Sarcopenia and liver transplant: The relevance of too little muscle mass

Table 1 Methods used to define muscle mass or function in cirrhosis

Method	Measurements obtained	Limitations	Notations
Multiple (four) compartment model	Total body water	Requires combinations of methods (such as water dilution, densitometry)	Best model for cirrhosis when fluid overload is present
	Body fat
	Fat free dry matter (protein)
	Bone mineral content
Dual X-ray Absorptiometry	Body fat Fat free body mass Bone mineral content	Limited ability to differentiate between lean tissue and body water with excess body water resulting in overestimation of fat free mass[1]	Peripheral measures of lean tissue are less impacted by ascites[24,26]
		Ascites, especially more than 4 liters, can significantly impact truncal measures[26,27]
Cross sectional imaging	Estimate skeletal muscle volume	Can be used to determine differences in skeletal muscle between groups[31] Studies use different muscle groups, anatomic levels and cutoff values to diagnose sarcopenia	Measurements 5 cm above the level of the 4th-5th lumbar vertebra had the highest correlation with total body skeletal volume[31]
Biochemical methods	Skeletal muscle mass	Total body protein is a measure of functional muscle mass and can be done through techniques such as in vivo neutron activation analysis 24 h urine creatinine is one method, but is limited in cirrhosis where renal insufficiency is common[93]	Use calculations based on these methods to quantify muscle mass
Bioelectrical impedence analysis	Fat free mass Fat mass	Guidelines recommend against routine use of BIA under states of altered hydration[94]	Segmental BIA was found to have a better correlation and lower standard error in estimating body cell mass in the setting of cirrhosis without ascites but still performs poorly with ascites present[95]
		BIA estimates of total body water were found to be accurate in cirrhotic patients without ascites, but performed poorly when ascites was present[23]	Phase angle can be used. In a study including participants with a wide range in severity of liver disease, phase angle was positively correlated with total body protein, muscle mass and muscle strength[96]
Anthropometry	Estimate muscle mass	Edema alters results of anthropometry overestimating muscle mass[8]	Mid arm circumference was found to be one of the most accurate anthropometric measures[29] and was most predictive of clinical outcomes[97]
Functional Measures	Measures ability to perform physical task, for which muscle function is one component	Common functional measures assess all systems involved in exercise including cardiovascular, pulmonary, hematologic, neurologic and musculoskeletal[33]	Include tests such as submaximal cardiopulmonary fitness tests, six minute walk test, hand grip strength and isokinetic strength of flexion and extension at different joints Often simple tests such as hand grip correlate with measures of skeletal muscle[29]

BIA: Bioelectrical impedance analysis.

Table 2 Measures, definitions and outcomes relating to sarcopenia in the setting of cirrhosis and liver transplant

Study	Method	Definitions used/proposed	Outcomes	Notes/Limitations
Selberg et al[96]	BIA, phase angle	> 5.4° normal 4.4°-5.4 borderline < 4.4° abnormal	Phase angle < 5.4° associated with significantly lower survival	Phase angle may remain normal in cases of severe tissue loss when proportional losses of extracellular mass and body cell mass may occur
Kaido et al[11]	BIA, multiphase device (InBody 720; BioSpace, Tokyo, Japan)	< 90% skeletal muscle mass compared to standard or body cell mass below 23.0 kg	Survival was significantly decreased in recipients with low skeletal muscle mass or low body cell mass	No data is provided on volume status, although Child-Pugh classification is given
	Percent skeletal muscle mass against a standard and calculated body cell mass			Nutritional supplementation with branched chain amino acids improved survival in those with low skeletal muscle mass
Englesbe et al[15]	CT, combined area of right and left psoas muscle area at the highest level of the 4th lumbar vertebra Control population was 248 trauma patients	Percentile cutoffs for total psoas area in transplant population 1910 mm2 50th percentile 1420 mm2 25th percentile 950 mm2 5th percentile	Decreased psoas muscle area associated with higher risk of mortality 25th percentile HR = 1.88 5th percentile HR = 3.46	Retrospective definitions of sarcopenia were not derived from the control trauma patients, but were based on percentiles from the transplant population Included CT scans either 90 d before or after transplant; majority of scans were after transplant
Tandon et al[12]	CT or MRI, cross sectional area of muscle at 3rd lumbar vertebra (psoas, paraspinals, transversus abdominis, rectus abdominis and internal and external obliques)	Total L3 skeletal muscle area ≤ 52.4 cm2/m2 in males ≤ 38.5 cm2/m2 in females	Sarcopenia present in 41% of wait listed candidates Higher wait-list mortality with sarcopenia (HR = 2.36, 95%CI: 1.23-4.53) Greatest effect was in those with low MELD score	Retrospective Only study to report use of both MRI and CT
Montano-Loza et al[18]	CT cross sectional area of muscle at 3rd lumbar vertebra (psoas, paraspinals, transversus abdominis, rectus abdominis and internal and external obliques) Muscle identified by Housfield unit between -29 and + 150	Total L3 skeletal muscle area ≤ 52.4 cm2/m2 in males ≤ 38.5 cm2/m2 in females	Sarcopenia present in 40% of cirrhotics Sarcopenia was independent risk factor for mortality (HR = 2.28, P = 0.008) One year survival for cirrhosis with sarcopenia was 53% compared to 83% in cirrhosis without sarcopenia	Prospective data
Hamaguchi et al[14]	CT, cross sectional psoas muscle area at level of umbilicus Intramuscular fat accumulation of multifidus muscle (multifidus muscle Housfield units/subcutaneous fat Housfield units)	ROC curves selected from study data for best accuracy in predicting death Intramuscular adipose tissue content -0.375 in males and -0.216 in females Psoas muscle mass normalized for height ≤ 6.868 cm2/m2 in males ≤ 4.117 cm2/m2 in females	Pretransplant increased intramuscular adipose tissue content (OR = 3.898, 95%CI: 2.025-7.757) and decreased psoas muscle mass (OR = 3.635, 95%CI: 1.896-7.174) were associated with mortality	Used umbilical level which can vary based on body habitus Constructed cutoffs based on diseased population Included intramuscular fat content as a measure of muscle quality
Tsien et al[13]	CT cross sectional at mid 4th vertebra level	Psoas muscle area normalized 5th percentile cutoffs ≤ 12.27 cm2/m2 in males less than 50 yr of age	Sarcopenia was seen in 62.3% prior to transplant and increased to 86.8% after transplant	Includes serial measures in the same patients
	Total cross sectional area of psoas, paraspinals and abdominal wall muscles (rectus abdominis, oblique and transversus abdominis) normalized to height	≤ 10.12 cm2/m2 in males more than 50 yr of age ≤ 10.47 cm2/m2 in females less than 50 yr of age ≤ 10.33 cm2/m2 in females more than 50 yr of age	Only 6.1% had reversal of sarcopenia after transplant and 75% without pretransplant sarcopenia developed it after transplant	Mean time from transplant to post-transplant CT was about one year (13.1 ± 8.0 mo)
	Reference ranges derived from 109 healthy control subjects undergoing CT for unspecified abdominal pain	Total abdominal muscle area normalized 5th percentile cutoffs ≤ 60.09 cm2/m2 in males less than 50 yr of age ≤ 48.97 cm2/m2 in males more than 50 yr of age ≤ 53.43 cm2/m2 in females less than 50 yr of age ≤ 41.28 cm2/m2 in females more than 50 yr of age	Reduction in muscle after transplant was associated with new onset diabetes mellitus	Since follow up scan was done for indications (ie HCC surveillance, infection, pain, increased aminotransferases) the potential for significant selection bias exists
Masuda et al[9]	Cross sectional CT of psoas muscle at L3 Calculated area by multiplying major and minor axis of psoas (a × b ×∏)	< 800 cm in men < 380 cm in women	3 and 5 yr survival with sarcopenia was 74.5% and 69.7% respectively, without sarcopenia was 88.9% and 85.4% respectively (P = 0.02)	Enteral nutrition given in immediate post operative period appeared to decrease risk of sepsis when sarcopenia was present
	Compared to a reference group of healthy donors		Sepsis was seen in 17.7% with sarcopena, 7.4% without sarcopenia (P = 0.03)

MELD: Model for end stage liver disease; CT: Computed tomography.

Table 3 Methods and outcomes when measuring muscle function in the setting of cirrhosis and liver transplant

Study	Method	Outcomes	Notes/limitations
Andersen et al[37]	Isokinetic strength of flexion and extension of six joints	Upper and lower extremity strength was decreased in cirrhotics vs controls	Only included patients with alcohol related cirrhosis
		Lower extremity strength was associated with lean body mass and mid arm circumference, an effect independent of severity of liver disease, neuropathy, biochemical data and recent alcohol use	The majority of patients had Child-Pugh A or B classification Included 24 cirrhotics and 24 controls
Tarter et al[98]	Isokinetic strength measured by upper and lower extremity peak force, peak torque, total work and power	Most measures of strength were decreased in cirrhotic patients vs controls	Study included 49 with alcoholic cirrhosis, 42 with non-acoholic cirrhosis and 50 controls
		There was no difference in any measure between those with alcohol vs non-alcohol related cirrhosis	No patient had consumed alcohol in greater than one year prior to testing
Beyer et al[35]	Maximal oxygen uptake measured on a cycle ergometer SMWT Isokinetic knee flexion and extension	Maximal oxgen uptake, SMWT and isokinetic knee strength increased over the first six months after transplant compared to pretransplant values No changes were noted between six and 12 mo after transplant	Small study with only 17 patients having post transplant data and 13 patients completing both pretransplant and posttransplant assessment of maximal oxygen uptake Used a supervised exercise program after transplant
Epstein et al[38]	Symptom limited cardiopulmonary testing on a cycle ergometer	When examining patients that went on to transplant, a significantly higher proportion of patients that died within the first 100 post-operative days had a peak oxygen consumption < 60% predicted and had oxygen consumption at the anaerobic threshold < 50% predicted peak oxygen consumption	Median MELD at the time of exercise testing was low (7-12) The median time from exercise testing to transplant was long (471 ± 300 d)
Prentis et al[39]	Symptom or exertional limited cardiopulmonary testing on a cycle ergometer	Sixty tested patients went on to liver transplant with a 10% 90 d mortality	Mean MELD at transplant was low (< 20)
		Mean aerobic threshold was higher in survivors and was only variable in multivariate analysis that was associated with mortality	Compared to above study (Epstein 2004[38]), the authors did not make comparisons to population based reference values, but used ROC curve analysis to define thresholds associated with outcomes
		Optimal anaerobic threshold associated with survival was > 9 mL/min per kg
		Anaerboic threshold > 11 mL/min per kg was associated with shorter stay in critical care setting
Carey et al[34]	Six minute walk test	Candidates awaiting liver transplant had decreased SMWT distance (369 ± 122 m), significantly lower than reference values	Included patients too ill to walk, and designated zero m for this group
		When controlling for other factors including age and MELD, SMWT distance was significantly associated with wait list mortality (HR = 0.58, 95%CI: 0.37-0.93)
		ROC analysis found cut off value of 250 m having the highest sensitivity and specificity for mortality	Designated patients removed from the list as a waitlist death
Alameri et al[36]	Six minute walk test	Patients with cirrhosis had significantly diminished SMWT distance (306 ± 111 vs 421 ± 47 m, P < 0.0001)	Used Child-Pugh to assess severity of liver disease, no data on MELD
		SMWT was an independent predictor of survival and was inversely correlated with Child-Pugh classification
		The lowest quartile walked < 250 m

SMWT: Six minute walk test; MELD: Model for end stage liver disease.