Liver transplantation in the patient with physical frailty

Summary

Frailty is a decline in functional reserve across multiple physiological systems. A key component of frailty is sarcopenia, which denotes a loss of skeletal muscle mass and impaired contractile function that ultimately result in physical frailty. Physical frailty/sarcopenia are frequent and contribute to adverse clinical outcomes before and after liver transplantation. Frailty indices, including the liver frailty index, focus on contractile dysfunction (physical frailty), while cross-sectional image analysis of muscle area is the most accepted and reproducible measure to define sarcopenia. Thus, physical frailty and sarcopenia are interrelated. The prevalence of physical frailty/sarcopenia is high in liver transplant candidates and these conditions have been shown to adversely impact clinical outcomes including mortality, hospitalisations, infections, and cost of care both before and after transplantation. Data on the prevalence of frailty/sarcopenia and their sex- and age-dependent impact on outcomes are not consistent in patients on the liver transplant waitlist. Physical frailty and sarcopenic obesity are frequent in the obese patient with cirrhosis, and adversely affect outcomes after liver transplantation. Nutritional interventions and physical activity remain the mainstay of management before and after transplantation, despite limited data from large scale trials. In addition to physical frailty, there is recognition that a global evaluation including a multidisciplinary approach to other components of frailty (e.g., cognition, emotional, psychosocial) also need to be addressed in patients on the transplant waitlist. Recent advances in our understanding of the underlying mechanisms of sarcopenia and contractile dysfunction have helped identify novel therapeutic targets.

Defining physical frailty and sarcopenia in patients with cirrhosis

Frailty is a multidimensional syndrome characterised by derangements in multiple physiological systems that was initially described in the context of aging¹ and subsequently in cirrhosis.² As a multidimensional syndrome, general frailty not only includes physical frailty but also cognitive, emotional, and psychosocial domains.^1,2 Published data in the setting of cirrhosis and liver transplant(ation) (LT) have predominantly focused on functional impairment (i.e., physical frailty).³ Physical frailty includes a progressive continuum that starts from mild impairment affecting mostly leisure activities (i.e.: “pre-frail”) to an end-stage condition of advanced frailty.⁴ Other components of frailty, including cognitive frailty and psychological frailty, are more challenging to study in the LT patient population due to complications of liver disease such as sleep disturbance and encephalopathy. In a study of adult patients awaiting LT (n = 1,623), impaired cognition as measured using the number connection test was associated with higher rates of physical frailty but was not associated with waitlist mortality.³ In another study in cirrhosis (n = 355) where frailty was identified by the clinical frailty scale and impaired cognition by the Montreal cognitive assessment, a composite score including both tools predicted hospitalisation and lower quality of life.⁵ Psychological factors associated with frailty include illnesses such as depression, as well as traits such as resilience, and social components such as isolation and loneliness. A recent study in patients with cirrhosis reported an association between low resilience and frailty, but there are few studies in the LT population.⁶

Primary sarcopenia is defined as the progressive loss of muscle strength associated with aging,⁷ secondary sarcopenia occurs with chronic diseases,⁸ and compound sarcopenia refers to sarcopenia in older patients with chronic diseases.⁹ Sarcopenic obesity may also occur in LT patients and is defined by excess body fat mass in combination with low skeletal muscle mass and reduced muscle function; the risk of functional impairment is synergistically higher in patients with sarcopenic obesity than in those with either condition alone.¹⁰ Cirrhosis, particularly once decompensated, predisposes individuals to both physical frailty and sarcopenia, which are interrelated and often observed simultaneously in an individual patient.² Even though most studies to date have focused on a single measure of musclemass, there is increasing recognition that sarcopenia is progressive in non-transplant patients and dynamic measures will help define responses to interventions.² Defining these entities, as well as identifying optimal methods of measurement and underlying mechanisms, will help with the development of specific interventions.¹¹

Prevalence and pathogenesis of frailty and sarcopenia

In clinically stable patients with cirrhosis, the prevalence of frailty is estimated to be between 18% and 43%, depending on the severity of liver disease, comorbidities and the measurement tools that are used.² In individuals with acutely decompensated cirrhosis,¹² the prevalence is higher.¹³ To date, there are no specific reports on the prevalence and characteristics of objectively evaluated physical frailty in acute-on-chronic liver failure (ACLF). The prevalence of frailty for patients awaiting LT is directly impacted by centre selection criteria.

The major drivers of physical frailty are skeletal muscle loss (sarcopenia), malnutrition, and progressive immobility.^2,11 In decompensated cirrhosis, these contributors act synergistically in a vicious cycle that is further accelerated/exacerbated by liver-specific factors. Systemic inflammation and hepatic metabolic dysfunction contribute to a persistent hypercatabolic state, and severe anorexia and early satiety result from ascites and other complications of cirrhosis.^14,15 Additional non-liver-specific factors which contribute to frailty in these patients include age-related loss of muscle and contractile strength (~1% per year up the age of 70, ~1.5–2.5% thereafter),¹⁶ as well as comorbidities including diabetes mellitus, cardiovascular, pulmonary, and chronic kidney diseases. In the pre-transplant patient, additional factors such as alcohol intake¹⁷ and non-alcohol associated fatty liver disease are associated with both liver and muscle toxicity.¹¹ In assessing the severity of physical frailty/sarcopenia and the potential clinical implications, these aggravating factors are relevant because of their potential to deteriorate after LT, increasing morbidity and mortality in recipients.¹⁴

Despite the high clinical significance of physical frailty and sarcopenia, most published studies are descriptive, and provide limited insight into underlying pathological mechanisms. Since nearly 40% of skeletal muscle is protein (structural and contractile), the balance between protein synthesis and proteolysis is a major regulator of both muscle mass and contractile function.¹⁸ Both skeletal muscle contractile function and the initiation of mRNA translation during protein synthesis are cellular functions which require high amounts of energy.¹⁹ In addition to the relatively energy-deficient state in cirrhosis, a state of accelerated starvation, a number of molecular and metabolic factors have recently been identified to contribute to sarcopenia and physical frailty in these patients, as outlined in Fig. 1.^20,21 Focused studies have identified upstream mediators and downstream molecular and metabolic targets that contribute to sarcopenia and frailty in cirrhosis.^18,21–25 More recently, multiomics data from skeletal muscle in preclinical models and humans have provided insights into the novel mechanisms of sarcopenia and physical frailty.^26–29

Fig. 1.

Mediators of physical frailty/sarcopenia in cirrhosis.

Initial human studies in cirrhosis using whole-body tracer kinetics showed lower or unaltered protein synthesis with increased proteolysis.^20,30 However, there was little to no reversal of muscle loss in response to dietary supplementation, including of calories and amino acids.^30,31 Subsequent molecular and metabolic studies in muscle tissue helped identify changes in signalling molecules, including increased expression of myostatin (a TGFβ [transforming growth factor-β] superfamily member), that cause muscle loss.^{18,20,24,32,33} Myostatin binds to a TGFβ type II receptor, activin IIB, that subsequently recruits downstream type 1 receptors that function as kinases.³⁴ Myostatin binding to its receptor impairs mTORC1 (mammalian target of rapamycin complex-1) signalling, which is a critical regulator of protein synthesis.^18,32 Additionally, phosphorylation of eIF2α (eukaryotic initiation factor 2 α), another component of the integrated stress response that also inhibits global protein synthesis, is higher in skeletal muscle in patients with cirrhosis.^18,23 Molecular and functional studies have shown a limited role for the ubiquitin proteasome system, but a major role for increased autophagy in muscle proteolysis in cirrhosis.²⁵ These data have also revealed potential molecular targets for the treatment of sarcopenia and physical frailty.

There are also metabolic alterations in liver disease which impact proteolysis. Ammonia, a cytotoxic molecule generated during a number of cellular processes, including transamination, purine breakdown, and gut microbial metabolism, has been shown to be a mediator of the “liver-muscle” axis.^20,30 Decreased hepatocyte ureagenesis, the major mechanism of ammonia disposal in healthy individuals, results in increased skeletal muscle uptake of ammonia to prevent the systemic effects of hyperammonaemia, which causes multiple molecular, metabolic, and organelle perturbations.²⁴ Hyperammonaemia causes upregulation of myostatin²⁴ and impairs β-catenin and consequent ribosomal biogenesis, which subsequently contributes to impaired mRNA translation and protein synthesis.³⁵ Skeletal muscle ammonia disposal contributes to mitochondrial dysfunction which reduces ATP production.^22,26 The consequent low ATP levels and increased free radical generation with oxidative stress in the skeletal muscle are accompanied by accelerated post-mitotic senescence in the skeletal muscle.²⁶

Other potential mediators of sarcopenia in cirrhosis include low testosterone levels, as well as endotoxemia due to a combination of gut dysbiosis, decreased gut barrier function, and impaired immune function.²⁰ Lower levels of plasma branched-chain amino acids (BCAAs) were initially thought to be a mediator of sarcopenia, but several prior human studies of BCAA supplementation demonstrated no significant benefit in reversing sarcopenia.^18,23 Increased muscle utilisation of BCAAs in a hyperammonaemic stress response is a potential explanation for the lack of benefit observed with amino acid supplementation.²³ Additionally, muscle catabolism of BCAAs generates ammonia that can also limit beneficial responses to L-leucine, a potent anabolic amino acid, though a leucine metabolite, β-hydroxymethyl butyrate, has been reported to be effective in preclinical models of liver disease,²⁷ while a recent randomised trial (n = 32 patients) of BCAA supplementation in patients with cirrhosis and sarcopenia diagnosed on CT imaging demonstrated improved muscle mass and improved liver frailty index (LFI) for the intervention group.³⁶ Hyperammonaemia also causes phosphorylation of GCN2 (general control non-derepressible-2), which results in impaired protein synthesis.²³

Impaired contractile function has also been studied using multiomic analyses including integration of data across species (horizontal integration) which has identified a number of novel mechanisms including perturbations in protein kinase A signalling and Hippo pathways in the skeletal muscle of patients with cirrhosis and hyperammonaemia.^26–29 Both low ATP and modifications of skeletal muscle contractile proteins cause skeletal muscle fatigue due to lower maximum and repetitive contractile strength.²¹

Diagnosis and measurement of physical frailty and sarcopenia

Multiple methods have been proposed for the assessment of physical frailty and sarcopenia in individuals at risk (Table 1).^20,37,38 The in-person assessment of frailty in cirrhosis includes the use of provider-evaluated scales and objective measures. Provider-evaluated scales such as the clinical frailty scale (CFS) and Karnofsky performance scale (KPS) have been used in the geriatric population to define frailty. Though not as well evaluated in patients with liver disease, these tools have shown promise as predictors of clinical outcome in cirrhosis^39,40 (Table 1). An advantage is that these tools include determination of physical limitations due to both chronic medical conditions and cognitive impairment.^41,42 Notably, the KPS is currently a required data element for waitlisted patients in the US, and thus is readily available. An intrinsic limitation of these measures, however, is the inclusion of subjective items, which may lead to underestimation of frailty.^43,44

Table 1.

Methods to assess sarcopenia and frailty in the liver transplant candidates.

Methods	Components	Clinical accessibility	Includes subjective variables	Cut-offs associated with clinical outcomes
Sarcopenia
CT scan	Skeletal muscle index (SMI): total muscle area at the 3rd lumbar vertebrae/height2	Easy though the analysis requires a dedicated software	No	SMI <50 cm2/m2 in male and <39 cm2/m2 in female
Frailty
Activities of daily living	Questionnaire-based; assesses the individual’s difficulty in performing daily self-care activities	Easy	Yes*	12
Karnofsky performance status	Incorporates activities of daily living + limitations one may experience with chronic diseases as well as cognitive decline.	Easy	Yes*	B
Clinical frailty scale	Same as Karnofsky performance status	Easy	Yes*	>4
Fried frailty phenotype°	Weight loss, self-reported exhaustion, weakness, slow walking speed, decreased physical activity levels	Moderate	Yes	≥3
Gait speed test	Measures gait speed (4 or 10 m walk)	Easy	No	≤0.8 m/s
6-minute walk test	Measure the distance an individual is capable of walking on a flat in 6 min.	Easy	No	<250 m
Cardiopulmonary exercise testing	Objectively assesses cardiovascular, pulmonary, and skeletal muscle systems capacity during exercise (maximal O2 uptake).	Poor/very poor	No	<60%
Short physical performance battery	Assesses lower limb function via: standing balance, gait speed, continuous chair stands	Easy	No	<10
Liver frailty indexˆ	Grip strength, chair stands, balance testing	Moderate	No	≥4.5

^*Both from self-reported information and judgment from the personnel administering the test.

^°Extensively used in geriatric populations but likely underestimates functionality in cirrhosis due to symptoms commonly seen in patients with cirrhosis, which would lead to inaccurate assessment of frailty.

^ˆThe one most rigorously validated in patients with cirrhosis though no data on improvement following an intervention is yet available.

The Fried frailty phenotype (FFP) is the most commonly used objective measurement battery for the assessment of frailty in geriatric patients,⁴⁵ and classifies an individual as frail (≥3), pre-frail (1–2), and robust (none) using patient-reported weight loss, exhaustion, amount of weekly activity plus measured walk speed and grip strength. Although the FFP has been associated with clinically relevant outcomes in multiple cohorts of patients without liver disease,^46,47 its usefulness in patients with decompensated cirrhosis being considered for LT seems more limited^48,49 since the characteristics evaluated by FFP are affected by complications of cirrhosis, such as encephalopathy, which may result in underestimation of functional capacity and overestimation of frailty.

The 6-minute walk test, gait speed test, and grip strength are practical, objective screening tests for physical frailty and can be used to estimate the risk of hospitalisation in patients.^50–52 The short physical performance battery (SPPB) includes repeated chair stands, balance testing, and 4-meter walk.^53,54 As an adaptation of the SPPB, the LFI includes chair stands, balance testing and grip strength and has been proposed as an objective tool to evaluate frailty and the effectiveness of tailored interventions in cirrhosis and LT.⁵⁵ There is still uncertainty as to which of these measures should be incorporated into the clinical care of LT patients.⁵⁶ The ideal test must be accurate, reproducible, simple, and predictive of clinical outcome independently of MELD (model for end-stage liver disease) score or Child-Pugh stage.^57,58 A recent prospective study indicates that the LFI, SPPB, FFP, and CFS are comparable for the assessment of frailty in patients with cirrhosis, which allows for some flexibility in their clinical application.⁵⁹ Among these tools, the LFI is the most carefully validated in multicentre studies, at least in North American populations.⁶⁰

The need for patient participation in frailty assessment may limit the use of LFI and other objective tools in very ill, hospitalised patients, such as those with acute decompensation or ACLF.⁶¹ In ACLF, a retrospective study from the Veterans Administration did not find any association between frailty, as defined by an administrative algorithm, the hospital frailty risk score, and short-term survival.⁶² It could be that patients with advanced ACLF are so sick that the presence (or absence) of frailty has limited impact on their survival. However, it should be highlighted that the hospital frailty risk score has not been validated in cirrhosis; thus, a better definition and more granular understanding of frailty in ACLF is still awaited. It is also emerging that ACLF is a distinct syndrome wherein traditional scores are not effective to predict the risk of mortality or futile transplantation.⁶³ Whether a better assessment of physical frailty in ACLF would provide any additional information for clinical decision making, particularly regarding suitability for transplantation, requires further investigation.

For sarcopenia, anthropometric measures are the lowest cost approach to determine muscle mass, but inter- and intraobserver variability limit their accuracy and application in clinical practice. Fat-free or lean body mass determined by dual energy X-ray absorptiometry and bioelectric impedance analyses have been used as measures of whole-body muscle mass but logistical issues including reproducibility, as well as the influence of volume status, limit their routine application.² Additional techniques for the assessment of sarcopenia, including the pros and cons of each technique, have recently been reviewed elsewhere.³⁷ Cross-sectional imaging with CT scans (MRI can also be used, though CT is more commonly used) is reproducible and allows for the measurement of muscle area directly. Although other areas of measurement have been used, measures of muscle area at the 3^rd lumbar vertebra on CT (or MR slice) normalised for height, i.e. the skeletal muscle index (SMI), is used most often across centres. Based on North American data, sarcopenia is typically defined as a SMI <50 cm²/m² in men and <39 cm²/m² in women.⁶⁴ Repeated use and rapid interpretation in clinical practice is limited by cost and availability.⁶⁵

Based on the currently available evidence in the LT patient population, we recommend using an objective screening test for physical frailty, such as the LFI, combined with cross-sectional imaging to identify sarcopenia in the outpatient setting. We do not recommend using the currently available objective screening tests for physical frailty in severely ill hospitalised patients because they require patient participation, which may not be possible depending on severity of illness, and because of limited validation in this setting. The CFS, as assessed by the clinical provider (<1 min to perform), was the most commonly used frailty assessment tool across a systematic review of patients (n = 253,376) admitted to an intensive care unit (ICU),⁶⁶ but this tool has not been validated in large studies in cirrhosis. Screening for sarcopenia with cross-sectional imaging is possible in hospitalised patients, but how the information should be incorporated into the care of severely ill patients being considered for LT has not been defined.

Clinical significance of frailty/sarcopenia on waitlist and post-LT outcomes

In both outpatient and hospitalised patients with cirrhosis, physical frailty is associated with a higher risk of mortality.^{11,13,48,49,51,55,58,67–73} In a large multicentre study, a LFI ≥4.5 was associated with an 82% increase in the risk of waitlist mortality in LT outpatient candidates independent of the MELD-Na score, ascites and hepatic encephalopathy.¹⁴ During time on the LT waitlist, ~50% of LT candidates may experience a worsening of LFI,⁷⁴ and an increase of LFI ≥0.1 after 3 months has been associated with a twofold increase in the risk of mortality on the LT waitlist.

Besides waitlist mortality, physical frailty – as defined by LFI – has been associated with the risk of decompensation,⁷³ mortality after LT,⁷⁵ length of hospital stay after LT⁷⁵ and impaired health-related quality of life 1 year after LT.⁷⁶ Importantly, pre-transplant physical frailty was shown to be significantly associated with post-transplant mortality (adjusted hazard ratio [HR] 2.1; 95% CI 1.4–3.3).⁷⁵ Physical frailty also results in increased healthcare resource utilisation and is an independent predictor of post-transplant intensive care unit stays ≥4 days, post-transplant hospital stays ≥12 days and non-home discharge.⁷⁵ Finally, physical frailty before LT has been associated with worse health-related quality of life 1 year after LT (mainly related to the physical component of healthcare questionnaires).⁷⁶

Independent lines of investigation assessing physical frailty by other tools such as FFP,^48,49,70 CFS,⁴⁹ SPPB,^48,49,68,77 6-minute walk test,^78,79 and gait speed,⁵¹ further confirm the association between functional impairment and mortality risk in cirrhosis. Activities of daily living are independently associated with in-hospital mortality, non-home discharge, and impaired quality of life.^71,72 In LT candidates, deficits in activities of daily living were independent predictors of waitlist mortality.⁸⁰

KPS predicted in-hospital mortality and post-discharge mortality in hospitalised patients with cirrhosis as well as waitlist mortality in LT candidates.^81,82 After LT, a lower pre-LT KPS (i.e. ≤40%) was associated with a 38% higher risk of graft failure and 43% higher risk of mortality.⁸³ Higher KPS was also associated with shorter intubation time after LT.⁸⁴ Moreover, a KPS ≤40% was associated with longer hospital stay after LT and higher costs in the first year after LT.⁸⁵

Sarcopenia has been associated with morbidity and mortality in patients with cirrhosis. Most studies in patients on the LT waitlist have evaluated sarcopenia using skeletal muscle mass measured by CT imaging. Patients with cirrhosis and sarcopenia showed lower probability of survival than those without sarcopenia.^65,86–91 Moreover, sarcopenia was found to be an independent risk factor for hepatic encephalopathy,⁹² bacterial infections^93–95 and ACLF.⁸⁷ In patients with sarcopenia before LT, survival rates were lower,^96–98 ICU and overall hospital stays were longer,^96,98,99 and the risk of infections was higher after LT.^98,100 A meta-analysis of 19 studies including 3,803 LT candidates found that sarcopenia (assessed by CT scan) was associated with an 84% increase in the risk of mortality on the waitlist and a 72% increase in the risk of post-LT mortality.¹⁰¹

Several risk factors including sarcopenia, frailty, older age, severe ACLF and/or grade 2–3 obesity can occur concurrently. The use of these diagnoses alone for waitlist prioritisation or listing is challenging for three reasons: a) the mortality risk associated with frailty and sarcopenia is similar in the pre- and post-LT setting; b) sarcopenia and frailty are not completely or consistently reversed post-LT, c) there is inconsistent use of standardised interventions to treat frailty and sarcopenia in the pre- and post-transplant period. Therefore, high prioritisation of patients who are frail/sarcopenic may result in reduced post-transplant survival, especially without a standardised intervention. The impact of these strategies on post-LT outcomes and resource utilisation needs to be evaluated in future prospective studies.

The post-liver transplant course of physical frailty and sarcopenia

There is limited but evolving data describing the post-LT course of physical frailty and sarcopenia and its determinants. Determinants of frailty and sarcopenia such as increasing age and non-cirrhosis-related comorbidities are not reversible with LT. Post-LT immunosuppression can also contribute to impaired muscle health. Additional determinants including post-LT complications, sedentary behaviour, nutritional intake and others deserve more detailed evaluation in future studies. In a series of patients who underwent LT (n = 214), frailty scores worsened at 3 months post-LT, and modestly improved by 12 months post-transplant, but only 40% of patients were physically robust by 1 year after transplant.¹⁰⁷ The strongest predictor of being robust post-LT was high physical function pre-LT, supporting the role of pre-transplant prehabilitation.¹⁰⁷ As for the post-LT course of sarcopenia, Plank et al. found that during the first 10 days after LT there is a 10% loss in total body protein, mostly from skeletal muscle and only 54% of this loss was restored by 12 months after transplant.¹⁰⁸ Bhanji et al. reported that, following transplant, sarcopenia was only resolved in 6% of patients with pre-transplant sarcopenia.¹⁰⁹ In transplant recipients with a CT scan within the first year of LT, there was a further decrease in muscle mass and an increase in myosteatosis. Other studies confirmed a low rate of recovery from sarcopenia after LT and a further decrease in muscle mass.^110–112 Risk factors for sarcopenia after LT were pre-LT sarcopenia, post-LT length of stay, biliary complications and having been transplanted for decompensated cirrhosis instead of hepatocellular carcinoma.^110,112 The association between post-LT sarcopenia and mortality is controversial, with some studies showing a significant association¹¹⁰ and others not.^109,111

Frailty/sarcopenia in special patient populations

Frailty/sarcopenia in female LT candidates

Multiple studies have analysed the impact of physical frailty/sarcopenia on waitlist and post-LT survival in males vs. females to determine sex-dependent effects. A Canadian retrospective study of patients listed for LT (n = 142) and screened for sarcopenia using CT imaging demonstrated a higher prevalence in males (54% vs. 21%, p <0.001), with sarcopenia being an independent predictor of mortality (HR 2.36; 95% CI 1.23–4.53).⁹⁰ A subsequent single-centre analysis examining the rates of physical frailty and sarcopenia in patients waitlisted for LT with a diagnosis of non-alcoholic steatohepatitis (NASH) vs. alcohol-related liver disease (ALD) also found no differences in the incidence of frailty between men and women, though sarcopenia was more common in males.¹¹³ However, in a large European analysis (n = 585 patients awaiting LT) the rate of sarcopenia was similar in males and females, with sarcopenia being associated with inferior waitlist survival in a sex-independent manner.⁹¹ Sarcopenia was recently assessed as a prognostic indicator specifically for hospitalised patients with decompensated cirrhosis; it was more commonly identified in males and was predictive of post-transplant mortality for males with sarcopenia, who experienced higher rates of death at 1 year (86% vs. 95%) and 3 years (73% vs. 95%) (log-rank p = 0.01) (115). Interestingly, sarcopenia was not associated with post-LT mortality in women (HR 1.02; 95% CI 0.96–1.09).¹¹⁴

Physical frailty (though not sarcopenia) was assessed in a recent prospective study by Lai et al. on patients undergoing outpatient LT evaluation, which compared the LFI in waitlisted women vs. men (nine separate US transplant centres, n = 1,405 waitlisted patients),.¹¹⁵ In this analysis, the LFI was higher for women than men (mean, 4.12 [0.85] vs. 4.00 [0.82]; p = 0.005), though the actual incidence of frailty in female vs. male candidates was not provided in this analysis. In the most recent publication by this multicentre US cohort which focused on LT recipients (eight centres, n = 1,166 LT recipients), a larger proportion of frail patients were female (40% vs. 33% in the non-frail group), with frailty defined as an LFI ≥4.5.⁷⁵ On multivariate analysis, pre-transplant frailty remained significantly associated with post-transplant mortality (HR 2.13; 95% CI 1.39–3.26; p <0.001), though importantly, the overall survival for this patient cohort was still excellent for both groups, with 5-year survival rates of 90% vs. 84%, respectively. A small study in cirrhosis (n = 291) revealed that sarcopenia was associated with increased odds of frailty in males but not females.¹¹⁶ However, a single-centre report on dynamic changes in muscle mass in patients with cirrhosis did not find any sex-dependent differences.¹¹⁷ The potential for differential drivers of frailty based on sex requires further evaluation.

Frailty/sarcopenia in elderly LT candidates

Although first defined in geriatric populations and initially viewed as an age-specific condition, frailty is now recognised to impact younger patients with chronic medical conditions. The determination of whether there is a differential impact of frailty for LT patients based on age is an area of ongoing investigation. A recent study involving 882 patients listed at two centres specifically analysed the prevalence of physical frailty in LT candidates aged >65 vs. 18–65, using the LFI; they found that older candidates were more likely to be frail, 33.3% vs. 21.7%, p = 0.002.¹¹⁸ Notably, the adverse impact of frailty on waitlist outcomes was similar in both younger patients and older patients and frailty was associated with a nearly twofold increased risk of waitlist mortality. Other analyses including the large US multicentre cohort which published sequential reports on patients waitlisted at nine US centres, and of patients from eight centres who subsequently underwent LT, found that the age of frail and non-frail patients was similar, which may be at least partly related to the selection criteria used.^58,75

Like physical frailty, assessments of the relationship between sarcopenia and age in LT patients (compound sarcopenia¹¹⁹) are subject to selection criteria, though in a recent single-centre analysis of the evolution of sarcopenia pre- and post-LT (n = 293 LT candidates), there was no association between age and the presence of sarcopenia.¹⁰⁹ Similarly, in earlier single-centre reports of sarcopenia in waitlisted and transplanted patients,⁹⁰ as well as a recent large multicentre European study on sarcopenia in waitlisted patients, there was no association between age and sarcopenia.⁹¹

Frailty/sarcopenia in obese LT candidates

The worldwide epidemic of obesity and the subsequent increased prevalence of non-alcoholic fatty liver disease has led to an increased frequency of obesity in candidates awaiting LT.

Waitlist mortality is higher and transplant rates are lower for obese LT candidates, though they still experience a significant benefit from LT and achieve similar post-LT survival as non-obese LT recipients.^120,121 However, as in the general population, frailty and sarcopenia also affect obese LT candidates and may be more easily overlooked. A large study of the general (non-LT) population using the NHANES and SHARE data sets (US and Europe) has determined that frailty is more common in obese patients than normal BMI patients, though interestingly, frail patients with higher BMI had a lower risk of mortality vs. frail patients with a normal BMI.¹²² The multicentre US frailty cohort assessed the incidence and impact of physical frailty on waitlist outcomes using the LFI, and found that the prevalence of physical frailty was similar in non-obese, class I obese, and class II obese candidates at 25.4%, 26.0%, and 29.2%, respectively (p = 0.57).¹¹⁸ In this analysis, physical frailty was associated with a twofold higher risk of waitlist mortality for both non-obese and class I obese candidates, while mortality risk increased further for frail LT candidates with class 2 or greater obesity.

Patients with obesity are not exempt from sarcopenia. Indeed, sarcopenic obesity has been observed in 27% of patients with cirrhosis and obesity¹²³ and it has been associated with poor survival both before and after LT.^114,123 In addition to muscle mass, muscle quality also matters. Myosteatosis has been associated with poor survival before and after LT,^123,124 although clear thresholds to define this condition are still to be determined.¹²⁵ An interesting single-centre analysis of NASH vs. ALD found that NASH was associated with a lower prevalence of sarcopenia (22% vs. 47%; p <0.001) but a higher prevalence of frailty (49% vs. 34%; p = 0.03) than ALD at the time of listing. In this analysis, sarcopenia was not associated with adverse events in patients with NASH, but a higher frailty score was associated with an increased length of hospitalisation (p = 0.05) and an increased risk of delisting (p = 0.02) in the NASH cohort.¹¹³ In an analysis of hospitalised patients who underwent urgent evaluation and LT, the assessment of a combination of sarcopenia and obesity was associated with increased mortality after LT (univariate HR 2.92, 95% CI 1.04–8.23; multivariate HR 3.50, 95% CI 1.10–11.15).¹¹⁴

Optimal management of frailty/sarcopenia in patients before and after transplant

Across the spectrum of care from the waitlist period into the immediate post-transplant period, as summarised in Fig. 2, the current mainstay of treatment for physical frailty and sarcopenia involves the optimisation of muscle health with nutrition and physical activity, and the reduction of potentially myotoxic insults. In the ideal setting, this involves access to practical educational material (see www.cirrhosiscare.ca), and multidisciplinary support by team members including dietitians and exercise professionals.¹¹ A number of pharmacological therapies are under investigation and are discussed in the future directions section, as well as in another recent review on the topic.¹²⁶

Fig. 2.

Interventions for physical frailty/sarcopenia across the liver transplant continuum.

Nutritional considerations

Detailed, evidence-based recommendations on the nutritional management of patients with liver disease have recently been published by both EASL and ESPEN.^127–129 We provide recommendations that focus on LT candidates with frailty/sarcopenia.

1. Waitlist period –The nutritional prescription during this period involves three main areas of focus.

   1. Energy requirements - Where possible, energy needs should be evaluated with indirect calorimetry given the inaccuracy of available predictive equations (e.g., Harris-Benedict) and weight-based equations (e.g. 35 kcal/kg/day).^11,130 If indirect calorimetry is not available, either predictive equations or BMI-adjusted energy intake goals can be used: 35 kcal/kg/day with a BMI of <30 kg/m², 25–35 kcal/kg/day with a BMI of 30–40 kg/m² and 20–25 kcal/kg/day with a BMI of ≥40 kg/m².^11,131,132 Alternatively, for individuals with obesity (BMI ≥30), a tailored, moderately hypocaloric diet (−500–800 kcal/day) can be prescribed alongside adequate protein intake.¹³³ Energy targets should be adjusted based on the goal of treatment (i.e., improvement or maintenance of nutritional status). Barriers to food intake should be evaluated and targeted, including loosening restrictions on sodium intake,¹³⁴ getting assistance for food insecurity and attempting to ameliorate ascites or hepatic encephalopathy.¹³² Enteral supplementation via a nasoenteric tube should be considered for high-risk outpatients who do not meet energy requirements with oral intake alone.
   2. Protein requirements – Patients with cirrhosis require a protein intake of at least 1.2 g/kg/day to achieve a positive protein balance.^135,136 Guidelines support a protein intake target of 1.5 g/kg/day in malnourished patients (calculated using the ideal body weight for obese patients), with increases to 2 g/kg/day in hospitalised and critically ill patients.^{11,128,132,133} A diverse range of protein sources including both meat and non-meat sources can be considered to avoid food boredom and to take advantage of the potentially beneficial impact of vegetable and casein-based protein diets on hepatic encephalopathy.^{11,128,132,133,137,138} Guideline recommendations on BCAAs vary,^{11,128,132,133} but if it is not possible for a patient to meet protein intake targets via the diet, BCAAs can be prescribed at a dose of 0.25 g/kg/day.¹²⁸ A recent systematic review on the use of BCAA supplementation suggests potential benefit in improving muscle mass but not contractile function. However, given the potential for increased ammoniagenesis, firm recommendations require critical, mechanistic studies on BCAAs because the composition, dose and duration of use, and outcome measures continue to remain variable across studies.¹³⁹
   3. Reduction of the fasting interval – Given the hepatic glycogen depletion and rapid transition to protein catabolism for gluconeogenesis (i.e., accelerated starvation) that accompanies cirrhosis, patients should be advised to eat every 3–4 h while awake, and to consume a late evening snack ± early breakfast.^{11,128,132,133} A landmark trial by Plank et al. reported significant improvements in total body protein and fat-free mass across all Child-Pugh classes when an isocaloric supplemental nutrition intervention was administered during the evening time vs. the daytime period.¹⁴⁰

2. Immediate pre-operative and post-operative periods – In the immediate pre-operative period, based on enhanced recovery after surgery (ERAS) recommendations for LT, pre-operative fasting does not need to exceed 6 h for solids and 2 h for liquids. Carbohydrate loading may be considered at least 2 h before the induction of anaesthesia. For both of these recommendations, caution should be exercised if there are risk factors for delayed gastric emptying (e.g., tense ascites, diabetes, autonomic dysfunction).¹⁴¹ ERAS recommendations acknowledge the profound muscle mass loss that can occur after transplant, the equivalent of 10% of total body protein stores in the first 2 weeks after transplant.¹⁰⁸ In the acute post-operative phase, the goal is to avoid protein breakdown, with guidelines recommending energy intake targets of 30–35 kcal/kg/day and protein intake targets of 1.2–1.5 g/kg/day.¹²⁸ As per ERAS recommendations, according to the patient’s tolerance, enteral nutrition can be started within 12–24 h after transplantation and titrated as tolerated with parenteral nutrition considered if enteral nutrition is not possible.^128,141 The timing, route and progression of post-operative feeding is individualised based on nutrition risk (i.e. degree of malnutrition), the surgical intervention (e.g., Roux vs. duct-to-duct anastomosis, where feeding may be delayed with a Roux limb), intensity of ICU support needs (e.g., need for high-dose vasopressors, intubation which may impact feeding) and the practice patterns of the involved transplant and intensivist teams.

3. Longer term post-operative period – Alongside incomplete and delayed recovery of muscle mass and function, weight gain is a common challenge after LT, with a median weight gain of 5.1 kg and 9.5 kg at 1 and 3 years, respectively, in one post-transplant series.¹⁴² Although there is limited evidence to guide standardised post-operative recommendations on nutrition for the patient in the long-term post-transplant period,¹⁴³ nutritional interventions should focus on the optimisation of body composition by reducing gain in fat mass while increasing muscle mass and function.¹⁴⁴

Physical activity considerations

While data quantifying the long-term benefits of physical activity across the transplant continuum are promising and still evolving,^145–147 pragmatically, patients should be supported to engage in a combination of aerobic and resistance exercises to increase both cardiopulmonary fitness as well as skeletal muscle strength and mass. In the pre-transplant period, given the acute increases in portal pressure that can occur with exercise,^148,149 primary or secondary variceal prophylaxis should be initiated if indicated.¹⁵⁰ Interestingly, in patients who are overweight/obese and have portal hypertension, the combination of exercise and variceal prophylaxis is safe and reduces portal pressure.¹⁵¹ Recommendations for prehabilitation support a “start low and go slow” approach which targets 150 min of moderate intensity exercise per week and at least 2 days of resistance exercise per week.^{11,132,133,149} Similarly, although specific dosing recommendations are not available, inpatient rehabilitation appears to be safe, tolerable, feasible and improve functional outcomes after LT.^152,153 The limited post-transplant exercise studies that are available have been associated with a promising impact on exercise capacity and physical function.^154–156 Across the transplant continuum, given the vulnerable state of patients with frailty and/or sarcopenia, the involvement of a certified exercise professional is an important consideration to allow for the development of a tailored physical activity programme with adjustments as the patient’s condition evolves.

Post-transplant immunosuppression

Calcineurin inhibitors upregulate slow-fibre-specific gene promoters and increase myostatin expression, thereby inhibiting muscle growth and regeneration.^102–104 mTOR inhibitors block muscle hypertrophy.¹⁰⁵ Steroids result in type II muscle fibre atrophy and myopathy.¹⁰⁶ While immunosuppressants are essential to maintaining the health of the graft, it is anticipated that future work will provide guidance on how to tailor these regimens based on individual risk profiles.

An expanded approach beyond the constructs of physical frailty and sarcopenia

While this review is focused on physical frailty and sarcopenia, it is essential to acknowledge that the biological and chronological age of patients awaiting LT is increasing, as is the prevalence of associated comorbidities.¹⁵⁷ Additionally, our patients are physiologically older than their biological age; a study of 1,107 patients awaiting LT (mean age 55 years) revealed a peak baseline VO₂ of 17.4 ml/kg/min, consistent with what would be expected for a sedentary adult in the 8^th decade of life.¹⁵⁸ Thus, a whole-person, multidisciplinary, approach that extends beyond physical aspects of frailty and sarcopenia may be best for these often “geriatric” patients. Guidelines by the American College of Surgeons (ACS) National Surgical Quality Improvement Project and the American Geriatrics Society on the optimal perioperative assessment¹⁵⁹ and management of the geriatric patient¹⁶⁰ are useful frame-works to guide this whole-person approach.¹⁵³

In brief, the ACS assessment involves the identification of 12 factors that can be screened for and then individualised for optimisation: (i) cognitive ability and capacity, (ii) depression, (iii) alcohol and other substance abuse/dependence, (iv) pre-operative cardiac evaluation, (v) risk factors for post-operative pulmonary complications, (vii) functional status and fall history, (viii) frailty score, (ix) nutritional status, (x) polypharmacy, (xi) treatment goals and expectations and (xii) family and social support system.^159,160

The tenets of perioperative management in the immediate peri-operative period are similarly detailed by the ACS and involve: (i) the documentation of the patient goals and treatment preferences and the surrogate decision maker, (ii) short-ened fasting pre-operatively, (iii) antibiotic and venous thromboembolism best practices and (iv) medication optimisation. Post-operatively, relevant considerations include focusing on strategies to: (i) reduce delirium/cognitive impairment, (ii) reduce perioperative acute pain, pulmonary complications, urinary tract infections, pressure ulcers and falls, (iii) maintain adequate nutrition and (iv) reduce functional decline.¹⁶⁰

Future directions

Advances in integrated molecular-metabolic approaches complemented by a number of multiomics analyses could facilitate the identification of biomarkers and mechanism-based therapeutic targets.^27–29 Plasma proteomics and metabolomics analyses which integrate bioinformatics and machine learning/artificial intelligence-based approaches can help identify biomarkers of progression and therapeutic responses. Mechanism-based therapies need to replace the current “deficiency-replacement approaches” that focus on supplementing deficient factors and removal of factors that are higher in disease. Ammonia is one of the best studied mediators of sarcopenia and long-term ammonia lowering is beneficial in preclinical models.^26,161 Since ammonia-lowering measures are used clinically for patients with symptomatic hepatic encephalopathy, extending these therapies to patients with sarcopenia and physical frailty is a rapidly translatable option. Targeting GCN2 also has the potential to reverse the hyperammonaemic stress response and prevent progression of sarcopenia. The leucine metabolite, β-hydroxymethyl butyrate, is beneficial in improving outcomes in the elderly, following orthopaedic procedures and in nursing home residents;¹⁶² however, despite compelling data in preclinical models,²⁷ studies in patients with cirrhosis are still needed. Other approaches, including myostatin antagonists like follistatin,³³ require further evaluation given the limited benefit of myostatin-blocking antibodies in clinical studies.^126,163 Simultaneous targeting of causal factors and mediators of sarcopenia/physical frailty is likely to have the greatest impact on clinical outcomes given the multicomponent mechanistic models identified to date. Additionally, dynamic, rather than static, measures of sarcopenia/frailty are likely to be more appropriate outcome measures to guide improved clinical care.

Key points.

• Frailty, a decline in functional reserve across multiple physiological systems, and sarcopenia, a loss of skeletal muslce and impaired contractile function resulting in physical frailty, are common in patients with decompensated liver disease and may impact outcomes for transplant patients.

• Nutritional support and physicial activity are the primary interventions though there are limited data from large-scale trials.

• Future interventions based on recently identified potential theraputic targets may provide more effective treatments for sarcopenia.

Abbreviations

ACLF

acute-on-chronic liver failure

ACS

American College of Surgeons

ALD

alcohol-related liver disease

BCAAs

branched-chain amino acids

CFS

clinical frailty scale

ERAS

enhanced recovery after surgery

FFP

Fried frailty phenotype

ICU

intensive care unit

KPS

Karnofsky performance scale

LFI

liver frailty index

MELD

model for end-stage liver disease

NASH

non-alcoholic steatohepatitis

SMI

skeletal muscle index

SPPB

short physical performance battery