Clinical Profile and Outcome of Patients Presenting With Acute-on-Chronic Liver Failure: A Single-Center Experience

Background and aim: We aimed to study the etiologies and clinical profile and to describe the factors associated with mortality in acute-on-chronic liver failure (ACLF) patients at our center. Methods: Patients meeting the Asian Pacific Association for the Study of the Liver (APASL) definition of ACLF were included. We studied etiologies and clinical profile and analyzed the factors associated with mortality in patients with ACLF. We also analyzed the mortality rates based on the number of organ failures and the grade of ACLF. Results: 114 patients were included. Alcohol (82, 71.9%), drugs (22, 19.3%), and viral hepatitis (17, 14.9%) were the commonest precipitating factors of ACLF. The commonest cause of chronic disease was alcohol (83, 72.8%). Fifty-three (46.5%), 60 (52.6%), 44 (38.6%), 32 (28.1%), and 24 (21.1%) experienced renal, coagulation, cerebral, respiratory, and circulation failures, respectively. Overall, the in-hospital mortality rate stood at 54 (48.6%), with a median stay of eight days. Advanced hepatic encephalopathy and ventilator support independently predicted mortality. The Sequential Organ Failure Assessment (SOFA) score outperformed all other prognostic scores in predicting mortality in ACLF. Conclusion: Alcohol was the most common precipitating factor for ACLF. The in-hospital mortality rate was 48.6%. Advanced hepatic encephalopathy and ventilator support independently predicted mortality. The SOFA score is a more accurate predictor of mortality in ACLF when compared to other prognostic scores.


Introduction
Acute-on-chronic liver failure (ACLF) is a condition characterized by the sudden deterioration of liver function in patients with previously diagnosed or undiagnosed chronic liver disease.ACLF occurs when two liver insults happen at the same time [1].The mortality rate of patients with ACLF has been stable over the previous two decades, with a rate exceeding 50% [2].ACLF, as defined by the European Association for the Study of the Liver and the American Association for the Study of Liver Diseases (EASL-AASLD), refers to the sudden worsening of an existing chronic liver disease, typically caused by a specific triggering event, and is characterized by a higher risk of death within three months due to multiple organ failure [3].According to the Asian Pacific Association for the Study of the Liver (APASL), ACLF is defined as a sudden liver injury that occurs in a patient with pre-existing diagnosed or undiagnosed chronic liver disease.It is characterized by jaundice (bilirubin levels above 5 mg/dl) and coagulopathy (international normalized ratio (INR) >1.5) and is further complicated by the presence of ascites and/or encephalopathy within four weeks [4].The APASL definition includes both persons with cirrhosis and without cirrhosis, while the EASL-AASLD definition only applies to persons with cirrhosis.Recently, the World Health Organization proposed an alternative definition that merges the requirements of APASL and EASL-AASLD.This study aimed to investigate the causes, clinical spectrum, and factors associated with mortality in patients with ACLF at our facility.We also analyzed the mortality rates based on the number of organ failures and the grade of ACLF.

Materials And Methods
All patients with ACLF who met the definition provided by the APASL and were admitted to the Department of Gastroenterology at the Govind Ballabh Pant Institute of Post-Graduate Medical Education and Research in New Delhi between September 2018 and October 2023 were included in the study.We retrospectively gathered patient data from our database, which was originally maintained prospectively.In our analysis, we included patients who had non-hepatic acute insults and pre-existing decompensation, although they were excluded under the APASL-ACLF criterion.This decision was made because such criteria were followed in many other studies [5,6].As our investigation was conducted retrospectively, ethical approval was waived.Patients who fulfilled the APASL criteria and experienced either liver or non-liver insults were included, regardless of whether they had previously experienced decompensation.Patients who were pregnant and had disseminated malignancy, portal vein thrombosis, ambiguous diagnoses, and insufficient data were excluded from the study.
The presence of chronic liver disease was determined through a comprehensive assessment that included a thorough clinical history, physical examination, laboratory tests, and radiological findings such as abnormal liver texture, shrinkage of the liver, nodular liver outline, and signs of portal hypertension such as dilated portal vein, ascites, and splenomegaly [7].Additionally, the diagnosis was confirmed through the endoscopic visualization of varices or by examining the liver tissue obtained through a biopsy, which revealed characteristic changes associated with cirrhosis.Prognostic scores such as Sequential Organ Failure Assessment (SOFA) [8], Chronic Liver Failure Consortium Acute on Chronic Liver Failure (CLIF-C ACLF) [9], Acute Physiology and Chronic Health Evaluation (APACHE II) [10], Child-Turcotte-Pugh (CTP) [11], and Model for End-Stage Liver Disease Sodium (MELD-Na) [12] were computed.Each case was assessed for potential precipitating factors, such as persistent alcohol consumption (alcohol consumption within 28 days prior to the onset of jaundice), ingestion of hepatotoxic drugs, infections caused by hepatotropic viruses (hepatitis A, B, D, and E), recent major surgery, sepsis, or variceal bleeding.Assessment for autoimmune hepatitis and Wilson's disease was also performed if the etiology of the chronic liver disease could not be determined.When multiple acute triggering events were present, they were considered multiple acute insults.Organ failures and grading of ACLF were defined as per the EASL-CLIF Consortium.
Standard care therapy was followed for all patients.Patients diagnosed with hepatitis B virus (HBV) or hepatitis C virus (HCV) were prescribed antiviral medications.Patients diagnosed with Wilson's disease were initiated on penicillamine and zinc supplementation, whereas those diagnosed with autoimmune hepatitis were initiated on steroid treatment.Renal replacement therapy was administered as needed.Patients with grade 3 ascites underwent a large-volume paracentesis.Patients with hepatic encephalopathy were administered anti-coma medicines, such as rifaximin and lactulose.The usual protocols are used to treat hepatorenal syndrome and spontaneous bacterial peritonitis.Endoscopic variceal ligation (EVL) was performed as required.The decision to start and adjust the dosage of antibiotics, as well as the determination of the necessity for ventilator support, was made by the treating clinicians based on the clinical situation.None of the patients underwent liver transplantation.
Quantitative data are expressed as either the average (mean) or the middle value (median).The proportions were used to represent categorical data.A univariate analysis was conducted to compare survivors and nonsurvivors.An independent t-test, or Mann-Whitney U-test, was used to analyze continuous data.The chisquared test was used to analyze categorical variables.We used the Cox proportional hazards model to predict patient survival.A p-value less than 0.05 was used to assess statistical significance.

Results
In   The most frequent precipitating event was active alcoholism in 82 patients (71.9%), followed by druginduced liver injury (DILI) and hepatotropic viral infections (hepatitis A, B, and E) in 22 (19.3%) and 17 (14.9%)patients, respectively.Sepsis and variceal bleeding were the most common extrahepatic precipitating events observed in six (5.3%) and three (2.6%) patients, respectively.

Discussion
ACLF can be reversed if the acute inciting event has been resolved.The acute triggering events of ACLF may originate from the liver itself (hepatic) or from outside the liver (extrahepatic).Continuous alcohol use was the most frequent cause of ACLF in our study, accounting for 71.9% of the cases.This finding is consistent with previous Indian studies conducted by Pati et al. [13] and Shalimar et al. [6].In our investigation, bacterial infections and variceal bleeding constituted 5.3% and 2.6% of cases, respectively.These results closely align with the data observed in the western region.The CANONIC investigation, which involved 303 patients with ACLF, found that the most frequent causes of acute insults were bacterial infections (32.6%), alcoholism (24.5%), and gastrointestinal hemorrhage (13.2%).However, no acute insult was identified in 43.6% of the patients [14].Our investigation found that viral hepatitis was a contributing factor for 14  [6,13].The higher prevalence of viral hepatitis as a cause of acute insult for ACLF in India is likely attributed to the endemic presence of hepatitis A virus (HAV) and hepatitis E virus (HEV) in the country.In this study, we also identified a case of systemic lupus erythematosus presenting for the first time as severe liver injury leading to ACLF, which is quite rarely reported in previous literature.The most prevalent cause of cirrhosis in our analysis was alcohol consumption, which accounted for 72.8% of the cases.This finding is consistent with those of other investigations, such as the CANONIC study [14] and Duseja et al. [15].In addition, approximately half of the individuals (45%) with active alcoholism died in our study.This underscores the significant burden of alcoholic liver disease in our nation, emphasizing the need to address this escalating threat to public health.According to the CANONIC study, the presence of organ failure is necessary to define an ACLF.The CANONIC study [14] found that the mortality rates at 28 and 90 days in patients with ACLF were directly correlated with the severity of ACLF.Our study found that the inhospital mortality rate increased from 25% to 69% when the ACLF grade went from 1 to 3.This trend has also been observed in other studies conducted in India [6].In our investigation, the occurrence of liver failure was documented at a rate of 73.7%, whereas respiratory failure was reported at a rate of 28.1%.In comparison, the CANONIC trial reported liver failure rates of 43.6% and respiratory failure rates of 9.2%.The greater prevalence of organ failure in our study suggests that the included patients were in a more severe state of illness.The overall fatality rate in our study was 48.6%, with a median hospital stay of eight days.This rate is similar to that reported in the study conducted by Shalimar et al. [6].Non-survivors had significantly elevated levels of serum bilirubin and serum creatinine compared with survivors.The nonsurvivors exhibited better prognostic scores.Mortality was independently predicted using advanced HE and ventilator support.Additionally, we found that the SOFA score outperformed all other prognostic indicators in accurately predicting ACLF mortality.
Strengths of our study include a large sample size and comprehensive evaluation of demographic, clinical, laboratory, and prognostic factors of patients with ACLF.However, our study was constrained by its retrospective approach, which might cause potential selection bias, which in turn might influence the conclusions of our study.Also, we did not assess mortality rates at 28 days or in the long term.We exclusively assessed mortality in the hospital setting.Patients were diagnosed with ACLF based on the diagnosis provided by the APASL.However, for the purpose of this investigation, organ failure was characterized according to the criteria established in the CANONIC study.

Conclusions
Alcohol was the most common precipitating factor for ACLF (71.9%) and also the commonest cause of underlying chronic liver disease (72.8%) in our study.Total in-hospital mortality was 48.6%.Mortality proportionately increased with an increase in the number of organ failures (6.7% with no organ failure to 100% with six organ failures) and an increase in the grade of ACLF (25% with grade 1 ACLF to 69.1% with grade 3 ACLF).Advanced hepatic encephalopathy (HR: 2.762) and the need for a ventilator (HR: 2.225) independently predicted mortality in ACLF patients.The SOFA score (AUROC: 0.775) outperformed all other prognostic scores in predicting mortality in ACLF patients.

FIGURE 2 :
FIGURE 2: In-hospital mortality according to the number of organ failures (n=54) n: sample size; %: percentage

FIGURE 4 :
FIGURE 4: ROC curves for various prognostic indices: CTP, MELD-Na, CLIF-C ACLF, SOFA, and APACHE II scores APACHE II: Acute Physiology and Chronic Health Evaluation; CLIF-C ACLF: Chronic Liver Failure Consortium Acute on Chronic Liver Failure; CTP: Child-Turcotte-Pugh; MELD-Na: Model for End-Stage Liver Disease Sodium; ROC curve: receiver operating characteristic curve; SOFA: Sequential Organ Failure Assessment