Abstract
Background & Aims:
Frailty, a clinical phenotype of decreased physiologic reserve, is a strong determinant of adverse health outcomes in patients with cirrhosis. The only cirrhosis-specific frailty metric is the Liver Frailty Index (LFI) which must be administered in person and may not be feasible for every clinical scenario. We sought to discover candidate serum/plasma protein biomarkers that could differentiate frail from robust patients with cirrhosis.
Approach & Results:
Included were 140 adults with cirrhosis awaiting liver transplantation in the ambulatory setting with LFI assessments and available serum/plasma samples. We selected 70 pairs of patients on opposite ends of the frailty spectrum (LFI>4.4 for frail and LFI<3.2 for robust) who were matched by age, sex, etiology, HCC, and MELDNa. Twenty-five biomarkers with biologically plausible associations with frailty were analyzed by ELISA by a single laboratory. Conditional logistic regression was used to examine their association with frailty. Of the 25 biomarkers analyzed, we identified 7 proteins that were differentially expressed between frail and robust. We observed differences in 6 of the 7 proteins in the expected direction: a) higher median values in frail vs robust with growth-differentiation factor-15 (3682 vs 2249 pg/mL), interleukin-6 (17.4 vs 6.4 pg/mL), tumor necrosis factor-alpha receptor 1 (2062 vs 1627 pg/mL), leucine-rich alpha-2 glycoprotein (44.0 vs 38.6 ug/mL), and myostatin (4066 vs 6006 ng/mL), and b) lower median values in frail vs robust with alpha-2-Heremans-Schmid glycoprotein (0.11 vs 0.13 mg/mL) and free total testosterone (1.2 vs 2.4 ng/mL).
Conclusions:
These biomarkers represent inflammatory, musculoskeletal, and endocrine/metabolic systems, reflecting the multiple physiologic derangements observed in frailty. These data lay the foundation for confirmatory work and development of a laboratory frailty index for patients with cirrhosis to improve diagnosis and prognostication.
Keywords: alpha-2-Heremans-Schmid glycoprotein, growth-differentiation factor-15, interleukin-6, leucine-rich alpha-2 glycoprotein, testosterone, tumor necrosis factor-alpha receptor 1
INTRODUCTION
Cirrhosis is the terminal complication of a multitude of chronic liver conditions that can lead to progressive liver failure and death. Liver transplant continues to be the only known “cure” for cirrhosis but is a limited resource requiring accurate prioritization of livers to those in greatest need, thus, ongoing improvement of prioritization tools is necessary. Currently, prioritization of patients with cirrhosis for liver transplantation is based on their risk of death in the next 90 days, as determined entirely by a laboratory-based metric, Model for End-Stage Liver Disease-Sodium (MELDNa). (1–3) While MELDNa accurately predicts 90-day mortality in many patients with cirrhosis, it is particularly vulnerable to underestimating mortality risk in a subgroup of patients with a high degree of “extra-hepatic” manifestations—including malnutrition, muscle wasting, and functional decline. These extra-hepatic manifestations have come to be known in the hepatology community as “frailty” and, conceptually, are poorly represented by laboratory markers of liver and kidney function included in the MELDNa score. (4) In fact, studies have shown that frailty predicts mortality independent of liver disease severity, and that available clinical tools to assess frailty improves mortality risk prediction over MELDNa alone. (4–7)
While frailty is a critical determinant of mortality in patients with cirrhosis, tools to assess frailty must be administered—whether it be having the patient perform tests, asking the patient to answer questions, or having the clinician make a judgment on the patient’s status. (8) This limits frailty assessment to patient-provider touchpoints (e.g., clinic visits, telehealth assessments) and therefore, the possibility of integrating the construct of frailty into national allocation systems, which necessitate continuous updates of mortality risk assessments (currently by the MELDNa score) at intervals as short as every 7 days. It would, therefore, be valuable to identify blood-based protein biomarkers of frailty in patients with cirrhosis.
To set the foundation for this quest, we embarked upon this study to identify serum and plasma protein biomarkers that could serve as surrogate markers for frailty in patients with cirrhosis.
MATERIAL AND METHODS
Study design and patient selection
This is an ancillary study of the Functional Assessment in Liver Transplantation (FrAILT) Study, a prospective cohort study that enrolls adult patients (age ≥18 years old) with cirrhosis who are listed for liver transplantation and are seen in the ambulatory setting. (4) Patients in the FrAILT Study enrolled at the University of California, San Francisco (UCSF) undergo biospecimen collection at the time of enrollment and clinical frailty assessment. For this specific case-control study, we selected cases and controls from our cohort of 899 participants who had been enrolled in the FrAILT Study at UCSF between April 2015 and July 2020 and had available clinical frailty assessment and biospecimens. Patients were initially categorized as frail or robust by a Liver Frailty Index (LFI) of >4.4 or <3.2, respectively. From these groups, frail cases and robust controls were then selected, matched in pairs by sex, age (±5 years), cirrhosis etiology, hepatocellular carcinoma (HCC) status, and MELDNa (±5 points) in order to minimize confounding of biomarkers results based on these factors. Demographic and clinical data, including primary liver disease etiology, comorbid conditions, and laboratory data were manually chart reviewed and abstracted from the electronic health records by trained study personnel who were blinded to all biomarker data. Clinical evidence of ascites and hepatic encephalopathy were ascertained through a chart review by trained study personnel at the time of the study visit.
All research was conducted in accordance with both the Declarations of Helsinki and Istanbul. Written consent was provided by all subjects. This study was approved by the Institutional Review Board at the University of California, San Francisco (San Francisco, CA, USA).
Measurement of frailty
All patients underwent objective measurement of frailty using the LFI at time of enrollment. The LFI consists of three performance-based tests: 1) grip strength, 2) chair stands, 3) balance testing. (4) For classification of frailty, we used previously established cutoffs of the LFI to define robust (<3.2) and frail (>4.4). (4, 7)
Selection and measurement of protein biomarkers
Blood samples were collected from study participants on the same day as clinical frailty testing, delivered and processed within one hour to the UCSF Clinical and Translational Science Institute Clinical Research Services Sampling Processing Laboratory using standard protocols for serum processing. All samples were labeled using a formal laboratory cataloguing system and stored at minus 80 degree Celsius at the UCSF Biospecimen Resource Program (BIOS) in San Francisco, California. Given available literature on objective and validated biomarkers of frailty in cirrhosis remains limited, potential candidate biomarkers were reviewed and considered based on their biologically plausible associations with aging and age-related conditions, in particular, frailty. (9–12) By focusing on physical frailty, an intermediate phenotype that we have previously demonstrated to be strongly predictive of mortality, we gain the advantages of narrowing the list of potential biomarkers for investigation to those with biological plausibility and therefore, most likely predictive—thus, reducing the likelihood of false positive associations. We employed this biologically rational dimension-reduction approach to more efficiently identify protein biomarkers that may correlate with the frail phenotype, and ultimately enhance our ability to predict waitlist mortality in this population. We then selected twenty-five promising candidate biomarkers based on whether (1) they have been previously linked or associated with frailty, (2) readily secreted and measurable with (3) commercially available enzyme-linked immunosorbent assay (ELISA) assays, and (4) derived from diverse pathophysiological pathways associated with age-related pathologies including inflammatory, musculoskeletal, cardiovascular, and endocrine and metabolic systems. Stored plasma and serum samples were sent from BIOS to a single laboratory, BioBank and Research Laboratory Services, Maine Medical Center Research Institute in Scarborogh, Maine for analysis. All laboratory personnel were blinded to clinical data including clinical frailty status. The following commercially available ELISA assays were utilized for this study: alpha-1 acid glycoprotein (R&D Systems, DAGP00), alpha-2-Heremans-Schmid glycoprotein (ALPCO, 43-FETHU-E01), alpha-2 macroglobulin (R&D Systems, DY1938), apolipoprotein A1 (ALPCO, 41-APOHU-E01), C-reactive protein (R&D Systems, DY1707), estradiol (ALPCO, 11-ESTHU-E01), glutathione peroxidase 3 (MyBioSource, MBS765842), growth differentiation factor-15 (R&D Systems, DGD150), haptoglobin (R&D Systems, DHAPG0), inhibitor of nuclear factor kappa-beta kinase (MyBioSource, MBS260718), insulin growth factor-1 (IDS, IS-3900), insulin growth factor binding protein-3 (R&D Systems, DY675), interleukin-6 (R&D Systems, D6050), keratin 18 (ThermoFisher Invitrogen, EH145RB), leptin (R&D Systems, DLP00), leucine-rich alpha-2 glycoprotein (MyBioSource, MBS701497), myostatin (R&D Systems, DGDF80), nuclear factor-kappa light chain-enhancer of activated B cells (MyBioSource, MBS260718), sex hormone binding globulin (R&D Systems, DSHBG0B), total testosterone (ALPCO, 11-TESHU-E01), transthyretin (MyBioSource, MBS762549), tumor necrosis factor-alpha (R&D Systems, DY210), tumor necrosis factor-alpha receptor 1 (R&D Systems, DY225), vitronectin (ThermoFisher Invitrogen, EHVTN), and zinc alpha-2 glycoprotein (BioVendor, RD191093100R).
Statistical analysis
Data were summarized using counts and percentages (%) for categorical variables or medians and interquartile ranges (IQR) for continuous variables. Variables were compared between groups using Wilcoxon rank-sum and Pearson’s chi-square tests, as appropriate. Difference in protein quantity between frail and robust cohorts were compared using Wilcoxon rank-sum. Quantitative biomarker variables were log-transformed as appropriate to attenuate the influence of extreme laboratory values. Given our frail and robust cohort were non-loosely matched in pairs based on their sex, age ± 5 years, etiology, HCC, and MELDNa ± 5 points, conditional logistic regression was used to examine associations between the 25 individual protein biomarkers quantified on ELISA and clinical frailty on LFI assessment. All biomarkers that were associated with frailty in univariate analysis with p-value <0.10 were evaluated for inclusion in the final multivariate model. The final multivariate model was developed using backwards elimination of variables until all variables included were associated with a p-value of <0.05. A two-sided p-value <0.05 was considered statistically significant. Analyses were performed using STATA, version 15.1 (StataCorp, College Station, TX, USA).
RESULTS
Baseline patient characteristics
Included were 140 patients: 70 were frail with a median (IQR) LFI of 4.9 (4.6–5.1) and 70 were robust with a median (IQR) LFI of 2.9 (2.6–3.0). Baseline characteristics are shown in Table 1. Both groups were well matched by sex (66% vs 66% men), median age (58 vs 57 years, p=0.59), HCC (34% vs 34%), median MELDNa (14 vs 14), and etiologies: hepatitis C (34% vs 30%), hepatitis B (1% vs 3%), alcohol-associated (33% vs 33%), nonalcoholic fatty liver disease/nonalcoholic steatohepatitis (23% vs 23%), autoimmune/cholestatic disease (9% vs 11%). Compared to robust patients, frail patients had higher prevalence of hepatic encephalopathy (70% vs 39%) and ascites (64% vs 51%), with a greater proportion of frail patients experiencing severe/refractory ascites (31% vs 13%). Rates of hypertension (51% vs 44%) and diabetes (39% vs 29%) were slightly higher in frail compared to robust patients.
Table 1.
Baseline demographic and clinical characteristics among frail and robust patients with cirrhosis
| Overall n=140 | Frail n=70 | Robust n=70 | p-value | |
|---|---|---|---|---|
|
| ||||
| Sex (male) | 92 (66) | 46 (66) | 46 (66) | >0.99 |
| Age (years) | 57 (51–63) | 58 (51–63) | 57 (51–63) | 0.59 |
| Body mass index (kg/m2) | 28.2 (25.0–32.6) | 28.9 (24.7–33.9) | 27.8 (25.5–32.3) | 0.47 |
| Ethnicity/race | 0.32 | |||
| White | 121 (86) | 62 (89) | 59 (84) | |
| Black | 7 (5) | 4 (6) | 3 (4) | |
| Asian | 11 (8) | 3 (4) | 8 (11) | |
| Indian/Native | 1 (1) | 1 (1) | 0 (0) | |
| Etiology | 0.94 | |||
| Hepatitis C virus | 45 (32) | 24 (34) | 21 (30) | |
| Alcohol | 46 (33) | 23 (33) | 23 (33) | |
| NAFLD/NASH | 32 (23) | 16 (23) | 16 (23) | |
| AIH/PSC/PBC | 14 (10) | 6 (9) | 8 (11) | |
| Hepatitis B virus | 3 (2) | 1 (1) | 2 (3) | |
| Hepatocellular carcinoma | 48 (34) | 24 (34) | 24 (34) | >0.99 |
| Comorbidities | ||||
| Hypertension | 67 (48) | 36 (51) | 31 (44) | 0.40 |
| Diabetes | 47 (34) | 27 (39) | 20 (29) | 0.21 |
| Coronary artery disease | 4 (3) | 2 (3) | 2 (3) | >0.99 |
| Ascites | 0.03 | |||
| Mild/moderate | 31 (22) | 22 (31) | 9 (13) | |
| Severe/refractory | 31 (22) | 22 (31) | 9 (13) | |
| Hepatic encephalopathy | 76 (54) | 49 (70) | 27 (39) | <0.001 |
| MELDNa | 14 (12–17) | 14 (13–18) | 14 (11–17) | 0.41 |
| Albumin (mg/dL) | 3.2 (2.8–3.6) | 3.1 (2.8–3.5) | 3.4 (2.8–3.7) | 0.06 |
| Liver Frailty Index | 3.8 (2.9–4.9) | 4.9 (4.6–5.1) | 2.9 (2.6–3.0) | <0.001 |
Values reported in median (IQR) or number (percentage). Abbreviations: AIH/PBC/PSC, autoimmune hepatitis/primary biliary cholangitis/primary sclerosing cholangitis; MELDNa, model for end-stage liver disease-sodium; NAFLD/NASH, nonalcoholic fatty liver disease/nonalcoholic steatohepatitis
Quantification and association of protein expression in frail vs robust patients
ELISA-measured protein concentrations of 25 biomarkers in frail and robust patients and its expected direction in relationship to frailty (increased or decreased) are shown in Table 2. Among inflammatory candidate biomarkers, only two were significantly expressed differentially among frail and robust patients: interleukin-6 (IL-6, p<0.001) and growth differentiation factor-15 (GDF-15, p<0.001); with tumor necrosis factor-alpha receptor 1 (TNFαR1, p=0.07) meeting our predefined criteria (p<0.10) for inclusion in our conditional regression analysis. In the remaining inflammatory candidate biomarkers, there were no significant differences between the two cohorts in terms of alpha-2 macroglobulin, C-reactive protein, glutathione peroxidase 3, haptoglobin, inhibitor of nuclear factor kappa-B kinase, keratin 18, nuclear factor kappa-light chain-enhancer of activated B cells, tumor necrosis factor-alpha (TNFα), and vitronectin. Among musculoskeletal candidate biomarkers, two were significantly expressed differentially among frail and robust patients: leucine-rich alpha-2 glycoprotein (LRG, p=0.01) and myostatin (p=0.02); there was a trend with alpha-2-Heremans-Schmid glycoprotein (AHSG, p=0.08) differentially expressed lower in frail patients. There was no significant difference in levels of zinc alpha-2 glycoprotein among the two cohorts. Among endocrine and metabolic candidate biomarkers, only free total testosterone was expressed differentially between frail and robust patients (p=0.02). Though not statistically significant, increased levels of leptin and sex hormone binding globulin along with decreased levels of transthyretin was observed in frail compared to robust patients. The difference in measured estradiol, insulin growth factor 1, and insulin growth factor binding protein-3 between the two cohorts did not reach statistical significance.
Table 2.
Serum and plasma protein biomarkers level among frail and robust patients with cirrhosis and univariable conditional logistic regression for predictors of frailty
| System | Protein | Frail n=70 | Robust n=70 | p-value | Expected in Frail | OR (95% CI) | Biological function |
|---|---|---|---|---|---|---|---|
|
| |||||||
| Inflammatory | Alpha-2 Macroglobulin (g/L) | 2.10 (1.58–2.60) | 2.09 (1.68–2.50) | 0.65 | ↑ | 0.34 (0.02–5.45) | Acute phase protein, antiprotease, and inhibit fibrinolysis |
| C-Reactive Protein (mg/L) | 5.9 (2.0–14.4) | 4.1 (1.9–14.9) | 0.50 | ↑ | 1.20 (0.70–2.06) | Acute phase protein associated with inflammation | |
| Glutathione Peroxidase 3 (ng/mL) | 8.0 (2.6–55.5) | 5.8 (1.1–72.4) | 0.50 | ↓ | 0.99 (0.76–1.31) | Prevent oxidative damage | |
| Growth Differentiation Factor-15 (pg/mL) | 3682 (2052–5614) | 2249 (1604–3486) | <0.001 | ↑ | 27.3 (4.46–167.09) | Associated with chronic inflammation and marker for mitochondrial dysfunction | |
| Haptoglobin (ng/ml) | 80413 (1440–389617) | 82033 (1117–375459) | 0.58 | ↑ | 1.08 (0.79–1.47) | Antioxidant that binds to oxidatively active heme, indicating physiologic stress and inflammation | |
| Inhibitor of Nuclear Factor Kappa-Beta Kinase (ng/mL) | 9.7 (6.7–13.4) | 9.8 (7.2–15.5) | 0.46 | ↓ | 0.99 (0.96–1.02) | Regulate inflammatory response | |
| Interleukin 6 (pg/mL) | 17.4 (7.0–31.7) | 6.4 (4.4–16.7) | <0.001 | ↑ | 5.84 (2.14–15.96) | Acute phase protein associated with inflammation | |
| Keratin 18 (ng/mL) | 1.2 (0.4–4.4) | 1.0 (0.3–3.7) | 0.91 | ↑ | 0.92 (0.57–1.49) | Major intermediate filament protein involved in cytoskeletal organization | |
| Nuclear Factor Kappa-Light Chain-Enhancer of Activated B Cells (ng/mL) | 0.34 (0.22–0.43) | 0.34 (0.20–0.43) | 0.85 | ↑ | 0.90 (0.12–6.53) | Involve in cellular response to stress | |
| Tumor Necrosis Factor Alpha (pg/mL) | 6.5 (0.0–136.4) | 15.2 (0.0–595.7) | 0.41 | ↑ | 0.88 (0.71–1.09) | Acute inflammatory regulatory protein leading to necrosis or apoptosis | |
| Tumor Necrosis Factor Alpha Receptor 1 (pg/mL) | 2062 (1426–3725) | 1627 (1244–2437) | 0.07 | ↑ | 1.21 (0.53–2.77) | Maintain immune homeostasis by promoting apoptosis, cell survival, differentiation, and inflammation | |
| Vitronectin (g/L) | 0.02 (0.01–0.03) | 0.02 (0.02–0.03) | 0.33 | ↑ | 0.48 (0.13–1.77) | Involve in cell migration, tissue repair and regulation of membrane attack complex | |
|
| |||||||
| Cardiovascular | Alpha-1 Acid Glycoprotein 1 (mg/ml) | 0.31 (0.20–0.49) | 0.29 (0.20–0.46) | 0.37 | ↑ | 1.90 (0.58–6.22) | Acute phase reactant that is upregulated in response to inflammation or infection |
| Apolipoprotein A1 (mg/mL) | 0.13 (0.10–0.16) | 0.13 (0.11–0.16) | 0.64 | ↑ | 0.37 (0.04–3.40) | Involve in immunity, inflammation, and apoptosis | |
|
| |||||||
| Musculoskeletal | Alpha-2-Heremans-Schmid Glycoprotein (mg/mL) | 0.11 (0.09–0.14) | 0.13 (0.10–0.16) | 0.08 | ↓ | 0.12 (0.01–1.33) | Regulate inflammatory response and phagocytosis |
| Leucine-rich Alpha-2 Glycoprotein (ug/mL) | 44.0 (34.1–52.8) | 38.6 (27.6–48.0) | 0.01 | ↑ | 1.03 (1.01–1.06) | Involve in signal-transduction, cell adhesions, neovascularization, and activator of mammalian target of rapamycin (mTOR) | |
| Myostatin (ng/mL) | 4066 (2554–7170) | 6006 (3370–9240) | 0.02 | ↑ | 0.18 (0.05–0.65) | Inhibit myogenesis for muscle cell growth and differentiation; impairs mTOR signaling | |
| Zinc Alpha-2 Glycoprotein (ug/mL) | 30.9 (23.8–39.5) | 28.7 (23.7–40.2) | 0.54 | ↑ | 2.46 (0.25–23.74) | Stimulate lipolysis and proteolysis-inducing factor associated with degradation of adipose and skeletal muscle | |
|
| |||||||
| Endocrine and Metabolic | Estradiol (pg/mL) | 57.4 (35.9–95.5) | 51.0 (29.5–89.7) | 0.60 | ↑ | 0.92 (0.40–2.10) | Associated with frailty, inflammation, and functional impairment |
| Insulin Growth Factor 1 (ng/mL) | 42.9 (31.6–60.7) | 48.0 (30.4–63.0) | 0.32 | ↓ | 0.29 (0.07–1.21) | Major mediator of growth hormone; promote growth of bones and tissues | |
| Insulin Growth Factor Binding Protein-3 (mg/L) | 0.66 (0.36–0.91) | 0.69 (0.43–1.11) | 0.42 | ↑ | 0.75 (0.45–1.26) | Modulate bioavailability of IGF-1; key protein in the insulin-like growth factor pathway | |
| Leptin (ng/mL) | 13.2 (5.1–26.4) | 8.6 (4.2–17.4) | 0.19 | ↑ | 1.31 (0.65–2.62) | Regulate body weight via food intake and energy expenditure | |
| Sex Hormone Binding Globulin (nmol/L) | 82.3 (53.9–103.3) | 74.2 (53.0–90.5) | 0.15 | ↑ | 3.68 (0.46–29.22) | Regulate and bind to sex hormone | |
| Total Testosterone (ng/mL) | 1.2 (0.2–3.6) | 2.4 (0.8–5.7) | 0.02 | ↓ | 0.60 (0.40–0.92) | Regulate bone mass, fat distribution, muscle mass and strength | |
| Transthyretin (ng/mL) | 3971 (3018–6082) | 3617 (2501–5173) | 0.17 | ↓ | 2.41 (0.54–10.85) | Transport protein of thyroxine and retinol-binding protein; associated with protein-caloric malnutrition and sarcopenia | |
Values reported in median (IQR).
In summary, we identified 7 protein biomarkers among 25 candidate biomarkers that were differentially expressed among the pairs of frail and robust patients with cirrhosis (p<0.10). Six of the 7 proteins differed between the two cohorts in the expected direction: a) higher median values in frail vs robust with GDF-15 (3682 vs 2249 pg/mL), IL-6 (17.4 vs 6.4 pg/mL), TNFαR1 (2062 vs 1627 pg/mL), and LRG (44.0 vs 38.6 ug/mL), and b) lower median values in frail vs robust with AHSG (0.11 vs 0.13 mg/mL) and total testosterone (1.2 vs 2.4 ng/mL); unexpectedly, myostatin was observed to be lower in frail compared to robust patients (4066 vs 6006 ng/mL) and was not associated with frailty. Despite the disproportionate number of patients with ascites and hepatic encephalopathy among frail and robust patients, we observed similar findings of protein biomarkers differentially expressed among the pairs of frail and robust patients with cirrhosis according to the presence of ascites or hepatic encephalopathy (Supplementary Tables 1 to 4). We then conducted an exploratory analysis to identify a parsimonious combination of biomarkers that would most strongly correlate with frailty. In this exploratory multivariable analysis, we found that GDF-15 and IL-6 were associated with frailty (Table 3). Additionally, interaction terms for GDF-15 and IL-6 with ascites (p=0.65) and GDF-15 and IL-6 with hepatic encephalopathy (p=0.69) were not statistically significant.
Table 3.
Exploratory univariable and multivariable conditional logistic regression model for predictors of frailty
| Univariable analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p-value | OR | 95% CI | p-value | |
|
| ||||||
| Alpha-2-Heremans-Schmid Glycoprotein (Log10 ng/mL) | 0.12 | 0.01–1.33 | 0.08 | |||
| Leucine-rich Alpha-2 Glycoprotein (ug/mL) | 1.03 | 1.01–1.06 | 0.01 | |||
| Growth Differentiation Factor-15 (Log10 pg/mL) | 27.3 | 4.46–167.09 | <0.001 | 12.77 | 1.72–94.87 | 0.01 |
| Interleukin 6 (Log10 pg/mL) | 5.84 | 2.14–15.96 | 0.001 | 3.89 | 1.29–11.75 | 0.02 |
| Myostatin (Log10 ng/mL) | 0.18 | 0.05–0.65 | 0.01 | |||
| Total Testosterone (Log10 ng/mL) | 0.60 | 0.40–0.92 | 0.02 | |||
| Tumor Necrosis Factor Alpha Receptor 1 (Log10 pg/mL) | 1.21 | 0.53–2.77 | 0.66 | |||
DISCUSSION
Frailty is a complex multidimensional syndrome characterized by functional decline and reduced physiological reserve. Despite the increasing prevalence and clinical impact of frailty in patients with cirrhosis, available laboratory protein biomarkers for assessing frailty remain elusive in this population. In this prospective cohort study of patients with cirrhosis awaiting liver transplantation in an ambulatory setting, we observed seven protein biomarkers among twenty-five biomarker candidates that differed between frail and robust cirrhosis patients. Similar to the multitude of complications of frailty, these biomarkers are involved in derangement of one or more physiological systems including musculoskeletal, endocrine/metabolic, cardiovascular, neurological, and immune system, further offering important insights into the underlying mechanisms of frailty in cirrhosis.
The first broad category of protein biomarkers is the inflammatory system including TNFαR1, IL-6, and GDF-15, which were all found at higher concentrations in frail compared to robust cirrhosis patients. Patients with cirrhosis are considered to be in a chronic low-inflammatory state due to dysregulation and increased circulatory of proinflammatory cytokine leading to development of subsequent complications including frailty. (13–15) TNFα is a proinflammatory cytokine that has been shown to be elevated in patients with acute and chronic liver disease regardless of etiology (16, 17). Its receptor, TNFαR1, is considered to be a more suitable biomarker of inflammation given its longer half-life compared to TNFα (18, 19). In this study, frail patients were observed to have higher levels of TNFαR1 compared to robust patients, with no significant difference in levels of TNFα. Similar to TNFαR1, IL-6 is a proinflammatory cytokine that has been associated with lower muscle mass and strength (20), reduced grip strengths (21) and physical function (22, 23), and impaired muscle protein synthesis (24). GDF-15 is a stress-responsive member of the transforming growth factor beta superfamily involved in inflammation, fibrosis, and disease progression, which has been shown to be associated with various viral and autoimmune/cholestatic liver diseases, cirrhosis, and HCC. (25, 26) Overexpression of GDF-15 due to chronic tissue injury and gene regulatory dysfunction by activated macrophage has been described as a major pathogenesis of organ fibrosis. (25) It is the prolonged stimulation of hepatic stellate cells due to chronic liver injury that leads to stimulation of GDF-15. (25, 26) Although an increased level of GDF-15 was expected in this study given all patients had cirrhosis, we observed significantly higher levels of GDF-15 in frail compared to robust patients. Several mechanisms by which GDF-15 may contribute to frailty include mitochondrial dysfunction leading to muscle-mitochondrial stress (27) and malnutrition from dysregulation of energy homeostasis (28–31). Additionally, TNFα and IL-6 has been associated with increased muscle breakdown, likely as a compensatory mechanism in the setting of inflammation to yield amino acids that can be used for subsequent energy source. (32) Though an effective response mechanism to inflammation, dysregulation of this system further potentiates muscle wasting. These findings infer that inflammation remains a key contributor of frailty in patients with cirrhosis and may be earlier indicators of development and/or worsening of frailty.
The second broad category of protein biomarkers is the musculoskeletal system. Sarcopenia, defined as generalized loss of muscle mass, regularly accompanies frailty and are closely related concepts that share common clinical significance. (8, 33, 34) Metabolic derangement in the setting of physiologic stress and malnutrition is compensated by a shift in homeostatic mechanisms towards catabolic utilization of muscle mass, contributing to muscle mass degradation leading to sarcopenia. (35) Furthermore, disturbances of ammonia metabolism in cirrhosis impairs muscle contractility while accelerating muscle autophagy and wasting, the latter due to the upregulation of myostatin by ammonia. (34) Myostatin is a muscle-specific protein that inhibits myogenesis, muscle cell growth and differentiation, and has been shown to be elevated in patients with cirrhosis. (36, 37) Unexpectedly, myostatin was observed to be lower in frail compared to robust patients and not associated with frailty in this study; notably, it was the only protein out of the seven biomarkers with an unexpected divergent median concentration. We postulate that this finding may be due to the unstable nature of myostatin in serum samples along with its other weaknesses as a potential biomarker including various confounding factors affecting serum level – age, sex, nutrition, metabolic disorders, inflammation, cardiovascular/kidney disease, and physical activity – and differences in laboratory measurements resulting in conflicting studies. (38) Contrary, LRG and AHSG were observed in a pattern consistent with frailty. LRG is an inflammatory protein that is synthesized in the liver and has been associated with infection, malignancy, and various inflammation-associated diseases including inflammatory musculoskeletal disease (39, 40). Given that increased level of LRG was associated with frailty in this study, it is plausible that this protein biomarker may be upregulated in frail patients with cirrhosis. AHSG is a multifunctional protein and negative acute phase reactant that is synthesized in the liver and has been associated with metabolic derangement such as insulin resistance, bone remodeling and calcium homeostasis. (41) As it is synthesized in the liver, prior studies have suggested that reduced levels of AHSG observed in alcohol-associated and fatty liver disease, cirrhosis, and HCC, may be due to reduced hepatic synthetic function. (42) In this study, AHSG levels were observed to be lower in frail compared to robust patients. Although both cohorts were matched by etiology, HCC, and MELDNa, a higher proportion of frail patients had severe/refractory ascites and hepatic encephalopathy along with lower serum albumin, therefore, it is possible that lower levels of AHSG may reflect changes in hepatic synthetic function not otherwise captured by MELDNa. In contrast, other hepatic synthesized proteins such as LRG was found to be higher in frail patients despite reduced hepatic synthetic function, suggesting that frailty extends beyond just hepatic synthetic dysfunction but rather due to complex multiorgan system derangements. Further studies are needed into the potential clinical utility of AHSG, or other hepatic synthesized biomarkers as potential measures of impaired hepatic function compared to other synthetic biomarkers such as albumin and international normalized ratio, which may allow for earlier detection of worsening hepatic dysfunction for possible intervention.
The third broad category of protein biomarkers is the endocrine and metabolic system. Given testosterone is known to support muscle function (43), improve nutrition, and maintain bone health (44), reduced levels of free testosterone have been shown to be associated with frailty in both men and women (45, 46). Within the construct of cirrhosis, low testosterone levels were associated with sarcopenia in men (47) which improved after administration of testosterone (48). In our study, frail patients had significantly lower levels of free testosterone compared to robust patients with cirrhosis and was a significant predictor of frailty.
While there is clearly shared pathophysiology between age-related frailty and cirrhosis-related frailty, there are also drivers specific to liver disease that lead to the frail phenotype in patients with cirrhosis. (49) Ascites results in early satiety and reduced protein intake, accelerating frailty, and hepatic encephalopathy leads to significant reductions in physical activity (50, 51) In addition, the etiology of liver disease has also been associated with differences in the prevalence of frailty and/or sarcopenia. For example, patients with alcohol-associated liver disease exhibit greater loss of muscle mass due to ethanol-induced sensitization of skeletal muscle to hyperammonemia, muscle autophagy, and decrease anabolic hormone. (52, 53) Moreover, patients with nonalcoholic steatohepatitis have increased insulin resistance and chronic inflammation, which are known frailty contributors. (54) In the present study, frail and robust patients were well-matched to minimize confounders of frailty—including age-, sex-, and etiology-related factors—to study a relatively homogeneous cohort of patients with a single dominate condition (i.e., cirrhosis) that leads to premature aging through well-established pathophysiologic pathways observed in frailty.
We acknowledge several limitations to our study. First, only patients with cirrhosis who were seen in the ambulatory setting were included; as such, our data may not be as generalizable to patients with cirrhosis who are acutely ill. However, the median MELDNa score of 14 in our patients is similar to the proportion of initial MELDNa score of <15 seen in the majority of patients awaiting liver transplantation in the United States, so our data are still of high clinical relevance. (55) Given that the majority race/ethnicity was white, findings from this study may not be directly generalizable to more diverse populations for which protein biomarker levels and association with frailty may differ. Moreover, protein biomarkers were measured at time of frailty assessment and may not be representative of the dynamic changes of frailty; thus, future studies with protein biomarkers measured on a pre-planned schedule to correlate with frailty is warranted. Nonetheless, we would expect similar patterns in protein biomarkers (i.e., increased or decreased) based on frailty status given its associated biological plausibility as shown in Table 2. Our study population was comprised of both men and women of various age with different etiologies and severity of cirrhosis or presence of HCC, all of which may factor or contribute differently to the underlying mechanism of frailty. However, these subpopulations were well matched to control for any possible confounders to demonstrate correlations between protein biomarkers and frailty. Additionally, we dichotomized the study population by matched pairs on opposite ends of the frailty spectrum—frail (LFI >4.4) or robust (LFI <3.2)—to increase the likelihood of identifying protein biomarkers associated with frailty—but did not include patients with intermediate risk (LFI 3.2 to 4.4) or “pre-frail”. A larger validation diverse cohort consisting of patients across the frailty spectrum—robust, pre-frail, and frail—is needed to develop a laboratory frailty index for this population, but this study represents critical foundational work for that next study.
While our primary goal is to improve mortality prediction in patients with cirrhosis, we proposed to identify biomarkers of the phenotype of frailty, an intermediary of mortality, to simultaneously achieve a secondary goal—to provide the hepatology and liver transplant communities with laboratory-based markers to allow a clinician to approximate frailty, independent of liver disease severity. This is important for several reasons. First, patients awaiting liver transplantation frequently live hours away from their transplant center, so frequent in-person assessment is not feasible; MELDNa, for example, must be updated as frequently as every 7 days. Second, certain aspects of the frail phenotype, unlike the MELDNa score, may be modifiable through intensive nutrition and physical therapy support. And while laboratory frailty markers themselves may not be modifiable, it may help identify those in greatest need for intensive support to improve clinical outcomes.
In conclusion, we identified several protein biomarkers that were differentially expressed in frail and robust patients with cirrhosis in a well-matched cohort. These biomarkers were representative of musculoskeletal, endocrine/metabolic, cardiovascular, and inflammatory systems, reflecting the multiple physiologic derangements observed in frailty. Our exploratory data lay the foundation for confirmatory work and external validation, with subsequent development of a laboratory frailty index specific to patients with cirrhosis for improved diagnosis and prognostication for use in a national liver allocation system.
Supplementary Material
Financial Support:
This study was funded by NIH R01AG059183 (Lai), NIH P30DK026743 (Huang, Lai, Shui), NIH R21AG067554 (Lai), NIH T32DK060414 (Ha), NIH P30AG044281 (Lai), and NIH P30AG021334 (Walston). These funding agencies played no role in the analysis of the data or the preparation of this manuscript.
List of Abbreviations:
- AHSG
alpha-2-Heremans-Schmid glycoprotein
- ELISA
enzyme-linked immunosorbent assay
- GDF-15
growth-differentiation factor-15
- HCC
hepatocellular carcinoma
- IL-6
interleukin-6
- IQR
interquartile range
- LRG
leucine-rich alpha-2 glycoprotein
- LFI
liver frailty index
- MELDNa
model for end-stage liver disease sodium
- TNFαR1
tumor necrosis factor-alpha receptor 1
Footnotes
Author COI:
Jennifer C. Lai advises for Novo Nordisk, consults for Genfit and has received grants from Axcella Health, Nestle Nutrition Institute, Pliant, and Vir biotechnologies.
REFERENCES
- 1.Kamath PS, Wiesner RH, Malinchoc M, Kremers W, Therneau TM, Kosberg CL, et al. A model to predict survival in patients with end-stage liver disease. Hepatology. 2001;33(2):464–70. [DOI] [PubMed] [Google Scholar]
- 2.Kim WR, Biggins SW, Kremers WK, Wiesner RH, Kamath PS, Benson JT, et al. Hyponatremia and mortality among patients on the liver-transplant waiting list. N Engl J Med. 2008;359(10):1018–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wiesner R, Edwards E, Freeman R, Harper A, Kim R, Kamath P, et al. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology. 2003;124(1):91–6. [DOI] [PubMed] [Google Scholar]
- 4.Lai JC, Covinsky KE, Dodge JL, Boscardin WJ, Segev DL, Roberts JP, et al. Development of a novel frailty index to predict mortality in patients with end-stage liver disease. Hepatology. 2017;66(2):564–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Haugen CE, McAdams-DeMarco M, Holscher CM, Ying H, Gurakar AO, Garonzik-Wang J, et al. Multicenter Study of Age, Frailty, and Waitlist Mortality Among Liver Transplant Candidates. Ann Surg. 2020;271(6):1132–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lai JC, Feng S, Terrault NA, Lizaola B, Hayssen H, Covinsky K. Frailty predicts waitlist mortality in liver transplant candidates. Am J Transplant. 2014;14(8):1870–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kardashian A, Ge J, McCulloch CE, Kappus MR, Dunn MA, Duarte-Rojo A, et al. Identifying an Optimal Liver Frailty Index Cutoff to Predict Waitlist Mortality in Liver Transplant Candidates. Hepatology. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lai JC, Tandon P, Bernal W, Tapper EB, Ekong U, Dasarathy S, et al. Malnutrition, Frailty, and Sarcopenia in Patients With Cirrhosis: 2021 Practice Guidance by the American Association for the Study of Liver Diseases. Hepatology. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cardoso AL, Fernandes A, Aguilar-Pimentel JA, de Angelis MH, Guedes JR, Brito MA, et al. Towards frailty biomarkers: Candidates from genes and pathways regulated in aging and age-related diseases. Ageing Res Rev. 2018;47:214–77. [DOI] [PubMed] [Google Scholar]
- 10.Ferrucci L, Cavazzini C, Corsi A, Bartali B, Russo CR, Lauretani F, et al. Biomarkers of frailty in older persons. J Endocrinol Invest. 2002;25(10 Suppl):10–5. [PubMed] [Google Scholar]
- 11.Mitnitski A, Collerton J, Martin-Ruiz C, Jagger C, von Zglinicki T, Rockwood K, et al. Age-related frailty and its association with biological markers of ageing. BMC Med. 2015;13:161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Puts MT, Visser M, Twisk JW, Deeg DJ, Lips P. Endocrine and inflammatory markers as predictors of frailty. Clin Endocrinol (Oxf). 2005;63(4):403–11. [DOI] [PubMed] [Google Scholar]
- 13.Albillos A, Lario M, Alvarez-Mon M. Cirrhosis-associated immune dysfunction: distinctive features and clinical relevance. J Hepatol. 2014;61(6):1385–96. [DOI] [PubMed] [Google Scholar]
- 14.Arroyo V, Garcia-Martinez R, Salvatella X. Human serum albumin, systemic inflammation, and cirrhosis. J Hepatol. 2014;61(2):396–407. [DOI] [PubMed] [Google Scholar]
- 15.Hubbard RE, O’Mahony MS, Savva GM, Calver BL, Woodhouse KW. Inflammation and frailty measures in older people. J Cell Mol Med. 2009;13(9B):3103–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Odeh M, Sabo E, Srugo I, Oliven A. Serum levels of tumor necrosis factor-alpha correlate with severity of hepatic encephalopathy due to chronic liver failure. Liver Int. 2004;24(2):110–6. [DOI] [PubMed] [Google Scholar]
- 17.Tilg H, Wilmer A, Vogel W, Herold M, Nolchen B, Judmaier G, et al. Serum levels of cytokines in chronic liver diseases. Gastroenterology. 1992;103(1):264–74. [DOI] [PubMed] [Google Scholar]
- 18.Riordan SM, Skinner N, Nagree A, McCallum H, McIver CJ, Kurtovic J, et al. Peripheral blood mononuclear cell expression of toll-like receptors and relation to cytokine levels in cirrhosis. Hepatology. 2003;37(5):1154–64. [DOI] [PubMed] [Google Scholar]
- 19.Tarrats N, Moles A, Morales A, Garcia-Ruiz C, Fernandez-Checa JC, Mari M. Critical role of tumor necrosis factor receptor 1, but not 2, in hepatic stellate cell proliferation, extracellular matrix remodeling, and liver fibrogenesis. Hepatology. 2011;54(1):319–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Visser M, Pahor M, Taaffe DR, Goodpaster BH, Simonsick EM, Newman AB, et al. Relationship of interleukin-6 and tumor necrosis factor-alpha with muscle mass and muscle strength in elderly men and women: the Health ABC Study. J Gerontol A Biol Sci Med Sci. 2002;57(5):M326–32. [DOI] [PubMed] [Google Scholar]
- 21.Bautmans I, Gorus E, Njemini R, Mets T. Handgrip performance in relation to self-perceived fatigue, physical functioning and circulating IL-6 in elderly persons without inflammation. BMC Geriatr. 2007;7:5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ferrucci L, Penninx BW, Volpato S, Harris TB, Bandeen-Roche K, Balfour J, et al. Change in muscle strength explains accelerated decline of physical function in older women with high interleukin-6 serum levels. J Am Geriatr Soc. 2002;50(12):1947–54. [DOI] [PubMed] [Google Scholar]
- 23.Leng S, Chaves P, Koenig K, Walston J. Serum interleukin-6 and hemoglobin as physiological correlates in the geriatric syndrome of frailty: a pilot study. J Am Geriatr Soc. 2002;50(7):1268–71. [DOI] [PubMed] [Google Scholar]
- 24.Toth MJ, Matthews DE, Tracy RP, Previs MJ. Age-related differences in skeletal muscle protein synthesis: relation to markers of immune activation. Am J Physiol Endocrinol Metab. 2005;288(5):E883–91. [DOI] [PubMed] [Google Scholar]
- 25.Lee ES, Kim SH, Kim HJ, Kim KH, Lee BS, Ku BJ. Growth Differentiation Factor 15 Predicts Chronic Liver Disease Severity. Gut Liver. 2017;11(2):276–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Liu X, Chi X, Gong Q, Gao L, Niu Y, Chi X, et al. Association of serum level of growth differentiation factor 15 with liver cirrhosis and hepatocellular carcinoma. PLoS One. 2015;10(5):e0127518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ost M, Igual Gil C, Coleman V, Keipert S, Efstathiou S, Vidic V, et al. Muscle-derived GDF15 drives diurnal anorexia and systemic metabolic remodeling during mitochondrial stress. EMBO Rep. 2020;21(3):e48804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Coll AP, Chen M, Taskar P, Rimmington D, Patel S, Tadross JA, et al. GDF15 mediates the effects of metformin on body weight and energy balance. Nature. 2020;578(7795):444–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kim M, Walston JD, Won CW. Associations Between Elevated Growth Differentiation Factor-15 and Sarcopenia Among Community-dwelling Older Adults. J Gerontol A Biol Sci Med Sci. 2022;77(4):770–80. [DOI] [PubMed] [Google Scholar]
- 30.Tsai VWW, Husaini Y, Sainsbury A, Brown DA, Breit SN. The MIC-1/GDF15-GFRAL Pathway in Energy Homeostasis: Implications for Obesity, Cachexia, and Other Associated Diseases. Cell Metab. 2018;28(3):353–68. [DOI] [PubMed] [Google Scholar]
- 31.Tsutsumimoto K, Doi T, Nakakubo S, Kim M, Kurita S, Ishii H, et al. Association between anorexia of ageing and sarcopenia among Japanese older adults. J Cachexia Sarcopenia Muscle. 2020;11(5):1250–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Spate U, Schulze PC. Proinflammatory cytokines and skeletal muscle. Curr Opin Clin Nutr Metab Care. 2004;7(3):265–9. [DOI] [PubMed] [Google Scholar]
- 33.Chen HW, Dunn MA. Arresting frailty and sarcopenia in cirrhosis: Future prospects. Clin Liver Dis (Hoboken). 2018;11(2):52–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Dasarathy S, Merli M. Sarcopenia from mechanism to diagnosis and treatment in liver disease. J Hepatol. 2016;65(6):1232–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Sinclair M, Gow PJ, Grossmann M, Angus PW. Review article: sarcopenia in cirrhosis--aetiology, implications and potential therapeutic interventions. Aliment Pharmacol Ther. 2016;43(7):765–77. [DOI] [PubMed] [Google Scholar]
- 36.Merli M, Giusto M, Molfino A, Bonetto A, Rossi M, Ginanni Corradini S, et al. MuRF-1 and p-GSK3beta expression in muscle atrophy of cirrhosis. Liver Int. 2013;33(5):714–21. [DOI] [PubMed] [Google Scholar]
- 37.Nishikawa H, Enomoto H, Ishii A, Iwata Y, Miyamoto Y, Ishii N, et al. Elevated serum myostatin level is associated with worse survival in patients with liver cirrhosis. J Cachexia Sarcopenia Muscle. 2017;8(6):915–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Baczek J, Silkiewicz M, Wojszel ZB. Myostatin as a Biomarker of Muscle Wasting and other Pathologies-State of the Art and Knowledge Gaps. Nutrients. 2020;12(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ghafouri B, Carlsson A, Holmberg S, Thelin A, Tagesson C. Biomarkers of systemic inflammation in farmers with musculoskeletal disorders; a plasma proteomic study. BMC Musculoskelet Disord. 2016;17:206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Serada S, Fujimoto M, Ogata A, Terabe F, Hirano T, Iijima H, et al. iTRAQ-based proteomic identification of leucine-rich alpha-2 glycoprotein as a novel inflammatory biomarker in autoimmune diseases. Ann Rheum Dis. 2010;69(4):770–4. [DOI] [PubMed] [Google Scholar]
- 41.Schafer C, Heiss A, Schwarz A, Westenfeld R, Ketteler M, Floege J, et al. The serum protein alpha 2-Heremans-Schmid glycoprotein/fetuin-A is a systemically acting inhibitor of ectopic calcification. J Clin Invest. 2003;112(3):357–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Kalabay L, Jakab L, Prohaszka Z, Fust G, Benko Z, Telegdy L, et al. Human fetuin/alpha2HS-glycoprotein level as a novel indicator of liver cell function and short-term mortality in patients with liver cirrhosis and liver cancer. Eur J Gastroenterol Hepatol. 2002;14(4):389–94. [DOI] [PubMed] [Google Scholar]
- 43.Herbst KL, Bhasin S. Testosterone action on skeletal muscle. Curr Opin Clin Nutr Metab Care. 2004;7(3):271–7. [DOI] [PubMed] [Google Scholar]
- 44.Tracz MJ, Sideras K, Bolona ER, Haddad RM, Kennedy CC, Uraga MV, et al. Testosterone use in men and its effects on bone health. A systematic review and meta-analysis of randomized placebo-controlled trials. J Clin Endocrinol Metab. 2006;91(6):2011–6. [DOI] [PubMed] [Google Scholar]
- 45.Hyde Z, Flicker L, Almeida OP, Hankey GJ, McCaul KA, Chubb SA, et al. Low free testosterone predicts frailty in older men: the health in men study. J Clin Endocrinol Metab. 2010;95(7):3165–72. [DOI] [PubMed] [Google Scholar]
- 46.Wu IC, Lin XZ, Liu PF, Tsai WL, Shiesh SC. Low serum testosterone and frailty in older men and women. Maturitas. 2010;67(4):348–52. [DOI] [PubMed] [Google Scholar]
- 47.Moctezuma-Velazquez C, Low G, Mourtzakis M, Ma M, Burak KW, Tandon P, et al. Association between Low Testosterone Levels and Sarcopenia in Cirrhosis: A Cross-sectional Study. Ann Hepatol. 2018;17(4):615–23. [DOI] [PubMed] [Google Scholar]
- 48.Sinclair M, Grossmann M, Hoermann R, Angus PW, Gow PJ. Testosterone therapy increases muscle mass in men with cirrhosis and low testosterone: A randomised controlled trial. J Hepatol. 2016;65(5):906–13. [DOI] [PubMed] [Google Scholar]
- 49.Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56(3):M146–56. [DOI] [PubMed] [Google Scholar]
- 50.Laube R, Wang H, Park L, Heyman JK, Vidot H, Majumdar A, et al. Frailty in advanced liver disease. Liver Int. 2018;38(12):2117–28. [DOI] [PubMed] [Google Scholar]
- 51.Merli M, Giusto M, Lucidi C, Giannelli V, Pentassuglio I, Di Gregorio V, et al. Muscle depletion increases the risk of overt and minimal hepatic encephalopathy: results of a prospective study. Metab Brain Dis. 2013;28(2):281–4. [DOI] [PubMed] [Google Scholar]
- 52.Thapaliya S, Runkana A, McMullen MR, Nagy LE, McDonald C, Naga Prasad SV, et al. Alcohol-induced autophagy contributes to loss in skeletal muscle mass. Autophagy. 2014;10(4):677–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Welch N, Dasarathy J, Runkana A, Penumatsa R, Bellar A, Reen J, et al. Continued muscle loss increases mortality in cirrhosis: Impact of aetiology of liver disease. Liver Int. 2020;40(5):1178–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Bhanji RA, Narayanan P, Allen AM, Malhi H, Watt KD. Sarcopenia in hiding: The risk and consequence of underestimating muscle dysfunction in nonalcoholic steatohepatitis. Hepatology. 2017;66(6):2055–65. [DOI] [PubMed] [Google Scholar]
- 55.Kwong AJ, Kim WR, Lake JR, Smith JM, Schladt DP, Skeans MA, et al. OPTN/SRTR 2019 Annual Data Report: Liver. Am J Transplant. 2021;21 Suppl 2:208–315. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.