Prediction and Risk Factors for Prognosis of Cirrhotic Patients with Hepatic Encephalopathy

Hepatic encephalopathy post-TIPS: Current status and prospects in predictive assessment

Transjugular intrahepatic portosystemic shunt (TIPS) is an essential procedure for the treatment of portal hypertension but can result in hepatic encephalopathy (HE), a serious complication that worsens patient outcomes. Investigating predictors of HE after TIPS is essential to improve prognosis. This review analyzes risk factors and compares predictive models, weighing traditional scores such as Child-Pugh, Model for End-Stage Liver Disease (MELD), and albumin-bilirubin (ALBI) against emerging artificial intelligence (AI) techniques. While traditional scores provide initial insights into HE risk, they have limitations in dealing with clinical complexity. Advances in machine learning (ML), particularly when integrated with imaging and clinical data, offer refined assessments. These innovations suggest the potential for AI to significantly improve the prediction of post-TIPS HE. The study provides clinicians with a comprehensive overview of current prediction methods, while advocating for the integration of AI to increase the accuracy of post-TIPS HE assessments. By harnessing the power of AI, clinicians can better manage the risks associated with TIPS and tailor interventions to individual patient needs. Future research should therefore prioritize the development of advanced AI frameworks that can assimilate diverse data streams to support clinical decision-making. The goal is not only to more accurately predict HE, but also to improve overall patient care and quality of life.

Optic nerve sheath diameters predict mortality and severity in hepatic encephalopathy

OBJECTIVE

The aim of this study was to compare the predictive power of optic nerve sheath diameter (ONSD) measured by brain computed tomography (CT) in patients diagnosed with hepatic encephalopathy (HE) in the emergency department, with other factors for mortality and disease severity.

MATERIALS AND METHODS

A total of 217 patients aged 18 years and older with acute decompensation of cirrhosis diagnosed with HE in the emergency department were included in the study. To compare with patients diagnosed with HE, a total of 217 individuals were included in the study as the healthy control group. ONSD measurements were performed on both the HE patients and the healthy control group in the brain CT.

RESULTS

The mortality rate of HE patients was 32.7%. Regarding the severity of the disease, 53% of the patients had late-stage HE. The presence of acute-on-chronic liver failure was detected in 51.4% of patients. The mortality rate among acute-on-chronic liver failure patients was 56.6%. According to the study data, ONSD, creatinine, lactate, and procalcitonin were independent predictors of mortality. Meanwhile, Child-Pugh score, direct bilirubin, ONSD, ammonia, and total bilirubin were independent predictors of disease severity. In the receiver operating characteristic curve analysis, the ONSD had the highest predictive power for mortality and disease severity among the determined predictive values.

CONCLUSION

The data from the study suggests that assessing the ONSD through brain CT scans in individuals diagnosed with HE in the emergency department may provide valuable insights for clinicians, aiding in the prediction of both mortality rates and the severity of the disease.

Construction of a prediction model for hepatic encephalopathy in acute-on-chronic liver failure patients

OBJECTIVE

Hepatic encephalopathy (HE) is a serious complication of acute-on-chronic liver failure (ACLF) that requires early detection and intervention to positively impact patient prognosis. This study aimed to develop a reliable model to predict HE in ACLF patients during hospitalization.

METHODS

Retrospectively recruiting 255 hepatitis B-related ACLF patients, including 67 who developed HE during hospitalization, the study analysed clinical data and biochemical indices collected during the first week of admission. The least absolute shrinkage and selection operator (LASSO) was used to identify characteristic predictors for hospitalization HE events, and a logistic regression model was subsequently developed. Receiver operating characteristic (ROC) curves, calibration curves, and bootstrap resampling were used to evaluate the model's discrimination, consistency, and accuracy, and a nomogram was created to visualize the model. An external validation cohort of 236 liver failure patients collected from the same medical centre between 2007 and 2010 was used to validate the model.

RESULTS

The study found that blood urea nitrogen (BUN), alpha-fetoprotein (AFP), international normalized ratio (INR), serum ammonia, and infection complications during hospitalization were risk factors for HE in ACLF patients. The new model predicted the development of HE in ACLF patients with an area under the receiver operating characteristic curve (AUROC) of 85.2%, which was superior to other models. The best threshold for the new model was 0.28, resulting in a specificity of 81.4% and a sensitivity of 80.6%. In the validation group, the new model showed similar results, with an AUROC of 79% and a specificity of 83.6% and a sensitivity of 56.6%.

CONCLUSION

This study developed and validated a new prediction model for HE in ACLF patients offering a useful tool for early identification of patients with a high risk of HE in clinical settings. However, to ascertain the model's general effectiveness, future prospective multicentre studies are warranted.

Recurrence of portosystemic encephalopathy in cirrhotic patients and its risk factors

Objectives

Recurrent episodes of Portal Systemic Encephalopathy (PSE), poses a significant burden of illness on the patients and healthcare system. The objective of this study was to assess the recurrence of PSE in cirrhotic patients after index episode of PSE and to identify various risk factors associated with it.

Methods

A retrospective, single-centre study was conducted at Aga Khan University Hospital over a span of one year. Patients who were admitted first time with PSE and admitted within three months of index PSE were enrolled in the study. Variables assessed were demographic data, associated comorbid conditions, aetiology of cirrhosis, Child-Turcotte-Pugh (CTP) score, Model of End-Stage Liver Disease (MELD) score, PSE grade, laboratory tests, ascites with spontaneous bacterial peritonitis (SBP), variceal bleeding. Statistical analysis was done and variables of those who developed recurrence were compared with those who did not.

Results

Fifty one patients were recruited. Thirty three (64.7%) were readmitted with PSE. On comparative analysis of both groups; infection, Meld score, low albumin, and raised total bilirubin showed significant P-value (<0.05).

Conclusion

Identification of risk factors during assessment can reduce the recurrence of PSE. We would recommend to validate result of our study on a large scale prospectively.

General anesthesia in patients with hepatic encephalopathy and acute variceal bleeding undergoing endoscopic treatment: A retrospective study

The management of cirrhotic patient with encephalopathy and acute variceal bleeding (AVB) remains a clinical challenge with a high mortality. Early endoscopic therapies are frequently applied in patients with AVB. However, the application of general anesthesia in endoscopic surgery for patients with hepatic encephalopathy (HE) is pretty challenging. The present study aimed to evaluate the possible effect of general anesthesia with tracheal intubation on patient complicated with encephalopathy and AVB during endoscopic procedure. Thirty-six cirrhotic patients with encephalopathy and AVB were retrospectively studied, 14 patients underwent endoscopic treatment under general anesthesia with tracheal intubation, and 22 patients received pharmacological treatment, or and endoscopic therapy without general anesthesia served as the control group. Routine clinical and laboratory data were collected. The total mortality rate was 13.9% (5/36), 2 (14.3%) in the anesthesia group, 3 (13.6%) in the control group. The child-Pugh class of death cases were all grade C, 3 (60%) of them were in the stage IV of HE. The stage of HE was all improved in the 2 groups, 12 (85.8 %) patients were totally recovered from consciousness disturbance in the anesthesia group and 16 (72.7 %) in the control group respectively, the difference between the 2 groups was not significant (P>0.05). Except the death cases, there were still 3 patients in the control group had impaired consciousness at discharge. Child-Pugh score, Child-Pugh class and the stage of HE in the anesthesia group were significantly improved at discharge compared with those before operation. General anesthesia does not aggravate the severity of encephalopathy, and endoscopic treatment under general anesthesia with tracheal intubation is effective for HE patients complicated with AVB.

Myocardial injury is a risk factor for 6-week mortality in liver cirrhosis associated esophagogastric variceal bleeding

This study sought to investigate risk factors for 6-week mortality of patients with decompensated liver cirrhosis associated esophagogastric variceal bleeding (EGVB) and clinical characteristics of myocardial injury in cirrhotic patients with EGVB. This retrospective cohort study included 249 patients with decompensated liver cirrhosis associated EGVB in the Department of Emergency. Patients were divided into two groups including liver cirrhosis associated EGVB without myocardial injury and liver cirrhosis associated EGVB with myocardial injury. Myocardial injury, recurrent bleeding, total bilirubin (TBIL) level and dyslipidemia are independent risk factors for 6-week mortality in liver cirrhosis associated EGVB. Among all patients with liver cirrhosis associated EGVB, 90 (36.2%) had myocardial injury and 159 individuals (63.8%) not. The 6-week mortality in the group with myocardial injury was 21%, which was significantly higher than that of 7% in the group without myocardial injury. More patients in the myocardial injury group smoked, had moderate to severe esophageal varices, liver failure, and Child-Pugh C liver function compared to the non-myocardial injury group. Myocardial injury, recurrent bleeding, TBIL level and dyslipidemia are independent risk factors for death within 6 weeks in liver cirrhosis associated EGVB. The 6-week mortality is considerably higher in patients with myocardial injury in liver cirrhosis associated EGVB than those without myocardial injury.

Using machine learning methods to predict 28-day mortality in patients with hepatic encephalopathy

BACKGROUND

Hepatic encephalopathy (HE) is associated with marked increases in morbidity and mortality for cirrhosis patients. This study aimed to develop and validate machine learning (ML) models to predict 28-day mortality for patients with HE.

METHODS

A retrospective cohort study was conducted in the Medical Information Mart for Intensive Care (MIMIC)-IV database. Patients from MIMIC-IV were randomized into training and validation cohorts in a ratio of 7:3. Training cohort was used for establishing the model while validation cohort was used for validation. The outcome was defined as 28-day mortality. Predictors were identified by recursive feature elimination (RFE) within 24 h of intensive care unit (ICU) admission. The area under the curve (AUC) and calibration curve were used to determine the predictive performance of different ML models.

RESULTS

In the MIMIC-IV database, 601 patients were eventually diagnosed with HE. Of these, 112 (18.64%) experienced death within 28 days. Acute physiology score III (APSIII), sepsis related organ failure assessment (SOFA), international normalized ratio (INR), total bilirubin (TBIL), albumin, blood urea nitrogen (BUN), acute kidney injury (AKI) and mechanical ventilation were identified as independent risk factors. Validation set indicated that the artificial neural network (NNET) model had the highest AUC of 0.837 (95% CI:0.774-0.901). Furthermore, in the calibration curve, the NNET model was also well-calibrated (P = 0.323), which means that it can better predict the 28-day mortality in HE patients. Additionally, the performance of the NNET is superior to existing scores, including Model for End-Stage Liver Disease (MELD) and Model for End-Stage Liver Disease-Sodium (MELD-Na).

CONCLUSIONS

In this study, the NNET model demonstrated better discrimination in predicting 28-day mortality as compared to other models. This developed model could potentially improve the early detection of HE with high mortality, subsequently improving clinical outcomes in these patients with HE, but further external prospective validation is still required.

A novel nomogram to predict 90-day mortality in patients with hepatitis B virus-related acute-on-chronic liver failure: a single-center retrospective study

BACKGROUND

Acute-on-chronic liver failure (ACLF) is a critical illness with high mortality. Herein, we developed and validated a new and simple prognostic nomogram to predict 90-day mortality in hepatitis B virus-related ACLF (HBV-ACLF) patients.

METHODS

This single-center retrospective study collected data from 181 HBV-ACLF patients treated between June 2018 and March 2020. The correlation between clinical data and 90-day mortality in patients with HBV-ACLF was assessed using univariate and multivariate logistic regression analyses.

RESULTS

Multivariate logistic regression analysis showed that age (p = 0.011), hepatic encephalopathy (p = 0.001), total bilirubin (p = 0.007), international normalized ratio (p = 0.006), and high-density lipoprotein cholesterol (p = 0.011) were independent predictors of 90-day mortality in HBV-ACLF patients. A nomogram was created to predict 90-day mortality using these risk factors. The C-index for the prognostic nomogram was calculated as 0.866, and confirmed to be 0.854 via bootstrapping verification. The area under the curve was 0.870 in the external validation cohort. The predictive value of the nomogram was similar to that of the Chinese Group on the Study of Severe Hepatitis B score, and exceeded the performance of other prognostic scores.

CONCLUSION

The prognostic nomogram constructed using the factors identified in multivariate regression analysis might serve as a beneficial tool to predict 90-day mortality in HBV-ACLF patients.

Validation of prognostic scores predicting mortality in acute liver decompensation or acute-on-chronic liver failure: A Thailand multicenter study

BACKGROUND & OBJECTIVES

Cirrhosis patients with worsening of the liver function are defined as acute decompensation (AD) and those who develop extrahepatic organ failure are defined as acute-on-chronic liver failure (ACLF). Both AD and ACLF have an extremely poor prognosis. However, information regarding prognostic predictors is still lacking in Asian populations. We aimed to identify prognostic factors for 30-day and 90-day mortality in cirrhosis patients who develop AD with or without ACLF.

METHODS

We included 9 tertiary hospitals from Thailand in a retrospective observational study enrolling hospitalized cirrhosis patients with AD. ACLF was diagnosed according to the EASL-CLIF criteria, which defined as AD patients who have kidney failure or a combination of at least two non-kidney organ failure. Outcomes were clinical parameters and prognostic scores associated with mortality evaluated at 30 days and 90 days.

RESULTS

Between 2015 and 2020, 602 patients (301 for each group) were included. The 30-day and 90-day mortality rates of ACLF vs. AD were 57.48% vs. 25.50% (p<0.001) and 67.44% vs. 32.78% (p<0.001), respectively. For ACLF patients, logistic regression analysis adjusted for demographic data, and clinical information showed that increasing creatinine was a predictor for 30-day mortality (p = 0.038), while the CLIF-C OF score predicted both 30-day (p = 0.018) and 90-day (p = 0.037) mortalities, achieving the best discriminatory power with AUROCs of 0.705 and 0.709, respectively. For AD patients, none of the parameters was found to be significantly associated with 30-day mortality, while bacterial infection, CLIF-AD score and Child-Turcotte-Pugh score were independent parameters associated with 90-day mortality, with p values of 0.041, 0.024 and 0.024. However, their predictive performance became nonsignificant after adjustment by multivariate regression analysis.

CONCLUSIONS

Regarding Thai patients, the CLIF-C OF score was the best predictor for 30-day and 90-day mortalities in ACLF patients, while appropriate prognostic factors for AD patients remained inconclusive.