**3. Malnutrition in hemodialysis patients**

### **3.1 Albumin as a marker of nutritional status**

Albumin is the most abundant plasma protein with a half-life of about twenty-one days. It is a negative acute-phase protein that functions to maintain oncotic pressure and act as a transport protein (Carlson, 2004). Normal serum albumin concentration is between 3.5 and 5.2 g/dL, and this reference range is often utilized as a marker of nutritional status in healthy older populations (Carlson, 2004; Covinsky, Covinsky, Palmer, & Sehgal, 2002). The cutoff point for low serum albumin concentration (hypoalbuminemia) has been proposed to be even higher at about 3.9 g/dL in HP (Trivedi, Xiang, & Klein, 2009). However, it must be noted that conclusions derived from measures of serum albumin concentration are not always in agreement with the conclusions derived from clinical assessments regarding nutritional status (Covinsky, et al., 2002). Forty HP with CRP levels below 0.80 mg/dL were evaluated for malnutrition by measuring serum albumin and through the use of the Subjective Global Assessment (SGA) described by Detsky, McLaughlin, and Baker et al. (1987). The SGA was used to determine whether a patient was classified as well-nourished

Malnutrition, Inflammation and Reverse Epidemiology in Hemodialysis Patients 301

mortality rate among HP (Kalantar-Zadeh et al., 2010). Supplementation with 15 grams of liquid hydrolyzed collagen protein three times per week after each hemodialysis treatment in one crossover group increased serum albumin by month 3 of supplementation. However, this change was small (+0.03 g/dL) and was not sustained throughout the remaining 3 months of treatment (Moretti, et al., 2009). Conversely, two pilot studies have shown promising effects of nutritional supplementation on serum albumin. In one study, maintenance HP consumed eight ounces of egg whites (egg whites are low in phosphorus) once per day for six weeks (Taylor, et al., 2011). Mean serum albumin concentrations increased by 0.19 g/dL along with a fall in mean serum phosphorus of 0.94 mg/dL. In the other pilot study, four grams of oral amino acid supplementation three times daily increased mean serum albumin concentration by 0.50 g/dL after 3 months of treatment (Covinsky, et al., 2002). Also, inflammation was attenuated in the study group as demonstrated by a decrease in CRP levels. Based on these pilot studies, protein and amino acid supplementation may benefit HP, but more research including larger sample sizes with controlled trials is needed before a definite conclusion or treatment protocol can be

The number of HP patients in the United States is approximately 350,000 with an expectation of reaching 1.5 million by 2016. Most HP have a significant decline in quality of life with two-thirds dying within 5 years of dialysis initiation; a survival rate worse than most cancers (Kilpatrick et al., 2007). Cardiovascular disease (CVD) is a leading cause of death in HP with rates higher than the general population and risk primarily associated with elevated lipids, inflammation, malnutrition, hypertension, hyperhomocysteinemia, obesity, and insulin resistance (Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) Final Report, 2002). Most HP patients' elevated lipid profiles are associated with a higher incidence of CVD morbidity and mortality (Vaziri, 2009) and allcause mortality (Tsirpanlis, et al., 2009). Additionally, non-fatal CVD is 10-30 times higher in HP suggesting this population is more prone to heart disease (Nanayakkara & Gaillard, 2010). Though traditional treatment of CVD in HP has demonstrated promise, many large randomized trials in HP have not demonstrated a survival benefit from traditional treatment strategies (Nanayakkara & Gaillard, 2010). The challenge becomes understanding why traditional risk factors are less predictive and whether the progression of disease is so advanced in HP that risk factor management should be different in this patient population. It should be noted however, that traditional and non-traditional risk factors for CVD in HP do have some crossover, complicating our understanding of how risk factor management may assist in disease causation and association. Though this chapter focuses on HP, chronic kidney disease patients who are predialysis share some of the same counter-intuitive findings. Therefore it should be noted that progression of CVD begins well before HP begin dialysis and could be detected as early as stage I CKD. Finally, it has been suggested that plaque accumulation may occur differently in HP. In the general population arterial plaque is more associated with lipid accumulation, but with HP it is more likely to be associated with calcified plaques and increased arterial stenosis (Diepeveen, Wetzels, Bilo, van Tits, & Stalenhoef, 2008). Various research groups, including the National Cholesterol Education Program (NCEP) (Third Report of the National Cholesterol Education Program Expert Panel on Detection,

**4. Cardiovascular disease and hemodialysis patients** 

formulated.

or malnourished. Using 3.5 g/dL as the cutoff point for hypoalbuminemia resulted in a sensitivity of just 14.3% compared to the results of the SGA. Raising the hypoalbuminemia cutoff point to 4.1 g/dL increased the sensitivity of the measurement to 64%. Furthermore, the mean albumin concentrations for the well-nourished and malnourished groups were 4.3 g/dL and 4.0 g/dL, respectively, with considerable overlap between the two groups. These data suggest that serum albumin alone may not be a sensitive marker of malnutrition in the absence of inflammation in HP (Santos et al., 2003). Regardless, serum albumin is a very important marker of mortality risk in HP (Iseki, Kawazoe, & Fukiyama, 1993). Despite the lack of total agreement with clinical assessments of nutritional status, hypoalbuminemia has been demonstrated to be an independent risk factor for all-cause mortality in older persons especially when combined with measures of physical disability (Corti, Guralnik, Salive, & Sorkin, 1994). Additionally, hypoalbuminemia is associated with mortality in various disease populations including cardiovascular, cancer, and HP (Iseki, et al., 1993; Phillips, Shaper, & Whincup, 1989). A 1 g/dL reduction in serum albumin has been associated with a 47% greater risk of mortality in HP, with the serum albumin concentration in these particular HP being linked to inflammation more so than the presence of malnutrition (de Mutsert et al., 2009).

### **3.2 Regulation of serum albumin: malnutrition and inflammation**

Serum albumin concentration is controlled by the rate of its synthesis, fractional catabolic rate (FCR), and distribution between intra and extravascular compartments. These three variables controlling albumin concentration are heavily influenced by both nutritional status and inflammation (Kaysen, 2003). In healthy individuals and HP without inflammation that are malnourished, albumin levels usually stay within a normal range until the degree of starvation is preterminal (Kaysen, 2009). Renal disease is associated with anorexia and PEM due to the build-up of uremic toxins. Additionally, hemodialysis for the removal of these toxins is also associated with anorexia and PEM because of the resulting nausea and postdialysis fatigue (Bergstrom, 1996). PEM leads to a decreased rate of albumin synthesis. In normal individuals, the FCR of albumin and resting energy expenditure are also downregulated in order to compensate for its decreased synthesis during periods of PEM. For HP that are in an inflammatory state, normal down-regulation of FCR is blunted leading to an imbalance between albumin synthesis and catabolism that result in hypoalbuminemia (Kaysen, 2009). Even in the absence of malnutrition, positive acute phase proteins that result from the production of pro-inflammatory cytokines are associated with decreased albumin synthesis. Additionally, inflammation leads to a greater than normal albumin FCR for a given serum albumin level (Kaysen, 2003). On top of the challenges presented by malnutrition and inflammation in HP, amino acid loss from hemodialysis itself may contribute further to nitrogen restriction and hypoalbuminemia (Kaysen, 2009).

#### **3.3 Nutritional supplementation and hemodialysis patients**

Recently, studies have investigated the effects of protein and amino acid supplements on serum albumin levels in HP (Bolasco, Caria, Cupisti, Secci, & Saverio Dioguardi, 2011; Moretti, Johnson, & Keeling-Hathaway, 2009; Taylor et al., 2011). When selecting an appropriate nutritional supplement for HP, phosphorus levels must be taken into consideration as it has been demonstrated that high-protein intake with concurrent lowphosphorus ingestion and normal serum phosphorus levels is associated with the lowest

or malnourished. Using 3.5 g/dL as the cutoff point for hypoalbuminemia resulted in a sensitivity of just 14.3% compared to the results of the SGA. Raising the hypoalbuminemia cutoff point to 4.1 g/dL increased the sensitivity of the measurement to 64%. Furthermore, the mean albumin concentrations for the well-nourished and malnourished groups were 4.3 g/dL and 4.0 g/dL, respectively, with considerable overlap between the two groups. These data suggest that serum albumin alone may not be a sensitive marker of malnutrition in the absence of inflammation in HP (Santos et al., 2003). Regardless, serum albumin is a very important marker of mortality risk in HP (Iseki, Kawazoe, & Fukiyama, 1993). Despite the lack of total agreement with clinical assessments of nutritional status, hypoalbuminemia has been demonstrated to be an independent risk factor for all-cause mortality in older persons especially when combined with measures of physical disability (Corti, Guralnik, Salive, & Sorkin, 1994). Additionally, hypoalbuminemia is associated with mortality in various disease populations including cardiovascular, cancer, and HP (Iseki, et al., 1993; Phillips, Shaper, & Whincup, 1989). A 1 g/dL reduction in serum albumin has been associated with a 47% greater risk of mortality in HP, with the serum albumin concentration in these particular HP being linked to inflammation more so than the presence of malnutrition (de

Serum albumin concentration is controlled by the rate of its synthesis, fractional catabolic rate (FCR), and distribution between intra and extravascular compartments. These three variables controlling albumin concentration are heavily influenced by both nutritional status and inflammation (Kaysen, 2003). In healthy individuals and HP without inflammation that are malnourished, albumin levels usually stay within a normal range until the degree of starvation is preterminal (Kaysen, 2009). Renal disease is associated with anorexia and PEM due to the build-up of uremic toxins. Additionally, hemodialysis for the removal of these toxins is also associated with anorexia and PEM because of the resulting nausea and postdialysis fatigue (Bergstrom, 1996). PEM leads to a decreased rate of albumin synthesis. In normal individuals, the FCR of albumin and resting energy expenditure are also downregulated in order to compensate for its decreased synthesis during periods of PEM. For HP that are in an inflammatory state, normal down-regulation of FCR is blunted leading to an imbalance between albumin synthesis and catabolism that result in hypoalbuminemia (Kaysen, 2009). Even in the absence of malnutrition, positive acute phase proteins that result from the production of pro-inflammatory cytokines are associated with decreased albumin synthesis. Additionally, inflammation leads to a greater than normal albumin FCR for a given serum albumin level (Kaysen, 2003). On top of the challenges presented by malnutrition and inflammation in HP, amino acid loss from hemodialysis itself may

**3.2 Regulation of serum albumin: malnutrition and inflammation** 

contribute further to nitrogen restriction and hypoalbuminemia (Kaysen, 2009).

Recently, studies have investigated the effects of protein and amino acid supplements on serum albumin levels in HP (Bolasco, Caria, Cupisti, Secci, & Saverio Dioguardi, 2011; Moretti, Johnson, & Keeling-Hathaway, 2009; Taylor et al., 2011). When selecting an appropriate nutritional supplement for HP, phosphorus levels must be taken into consideration as it has been demonstrated that high-protein intake with concurrent lowphosphorus ingestion and normal serum phosphorus levels is associated with the lowest

**3.3 Nutritional supplementation and hemodialysis patients** 

Mutsert et al., 2009).

mortality rate among HP (Kalantar-Zadeh et al., 2010). Supplementation with 15 grams of liquid hydrolyzed collagen protein three times per week after each hemodialysis treatment in one crossover group increased serum albumin by month 3 of supplementation. However, this change was small (+0.03 g/dL) and was not sustained throughout the remaining 3 months of treatment (Moretti, et al., 2009). Conversely, two pilot studies have shown promising effects of nutritional supplementation on serum albumin. In one study, maintenance HP consumed eight ounces of egg whites (egg whites are low in phosphorus) once per day for six weeks (Taylor, et al., 2011). Mean serum albumin concentrations increased by 0.19 g/dL along with a fall in mean serum phosphorus of 0.94 mg/dL. In the other pilot study, four grams of oral amino acid supplementation three times daily increased mean serum albumin concentration by 0.50 g/dL after 3 months of treatment (Covinsky, et al., 2002). Also, inflammation was attenuated in the study group as demonstrated by a decrease in CRP levels. Based on these pilot studies, protein and amino acid supplementation may benefit HP, but more research including larger sample sizes with controlled trials is needed before a definite conclusion or treatment protocol can be formulated.
