Background
Sarcopenia, defined as the loss of skeletal muscle mass and strength, is a common complication of liver cirrhosis (LC) [
1]. Originally, the term ‘sarcopenia’ was used to describe age-related decreases in muscle mass [
2]. Later, the European Working Group on Sarcopenia in Older People (EWGSOP) defined sarcopenia as a syndrome characterized by decreases in both skeletal muscle mass and strength, which are associated with physical disability, poor quality of life (QOL), and high mortality [
3‐
6]. Furthermore, the Working Group classified sarcopenia into two categories as follows: ‘primary’ (or age-related), when sarcopenia is caused by aging itself; and ‘secondary’, when one or more other causes are evident, such as chronic debilitating disease, including LC. In patients with LC, secondary sarcopenia is an independent predictor of minimal hepatic encephalopathy and poor prognosis [
7,
8]. However, with advances in the conservative treatment for LC, patients live to a greater age than before, and are also aging among Japan’s graying population. Therefore, sarcopenia is increasingly prevalent and attracts attention especially in aging patients.
The definition of sarcopenia is changing with time and varies according to which academic society diagnostic criteria are used. In Europe, the EWGSOP established the definition of sarcopenia in 2010 and updated the criteria in 2018 (EWGSOP2) based on accumulated evidence [
3,
9]. Asia has a large, rapidly aging population; therefore, the Asian Working Group for Sarcopenia (AWGS) developed the definition of sarcopenia for Asian people in 2014 [
10]. These definitions included impaired physical performance and loss of muscle mass and strength in the aged population. Meanwhile, the Japan Society of Hepatology (JSH) proposed sarcopenia criteria for patients with chronic liver disease (CLD) in 2015 [
1]. The JSH criteria use identical cut-off values for muscle mass and strength as the AWGS criteria, but the age-related criterion and assessment of physical performance (gait speed) are omitted. Different definitions, depending on which criteria are used (JSH, AWGS, or EWGSOP2), may alter the diagnosis of sarcopenia, resulting in the variation in disease prevalence. Therefore, research to analyze the actual situation based on several sarcopenia criteria is required in order to take preventative measures against sarcopenia in patients with LC.
Osteoporosis, a metabolic bone disorder characterized by compromised bone strength, is a common complication in patients with LC and affects the QOL due to chronic pain and immobility [
11]. Reportedly, the prevalence of osteoporosis among patients with LC varies from 12 to 55% in the West [
12], whereas the prevalence remains uncertain in Japan. Patients with osteoporosis are susceptible to fractures of bones, such as vertebrae, the femoral neck, and peripheral bones. Specifically, vertebral fracture occurs frequently in patients with LC, and its prevalence ranges from 7 to 35% [
11]. However, as vertebral fracture often develops without symptoms, the prevalence of symptomatic and asymptomatic vertebral fracture in Japanese patients remains unclear.
A previous report demonstrated the relationship between sarcopenia and osteoporosis in a Japanese elderly community-based population [
13]. The prevalence of sarcopenia among subjects aged ≥60 years was 8.2%, and the cumulative incidence of sarcopenia was 2.0% per year. The prevalence of osteoporosis among patients with sarcopenia was 57.8%, which was significantly higher than those without sarcopenia [
13]. On the other hand, the prevalence of osteoporosis among subjects aged ≥60 years was 24.9%, and the prevalence of sarcopenia among those with osteoporosis was 19.1%. Moreover, osteoporosis was an independent predictor of the occurrence of sarcopenia [
13]. Sarcopenia and osteoporosis are interrelated and closely linked in terms of common risk factors and biological pathways; therefore, the term ‘osteosarcopenia’ (originating from the term ‘sarco-osteopenia’) was defined as when the two diseases coexist [
14,
15]. Patients with osteosarcopenia have a higher risk of falls, fractures, and frailty [
14‐
16].
Therefore, comprehensive assessments and strategies for skeletal muscle and bone disorders, such as sarcopenia and osteoporosis, are essential to improve QOL and morbidity in patients with LC. However, there are few reports evaluating the association between low skeletal muscle mass and strength, osteoporosis, and vertebral fracture in the same individuals with LC. The aim of this study was to investigate the actual situation of sarcopenia according to the JSH, AWGS, and EWGSOP2 criteria and to clarify the relationship between sarcopenia, osteoporosis, osteosarcopenia, and vertebral fractures in Japanese patients with LC.
Discussion
In this study, we compared the prevalence of sarcopenia using three different diagnostic criteria. The cut-off values for low muscle mass in the JSH criteria were identical to those in the AWGS criteria and similar to those in the EWGSOP 2 criteria. Physical performance was included as an essential requirement in the AWGS criteria and used as an indicator of disease severity in the EWGSOP2 criteria, whereas it was omitted in the JSH criteria. We found that the prevalence of sarcopenia was 33.8% for both the JSH and AWGS criteria and 28.2% for the EWGSOP2 criteria. Notably, the diagnostic outcome using the JSH criteria was identical to that using the AWGS criteria. Furthermore, male patients diagnosed with sarcopenia were same using any of the three criteria. Among 42 patients with low physical performance diagnosed using the AWGS or EWGSOP2 criteria, 36 (85.7%) had low muscle strength and the remaining 6 (14.3%) did not show low muscle mass and thus did not reach a diagnosis of sarcopenia (data not shown). In future, these different diagnostic criteria should be harmonized to achieve more universal definitions.
Although each patient’s daily exercise was not fully quantified in the present study, most patients appeared to lack an exercise routine, and these circumstances may have increased the apparent prevalence of sarcopenia and osteoporosis. Moreover, as HCV infection was highly prevalent in the area (Fuji city) in the beginning of the 1900s, patients with HCV-related LC frequently reach an advanced age. In contrast, alcohol-induced LC is most common among newly diagnosed LC patients. Therefore, alcoholic patients were significantly younger than HCV patients (Additional file
1: Table S7). Furthermore, as a result of the rapid exacerbation of alcoholic liver dysfunction, decompensated LC developed more frequently in alcoholic patients than in HCV patients. In such situations, the present criteria proposed by the JSH have some limitations for diagnosing sarcopenia or evaluating disease severity in a heterogeneous population; these include younger patients with severe liver impairment and older patients with less severe liver impairment. Therefore, revisions of the diagnostic criteria, such as stratification by age, disease stage, and/or etiology, are necessary for the accurate diagnosis and assessment of sarcopenia.
In the present study, decreased levels of BCAAs and IGF-1 were significant independent factors associated with sarcopenia. Branched-chain amino acids, consisting of leucine, isoleucine, and valine, are essential for increasing and maintaining muscle mass [
20,
21]. As the liver is a multifunctional organ involved in carbohydrate, protein, and lipid metabolism, LC is complicated by PEM and hyperammonemia and can lead to the consumption of BCAAs by skeletal muscles for energy production and ammonia metabolism [
22]. IGF-1, produced by hepatocytes and myocytes, is involved in muscle protein synthesis, and the mammalian target of rapamycin (mTOR), activated by protein kinase B (PCK/AKT), stimulates protein synthesis [
20]. The mTOR pathway is activated by IGF-1, BCAAs (particularly leucine), and exercise. The proliferation of satellite cells, which are the precursors of new muscle fibers, is essential for muscle growth. Satellite cell activation is stimulated by protein kinase B and promoted by IGF-1 and BCAAs. On the other hand, the proliferation of satellite cells is suppressed by myostatin, which is a cytokine belonging to the transforming growth factor-β family. The IGF-1 signaling pathway inhibits myostatin and stimulates muscle growth. Reportedly, higher myostatin levels correlate with a loss of muscle mass and reduced survival in patients with LC [
23]. Taken together, our results are theoretically reasonable and support the notion that decreased levels of both BCAA and IGF-1 are associated with the development and progression of sarcopenia in patients with LC.
Since osteoporosis is one of the most common complications of LC, the term ‘hepatic osteodystrophy’ is often used to describe bone disorders in subjects with CLD. According to the WHO criteria, the prevalence rates of osteoporosis, osteopenia, and normal BMD in the present study were 34.5, 40.1, and 25.4%, respectively (data not shown). Although the pathogenesis of osteoporosis is not entirely understood, an imbalance in bone remodeling is known to affect the development of bone loss in CLD. IGF-1 is essential for bone remodeling and the maintenance of bone mass and strength due to its stimulation of osteoblast differentiation and proliferation, and its regulation of diaphyseal growth [
24,
25]. The present study also showed that low IGF-1 levels were associated with osteoporosis, although there was no statistical significance in the multivariate analysis.
The prevalence of Child-Pugh class B/C (decompensated LC) and the values of total bilirubin and PT-INR were higher in non-osteoporosis patients than in osteoporosis patients. One possible explanation for these paradoxical results is that the etiology components differed between the two groups. Hepatitis C virus was the most frequent etiology (42.9%) in osteoporosis patients, whereas alcohol was the most frequent etiology (43.0%) in non-osteoporosis patients. As described above, alcoholic patients were significantly younger than HCV patients and more frequently developed decompensated LC. As the total bilirubin and PT-INR values were significantly higher in alcoholic patients than in HCV patients, a higher frequency of alcoholic etiology may cause higher total bilirubin and PT-INR values in non-osteoporosis patients.
Sarcopenia and rapid skeletal muscle wasting are associated with mortality and reduced QOL in patients with LC [
8,
26,
27]. Osteoporosis predisposes patients to fragility fractures, which affect both morbidity and QOL, and the early diagnosis and treatment of sarcopenia and osteoporosis are important. Recently, the term ‘osteosarcopenia’ was defined when sarcopenia and osteoporosis coexist [
14,
15]. Reportedly, the prevalence of osteosarcopenia was 28.7% in patients aged ≥60 years with hip fractures, and the 1-year mortality in osteosarcopenia patients was higher than that in other groups: normal, 7.8%; osteoporosis only, 5.1%; and sarcopenia only, 10.3% [
28]. A study on community-dwelling Chinese elders showed that the prevalence of osteosarcopenia was 10.4% in males and 15.1% in females [
29], and that patients with osteosarcopenia were more susceptible to fragility fractures, frailty, and mortality [
28‐
30]. In our study, the prevalence of osteosarcopenia was 21.8% (31/142) for all patients; 15.6% (14/90) for males and 32.7% (17/52) for females. The prevalence of vertebral fractures in osteosarcopenia patients was 61.3% (31/19), which was the highest among the four patient groups. Thus, osteosarcopenia may increase the frequency of vertebral fractures. Importantly, more than half (24/41; 58.5%) of the patients with vertebral fracture were newly diagnosed in the present study; these patients were either asymptomatic or had no opportunity to evaluate the vertebral fracture using spinal X-rays. Therefore, we should carefully interview patients regarding fracture-related symptoms and assess vertebral fracture using radiological imaging tests, especially in LC patients with osteosarcopenia.
Regarding the association between muscle and bone, recent reports have shown that sarcopenia is independently associated with low BMD in patients with CLD [
31,
32].
In the present study, the SMI and handgrip strength values were significantly correlated with the BMD of the lumbar spine, femoral neck, and total hip. In the ROC curve analysis, the SMI cut-off values for predicting osteoporosis were 7.05 kg/m2 in males and 5.88 kg/m2 in females. Similarly, the handgrip strength cut-off values for predicting osteoporosis were 27.9 kg in males and 20.1 kg in females. Intriguingly, these cut-off values almost coincided with those for the sarcopenia diagnostic criteria, which were proposed by the JSH, AWGS, and EWGSOP2. These findings suggest that a diagnosis of sarcopenia may be useful for predicting the presence of osteoporosis in patients with LC.
Recently, the EWGSOP proposed the concept of malnutrition-associated sarcopenia whereby a sarcopenia phenotype is related to malnutrition irrespective of the cause (reduced food intake, low nutrient bioavailability, or high nutrient requirements including inflammatory diseases such as CLD and malignancy) [
9,
33,
34]. The definition of malnutrition by the Global Leadership Initiative on Malnutrition (GLIM) recommends low muscle mass as one of its criterion items [
9,
34]. Hence, low muscle mass is a common feature in both malnutrition and sarcopenia. Reportedly, patients with sarcopenia showed an increased risk of malnutrition, and conversely, hospitalized older patients with malnutrition were more susceptible to sarcopenia [
35]. As the liver is an organ central to nutrient metabolism, malnutrition and sarcopenia are frequently observed and closely linked together in patients with LC. Therefore, assessment of the nutrition status and malnutrition risk is important to ascertain the causes of loss muscle mass and its pathological mechanisms.
Branched-chain amino acid supplementation stimulates albumin and protein synthesis in skeletal muscle and has the potential to improve the prognosis of LC patients with sarcopenia [
8,
21]. In particular, the administration of leucine-enriched foods activates the mTOR pathway and increases muscle protein synthesis [
36]. Reportedly, decreased physical activity and insufficient energy intake are associated with sarcopenia in patients with LC, regardless of disease progression [
37]. Therefore, exercise regimens involving walking ≥5000 steps/day with a total energy intake of approximately 30 kcal/ideal body weight are recommended [
37]. In another report, moderate physical exercise together with leucine supplements improved exercise capacity, leg muscle mass, and health-related QOL [
38]. Regarding the pharmacological treatment of osteoporosis, weekly alendronate or monthly ibandronate has been shown to increase bone mass in PBC patients. Cyclical etidronate, a bisphosphonate with an antiresorptive effect, increases bone mass in female patients with LC [
39]. However, the majority of previous studies only included a small number of patients and predominantly targeted PBC patients or transplant recipients [
40‐
42]. Thus, the research regarding osteoporosis treatment for patients with LC has not reached a definite conclusion. A large-scale trial for new drugs is needed in order to establish the treatment for osteoporosis in patients with LC.
This study had some limitations. First, we used the BIA method for the assessment of muscle mass, as per the JSH criteria recommendation (as well as CT). BIA equipment, although safe, non-invasive, and easy to use, does not measure muscle mass directly and is sensitive to patients’ conditions, such as hydration and ascites [
43]. In addition, the CT method needs proprietary software for analysis. Second, this study did not include a nutritional intake assessment. In future, conducting a nutritional intake assessment by an expert nutritionist will aid in the investigation of the relationship between malnutrition and sarcopenia.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.