Higher subcutaneous adipose tissue radiodensity is associated with an increased risk of the onset of Acute-on-Chronic Liver Failure

doi:10.21203/rs.3.rs-4881664/v1

Download PDF

Research Article

Higher subcutaneous adipose tissue radiodensity is associated with an increased risk of the onset of Acute-on-Chronic Liver Failure

https://doi.org/10.21203/rs.3.rs-4881664/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Body composition plays a significant role in the development and progression of disease in patients with acute decompensation of cirrhosis. Previous studies have identified the psoas muscle index (PMI) as an independent risk factor for predicting one-year mortality in patients with acute-on-chronic liver failure (ACLF). However, the relationship between adipose tissue, another component of body composition, and disease progression in patients with ACLF remains unclear. This study aimed to investigate the association between computed tomography (CT)-derived adipose tissue characteristics and the occurrence of ACLF and to develop a predictive model for ACLF.

Methods

This study included 343 adult patients with acute decompensation of cirrhosis who underwent abdominal CT examinations at our center between 2018 and 2022. Clinical laboratory test results and CT-derived adipose tissue characteristics were analyzed. Disease progression within 7 days was monitored, and a predictive model for ACLF occurrence incorporating adipose tissue information was developed.

Results

A total of 42 patients progressed to ACLF within 7 days. Significant differences in adipose tissue density were observed between the ACLF occurrence and non-occurrence groups. The subcutaneous adipose tissue (SAT) radiodensity in the ACLF occurrence group was − 75.79 (SD ± 16.35) HU, compared to -86.08 (SD ± 14.35) HU in the non-occurrence group. Ultimately, the predictive model comprising four variables—direct bilirubin (DBIL), prothrombin activity (PTA), sodium ion (NA+), and SAT radiodensity—demonstrated an area under the curve (AUC) of 0.93 and 0.91 in the training and test sets, respectively.

Conclusions

Patients with high SAT radiodensity are at a higher risk of developing ACLF. The newly established model can accurately identify individuals at higher risk of ACLF among patients with acute decompensation of cirrhosis, thereby facilitating early clinical intervention and improving patient survival rates.

Acute-on-chronic liver failure

Subcutaneous adipose tissue

Visceral adipose tissue

Acute decompensation of cirrhosis

Radiodensity

Acute-on-chronic liver failure (ACLF) is a complex syndrome characterized by the acute decompensation of cirrhosis, presenting with severe clinical symptoms, rapid progression, and high short-term mortality rates(1). Liver transplantation remains the only definitive and effective treatment; however, its widespread application is limited by the scarcity of donor livers and high costs. Early and accurate prediction of clinical progression is crucial for optimizing the timing of liver transplantation and improving patient survival rates(2).

Over the past decade, research has increasingly demonstrated a correlation between body composition assessment and the onset and prognosis of chronic liver diseases(3). Computed tomography (CT) provides specific quantitative information regarding body composition, particularly adipose tissue, which can be categorized into subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT). Previous studies have indicated that the psoas muscle index (PMI) can predict long-term mortality in patients with ACLF(4). However, no studies have yet explored the relationship between the imaging parameters of adipose tissue, which reflect biological tissue characteristics, and ACLF. Existing literature suggests that imaging parameters of adipose tissue, including area, density, and normalized index, are significantly associated with disease progression in patients with cirrhosis(5).

This study aimed to analyze whether CT imaging-based adipose tissue parameters are associated with the occurrence and progression of ACLF in patients with acute decompensation of cirrhosis and to identify body composition characteristics of patients at high risk of developing ACLF. Based on these findings, we sought to develop a predictive model for ACLF that includes adipose tissue imaging parameters.

2.1 Patients

This study collected data from patients admitted to the First Hospital of Jilin University for acute decompensation of cirrhosis between September 2018 and October 2022 (Fig. 1). The inclusion criteria were: (1) age between 18 and 80 years, (2) acute decompensation of cirrhosis at the time of admission (including ascites, hepatic encephalopathy, upper gastrointestinal bleeding, bacterial infection, or any combination of these conditions(6)), and (3) underwent abdominal CT and laboratory examinations during the acute decompensation state. The exclusion criteria were: (1) presence of severe diseases in other systems during hospitalization, (2) incomplete laboratory examination data, (3) poor-quality CT images, and (4) hepatocellular carcinoma beyond the Milan criteria. Clinical laboratory examination data and CT images were collected at admission, and disease progression within 7 days of admission was recorded. The study’s primary endpoint was the progression of patients from acute decompensation of cirrhosis to ACLF.

All patients’ clinical data were retrieved from electronic medical records, and the original CT images were downloaded from the Picture Archiving and Communication System (PACS). This study was approved by the Ethics Committee of the First Hospital of Jilin University. Due to the study’s retrospective nature, informed consent was not required.

2.2 CT image analysis

Adipose tissue was assessed by analyzing abdominal CT images at the L3 vertebral level. SAT and VAT were defined as the areas outside and inside the abdominal muscle wall, respectively. These tissue regions were identified using RIASEG software (version 1.0; RIASEG Technologies, Shenzhen, China), and an automatic recognition model was constructed based on a 2D Attention UNet network using a combination of artificial intelligence recognition and manual modification (Fig. 2). The CT images of patients at the L3 vertebra were annotated for SAT and VAT regions of interest (ROIs). The L3 level was selected because the fat area obtained from single-slice CT images at this level correlates best with the total body fat mass(7). These tissue areas were normalized by dividing by the square of the height in meters and described as subcutaneous adipose tissue index (SATI) and visceral adipose tissue index (VATI). The radiodensities of the SAT and VAT were measured as the mean value of the entire ROI. Two radiologists performed image annotation. After all annotations were completed, 30 patients were randomly selected by both radiologists to calculate the interobserver coefficient.

2.3 Study design

After a comprehensive summarization of clinical laboratory examination data and adipose tissue imaging parameters obtained from CT scans, the patients were divided into two groups based on whether they progressed to ACLF within 7 days. Differential analysis was conducted, and variables showing intergroup differences were included in a single-factor logistic regression analysis. Variables with a p-value < 0.05 were further analyzed using multivariable logistic regression analysis. Finally, variables with a p-value < 0.05 were selected as indicators for model establishment. The data were divided into training and testing sets in a ratio of 7:3, and a predictive model for ACLF incorporating adipose tissue imaging parameters was developed. Model performance was evaluated using the area under the curve (AUC) value, sensitivity, specificity, accuracy, precision, and F1 score. These evaluation metrics were used to comprehensively assess the model’s predictive performance and clinical utility.

2.4 Statistical analysis

Continuous variables were presented as mean ± standard deviation and intergroup differences were compared using the independent samples t-test or Mann-Whitney U test. Descriptive statistics for categorical variables were presented as frequencies, and intergroup comparisons were performed using the Chi-square test. All statistical tests were two-tailed, and p-values < 0.05 were considered statistically significant. This study focused on the progression of acute decompensation of cirrhosis to ACLF and analyzed the association between collected clinical laboratory examination data, adipose tissue imaging parameters, and ACLF occurrence. Differential analysis was first conducted for all variables, and significant indicators were included in a single-factor logistic regression analysis. Variables with a p-value < 0.05 in the single-factor analysis were further analyzed using multivariable logistic regression analysis. Finally, variables with a p-value < 0.05 in the multivariable regression analysis were considered independent predictors of ACLF occurrence in patients with acute decompensation of cirrhosis, and a predictive model for the risk of developing ACLF was developed based on these variables.

3.1 Characteristics of the deriving cohort

This study included 342 patients with acute decompensation of cirrhosis. Table 1 summarizes the clinical characteristics of the enrolled patients. The mean age of the patients was 56.23 (SD ± 11.31) years, with males constituting 64.6% and females 35.4%. Hepatitis B was the most common etiology (41.5%), followed by alcoholic hepatitis (12.0%). Of these patients, 48 patients (14.3%) progressed to ACLF. Statistical analysis revealed that the mean body mass index (BMI) of patients who progressed to ACLF was 18.35 (SD ± 10.90), whereas the mean BMI of patients who did not progress was 24.16 (SD ± 4.05). These preliminary results indicated that patients who progressed to ACLF had lower BMIs and poorer nutritional statuses.

3.2 Patients with high SAT radiodensity are at a higher risk of developing ACLF

To explore the laboratory and body composition characteristics of patients at higher risk of developing ACLF, we included 26 clinical, laboratory, and imaging parameters in a single-factor logistic regression analysis. This analysis identified 13 variables that remained significant predictors of short-term disease occurrence, including the radiodensity of SAT, aspartate aminotransferase (AST), alanine aminotransferase (ALT), cholinesterase (CHE), albumin (ALB), total bilirubin (TBIL), direct bilirubin (DBIL), activated partial thromboplastin time (APTT), prothrombin time (PT), international normalized ratio (INR), prothrombin activity (PTA), serum creatinine (SCR), and sodium ion (NA+). Subsequently, the results obtained from the single-factor logistic regression analysis were subjected to a multivariable logistic regression analysis. The multivariable logistic regression analysis revealed that DBIL, PTA, NA+, andradiodensity of SAT were significantly associated with ACLF occurrence. The radiodensity of SAT in the ACLF occurrence group was − 75.79 (SD ± 16.35) HU compared to -86.08 (SD ± 14.35) HU in the non-occurrence group (Fig. 3). These findings indicate that patients with higher radiodensity of SAT are at an increased risk of developing ACLF, according to adipose tissue imaging parameters.

3.3 The association between VAT characteristics and the occurrence of ACLF

Statistical analysis showed that the VAT radiodensity of patients in the ACLF occurrence group was − 69.81 (SD ± 9.67) HU and − 71.95 (SD ± 10.77) HU in the non-occurrence group. However, neither single-factor nor multivariable logistic regression analyses found these VAT values to be statistically significant, thus excluding them as predictive factors for disease occurrence risk.

3.4 Variable selection and prognostic model construction

In our study, imaging variables after undergoing multivariable regression analysis revealed that high SAT radiodensity was an independent predictive factor of ACLF occurrence in patients with acute decompensation of cirrhosis. Consequently, DBIL, PTA, NA+, and SAT radiodensity were selected for model development. The data were divided into training and testing sets with a 7:3 ratio, and a new ACLF occurrence risk prediction model was developed. The new model achieved an accuracy of 0.91, a precision of 0.75, a recall of 0.60, and an F1 score of 0.67. The AUC was 0.93 and 0.91 for the training and testing sets, respectively (Fig. 4). The newly established model demonstrated strong performance and effectively identified individuals at a higher risk of ACLF among patients with acute decompensation of cirrhosis.

This study investigated the relationship between adipose tissue imaging parameters and the risk of developing ACLF in patients with acute decompensation of cirrhosis. The results indicate that individuals with higher SAT radiodensity are more susceptible to developing ACLF in the context of acute decompensation of cirrhosis(8). This finding introduces a novel predictive factor for ACLF occurrence. Similar associations have also been observed in prognostic studies involving patients with cirrhosis. Therefore, this study proposes SAT radiodensity as a new independent predictive factor for ACLF. Furthermore, a predictive model integrating clinical laboratory examinations and SAT density was developed to identify individuals at higher risk of ACLF among those with acute decompensation of cirrhosis. This model aims to facilitate early clinical intervention and improve patient survival rates.

The pathogenesis of ACLF is exceedingly complex, making it challenging to accurately and comprehensively screen patients at higher risk of occurrence based solely on clinical indicators. Numerous studies have investigated the liver morphological indicators and quantitative parameters of skeletal muscles in patients with cirrhosis to establish predictive models for ACLF occurrence and prognosis(4, 10). However, SAT— an important predictive factor reflecting patient body composition and nutritional status—has not been incorporated in previous predictive models.

Previous studies have indicated that sarcopenia is an independent predictor of mortality in patients with ACLF(10). However, the significance of adipose tissue in ACLF occurrence and progression remains unclear. Total body adipose tissue comprises VAT and SAT, each potentially playing an independent role, necessitating separate investigations of their relationships with ACLF. The radiodensity of adipose tissue can be objectively measured in HU using CT scans. The mean radiodensity of adipose tissue obtained from CT scans has been introduced as an indirect surrogate marker of adipose tissue quality(11). The imaging parameters of SAT obtained from CT scans serve as good indicators for assessing the body composition and nutritional status of patients with acute decompensation of cirrhosis. HU not only represents triglyceride storage in adipose tissue but also includes other aspects of tissue structure, such as water, blood, and residual fat cell components(11).

We believe that SAT, to some extent, reflects the nutritional status of patients with cirrhosis, which is crucial for their ability to withstand disease progression. Unlike BMI, a traditional indicator for assessing body composition, which can be significantly influenced by the general condition of patients, such as the presence of ascites or the use of diuretics(12), adipose tissue parameters obtained from CT scans are not affected by these conditions. Furthermore, patients with cirrhosis typically undergo CT scans, making it convenient to obtain adipose tissue parameters. Additionally, while contrast-enhanced CT scans can also extract adipose tissue parameters, they are unsuitable for patients with acute decompensation of cirrhosis. When measuring SAT density, the values obtained from plain and contrast-enhanced CT scans exhibited high consistency(13). However, contrast-enhanced scans impose a greater financial burden and higher radiation doses on patients.

Previous studies have primarily focused on SAT radiodensity in patients with cirrhosis. For instance, Ebadi et al. reported that patients with cirrhosis had relatively higher SAT radiodensity, which was associated with higher mortality rates. The reshaping of SAT morphology in patients with high SAT radiodensity may indicate reduced fat storage and changes in tissue characteristics(9). In another study, Ebadi et al. demonstrated that a lower SATI was associated with higher mortality rates in female patients with cirrhosis, potentially reflecting the depletion of the body's primary energy reservoir with a good metabolic status, leading to poor clinical outcomes(12). Furthermore, related research in malignancies, such as liver and colon cancer, suggests that SAT radiodensity is associated with disease prognosis(14, 15).

SAT plays a crucial role in lipid storage and energy balance. High SAT radiodensity may reflect severe energy consumption induced by cirrhosis, leading to adverse clinical outcomes. Recent studies have highlighted the comprehensive endocrine functions of adipose tissue, where adipose factors influence the regulation of various metabolic and inflammatory states. The cellular composition of adipose tissue, including adipocytes and macrophages, plays an important role in regulating responses to metabolic states, the release of adipose factors, and potential effects on other tissues(16). Therefore, high SAT radiodensity not only signifies low SAT volume but also indicates adipose tissue remodeling. Its morphological features include small atrophied adipocytes, expanded interstitial spaces, and mononuclear cell infiltration. Moreover, increased SAT radiodensity is associated with a higher risk of decompensation in patients with cirrhosis(17).

In this study, we delineated two ROIs and examined the influence of SAT and VAT characteristics on the likelihood of ACLF development. Despite conducting quantitative and normalized index analyses of SAT and VAT, no significant results were found. We speculate two plausible reasons for this outcome. First, the small number of patients in the ACLF occurrence group may have led to inconclusive results in the statistical analysis of both variables. Second, existing research has suggested a U-shaped relationship between SATI, VATI, and the prognosis of patients with cirrhosis, in which excessively high or low indices may lead to unfavorable outcomes(18). Hence, we hypothesized that a similar U-shaped relationship exists between these variables and the risk of developing ACLF. Further investigation with larger sample sizes is needed to explore this relationship.

This study also had several limitations. First, previous research on body composition assessment in patients with cirrhosis found that fat depletion was more common in female patients, while muscle depletion was more common in male patients(12, 19). However, in our study, the number of patients in the ACLF occurrence group was insufficient to support subgroup analysis based on sex due to the need to collect patient CT scan data. Second, viral hepatitis was the predominant etiology among the enrolled patients; therefore, future studies should include patients with other etiologies. Lastly, our study was a retrospective cohort study, and further validation of the predictive accuracy of SAT radiodensity for ACLF occurrence in patients with acute decompensation of cirrhosis may be needed in other cohorts or prospective studies.

This study identified SAT radiodensity as a novel predictor of ACLF onset and developed a new predictive model. This model aimed to identify individuals at a higher risk of ACLF among patients with acute decompensation of cirrhosis, thereby facilitating early clinical intervention and improving patient survival rates.

ACLF: Acute-on-chronic liver failure

PMI: Psoas muscle index

CT: Computed tomography

SAT: Subcutaneous adipose tissue

VAT: Visceral adipose tissue

AUC: Area under the curve

DBIL: Direct bilirubin

PTA: Prothrombin activity

NA+: Sodium ion

HU: Hounsfield units

BMI: Body mass index

AST: Aspartate aminotransferase

ALT: Alanine aminotransferase

CHE: Cholinesterase

ALB: Albumin

TBIL: Total bilirubin

APTT: Activated partial thromboplastin time

PT: Prothrombin time

INR: International normalized ratio

SCR: Serum creatinine

PACS: Picture Archiving and Communication System

ROIs: Regions of interest

SATI: Subcutaneous adipose tissue index

VATI: Visceral adipose tissue index

Ethics approval and consent to participate:The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the ethics committee of the First Hospital of Jilin University (Approval Number: 2024-599). The need for individual consent was waived by the committee.

Availability of data and materials:

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Competing interests:

All authors have completed the ICMJE uniform disclosure form. The authors have no conflicts of interest to declare.

Funding:

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors' contributions:

(I) Conception and design: Weiling Xu

(II) Administrative support: Yu-Hang Yuan, Jing-Yu Wang

(III) Provision of study materials or patients: Weiling Xu, Yu-Hang Yuan

(IV) Collection and assembly of data: Yu-Hang Yuan,Wu-Xi Zhang,Han Xiao

(V) Data analysis and interpretation: Yu-Hang Yuan

(VI) Manuscript writing: Yu-Hang Yuan

(VII) Final approval of manuscript: All authors.

Acknowledgments:

The authors would like to thank all study participants who were enrolled in this study.

Sarin SK, Choudhury A, Sharma MK, Maiwall R, Al Mahtab M, Rahman S, et al. Acute-on-chronic liver failure: consensus recommendations of the Asian Pacific association for the study of the liver (APASL): an update. Hep Intl. 2019;13 (4):353–90.
Gustot T, Fernandez J, Garcia E, Morando F, Caraceni P,Alessandria C, et al. Clinical course of acute-on-chronic liver failure syndrome and effects on prognosis. Hepatology (Baltimore, MD). 2015;62(1):243–52.
Ebadi M, Bhanji RA, Tandon P, Mazurak V, Baracos VE, Montano-Loza AJ.Review article: prognostic significance of body composition abnormalities in patients with cirrhosis. Aliment Pharmacol Ther 2020;52:600–618
Xu M, Li T, Kong M, et al. Psoas Muscle Index Can Be Used to Predict Long-Term Mortality in Young Male Patients With Acute-on-Chronic Liver Failure. Front Nutr. 2022;9:811826. Published 2022 Feb 18. doi:10.3389/fnut.2022.811826
Ebadi M, Bhanji RA, Tandon P, Mazurak V, Baracos VE, Montano-Loza AJ. Review article: prognostic significance of body composition abnormalities in patients with cirrhosis. Aliment Pharmacol Ther. 2020;52(4):600-618. doi:10.1111/apt.15927
Moreau R, Jalan R, Gines P, et al. Acute-on-chronic liver failure is a distinct syndrome that develops in patients with acute decompensation of cirrhosis. Gastroenterology. 2013;144(7):1426-1437.e14379. doi:10.1053/j.gastro.2013.02.042
Shen W, Punyanitya M, Wang ZM, Gallagher D, et al. Total body skeletal muscle and adipose tissue volumes: estimation from asingle abdominal cross-sectional image. J Appl Physiol (1985)2004;97:2333–2338.
Ebadi M, Dunichand-Hoedl AR, Rider E, et al. Higher subcutaneous adipose tissue radiodensity is associated with increased mortality in patients with cirrhosis. JHEP Rep. 2022;4(7):100495. Published 2022 Apr 27. doi:10.1016/j.jhepr.2022.100495
Hu C, Jiang N, Zheng J, et al. Liver volume based prediction model for patients with hepatitis B virus-related acute-on-chronic liver failure. J Hepatobiliary Pancreat Sci. 2022;29(12):1253-1263. doi:10.1002/jhbp.1112
Hanley C, Shields KJ, Matthews KA, Brooks MM, Janssen I, Budoff MJ, et al.Associations of cardiovascular fat radiodensity and vascular calcification in midlife women: the SWAN cardiovascular fat ancillary study. Atherosclerosis 2018;279:114–121.
Din MU, Raiko J, Saari T, Saunavaara V, Kudomi N, Solin O, et al. Human brown fat radiodensity indicates underlying tissue composition and systemic metabolic health. J Clin Endocrinol Metab 2017;102:2258–2267.
Ebadi M, Tandon P, Moctezuma-Velazquez C, et al. Low subcutaneous adiposity associates with higher mortality in female patients with cirrhosis. J Hepatol. 2018;69(3):608-616. doi:10.1016/j.jhep.2018.04.015
Monti CB, Capra D, Malavazos A, Florini G, Parietti C, Schiaffino S, et al.Subcutaneous, paracardiac, and epicardial fat CT density before/after contrast injection: any correlation with CAD? J Clin Med 2021:10.
von Hessen L, Roumet M, Maurer MH, et al. High subcutaneous adipose tissue density correlates negatively with survival in patients with hepatocellular carcinoma. Liver Int. 2021;41(4):828-836. doi:10.1111/liv.14755
Feliciano EMC, Winkels RM, Meyerhardt JA, Prado CM, Afman LA, Caan BJ. Abdominal adipose tissue radiodensity is associated with survival after colorectal cancer. Am J Clin Nutr. 2021;114(6):1917-1924. doi:10.1093/ajcn/nqab285
Ibrahim MM. Subcutaneous and visceral adipose tissue: structural and functional differences. Obes Rev 2010;11:11–18.
Tapper EB, Zhang P, Garg R, Nault T, Leary K, Krishnamurthy V, et al. Body composition predicts mortality and decompensation in compensated cirrhosis patients: a prospective cohort study. JHEP Rep 2020;2:100061.
Zhu M, Li H, Yin Y, et al. U-shaped relationship between subcutaneous adipose tissue index and mortality in liver cirrhosis. J Cachexia Sarcopenia Muscle. 2023;14(1):508-516. doi:10.1002/jcsm.13154
Carey EJ, Lai JC, Wang CW, et al. A multicenter study to define sarcopenia in patients with end-stage liver disease. Liver Transpl. 2017;23(5):625-633. doi:10.1002/lt.24750

No competing interests reported.

table1.png
Table 1 Baseline characteristics of patients
table2.png
Table 2 Identification of the optimum predictive indicators by logistic regression. Abbreviations: HR, hazard ratio; CI, confidence interval; SATI, subcutaneous adipose tissue index; VATI, visceral adipose tissue index

Download PDF

Version 1

posted

You are reading this latest preprint version

Higher subcutaneous adipose tissue radiodensity is associated with an increased risk of the onset of Acute-on-Chronic Liver Failure

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

1. BACKGROUND

2. METHODS

2.1 Patients

2.2 CT image analysis

2.3 Study design

2.4 Statistical analysis

3. RESULTS

3.1 Characteristics of the deriving cohort

3.2 Patients with high SAT radiodensity are at a higher risk of developing ACLF

3.3 The association between VAT characteristics and the occurrence of ACLF

3.4 Variable selection and prognostic model construction

4. DISCUSSION

5. CONCLUSIONS

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1