Towards a new definition of decompensated cirrhosis

A cholestatic pattern predicts major liver-related outcomes in patients with non-alcoholic fatty liver disease

BACKGROUND & AIMS

NAFLD patients usually have an increase in AST/ALT levels, but cholestasis can also be observed. We aimed to assess in subjects with NAFLD the impact of the (cholestatic) C pattern on the likelihood of developing major liver-related outcomes (MALO).

METHODS

Five hundred and eighty-two consecutive patients with biopsy-proven NAFLD or a clinical diagnosis of NAFLD-related compensated cirrhosis were classified as hepatocellular (H), C and mixed (M) patterns, by using the formula (ALT/ALT Upper Limit of Normal-ULN)/(ALP/ALP ULN). MALO were recorded during follow-up. An external cohort of 1281 biopsy-proven NAFLD patients was enrolled as validation set.

RESULTS

H, M and C patterns were found in 153 (26.3%), 272 (46.7%) and 157 (27%) patients respectively. During a median follow-up of 78 months, only 1 (0.6%) patient with H pattern experienced MALO, whilst 15 (5.5%) and 38 (24.2%) patients in M and C groups had MALO. At multivariate Cox regression analysis, age >55 years (HR 2.55, 95% CI 1.17-5.54; p = .01), platelets <150 000/mmc (HR 0.14, 95% CI 0.06-0.32; p < .001), albumin <4 g/L(HR 0.62, 95% CI 0.35-1.08; p = .09), C versus M pattern (HR 7.86, 95% CI 1.03-60.1; p = .04), C versus H pattern(HR 12.1, 95% CI 1.61-90.9; p = .01) and fibrosis F3-F4(HR 35.8, 95% CI 4.65-275.2; p < .001) were independent risk factors for MALO occurrence. C versus M pattern(HR 14.3, 95% CI 1.90-105.6; p = .008) and C versus H pattern (HR 15.6, 95% CI 2.10-115.1; p = .0068) were confirmed independently associated with MALO occurrence in the validation set. The immunohistochemical analysis found a significantly higher prevalence of moderate-high-grade ductular metaplasia combined with low-grade ductular proliferation in C pattern when compared with the biochemical H pattern. Gene expression analysis showed a lower expression of NR1H3, RXRα and VCAM1 in patients with the C pattern.

CONCLUSIONS

The presence of a cholestatic pattern in patients with NAFLD predicts a higher risk of MALO independently from other features of liver disease.

Clinical Characteristics and Outcomes of Decompensated Cirrhosis Patients Admitted to Hospitals With Acute Pulmonary Embolisms: A Nationwide Analysis

Introduction: Cirrhosis is a significant cause of mortality and morbidity worldwide. Recent studies suggested that cirrhosis is associated with an increased risk of venous thromboembolism (VTE), which disproves the old belief that chronic liver disease coagulopathy is considered protective against VTE. We conducted a retrospective study which is to our knowledge the first of its kind to assess clinical characteristics and outcomes of decompensated cirrhosis (DC) patients admitted with acute pulmonary embolism (APE).

Methodology: We used the National Inpatient Sample database for the years 2016-2019. All adults admitted to the hospitals with a primary diagnosis of APE were included. Patients less than 18 years old, missing race, gender, or age were excluded. Patients were divided into two groups, either having DC or not. A multivariate logistic regression model was built by using only variables associated with the outcome of interest on univariable regression analysis at P < 0.05.

Results: 142 million discharges were included in the NIS database between the years 2016 and 2019, of which 1,294,039 met the study inclusion criteria, 6,200 patients (0.5%) had DC. For adult patients admitted to the hospitals with APE, odds of inpatient all-cause mortality were higher in the DC group than in patients without DC; OR of 1.996 (95% CI, 1.691-2.356, P-value < 0.000). Also, vasopressor use, mechanical ventilation, and cardiac arrest were more likely to occur in the DC group, OR of 1.506 (95% CI, 1.254-1.809, P-value < 0.000), OR of 1.479 (95% CI, 1.026-2.132, P-value 0.036), OR of 1.362 (95% CI, 1.050-1.767, P-value 0.020), respectively. In addition, DC patients tend to have higher total hospital charges and longer hospital length of stay, coefficient of 14521 (95% CI, 6752-22289, P-value < 0.000), and a coefficient of 1.399 (95% CI, 0.848-1.950, P-value < 0.000), respectively.

Conclusion: This study demonstrates that DC is a powerful predictor of worse hospital outcomes in patients admitted with APE. An imbalance between clotting factors and natural anticoagulants produced by the liver is believed to be the primary etiology of thrombosis in patients with DC. The burden of APE can be much more catastrophic in cirrhotic than in non-cirrhotic patients; therefore, those patients require closer monitoring and more aggressive treatment.