.

Another evaluation of 540 patients with COVID-19 suggested that increasing laboratory serum markers of liver injury may reflect the severity of COVID-19, suggesting, the case authors note, that this correlation supports consideration of pre-existing liver disease as a severity factor, mainly due to the association with immunological injury.

The positive outcome in this patient, who was receiving oxygen through a nasal catheter plus dexamethasone 6 mg/day for infection, may be due to the treatment approach used in the RECOVERY study, in which dexamethasone was shown to reduce deaths by one-third in patients who received invasive mechanical ventilation and by one-fifth in patients who received auxiliary oxygen only.

As learned from previous viral pandemics, bacterial co-infections increase the risk of death -- accounting, for example, for up to 34% of all deaths in the 2009 H1N1 influenza pandemic. This is probably because virus-mediated immunosuppression of the hosts' innate immune response enables opportunistic bacteria to colonize the host, the case authors speculate.

They note that the incidence of liver damage in COVID-19 patients (related primarily to elevated levels of ALT and AST) has ranged in published studies from about 15% to 78%, although a direct causative link has yet to be clearly identified.

Reports are inconclusive about whether COVID-19 causes direct liver damage; this may occur when the virus invades target cells by binding to the angiotensin-converting enzyme 2 receptor, which is expressed in approximately 60% of cholangiocytes.

However, the case authors caution that despite observation of increased ALT and AST in severe COVID-19 cases, evidence of the link with liver function disorder remains inconclusive or conflicting, and likely remains little cause for concern.

Other possible explanations for the association are that COVID-19 may induce myositis, as has been observed in infections by other respiratory viruses, such as herpes, parvovirus, and adenovirus, the case authors note.

They explain that specific viral CD8+ T cells resulting from a viral infection outside the liver can independently trigger T cell-mediated hepatitis. Infection-related lymphopenia resulting in increased levels of interleukins 6, 10, and 2, and gamma interferon worsen inflammatory responses in various organs, including the liver.

Such an inflammatory insult in the body could explain the changes in liver biochemistry and the association between increasing severity of COVID-19 with higher levels of liver injury markers, the case authors write, adding that it is also possible that these hypotheses would interpret liver damage as collateral damage not directly due to COVID-19.

While epidemiological data on COVID-19 in patients with underlying liver disease remains sparse, there are two databases that have been collecting information -- SECURE-Cirrhosis and the COVID-HEP -- suggesting that of 508 such cases, 88% of the patients were hospitalized, 27% were admitted to the intensive care unit (ICU), 19% required mechanical ventilation, and 31% died.

In another international database study of 745 COVID-19 patients with chronic liver disease, 90% were hospitalized, 24% were admitted to the ICU, 18% received invasive mechanical ventilation, 4% started renal replacement therapy, and 20% died.

Mortality reported among patients with chronic liver disease with and without cirrhosis was 32% and 8%, respectively -- a cirrhosis-related increase of 24%. And not surprisingly, there was a directly proportional increase related to CTP scores: CTP-A 33 (19%), CTP-B 44 (35%), and CTP-C 46 (51%).

The risk factors significantly associated with mortality in univariable analysis were age (OR 1.03 per year, 95% CI 1.01-1.04, P<0.001); heart disease (OR 1.76, 95% CI 1.16-2.66, P=0.008); white ethnicity (OR 2.52, 95% CI 1.73-3.68, P<0.001); and baseline serum creatinine (OR 1.19/mg/dL, 95% CI 1.04-1.38 mg/dL, P=0.014); the risk associated with hepatitis B infection did not reach significance.

The presence of cirrhosis versus chronic liver disease without cirrhosis was significant in univariable analysis (OR 1.98, 95% CI 1.52-2.59). However, the predominant cause of death was COVID-19 lung injury, with only 19% of the patients with cirrhosis who died accounted for by liver-related complications.

Patients with liver cirrhosis, especially after decompensation, may be more susceptible to infection by SARS-CoV-2 due to systemic immune dysfunction, the case authors explained, pointing to preliminary data suggesting preexisting liver disease in 2-11% of COVID-19 patients.

In the setting of the COVID-19 pandemic, clinicians must strike a careful balance between protecting patients with cirrhosis from exposure to the virus and instituting treatment to prevent progression of liver disease and treating the cirrhosis complications, the authors state.

Conclusion

They conclude that in patients with cirrhotic immunosuppression, it is extremely important to pay close attention to the complications of coronavirus infection such as desaturation and the risk of secondary infection. If secondary bacterial infection is detected, as in this patient, broad-spectrum antibiotic therapy should be instituted. However, further studies in the population of COVID-19 patients with chronic liver disease are necessary for health teams to understand the challenges of this new disease, mainly in terms of pathophysiology and therapy. The results based on large virtual databases are providing useful epidemiologic information, the case authors note.

Comment