A model to predict 3-month mortality risk of acute-on-chronic hepatitis B liver failure using artificial neural network

Sarcopenia is associated with short- and long-term mortality in patients with acute-on-chronic liver failure

BACKGROUND

While sarcopenia is recognized as a predictor of mortality in cirrhosis, its influence on acute-on-chronic liver failure (ACLF) remains uncertain. Despite multiple studies examining the impact of sarcopenia on short-term mortality in patients with ACLF, the sample size of these studies was limited, and their outcomes were inconsistent. Therefore, this study aimed to explore the impact of sarcopenia on both short- and long-term mortality in patients with ACLF.

METHODS

This retrospective cohort study included 414 patients with ACLF that were treated between January 2016 and September 2022. Sarcopenia was diagnosed based on the measurement of the skeletal muscle index at the third lumbar vertebra (L3-SMI). Subsequently, the patients were divided into sarcopenia and non-sarcopenia groups. We analysed the basic clinical data of the two groups. Multivariate Cox proportional analysis was used to analyse short-term (28 days) and long-term (1 year and overall) mortality rates.

RESULTS

A total of 414 patients were included, with a mean age of 52.88 ± 13.41 years. Among them, 318 (76.8%) were male, and 239 (57.7%) had sarcopenia. A total of 280 (67.6%) patients died during the study period. Among them, 153 patients died within 28 days (37%) and 209 patients died within 1 year (50.5%). We found that the 28-day, 1-year and overall mortality rates in the sarcopenia group were significantly higher than those in the non-sarcopenia group (37% vs. 22.3%, P < 0.01; 50.5% vs. 34.9%, P < 0.01; and 67.6% vs. 53.1%, P < 0.01, respectively). Multivariate Cox regression analysis revealed that sarcopenia was significantly associated with increased mortality. The hazard ratios for sarcopenia were 2.05 (95% confidence interval [CI] 1.41-3.00, P < 0.01) for 28-day mortality, 1.81 (95% CI 1.29-2.54, P < 0.01) for 1-year mortality and 1.82 (95% CI 1.30-2.55, P < 0.01) for overall mortality. In addition, muscle density and international normalized ratio were associated with short- and long-term mortality.

CONCLUSIONS

Sarcopenia is associated with both short- and long-term mortality in patients with ACLF. Therefore, regular monitoring for sarcopenia is important for these patients.

Facial nerve palsy following parotid gland surgery: A machine learning prediction outcome approach

Introduction

Machine learning (ML)-based facial nerve injury (FNI) forecasting grounded on multicentric data has not been released up to now. Three distinct ML models, random forest (RF), K-nearest neighbor, and artificial neural network (ANN), for the prediction of FNI were evaluated in this mode.

Methods

A retrospective, longitudinal, multicentric study was performed, including patients who went through parotid gland surgery for benign tumors at three different university hospitals.

Results

Seven hundred and thirty-six patients were included. The most compelling aspects related to risk escalation of FNI were as follows: (1) location, in the mid-portion of the gland, near to or above the main trunk of the facial nerve and at the top part, over the frontal or the orbital branch of the facial nerve; (2) tumor volume in the anteroposterior axis; (3) the necessity to simultaneously dissect more than one level; and (4) the requirement of an extended resection compared to a lesser extended resection. By contrast, in accordance with the ML analysis, the size of the tumor (>3 cm), as well as gender and age did not result in a determining favor in relation to the risk of FNI.

Discussion

The findings of this research conclude that ML models such as RF and ANN may serve evidence-based predictions from multicentric data regarding the risk of FNI.

Conclusion

Along with the advent of ML technology, an improvement of the information regarding the potential risks of FNI associated with patients before each procedure may be achieved with the implementation of clinical, radiological, histological, and/or cytological data.

Implications of anaemia and response to anaemia treatment on outcomes in patients with cirrhosis

Background & Aims

Anaemia is frequently observed in patients with cirrhosis and was identified as a predictor of adverse outcomes, such as increased mortality and occurrence of acute-on-chronic liver failure. To date, the possible effects of iron supplementation on these adverse outcomes are not well described. We therefore aimed to assess the role of iron supplementation in patients with cirrhosis and its capability to improve prognosis.

Methods

Laboratory diagnostics were performed in consecutive outpatients with cirrhosis admitted between July 2018 and December 2019 to the University Hospital Essen. Associations with transplant-free survival were assessed in regression models.

Results

A total of 317 outpatients with cirrhosis were included, of whom 61 received a liver transplant (n = 19) or died (n = 42). In multivariate Cox regression analysis, male sex (hazard ratio [HR] = 3.33, 95% CI [1.59, 6.99], p = 0.001), model for end-stage liver disease score (HR = 1.19, 95% CI [1.11, 1.27], p <0.001) and the increase of haemoglobin levels within 6 months (ΔHb6) (HR = 0.72, 95% CI [0.63, 0.83], p <0.001) were associated with transplant-free survival. Regarding the prediction of haemoglobin increase, intake of rifaximin (beta = 0.50, SD beta = 0.19, p = 0.007) and iron supplementation (beta = 0.79, SD beta = 0.26, p = 0.003) were significant predictors in multivariate analysis.

Conclusions

An increase of haemoglobin levels is associated with improvement of transplant-free survival in patients with cirrhosis. Because the prediction of haemoglobin increase significantly depends on rifaximin and iron supplementation, application of these two medications can have an important impact on the outcome of these patients.

Impact and implications

Anaemia is very common in patients with cirrhosis and is known to be a predictor of negative outcomes, but little is known about the effect of iron substitution in these individuals. In our cohort, increase of haemoglobin levels improved transplant-free survival of patients with cirrhosis. The increase of haemoglobin levels was mainly induced by iron supplementation and was even stronger in the case of concomitant use of iron and rifaximin.

Clinical trial registration

UME-ID-10042.

Acute on chronic liver failure: prognostic models and artificial intelligence applications

Critically ill patients presenting with acute on chronic liver failure (ACLF) represent a particularly vulnerable population due to various considerations surrounding the syndrome definition, lack of robust prospective evaluation of outcomes, and allocation of resources such as organs for transplantation. Ninety-day mortality related to ACLF is high and patients who do leave the hospital are frequently readmitted. Artificial intelligence (AI), which encompasses various classical and modern machine learning techniques, natural language processing, and other methods of predictive, prognostic, probabilistic, and simulation modeling, has emerged as an effective tool in various areas of healthcare. These methods are now being leveraged to potentially minimize physician and provider cognitive load and impact both short-term and long-term patient outcomes. However, the enthusiasm is tempered by ethical considerations and a current lack of proven benefits. In addition to prognostic applications, AI models can likely help improve the understanding of various mechanisms of morbidity and mortality in ACLF. Their overall impact on patient-centered outcomes and countless other aspects of patient care remains unclear. In this review, we discuss various AI approaches being utilized in healthcare and discuss the recent and expected future impact of AI on patients with ACLF through prognostic modeling and AI-based approaches.

A Prognostic Nomogram with High Accuracy Based on 2D-SWE in Patients with Acute-on-chronic Liver Failure

BACKGROUND AND AIMS

Acute-on-chronic liver failure (ACLF) is associated with very high mortality. Accurate prediction of prognosis is critical in navigating optimal treatment decisions to improve patient survival. This study was aimed to develop a new nomogram integrating two-dimensional shear wave elastography (2D-SWE) values with other independent prognostic factors to improve the precision of predicting ACLF patient outcomes.

METHODS

A total of 449 consecutive patients with ACLF were recruited and randomly allocated to a training cohort (n=315) or a test cohort (n=134). 2D-SWE values, conventional ultrasound features, laboratory tests, and other clinical characteristics were included in univariate and multivariate analysis. Factors with prognostic value were then used to construct a novel prognostic nomogram. Receiver operating curves (ROCs) were generated to evaluate and compare the performance of the novel and published models including the Model for End-Stage Liver Disease (MELD), MELD combined with sodium (MELD-Na), and Jin's model. The model was validated in a prospective cohort (n=102).

RESULTS

A ACLF prognostic nomogram was developed with independent prognostic factors, including 2D-SWE, age, total bilirubin (TB), neutrophils (Neu), and the international normalized ratio (INR). The area under the ROC curve (AUC) was 0.849 for the new model in the training cohort and 0.861 in the prospective validation cohort, which were significantly greater than those for MELD (0.758), MELD-Na (0.750), and Jin's model (0.777, all p <0.05). Calibration curve analysis revealed good agreement between the predicted and observed probabilities. The new nomogram had superior overall net benefit and clinical utility.

CONCLUSIONS

We established and validated a 2D-SWE-based noninvasive nomogram to predict the prognosis of ACLF patients that was more accurate than other prognostic models.

Reversibility of acute-on-chronic liver failure syndrome in hepatitis B virus-infected patients with and without prior decompensation

Acute-on-chronic liver failure (ACLF) is a severe clinical syndrome associated with high short-term mortality and reversibility. This study aimed to compare the characteristics of survival and reversibility in hepatitis B virus (HBV)-related ACLF (HBV-ACLF) patients with and without previous decompensation. Overall, 1044 patients who fulfilled the acute hepatic insult criteria of the APASL-ACLF Research Consortium (AARC) definition were enrolled from a prospectively established cohort of HBV-related liver failure patients. These patients were divided into the AARC ACLF group and the non-AARC ACLF group according to prior decompensation. Mortality, reversibility of ACLF syndrome, and predicted factors associated with reversibility were evaluated. Liver transplantation-free mortality of the AARC ACLF group was significantly lower than that of the non-AARC ACLF group (28 days: 28.2% vs. 40.3%, p = .012; 90 days: 41.7% vs. 65.4%, p &lt; .001). The 5-year cumulative reversal rates of ACLF syndrome were 88.0% (374/425) and 66.0% (31/47) in the AARC and non-AARC ACLF groups, respectively, (p = .039). Following reversibility of ACLF syndrome, 340/374 (90.9%) and 21/31 (67.7%) patients in the AARC and non-AARC ACLF groups, respectively, maintained a stable status within 5 years. Although prior decompensation indicated poor reversibility of ACLF syndrome, HBV-infected patients with prior decompensation who fulfilled the acute hepatic insult criteria of the AARC definition showed favourable reversibility and maintained a stable status after receiving nucleoside analogues. The AARC ACLF definition identified HBV-ACLF as a distinct syndrome with good reversibility. HBV-infected patients with prior decompensation could be included in the AARC ACLF management.

Influence of ambient air pollution on successful pregnancy with frozen embryo transfer: A machine learning prediction model

Numerous air pollutants have been reported to influence the outcomes of in vitro fertilization (IVF). However, whether air pollution affects implantation in frozen embryo transfer (FET) process is under debate. We aimed to find the association between ambient air pollution and implantation potential of FET and test the value of adding air pollution data to a random forest model (RFM) predicting intrauterine pregnancy. Using a retrospective study of a 4-year single-center design，we analyzed 3698 cycles of women living in Shanghai who underwent FET between 2015 and 2018. To estimate patients' individual exposure to air pollution, we computed averages of daily concentrations of six air pollutants including PM_2.5, PM₁₀, SO₂, CO, NO₂, and O₃ measured at 9 monitoring stations in Shanghai for the exposure period (one month before FET). Moreover, A predictive model of 15 variables was established using RFM. Air pollutants levels of patients with or without intrauterine pregnancy were compared. Our results indicated that for exposure periods before FET, NO₂ were negatively associated with intrauterine pregnancy (OR: 0.906, CI: 0.816-0.989). AUROC increased from 0.712 to 0.771 as air pollutants features were added. Overall, our findings demonstrate that exposure to NO₂ before transfer has an adverse effect on clinical pregnancy. The performance to predict intrauterine pregnancy will improve with the use of air pollution data in RFM.

Application of artificial intelligence in non-alcoholic fatty liver disease and viral hepatitis

Non-alcoholic fatty liver disease (NAFLD) and chronic viral hepatitis are among the most significant causes of liver-related mortality worldwide. It is critical to develop reliable methods of predicting progression to fibrosis, cirrhosis, and decompensated liver disease. Current screening methods such as biopsy and transient elastography are limited by invasiveness and observer variation in analysis of data. Artificial intelligence (AI) provides a unique opportunity to more accurately diagnose NAFLD and viral hepatitis, and to identify patients at high risk for disease progression. We conducted a literature review of existing evidence for AI in NAFLD and viral hepatitis. Thirteen articles on AI in NAFLD and 14 on viral hepatitis were included in our analysis. We found that machine learning algorithms were comparable in accuracy to current methods for diagnosis and fibrosis prediction (MELD-Na score, liver biopsy, FIB-4 score, and biomarkers). They also reliably predicted hepatitis C treatment failure and hepatic encephalopathy, for which there are currently no established prediction tools. These studies show that AI could be a helpful adjunct to existing techniques for diagnosing, monitoring, and treating both NAFLD and viral hepatitis.

Machine learning in liver transplantation: a tool for some unsolved questions?

Machine learning has recently been proposed as a useful tool in many fields of Medicine, with the aim of increasing diagnostic and prognostic accuracy. Models based on machine learning have been introduced in the setting of solid organ transplantation too, where prognosis depends on a complex, multidimensional and nonlinear relationship between variables pertaining to the donor, the recipient and the surgical procedure. In the setting of liver transplantation, machine learning models have been developed to predict pretransplant survival in patients with cirrhosis, to assess the best donor-to-recipient match during allocation processes, and to foresee postoperative complications and outcomes. This is a narrative review on the role of machine learning in the field of liver transplantation, highlighting strengths and pitfalls, and future perspectives.

Artificial Neural Network as a Tool to Predict Facial Nerve Palsy in Parotid Gland Surgery for Benign Tumors

(1) Background: Despite the increasing use of intraoperative facial nerve monitoring during parotid gland surgery or the improvement in the preoperative radiological assessment, facial nerve injury (FNI) continues to be the most feared complication; (2) Methods: patients who underwent parotid gland surgery for benign tumors between June 2010 and June 2019 were included in this study aiming to make a proof of concept about the reliability of an artificial neural networks (AAN) algorithm for prediction of FNI and compared with a multivariate linear regression (MLR); (3) Results: Concerning prediction accuracy and performance, the ANN achieved the highest sensitivity (86.53% vs 46.23%), specificity (95.67% vs 92.59%), PPV (87.28% vs 66.94%), NPV (95.68% vs 83.37%), ROC–AUC (0.960 vs 0.769) and accuracy (93.42 vs 80.42) than MLR; and (4) Conclusions: ANN prediction models can be useful for otolaryngologists—head and neck surgeons—and patients to provide evidence-based predictions about the risk of FNI. As an advantage, the possibility to develop a calculator using clinical, radiological and histological or cytological information can improve our ability to generate patients counselling before surgery.

Artificial neural network-based models used for predicting 28- and 90-day mortality of patients with hepatitis B-associated acute-on-chronic liver failure

Background

This study aimed to develop prognostic models for predicting 28- and 90-day mortality rates of hepatitis B virus (HBV)-associated acute-on-chronic liver failure (HBV-ACLF) through artificial neural network (ANN) systems.

Methods

Six hundred and eight-four cases of consecutive HBV-ACLF patients were retrospectively reviewed. Four hundred and twenty-three cases were used for training and constructing ANN models, and the remaining 261 cases were for validating the established models. Predictors associated with mortality were determined by univariate analysis and were then included in ANN models for predicting prognosis of mortality. The receiver operating characteristic curve analysis was used to evaluate the predictive performance of the ANN models in comparison with various current prognostic models.

Results

Variables with statistically significant difference or important clinical characteristics were input in the ANN training process, and eight independent risk factors, including age, hepatic encephalopathy, serum sodium, prothrombin activity, γ-glutamyltransferase, hepatitis B e antigen, alkaline phosphatase and total bilirubin, were eventually used to establish ANN models. For 28-day mortality in the training cohort, the model’s predictive accuracy (AUR 0.948, 95% CI 0.925–0.970) was significantly higher than that of the Model for End-stage Liver Disease (MELD), MELD-sodium (MELD-Na), Chronic Liver Failure-ACLF (CLIF-ACLF), and Child-Turcotte-Pugh (CTP) (all p < 0.001). In the validation cohorts the predictive accuracy of ANN model (AUR 0.748, 95% CI: 0.673–0.822) was significantly higher than that of MELD (p = 0.0099) and insignificantly higher than that of MELD-Na, CTP and CLIF-ACLF (p > 0.05). For 90-day mortality in the training cohort, the model’s predictive accuracy (AUR 0.913, 95% CI 0.887–0.938) was significantly higher than that of MELD, MELD-Na, CTP and CLIF-ACLF (all p < 0.001). In the validation cohorts, the prediction accuracy of the ANN model (AUR 0.754, 95% CI: 0.697–0.812 was significantly higher than that of MELD (p = 0.019) and insignificantly higher than MELD-Na, CTP and CLIF-ACLF (p > 0.05).

Conclusions

The established ANN models can more accurately predict short-term mortality risk in patients with HBV- ACLF.

The main content has been postered as an abstract at the AASLD Hepatology Conference (10.1002/hep.30257).

Machine learning application for prediction of locoregional recurrences in early oral tongue cancer: a Web-based prognostic tool

Estimation of risk of recurrence in early-stage oral tongue squamous cell carcinoma (OTSCC) remains a challenge in the field of head and neck oncology. We examined the use of artificial neural networks (ANNs) to predict recurrences in early-stage OTSCC. A Web-based tool available for public use was also developed. A feedforward neural network was trained for prediction of locoregional recurrences in early OTSCC. The trained network was used to evaluate several prognostic parameters (age, gender, T stage, WHO histologic grade, depth of invasion, tumor budding, worst pattern of invasion, perineural invasion, and lymphocytic host response). Our neural network model identified tumor budding and depth of invasion as the most important prognosticators to predict locoregional recurrence. The accuracy of the neural network was 92.7%, which was higher than that of the logistic regression model (86.5%). Our online tool provided 88.2% accuracy, 71.2% sensitivity, and 98.9% specificity. In conclusion, ANN seems to offer a unique decision-making support predicting recurrences and thus adding value for the management of early OTSCC. To the best of our knowledge, this is the first study that applied ANN for prediction of recurrence in early OTSCC and provided a Web-based tool.

Prediction of nosocomial acute-on-chronic liver failure in patients with cirrhosis admitted to hospital with acute decompensation

Nosocomial acute-on-chronic liver failure (nACLF) develops in at least 10% of patients with cirrhosis hospitalized for acute decompensation (AD), greatly worsening their prognosis. In this prospective observational study, we aimed to identify rapidly obtainable predictors at admission, which allow for the early recognition and stratification of patients at risk of nACLF.

Methods

A total of 516 consecutive patients hospitalized for AD of cirrhosis were screened: those who did not present ACLF at admission (410) were enrolled and surveilled for the development of nACLF.

Results

Fifty-nine (14%) patients developed nALCF after a median of 7 (IQR 4-18) days. At admission, they presented a more severe disease and higher degrees of systemic inflammation and anemia than those (351; 86%) who remained free from nACLF. Competing risk multivariable regression analysis showed that baseline MELD score (sub-distribution hazard ratio [sHR] 1.15; 95% CI 1.10-1.21; p ≪0.001), hemoglobin level (sHR 0.81; 95% CI 0.68-0.96; p = 0.018), and leukocyte count (sHR 1.11; 95% CI 1.06-1.16; p ≪0.001) independently predicted nACLF. Their optimal cut-off points, determined by receiver-operating characteristic curve analysis, were: 13 points for MELD score, 9.8 g/dl for hemoglobin, and 5.6x10⁹/L for leukocyte count. These thresholds were used to stratify patients according to the cumulative incidence of nACLF, being 0, 6, 21 and 59% in the presence of 0, 1, 2 or 3 risk factors (p ≪0.001). Nosocomial bacterial infections only increased the probability of developing nACLF in patients with at least 1 risk factor, rising from 3% to 29%, 16% to 50% and 52% to 83% in patients with 1, 2 or 3 risk factors, respectively.

Conclusions

Easily available laboratory parameters, related to disease severity, systemic inflammation, and anemia, can be used to identify, at admission, hospitalized patients with AD at increased risk of developing nACLF.

Lay summary

More than 10% of patients with cirrhosis hospitalized because of an acute decompensation develop acute-on-chronic liver failure, which is associated with high short-term mortality, during their hospital stay. We found that the combination of 3 easily obtainable variables (model for end-stage liver disease score, leukocyte count and hemoglobin level) help to identify and stratify patients according to their risk of developing nosocomial acute-on-chronic liver failure, from nil to 59%. Moreover, if a nosocomial bacterial infection occurs, such an incidence proportionally increases from nil to 83%. This simple approach helps to identify patients at risk of developing nosocomial acute-on-chronic liver failure at admission to hospital, enabling clinicians to put in place preventive measures.

Comparative Study of Back Propagation Artificial Neural Networks and Logistic Regression Model in Predicting Poor Prognosis after Acute Ischemic Stroke

Objective

To investigate the predictive value of clinical variables on the poor prognosis at 90-day follow-up from acute stroke onset, and compare the diagnostic performance between back propagation artificial neural networks (BP ANNs) and Logistic regression (LR) models in predicting the prognosis.

Methods

We studied the association between clinical variables and the functional recovery of 435 acute ischemic stroke patients. The patients were divided into 2 groups according to modified Rankin Scale scores evaluated on the 90th day after stroke onset. Both BP ANNs and LR models were established for predicting the poor outcome and their diagnostic performance were compared by receiver operating curve.

Results

Age, free fatty acid, homocysteine and alkaline phosphatase were closely related with the poor outcome in acute ischemic stroke patients and finally enrolled in models. The accuracy, sensitivity and specificity of BP ANNs were 80.15%, 75.64% and 82.07% respectively. For the LR model, the accuracy, sensitivity and specificity was 70.61%, 88.46% and 63.04% respectively. The area under the ROC curve of the BP ANNs and LR model was 0.881and 0.809.

Conclusions

Both BP ANNs and LR model were promising for the prediction of poor outcome by combining age, free fatty acid, homocysteine and alkaline phosphatase. However, BP ANNs model showed better performance than LR model in predicting the prognosis.

Prognostic factors and treatment effect of standard-volume plasma exchange for acute and acute-on-chronic liver failure: A single-center retrospective study

Patients with acute liver failure (ALF) and acute-on-chronic liver failure (ACLF) have a high risk of mortality. Few studies have reported prognostic factors for patients receiving plasma exchange (PE) for liver support. We conducted a retrospective analysis using data of 55 patients with severe ACLF (n = 45) and ALF (n = 10) who received standard-volume PE (1-1.5 plasma volume) in the ICU. Hepatitis B virus infection accounts for the majority of ACLF (87%) and ALF (50%) patients. PE significantly improved the levels of total bilirubin, prothrombin time and liver enzymes (P<0.05). Thirteen ACLF patients (29%) and one ALF patient (10%) underwent liver transplantation. Two ALF patients (20%) recovered spontaneously without transplantation. The overall in-hospital survival rates for ACLF and ALF patients were 24% and 30%, and the transplant-free survival rates were 0% and 20%, respectively. For the 14 transplanted patients, the one-year survival rate was 86%. Multivariate analysis showed that pre-PE hemoglobin (P = 0.008), post-PE hemoglobin (P = 0.039), and post-PE CLIF-C ACLF scores (P = 0.061) were independent predictors of survival in ACLF. The post-PE CLIF-C ACLF scores ≥59 were a discriminator predicting the in-hospital mortality (area under the curve = 0.719, P = 0.030). Cumulative survival rates differed significantly between patients with CLIF-C ACLF scores ≤ 58 and those with CLIF-C ACLF scores ≥ 59 after PE (P< 0.05). The findings suggest that PE is mainly a bridge for liver transplantation and spontaneous recovery is exceptional even in patients treated with PE. A higher improvement in the post-PE CLIF-C ACLF score is associated with a superior in-hospital survival rate.

Comparison of three data mining models for prediction of advanced schistosomiasis prognosis in the Hubei province

BACKGROUND

In order to better assist medical professionals, this study aimed to develop and compare the performance of three models-a multivariate logistic regression (LR) model, an artificial neural network (ANN) model, and a decision tree (DT) model-to predict the prognosis of patients with advanced schistosomiasis residing in the Hubei province.

METHODOLOGY/PRINCIPAL FINDINGS

Schistosomiasis surveillance data were collected from a previous study based on a Hubei population sample including 4136 advanced schistosomiasis cases. The predictive models use LR, ANN, and DT methods. From each of the three groups, 70% of the cases (2896 cases) were used as training data for the predictive models. The remaining 30% of the cases (1240 cases) were used as validation groups for performance comparisons between the three models. Prediction performance was evaluated using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. Univariate analysis indicated that 16 risk factors were significantly associated with a patient's outcome of prognosis. In the training group, the mean AUC was 0.8276 for LR, 0.9267 for ANN, and 0.8229 for DT. In the validation group, the mean AUC was 0.8349 for LR, 0.8318 for ANN, and 0.8148 for DT. The three models yielded similar results in terms of accuracy, sensitivity, and specificity.

CONCLUSIONS/SIGNIFICANCE

Predictive models for advanced schistosomiasis prognosis, respectively using LR, ANN and DT models were proved to be effective approaches based on our dataset. The ANN model outperformed the LR and DT models in terms of AUC.

Prognostic nomogram for acute-on-chronic hepatitis B liver failure

Background & Aims

To establish an effective prognostic nomogram for acute-on-chronic hepatitis B liver failure (ACHBLF).

Materials and Methods

The nomogram was based on clinical data of 203 ACHBLF patients who admitted to the First Affiliated Hospital of Fujian Medical University from 2009 to 2014. The area under the receiver-operating characteristic curve (AUC) and calibration curve were carried out to verify the predictive accuracy ability of the nomogram. The result was validated in internal and external validation cohorts. Kaplan-Meier survival curve was used in survival analysis.

Results

We developed a new prognostic nomogram to predict 3-month mortality based on risk factors selected by multivariate analysis. This nomogram consisted three independent factors: age, liver to abdominal area ratio (LAAR) and model for end-stage liver disease (MELD) score. The AUC of this nomogram for survival prediction was 0.877 (95% CI 0.831–0.923), which was higher than that of MELD score, MELD-Na and Child-Turcotte-Pugh (CTP). Good agreement of calibration plot for the probability of survival at 3-month was shown between the prediction by nomogram and actual observation. These results were supported by internal and external validation studies.

Conclusions

The ACHBLF nomogram could predict the short-term survival for ACHBLF patients.

Aberrant GSTP1 promoter methylation predicts poor prognosis of acute-on-chronic hepatitis B pre-liver failure

It has been demonstrated that glutathione-S-transferase P1 (GSTP1) could protect cells from DNA damage mediated by oxidizing agents or electrophiles in hepatic inflammatory response. Our study evaluated the methylation status and the predictive value for prognosis of GSTP1 promoter region in patients with acute-on-chronic hepatitis B pre-liver failure (pre-ACHBLF). Methylation status of GSTP1 promoter in peripheral blood mononuclear cells (PBMCs) and plasma was measured in 103 patients with pre-ACHBLF, 80 patients with chronic hepatitis B (CHB) and 30 healthy controls (HCs) by methylation-specific polymerase chain reaction. The mRNA level of GSTP1 was detected by quantitative real-time polymerase chain reaction. The methylation frequency of GSTP1 promoter region in patients with pre-ACHBLF (35/103 in PBMCs and 33/103 in plasma) was significantly higher than CHB (2/80) and HCs (0/30), respectively. The mRNA level of GSTP1 in patients with pre-ACHBLF was significantly lower than CHB and HCs. Additionally, pre-ACHBLF patients with methylated GSTP1 presented strikingly higher incidence of ACHBLF than those without. Of note, GSTP1 methylation presented distinctly better performance than model for end-stage liver disease score [area under the receiver operating characteristic curves (AUCs) 0.825 in PBMCs and 0.798 in plasma VS 0.589; AUC 0.804 in PBMCs and 0.779 in plasma VS 0.622; AUC 0.767 in PBMCs and 0.744 in plasma VS 0.602, respectively] when used to predict the 1-, 2- or 3-month incidence of ACHBLF in patients with pre-ACHBLF. Aberrant methylation of GSTP1 has potential to be a prognostic biomarker for pre-ACHBLF.

Validation of prognostic scores to predict short-term mortality in patients with HBV-related acute-on-chronic liver failure: The CLIF-C OF is superior to MELD, CLIF SOFA, and CLIF-C ACLF

Acute-on-chronic liver failure (ACLF) in chronic hepatitis B (CHB) patients has a high short-term mortality. Identification of effective models to predict the short-term mortality may enable early intervention and improve patients' prognosis. We aim to assess the performance of the CLIF Consortium Organ Failure score (CLIF-C OFs), CLIF sequential organ failure assessment score (CLIF-SOFAs), CLIF Consortium ACLF score (CLIF-C ACLFs), ACLF grade, and model for end-stage liver disease score (MELDs) in predicting the short-term mortality in CHB patients with ACLF.Among the 155 consecutive adult patients with liver failure as a discharge diagnosis were screened, and all the patients were treated at the Department of Infectious Diseases, Huashan Hospital, Fudan University (Shanghai, China) from January 2010 to February 2016. The diagnosis of ACLF was based on the criteria formalized by the ACLF consensus recommendations of the Asian Pacific Association for the Study of the Liver (APASL). Diagnostic accuracy for predicting short-term (28-day) mortality was calculated for CLIF-C OFs, CLIF-SOFAs, CLIF-C ACLFs, ACLF grade, and MELDs in all patients.One hundred fifty-five consecutive adult liver failure patients were screened and 85 patients including 73 males and 12 females were enrolled. Overall, the 28-day transplant-free mortality was 32% in all patients, and 100% in those with severe early course (ACLF-3). The area under the receiver operating characteristic curve (AUROC) of CLIF-C OFs (AUROC: 0.906, P = .0306, compared with MELDs) was higher than those of CLIF-SOFAs (AUROC: 0.876), CLIF-C ACLFs (AUROC: 0.858), ACLF grade (AUROC: 0.857), and MELDs (AUROC: 0.838) for predicting short-term mortality. The cut-point for baseline CLIF-C OFs in predicting death was 8.5, with 67% sensitivity, 90% specificity, and AUROC of 0.906 (95% CI: 0.8450-0.9679).The results indicate that short-term mortality is high in patients with ACLF and CLIF Consortium Organ Failure score is superior to MELD, CLIF SOFA, and CLIF-C ACLF in predicting its short-term mortality.

Risk stratification of decompensated cirrhosis patients by Chronic Liver Failure Consortium scores: Classification and regression tree analysis

AIM

Decompensated cirrhosis patients have greatly variable prognosis. The aim of the study was to carry out a risk stratification for those patients by Chronic Liver Failure (CLIF) Consortium scores.

METHODS

The performance of CLIF Consortium acute-on-chronic liver failure scores (CLIF-C ACLFs) and CLIF Consortium Acute Decompensation scores (CLIF-C ADs) were validated in 209 patients with ACLF and 1245 patients without ACLF at admission from the Ningbo Cohort. A classification and regression tree (CRT) analysis by CLIF-C ACLFs/CLIF-C ADs was carried out to stratify death risk among patients.

RESULTS

The CLIF-C ACLFs and CLIF-C ADs showed higher predictive accuracy than Model for End-stage Liver Disease (MELD) scores, MELD plus serum sodium (MELD-Na) scores, and Child-Turcotte-Pugh classification (CP) at main time points (28, 90, 180, and 365 days), determined by area under the receiver-operating characteristic curve and concordance index in ACLF and no-ACLF patients at admission. The CRT analysis categorized ACLF patients into two groups (advanced and early ACLF), and no-ACLF patients into three groups (high-, medium-, and low-risk AD) according to risk of death. However, early ACLF and high-risk AD patients had comparable mortality at the main time points. The CRT model had a higher area under the receiver-operating characteristic curve than MELDs, MELD-Nas, and CPs in predicting prognosis in all patients.

CONCLUSIONS

The CLIF-C ACLF and CLIF-C AD are better prognostic scores than MELD, MELD-Na, and CP in predicting mortality of ACLF and no-ACLF patients. A combined use of CLIF- Sequential Organ Failure Assessment, CLIF-C ACLFs, and CLIF-C ADs could identify cirrhosis patients at high death risk and assist clinical decisions for management.

Development and validation of a prognostic nomogram for acute-on-chronic hepatitis B liver failure

BACKGROUND AND AIM

Determining individual risk of short-term mortality in patients with acute-on-chronic hepatitis B liver failure (ACHBLF) is a difficult task. We aimed to develop and externally validate a prognostic nomogram for ACHBLF patients.

METHODS

The nomogram was built to estimate the probability of 30-day, 60-day, 90-day, and 60-month survival based on an internal cohort of 246 patients with ACHBLF. The predictive accuracy and discriminative ability of nomogram were determined by a concordance index (C-index), calibration curve, and time-dependent receiver operating characteristics (tdROC), comparing with model for end-stage liver disease (MELD) score. The results were validated using bootstrap resampling and an external cohort of 138 patients. Furthermore, we plotted decision curves to evaluate the clinical usefulness of nomogram.

RESULTS

Independent factors derived from multivariable Cox analysis of training cohort to predict mortality were age, total bilirubin, serum sodium, and prothrombin activity, which were all assembled into nomogram. The calibration curves for probability of survival showed optimal agreement between nomogram prediction and actual observation. The C-index of nomogram was higher than that of MELD score for predicting survival (30-day, 0.809 vs 0.717, P < 0.001; 60-day, 0.792 vs 0.685, P < 0.001; 90-day, 0.779 vs 0.678, P < 0.001; 6-month, 0.781 vs 0.677, P < 0.001). Additionally, tdROC and decision curves also showed that nomogram was superior to MELD score. The results were confirmed in validation cohort.

CONCLUSIONS

The prognostic nomogram provided an individualized risk estimate of short-term survival in patients with ACHBLF, offering to clinicians to improve their abilities to assess patient prognosis.

Classification and regression tree analysis of acute-on-chronic hepatitis B liver failure: Seeing the forest for the trees

At present, there is no ideal model for predicting the short-term outcome of patients with acute-on-chronic hepatitis B liver failure (ACHBLF). This study aimed to establish and validate a prognostic model by using the classification and regression tree (CART) analysis. A total of 1047 patients from two separate medical centres with suspected ACHBLF were screened in the study, which were recognized as derivation cohort and validation cohort, respectively. CART analysis was applied to predict the 3-month mortality of patients with ACHBLF. The accuracy of the CART model was tested using the area under the receiver operating characteristic curve, which was compared with the model for end-stage liver disease (MELD) score and a new logistic regression model. CART analysis identified four variables as prognostic factors of ACHBLF: total bilirubin, age, serum sodium and INR, and three distinct risk groups: low risk (4.2%), intermediate risk (30.2%-53.2%) and high risk (81.4%-96.9%). The new logistic regression model was constructed with four independent factors, including age, total bilirubin, serum sodium and prothrombin activity by multivariate logistic regression analysis. The performances of the CART model (0.896), similar to the logistic regression model (0.914, P=.382), exceeded that of MELD score (0.667, P<.001). The results were confirmed in the validation cohort. We have developed and validated a novel CART model superior to MELD for predicting three-month mortality of patients with ACHBLF. Thus, the CART model could facilitate medical decision-making and provide clinicians with a validated practical bedside tool for ACHBLF risk stratification.

Conditional survival estimate of acute-on-chronic hepatitis B liver failure: a dynamic prediction based on a multicenter cohort

Objectives

Counseling patients with acute-on-chronic hepatitis B liver failure (ACHBLF) on their individual risk of short-term mortality is challenging. This study aimed to develop a conditional survival estimate (CSE) for predicting individualized mortality risk in ACHBLF patients.

Methods

We performed a large prospective cohort study of 278 ACHBLF patients from December 2010 to December 2013 at three participating medical centers. The Kaplan-Meier method was used to calculate the cumulative overall survival (OS). Cox proportional hazard regression models were used to analyze the risk factors associated with OS. 4-week CSE at “X” week after diagnostic established were calculated as CS₄ = OS_(X+4)/OS_(X).

Results

The actual OS at 2, 4, 6, 8, 12 weeks were 80.5%, 71.8%, 69.3%, 66.0% and 63.7%, respectively. Using CSE, the probability of surviving an additional 4 weeks, given that the patient had survived for 1, 3, 5, 7, 9 weeks was 74%, 86%, 92%, 93%, 97%, respectively. Patients with worse prognostic feathers, including MELD > 25, Child grade C, age > 45, HE, INR > 2.5, demonstrated the greatest increase in CSE over time, when compared with the “favorable” one (Δ36% vs. Δ10%; Δ28% vs. Δ16%; Δ29% vs. Δ15%; Δ60% vs. Δ12%; Δ33% vs. Δ12%; all P < 0.001; respectively).

Conclusions

This easy-to-use CSE can accurately predict the changing probability of survival over time. It may facilitate risk communication between patients and physicians.

TIPE2 mRNA Level in PBMCs Serves as a Novel Biomarker for Predicting Short-Term Mortality of Acute-on-Chronic Hepatitis B Liver Failure: A Prospective Single-Center Study

It remains difficult to accurately predicate short-term mortality of acute-on-chronic hepatitis B liver failure (ACHBLF). Tumor necrosis factor-α-induced protein 8-like 2 (TIPE2) is a novel identified negative regulator of immune response and we have previously demonstrated TIPE2 play an essential role in the pathogenesis of ACHBLF. We therefore aimed to evaluate the diagnosis value of TIPE2 mRNA in peripheral blood mononuclear cells (PBMCs) for predicting 3-month mortality of ACHBLF patients. This prospective study consisted of 108 ACHBLF patients from March 2009 to May 2013 as training cohort and 63 ACHBLF patients from June 2013 to December 2014 as validation cohort. Forty-two patients with chronic hepatitis B (CHB) and 22 healthy volunteers were also included as controls. The mRNA level of TIPE2 in PBMCs was determined using quantitative real-time polymerase chain reaction. Univariate analysis and Cox proportional hazard regression analysis were performed to identify independent risk factors to 3-month mortality. Area under the receptor operating characteristic curve (AUROC) was performed to assess diagnostic value of TIPE2 mRNA in training and validation cohort. The level of TIPE2 mRNA was significantly higher in ACHBLF patients (median (interquartile): 6.5 [3.7, 9.6]) compared with CHB (2.3 [1.6, 3.7]) and healthy controls (0.4 [0.3, 0.6]; both P < 0.05). Cox proportional hazards regression analyses showed 5 independent risk factors associated with 3-month mortality of ACHBLF: white blood cells (HR = 1.058, 95% CI: 1.023-1.095), spontaneous bacterial peritonitis (HR = 2.541, 95% CI: 1.378-4.686), hepatic encephalopathy (HR = 1.848, 95% CI: 1.028-3.321), model for end-stage liver diseases (MELD) score (HR = 1.062, 95% CI: 1.009-1.118), and TIPE2 mRNA (HR = 1.081, 95% CI: 1.009-1.159). An optimal cut-off point 6.54 of TIPE2 mRNA showed sensitivity of 74.63%, specificity of 90.24%, positive predictive value of 92.5%, and negative predictive value of 67.3% for predicting 3-month mortality in training cohort. Furthermore, TIPE2 mRNA plus MELD performed better than MELD alone for predicting 3-month mortality in training (AUROC, 0.853 vs 0.722, P < 0.05) and validation cohort (AUROC, 0.909 vs 0.717, P < 0.001). TIPE2 mRNA level might be a novel biomarker in predicting 3-month mortality of ACHBLF. Combination of TIPE2 mRNA and MELD would improve the diagnostic value of MELD alone in predicting 3-month mortality of patients with ACHBLF.

Seasonal variation in onset and relapse of IBD and a model to predict the frequency of onset, relapse, and severity of IBD based on artificial neural network

BACKGROUND

Previous research has yielded conflicting data as to whether the natural history of inflammatory bowel disease follows a seasonal pattern. The purpose of this study was (1) to determine whether the frequency of onset and relapse of inflammatory bowel disease follows a seasonal pattern and (2) to establish a model to predict the frequency of onset, relapse, and severity of inflammatory bowel disease (IBD) with meteorological data based on artificial neural network (ANN).

METHOD

Patients with diagnosis of ulcerative colitis (UC) or Crohn's disease (CD) between 2003 and 2011 were investigated according to the occurrence of onset and flares of symptoms. The expected onset or relapse was calculated on a monthly basis over the study period. For artificial neural network (ANN), patients from 2003 to 2010 were assigned as training cohort and patients in 2011 were assigned as validation cohort. Mean square error (MSE) and mean absolute percentage error (MAPE) were used to evaluate the predictive accuracy.

RESULTS

We found no seasonal pattern of onset (P = 0.248) and relapse (P = 0.394) among UC patients. But, the onset (P = 0.015) and relapse (P = 0.004) of CD were associated with seasonal pattern, with a peak in July and August. ANN had average accuracy to predict the frequency of onset (MSE = 0.076, MAPE = 37.58%) and severity of IBD (MSE = 0.065, MAPE = 42.15%) but high accuracy in predicting the frequency of relapse of IBD (MSE = 0.009, MAPE = 17.1%).

CONCLUSION

The frequency of onset and relapse in IBD showed seasonality only in CD, with a peak in July and August, but not in UC. ANN may have its value in predicting the frequency of relapse among patients with IBD.

Aberrant GSTP1 promoter methylation predicts short-term prognosis in acute-on-chronic hepatitis B liver failure

BACKGROUND

Glutathione-S-transferase P1 (GSTP1) methylation has been demonstrated to be associated with oxidative stress induced liver damage in acute-on-chronic hepatitis B liver failure (ACHBLF).

AIM

To evaluate the methylation level of GSTP1 promoter in acute-on-chronic hepatitis B liver failure and determine its predictive value for prognosis.

METHODS

One hundred and five patients with acute-on-chronic hepatitis B liver failure, 86 with chronic hepatitis B (CHB) and 30 healthy controls (HC) were retrospectively enrolled. GSTP1 methylation level in peripheral mononuclear cells (PBMC) was detected by MethyLight. Clinical and laboratory parameters were obtained.

RESULTS

GSTP1 methylation levels were significantly higher in patients with acute-on-chronic hepatitis B liver failure (median 16.84%, interquartile range 1.83-59.05%) than those with CHB (median 1.25%, interquartile range 0.48-2.47%; P < 0.01) and HC (median 0.80%, interquartile range 0.67-1.27%; P < 0.01). In acute-on-chronic hepatitis B liver failure group, nonsurvivors showed significantly higher GSTP1 methylation levels (P < 0.05) than survivors. GSTP1 methylation level was significantly correlated with total bilirubin (r = 0.29, P < 0.01), prothrombin time activity (r = -0.24, P = 0.01) and model for end-stage liver disease (MELD) score (r = 0.26, P = 0.01). When used to predict 1- or 2-month mortality of acute-on-chronic hepatitis B liver failure, GSTP1 methylation showed significantly better predictive value than MELD score [area under the receiver operating characteristic curve (AUC) 0.89 vs. 0.72, P < 0.01; AUC 0.83 vs. 0.70, P < 0.05 respectively]. Meanwhile, patients with GSTP1 methylation levels above the cut-off points showed significantly poorer survival than those below (P < 0.05).

CONCLUSIONS

Aberrant GSTP1 promoter methylation exists in acute-on-chronic hepatitis B liver failure and shows high predictive value for short-term mortality. It might serve as a potential prognostic marker for acute-on-chronic hepatitis B liver failure.

Overexpression of serum sST2 is associated with poor prognosis in acute-on-chronic hepatitis B liver failure

BACKGROUND AND OBJECTIVE

Interleukin-33 (IL-33) and soluble ST2 (sST2) have been demonstrated to be involved in liver injury. The present study aims to evaluate serum IL-33 and sST2 level in acute-on-chronic hepatitis B liver failure (ACHBLF) and determine their predictive value for prognosis.

METHODS

Serum IL-33 and sST2 level in patients with ACHBLF, chronic hepatitis B (CHB) and healthy controls (HCs) were determined by enzyme-linked immunosorbent assay (ELISA). Clinical and laboratory parameters were obtained.

RESULTS

Serum IL-33 was significantly higher in patients with ACHBLF (313.10±419.97pg/ml) than those with CHB (97.25±174.67pg/ml, P<0.01) and HCs (28.39±6.53pg/ml, P<0.01). Serum sST2 was significantly higher in patients with ACHBLF (1545.87±1135.70pg/ml) than those with CHB (152.55±93.28pg/ml, P<0.01) and HCs (149.27±104.90pg/ml, P<0.01). In all participants, serum IL-33 was significantly correlated with sST2 (r=0.43, P<0.01). In patients with ACHBLF, serum IL-33 was significantly correlated with alanine aminotransferase (ALT; r=0.26, P=0.04). Serum sST2 was significantly correlated with total bilirubin (TBIL; r=0.59, P<0.01), Log10 [HBV DNA] (r=-0.47, P<0.01) and model for end-stage liver diseases (MELD; r=0.28, P=0.03). Serum sST2 had an area under the receiver operating characteristic curve (AUC) of 0.81 in predicting 3-month mortality of ACHBLF. Patients with ACHBLF who had sST2 >1507pg/ml showed significantly poorer survival than those who had sST2 ≤1507pg/ml (P<0.01). Moreover, measurement of sST2 and MELD together significantly improved the diagnostic value of MELD alone (P<0.05).

CONCLUSIONS

Our study showed that serum IL-33 and sST2 were overexpressed in ACHBLF and sST2 might potentially serve as a prognostic marker for it.

Model for end-stage liver disease score as a predictor of short-term outcome in patients with drug-induced liver injury

OBJECTIVES

The purposes of this study were to investigate the clinical presentation, cause and outcome of drug-induced liver injury (DILI) and to evaluate the predictive value of the model for end-stage liver disease (MELD) score in DILI.

METHODS

Patients diagnosed with DILI between January 1, 2010 and December 31, 2012 in the Emergency Department at Asan Medical Center in Seoul, Korea were analyzed retrospectively. The primary end point was poor outcome, defined as liver transplantation or death within 30 days of the initial hospital visit.

RESULTS

Of 213 patients, 13.1% had a 30-day poor outcome. Folk remedies were the most common cause of DILI in 147 patients (69%). Univariate logistic regression analysis showed that multiple drugs (odds ratio [OR] 2.30, 95% confidence interval [CI]: 1.03-5.15), concurrent alcohol consumption (OR 3.69, 95% CI: 1.03-13.18), white blood cell (WBC) count (OR 1.17, 95% CI: 1.07-1.28), hemoglobin (Hb) (OR 0.60, 95% CI: 0.49-0.74), platelet count (OR 0.993, 95% CI: 0.987-0.998), total bilirubin (OR 1.09, 95% CI: 1.06-1.13) and MELD (OR 1.23, 95% CI: 1.15-1.32) were significantly associated with 30-day poor outcomes. Multivariate analysis showed that the MELD (OR 1.21, 95% CI: 1.12-1.30) and Hb (OR 0.77, 95% CI: 0.61-0.98) were independent predictors of poor outcome. For 30-day mortality, the c-statistics for MELD alone and for combination of MELD and Hb were 0.93 (95% CI: 0.89-0.97) and 0.94 (95% CI: 0.90-0.97), respectively.

CONCLUSION

The outcome of patients with DILI was poor. MELD score and Hb were reliable predictors of short-term outcome in patients with DILI.

Increased delayed mortality in patients with acute-on-chronic liver failure who have prior decompensation

BACKGROUND AND AIM

Patients with acute-on-chronic liver failure (ACLF) represent a complex population with differential prognosis. The aim of the study was to categorize ACLF according to the severity of underlying chronic liver diseases.

METHODS

A total of 540 ACLF patients were recruited, including 127 with prior decompensated cirrhosis and 413 without prior decompensation (PD) including 193 with underlying chronic hepatitis and 220 with prior compensated cirrhosis. The clinical characteristics and prognosis of subgroups were compared. Cox proportional hazard model and multinominal logistic regression analysis were performed to identify significant prognostic parameters.

RESULTS

The 28-day, 3-month and 1-year survival of ACLF patients with or without PD were 58.9% versus 61.4%, 36.2 versus 52.5%, and 29.1% versus 49.6%, respectively. On multinominal logistic regression analysis or time-to-death analysis by Cox proportional hazard model, PD was significantly associated with post-28-day mortality but not within-28-day mortality. On multivariate time-to-death analysis, older age, high international normalized ratio (INR) and serum bilirubin, low levels of serum sodium and platelet count, and presence of hepatic encephalopathy (HE), upper gastrointestinal bleeding, and respiratory or circulation dysfunction were predictors of within-28-day mortality in patients without PD, whereas the risk factors in patients with PD were high INR, creatinine, presence of HE, and respiratory or circulation dysfunction.

CONCLUSION

ACLF patients with or without PD had comparable short-term prognosis but differential 1-year mortality. ACLF patients with PD were distinct from those without PD in age, types of acute insults, severity of hepatic damage, and distribution of complications, and the former group was characterized by increased delayed mortality.

Aberrant DNA methylation of G-protein-coupled bile acid receptor Gpbar1 predicts prognosis of acute-on-chronic hepatitis B liver failure

The G-protein-coupled bile acid receptor Gpbar1 (TGR5) has been demonstrated to be able to negatively regulate hepatic inflammatory response. In this study, we aimed to determine the methylation status of TGR5 promoter in patients with acute-on-chronic hepatitis B liver failure (ACHBLF) and its predictive value for prognosis. We enrolled 76 consecutive ACHBLF patients, 80 chronic hepatitis B (CHB) patients and 30 healthy controls (HCs). Methylation status of TGR5 promoter in peripheral mononuclear cell (PBMC) was detected by methylation-specific polymerase chain reaction (MSP). The mRNA level of TGR5 was determined by quantitative real-time polymerase chain reaction (RT-qPCR). We found that the frequency of TGR5 promoter methylation was significantly higher in ACHBLF (35/76, 46.05%) than CHB patients (5/80, 6.25%; χ(2) = 32.38, P < 0.01) and HCs (1/30, 3.33%; χ(2) = 17.50, P < 0.01). TGR5 mRNA level was significantly lower (Z = -9.12, P < 0.01) in participants with aberrant methylation than those without. TGR5 methylation showed a sensitivity of 46.05% (35/76), specificity of 93.75% (75/80), positive predictive value (PPV) of 87.5% (35/40) and negative predictive value (NPV) of 64.66% (75/116) in discriminating ACHBLF from CHB patients. ACHBLF patients with methylated TGR5 showed significantly poor survival than those without (P < 0.01). When used to predict 3-month mortality of ACHBLF, TGR5 methylation [area under the receiver operating characteristic curve (AUC) = 0.75] performed significantly better than model for end-stage liver diseases (MELD) score (AUC = 0.65; P < 0.05). Therefore, our study demonstrated that aberrant TGR5 promoter methylation occurred in ACHBLF and might be a potential prognostic marker for the disease.

Scoring systems predict the prognosis of acute-on-chronic hepatitis B liver failure: an evidence-based review

Acute-on-chronic hepatitis B liver failure is a devastating condition that is associated with mortality rates of over 50% and is consequent to acute exacerbation of chronic hepatitis B in patients with previously diagnosed or undiagnosed chronic liver disease. Liver transplantation is the definitive treatment to lower mortality rate, but there is a great imbalance between donation and potential recipients. An early and accurate prognostic system based on the integration of laboratory indicators, clinical events and some mathematic logistic equations is needed to optimize treatment for patients. As parts of the scoring systems, the MELD was the most common and the donor-MELD was the most innovative for patients on the waiting list for liver transplantation. This review aims to highlight the various features and prognostic capabilities of these scoring systems.

Artificial neural network accurately predicts hepatitis B surface antigen seroclearance

BACKGROUND & AIMS

Hepatitis B surface antigen (HBsAg) seroclearance and seroconversion are regarded as favorable outcomes of chronic hepatitis B (CHB). This study aimed to develop artificial neural networks (ANNs) that could accurately predict HBsAg seroclearance or seroconversion on the basis of available serum variables.

METHODS

Data from 203 untreated, HBeAg-negative CHB patients with spontaneous HBsAg seroclearance (63 with HBsAg seroconversion), and 203 age- and sex-matched HBeAg-negative controls were analyzed. ANNs and logistic regression models (LRMs) were built and tested according to HBsAg seroclearance and seroconversion. Predictive accuracy was assessed with area under the receiver operating characteristic curve (AUROC).

RESULTS

Serum quantitative HBsAg (qHBsAg) and HBV DNA levels, qHBsAg and HBV DNA reduction were related to HBsAg seroclearance (P<0.001) and were used for ANN/LRM-HBsAg seroclearance building, whereas, qHBsAg reduction was not associated with ANN-HBsAg seroconversion (P = 0.197) and LRM-HBsAg seroconversion was solely based on qHBsAg (P = 0.01). For HBsAg seroclearance, AUROCs of ANN were 0.96, 0.93 and 0.95 for the training, testing and genotype B subgroups respectively. They were significantly higher than those of LRM, qHBsAg and HBV DNA (all P<0.05). Although the performance of ANN-HBsAg seroconversion (AUROC 0.757) was inferior to that for HBsAg seroclearance, it tended to be better than those of LRM, qHBsAg and HBV DNA.

CONCLUSIONS

ANN identifies spontaneous HBsAg seroclearance in HBeAg-negative CHB patients with better accuracy, on the basis of easily available serum data. More useful predictors for HBsAg seroconversion are still needed to be explored in the future.

Comparison between artificial neural network and Cox regression model in predicting the survival rate of gastric cancer patients

The aim of this study was to determine the prognostic factors and their significance in gastric cancer (GC) patients, using the artificial neural network (ANN) and Cox regression hazard (CPH) models. A retrospective analysis was undertaken, including 289 patients with GC who had undergone gastrectomy between 2006 and 2007. According to the CPH analysis, disease stage, peritoneal dissemination, radical surgery and body mass index (BMI) were selected as the significant variables. According to the ANN model, disease stage, radical surgery, serum CA19-9 levels, peritoneal dissemination and BMI were selected as the significant variables. The true prediction of the ANN was 85.3% and of the CPH model 81.9%. In conclusion, the present study demonstrated that the ANN model is a more powerful tool in determining the significant prognostic variables for GC patients, compared to the CPH model. Therefore, this model is recommended for determining the risk factors of such patients.