Sarcopenia from mechanism to diagnosis and treatment in liver disease

Abstract

Sarcopenia or loss of skeletal muscle mass is the major component of malnutrition and is a frequent complication in cirrhosis that adversely affects clinical outcomes including survival, quality of life, development of other complications and post liver transplantation survival. Radiological image analysis is currently used to diagnose sarcopenia in cirrhosis. Nutrient supplementation and physical activity are used to counter sarcopenia but have not been consistently effective because the underlying molecular and metabolic abnormalities persist or are not influenced by these treatments. Even though alterations in food intake, hypermetabolism, alterations in amino acid profiles, endotoxemia, accelerated starvation and decreased mobility may all contribute to sarcopenia in cirrhosis, hyperammonemia as a possible mediator of the liver-muscle axis has recently gained attention Increased muscle ammonia causes: cataplerosis of α-ketoglutarate, increased transport of leucine in exchange for glutamine, impaired signaling by leucine, increased expression of myostatin, a TGFβ superfamily member and an increased phosphorylation of eukaryotic initiation factor 2α, mitochondrial dysfunction, increased reactive oxygen species that decrease protein synthesis and increased autophagy mediated proteolysis. These molecular and metabolic alterations may contribute to the anabolic resistance and inadequate response to nutrient supplementation in cirrhosis. Central and skeletal muscle fatigue contributes to impaired exercise capacity and responses. Use of proteins with low ammoniagenic potential, leucine enriched amino acid supplementation, long term ammonia lowering strategies and a combination of resistance and endurance exercise to increase muscle mass and function may target the molecular abnormalities in the muscle. Strategies targeting endotoxemia and the gut microbiome need further evaluation.

Introduction

Malnutrition in liver disease has been used for decades to describe the phenotype of skeletal muscle loss with or without loss of fat mass[1]. The majority of “malnourished” patients with cirrhosis experience skeletal muscle wasting or sarcopenia, a major predictor of adverse clinical outcomes[2–4]. Although alterations in body composition in cirrhosis have been reported using a number of methods, radiographic image analysis is believed to be the most precise technique to quantify muscle mass and define sarcopenia[5, 6]. Over the past few years, a number of investigators have reported that sarcopenia occurs in 30–70% of cirrhotic patients[2, 7–10]. The clinical significance of sarcopenia in liver disease, primarily cirrhosis, is due to the high prevalence and adverse impact on clinical outcome measures including survival, quality of life, development of other complications of cirrhosis, and post liver transplant outcomes[1, 4, 10–14]. Etiology and severity of the underlying liver disease, duration of illness, age and co-morbidities contribute to the severity of sarcopenia[1, 4, 9, 15, 16]. Despite being widely recognized as a major complication of cirrhosis, most therapies to date have focused on the principle of “deficiency replacement” rather than targeted treatments and have generally been ineffective[17]. Therapies have focused on nutritional supplementation because reduced dietary intake was believed to be the major cause of malnutrition and sarcopenia but these approaches have been frequently inadequate in improving survival[18–20]. An integrated metabolic-molecular approach in a comprehensive array of models has shown that hyperammonemia is a mediator of the liver muscle axis[21, 22]. Physical activity has been suggested to improve functional capacity but the effect on skeletal muscle mass is still unclear [23]. In recent years, a combination of sarcopenia with obesity is being increasingly recognized, especially in patients with non-alcoholic fatty liver disease (NAFLD) and post liver transplantation, but whether sarcopenia is mechanistically related to obesity and NAFLD is still controversial[24, 25]. The major deficiency in the field of sarcopenia in cirrhosis is the lack of understanding of the mechanisms involved. A number of excellent recent reviews have described the clinical relevance of sarcopenia in cirrhosis but have not focused on the possible mechanisms and on the relevance of novel therapeutic targets that have the potential for clinical translation[1, 17, 26–29].

In the present review we will provide an overview of the clinical relevance of sarcopenia in liver cirrhosis, but the emphasis will be on the possible molecular and metabolic perturbations involved and the promising novel therapeutic approaches that could be made possible by these discoveries.

Diagnosis of Sarcopenia in cirrhosis

Most studies to date have used the term “malnutrition” to identify primarily skeletal muscle loss determined by one or more criteria that are not always uniform or precise and an alteration in energy metabolism and potentially fat mass depletion. The diagnosis of skeletal muscle loss requires analysis of the body composition using one or more of a number of available techniques (Table 1) as well as the normal values to define the appropriate cut-off values for sarcopenia[3, 6, 29]. Even though few studies have directly compared different methods, computed tomography (Supplementary Figure 1) with one of the image analysis programs is being increasingly used since skeletal muscle can be directly viewed and quantified[5, 10, 30–33]. Magnetic Resonance Imaging (MRI) has also been proposed as a valuable method although objective data in cirrhosis are scarce[34]. Although abdominal CT and MRI scans would be difficult to justify for quantifying muscle mass due to the cost and/or radiation exposure, most cirrhotic patients have surveillance scans for liver focal lesions, hepatocellular carcinoma, vascular disease and pre-transplant evaluation.

Table 1.

Methods to quantify skeletal muscle evaluation in cirrhosis

	Methods for quantification
Single muscle or Groups of muscle	Anthropometry, DEXA, bioelectrical impedance analysis, impedance plethysmography, ultrasonography, CT or MRI,
Quality of muscle	CT scan attenuation
Muscle function	Handgrip strength
Fiber type	Muscle biopsy
Contractile function	Measurement of Maximum strength, Maintenance of strength, Fatigability

Muscle mass depends on gender (lower in females) and age (lower with increasing age), cutoff values for gender and age have been recently reported[32]. Handgrip strength (a measure of muscle function) has been utilized in cirrhotic patients but when normalized for body mass index may not be accurate in cirrhosis due to fluctuations in body weight related to changes in body water.

Quantifying muscle mass by measurements in a single anatomic area like the limb or abdominal muscles are believed to provide a reasonably accurate measure of whole body muscle mass[35]. In cirrhosis, as in most chronic disease, a preferential loss of type II or fast fibers is expected but in vivo measurements of fiber type loss in cirrhotic patients is still lacking. Appendicular muscle mass (limb muscles) is strongly influenced by the activity level. Measurements of psoas and abdominal muscle mass on CT images at L3 or L4 vertebra are used due to their relative independence from the activity level however these muscles contain both type I and type IIA fibers[36] which also needs to be considered. Another consideration is the quality of skeletal muscle that has been reported based on the CT attenuation that is lower in the muscles of cirrhotics compared to controls[31] and is indicative of fatty infiltration with adverse clinical outcomes[37, 38]. Whether muscle quality can be determined by measuring contractile function or by the CT attenuation values need to be ascertained (Table 2). The possible impact of these parameters on clinical outcomes has not been systematically evaluated.

Table 2.

Sarcopenia adversely impacts outcome in cirrhosis.

Author (year)	N	Method to define sarcopenia	Outcome
Wang (2016)[39]	292	Effect of grip strength, muscle mass, muscle quality, SPPB on transplant wait list mortality	Grip strength (HR 0.74), SPPB (HR 0.89), muscle quality (0.77) but not muscle mass (0.91) decreased survival
Kalafateli (2016)[40]	232	L3 psoas area and Royal Free Hospital Global Assessment CT measure of psoas muscle area at L3/L4 CT measure of psoas at umbilicus	Post OLT infection (OR 6.55), ventilator requirement (OR 8.5), ICU stay >5 d (OR 7.46) higher in sarcopenic patients.
Hanai (2016)[41]	149		Greater rate of muscle (>3.1%/year) loss increases mortality (HR 2.73)
Durand (2014)[33]	376		Increased mortality for each unit decrease in muscle area
Kim (2014)[42]	89		Increased mortality hazard risk 5.4 for sarcopenia
Masuda (2014)[43]	204		Sarcopenia 2 fold increased risk of death, 5.3 fold increased risk of sepsis
DiMartini (2013)[44]	338		Increased mortality only in men for each unit decrease in skeletal muscle index
Montano-Loza (2012)[10]	112		Increased mortality hazard risk 2.26 for sarcopenia
Tandon (2012)[7]	142		Increased mortality, hazard risk 2.36 for sarcopenia
Englesbe (2010)[45]	163		Lower post OLT survival, HR 3.7/1000mm2 psoas area.
Hamaguchi (2014)[46]	200		Median post OLT survival in sarcopenic patients 17.6 m and in non-sarcopenic patients 33.9 m
Hara (2016)[47]	161	Bioelectrical impedance analysis	73 deaths over mean 1005 days follow up
Kaido (2013)[48]	124		Post living donor transplant lower with sarcopenia
Selberg (2002)[142]	305		Survival lower with phase angle <5.4°
Merli (2010)[9]	38	Anthropometrics (MAMA/TSF)	MAMC <5th percentile: relative risk of death 1.79.
Shahid (2005)[49]	61		Increased postoperative mortality.
Lai (2014)[50]	50	Frailty index, SPPB	45% greater mortality for each point increase in frailty index 19% increase in mortality for each point decrease in physical performance
Carey (2010)[51]	294	6 min walk test	Each 100m reduction in 6 min walk test reduces survival hazard risk 0.48
Alvares da-Silva (2005)[52]	121	Hand grip, anthropometrics	Increased mortality for lower hand grip strength

Clinical impact of sarcopenia in cirrhosis

A number of cross sectional and longitudinal studies using different methods to quantify muscle mass have reported that median survival and probability of survival are lower in patients who have cirrhosis with sarcopenia than those without sarcopenia (Table 2)[7, 9, 10, 33, 39–52]. Some of these reports suggest that sarcopenia adds to the prognostic value of the MELD scoring system[40, 53]. The cause(s) of higher mortality is however not as evident though both increased risk of infection and encephalopathy may be contributory factors[54]. Sarcopenia may also impair diaphragmatic work due to reduced muscle mass and this event may favor pulmonary complications especially in the context of surgery (liver resection or liver transplantation).

Sepsis related mortality is higher in sarcopenic than non-sarcopenic cirrhosis [10, 13, 55]. For appropriate antibody and cytokine responses, adequate amino acid supply is necessary that is impaired when skeletal muscle mass is decreased but a direct causal or mechanistic link between sarcopenia and impaired immune function has not been shown[56]. Furthermore, it is also possible that factors that cause sarcopenia, including hormonal and biochemical alterations as well as circulating endotoxins, also contribute to the impaired immune function and increase the risk of infection. Lack of mobility or frailty in sarcopenia may also play a role[57]. Interestingly, cirrhotic patients with refractory ascites seem particularly prone to malnutrition and sarcopenia. Ascites is known to increase resting energy expenditure[58] while food intake is decreased due to raised abdominal pressure and early satiety. Treating refractory ascites by transjugular intrahepatic portosystemic shunt (TIPS) has been shown to improve body composition in malnourished cirrhotic patients[6, 31].

Quality of life is lower in cirrhosis but whether this is due to the loss of muscle mass or impaired contractile function and consequent limited mobility, or increased risk of other complications in sarcopenic cirrhotics is still a field that needs well-designed studies [1, 11, 12]. All domains of the quality of life are lower in malnourished patients when measures that primarily quantify skeletal muscle mass are utilized[1].

Hepatocellular carcinoma (HCC) is a frequent complication in the natural history of chronic liver disease and recent studies have reported that sarcopenia is an independent prognostic factor decreasing survival and increasing treatment related mortality in patients with HCC[37, 59].

Liver transplantation is currently the definitive therapy to cure end stage liver disease and sarcopenia adversely impacts outcomes in patients on the transplant list, in the peri-transplant period and post transplantation[7, 9, 45, 60]. Survival is lower in sarcopenic cirrhotic patients before liver transplantation while increased length of hospitalization, prolonged ICU stay, and longer time of intubation have been reported after transplantation compared to patients without sarcopenia [9, 27, 45].

It is important to emphasize that clinical outcomes depend on other factors also, but sarcopenia is recognized as a major contributor to adverse outcomes in the management of the cirrhotic patient undergoing liver transplantation.

Mechanisms of skeletal muscle loss in cirrhosis

Alterations in protein turnover, energy disposal and metabolic changes induce muscle depletion in cirrhotic patients

As seen above, a number of studies and reviews have provided descriptive data on the high prevalence and adverse clinical impact of sarcopenia in cirrhosis[1, 4, 7, 10, 26, 60]. Skeletal muscle is the major protein store in the human body[61]. Skeletal muscle mass is maintained by a balance between protein synthesis, protein breakdown and regenerative capacity regulated by muscle satellite cell function[1]. Broadly, two types of studies have contributed to the current understanding of the pathogenesis of sarcopenia in cirrhosis: metabolic-tracer kinetics and molecular-signaling pathway studies[21, 22, 62–66]. An integrated approach using both strategies to examine how metabolic perturbations alter molecular signaling and vice-versa has allowed identification of novel potential therapeutic targets.

Whole-body turnover studies using labeled phenylalanine and leucine as primed constant infusion have yielded conflicting results with unaltered, increased or decreased protein breakdown and protein synthesis[62–64]. Arteriovenous difference studies and release of 3-methylhistidine to quantify protein synthesis and breakdown suggest impaired skeletal muscle protein synthesis[67]. Explanations for these conflicting observations included heterogeneity in etiology, duration, age, and severity of liver disease. Heterogeneity in methods used to determine protein turnover and in the contribution of different organs to whole body turnover also explain these differences. Whole-body substrate utilization studies using indirect calorimetry have shown that cirrhosis is a state of accelerated starvation because fatty acid oxidation and gluconeogenesis are increased early in the postabsorptive or fasting state[30, 68, 69]. Since glucose is a preferred substrate in many tissues, and fatty acid carbon cannot be used for gluconeogenesis, amino acids are used for gluconeogenesis[70]. The primary source of amino acids for gluconeogenesis is proteolysis in the skeletal muscle that generates both aromatic and branched chain amino acids (BCAA). Only BCAA are catabolized in the skeletal muscle due to the localization of the branched chain ketodehydrogenase and oxidation of the carbon skeleton as an energy source[71]. As a consequence, plasma BCAA concentrations are lower in cirrhotic patients. In contrast, aromatic amino acids are primarily metabolized in the liver but due to both portosystemic shunting and hepatocellular dysfunction, their plasma concentrations are increased in chronic liver disease[62, 72–75]. This interpretation that accelerated starvation and increased gluconeogenesis are bioenergetics perturbations in cirrhosis is supported by the low respiratory quotient in sarcopenia cirrhotics[30]. Most therapies to date have focused on treating the amino acid imbalance rather than targeting the mechanisms that contribute to these alterations that finally result in sarcopenia.

Potential mediators of the liver – muscle axis in cirrhosis

One of the major reasons for the very limited understanding of sarcopenia in cirrhosis has been the difficulty in identifying the mediator(s) of the liver-muscle axis. A number of potential mediators have been proposed including increased ammonia, decreased testosterone and growth hormone, and endotoxemia[21, 22, 76, 77]. Even though there is evidence to support each of these potential mediators, hyperammonemia has been studied most extensively[17, 21, 22, 78].

Of the hepatic metabolic functions, ammonia disposal by ureagenesis is critical. Both hepatocellular dysfunction and portosystemic shunting that are components of the pathophysiological changes in cirrhosis contribute to impaired ureagenesis[79]. Ammonia is generated by a number of mechanisms including amino acid metabolism, purine metabolism, enterocyte glutaminase activity and urealysis in the gut[80]. Neurotoxicity is the best-studied cytotoxic effect of ammonia[80, 81]. Independent investigators have reported increased skeletal muscle ammonia uptake and conversion to glutamate and glutamine in patients and models of liver disease[82–85]. Despite the well recognized cytotoxic effects of ammonia in the neurons and astrocytes, skeletal muscle effects have only been recently reported[21, 22, 86, 87]. Studies in human skeletal muscle, the hyperammonemic portacaval anastomosis (PCA) rat, mice during hyperammonemia and in vitro studies in myotubes cultures suggest that ammonia accumulates in the skeletal muscle and activates a program of molecular alterations that contribute to sarcopenia[21, 22, 86, 87]. Even though mechanism of entry of ammonia into the skeletal muscle has not been well studied, ammonia transporters including the Rh B and C proteins are expressed in the muscle [88]. Following entry, ammonia activates a series of signaling responses whose exact mechanisms are at yet unclear.

Hyperammonemia contributes to muscle depletion: intracellular signaling

In murine myotubes and murine cells cultures the response to hyperammonemia-mediated activation of p65NFkB is an increased expression of myostatin, a TGFβ superfamily member (Figure 1)[22, 89]. Increased expression of myostatin in the skeletal muscle and plasma of cirrhotic patients has been reported [89, 90] and these results should be confirmed in future studies. Myostatin is a known inhibitor of protein synthesis and potentially activates the ubiquitin proteasome and autophagy mediated proteolysis[21, 22, 91]. The ubiquitin-mediated proteolysis is not activated but autophagy has been found to be increased in muscle In experimental models of cirrhosis or during hyperammonemia[21, 22]. Other potential mechanisms for activation of autophagy include ammonia mediated mitochondrial dysfunction and generation of reactive oxygen species[92]. Even though these molecular signaling responses have been reported only in neural tissue, similar perturbations may occur in the skeletal muscle also[93].

Figure 1. Myostatin is transcriptionally upregulated by hyperammonemia in the skeletal muscle.

Ammonia enters the skeletal muscle via the transport proteins Rh B and G. In the muscle, ammonia activates transforming growth factor β activated kinase 1 (TAK1) that activates TRAF6. Activated TRAF6 (k63 polyubiquitination) activates IK kinase (IKK) that in turn phosphorylates NFkB inhibitor protein IKB. Phospho IKB is degraded via a proteasome pathway releasing p65NFkB that enters the nucleus and transcriptionally upregulates myostatin.

Interestingly, skeletal muscle metabolic responses to hyperammonemia are being increasingly recognized albeit in preliminary data[93]. Physiologically, glutamine and glutamate serve as anaplerotic substrates to generate α ketoglutarate (αKG) and ammonia in most tissues to maintain sufficient concentrations of the tricarboxylic acid (TCA) cycle intermediates[94]. This reaction is catalyzed by the bidirectional enzyme, glutamate dehydrogenase (GDH). The reaction preferentially occurs in the direction generating αKG, because the GDH Km for ammonia is very high (~1mM), a value that is significantly supraphysiological[95]. However, in cirrhosis, due to impaired ureagenesis and decreased hepatic ammonia disposal, the skeletal muscle functions as a metabolic partner for the liver and skeletal muscle ammonia concentrations are much higher potentially favoring cataplerosis or loss of critical TCA cycle intermediate, αKG[22]. This results in a number of potential consequences including lower flux of the TCA cycle, impaired mitochondrial function and decreased ATP synthesis. Since protein synthesis, especially translation initiation, is an energy intense process, low ATP concentrations may also cause reduced protein synthesis. Another consequence of hyperammonemia that can explain a number of clinical observations is that oxodehydrogenases, pyruvate dehydrogenase, that catalyzes the conversion of pyruvate to acetyl CoA, and αKG dehydrogenase that catalyzes conversion of αKG to succinyl CoA are inhibited by ammonia in a tissue specific manner[96]. An overview of these pathways is shown in Figure 2. A number of clinical studies and metaanalyses have failed to show significant benefit of nutritional supplementation in malnourished cirrhotic patients [18–20, 26]. This may be due to the impaired acetyl CoA generation that necessitates formation of acetyl CoA from non-pyruvate sources including amino acids and fatty acids. Continued mitochondrial dysfunction, generation of ROS, and impaired bioenergetics in the skeletal muscle all contribute to impaired protein synthesis and activate a metabolic, adaptive response, autophagy.

Figure 2. Biochemical abnormalities in the skeletal muscle that contribute impaired protein synthesis and increased autophagy with consequent sarcopenia.

Metabolic and molecular perturbations that can be potentially reversed by intervention at targeted sites. 1. Long term ammonia lowering strategies. 2. Myostatin blocking agent including antagomirs. 3. L-leucine provides acetyl-CoA, activates mTORC1 and protein synthesis. 4. Glucogenic amino acids can be a source of anaplerotic input to provide succinyl CoA replacing the loss of (cataplerosis) of αKG that is converted to glutamate during hyperammonemia (since skeletal muscle cannot generate urea). 5. Cell permeable esters of αKG are a potential strategy to reverse cataplerosis and a novel method to increase muscle ammonia disposal. 6. Physical activity stimulates mTORC1 via phosphatidic acid.

Reduced ATP in the muscle, impaired mitochondrial function, low concentrations of TCA cycle intermediates, increased gluconeogenesis, increased fatty acid oxidation in the skeletal muscle during hyperammonemia suggest a bioenergetics crisis with a starvation like response. Decreased cellular ATP is consistent with activation of the cellular energy sensor, AMPK and impaired mTORC1 signaling[66].

Increased cataplerosis and muscle catabolism of branched chain amino acids as a source of energy may be responsible for low circulating branched chain amino acids with skeletal muscle concentrations of BCAA expected to be decreased in the muscle of cirrhotics due to increased utilization. Reduced cellular amino acid concentrations activate adaptive responses that include increased skeletal muscle autophagy that has been reported in both cirrhosis and hyperammonemia in myotubes. Another response to intracellular amino acid deficiency is the integrated stress response mediated by activation of amino acid deficiency sensor, general control nondepressed 2 (GCN2) via phosphorylation of eukaryotic initiation factor 2 that are increased during hyperammonemia and cirrhosis. Surprisingly, skeletal muscle concentrations of branched chain have been mostly reported to be unaltered except for a single study that reported lower muscle concentrations of BCAA[73–75]. Preliminary studies in hyperammonemic myotubes increased cellular transport and concentrations of leucine despite which supplementation with leucine enriched BCAA resulted in reversal of GCN2 activation and rescue of impaired mTORC1 signaling in patients with cirrhosis[66] and in myotubes during hyperammonemia. Other amino acids with therapeutic potential include L citrulline that is a precursor for L arginine and stimulates mTORC1 and protein synthesis[97]. The beneficial effects of citrulline are believed to be due to decreased ureagenesis resulting in amino acid sparing, it is not known if impaired ureagenesis will aggravate hyperammonemia and its consequences in cirrhosis and need to be studied systematically.

Published data suggest that hyperammonemia is a mediator of the liver muscle axis and the skeletal muscle does not function only as a metabolic sink for ammonia[22]. Ammonia uptake and disposal via glutamine synthesis in the muscle and transport into the circulation may be involved in sarcopenia. At the same time, if there is low muscle mass, non-hepatic disposal of ammonia is impaired which may cause further adverse effects. Consistently, some investigators have reported that encephalopathy is more frequent in sarcopenic than non-sarcopenic cirrhotics[12, 14].

Other potential mediators of the liver – muscle axis in cirrhosis: testosterone, growth hormone

Other mediators of the liver muscle axis include the low testosterone due to increased aromatase activity in liver disease[98]. Decreased growth hormone concentrations or impaired growth hormone response in the muscle are also likely contributors to sarcopenia in cirrhosis[99, 100]. Both growth hormone and testosterone are known to inhibit myostatin expression and signaling responses [101, 102] but it is not known if these hormonal alterations of cirrhosis also contribute to the impaired protein synthesis and increased myostatin expression in cirrhosis. A recent randomized trial showed that testosterone supplementation in male cirrhotics did result in an increase in lean body mass but not survival [103].

Hepatocellular and immune dysfunction as well as portosystemic shunting worsen the endotoxemia due to impaired gut barrier function and potentially altered gut microbiome in cirrhosis[104]. Endotoxemia via TNFα dependent and potentially TNF independent pathways may also impair protein synthesis and potentially activate autophagy[105, 106]. Careful molecular studies on these mediators are not available and the cross talk between hyperammonemia and other putative mediators such as those described above are not presently known. The next decade is likely to see major advances in our understanding of the molecular-metabolic interaction and how it contributes to or causes sarcopenia in liver disease.

Finally, sarcopenic obesity has been reported in patients with non-alcoholic fatty liver disease and after liver transplantation[24, 25, 107]. It is possible that the combination of skeletal muscle loss and increased fat mass may contribute to the development of metabolic components including insulin resistance, diabetes mellitus, hyperlipidemia and possibly NAFLD but whether there is a common underlying mechanism for both sarcopenia and obesity is still not known[108].

Management Strategies

There is compelling evidence that sarcopenia is associated with adverse consequences while there are limited data showing that increasing muscle mass improves survival in the non-transplanted and post liver transplant population of cirrhotics[31, 32]. Therefore, reversing muscle mass is a priority area for therapeutic interventions in cirrhotic patients. Interventions that focus only on deficiency replacement have generally been ineffective while targeted therapies have the potential to reverse muscle loss[1, 18–20, 26, 66, 87]. The major strategies that have been used to improve muscle mass include supplemental calorie and protein intake, increased physical activity, supplemental hormone therapy, and mechanistic targeted treatments[17, 26, 109–111]. The critical outcome measures include survival, hospitalization, quality of life, development of and recovery from other complications of cirrhosis. It is not clear if the improved clinical outcomes are due to an increase in muscle mass, amelioration in skeletal muscle contractile dysfunction or a combination of the two. Despite the current focus being on reversing sarcopenia, it is also important to take into consideration skeletal muscle function that include maximum contractile strength, maintenance of contraction, and muscle fatigue in response to persistent and repetitive contraction[78].

Supplemental nutrition

Since caloric and protein intake are frequently decreased in cirrhosis, Guidelines and Consensus papers have consistently recommended to provide adequate amounts of calories and proteins either by frequent feeding, through oral dietary supplementation or when indicated, by enteral or parenteral nutrition [112–115]. Regimens providing extra calories via high caloric feeding, and/or enteral feeding have been extensively studied (Table 3)[114, 116–123]. Interestingly, few studies suggest improvement in nitrogen retention or nutritional status using very heterogeneous criteria that measure primarily fat and non-fat mass[17, 19, 20, 122, 124, 125]. On the other hand, a recent randomized controlled trial utilizing perioperative immunonutrition enriched in n-3 fatty acids, arginine, and nucleotides vs. an isocaloric diet in patients undergoing liver transplantation could not find any change in total body protein (TBP), measured by neutron activation analysis, from study entry until immediately prior to LT. Even the postoperative outcomes were not influenced by the nutritional supplementation[114].

TABLE 3.

Studies about nutritional intervention in adult liver cirrhosis reporting data about changes in parameters dealing with muscle mass

Author	Treatment	Setting	Duration	Patients (n)	Proteins g/day	Calories Kcal/d ay	Outcome on Nutrition al parameters	Outcome
Hirsch 1993[116]	Oral supplement vs. control	Cirrhotic patients of alcoholic origin, outpatients	1 Year	26 nutritional supplement vs. 25 controls	45±10 + 34g supplement	1580 ± 500 + 1000 kcal supplement	Similar improvement in both groups	Reduced severe infections Reduced hospital admission Similar survival
De Ledinghen 1997[117]	Short term Enteral nutrition vs. fasting	Cirrhotic patients after bleeding from esophageal varices	3 days Follow up 5 weeks	12 enteral vs. 10 controls	74	2090	No change in nutritional parameters	No change in outcome or rebleeding
Le Cornu 2000[118]	Oral supplementation to diet vs. diet	Malnourished cirrhotic patients in the waiting list for liver transplantation	Variable 77 days	42 supplementation vs. 40 controls	80 g	2419 Kcal	Treated improved arm circumference and arm muscle circumference + handgrip strength	No difference in outcomes or survival
Marchesini 2003[119]	Oral supplement of BCAA vs. isocaloric and isonitrogen supplement	Advanced cirrhotic outpatients	12 months	59 BCAA vs. 56 L-alb vs. 56 maltodextrin	0,8 g/kg/day + BCAA 14 g /day Or L-alb 14 g /day Or No protein supplement	30 kcal/die + 200 kcal	Significant Improvement of mid arm muscle area after BCAA supplement	Lower hospital admission in BCAA
Hu 2003[120]	Enteral vs. Parenteral vs. controls	Postoperative patients with poor liver function	7 days	65 enteral vs. 40 parenteral vs. 30 controls	0.16 g N/kg	30 kcal/kg/day	Enteral nutrition caused improved nitrogen balance Minor changes in body weight and Arm circumference	Enteral nutrition caused improvement in gut barrier
Dupont 2012[122]	Enteral 1 month and oral 2 months vs. Conventional treatment	Alcoholic cirrhotic patients with jaundice but no severe acute alcoholic hepatitis	12 month	44 enteral and oral nutritional supplementation vs. 55 controls	Oral diet 60 g Enteral 1.2 g/kg Oral supplement protein 20 g Three times a day	Oral diet 1800 kcal Enteral 3035/Kcal /kg Oral supplement 320 kcal+Three times a day	No change in arm muscle circumference	Similar complications and survival
Sorrentino 2012[123]	Parenteral Nutrition post paracentesis (PNPS) and late evening snack (LES)	Cirrhotic patients with refractory ascites	12 months	40 PNPS and LES vs. 40 only LES Vs 40 controls low sodium diet	1.2–1.3 g/kg BW + PN 1,5g/kg/bw LES 13,5 g /day	30 kcal/kg/day	Maintenance of arm muscle circumference in treated patients vs. deterioration of arm muscle circumference at 6 and 12 months in +LES and controls	Lower number of paracenthesis and better survival in patients treated with PN and LES
Plank 2015[114]	Oral/enter al immune nutrition vs. isocaloric control	Before LT and postoperative	Variable before transplant and 5 days postoperatively Follow up 12 months	52 immunonutrition Vs 49 isocaloric controls	80 g + Supplement of 14 g arginine, 4g omega 3 fatty acids, 1,6 ribonucleic acid	1860–1900 Kcal	Total Body Protein unchanged at 12 months	Similar outcomes In both groups

Another approach has been to shorten the duration of postabsorptive or fasting state in cirrhosis because of the accelerated starvation that results in proteolysis because after food intake, recovery of muscle mass is incomplete[68]. Daytime and nocturnal feeding have been evaluated and there is evidence that a late evening snack has the most beneficial effects and it is currently believed that a late evening and an early morning protein supplement are likely to have the greatest benefit on preventing continued muscle loss in cirrhosis[68, 126]. Meta analyses of supplemental nutrition in patients with alcoholic hepatitis and those with cirrhosis were, however, disappointing as nutritional supplementation by various routes did not improve survival[18–20]. Even though the exact reason for very limited improvement in sarcopenia with nutritional supplementation is not yet clear, cirrhosis can be seen as a state of anabolic resistance and caloric supplementation alone seems to be inadequate. As mentioned earlier, despite providing calories, impaired mitochondrial function and bioenergetics in combination with impaired molecular responses to nutrient administration in muscle are potential reasons for failure. Whether other outcomes including encephalopathy, sepsis and quality of life improve with reversal of sarcopenia are currently unknown.

Protein supplementation is another alternative to improve the availability of essential amino acids. However, cirrhosis and hyperammonemia may accelerate amino acids catabolism with further generation of skeletal muscle ammonia that can impair protein synthesis and increase autophagy further with little or no benefit in reversing sarcopenia. Animal proteins have the added disadvantage of being rich in aromatic amino acids that are not metabolized by the skeletal muscle and may worsen encephalopathy[72, 127]. Vegetable proteins are rich in BCAA and may have a beneficial effect by removing one mole of ammonia per mole of BCAA via the αKG→glutamate→glutamine pathway. Therefore, instead of protein supplementation, BCAA have been used in the past as treatment for hepatic encephalopathy in a number of acute and long-term studies[128–130]. A recent Cochrane review suggested benefit in the primary outcome, hepatic encephalopathy but not on survival, quality of life or nutritional parameters[131]. Lack of benefit in nutritional parameters was counter to expected outcomes, since BCAA provide a source of energy to the muscle in addition to being substrates for protein synthesis. Another mechanism by which BCAA may function is by inhibiting the amino acid deficiency sensor, GCN2 and reversing eIF2α phosphorylation[132], impaired protein synthesis and improve muscle mass. Finally, leucine directly activates mTORC1 that stimulates protein synthesis and decreases autophagy[133], both of which have the potential to improve muscle mass. A recent study in human cirrhosis reported that a leucine enriched BCAA mixture was able to reverse the molecular perturbations in the skeletal muscle downstream of myostatin in cirrhotic patients[66]. Tracer kinetic studies with direct quantification of muscle protein synthesis showed similar rates of protein synthesis in response to a single oral dose of leucine enriched BCAA mixture did reverse the GCN2-eIF2α mediated impaired protein synthesis and increased mTORC1 signaling[66]. These data provide the first direct evidence of molecular perturbations in the skeletal muscle in cirrhosis and in combination with animal and in vitro cell culture data support the role of hyperammonemia as a mediator of the liver-muscle axis.

Exercise and physical activity

The type of exercise determines the muscle related outcomes[134]. Resistance exercise increases skeletal muscle mass by inducing muscle injury and regeneration and protein synthesis[135]. Endurance exercise improves functional capacity but does not necessarily reverse sarcopenia. A combination of resistance and endurance exercise have the potential to improve muscle mass and functional capacity but such studies have not been performed in cirrhosis. Randomized studies have reported improvement in short-term outcomes in response to exercise in cirrhotics[23]. Since direct comparisons of outcomes in healthy and cirrhotic patients in response to exercise have not been reported, it is not possible to determine if the anabolic resistance to nutrients is also observed with exercise. There is evidence that protein kinase Cζ –phosphatidic acid mediates signal transduction of mechanical activity to signaling responses by activating mTORC1 signaling and protein synthesis[136]. However, it is not known if these physiological responses are blunted in cirrhosis and if ammonia is the mediator of such blunted responses. A recent study in a comprehensive array of models including hyperammonemic rats, human subjects and ex vivo muscle preparations does suggest that hyperammonemia also alters contractile response and increases fatigue in cirrhosis[78]. Whether immobilization and injury responses in the cirrhotic skeletal muscle are altered has not been studied but may explain the rapid deconditioning observed during hospitalization.

Anabolic hormones

Testosterone and growth hormone have been used in the past to improve nutritional status and, potentially, muscle mass in cirrhosis but have not been beneficial[99, 100, 137, 138]. Despite adverse effects, these therapies are not effective in reversing nutritional status or sarcopenia. Increased aromatase activity contributes to conversion of testosterone to estradiol that blunts its effect[98]. Aromatase resistant androgens like oxandrolone may therefore be beneficial but have not been borne out in clinical practice [137]. Lack of therapeutic benefit with hormone replacement may also be due to impaired signaling responses including mTORC1 response downstream of androgen and growth hormone receptors may be responsible for failure of these therapies. Increasing the understanding of molecular and metabolic perturbations in the skeletal muscle not only provides explanations for the lack of clinical benefit of standard therapies but also is likely to help identify novel, specific therapeutic targets for reversing sarcopenia.

Ammonia lowering strategies

Current methods to lower ammonia include non-absorbable disaccharides and antibiotics to prevent gut generation of ammonia[139]. The primary outcomes of these treatments are reversal of encephalopathy and lowering of blood ammonia concentrations. It is however well known that blood ammonia concentrations do not always correlate with the severity of encephalopathy, the most studied response[140]. Skeletal muscle turnover is a slow process and lowering ammonia transiently may not necessarily lower muscle ammonia concentrations or reverse the ongoing metabolic and molecular perturbations rapidly. Studies on long-term ammonia lowering strategies, quantifying muscle ammonia concentrations and signaling responses to these interventions are necessary before such an approach can be used to reverse muscle loss and impaired contractile function. Novel and potential methods to lower muscle ammonia include the use of cell permeable esters of αKG that can provide a direct anaplerotic influx with removal of ammonia as glutamine. However, glutamine disposal will then become limiting and strategies for long-term ammonia disposal to protect the skeletal muscle are necessary. Isoleucine and valine as anaplerotic substrates have been suggested because they can remove one mole of ammonia per mole of amino acid but the molecular and functional responses to these interventions have not been evaluated in preclinical or clinical studies to lower muscle ammonia or reverse sarcopenia[82, 83].

Novel molecular targeted strategies

Myostatin antagonists[91], direct mTORC1 activators[66, 133], antioxidants, and mitochondrial protective agents have the potential to benefit skeletal muscle protein turnover but have not been adequately evaluated. Careful mechanistic studies are necessary with preclinical testing before these interventions can be translated to clinical practice.

Post liver transplantation sarcopenia

The underlying molecular mechanisms and mediators need to be ascertained before therapies can be recommended. Direct mTORC1 inhibitors that block protein synthesis responses and accelerate autophagy are largely used after liver transplantation, at least in the United States [17]. Calcineurin inhibits muscle growth and hypertrophy[141] and calcineurin inhibitors are used in the vast majority of post transplant patients. The contribution of these medications to post transplant sarcopenia and sarcopenic obesity needs to be evaluated. Whether anabolic resistance of cirrhosis is reversed by liver transplantation is not known and integrated metabolic-molecular studies with muscle biopsies are needed before specific therapies and preventive measures can be developed. Finally, the reversibility of hyperammonemia induced signaling responses and impaired protein synthesis after liver transplantation is not known and it is possible that epigenetic changes in the regulatory molecules result in long term or persistent sarcopenia even after transplantation or ammonia lowering therapies.

Conclusion

In summary, there is compelling evidence to show that sarcopenia is the major complication of cirrhosis and adversely affects outcomes during the entire course of a cirrhotic patients life. Evidence to show that sarcopenia can be reversed is much more limited and it is not clear if reversing sarcopenia will indeed improve outcomes as expected. Nutritional supplementation is not consistently effective in improving outcomes but long term BCAA with leucine are promising therapies to prevent and treat sarcopenia in cirrhosis. Long-term reduction of muscle ammonia, novel approaches to enhance muscle ammonia disposal, and strategies to block myostatin hold potential for the future. Identification of molecular and metabolic perturbations in the cirrhotic skeletal muscle will allow development of targeted therapies that focus in reversing the anabolic resistance in these patients.

Figure 3. Overview of strategies to reverse sarcopenia and potentially contractile dysfunction in cirrhosis.

Molecular targets are in bold and encircled and putative interventions are italicized.

Key points.

• Sarcopenia is most frequent complication in cirrhosis and the major component of malnutrition that is not reversed after liver transplantation and may actually worsen.

• Therapies including nutrient supplementation and exercise not consistently effective since they target replacing deficiency rather than the underlying mechanisms.

• Hyperammonemia mediated upregulation of myostatin is believed to be a mechanism of impaired protein synthesis and increased autophagy, that contribute to sarcopenia.

• Other perturbations that contribute to sarcopenia include endotoxemia, increased aromatase activity to lower testosterone, and mitochondrial dysfunction

• Therapies targeting mitochondrial function including mitochondrial antioxidants, mTORC1 signaling, and myostatin hold promise for the future.