Comparison of General and Liver-Specific Prognostic Scores in Their Ability to Predict Mortality in Cirrhotic Patients Admitted to the Intensive Care Unit

Development and validation of a new prognostic model for patients with acute-on-chronic liver failure in intensive care unit

BACKGROUND

Cirrhotic patients with acute-on-chronic liver failure (ACLF) in the intensive care unit (ICU) have a poor but variable prognoses. Accurate prognosis evaluation can guide the rational management of patients with ACLF. However, existing prognostic scores for ACLF in the ICU environment lack sufficient accuracy.

AIM

To develop a new prognostic model for patients with ACLF in ICU.

METHODS

Data from 938 ACLF patients in the Medical Information Mart for Intensive Care (MIMIC) database were used to develop a new prognostic model (MIMIC ACLF) for ACLF. Discrimination, calibration and clinical utility of MIMIC ACLF were assessed by area under receiver operating characteristic curve (AUROC), calibration curve and decision curve analysis (DCA), respectively. MIMIC ACLF was then externally validated in a multiple-center cohort, the Electronic Intensive Care Collaborative Research Database and a single-center cohort from the Second Hospital of Hebei Medical University in China.

RESULTS

The MIMIC ACLF score was determined using nine variables: ln (age) × 2.2 + ln (white blood cell count) × 0.22 - ln (mean arterial pressure) × 2.7 + respiratory failure × 0.6 + renal failure × 0.51 + cerebral failure × 0.31 + ln (total bilirubin) × 0.44 + ln (internationalized normal ratio) × 0.59 + ln (serum potassium) × 0.59. In MIMIC cohort, the AUROC (0.81/0.79) for MIMIC ACLF for 28/90-day ACLF mortality were significantly greater than those of Chronic Liver Failure Consortium ACLF (0.76/0.74), Model for End-stage Liver Disease (MELD; 0.73/0.71) and MELD-Na (0.72/0.70) (all P < 0.001). The consistency between actual and predicted 28/90-day survival rates of patients according to MIMIC ACLF score was excellent and superior to that of existing scores. The net benefit of MIMIC ACLF was greater than that achieved using existing scores within the 50% threshold probability. The superior predictive accuracy and clinical utility of MIMIC ACLF were validated in the external cohorts.

CONCLUSION

We developed and validated a new prognostic model with satisfactory accuracy for cirrhotic patients with ACLF hospitalized in the ICU. The model-based risk stratification and online calculator might facilitate the rational management of patients with ACLF.

Critical care outcomes in decompensated cirrhosis: a United States national inpatient sample cross-sectional study

BACKGROUND

Prior assessments of critical care outcomes in patients with cirrhosis have shown conflicting results. We aimed to provide nationwide generalizable results of critical care outcomes in patients with decompensated cirrhosis.

METHODS

This is a retrospective study using the National Inpatient Sample from 2016 to 2019. Adults with cirrhosis who required respiratory intubation, central venous catheter placement or both (n = 12,945) with principal diagnoses including: esophageal variceal hemorrhage (EVH, 24%), hepatic encephalopathy (58%), hepatorenal syndrome (HRS, 14%) or spontaneous bacterial peritonitis (4%) were included. A comparison cohort of patients without cirrhosis requiring intubation or central line placement for any principal diagnosis was included.

RESULTS

Those with cirrhosis were younger (mean 58 vs. 63 years, p < 0.001) and more likely to be male (62% vs. 54%, p < 0.001). In-hospital mortality was higher in the cirrhosis cohort (33.1% vs. 26.6%, p < 0.001) and ranged from 26.7% in EVH to 50.6% HRS. Mortality when renal replacement therapy was utilized (n = 1580, 12.2%) was 46.5% in the cirrhosis cohort, compared to 32.3% in other hospitalizations (p < 0.001), and was lowest in EVH (25.7%) and highest in HRS (51.5%). Mortality when cardiopulmonary resuscitation was used was increased in the cirrhosis cohort (88.0% vs. 72.1%, p < 0.001) and highest in HRS (95.7%).

CONCLUSIONS

One-third of patients with cirrhosis requiring critical care did not survive to discharge in this U.S. nationwide assessment. While outcomes were worse than in patients without cirrhosis, the results do suggest better outcomes compared to previous studies.

Clinical Profile, Short-term Prognostic Accuracies of CLIF-C ACLF Score and Serial CLIF-C OF Scores in Acute-on-chronic Liver Failure Patients: A Prospective Observational Study

Background

Acute-on-chronic liver failure (ACLF) is a recently defined entity that carries high short-term mortality. The European Association for Study of Liver (EASL) has given a different definition for ACLF and derived two scores called Chronic Liver Failure-Consortium Organ Failure (CLIF-C OF) and CLIF-C ACLF to diagnose and predict the short-term outcome, respectively.

Materials and methods

This was the prospective observational study, included 40 ACLF patients diagnosed as per the EASL definition and calculated CLIF-C ACLF as well as other scores (CTP, MELD, MELD-Na, CLIF-C OF) on admission. Serial CLIF-C OF scores were also calculated (Day 3 and Day 7). The 28-day and 90-day mortality was recorded.

Results

Alcohol was the predominant etiology of cirrhosis (32 patients-80%). Infection was the chief precipitating factor in 19 patients (47.5%). The 28-day and 90-day mortality was 45% and 52.5%. Mean (SD) of CLIF-C ACLF scores of survivors and non-survivors on Day-90 were 44.11(6.62) and 53.86 (7.83). The prognostic accuracy of the CLIF-C ACLF score (Area Under Receiver Operating Characteristic Curve-AUROC) to predict 28-day and 90-day mortality was 0.86 and 0.84, respectively. MELD-Na and CLIF-C ACLF scores had higher AUROC for predicting 28-day and 90-day mortality, respectively. The AUROC of the CLIF-C OF score on Day 3 was found to be higher than the values of Day 1 and Day 7, but it was not statistically significant.

Conclusion

CLIF-C ACLF has good short-term prognostic accuracy and it is as good as other available scores. Serial CLIF-C OF scores were equally good in predicting in short-term mortality.

How to cite this article

Hareesh GJ, Ramadoss R. Clinical Profile, Short-term Prognostic Accuracies of CLIF-C ACLF Score and Serial CLIF-C OF Scores in Acute-on-chronic Liver Failure Patients: A Prospective Observational Study. Indian J Crit Care Med 2024;28(2):126-133.

An artificial neural network approach to diagnose and predict liver dysfunction and failure in the critical care setting

Background

Detecting liver dysfunction/failure in the intensive care unit poses a challenge as individuals afflicted with these conditions often appear symptom-free, thereby complicating early diagnoses and contributing to unfavorable patient outcomes. The objective of this endeavor was to improve the chances of early diagnosis of liver dysfunction/failure by creating a predictive model for the critical care setting. This model has been designed to produce an index that reflects the probability of severe liver dysfunction/failure for patients in intensive care units, utilizing machine learning techniques.

Materials and Methods

This effort used comprehensive open-access patient databases to build and validate machine learning-based models for predicting the likelihood of severe liver dysfunction/failure. Two artificial neural network model architectures that derived a novel 0-100 Liver Failure Risk Index were developed and validated using the comprehensive patient databases. Data used to train and develop the models included clinical (patient vital signs) and laboratory results related to liver function which included liver function test results. The performance of the developed models was compared in terms of sensitivity, specificity, and the mean lead time to diagnosis.

Results

The best model performance demonstrated an 83.3 % sensitivity and a specificity of 77.5 % in diagnosing severe liver dysfunction/failure. This model accurately identified these patients a median of 17.5 hours before their clinical diagnosis, as documented in their electronic health records. The predictive diagnostic capability of the developed models is crucial to the intensive care unit setting, where treatment and preventative interventions can be made to avoid severe liver dysfunction/failure.

Conclusion

Our machine learning approach facilitates early and timely intervention in the hepatic function of critically ill patients by their healthcare providers to prevent or minimize associated morbidity and mortality. HIPPOKRATIA 2024, 28 (1):1-10.

EASL Clinical Practice Guidelines on acute-on-chronic liver failure

Acute-on-chronic liver failure (ACLF), which was described relatively recently (2013), is a severe form of acutely decompensated cirrhosis characterised by the existence of organ system failure(s) and a high risk of short-term mortality. ACLF is caused by an excessive systemic inflammatory response triggered by precipitants that are clinically apparent (e.g., proven microbial infection with sepsis, severe alcohol-related hepatitis) or not. Since the description of ACLF, some important studies have suggested that patients with ACLF may benefit from liver transplantation and because of this, should be urgently stabilised for transplantation by receiving appropriate treatment of identified precipitants, and full general management, including support of organ systems in the intensive care unit (ICU). The objective of the present Clinical Practice Guidelines is to provide recommendations to help clinicians recognise ACLF, make triage decisions (ICU vs. no ICU), identify and manage acute precipitants, identify organ systems that require support or replacement, define potential criteria for futility of intensive care, and identify potential indications for liver transplantation. Based on an in-depth review of the relevant literature, we provide recommendations to navigate clinical dilemmas followed by supporting text. The recommendations are graded according to the Oxford Centre for Evidence-Based Medicine system and categorised as 'weak' or 'strong'. We aim to provide the best available evidence to aid the clinical decision-making process in the management of patients with ACLF.

Association and prediction of red blood cell distribution width to albumin ratio in all-cause mortality of acute kidney injury in critically ill patients

Aim

The progression of acute kidney injury (AKI) might be associated with systemic inflammation. Our study aims to explore the association and predictive value of the red blood cell distribution width (RDW) to human serum albumin (ALB) ratio (RDW/ALB ratio), an inflammation-related indicator, in the risk of all-cause mortality and renal replacement therapy (RRT) in AKI patients admitted in intensive care units (ICU).

Methods

A retrospective cohort study was designed, and data were extracted from the Medical Information Mart for Intensive Care III (MIMIC-III). The primary outcome was the risk of all-cause mortality (1-month, 3-month, and 12-month), and the secondary outcome was the risk of RRT. The association between the RDW/ALB ratio and the risk of all-cause mortality and RRT was assessed using the Cox regression analysis, with results shown as hazard ratio (HR) and 95% confidence intervals (CIs). The relationship between the RDW/ALB ratio and crude probability of all-cause mortality or RRT was assessed using restricted cubic splines (RCS). The concordance index (C-index) was used to assess the discrimination of the prediction model.

Results

A total of 13,856 patients were included in our study. In the fully adjusted Cox regression model, we found that a high RDW/ALB ratio was associated with an increased risk of 1-month, 3-month, and 12-month all-cause mortality and RRT (all p &lt; 0.05). Moreover, RCS curves showed the linear relationship between the RDW/ALB ratio and the probability of all-cause mortality and RRT, and the probability was elevated with the increase of the ratio. In addition, the RDW/ALB ratio showed a good predictive performance in the risk of 1-month all-cause mortality, 3-month all-cause mortality, 12-month all-cause mortality, and RRT, with a C-index of 0.728 (95%CI: 0.719–0.737), 0.728 (95%CI: 0.721–0.735), 0.719 (95%CI: 0.713–0.725), and 0.883 (95%CI: 0.876–0.890), respectively.

Conclusion

The RDW/ALB ratio performed well to predict the risk of all-cause mortality and RRT in critically ill patients with AKI, indicating that this combined inflammatory indicator might be effective in clinical practice.

CLIF-SOFA and CLIF-C scores for the prognostication of acute-on-chronic liver failure and acute decompensation of cirrhosis: A systematic review

BACKGROUND

Acute-on-chronic liver failure (ACLF) is a syndrome characterized by decompensation in individuals with chronic liver disease, generally secondary to one or more extra-hepatic organ failures, implying an elevated mortality rate. Acute decompensation (AD) is the term used for one or more significant consequences of liver disease in a short time and is the most common reason for hospital admission in cirrhotic patients. The European Association for the Study of Liver-Chronic-Liver Failure (EASL-CLIF) Group modified the intensive care Sequential Organ Failure Assessment score into CLIF-SOFA, which detects the presence of ACLF in patients with or without AD, classifying it into three grades.

AIM

To investigate the role of the EASL-CLIF definition for ACLF and the ability of CLIF-SOFA, CLIF-C ACLF, and CLIF-C AD scores for prognosticating ACLF or AD.

METHODS

This study is a literature review using a standardized search method, conducted using the steps following the guidelines for reporting systematic reviews set out by the PRISMA statement. For specific keywords, relevant articles were found by searching PubMed, ScienceDirect, and BioMed Central-BMC. The databases were searched using the search terms by one reviewer, and a list of potentially eligible studies was generated based on the titles and abstracts screened. The data were then extracted and assessed on the basis of the Reference Citation Analysis (https://www.referencecitationanalysis.com/).

RESULTS

Most of the included studies used the EASL-CLIF definition for ACLF to identify cirrhotic patients with a significant risk of short-term mortality. The primary outcome in all reviewed studies was mortality. Most of the study findings were based on an area under the receiver operating characteristic curve (AUROC) analysis, which revealed that CLIF-SOFA, CLIF-C ACLF, and CLIF-C AD scores were preferable to other models predicting 28-d mortality. Their AUROC scores were higher and able to predict all-cause mortality at 90, 180, and 365 d. A total of 50 articles were included in this study, which found that the CLIF-SOFA, CLIF-C ACLF and CLIF-C AD scores in more than half of the articles were able to predict short-term and long-term mortality in patients with either ACLF or AD.

CONCLUSION

CLIF-SOFA score surpasses other models in predicting mortality in ACLF patients, especially in the short-term. CLIF-SOFA, CLIF-C ACLF, and CLIF-C AD are accurate short-term and long-term mortality prognosticating scores.

Which model is superior in predicting ICU survival: artificial intelligence versus conventional approaches

Background

A disease severity classification system is widely used to predict the survival of patients admitted to the intensive care unit with different diagnoses. In the present study, conventional severity classification systems were compared with artificial intelligence predictive models (Artificial Neural Network and Decision Tree) in terms of the prediction of the survival rate of the patients admitted to the intensive care unit.

Methods

This retrospective cohort study was performed on the data of the patients admitted to the ICU of Ghaemshahr’s Razi Teaching Care Center from March 20th, 2017, to September 22nd, 2019. The required data for calculating conventional severity classification models (SOFA, SAPS II, APACHE II, and APACHE IV) were collected from the patients’ medical records. Subsequently, the score of each model was calculated. Artificial intelligence predictive models (Artificial Neural Network and Decision Tree) were developed in the next step. Lastly, the performance of each model in predicting the survival of the patients admitted to the intensive care unit was evaluated using the criteria of sensitivity, specificity, accuracy, F-measure, and area under the ROC curve. Also, each model was validated externally. The R program, version 4.1, was used to create the artificial intelligence models, and SPSS Statistics Software, version 21, was utilized to perform statistical analysis.

Results

The area under the ROC curve of SOFA, SAPS II, APACHE II, APACHE IV, multilayer perceptron artificial neural network, and CART decision tree were 76.0, 77.1, 80.3, 78.5, 84.1, and 80.0, respectively.

Conclusion

The results showed that although the APACHE II model had better results than other conventional models in predicting the survival rate of the patients admitted to the intensive care unit, the other conventional models provided acceptable results too. Moreover, the findings showed that the artificial neural network model had the best performance among all the studied models, indicating the discrimination power of this model in predicting patient survival compared to the other models.

Performance of Sequential Organ Failure Assessment and Simplified Acute Physiology Score II for Post-Cardiac Surgery Patients in Intensive Care Unit

Background: The aim of this study is to assess the performance of Sequential Organ Failure Assessment (SOFA) score and Simplified Acute Physiology Score (SAPS II) on outcomes of patients with cardiac surgery and identify the cutoff values to provide a reference for early intervention.

Methods: All data were extracted from MIMIC-III (Medical Information Mart for Intensive Care-III) database. Cutoff values were calculated by the receiver-operating characteristic curve and Youden indexes. Patients were grouped, respectively, according to the cutoff values of SOFA and SAPS II. A non-adjusted model and adjusted model were established to evaluate the prediction of risk. Comparison of clinical efficacy between two scoring systems was made by decision curve analysis (DCA). The primary outcomes of this study were in-hospital mortality, 28-day mortality, 90-day mortality, and 1-year mortality after cardiac surgery. The secondary outcomes included length of hospital stay and intensive care unit (ICU) stay and the incidence of acute kidney injury (AKI) within 7 days after ICU admission.

Results: A total of 6,122 patients were collected and divided into the H-SOFA group (SOFA ≥ 7) and L-SOFA group (SOFA < 7) or H-SAPS II group (SAPS II ≥ 43) and L-SAPS II group (SAPS II < 43). In-hospital mortality, 28-day mortality, 90-day mortality, and 1-year mortality were higher, the length of hospital and ICU stay were longer in the H-SOFA group than in the L-SOFA group (p < 0.05), while the incidence of AKI was not significantly different. In-hospital mortality, 28-day mortality, 90-day mortality, 1-year mortality, and the incidence of AKI were all significantly higher in the H-SAPS II group than in the L-SAPS II group (p < 0.05). Hospital stay and ICU stay were longer in the H-SAPS II group than in the L-SAPS II group (p < 0.05). According to DCA, the SAPS II scoring system had more net benefits on assessing the long-term mortality compared with the SOFA scoring system.

Conclusion: Exceeding the cutoff values of SOFA and SAPS II scores could lead to increased mortality and extended length of ICU and hospital stay. The SAPS II scoring system had a better discriminative performance of 90-day mortality and 1-year mortality in post-cardiac surgery patients than the SOFA scoring system. Emphasizing the critical value of the scoring system is of significance for timely treatment.