1. Definition and Diagnosis of Sarcopenia in the Research and Clinical Settings

Aldo J. Montano-Loza¹ and Maryam Ebadi¹

(1)

Department of Medicine, Division of Gastroenterology and Liver Unit, University of Alberta, Edmonton, AB, Canada

Aldo J. Montano-Loza

Email: montanol@ualberta.ca

Keywords

(5–8)Muscle wastingMuscle massMortalityLiver transplantationAdverse outcomes

Sarcopenia or muscle wasting is a common feature in patients with cirrhosis. Currently, it is well established that this complication is known to be a significant risk factor for overall mortality, mortality on the liver transplantation (LT) wait list, postoperative complication, and even post-LT death.

It is important to distinguish that sarcopenia is not the same as frailty. While sarcopenia is one of the main components of frailty, the latter is a syndrome characterized by decreased physiologic reserve and increased vulnerability to health stressors that predisposes patients to adverse health outcomes.

Muscle mass decreases with aging, with a loss of approximately 1% per year up to age 70, which increases later to 1.5% per year. However, the annual rate of muscle loss is twofold faster in patients with cirrhosis, and losses greater than 3% per year increase the risk of mortality, independent of the severity of liver disease [1].

In some cases, severe sarcopenia can be identified on clinical routine examination; however, early stages of muscle loss may not be visually evident, and the presence of ascites and obesity might make sarcopenia recognition difficult. As a result, objective, reproducible, and validated measures of muscle loss are essential. In addition, quantification of muscle mass provides objective data, which might be relevant for central decisions, such as candidacy for LT.

Sarcopenia refers to loss of muscle mass and function according to the European Working Group on Sarcopenia in Older People (EWGSOP). However, majority of studies of sarcopenia in patients with liver disease focus on the measurement of muscle mass alone. This might be related to the fact that function studies such as handgrip have not shown to correlate with CT muscle assessment in patients with cirrhosis [2]. In addition, muscle function can be affected by acute illness, particularly complicating inpatient assessment, whereas muscle mass is a more stable and objective parameter that is insensitive to acute changes in cognition or global physical function.

Sarcopenia provides an objective measurement of physical fitness that is applicable even to patients who cannot participate in bedside measures of frailty or for whom estimates of performance status may not reflect their overall health status.

In this chapter, we summarize the evidence from the medical literature to address definition of sarcopenia in patients with cirrhosis in the clinical and research setting.

Definition of Sarcopenia

Sarcopenia is a term which was first adopted by Rosenberg in 1989 in the geriatric literature [3]. Sarcopenia was initially defined as an age-related loss of skeletal muscle, generally described as lean appendicular mass (normalized to height squared) greater than two standard deviations below that typical for healthy young adults [4]. The term sarcopenia has since been expanded to reflect low muscle mass leading to negative effects on physical performance and clinical outcomes across a broad range of disease states.

The new International Statistical Classification of Diseases and Related Health Problems, 10th revision (ICD-10-CM) (M62.84), code for sarcopenia represents a significant recognition as a disease [5]. Great heterogeneity in the metric used to define sarcopenia exists among published study cohorts [6]. Furthermore, studies have varied in their use of a standard definition of sarcopenia defined by threshold values, versus evaluating sarcopenia by percentile among a specific population. The specific definition used for determining sarcopenia is obviously linked to the method of measurement, as not all metrics apply or have been validated across modalities.

Modalities to Evaluate Muscle Mass in Patients with Cirrhosis

Multiple modalities exist for muscle mass quantification, including indirect and direct techniques such as anthropometry, bioelectrical impedance analysis (BIA), dual-energy X-ray absorptiometry (DEXA), ultrasound (US), magnetic resonance imaging (MRI), and computed tomography (CT) (Table 1.1).

Table 1.1

Modalities evaluating sarcopenia in cirrhosis

	Advantages	Disadvantages
Sarcopenia tools for the clinical setting
Mid-arm muscle circumference (MAMC)	Cheap, noninvasive, wide availability	Low reproducibility, measurements are affected by subcutaneous adipose tissue loss, requires specialized training
Bioelectrical impedance analysis (BIA)	Noninvasive, quick, simple, inexpensive	Fluid retention, diuretic use, liquid and food intake before the test, and physical activity alter the results
Dual-energy X-ray absorptiometry (DEXA)	Noninvasive, inexpensive, wide availability, consistent results, and little radiation exposure	Influenced by lower limb edema
Sarcopenia tools for the research setting
Ultrasound (thigh muscle thickness)	Noninvasive, simple, reliable, inexpensive without radiation exposure, independent of edema	Unidentified reproducibility
CT/ MRI	Quick, precise, independent of ascites or edema, capability to quantify muscle and adipose tissue quality and quantity	Expensive, radiation exposure in longitudinal CT images

Choosing an appropriate assessment technique for clinical practice versus research depends on various features such as cost, accuracy, accessibility, and feasibility of the technique in the population of interest. While all methods may be applicable in the general population, many are not appropriate or accurate in cirrhosis. Mid-arm muscle circumference measurement, a form of anthropometry, requires specialized training and has a low reproducibility.

DEXA which has become the standard of care modality for measuring bone mineral density, may also be used to measure appendicular skeletal muscle mass. DEXA-estimated appendicular skeletal muscle mass, which is adjusted to body size, has been adopted by consensus groups (International Working Groups on Sarcopenia and the EWGSOP) to define criteria for the determination of sarcopenia [7]. However, these criteria have not been validated in larger data sets and diverse populations, and since it is a projection-based rather than cross-sectional modality, DEXA inherently suffers from challenges related to tissue overlap. In addition, the fluid shifts common to patients with decompensated cirrhosis reduce the accuracy of DEXA and BIA.

In contrast, US is a promising modality that deserves additional study as it is inexpensive, noninvasive, and free of radiation exposure, has applicability in a variety of practice settings, and can provide results independent of fluid retention [8]. However, more studies are needed before it can be recommended in clinical practice.

Cross-sectional imaging is a routine part of LT assessment in most centers to evaluate the vascular and biliary anatomy for surgical planning and as screening for hepatocellular carcinoma (HCC). The existence of routine cross-sectional imaging in patients evaluated for LT has provided an opportunity for assessment of muscle mass in patients with cirrhosis without the need for additional testing. Indeed, the majority of studies on sarcopenia in LT have used CT imaging.

MRI is also commonly used in LT patients and provides an additional opportunity to assess body composition. While preliminary reports suggest that CT and MRI-based imaging yield equivalent results [9], validity across both techniques needs to be established. Sensitivity and specificity of techniques for capturing longitudinal changes are inevitable criteria for predicting patients’ long-term outcomes.

Muscle area on cross-sectional CT images can be assessed by using the Hounsfield units from −29 to +150. Measurements can be obtained in a semiautomated way with the help of the software SliceOmatic (TomoVision, Montreal, Quebec, Canada). Besides SliceOmatic, other software has been used to determine the SMI (FatSeg, OsiriX, ImageJ).

Overall, CT image analysis can be used for opportunistic body composition assessment, and therefore, inclusion of cross-sectional imaging as part of LT assessment should be considered in the future to develop and validate consistent criteria for sarcopenia.

While individual investigators have made these techniques relatively high throughput in a research setting, an efficient clinical tool or process for measuring muscle mass in clinical practice has not been developed. “Single-slice” or abbreviated CT is a technique with obvious appeal, as it is based upon the modality with the most use in published research studies and is widely available [10]. While caution regarding radiation exposure is appropriate in general, the dose involved in such limited CT scans is miniscule compared with other medical imaging modalities (e.g., equivalent to a standard chest X-ray). Furthermore, such limited scans can be done without the need for intravenous contrast exposure. Importantly, since CT is often performed in these patients for other reasons, these images may already be available for analysis in a large proportion of patients. However, automation of measurement of muscle mass such that a diagnosis of sarcopenia could be readily made has not been established. Nevertheless, this modality is likely to be the most applicable to clinical practice, as it requires little additional work for the clinician, is not of undue burden to patients, and minimizes operator error in measurement. Similar research efforts have been attempted using single-slice MRI, which eliminates the concern of radiation exposure although may be logistically and financially more costly [11].

Determination of Sarcopenia in Patients with Cirrhosis

In the past decade, a blast of research on sarcopenia in cirrhosis has emerged. This information has significantly advanced our knowledge; however, the field is currently hampered by heterogeneous definitions, measurements, and study design. A main factor confounding the current literature is the use of different cutoffs to define sarcopenia in cirrhosis.

Englesbe et al. first reported sarcopenia using the cross-sectional area of the total psoas area (TPA) on CT imaging. Using the lowest quartile to define sarcopenia, an association between TPA and post-LT mortality was found [12]. Subsequent studies evaluating patients assessed for LT used CT cutoffs for sarcopenia associated with higher mortality risk derived from skeletal muscle index (SMI) in cancer patients [13]. CT-measured SMI, calculated as total muscle area at the third lumbar vertebrate (L3) normalized to height squared, is a strong surrogate of whole body muscle mass [10] and can be applied as a reliable marker of whole body muscularity. Among patients with cirrhosis, CT-measured SMI at L3 has been the most well-validated tool to define sarcopenia and correlates strongly with survival. To create standardized cutoff points for sarcopenia in cirrhosis, Carey et al. recently defined sarcopenia as SMI <50 cm²/m² in male and <39 cm²/m² in female patients with cirrhosis, listed for LT based on the data from North America centers, part of the Fitness, Life Enhancement, and Exercise in Liver Transplantation (FLEXIT) Consortium [14]. Defined in this way, sarcopenia was delineated based on sex-specific SMI values associated with mortality independent of age and MELD score [14]. To date, this is the only North American study to define sarcopenia using the data derived from a diverse population of patients with cirrhosis awaiting LT. Figure 1.1 highlights the total skeletal muscle quantification at L3 from two male patients.

../images/467095_1_En_1_Chapter/467095_1_En_1_Fig1_HTML.jpg — Fig. 1.1

Total muscle area quantification at the level of third lumbar vertebra using abdominal CT images from two male patients with cirrhosis. (a) and (b), respectively, present a patient who had low SMI (39 cm²/m²) and high SMI (54 cm²/m²)

Apart from total SMI at L3, various psoas muscle measurements including cross-sectional area [12, 15], index [16], and thickness [17] have been applied to predict mortality risk in patients with cirrhosis. However, it has been recently reported that psoas muscle index has limited performance for identifying patients with higher waitlist mortality risk in cirrhosis as compared to SMI [18].

Discrepancies in sarcopenia prevalence between studies likely result from divergent criteria for defining sarcopenia (values below 5th percentile of age- and sex-matched normal population or optimal cut points for the mortality discriminations), outcome of interest, and study population (Table 1.2). In the past, cut points derived from other populations, such as patients with cancer, have been applied, and these may not be appropriate for patients with cirrhosis. The use of different imaging software between studies does not appear to be a significant concern, as excellent agreement has been observed between various software programs when compared directly for quantification of abdominal skeletal muscle cross-sectional area [19].

Table 1.2

Summary of studies investigating sarcopenia diagnosis in cirrhosis

Author/year	Study population	Sarcopenia assessment	Sarcopenia cutoff	Criteria for cutoff
Merli et al. 2010 [30]	150 patients with cirrhosis hospitalized to a tertiary care	Mid-arm muscle circumference (MAMC)	NA	MAMC below the 5th percentile of age- and sex-matched normal population
Englesbe et al. 2010 [12]	163 LT recipients	CT-measured psoas muscle	TPA <1420 mm² at the level of the fourth lumbar vertebra (L4)	Lowest TPA quartile
Montano-Loza et al. 2012 [13]	112 patients with cirrhosis evaluated for LT	CT at the level of the third lumbar vertebrae	L3 SMI ≤38.5 cm²/m² for women and ≤52.4 cm²/m² for men	Mortality-associated SMI cutoffs in cancer [31]
Krell et al. 2013 [28]	207 LT recipients	CT-measured psoas muscle	TPA <1499 mm² for men and <954 mm² for women	Lowest TPA tertile
DiMartini et al. 2013 [25]	338 LT recipients	CT at the level of the third lumbar vertebrae	L3 SMI ≤38.5 cm²/m² for women and ≤52.4 cm²/m² for men	Mortality-associated SMI cutoffs in cancer [31]
Masuda et al. 2014 [27]	204 patients undergoing LT	CT-measured psoas muscle	PMA <800 cm² for men and <380 cm² for women	PMA below the 5th percentile for each sex
Durand et al. 2014 [17]	562 patients listed for LT	CT-measured psoas muscle	Transversal psoas muscle thickness [TPMT/height (mm/m)] at the level of the umbilicus ≤16.8 mm/m	Optimal cutoffs of TPMT/height to discriminate waiting list mortality
Hara et al. 2016 [32]	161 patients with cirrhosis	Bioelectrical impedance analysis (BIA)- measured upper limb skeletal muscle mass (kg)	1.2 kg/m² for women and 1.7 kg/m² for men	Values less than −1 standard deviation from the mean values in the control group (diabetic patients)
Carey et al. 2017 [14]	396 patients listed for LT	CT at the level of the third lumbar vertebrae	SMI <39 cm²/m² for women and <50 cm²/m² for men	Optimal cutoffs of SMI to discriminate waiting list mortality
Golse et al. 2017 [15]	256 LT recipients	CT-measured psoas muscle	PMA <1561 mm² for men and <1464 mm² for women	Optimal cutoffs of PMA to discriminate 1-year post-LT mortality
Tandon et al. 2016 [8]	159 patients evaluated for LT	CT at the level of the third lumbar vertebrae	L3 SMI ≤38.5 cm²/m² for women and ≤52.4 cm²/m² for men	Mortality-associated SMI cutoffs in cancer [31]
Belarmino et al. 2018 [33]	144 male patients with cirrhosis	Appendicular skeletal muscle index (ASMI) with dual-energy X-ray absorptiometry (DEXA)	DEXA-ASMI ≤7 kg/m²	Lowest tertiles of the DXA-ASMI
Van Vugt et al. 2018 [29]	224 patients listed for LT	CT at the level of the third lumbar vertebrae	L3 SMI <44.1 for men and <37.9 for women	Lowest sex-specific quartile of SMI
Praktiknjo et al. 2018 [34]		Magnetic resonance imaging (MRI)	Fat-free muscle area <2895 mm² for women and Fat-free muscle area <3197 mm² for men	Optimal cutoffs to discriminate 3-year survival