The goal of this observational study is to learn if new muscle attenuation cut-offs for the definition of myosteatosis can better predict survival in patients with cirrhosis. The main questions it aims to answer are:

• Can these new muscle attenuation cut-offs more accurately diagnose myosteatosis (fat infiltration in muscle mass ) in cirrhosis?
• Do these cut-offs work effectively across different patient groups, including men and women, those with obesity, and those with fluid retention in the abdomen?

Researchers will compare patients with cirrhosis from a retrospective Canadian cohort to those from a prospective Italian cohort to see if the new cut-offs predict survival better than existing ones**.

Due to the retrospective nature of the study, no action is required from participants.

Detailed Description This observational study seeks to redefine muscle attenuation cut-offs to better predict mortality in patients with cirrhosis, particularly focusing on myosteatosis-a condition characterized by the pathological accumulation of fat within skeletal muscles. Myosteatosis is a critical factor in cirrhosis, associated with poor outcomes, including increased risk of overt hepatic encephalopathy, prolonged hospitalization, post-transplant complications, and reduced survival. The current cut-offs for diagnosing myosteatosis are derived from studies in oncology populations, which may not be appropriate for cirrhosis patients due to specific complications like fluid retention (ascites) and differences in fat and muscle distribution between genders.

Background and Rationale Cirrhosis patients often experience muscle changes due to chronic inflammation and metabolic disturbances, leading to muscle fat infiltration or myosteatosis. This study builds on previous research by Ebadi et al., who proposed sex-specific cut-offs (<33 HU for men and <28 HU for women) for diagnosing myosteatosis based on muscle attenuation values measured via CT scans. However, these cut-offs were derived from a retrospective cohort and have not been validated externally. There is a clear need for liver-specific cut-offs that account for the unique characteristics of cirrhosis, including the impact of fluid retention on BMI and muscle attenuation measurements.

Study Objectives

The primary objective is to validate new muscle mass attenuation cut-offs for the diagnosis of myosteatosis using L3 CT scan evaluations to predict survival and outcomes in patients with cirrhosis. Secondary objectives include:

• Assessing the accuracy of these cut-offs across different patient groups (e.g., by gender, obesity status, and presence of ascites).
• Evaluating the prognostic significance of redefined myosteatosis when combined with sarcopenia (muscle wasting).

Study Design

The study involves a post-hoc analysis of data from two large cohorts:

1. Retrospective Cohort (Canada): This cohort consists of 1,104 patients evaluated for liver transplantation (LT) at the University of Alberta Hospital between January 2000 and August 2021. After applying exclusion criteria (e.g., absence of cirrhosis, multi-organ transplantation), 863 patients were included in the final analysis.
2. Prospective Cohort (Italy): This cohort includes 447 patients from 26 Italian centers, prospectively enrolled between January 2019 and January 2020, with cirrhosis who underwent abdominal CT scans. After excluding patients with incomplete data, 433 patients were included in the analysis.

Data Collection

Data were collected at the time of enrollment and included:

• Demographic characteristics: Age, sex
• Anthropometric measurements: Height, weight, BMI
• Cirrhosis etiology: Hepatitis C, alcohol-related liver disease, non-alcoholic steatohepatitis (NASH), etc.
• Comorbidities: Diabetes, hypertension, dyslipidemia
• Liver complications: Refractory ascites, variceal bleeding, hepatic encephalopathy, hepatocellular carcinoma (HCC)
• Liver classification and blood tests: Child-Pugh score, MELD score, MELD-Na score, albumin, sodium, creatinine, INR, bilirubin, hemoglobin, etc.
• Radiological evaluation: Muscle mass and quality assessments via CT scans at the L3 or L3/L4 vertebrae level, including muscle attenuation and skeletal muscle index (SMI).

Radiological Assessment Muscle mass and attenuation were measured using CT scans, with radiodensity (measured in Hounsfield Units, HU) being the primary indicator of myosteatosis. The HU thresholds for assessing skeletal muscle ranged from -29 to +150 HU. The muscle area was adjusted for height to obtain the L3 or L3/L4 skeletal muscle index (SMI). Sarcopenia was diagnosed using validated SMI cut-off values (<50 cm²/m² for men and <39 cm²/m² for women). The study specifically examines muscle attenuation in relation to BMI, gender, and ascites to refine the diagnostic criteria for myosteatosis.

Patient Classification

Patients were classified into four groups based on the presence of myosteatosis and sarcopenia:

1. No muscle changes: No myosteatosis, no sarcopenia.
2. Isolated myosteatosis: Myosteatosis without sarcopenia.
3. Isolated sarcopenia: Sarcopenia without myosteatosis.
4. Combined: Both myosteatosis and sarcopenia.

Statistical Analysis

To validate the new cut-offs, the study will:

• Perform univariate analyses on variables likely associated with mortality, such as muscle attenuation and muscle mass.
• Conduct multivariate analyses to assess the impact of confounding variables.
• Use Receiver Operating Characteristic (ROC) curves, stratified by sex, BMI, and ascites, to identify the muscle attenuation values most strongly associated with mortality.
• Assess non-causal concordance (kappa index) between CT scan measurements across the two cohorts by exchanging and reanalyzing a subset of data.

The study aims to confirm or refine the cut-offs proposed by Ebadi et al. and evaluate their adequacy in different cirrhotic populations, stratified by disease severity and history of decompensation.

Follow-Up and Outcomes In the retrospective cohort, patients were followed for an average of 24 ± 35 months, with outcomes such as death and liver transplantation recorded. In the prospective cohort, patients were followed for 12 months post-enrollment, with outcomes including survival, hospitalizations, and liver-related complications. Data from these follow-ups will be crucial in validating the predictive power of the new muscle attenuation cut-offs.

Ethical Considerations

The study protocol has been reviewed and approved by the Institutional Review Board of the University of Alberta (approval number: Pro00066572) for the retrospective cohort and by the Ethical Committee of the Rome's Center (EC n° 94/19 30/01/19) for the prospective cohort. Each participating center provided its own ethical approval, and all patients gave informed consent for their participation.

Conclusion This study aims to provide a more accurate and tailored method for diagnosing myosteatosis in patients with cirrhosis by refining muscle attenuation cut-offs. By validating these new thresholds, the study hopes to improve the prediction of patient outcomes and aid in the management and treatment strategies for cirrhosis, ultimately leading to better patient care and survival rates.

Inclusion Criteria:

Retrospective cohort:

• Clinical, radiological and/or histological diagnosis of cirrhosis
• Abdominal CT-scan obtained as part of the LT evaluation

Prospective cohort:

• Clinical, radiological and/or histological diagnosis of cirrhosis
• Age 40-75 years
• Abdominal CT-scan

Exclusion Criteria:

Retrospective cohort:

• Multi-organ transplantation
• Re-transplantation

Prospective cohort:

• Patient on LT waiting lists
• Hepatocellular carcinoma (HCC)
• History of LT
• Concomitant neuromuscular disease
• Current malignancy other than non-melanocytic skin cancer
• History of serious extrahepatic disease
• HIV infection.

In the retrospective cohort, 1104 patients were evaluated for LT between January 2000 and August 2021 in the University of Alberta Hospital (Canada), of whom 241 were excluded for not meeting the inclusion and exclusion criteria, so 863 patients were considered for the analysis.

In the prospective cohort, 447 patients were prospectively enrolled from 26 Italian centres, 14 of whom were excluded due to incomplete data, leaving 433 patients for the analysis.

Given the nature of the databases and the sharing of the project with another centre, we do not believe that individual data can be shared. If necessary, if requested by any author, we can agree on individual data sharing