11. Evidence for the Assessment of Physical Frailty and Sarcopenia in Hospitalized Patients and the Role of Assessing Changes Over Time
Patients with advanced liver cirrhosis require frequent hospitalization, mostly due to cirrhosis-related complications, especially complications of portal hypertension, infections, acute or acute-on-chronic kidney injury, and hepatocellular carcinoma .
Importantly, hospitalized patients with cirrhosis also have a high risk for readmission within 30 days ranging from 20 up to 37% [2–4]. As risk factors for readmission, Volk and colleagues identified severity of liver disease (based on MELD score), serum sodium, and number of medications as measures for the complexity of the medical regimen . Furthermore, in patients readmitted within 30 days, 90-day mortality was significantly higher than in patients that were not readmitted within 1 month (26.8% vs 9.8%) .
Due to the significant risk for further complications, readmission, and fatal outcome, hospitalized patients with cirrhosis should undergo a careful risk assessment as part of the standard work-up. Based on the individual risk profile, specific measures can be implemented with the final goal of reducing the hospital stay, preventing future complications, and decreasing the overall burden of cirrhosis and its complications.
Hospitalized Patients with Cirrhosis
Hospitalized patients with advanced liver cirrhosis have a significant burden of morbidity and mortality. The risk for further deterioration and for liver-associated complications is increased not only in patients with acute decompensation or acute on chronic liver failure, but also in patients with elective procedures like pre-transplant evaluation. No matter what the reason for the hospitalization is, it facilitates a thorough assessment regarding cirrhosis-associated risk factors and prognosis. Several scores have been developed in the past specifically for hospitalized patients but also more general for patients with cirrhosis.
Pathophysiology of Muscle Breakdown
Due to the decrease in glycogen production and storage, relatively short episodes of fasting, for instance, overnight fasting, lead to an increase in proteolysis and lipolysis, and amino acids from the muscle are used as source of gluconeogenesis to fuel tissues where fatty acid carbon cannot be used [5, 6]. During hospitalization, patients are confronted with additional procedure-related fasting episodes, for instance, during preparation for GI investigations or while on the intensive care unit. This may further accelerate muscle breakdown.
CPT and MELD
Conventional prognostic scores commonly used in patients with cirrhosis are the Child-Pugh-Turcotte (CPT) score and the Model for End-Stage Liver Disease (MELD) score, the former including the two subjective parameters: severity of ascites and hepatic encephalopathy. The latter had been developed to predict mortality after transjugular intrahepatic portosystemic shunt (TIPS) placement and then used as an allocation system for deceased donor liver grafts based on the objective laboratory parameters: creatinine, bilirubin, and INR . Later on, the MELD score had been further refined by adding sodium as the fourth parameter. The resulting MELD-Na score is today widely used in the organ allocation process for liver transplants [8, 9]. These scores can be applied during hospitalization as well as in the outpatient setting.
In 2013, Moreau and colleagues introduced the Chronic Liver Failure-Sequential Organ Failure Assessment (CLIF-SOFA) score to better characterize and stratify hospitalized cirrhotic patients with acute decompensation regarding severity of disease and risk of mortality . The CLIF-SOFA score has been developed as an adaptation of the not-cirrhosis-specific Sepsis Organ Failure Assessment (SOFA) score and includes six subscores to assess the liver, kidney, brain, circulation, coagulation, and respiration. Importantly, this score reflects expert opinion and is based on consensus rather than on data and did not lead to a significant improvement of prediction accuracy compared with the MELD and the MELD-Na score .
In 2014, the CLIF-C ACLF score was published, adding age and white blood cell count to the CLIF-C Organ Failure (CLIF-SOFA) score. This lead to a significant improvement of the prediction of mortality of patients that were hospitalized due to decompensation of cirrhosis as compared with MELD, MELD-Na, and Child-Pugh score . For patients in intensive care units, the Acute Physiology and Chronic Health Evaluation (APACHE) score can be used to predict mortality [12, 13].
However, not all complications of cirrhosis are adequately reflected in the above listed scores that either do not specifically target hospitalized patients (Child-Pugh, MELD, MELD-Na score) or explicitly address subpopulations (i.e., patients with ACLF or ICU patients), and some of these scores are [10–13] technically demanding and require a specific training.
How to Assess Physical Frailty
In recent years, physical frailty evolved as a new, important, independent, and potentially modifiable risk factor in patients with advanced chronic liver disease. In hospitalized patients with cirrhosis, frailty has not yet been investigated extensively, but a limited set of frailty assessment tools from other fields of medicine has been validated successfully in this patient population . Tapper et al. investigated three different metrics in the hospital setting (activities of daily living (ADL), Braden scale, and Morse fall scale) that enable a fast and reliable assessment of hospitalized patients with cirrhosis. Although not associated with 30-day hospital readmission, frailty measures were predictive for 90-day mortality, length of hospital stay, and rehabilitation needs . The three tools used for the assessment of physical frailty are described in more detail in the following sections.
Activities of Daily Living (ADL) Score
Activities of daily living is assessed as the self-reported ability to feed, toilet, bath, dress, bathe, and transfer. Each point is rated as “dependent” (1 point), “needs assistance” (2 points), and “independent” (3 points) summing up to a maximum of 15 points [15, 16]. In their study, Tapper et al. found an odds ratio (OR) of 1.83 (95% confidence interval [CI] 1.05–3.20) for 90-day mortality and an OR of 3.78 (95% CI 1.97–7.29) for discharge to a rehabilitation hospital if the ADL score was less than 12 of 15.
The Braden scale is a measure that is used for the assessment of pressure ulcer risk. It consists of a physical exam and the additional parameters: skin sensory perception, moisture, mobility, activity, nutrition, and friction (ability to hold a comfortable position in a chair and bed). While a score of 23 or more defines no risk for skin lesions, a score of less than 16 is indicative for an increased risk and consequently requires physical and nutritional therapy [15, 17]. For inpatients with decompensated cirrhosis, the OR for discharge to a rehabilitation hospital was 6.23 (95% CI 2.53–15.4) if the Braden scale was <16. In case of an intermediate Braden scale (16–18), an increase in mortality risk was identified (OR 1.62, 95% CI 1.03–2.56).
Morse Fall Scale
The risk to fall can be assessed with the Morse fall scale that includes history of falling, ambulatory aids, intravenous access, gait disturbances, secondary diagnoses, and mental status [15, 18]. However, the Morse fall scale did not predict the mortality risk in hospitalized patients with decompensated cirrhosis.
How to Assess Sarcopenia
Another important risk factor in patients with advanced cirrhosis is sarcopenia. The best objective method to assess sarcopenia in cirrhotic patients is based on cross-sectional CT or MR image analysis . Muscularity can be analyzed on abdominal CT or MR scans that have been performed as part of the routine assessment or on single slice images at the respective lumbar level. The need for special image analysis software as well as training to assess the images limits this type of analysis to specialized centers .
Sarcopenia Risk Scores
Since sarcopenia is not included in the standard risk assessment tools, new risk scores have been developed. The MELD-sarcopenia score  and the MELD-psoas score  both take into account the muscularity normalized by height with specific cutoffs for male and female patients. Especially for patients with a MELD score below 20, sarcopenia is an additional risk factor for unfavorable outcome . Although not reflected in the current risk scores, special attention should be given to obese patients with cirrhosis, since sarcopenic obesity and myosteatosis are associated with a higher mortality risk .
Treatment of Sarcopenia and Frailty
Ideally, a comprehensive assessment of frailty and sarcopenia is not only used to assess specific risks but also to implement a targeted treatment strategy. If the diagnosis of sarcopenia is confirmed, a nutritional supplementation strategy can be deployed, and in case of frailty, specific physical measures can be implemented. These interventions should address the inhospital period and the time after hospital discharge.
A cornerstone of treatment in patients with frailty and/or sarcopenia is the optimization of nutrition . Daily oral nutritional supplementation (1000 kcal and 34 g of proteins) in patients with alcoholic cirrhosis and treatment with branched-chain amino acids (BCAA) in patients with advanced cirrhosis each have been associated with a reduction in hospital admission rate [24, 25], and BCAA supplementation furthermore lead to an improvement of health-related quality of life . However, oral or enteral nutritional treatment did not affect overall survival in a meta-analysis performed by Ney et al. . The number of trials investigating the effect of physical exercise in mainly compensated cirrhotic patients is still small as is the total number of patients investigated. Although effects on survival and long-term benefit are unclear to date, a positive effect on peak VO2, quadriceps muscle, self-perceived health status, and 6-min walk test could be observed [27–29].
Specific recommendations regarding caloric and protein intake are given in the recently published EASL Clinical Practice Guideline “Nutrition in Chronic Liver Disease” . Daily caloric intake should be at least 35 kcal/kg body weight (BW) in nonobese patients, and the optimal protein intake consists of at least 1.2–1.5 g/kg BW. In sarcopenic obese patients, a moderate hypocaloric diet in combination with a high protein intake of >1.5 g/kg BW is recommended. In decompensated cirrhotic patients, a late evening snack should be added to compensate for catabolic episodes during the night. Prolonged fasting periods in the context of diagnostic or therapeutic procedures and during ICU stays should be avoided whenever possible.
It is important to note that frailty and sarcopenia as well as the associated treatment strategies should be reassessed in a longitudinal way not only during a hospital stay but also in the outpatient setting .
In the USA, the number of hospitalization due to cirrhosis increased from 2001 to 2011 from 350,000 to 650,000 per year, and in parallel, the hospitalization-related costs increased from 5 to 10 billion dollars.
Prevention of Readmissions
Patients with cirrhosis have a high risk for readmission ranging from 20% to 37% [2, 32, 33]. Typical disease-specific complications that require readmission are hepatic encephalopathy, spontaneous bacterial peritonitis, and esophageal variceal bleeding .
Due to the high and rising costs that are related to inhospital treatment and especially readmissions of patients with cirrhosis, prevention strategies are important to reduce the economic burden of this patient population. Intervention programs typically require a multidisciplinary approach and should target high-risk patient groups. Readmission reduction programs are not only medically meaningful, but also from a financial point of view .
Tapper et al. defined in their review five different strategies to decrease the readmission rate of patients with cirrhosis: (1) interventions should be integrated in the clinical workflow, (2) default options are more powerful than voluntary actions, (3) knowledge improvement should focus on the front line clinicians, (4) process improvements do not always translate into better outcomes, and (5) any successful intervention must include viable alternatives to hospitalization .
So far, almost no specific pharmacological treatment of sarcopenia and frailty is available. In male patients with low testosterone, testosterone substitution has demonstrated a positive effect on muscle mass . Ammonia-lowering agents or myostatin antagonists are promising new targets that have to be investigated in more detail in the future .
It is important to note that sarcopenia may persist or even progress after liver transplantation and that only a minority of patients show an improvement of sarcopenia posttransplant . Whether an optimal nutrition strategy, ideally in combination with a physical therapy, can improve the long-term and especially the posttransplant outcome regarding sarcopenia and frailty needs to be investigated in future trials.
To date, literature on sarcopenia and frailty assessment in hospitalized patients is still limited. Most techniques and measures used to assess sarcopenia and frailty can be applied in the hospital as well as in the outpatient setting. However, the significance of specific findings may differ significantly in a hospitalized patient, where length of hospital stay, rehabilitation needs after discharge, rehospitalization risk within the next months, and mortality risk are of special interest.
Sarcopenia is ideally assessed as normalized total muscle area or psoas muscle area in CT or MR images at the lumbar level L3; however, a correct assessment requires specific image analysis software and trained personnel. Other assessment techniques like DEXA scan analysis are less precise but easier to perform, do not require special software tools or training, and are associated with less radiation than full CT scans. For the assessment of frailty, clinical scores like the activities of daily living (ADL) score or the Braden scale that have been developed in other medical fields can be used in hospitalized patients with cirrhosis.
Sarcopenia has an impact on survival on the transplant list as well as on the posttransplant evolution. Whether it also affects rehospitalization rates or mortality risk in hospitalized patient with cirrhosis is currently not known. In contrast, frailty measured during hospitalization is associated with an increased risk for rehospitalization and an increased risk for death.
Several measures may help to improve the outcome of hospitalized sarcopenic and frail patients. Optimization of nutrition, if required with enteral tube feeding or parenteral nutrition, the maintenance of a sufficient protein intake, supplementation of vitamin and trace element deficits, and the implementation of specific training to reverse frailty are the most important measures in hospitalized patients with advanced cirrhosis. Whether these interventions effectively lead to less complication, shorter hospital stays, a decreased readmission rate, and a lower mortality risk is currently not fully investigated.
Successful interventions have the potential to improve the medical situation of patients with cirrhosis, to reduce hospital admission and especially readmission rate, and to decrease cirrhosis-related healthcare costs. However, these strategies need further evaluation in future prospective clinical trials.