Assessment of protein-energy nutritional status is one of the most common applications of biochemical assessment. The most commonly used proteins for this purpose are serum albumin, prealbumin, and transferrin. Other biochemical parameters are useful as screening tools: serum creatinine, cholesterol, and bicarbonate.
Serum proteins can be broadly divided into two categories, negative and positive acute phase proteins. Circulating levels of negative acute phase proteins, such as serum albumin and prealbumin, are at their highest in well-nourished, unstressed individuals and decline in the presence of inadequate nutrition, inflammatory stress, or both. In contrast, positive acute phase proteins, such as serum C-reactive protein (CRP), normally circulate at very low levels and rise dramatically in the presence of inflammatory stress. Because both nutrition and inflammation independently influence each of these nutritional markers and respond to nutritional interventions, both must be considered when interpreting serum protein levels as nutritional or inflammatory parameters.
The expected frequency of laboratory testing should be considered in selecting nutritional markers. Where nutritional assessments are performed monthly in generally stable individuals, longer half-life nutritional parameters are logical choices. In contrast, for patients undergoing acute changes in nutritional and inflammatory state, addition of short half-life nutritional and inflammatory markers allows adjustment of nutritional interventions based on acute patient response.
Albumin is the most abundant protein in the blood, readily available on most biochemistry panels, and is therefore widely used as a nutritional and inflammatory marker. The half-life of serum albumin is approximately 20 days, making it a good tool for use in monthly nutritional assessments but relatively unresponsive to acute changes in nutritional or inflammatory status.
Interpretation of albumin levels is challenging in people with CKD. Both modifiable and nonmodifiable predictors of serum albumin have been identified in people with CKD. Older age, female sex, white race, presence of several chronic diseases (chronic obstructive pulmonary disease, peripheral vascular disease, diabetes mellitus, and cancer), and being the first year of dialysis are nonmodifiable factors correlated with hypoalbuminemia. Modifiable factors associated with improved albumin include smoking cessation, use of arteriovenous fistulas, or biocompatible dialysis membranes (4). Longitudinal analyses of serum albumin show a decline in serum albumin in the months immediately preceding death and an improvement during the first year of dialysis (4).
Serum albumin is extensively distributed in both intravascular and extravascular compartments and can be redistributed into the extravascular space, resulting in hypoalbuminemia. Serum albumin levels are also reduced in patients with metabolic acidosis. Nutrient intake, particularly protein intake, is one of several factors determining serum albumin levels (5). However, serum albumin levels are maintained in otherwise healthy individuals until late in the course of protein-energy malnutrition (PEM) in the absence of underlying inflammatory stress. Serum albumin is also preserved with chronically low food intake because resting energy expenditure (REE) is simultaneously down-regulated. However, systemic inflammation inhibits this normal adaptation to protein-energy deficits. In addition, inflammation also both inhibits albumin synthesis and increases its fractional catabolic rate (5,6). Thus, dietary protein and inflammation have separate and opposing influences on serum albumin. Interpretation of hypoalbuminemia must be made in the context of both nutrient intake and presence of inflammation (6).
Subnormal serum albumin levels (<4.0 g/dL) have long been shown to predict both all-cause and cardiovascular mortality in people receiving maintenance dialysis (5,7). Without additional information to help differentiate between nutritional deficits, inflammation, or a combination of the two, hypoalbuminemia should not be presumed to be nutrition-related. The view that hypoalbuminemia is not primarily due to nutritional deficits appears to be supported by the limited success of enteral or parenteral intervention trials to effectively correct hypoal-buminemia (8). Others, however, have concluded that although both inflammation and malnutrition frequently coexist, subnormal levels of serum albumin and prealbumin are principally reflective of nutritional inadequacy (9).
Serum prealbumin, like albumin, is a negative acute phase protein. Prealbumin has a half-life of about 2 days and is therefore very responsive to recent events, especially calorie and protein deficits. Thus, for patients with acute illnesses or following initiation of nutritional interventions, prealbumin can be a useful, early directional indicator of changes in nutritional and inflammatory status.
Serum prealbumin, and a decline in serum prealbumin over time, predicts all-cause mortality independently of inflammation (based on CRP) (9). Predialysis serum prealbumin is directly correlated with other biochemical nutritional markers (serum albumin, creatinine, and cholesterol) as well as predialysis body weight and bioelectrical impedance (BIA)-derived reactance, body cell mass, body water, and phase angle (10). Prealbumin was predictive of hospitalization over the next 12 months in a univariate analyses, though when included in a multivariate analysis, prealbumin was no longer significant (11).
When the KDOQI nutrition guidelines were drafted, there was insufficient published data to conclude that prealbumin was a more sensitive or specific nutritional marker than serum albumin (3). However, subsequent publications indicate that predialysis serum prealbumin is a useful addition to nutritional profiles and provides additive information to serum albumin (10,12). Patients with a predialysis serum prealbumin of less than 30mg/dL should be evaluated for nutritional adequacy (3). This threshold is within the normal range
(~ 17-45 mg/dL) for people without kidney disease, reflecting decreased renal clearance of prealbumin in patients with CKD.
Serum creatinine is a nutritional screening parameter in people receiving maintenance dialysis (3). Predialysis serum creatinine concentration reflects the sum of creatinine by dietary origin (creatine and creatinine from meat) and that formed endogenously in skeletal muscle tissue less the creatinine removed by residual kidney function and dialysis. Creatinine is formed irreversibly from creatine in skeletal muscle at a fairly constant rate that is directly proportional to skeletal muscle mass. Thus, under steady state conditions of diet and dialysis, predialysis serum creatinine is roughly proportional to lean body mass. A declining predialysis serum creatinine over time in otherwise stable dialyzed patients indicates loss of skeletal muscle mass. Although not commonly used in clinical practice, creatinine index can be calculated to easily estimate fat-free body mass, especially in anuric patients (3).
Serum creatinine, and a decline in serum creatinine over time, predicts all-cause mortality independently of inflammation (as measured by CRP) (9). Serum creatinine levels are directly correlated with both serum albumin and serum prealbumin. The relationship between serum creatinine and mortality in maintenance dialysis patients is typically a backward "J" shape, with the lowest mortality occurring at a predialysis creatinine level of 9-11 mg/dL and rising significantly at lower levels and modestly at higher levels (7). Lower levels of predialysis serum creatinine reflect low dietary intake of creatinine and creatine as well as low lean body mass. High serum creatinine typically is suggestive of inadequate dialysis. The KDOQI nutrition guidelines recommend that dietary adequacy be evaluated in patients exhibiting serum creatinine levels of less than approximately 10 mg/dL (3).
Low serum total cholesterol is correlated with markers of protein nutritional status (serum albumin, prealbumin, and creatinine) and with mortality in most, but not all, trials. The relationship between nutrient intake and low serum total cholesterol is indirect. The presence of hypocholesterolemia, below 150-180 mg/dL, or a declining serum cholesterol concentration can be an indicator of chronically inadequate protein and energy intake (3).
Serum cholesterol is primarily useful as a screening tool because its sensitivity to, and specificity for, changes in protein and energy intake is poor. Serum cholesterol also is depressed with chronic inflammation. The relationship between mortality and serum cholesterol is usually "U" shaped, with lowest mortality occurring with serum cholesterol levels of about 200-220 mg/dL in most trials and increasing for higher or lower values. A relationship between CRP and serum cholesterol has been reported, with patients at both high and low extremes of the serum cholesterol distribution having higher CRP levels (13). Low levels of cholesterol and elevated CRP suggest the presence of inflammatory stress and anorexia, whereas elevated levels of both may be more reflective of cardiovascular disease (CVD).
Serum transferrin is frequently used as a marker of protein-energy nutritional status in people without CKD. Compared to serum albumin, it has the advantage of a shorter half-life (about 8.5 days) and a smaller pool size, making it more responsive to nutritional deficits. However, because transferrin is also influenced by the presence of anemia, a comorbidity prevalent in people with CKD requiring maintenance dialysis, it is not recommended for nutritional assessment in patients with stage 4 or 5 CKD because its specificity for nutritional deficits is low in this population (3). Transferrin is a useful nutritional parameter in patients with higher levels of kidney function. In the MDRD study, transferrin and nutrient intake gradually declined as kidney function deteriorated (14).
Was this article helpful?