2.1. Prevalence and Clinical Outcomes of Osteoporosis in Chronic Liver Disease

A number of studies on osteoporosis in chronic liver disease have focused on cholestatic liver disease, including primary biliary cholangitis (PBC), primary sclerosing cholangitis (PSC), and end-stage liver cirrhosis [18]. The prevalence of osteoporosis is reported to be 20–32% in PBC and 15% in PSC, with the severity of liver disease being a risk factor for in PBC patients [19,20,21]. About 10–20% patients with PBC experienced facture, and the risk of fracture in this patient group was 2-fold higher than in the general population [22,23]. Patient with non-cirrhotic chronic hepatitis B or C exhibited lower bone mineral density (BMD) and developed osteoporosis with a prevalence of 10–30% [24,25,26,27,28,29]. Hansen et al. reported that fracture risk was higher in HCV-exposed patients [29]. In those with liver cirrhosis, the prevalence of osteoporosis was 12–28%, with advanced-stage cirrhosis and alcohol-associated liver cirrhosis being significantly associated with osteoporosis [30,31,32,33]. Interestingly, liver transplantation improved BMD, especially in patients receiving less glucocorticoid treatment, without cholestasis, and exhibiting an elevation of vitamin D and parathyroid hormones 4–6 months after liver transplantation [34,35]. With regard to non-cholestatic liver disease, a retrospective study reported that osteoporotic fractures were 2.5-fold more common in patients with non-alcoholic fatty liver disease (NAFLD), and low BMD was more prominent in those with higher disease activity, such as in cases with non-alcoholic steatohepatitis (NASH), significant fibrosis, and a high fatty liver index [36,37,38,39]. As alcohol consumption is an independent risk factor for osteoporosis, patients with considerable alcohol consumption developed osteoporosis without liver cirrhosis [40]. About 30% and 36% of alcoholic patients showed osteoporosis and vertebral fracture upon radiologic examination, respectively [41]. Importantly, abstinence increased BMD and bone formation marker osteocalcin levels in alcoholics [42]. Osteoporosis was observed in 25–34% of patients with hereditary hemochromatosis, independent of cirrhosis or hypogonadism [43,44,45]. The prevalence of osteopenia and osteoporosis was 9% and 50%, respectively, in patients with Wilson’s disease, which was significantly higher than in the healthy population [46]. In addition, Quemeneur et al. reported that half of patients with Wilson’s disease suffered peripheral fractures 47].

2.2. Molecular Mechanism of Osteoporosis in Chronic Liver Disease

The pathophysiology of chronic liver disease-associated osteoporosis is complex. Increased bone resorption is an important pathological characteristic of osteoporosis, especially in patients with end-stage cholestatic liver disease, viral hepatitis, and NAFLD. Suggested molecular mechanisms underlying bone resorption in chronic liver disease include the receptor activator of nuclear factor kappa (RANK)-RANK ligand (RANKL)-osteoprotegerin (OPG) system, upregulation of proinflammatory cytokines, such as interleukin-1 (IL-1), IL-6, and tumor necrosis factor alpha (TNF-α), as well as low levels of testosterone [1,2,48,49]. Bone remodeling is tightly regulated by osteocytes and osteoblasts through different cytokines and hormones modulating the activation, resorption, reversal, formation, and termination phase [50]. Various types of signals are involved, including mechanical strain, bone damage, as well as hormone-induced bone remodeling via macrophage colony-stimulating factor (M-CSF), RANKL, and OPG secreted by osteoblasts. RANKL and OPG are members of the TNF superfamily and are crucial for the regulation of bone resorption [51]. RANKL, through its receptor RANK, activates osteoclast formation, activation, and survival, while OPG, which is another receptor for RANKL, restrains osteoclastogenesis and inhibits bone loss by binding RANKL to prevent the RANK-RANKL cascade [51]. RANKL is a type II transmembrane protein with a C-terminal extracellular domain. This ectodomain cleaved is cleaved by matrix metalloproteinases to yield soluble RANKL (sRANKL) in the extracellular environment and both membrane-bound and sRANKL bind to RANK [52]. In chronic liver disease, an imbalance between RANKL and OPG leads to high sRANKL levels, which increases bone turnover. Thus, the ratio of OPG/sRANKL might indicate a homeostatic response for bone mass preservation [53]. Moschen et al. reported that sRANKL levels were higher in patients with liver disease than those in controls, except those in the cirrhotic subgroup, while OPG levels were found to be proportional to the severity of liver disease and highest in the cirrhotic subgroup with osteoporosis and osteopenia, resulting in a greater OPG/sRNAKL ratio in the cirrhotic subgroup with osteoporosis and osteopenia than in cirrhotic patients with normal BMD. These results suggested that high sRNAKL levels corresponded to increased bone turnover in patients with liver disease, and that OPG was also increased to compensate for negative bone turnover. Therefore, the high OPG/sRANKL ratio could be explained a response to maintain bone homeostasis in these patients [54]. In addition, RANKL/OPG gene expression, indicative of osteoblast-related osteoclastogenesis, was increased in the serum of jaundice patients [55].

In a chronic inflammation state, proinflammatory cytokines, especially IL-6, IL-1, and TNF-α, contribute to osteoclast activation and subsequent bone resorption [56]. IL-6 and IL-1 directly modulated osteoclastogenesis by enhancing osteoclast function [57]. They was also shown to indirectly promote osteoclast activity by facilitating RANKL production in osteoblasts [56,57,58,59]. Experimental studies revealed that IL-6 was associated with disrupted osteogenesis of bone marrow stem cells in osteoporosis models, and suppression of the IL-6 receptor prevented osteoclast-mediated bone resorption [60,61]. As IL-6 increases during liver injury to stimulate liver regeneration, upregulated IL-6 could affect bone remodeling in various types of liver disease [62,63]. Additionally, ethanol seems to activate osteoclasts through the induction of IL-6 and TNF-α [64,65]. Another potent inflammatory cytokine, TNF-α, is also involved in inflammatory bone resorption by stimulating RANKL expression in osteoblasts and tissue stromal cells, in turn promoting osteoclast differentiation and activity [66]. In particular, TNF-α enhanced CSF-1 receptor gene expression during the initial stage of osteoclastogenesis and subsequently stimulated osteoblast precursors, which resulted in increased osteoclast formation independent of the RANKL pathway [67]. These proinflammatory cytokines had effects on osteoporosis in viral hepatitis and NASH. Gonzalez-Calvin et al. showed that serum soluble TNF receptor p55 levels were significantly higher in patients with viral cirrhosis with osteoporosis than those without osteoporosis and positively associated with bone resorption [68]. In addition, since obesity is considered a chronic inflammatory state, these proinflammatory cytokines contribute to osteoporosis in NAFLD [69]. Indeed, low BMD is significantly associated with NASH presenting elevated alanine aminotransferase (ALT) and C-reactive protein (CRP) than in simple steatosis [37]. Another study showed that TNF-α levels are elevated in pediatric NASH [70]. Furthermore, Kim et al. reported that liver fibrosis in NAFLD is significantly correlated with low BMD, suggesting an association between aggravation of hepatic inflammation, fibrosis, and bone loss in NAFLD patients [38]. These proinflammatory cytokines have also been proposed as involved in dysbiosis-induced osteoporosis associated with chronic liver disease. Altered short-chain fatty acid (SCFA) levels and increased gut permeability (“leaky-gut syndrome”) may affect bone remodeling by regulating inflammation and immune system [2]. In addition, impaired liver function and cholestasis result in decreased 25-hydroxylation and intestinal absorption of vitamin D. Vitamin D deficiency is associated with osteoporosis in patients with liver cirrhosis [30,33]. Calcium and vitamin D deficiencies in patients with cholestatic liver disease caused secondary hyperparathyroidism, subsequently enhancing bone resorption [71]. Vitamin K is known to be involved in osteoblast apoptosis inhibition and osteoclast differentiation [72,73,74]. Therefore, decreased vitamin K levels were suggested to affect bone metabolism in chronic liver disease. Vitamin K deficiency reduced bone matrix proteins such as osteocalcin and osteonectin in patients with PBC [75,76].

Bone formation is also compromised in chronic liver disease as a result of toxic materials, sclerostin, and decreased anabolic hormones, which contribute to osteoporosis in patients with PBC, advanced stage liver cirrhosis, hereditary hemochromatosis, and Wilson disease [1,2,48]. Direct or indirect toxic compounds such as bilirubin, alcohol, iron, and copper accumulate in specific liver diseases, which can impair bone formation by inhibiting osteoblast proliferation and differentiation as well as bone mineralization by osteoblasts [45,55]. Unconjugated bilirubin decreased the survival of osteoblast, and osteoblast differentiation was significantly reduced only in jaundiced patients, except in patients with normal bilirubin levels, whereas ursodeoxycholic acid compensated for the negative effect of cholestatsis on osteoblast survival, proliferation and mineralization [55,77]. Previous studies reported that the severity of liver disease including cholestasis is significantly associated with osteoporosis in patients with PBC [19,20,23]. In hereditary hemochromatosis, lumbar spine BMD is significantly decreased in parallel to an increase of iron and alkaline phosphatase (ALP) levels [45]. In addition, bone synthesis is lower in alcoholic patients with low levels of osteocalcin, which is secreted from osteoblasts, and plays a role in calcium homeostasis, bone matrix mineralization, and osteoblastic proliferation [53]. Under physiological conditions, after bone resorption, mesenchymal stem cells and early osteoblast progenitors differentiate into osteoblasts through Wnt, bone morphogenetic protein (BMP), and fibroblast growth factor (FGF) signaling, leading to bone formation [2,78]. As a regulator of bone formation, sclerostin is produced in osteocytes and hinders osteoblast differentiation and proliferation, subsequently restricting bone formation by antagonizing Wnt signaling via binding to low-density lipoprotein receptor-related proteins (LRP) 5/6 transmembrane receptors [79,80]. As a result of sclerostin expression, the Wnt receptor is blocked and glycogen synthase kinase 3 phosphorylates β-catenin, which is involved in ubiquitination and degradation through the proteasome pathway [81]. Guanabens et al. proposed sclerostin as a crucial regulator of the Wnt/β-catenin pathway in relation to bone formation in patients with PBC [82]. A cross-sectional study revealed that sclerostin levels were significantly increased in patients with advanced cirrhosis when compared to those with early cirrhosis or healthy controls [83].

Insulin-like growth factor (IGF-1), which is secreted from hepatocytes by growth hormone (GH) has an anabolic effect on bone growth by suppressing osteoblast apoptosis and enhancing osteoblastogenesis through stabilization of the Wnt/β-catenin pathway [84,85]. In addition, IGF-1 reduced bone resorption through the OPG and RANKL system [86]. In end-stage liver disease, hepatocellular dysfunction and reduced GH receptors lead to low serum IGF-1 levels, subsequently causing osteoporosis [87]. Insulin, an important hormone associated with NAFLD, also affects bone remodeling by activating collagen synthesis and stimulating osteoblast proliferation and differentiation [88]. In addition, hypogonadism, which results from hyperestrogenism of portal hypertension in males and suppression of the hypothalamic-pituitary-gonadal axis in females, is frequently observed in hemochromatosis, liver cirrhosis, and alcoholics liver disease [1,89]. Since testosterone directly modulates osteoblasts and osteocytes via the androgen receptor to stimulate trabecular bone formation and prevent its loss, low testosterone level owing to hypogonadism in male enhances osteoclast function and induces bone turnover [53,90]. Though estrogen levels are increased in patient with liver cirrhosis due to increased peripheral conversion of androgen to estrogen, altered estrogen metabolism in liver cirrhosis contributes to a decrease in degradation of estrogen metabolites [91]. Because the bone protective effect of these estrogens is weak, it is not enough to overcome post-menopausal osteoporosis in women and liver-disease related osteoporosis in men [92,93].

3.1. Definition of Sarcopenia

Sarcopenia, a condition characterized by the loss of skeletal muscle mass and strength, has been explored as a prognostic predictor for various diseases [94]. However, consensus criteria for the diagnosis of sarcopenia have not yet been established, and different definitions have been proposed by several groups [95,96]. The European Working Group on Sarcopenia in Older People (EWGSOP) first proposed the definition of sarcopenia in 2010, with muscle mass being a cardinal requirement for sarcopenia diagnosis [97]. The working group categorized sarcopenia into two categories. Primary sarcopenia referred to age-related muscle mass deterioration, while secondary sarcopenia was defined as caused by factors other than aging, such as inflammatory processes, disease-related disabilities, and inadequate energy or protein intake. Given the growing body of scientific data on sarcopenia, the EWGSOP recently updated diagnostic criteria so that muscle strength is the principal determinant for diagnosis with strict cut-off values (hand grip strength <27 kg for men and <16 kg for women) due to its significant correlation with clinical outcomes when compared to muscle mass [95]. According to the revised criteria, sarcopenia is defined as low muscle strength in parallel to low muscle mass or decreased muscle function. In contrast, the Asian Working Group for Sarcopenia (AWGS) regarded low muscle mass as the cardinal criterion for the diagnosis of sarcopenia. Therefore, AWGS defined sarcopenia as low muscle mass with low muscle strength or low physical performance [96]. Further, both EWGSOP2 and AWGS recognize entirely impaired muscle mass, muscle strength, and muscle function as diagnostic criteria for severe sarcopenia [95,96]. On the other hand, Clark et al. proposed in 2008 the concept of dynapenia, a state of age-associated decline in muscle strength which focused on other physiologic factors except for muscle mass loss in 2008 [98]. Therefore, dynapenia is similar to the revised EWGSOP definition of sarcoepnia in that decrease of skeletal muscle mass was not always necessary for the diagnosis of dynapenia, but sarcopenia is recognized to develop earlier in life, unlike dynapenia.

3.2. Prevalence and Clinical Outcomes of Sarcopenia in Chronic Liver Disease

Sarcopenia results from the accelerated loss of muscle mass and function, which in turn contributes to adverse clinical outcomes including fracture, frailty, and mortality. Although the prevalence of sarcopenia depends on the definition used, which is based on, for example, muscle mass cutoff points as well as other factors, the condition has been rigorously studied in the context of various chronic liver diseases [94]. With regard to non-alcoholic liver disease, previous studies reported a significant relationship between sarcopenia, sarcopenic obesity, and NAFLD, with this association being later described as independent of obesity and insulin resistance [99,100,101,102,103].

In addition, a longitudinal cohort study revealed that increased skeletal mass during the study period had a beneficial effect against the development of NAFLD or improved existing NAFLD at baseline [104]. Muscle mass loss was observed in 40–70% of liver cirrhosis patients, and most studies reported negative influences of sarcopenia on mortality and liver cirrhosis complications, such as hepatic encephalopathy [105,106,107,108,110,112,113]. The degree of sarcopenia was associated with Child-Pugh score in patients with cirrhosis, and sarcopenia was a predictor of survival independently or in combination with the model for end-stage liver disease (MELD) score, especially in patients with low scores (<15) [109,111,114]. Unfortunately, after liver transplantation, sarcopenia did not improve and even worsened because of immunosuppressive drugs such as steroids, calcineurin inhibitors, and mammalian target of rapamycin (mTOR) inhibitors [6]. Sarcopenia also had significant impacts on the development of diabetes mellitus, the risk of infection, the length of hospitalization, and mortality in patients who underwent liver transplantation [115,116,117,118,119,120,121,122]. Further, sarcopenia was shown to independently predict mortality, overall survival, and recurrence-free survival in patients with hepatocellular carcinoma [123,124,125].

3.3. Molecular Mechanism of Sarcopenia in Chronic Liver Disease

In normal physiology, skeletal muscle protein turnover is maintained as a balance between protein synthesis and breakdown. mTOR is an essential modulator of translational control, majorly involved in protein synthesis [126]. Exercise, branched chain amino acids (BCAAs), as well as various hormones, including testosterone, insulin, and IGF-1, activate the mTOR pathway in muscle cells through protein kinase B, which subsequently triggers several intracellular pathways for muscle protein synthesis, including p70 ribosomal S6 kinase 1 (S6K1) and eukaryotic initiation factor 4E binding protein (4E-BP1) [126]. In addition, the proliferation of muscle satellite cells, which are muscle fiber precursors, is critical for muscle growth and is activated in response to IGF-1 and BCAAs through protein kinase B [126,127].

In contrast, myostatin, a transforming growth factor beta (TGF-β) superfamily member produced in muscle, inhibits protein synthesis by maintaining satellite cells in a quiescent state, activating SMAD family transcription factors 2 and 3 (SMAD 2/3), and stimulating proteolysis via forkhead box O transcription factors (FoxOs) associated with the ubiquitin-proteasome pathway (UPP) and autophagy [128,129]. Impaired mitochondrial function, insulin resistance, and reactive oxygen species (ROS) also stimulate autophagy.