MELD 3.0 in Advanced Chronic Liver Disease.

The MELD (model for end-stage liver disease) 3.0 score was developed to replace the MELD-Na score that is currently used to prioritize liver allocation for cirrhotic patients awaiting liver transplantation in the United States. The MELD 3.0 calculator includes new inputs from patient sex and serum albumin levels and has new weights for serum sodium, bilirubin, international normalized ratio, and creatinine levels. It is expected that use of MELD 3.0 scores will reduce overall waitlist mortality modestly and improve access for female liver transplant candidates. The utility of MELD 3.0 and PELDcre (pediatric end-stage liver disease, creatinine) scores for risk stratification in cirrhotic patients undergoing major abdominal surgery, placement of a transjugular intrahepatic portosystemic shunt, and other interventions requires further study. This article reviews the background of the MELD score and the rationale to create MELD 3.0 as well as potential implications of using this newer risk stratification tool in clinical practice.

The Diagnosis and Non-pharmacological Management of Acute Kidney Injury in Patients with Cirrhosis.

Acute kidney injury in patients with cirrhosis is quite common, and is seen in up to 50% of patients hospitalized for decompensated cirrhosis. Causes of acute kidney injury include prerenal, renal, or postrenal etiologies. The diagnosis and early institution of nonpharmacologic and pharmacologic management are key to the recovery of renal function. The objective of this review is to provide a practical approach to the use of diagnostic biomarkers and highlight the nonpharmacologic management and prevention of acute kidney injury.

Using Artificial Intelligence to Predict Cirrhosis From Computed Tomography Scans.

INTRODUCTION: Undiagnosed cirrhosis remains a significant problem. In this study, we developed and tested an automated liver segmentation tool to predict the presence of cirrhosis in a population of patients with paired liver biopsy and computed tomography (CT) scans. METHODS: We used a cohort of 1,590 CT scans within the Morphomics database to train an automated liver segmentation model using 3D-U-Net and Google's DeeplLabv3+. Imaging features were then automatically calculated from an external test cohort of patients with chronic liver disease who had a paired liver biopsy and CT within 6 months of each other in January 2004-2012. Using gradient boosting decision trees, we developed multivariate models to predict the presence of histologic cirrhosis and evaluated with 5-fold cross-validated c-statistic. RESULTS: Our cohort had 351 patients; 96 patients had cirrhosis. Of the total cohort, 72 were postliver transplant. Both fibrosis (FIB)-4 and liver morphomics alone performed equally well with area under the receiving operating characteristics of 0.76 (95% confidence interval 0.70-0.81) and 0.71 (95% confidence interval 0.65-0.76), respectively ( P = 0.2). However, the combination of liver morphomics with laboratory values or liver morphomics with laboratory and demographic data resulted in significant improved performance with area under the receiving operating characteristics of 0.84 (0.80-0.89) and 0.85 (0.81-0.90), respectively, compared with FIB-4 alone ( P < 0.001). In a subgroup analysis, we also examined performance in patients without liver transplantation and saw similar augmentation of FIB-4. DISCUSSION: This proof-of-principle study demonstrates that automatically extracted features within CT scans can be combined with classic electronic medical record data to improve the prediction of cirrhosis in patients with liver disease. This tool may be used in both pretransplant and posttransplant patients and has the potential to improve our ability to detect undiagnosed cirrhosis.

Portal Venous Remodeling Determines the Pattern of Cirrhosis Decompensation: A Systems Analysis.

INTRODUCTION: As liver disease progresses, scarring results in worsening hemodynamics ultimately culminating in portal hypertension. This process has classically been quantified through the portosystemic pressure gradient (PSG), which is clinically estimated by hepatic venous pressure gradient (HVPG); however, PSG alone does not predict a given patient's clinical trajectory regarding the Baveno stage of cirrhosis. We hypothesize that a patient's PSG sensitivity to venous remodeling could explain disparate disease trajectories. METHODS: We created a computational model of the portal system in the context of worsening liver disease informed by physiologic measurements from the field of portal hypertension. We simulated progression of clinical complications, HVPG, and transjugular intrahepatic portosystemic shunt placement while only varying a patient's likelihood of portal venous remodeling. RESULTS: Our results unify hemodynamics, venous remodeling, and the clinical progression of liver disease into a mathematically consistent model of portal hypertension. We find that by varying how sensitive patients are to create venous collaterals with rising PSG we can explain variation in patterns of decompensation for patients with liver disease. Specifically, we find that patients who have higher proportions of portosystemic shunting earlier in disease have an attenuated rise in HVPG, delayed onset of ascites, and less hemodynamic shifting after transjugular intrahepatic portosystemic shunt placement. DISCUSSION: This article builds a computational model of portal hypertension which supports that patient-level differences in venous remodeling may explain disparate clinical trajectories of disease.

Personality Traits in Patients With Cirrhosis Are Different From Those of the General Population and Impact Likelihood of Liver Transplantation.

BACKGROUND: Personality traits influence clinical outcomes in chronic diseases, but their impact in cirrhosis is unknown. We studied the personality of patients with cirrhosis undergoing liver transplant (LT) evaluation and determined their correlation to clinical outcomes. METHODS: A multicenter' prospective study of adult patients undergoing LT evaluation was performed from January 2018 to October 2019. The "Big Five" personality traits of conscientiousness, extraversion, openness, neuroticism, and agreeableness plus agency were assessed with the Midlife Development Inventory Personality Scale and compared with the general population. Frailty was assessed with the Liver Frailty Index. RESULTS: Two hundred sixty-three LT candidates were enrolled. Twenty-four percent had hepatitis C virus, 25% nonalcoholic steatohepatitis, and 25% ethyl alcohol (mean model for end-stage liver disease = 15.7). Compared with the general population, LT candidates had higher openness (3.1 versus 2.9; P < 0.001), extraversion (3.2 versus 3.1; P < 0.001), agreeableness (3.5 versus 3.4; P = 0.04), agency (2.9 versus 2.6; P < 0.001), neuroticism (2.2 versus 2.1; P = 0.001), and lower conscientiousness (3.3 versus 3.4; P = 0.007). Patients with higher conscientiousness were more likely to receive an LT (HR = 2.76; P = 0.003). CONCLUSIONS: Personality traits in LT candidates differ significantly from the general population, with higher conscientiousness associated with a higher likelihood of receiving a transplant.

The answer at our fingertips: Volume status in cirrhosis determined by machine learning and pulse oximeter waveform.

OBJECTIVE: The objective of our study was to determine if the waveform from a simple pulse oximeter-like device could be used to accurately assess intravascular volume status in cirrhosis. METHODS: Patients with cirrhosis underwent waveform recording as well as serum brain natriuretic peptide (BNP) on the day of their cardiac catheterization where invasive cardiac pressures were measured. Waveforms were processed to generate features for machine learning models in order to predict the filling pressures (regression) or to classify the patients as volume overloaded or not (defined as an LVEDP>15). RESULTS: Nine of 26 patients (35%) had intravascular volume overload. Regression analysis using PPG features (R2  = 0.66) was superior to BNP (R2  = 0.22). Linear discriminant analysis correctly classified patients with an accuracy of 78%, sensitivity of 60%, positive predictive value of 90%, and an AUROC of 0.87. CONCLUSIONS: Machine learning-enhanced analysis of pulse ox waveforms can estimate intravascular volume overload with a higher accuracy than conventionally measured BNP.

Predicting mortality among patients with liver cirrhosis in electronic health records with machine learning.

OBJECTIVE: Liver cirrhosis is a leading cause of death and effects millions of people in the United States. Early mortality prediction among patients with cirrhosis might give healthcare providers more opportunity to effectively treat the condition. We hypothesized that laboratory test results and other related diagnoses would be associated with mortality in this population. Our another assumption was that a deep learning model could outperform the current Model for End Stage Liver disease (MELD) score in predicting mortality. MATERIALS AND METHODS: We utilized electronic health record data from 34,575 patients with a diagnosis of cirrhosis from a large medical center to study associations with mortality. Three time-windows of mortality (365 days, 180 days and 90 days) and two cases with different number of variables (all 41 available variables and 4 variables in MELD-NA) were studied. Missing values were imputed using multiple imputation for continuous variables and mode for categorical variables. Deep learning and machine learning algorithms, i.e., deep neural networks (DNN), random forest (RF) and logistic regression (LR) were employed to study the associations between baseline features such as laboratory measurements and diagnoses for each time window by 5-fold cross validation method. Metrics such as area under the receiver operating curve (AUC), overall accuracy, sensitivity, and specificity were used to evaluate models. RESULTS: Performance of models comprising all variables outperformed those with 4 MELD-NA variables for all prediction cases and the DNN model outperformed the LR and RF models. For example, the DNN model achieved an AUC of 0.88, 0.86, and 0.85 for 90, 180, and 365-day mortality respectively as compared to the MELD score, which resulted in corresponding AUCs of 0.81, 0.79, and 0.76 for the same instances. The DNN and LR models had a significantly better f1 score compared to MELD at all time points examined. CONCLUSION: Other variables such as alkaline phosphatase, alanine aminotransferase, and hemoglobin were also top informative features besides the 4 MELD-Na variables. Machine learning and deep learning models outperformed the current standard of risk prediction among patients with cirrhosis. Advanced informatics techniques showed promise for risk prediction in patients with cirrhosis.

Black Patients With Cirrhosis Have Higher Mortality and Lower Transplant Rates: Results From a Metropolitan Cohort Study.

BACKGROUND AND AIMS: Estimates of racial disparity in cirrhosis have been limited by lack of large-scale, longitudinal data, which track patients from diagnosis to death and/or transplant. APPROACH AND RESULTS: We analyzed a large, metropolitan, population-based electronic health record data set from seven large health systems linked to the state death registry and the national transplant database. Multivariate competing risk analyses, adjusted for sex, age, insurance status, Elixhauser score, etiology of cirrhosis, HCC, portal hypertensive complication, and Model for End-Stage Liver Disease-Sodium (MELD-Na), examined the relationship between race, transplant, and cause of death as defined by blinded death certificate review. During the study period, 11,277 patients met inclusion criteria, of whom 2,498 (22.2%) identified as Black. Compared to White patients, Black patients had similar age, sex, MELD-Na, and proportion of alcohol-associated liver disease, but higher comorbidity burden, lower rates of private insurance, and lower rates of portal hypertensive complications. Compared to White patients, Black patients had the highest rate all-cause mortality and non-liver-related death and were less likely to be listed or transplanted (P < 0.001 for all). In multivariate competing risk analysis, Black patients had a 26% increased hazard of liver-related death (subdistribution HR, 1.26; 95% CI, [1.15-1.38]; P < 0.001). CONCLUSIONS: Black patients with cirrhosis have discordant outcomes. Further research is needed to determine how to address these real disparities in the field of hepatology.

Amalgamation of cloud-based colonoscopy videos with patient-level metadata to facilitate large-scale machine learning.

Background and study aims Storage of full-length endoscopic procedures is becoming increasingly popular. To facilitate large-scale machine learning (ML) focused on clinical outcomes, these videos must be merged with the patient-level data in the electronic health record (EHR). Our aim was to present a method of accurately linking patient-level EHR data with cloud stored colonoscopy videos. Methods This study was conducted at a single academic medical center. Most procedure videos are automatically uploaded to the cloud server but are identified only by procedure time and procedure room. We developed and then tested an algorithm to match recorded videos with corresponding exams in the EHR based upon procedure time and room and subsequently extract frames of interest. Results Among 28,611 total colonoscopies performed over the study period, 21,170 colonoscopy videos in 20,420 unique patients (54.2 % male, median age 58) were matched to EHR data. Of 100 randomly sampled videos, appropriate matching was manually confirmed in all. In total, these videos represented 489,721 minutes of colonoscopy performed by 50 endoscopists (median 214 colonoscopies per endoscopist). The most common procedure indications were polyp screening (47.3 %), surveillance (28.9 %) and inflammatory bowel disease (9.4 %). From these videos, we extracted procedure highlights (identified by image capture; mean 8.5 per colonoscopy) and surrounding frames. Conclusions We report the successful merging of a large database of endoscopy videos stored with limited identifiers to rich patient-level data in a highly accurate manner. This technique facilitates the development of ML algorithms based upon relevant patient outcomes.

Liver-related mortality is similar among men and women with cirrhosis.

BACKGROUND & AIMS: Sex-based differences are known to significantly contribute to outcomes in patients with chronic liver diseases; however, the role of patient sex in cirrhosis is unclear. We aimed to study the relationship between patient sex and cirrhosis. METHODS: We analyzed a cohort of 20,045 patients with cirrhosis using a Chicago-wide electronic health record database that was linked with the United Network for Organ Sharing and cause of death data from the state death registry. Adjusted Cox survival analyses and competing risk analyses were performed to obtain subdistribution hazard ratios (HRs) for liver-related cause of death. RESULTS: Female and male patients had similar age, racial distribution, insurance status, and comorbidity status by Elixhauser score. Females had higher rates of cholestatic liver disease (17.1% vs. 6.2%, p <0.001) and non-alcoholic steatohepatitis (29.8% vs. 21.2%, p <0.001) than males. They were less likely to have portal hypertensive complications and had lower peak MELD-Na scores during follow-up. Female sex was associated with a decreased hazard of all-cause mortality (adjusted HR 0.85; 95% CI 0.80-0.90). This effect was attenuated when liver-related mortality was examined (subdistribution HR 0.93; 95% CI 0.87-1.00). No significant difference was noted for women who were 'ever-listed' in competing risk analyses for either all-cause mortality (subdistribution HR 1.09; 95% CI 0.88-1.35) or liver-related death (subdistribution HR 1.12; 95% CI 0.87-1.43), despite lower rates of listing (7.5% vs. 9.8%; p <0.001) and transplant (3.5% vs. 5.2%; p <0.001). CONCLUSIONS: In this longitudinal study of patients with cirrhosis, female sex was associated with a survival advantage likely driven by lower rates of non-liver-related death. Women were not at an increased risk of liver-related death despite lower rates of listing and transplantation. LAY SUMMARY: Patient sex is an important contributor in many chronic diseases, including cirrhosis. Prior studies have suggested that female sex is associated with worse outcomes. We analyzed a cohort of 20,045 patients with cirrhosis using a Chicago-wide electronic health record database. Using multivariate competing risk analyses, we found that female sex in cirrhosis is actually associated with a lower risk of all-cause mortality and has no association with liver-related mortality. Our findings are novel because we show that women with cirrhosis have a similar risk of liver-related death as their male counterparts, despite lower rates of listing and transplantation.

A Comprehensive Review of Outcome Predictors in Low MELD Patients.

Risk scoring for patients with cirrhosis has evolved greatly over the past several decades. However, patients with low Model for End-Stage Liver Disease-Sodium scores still suffer from liver-related morbidity and mortality. Unfortunately, it is not clear which of these low Model for End-Stage Liver Disease-Sodium score patients would benefit from earlier consideration of liver transplantation. This article reviews the literature of risk prediction in patients with cirrhosis, identifies which patients may benefit from earlier interventions, such as transplantation, and proposes directions for future research.

Patients With Persistently Low MELD-Na Scores Continue to Be at Risk of Liver-related Death.

BACKGROUND: The vast majority of patients with cirrhosis have low Model for End-Stage Liver Disease-Sodium (MELD-Na) scores; however, the ability for the MELD-Na score to predict patient outcomes at low scores is unclear. METHODS: Adult patients in a multicenter, Chicago-wide database of medical records with International Classification of Disease, Ninth Edition codes of cirrhosis and without a history of hepatocellular carcinoma were included. Records were linked with the state death registry, and death certificates were manually reviewed. Deaths were classified as "liver-related," "non-liver-related," and "non-descript" as adjudicated by a panel comprised of a transplant surgeon, a hepatologist, and an internist. A sensitivity analysis was performed where patients with hepatocellular carcinoma were included. RESULTS: Among 7922 identified patients, 3999 patients had MELD-Na scores that were never higher than 15. In total, 2137 (27%) patients died during the study period with higher mortality rates for the patients in the high MELD-Na group (19.4 (41.6%) versus 4.1 (12.6%) per 100 person-y, P < 0.001). The high MELD-Na group died of a liver-related cause in 1142 out of 1632 (70%) as compared to 240 out of 505 (47.5%) deaths in the low MELD-Na group. There was no difference in the distribution of subcategory of liver-related death between low and high MELD-Na groups. Among subclassification of liver-related deaths, the most common cause of death was "Infectious" in both groups. CONCLUSIONS: Despite persistently low MELD-Na scores, patients with cirrhosis still experience high rates of liver-related mortality.

The Hispanic Paradox in Patients With Liver Cirrhosis: Current Evidence From a Large Regional Retrospective Cohort Study.

BACKGROUND: Despite lower socioeconomic status, Hispanics in the United States paradoxically maintain equal or higher average survival rates compared to non-Hispanic Whites (NHW). METHODS: We used multivariable Cox regression to assess whether this "Hispanic paradox" applies to patients with liver cirrhosis using a retrospective cohort of twenty 121 patients in a Chicago-wide electronic health record database. RESULTS: Our study population included 3279 (16%) Hispanics, 9150 (45%) NHW, 4432 (22%) African Americans, 529 (3%) Asians, and 2731 (14%) of other races/ethnic groups. Compared to Hispanics, NHW (hazard ratio [HR] 1.26; 95% confidence interval [CI], 1.16-1.37), African American (HR 1.26; 95% CI, 1.15-1.39), and other races/ethnic groups (HR 1.55; 95% CI, 1.40-1.71) had an increased risk of death despite adjustment for age, sex, insurance status, etiology of cirrhosis, and comorbidities. On stratified analyses, a mortality advantage for Hispanics compared to NHW was seen for alcohol cirrhosis (HR for NHW 1.35; 95% CI, 1.19-1.52), hepatitis B (HR for NHW 1.35; 95% CI, 0.98-1.87), hepatitis C (HR for NHW 1.21; 95% CI, 1.06-1.38), and nonalcoholic steatohepatitis (HR for NHW 1.14; 95% CI, 0.94-1.39). There was no advantage associated with Hispanic race over NHW in cases of hepatocellular carcinoma or cholestatic liver disease. CONCLUSIONS: Hispanic patients with cirrhosis experience a survival advantage over many other racial groups despite adjustment for multiple covariates.