INTRODUCTION — Protein calorie malnutrition (PCM) has been described in 50 to 100 percent of patients with decompensated cirrhosis and at least 20 percent with compensated cirrhosis [1-5]. PCM is associated with a number of complications including development of variceal bleeding and ascites, increased surgical morbidity and mortality, reduced survival, and (in some studies) worsening hepatic function [2,6-11]. Patients with cirrhosis (particularly those with advanced disease) may also have micronutrient deficiencies. Recognition of macro- and micronutrient deficiencies is important since supplemental nutrition has been associated with a reduction in the risk of infection and in-hospital mortality and improved liver function parameters [12-15].

This topic will focus on nutritional assessment of patients with chronic liver disease.

PATHOGENESIS OF MALNUTRITION — The pathogenesis of malnutrition in cirrhosis is multifactorial [5]. Protein, carbohydrate, and lipid metabolism are all affected by liver disease. Contributing factors include inadequate dietary intake, impaired digestion and absorption, and altered metabolism.

●Anorexia, nausea, encephalopathy, gastritis, ascites, a sodium-restricted diet, and concurrent alcohol consumption can all contribute to a reduction in dietary intake.

●Malabsorption and maldigestion of nutrients can result from bile salt deficiency, bacterial overgrowth, altered intestinal motility, portal hypertensive changes to the intestine, mucosal injury, and increased intestinal permeability [16-20].

●Cirrhosis represents an accelerated state of starvation and as such, fuels other than glucose (protein, lipids) are used [21].

●There is an overall loss of protein from reduced synthesis of urea and hepatic proteins, reduced intestinal protein absorption, and increased urinary nitrogen excretion. Liver disease is associated with a lowered ratio of branched-chain to aromatic amino acids.

●Abnormal carbohydrate metabolism is associated with insulin resistance, impaired gluconeogenesis and reduced glycogen stores. As a result, lipids are preferentially oxidized for energy and the respiratory quotient (RQ) is less than in patients without chronic liver disease [22,23]. The RQ is defined as the ratio of the volume of CO2 production to the volume of O2 consumption.

Studies on the effect of chronic liver disease on the resting energy expenditure have mixed results [24-29]. One study suggested that energy expenditure in patients with cirrhosis was similar to controls after adjusting for body surface area [26]. In contrast, a cross-sectional study of 473 patients with cirrhosis found that 34 percent had hypermetabolism as measured by indirect calorimetry [30]. The increase in resting energy expenditure correlated with lean body mass but not with the severity or type of liver disease and was, in part, attributed to an increase in beta-adrenergic activity. Other studies have demonstrated that hypermetabolism persists at least one year after liver transplant and correlates with a reduction in survival [31,32].

It has also been suggested that ascites may contribute to an increase in energy expenditure and that its removal results in a significant reduction in the resting energy expenditure as measured by indirect calorimetry [27]. The mechanism underlying this observation is unclear [27].

GOALS OF NUTRITIONAL ASSESSMENT — The goal of a nutritional assessment is to identify nutritional risk that influences morbidity and mortality and which may be modifiable with targeted nutritional therapy. Nutritional assessment allows for a determination of the macronutrient (energy, protein, water) and micronutrient (electrolytes, minerals, vitamins, trace elements) state of a given individual. Body composition and muscle function analysis add supplemental information.

There is no gold standard for the assessment of nutritional status in patients with cirrhosis. It is not practical to attempt detailed nutritional assessment in all patients. We suggest a staged approach, beginning with a history and physical examination and proceeding with more detailed testing if needed.

Patients who meet any of the following criteria require further nutritional assessment [33,34]:

●Body mass index (BMI) of <18.5 kg/m²

●Child-Pugh C disease (decompensated cirrhosis)

For patients who do not have either of these risk factors, nutritional screening tools may identify additional high risk patients who may benefit from a comprehensive nutritional assessment. However, data are limited on the utility of nutritional screens for patients with cirrhosis. Small studies have evaluated the Liver Disease Undernutrition Screening Tool [35], the Nutritional Risk Screening-2002 [36], and the Royal Free Hospital-Nutritional Prioritizing Tool [37]. Although the latter screening tool shows the most promise, there are concerns about its subjectivity and low negative predictive value.

As a general rule, patients with compensated cirrhosis are more likely to be similar to a healthy population clinically and on laboratory investigations compared with those with decompensated cirrhosis. Thus, nutritional assessment is generally more detailed in patients with decompensated disease. However, many standard nutritional assessment tools have limitations when applied to patients with decompensated cirrhosis due to the presence of salt and water retention and advanced hepatic synthetic dysfunction.

HISTORY — Several key features should be evaluated in all patients, and especially for those with decompensated disease.

Weight history — Patients should be questioned about their usual weight, recent weight loss (over two weeks), and percentage of usual weight lost over six months. Unintentional weight loss of >10 percent over six months is considered severe. Although useful in compensated cirrhosis, the weight history is less accurate in patients with decompensated cirrhosis because of salt and water retention.

Dietary intake — Food intake can be assessed by the 24 hour dietary recall. The patient recounts meals and snacks on a typical day, intake of food from each of the food groups, and the use of nutritional supplements. Although the specific foods do not have to be documented in the chart, the process does assist with an overall appraisal of adequate balanced intake (see "Dietary assessment in adults").

Alcohol intake should also be quantified since it can limit the intake of food calories in alcoholic patients. Other barriers to intake in patients with cirrhosis, especially those who have decompensated cirrhosis, include anorexia, nausea, sodium restriction, and gastroesophageal reflux.

Gastrointestinal symptoms — The duration and frequency of comorbidities that may impact nutritional status, such as nausea, vomiting, diarrhea, and steatorrhea should be assessed. Symptoms that persist for more than two weeks and which are associated with a limitation in nutrient intake are concerning.

Liver disease — The nature and severity of liver disease should be categorized using standard techniques such as the Child-Pugh score, Model for End-stage Liver Disease score (MELD), or designation of compensated versus decompensated liver disease. The presence of ascites is one of the earliest features of decompensation. Standard nutritional assessment tools are limited by sodium and water retention and with hepatic synthetic dysfunction.

Micronutrient deficiency — Features suggestive of micronutrient deficiency should be sought. These include dermatitis (zinc, vitamin A, niacin), night blindness or photophobia (vitamin A), burning of the mouth or tongue (vitamin B12, folate), easy bruising (vitamin C, K), and paresthesias (thiamine, pyridoxine).

PHYSICAL EXAMINATION — We suggest performing a general physical examination with additional focus given to physical findings supportive of macro- or micronutrient deficiency.

●Self-reported weights can be inaccurate. Thus, we obtain a total body weight on a standard scale and use it to determine body mass index. The presence of pedal edema and ascites should be recorded with each weight measurement since weight fluctuates with salt and water retention.

●We evaluate patients for the presence of ankle or sacral edema, ascites, muscle wasting (as determined in the quadriceps and deltoids), and loss of subcutaneous fat (as determined in the triceps and chest).

●We examine patients for features supportive of micronutrient deficiency such as pallor (iron deficiency), hyperkeratosis (vitamin A), dermatitis (vitamin A), bruising (vitamin C, vitamin K), glossitis (vitamin B12, folate, niacin), angular stomatitis (vitamin B12), and reduced lower extremity deep tendon reflexes (vitamins B12, B1) (table 1).

SUBJECTIVE GLOBAL ASSESSMENT — Standardized batteries of questions and physical examination findings have been evaluated for nutritional assessment. An example is the subjective global assessment (SGA) tool, which can be administered at the bedside and includes a focused history and physical examination [38-40].

The historical components of the SGA include weight loss, change in dietary intake, presence of gastrointestinal symptoms, functional capacity and the metabolic demand associated with the disease state. Physical examination components include the presence of edema, ascites, muscle wasting, and subcutaneous fat loss. The presence of muscle wasting and subcutaneous fat loss is suggestive of severe malnutrition. Components are combined to obtain an SGA rating ranging from well-nourished to severely malnourished.

The inter-rater reliability of a modified SGA was assessed in a study of 20 liver transplant candidates; raters agreed on the nutritional status of patients 80 percent of the time [40]. Muscle wasting and fat depletion were the best predictors of the final SGA score.

The sensitivity of the SGA in patients with cirrhosis has been questioned in several studies [41-43]. In addition, although SGA correlates with post-operative outcomes in patients without cirrhosis [38-40], there is growing evidence that other assessments may be more useful for predicting other outcomes [41-46]:

●A study that included predominantly Child-Pugh A patients suggested that the standard SGA did not perform as well as assessment of handgrip strength in predicting complications of cirrhosis [41]. (See 'Anthropometry' below.)

●Another study of 79 patients with cirrhosis (60 percent Child-Pugh B or C) estimated the prevalence of malnutrition as 32 percent by the SGA (6 percent of Child A, 35 percent of Child B, and 72 percent of Child C) compared with 60 percent by body composition analysis (34 percent of Child A, 69 percent of Child B, 94 percent of Child C) [42]. However, whether body composition analysis represents a gold standard is unclear since the authors did not report clinical outcomes. In addition, 2 of 17 control patients were classified as being malnourished by body composition analysis [42,47].

●In a study of 315 patients with cirrhosis who were evaluated with skeletal muscle index (SMI) using cross sectional imaging and with SGA, there was weak concordance between sarcopenia (as determined by SMI) and the SGA [43]. However, concordance between the tests was not significant for patients who were overweight or obese. Moreover, while sarcopenia was associated with mortality in this study, SGA was not. This data supports muscle mass measurement as a useful prognostic variable in patients with cirrhosis.

A variation on the SGA, the Royal Free Hospital Global Assessment (RFH GA) tool, incorporates the body mass index, mid-arm muscle circumference measurements, and details of dietary intake to classify individuals as adequately nourished, moderately malnourished, or severely malnourished [48]. The RFH GA classification has differences between sexes. When compared with body composition data, it only significantly correlates with the depletion of total body protein stores or fat free mass in men, but not in women. Furthermore, the relationship between nutritional status and survival is significant in men, but not in women. In addition to clarifying its utility in women, these findings require external validation before the RFH GA can be used routinely.

LABORATORY EVALUATION — We obtain a serum albumin, creatinine and international normalized ratio (INR) in patients with compensated cirrhosis. Although these measures are still obtained in patients with decompensated cirrhosis, they are affected by hepatic synthetic dysfunction, and therefore are less likely to be independent markers of nutritional status. We obtain additional tests in certain settings (eg, alcohol use disorder, clinically evident malnutrition or micronutrient deficiency, cholestatic liver disease or anemia).

Plasma proteins — Several plasma proteins have been used to assess nutritional status such as albumin, pre-albumin, transferrin and coagulation factors (as measured by the prothrombin time or international normalized ratio). These proteins are synthesized in the liver and thus their deficiency can reflect either malnutrition or hepatic synthetic dysfunction. As a result, their assessment may be more helpful in early cirrhosis. Two small studies comparing plasma proteins to anthropometric measures (triceps skinfold thickness, arm muscle circumference) found a poor correlation in patients with advanced cirrhosis [49,50]. (See "Tests of the liver's biosynthetic capacity (eg, albumin, coagulation factors, prothrombin time)".)

The half-life of the plasma proteins should also be considered when using them for nutritional assessment. The long half-life of albumin (t1/2 approximately 20 days) makes it less reflective of acute changes in nutritional status while the relatively short half-life of pre-albumin (t1/2 approximately two to three days) makes it easily influenced by intercurrent illness.

Fat-soluble vitamins — Patients with advanced alcoholic liver disease and cholestatic liver disease (primary biliary cholangitis) commonly have fat-soluble vitamin deficiency. We routinely check plasma levels of vitamins A, D, E and obtain an INR in such patients. Vitamin D deficiency is common in patients with cirrhosis and worsens with the severity of liver dysfunction [51]. In addition to its role in skeletal health, vitamin D has been associated with anti-inflammatory, anti-fibrotic, and immunomodulatory properties [52]. (See "Vitamin D deficiency in adults: Definition, clinical manifestations, and treatment".)

Water-soluble vitamins and minerals — Water-soluble vitamin deficiency, especially thiamine, is common in alcoholic liver disease. Evaluation for treatable micronutrient deficiencies (selenium, magnesium, zinc, vitamin B12, folate) is performed in patients who are at high nutritional risk [53,54]. Plasma levels are a reasonable surrogate marker of body stores. In patients with alcoholic liver disease, decompensated cirrhosis, or clinically evident malnutrition, we routinely measure the above panel of water-soluble vitamins and minerals and advise daily multivitamin intake with additional supplementation based on specific deficiencies. It is advisable to utilize a non-iron containing multivitamin until iron stores can be evaluated. (See "Diagnostic approach to anemia in adults".)

Patients with chronic liver disease, including viral hepatitis, cirrhosis, and hepatocellular carcinoma, may have elevated serum vitamin B12 levels [55]. Vitamin B12 is stored in the liver and may be released during hepatic cytolysis. In addition, decreased vitamin B12 clearance may contribute to elevated levels. However, before attributing an elevated vitamin B12 level to chronic liver disease, other causes of an elevated serum B12 level, such as chronic myelogenous leukemia, promyelocytic leukemia, polycythemia vera, and the hypereosinophilic syndrome, should be considered.

Less reliable tests — The following tests are less reliable in the setting of chronic liver disease:

●Delayed hypersensitivity and total lymphocyte count – The accuracy of the delayed hypersensitivity skin test and total lymphocyte count are affected by multiple factors including cirrhosis. Thus, these are considered unreliable markers of nutritional status in chronic liver disease, and we do not routinely check them [6].

●Creatinine – Creatinine is a marker of lean tissue (protein) stores. Cirrhosis is associated with reduced hepatic creatine synthesis, decreased muscle mass, and increased tubular creatinine secretion. Because of these limitations, the validity measures based upon creatinine (such as urinary creatinine and the creatinine height index) as markers of nutritional status in cirrhotics is unclear [6,56].

ANCILLARY TESTS — Multiple nutritional assessment tools have been assessed in patients with cirrhosis. Although the use of these tools may vary from center to center, in our experience it is uncommon for clinicians to incorporate tools beyond the history and physical examination, subjective global assessment (SGA) tool, and laboratory investigations into routine clinical practice. These tools may be considered on an individual basis in those with evidence of malnutrition, although the subgroup of malnourished patients that may be missed by standard techniques has not been clearly defined. Referral to a dietitian experienced in evaluating patients with chronic liver disease can be helpful for patients who are malnourished by standard assessment or who are at very high risk of malnutrition (BMI <18.5 or those with Child Pugh C disease) [33].

Anthropometry — Changes in the composition of body compartments (water, fat, muscle) occur with liver disease and correlate with adverse events [57,58]. Anthropometry is a bedside tool used to assess body fat and lean tissue stores that is largely unaffected by salt and water overload that indirectly estimates body composition. In our experience, malnutrition is almost universally present and clinically obvious in patients with cirrhosis who have severe decompensation. Although anthropometry correlates with mortality in these patients, the subgroup in which anthropometric tools may, if at all, improve the detection of malnutrition above and beyond history, physical examination and global assessment tools such as the SGA has not been well defined.

Mid-arm muscle circumference (MAMC) is a marker of lean tissue stores. It incorporates both the mid-arm circumference (measured at the midpoint between the acromion and olecranon in the non-dominant arm) and skinfold thickness. Skinfold thickness (SFT) is used commonly to estimate fat reserve. The triceps skinfold thickness is measured by a caliper at the midpoint between the acromion and olecranon. For all of these measures, a value lower than the fifth percentile is considered diagnostic of severe malnutrition. Reference values used for this were established from the general population in 1981 [59]. Interobserver variability can be minimized by using a single trained observer [6]

Several studies have correlated these measures with a reference standard and with survival in patients with cirrhosis [45,60-64].

●A study of 69 patients with cirrhosis (88 percent Child-Pugh class A) evaluated the correlation of MAMC (and other anthropometric markers) with body cell mass depletion (BCM) using isotope dilution [60]. BCM is an established marker of protein calorie malnutrition in chronic liver disease and correlates with the development of nutritional complications [1,28,65,66]. The authors concluded that MAMC and handgrip strength were the best predictors of BCM depletion. A mid-arm muscle circumference of <23 centimeters in combination with a handgrip strength of <30 kilograms had a sensitivity of 94 percent and negative predictive value of 97 percent. By contrast, the SGA, weight-height measurement, triceps skinfold thickness, and biochemical/hematologic parameters did not correlate well with BCM.

●The accuracy of skinfold thickness was estimated in a study that included 40 patients with cirrhosis without ascites (60 percent Child-Pugh A) who were evaluated with a dual energy X-ray absorptiometry (DEXA) scan and skinfold measurements of the triceps, biceps, subscapular and suprailiac areas [61]. The results of both investigations were similar, suggesting that four site skinfold measurements could be used to estimate body fat in patients without ascites.

●Studies assessing the relationship between anthropometric measures and survival have shown mixed results [62-64].

Other assessments — Specialized body composition assessment methods have been evaluated, including cross-sectional imaging (computed tomography scan or magnetic resonance imaging) based muscle mass assessment, DEXA scan, deuterium oxide dilution, in vivo neutron activation analysis, and bioelectrical impedance analysis. However, routine use of these investigations in clinical practice is limited by cost, availability, and the impact of fluid retention.

Tests for detecting sarcopenia — In guidelines published by the European Association for the Study of the Liver (EASL), an assessment for sarcopenia is included as part of the nutritional assessment for patients with cirrhosis; however, the method may vary depending on local practice patterns and resources (eg, CT scan [if it is being performed for another clinical indication], anthropometry, DEXA, bioelectrical impedance analysis). (See 'Clinical practice guidelines' below and 'Anthropometry' above.)

Despite variation in the methods used to measure muscle mass and the definition of sarcopenia, cross-sectional imaging has demonstrated good correlation with pretransplant morbidity and mortality across studies [33,34,67]. (See 'Goals of nutritional assessment' above.)

However, cross-sectional imaging for assessing body composition is mainly used in research due to issues of availability, cost, radiation and contrast exposure, and the specialized equipment required for data analysis. Strategies to increase access to methods for analyzing body composition remain in development.

Bioelectrical impedance analysis — Bioelectrical impedance analysis (BIA) is performed by applying electrodes to one arm and one leg or by standing on a special scale. Impedance is proportional to the length of the conductor and inversely related to the cross-sectional area of the conductor. Accuracy in placement of electrodes is essential because even small variations can cause relatively large errors in the measurement of impedance and corresponding errors in the estimate of body water. A variety of formulas have been developed to convert the impedance, which measures body water, into an estimate of fat. Studies evaluating its accuracy in patients with cirrhosis have been mixed. (See "Determining body composition in adults".)

A study focusing on 41 patients with cirrhosis (20 with and 21 without ascites) compared results of BIA with mid-arm circumference (MAMC), triceps skinfold thickness (TSFT), and urinary creatinine excretion [68]. Total body potassium counting (TBP) was used as a reference standard for body cell mass. Although TSFT, MAMC, and urinary creatinine excretion correlated well with BIA in controls and cirrhotics without ascites, no correlation was found in cirrhotics with ascites. When compared to TBP, BIA was superior to TSFT, MAMC and urinary creatinine excretion in patients with cirrhosis. The correlation of BIA with TBP in patients with ascites was lower but still significant. However, other studies found BIA to be less accurate in both compensated and decompensated cirrhosis [69-71].

Dual energy X-ray absorptiometry (DEXA) — DEXA accurately estimates fat mass in patients with liver disease but does not provide as accurate an assessment of lean body mass [72,73]. This inaccuracy is most likely related to the increase in extracellular water in patients with decompensated liver disease [3,74]. DEXA retains utility in the diagnosis of osteoporosis and osteomalacia, particularly in patients with cholestatic liver disease [75].

Deuterium oxide dilution in vivo neutron activation analysis (IVNAA) — The deuterium oxide dilution can be used to accurately assess total body water, while the IVNAA can be used to accurately assess total body protein. However, these techniques are labor intensive and not routinely available [58].

Muscle function — Several studies have confirmed the importance of muscle strength as a predictive factor for malnutrition [60,76,77]. The measurement of muscle strength requires a handgrip dynamometer and is not currently part of routine nutritional status evaluation. In the study described above [41], hand-grip strength (result below mean minus 2 standard deviations) was a more sensitive predictor of malnutrition and the only method that was able to predict major complications (uncontrolled ascites, hepatic encephalopathy, spontaneous bacterial peritonitis, hepatorenal syndrome) at one year. The limitations of this study are that the SGA was used as the gold standard to obtain sensitivity, specificity and predictive values for handgrip strength. In addition, 88 percent of patients were Child-Pugh A and thus results may not be generalizable to those with more advanced disease [41].

The six-minute walk test and physical frailty assessments have also demonstrated prognostic value in patients with cirrhosis and may prove to be valuable adjuncts in the assessment of nutritional status [33].

ENERGY EXPENDITURE — Part of the assessment of nutritional needs requires an estimation of resting energy expenditure (REE), which can be assessed using standard formulas such as the Harris Benedict equation.

The accuracy of the Harris Benedict equation (HBE) in determining REE is subject to variability based upon the factors that make up the equation (weight, height, age, gender). The use of ideal body weight tends to underestimate energy requirements whereas the use of an actual weight may overestimate them. In one study comparing the HBE to indirect calorimetry, the HBE underestimated the mean energy requirements using either the actual or corrected (ideal) body weight [78]. Several other studies have noted underestimation of mean energy requirements of up to 18 percent using this and other prediction methods [21,28].

The gold standard for the assessment of REE is indirect calorimetry [79]. We do not do this routinely in all patients but instead reserve it for patients who do not meet nutritional expectations despite compliance with nutritional therapy. Indirect calorimetry may also be relevant in determining energy requirements for individuals with a BMI <20 in whom estimates from the HBE may be incorrect and in whom the implications of inappropriate feeding are more likely to have a negative impact on outcome.

An assessment of 268 patients with cirrhosis (34 percent Child-Pugh A) evaluated resting energy expenditure (indirect calorimetry), muscle function (using grip strength and a respiratory pressure transducer), and body composition (using neutron activation analysis and dual-energy X-ray absorptiometry) [80]. The following summarizes the major findings:

●Significant protein depletion was more common in men than in women (63 versus 28 percent) and occurred in half of all patients. These sex differences in metabolism require further study.

●Protein depletion was more common as Child-Pugh class increased and correlated with a reduction in muscle function.

●Hypermetabolism was identified in 15 percent of patients, a lower percentage than in previous studies. It was not associated with protein depletion, gender, severity or cause of liver disease, or concurrent ascites or tumor.

CLINICAL PRACTICE GUIDELINES — Nutritional intervention is required for patients who are undernourished, and multidisciplinary management includes input from a dietician and a clinician who is managing the patient’s liver disease [34]. Guidelines from the European Association for the Study of the Liver (EASL) and the European Society for Clinical Nutrition and Metabolism (ESPEN) address the following nutritional issues for patients with chronic liver disease [34,79]:

●Perform a nutritional screen in all patients with cirrhosis. Follow this with a detailed nutritional assessment to confirm a diagnosis of malnutrition and document its severity (see 'Goals of nutritional assessment' above).

●Include a sarcopenia assessment within the nutritional assessment (see 'Tests for detecting sarcopenia' above).

●For non-obese individuals (BMI ≤30 kg/m²), the minimum energy intake is 35 kcal/kg per day (actual body weight). For obese patients, energy intake is reduced by 500 to 800 kcal per day to provide a moderately hypocaloric diet [34]. An online tool for calculating the patient’s nutrition prescription can be found here [81].

●Protein intake target is 1.2 to 1.5 g/kg per day

●Use supplementary nourishment (oral or tube feeding) when patients cannot meet their caloric requirements with normal food:

•Whole protein formulas are generally recommended.

•Consider using more concentrated high-energy formulas in patients with ascites.

•Use BCAA-enriched formulas in patients with hepatic encephalopathy arising during enteral nutrition; the use of oral BCAA supplements can improve clinical outcomes in advanced cirrhosis.

●Use tube feeding if patients are not able to maintain adequate oral intake

●Avoid placement of feeding gastrostomy because of risk of complications

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Healthy diet in adults".)

SUMMARY AND RECOMMENDATIONS

●Protein calorie and micronutrient deficiencies are common in patients with cirrhosis, particularly in those with advanced disease. (See 'Introduction' above.)

●Assessment of nutritional status in patients with decompensated cirrhosis is challenging because of volume and sodium shifts, hepatic synthetic dysfunction, and changes in energy metabolism. (See 'Goals of nutritional assessment' above.)

●For all patients with cirrhosis, we take a history and perform a physical examination focusing on weight history, dietary intake, gastrointestinal symptoms, severity of liver disease and signs and symptoms of micronutrient deficiencies. (See 'History' above.)

For patients who are at high risk for malnutrition, we obtain further nutritional assessment including an evaluation for sarcopenia. (See 'Goals of nutritional assessment' above and 'Ancillary tests' above.)

●We obtain serum albumin, creatinine, and INR in patients with compensated cirrhosis. Although these measures are also obtained in patients with decompensated cirrhosis, they are affected by hepatic synthetic dysfunction and therefore are less likely to be independent markers of nutritional status.

We obtain additional tests (ie, selenium, magnesium, zinc, vitamin B12, folate, vitamin A, D, E) in certain settings (eg, decompensated cirrhosis, alcohol use disorder, clinically evident malnutrition or micronutrient deficiency, cholestatic liver disease or anemia). (See 'Laboratory evaluation' above.)

●Other specialized investigations, including the bioelectrical impedance analysis, dual energy X-ray absorptiometry (DEXA) scan, deuterium oxide dilution and in vivo neutron activation analysis are not measured routinely. (See 'Ancillary tests' above.)

●For patients requiring nutritional intervention, we calculate resting energy expenditure (REE) to estimate caloric requirements. We initially use prediction equations such as the Harris Benedict equation, recognizing that they can underestimate the REE. Indirect calorimetry is reserved for situations where, despite compliance to an estimated nutrient prescription, the patient is not making the expected nutritional gains. (See 'Energy expenditure' above.)