3. Deciphering the Cirrhotic Patient’s Present Status: The Overlap Between Physical Frailty, Disability, and Sarcopenia
KeywordsLiver diseaseCirrhosisPortal hypertensionHepatic encephalopathyMalnutrition
“It is often difficult to decide whether a trial of [chemotherapy] is indicated.”
“The decision should be based on a review of the patient’s history and an evaluation of his present status, so that an estimate can be made of the probable therapeutic effect of the agent in a particular situation.” – Dr. David Karnofsky 
Transplant, like many therapies, possesses a therapeutic window. In much the same way that Karnofsky described the performance status as a decision aid to define the therapeutic window for chemotherapy, clinicians often turn to factors beyond the Model for End-Stage Liver Disease (MELD) in order to classify patients who are and are not likely to benefit from transplantation. Among the known “non-MELD” indicators, frailty, disability, and sarcopenia are the concepts. As predictors, they are closely associated with adverse events including hospitalization, transplant delisting, and death . Each is captured using different measures. Frailty, a syndrome of decreased physiologic reserve, is best captured through demonstrations of physical function (e.g., chair stands, handgrip, balance); disability, or limited activity, is typically captured by patient report (e.g., activities of daily living [ADLs]). Sarcopenia or loss of muscle bulk can be quantified using multiple imaging modalities, anthropometrics, and bioimpedance. However, despite the differences in how each concept is measured, they share common roots. Taken together, each factor provides helpful information for clinicians seeking an enhanced understanding of the patient’s present status with an eye toward that which can (or cannot) be rehabilitated.
In this chapter we review the derivation and application of these concepts in the context of cirrhosis care with a focus on their shared pathogenesis.
A Conceptual Model
First, a patient with a rapid decline in health such as acute-on-chronic liver failure due to alcoholic hepatitis or hepatitis B flare may present as disabled and frail but with preserved muscle bulk. Conversely, a patient with nonalcoholic steatohepatitis (NASH)-related cirrhosis who has had refractory ascites for multiple years may present with preserved ability, mild or absent frailty, and severe sarcopenia. Relatedly, the patient’s present status is fundamentally a snapshot of a dynamic health status. It varies with the fluctuations of encephalopathy and ascites and changes drastically in response to treatment (e.g., alcohol abstinence, treatment of ascites/hepatic encephalopathy [HE]) and illness (e.g., infection, falls, hospitalization).
Second, and relatedly, the severity of liver disease is a key driver of poor present status. Above all, the presence of HE is a central governing factor in the patient’s clinical phenotype. HE is disabling and interferes with cognitive function, balance, and coordination , confounding the assessment of frailty . Hyperammonemia is myotoxic and causes sarcopenia . Ascites limits physical function, diminishes appetite, and causes excess caloric burning.
Third, extrahepatic comorbidities further define the patient’s status commensurate with their severity and the organs effected. Fourth, socioeconomic status reflects employment categories such as manual labor that may be associated with strength that could remain disproportionate to global muscle loss. It also speaks to food security and food choices which impact nutritional status.
Sarcopenia is common in patients with cirrhosis and is associated with worse outcomes before and after liver transplantation [5–7]. The mechanism for the sarcopenia’s association with mortality in patients with cirrhosis is unclear but appears most closely linked to an increased risk of infections . However, the current epidemiology and associations with sarcopenia are confounded in important ways. Accurate estimates of sarcopenia’s prevalence in patients with cirrhosis are complicated by multiple factors. First, studies of sarcopenia have employed highly heterogeneous methodologies ranging from the area examined (e.g., ultrasound, bioimpedance, total psoas area, psoas at L3, psoas at L4; adjusted for height using the skeletal muscle index). Second, the prevalence of sarcopenia varies with severity of liver disease as captured by Child classification or MELD score, factors which are heterogeneously distributed in cohorts that examine the clinical impact of sarcopenia. Third, there are few prospective validations of sarcopenia as a biomarker for risk in patients with cirrhosis. Sarcopenia is therefore determined from tests that have been ordered for other reasons. Although it is unclear, the reasons for the scan in which sarcopenia was disclosed may impact the risk of sarcopenia. At the same time many patients may not be candidates for the modality chosen. For example, chronic kidney disease is associated with sarcopenia as well . However, many CT (computed tomography) scans are performed only in patients with either preserved renal function (given the risks of iodinated contrast) or those who are undergoing hemodialysis. Similarly, patients with abdominal symptoms, perhaps leading to anorexia and malnutrition, may be more likely to receive a CT from which sarcopenia can be measured.
The value and clinical impact of frailty in patients with cirrhosis is largely derived from studies of the outpatient transplant-waitlisted population [2, 9–13]. This population is likely markedly different from and may not generalized to patients who are compensated, newly decompensated, actively encephalopathic, critically ill, or undergoing an evaluation as an inpatient. In contrast to sarcopenia, however, frailty measures (e.g., 6-minute walk distance, handgrip, chair stands) are not routine components of the clinical evaluation. Accordingly, although the clinical context in which the frailty measure is obtained shapes the characteristics of the included cohort, studies of frailty are generally prospective allowing for accurate adjudication of future risks and outcomes.
Disability is evaluated subjectively by a trained observer or by patient report. Patient-reported ADL performance obtained by clinical nurses (or research assistants) has also been associated with a host of pretransplant outcomes including death, delisting, discharge to a nursing facility, and 30-day readmission after discharge [13–15]. The prototypical assessment of disability, the Karnofsky Performance Scale (KPS), has been extensively validated in patients with cirrhosis. The KPS is both a scale from 0 (death) to 100 (perfect health). It has demonstrated enhanced risk prediction in evaluating the survival hospitalized patients with decompensated cirrhosis [16, 17], waitlist mortality , and posttransplant survival [18–20]. Disability can be obtained in most clinical contexts and can also be evaluated retrospectively where it is recorded as part of clinical care . Although it can be performed widely, it can be affected by acute changes in the patient’s present status and may need adjustment, for example, for the presence of overt HE or infections [15, 17].
Given the overlap of sarcopenia, frailty, and disability, interventions to forestall their associated complications share common features. Foremost among these is malnutrition . Poor nutritional status can lead to sarcopenia and functional outcomes such as frailty or disability. Unsurprisingly, protein-energy malnutrition is associated with mortality in patients with cirrhosis [21, 22]. Furthermore, many patients who are objectively frail may not be able to reliably prepare adequate nutrition. In order to address malnutrition, a suite of interventions are needed.
The European Society for Parenteral and Enteral Nutrition recommends 35–40 kcal/kg/day and 1.2–1.5 g/kg ideal body weight/day of protein in patients with liver disease [23, 24]. A convenient rule of thumb to calculate protein needs adjusting for the conversion from actual to ideal body weight is to use 1 g/kg. This is particularly important for patients with HE who benefit from nutritional supplementation and high-protein diets [25, 26]. Protein restriction is categorically not recommended as it can lead to sarcopenia. Branched-chain amino acids can be supplemented safely for this purpose as well, although availability is limited and the data are not conclusive [27, 28].
The timing of nutritional administration seems to influence overall nourishment in cirrhosis. Owing to poor hepatic glycogen reserve as well as increased catabolic drive, sleeping itself – overnight fasting – may cause a decline in nutritional status by inducing catabolism and proteolysis [29, 30]. In a randomized controlled trial, 103 patients with cirrhosis received either isocaloric daytime or nighttime supplementary nutrition (710 kcal/day) . The group receiving nighttime nutritional supplementation had significant increases in total body protein.
Patients with cirrhosis may also present with deficiencies in micronutrients that could impact the interpretation of frailty and disability. Particularly in the case of alcohol-related liver disease, zinc deficiency is often present . Zinc deficiency may be associated with HE which, as above, could confound the interpretation of frailty . Vitamin D deficiency is also common. Vitamin D deficiency is independently associated with falls and vitamin D deficiency-related osteoporosis is associated with fractures, both of which increase the risk of disability .
Disability, frailty, and sarcopenia may all be sensitive to change from increased physical activity. It is easy to recommend exercise, but specific considerations should be addressed first . We recommend beginning with 10-minute sessions once to twice daily and slowly increased the workload as tolerated. Subjects should note an increased heart rate and respiration or breathing rate, without experiencing anginal symptoms and allowing for unencumbered speech while exercising. While most non-disabled patients with cirrhosis can accomplish brisk walking, for those at high risk of falls or limited mobility, other approaches are needed. Consultation with physical therapy is often helpful. Beyond that, we refer patients to the videos produced by Go4Life, an exercise and physical activity campaign from the National Institute on Aging (go4life.nia.nih.gov/workout-videos/).
As above, a patient with cirrhosis can be frail for many reasons. When present, HE plays an important, potentially primary role in the frailty phenotype. Hyperammonemia, one of the sources of HE, causes muscle catabolism which leads to sarcopenia [4, 35]. HE is associated with anorexia, exacerbating malnutrition and sarcopenia . Similarly, HE leads to falls which can cause a fear of activity, increasing deconditioning. Above all, HE leads to hospitalization which worsens physical decline. We recently studied a large prospective cohort of patients assessed for frailty at the time of transplant evaluation . We found that HE appears to be an important determinant of both the frailty phenotype and adverse health outcomes associated with frailty. It is not clear whether treatment of HE, including early identification of covert HE, would forestall or improve physical function. Further research is needed, but it is also justified. Basic science supports that ammonia-lowering interventions may reduce muscle proteolysis . In patients, treatment of HE improves coordination in a driving simulator . Trials are needed, however, to show whether HE-directed therapy improves sarcopenia, frailty, and disability in patients.
It is increasingly recognized that non-MELD factors such as disability, frailty, and sarcopenia are important predictors of meaningful outcomes for patients with cirrhosis. Although each concept is measured in different ways, they share a common biology. Understanding the overlap between these measures informs practice in important ways. Underlying each factor is malnutrition, deconditioning, and HE. Interventions to forestall or improve disability, frailty, or sarcopenia will require a deliberate, multimodal approach to address each of their drivers.