An impaired nutritional status can have serious implications for patients in acute and chronic care settings recovering from illness or surgery.1 Deficits in the nutritional status can negatively impact any and all of the following: healing process, functional status, central nervous system, immune system, and the risk of morbidity and mortality in hospitalized patients.2 Up to 60% of patients over age 65 admitted to hospitals have some degree of malnutrition. Older patients are at higher risk because their intake is frequently inadequate to meet the increased metabolic needs that an acute illness or surgical intervention may require.3

[black small square] Importance and Purpose

Researchers have demonstrated the relationship between poor nutritional status and length of hospital stay, morbidity, and mortality in hospitalized patients.2 It is in the interest of both the hospital and patients to identify those at nutritional risk as early as possible so interventions can be initiated to prevent adverse outcomes. Routine screening of nutritional status often includes evaluation of height, weight, and body mass index (BMI).4 The patient who is not 10% or more below ideal body weight may not receive further assessment of nutritional status. Nutritional assessment tools are often subjective and dependent on the operator's consistency of anthropometric measurements. Objective data of weight, height, and BMI depend on a patient's history for assessment of change from prior nutritional status. For example, if a patient's preevaluation weight was 195 pounds (88.6 kg) and no history was present, how would you assess a 40-pound (18.2-kg) weight loss over 2 months? A significant health risk can occur after an involuntary weight loss of 10% of body weight over 6 months.5 The purpose of this article is to review recent studies to evaluate the validity and reliability of serum albumin and prealbumin levels as clinical markers for risk assessment of morbidity and mortality from malnutrition at the time of hospital admission.

[black small square] Search Method

Literature relating to the use of albumin and prealbumin for nutritional screening and evaluations was identified through electronic searches of the databases from 1999 to 2005 of Medline, cinahl, Health Source: Nursing/Academic Edition, Cochrane Library database, and SportsDiscus. Search terms included nutritional screening, nutrition assessment, malnutrition, albumin, prealbumin, nutritional interventions, and feeding assessments. A total of 152 unique articles were identified. Articles were excluded if they were review papers, editorials, or single-care reports (30), not available in English (27), animal studies (7), the population focus specifically had muscle wasting or protein loss (proteinuria) as a diagnostic criteria (end-stage renal failure, end-stage AIDS, acute eating disorders, burns over 50% of body surface area) (66), children under 6 years of age (5), and albumin or prealbumin levels were not included in initial nutritional assessment (9).

The remaining nine research publications met the inclusion criteria of use by nurse, dietitian, or healthcare provider in the clinical setting and use of albumin or prealbumin levels in the initial nutritional assessment.

[black small square] Current Practice

National standards. Higher healthcare cost, morbidity, and mortality are compelling reasons for hospitals to initiate nutritional assessments to identify at-risk patients. The Joint Commission has emphasized nutritional assessment "when warranted" on every patient within 24 hours of hospital admission for patients identified as at risk.6 Hospital leaders are expected to develop nutritional plans and manage, assess, and improve the quality of these plans.7 The Joint Commission standards leave the choice of nutritional assessment to the professionals in the individual hospitals. An inspector looks for documentation of an initial assessment, planned intervention, assessment of the intervention, and quality improvement process. However, The Joint Commission does not mandate the type of nutritional assessment to be utilized.6

[black small square] Using Biometric Markers for Malnutrition

Researchers have utilized serum albumin and prealbumin (transthyretin) as biometric markers of a patient's nutritional status upon admission into the hospital.8-10 Serum albumin is routinely obtained as a part of the comprehensive metabolic profile (CMP). Prealbumin is not a standard component of the CMP and must be ordered specifically.4

Serum Albumin

Glucose is the body's preferred energy source for metabolism. In its absence, protein can be utilized as a source of energy through catabolization of visceral and muscular skeletal systems.11 In a nutritional deficit, the body will initially utilize glycogen stores. When these reserves fail to meet the physiologic needs, gluconeogenesis from protein in organs and muscles becomes the primary source for glucose. Key adaptive mechanisms that take place when inadequate glucose stores exist are: conservation of energy, resulting in a depressed central nervous system, drop in metabolic rate, and depressed immune system, utilization of fat deposits as the main source of energy, and preservation of lean mass to minimize protein loss.5

Serum albumin is considered the most accurate indicator of a patient's protein store. Its half-life is approximately 20 days. Large amounts of albumin are contained in body stores, meaning a patient could be malnourished before a drop is noted in the serum albumin.5 Anorexic patients have been found to have normal serum album levels despite severe malnutrition due to this catabolization of visceral systems.4 While production of serum albumin can affect anorexic patients, synthesis is decreased, thus increasing redistribution and decreasing catabolism of muscles and visceral organs. Although there is no national standard or consensus, a serum albumin level of 3.5 grams/dL is usually considered low, and a level of 2.5 grams/dL and below has been identified as a significant marker for mortality in acutely ill patients.5 Medical conditions other than malnutrition can lower serum protein levels. These include sepsis, nephrotic syndrome, liver disease, trauma, burns, and large wounds.8

Prealbumin

Prealbumin has a half-life of 1 to 3 days; therefore, levels respond quickly to protein deprivation and nutritional intervention. It has become a vital objective tool in evaluating acute nutritional changes.9 Prealbumin has a much shorter half-life than serum albumin and is therefore considered a more reliable indicator of the patient's nutritional status.4 Levels of prealbumin have also been used as markers of morbidity and mortality in hospitalized patients.8 Prealbumin levels respond rapidly to nutritional support and are often used as an indicator of the effectiveness of the feeding intervention.10

[black small square] Other Biometric Measures

Bioimpedance. Bioimpedance is a technique that measures the changes in a small electrical current passed through the body. Water significantly improves conduction of the current. Because fatty tissue contains higher water content than other tissue, conduction is facilitated. Bioimpedance is useful to monitor trends in free fat mass, but must be used carefully in the evaluation of nutritional status because hydration status greatly affects the results.11

Free fat mass index. Researchers routinely use BMI as an indicator of nutritional status. However, BMI does not account for muscle mass; BMI only compares height with weight. If a person is heavier due to muscle mass, they may actually have an adequate nutritional status. The free fat mass index (FFMI) takes into account the amount of muscle mass a person is carrying related to their height by calculating body mass in meters squared instead of height in relationship to body weight. The FFMI can be skewed by hydration status and should be used to monitor trends.12

Inflammatory markers. Several of the studies reviewed included inflammatory markers in the assessment of nutritional status (C-reactive protein and lymphocyte count).13-15 As previously mentioned, conditions other than malnutrition can alter serum albumin levels.8 The use of inflammatory markers allows the researcher to consider these conditions in their nutritional assessment.

Retinol binding protein. Retinol binding protein (RBP) is the primary plasma transport for retinol (vitamin A). Retinol binding protein binds to prealbumin to deliver retinol to specific receptors. Immunoassays for RBP are used in the differential diagnosis of liver disease, retinol deficiencies, and low prealbumin states including malnutrition.16

[black small square] Biometric Measures of Nutrition in Research

Nine studies met the inclusion criteria for this review.2,4,12,14,15,17,19,20 A quasi-experimental study examined the effects of standard nutrition compared with increasing the protein source in meals for older women with hip fractures. All participants had low prealbumin levels and low to borderline albumin levels at baseline and after 3 months of intervention without significant changes in albumin or prealbumin. These researchers concluded that albumin and prealbumin were not recommended for evaluating acute nutritional changes.17

In two cross-sectional studies, albumin and prealbumin were compared with previously established nutritional assessment instruments.4,12 Reported results indicated that albumin and prealbumin did not recognize individuals with malnutrition that were identified by the nutritional assessment instruments and physical exams and therefore were not recommended for nutritional assessment.

Two retrospective studies were identified in the literature search. One examined preoperative and postoperative patient changes in several nutritional markers including albumin and prealbumin.14 No mention of postoperative hydration status presented. All measures of nutrition were noted to significantly decrease except BMI in only 48 hours postoperatively in a previously well-nourished population. Inflammation, fluid overload, and blood loss could account for some of these measures, but because albumin's half-life is much longer than 48 hours, it is doubtful that nutritional status was solely responsible for these changes.

In the other retrospective study, protein-calorie malnutrition (PCM) was diagnosed on the criteria of a prealbumin level of less than 160 mg/mL and compared with albumin levels. The effects of PCM were examined in terms of length of stay (LOS) and inhospital mortality. The researchers reported that individuals with PCM had longer LOS and higher mortality. They concluded that PCM is underdiagnosed, but can identify patients that would benefit from nutritional supplementation. The authors made a specific note that PCM can be the result of inflammatory states, and not nutritional status, and that inflammatory states present higher risks for increased LOS and mortality.18

Four prospective studies were identified in a literature search. Two utilized expert opinions to determine nutritional status and examined how well albumin and prealbumin (and other biometrics) correlated with these experts.13,19 Conflicting results were reported: One study reported that albumin appropriately identified malnourished patients13; the other stated it did not, but prealbumin was an indicator of nutritional status.19

The remaining two studies correlated a standard nutritional instrument2 and an illness severity scale15 with albumin, prealbumin, BMI, and markers of inflammation. These researchers reported that all biometric markers examined underestimated the number of patients with malnutrition.2,15

Most of these papers report that using albumin or prealbumin as markers of nutritional status would miss malnourished patients that were identified with standard instruments or physical exams.2,4,12, 14, 15, 17 Two researchers reported that prealbumin could improve identification of malnourished patients18,19, and one researcher reported that albumin was an acceptable indication of nutritional status.13 Rationales for albumin and prealbumin failing to appropriately identify nutritional status included: albumin levels are affected by chronic disease states (cirrhosis and nephrotic syndrome both decrease serum albumin levels), decreased albumin levels are found in inflammatory states, serum albumin levels can shift with hydration status, and finally corticosteroids can lower serum albumin even in well-nourished patients.2,4,13,17,18

[black small square] Summary and Recommendations

Serum albumin is often identified as a measure of nutritional status. Both serum albumin and prealbumin have been compared and contrasted with other measures of nutritional status (nutritional history, physical exam, BMI, anthropomorphic measurements of height, weight, triceps skin fold, bioimpedance, and skin fold measurements). Current research has not consistently supported the use of serum albumin or prealbumin as an independent biometric marker of malnutrition. Serum albumin levels are reflective of the patient's nutritional status 18 to 21 days prior to the sample and drop in response to chronic inflammation. Prealbumin has a half-life of 1 to 3 days; therefore, levels respond quickly to protein deprivation and nutritional intervention.

Because the use of any single biometric measure is not supported as specific or reliable for an assessment of malnutrition, the use of serum albumin or prealbumin as an independent biometric marker of malnutrition for hospitalized patients cannot be recommended. Biometric measures of albumin and prealbumin cannot be substituted for a comprehensive nutritional history and physical assessment for risk of malnutrition. Medical conditions other than malnutrition can lower serum protein levels, making serum albumin an unreliable tool for use as an independent measure of nutrition.

The application for serum albumin levels has been well documented in clinical trials as a clinical indicator for LOS, morbidity, and mortality in hospitalized patients. This alone makes it a valuable and cost-effective adjunct to the admission assessment of the hospitalized patient. Prealbumin has been shown to be an evidence-based, objective biometric measure for the monitoring of nutritional interventions in the hospital setting.

Additional research is needed to identify a tool, or combination of tools, that will provide nursing a simple, rapid, and reliable way to assess the patient's nutritional status within 24 hours of admission. Until that time, there is no acceptable substitution for a complete nutritional history and physical assessment to determine the nutritional status and needs of the hospitalized patient.

REFERENCES