- Email: [email protected]
Serum Concentration of Amino Acids Versus Nutritional Status in Hemodialysis Patients
Serum Concentration of Amino Acids Versus Nutritional Status in Hemodialysis Patients lizien, MD, PhD,† Sylwia Małgorzewicz, MD, PhD,* Alicja D˛ebska-S Bolesław Rutkowski, MD, PhD,† and Wiesława Łysiak-Szydłowska, PhD* Objective: The aim of this study was to evaluate the correlation between the serum concentration of amino acids (AAs) and nutritional status in hemodialysis (HD) patients. Methods: This study was performed in 22 HD patients dialyzed for 10 to 288 months, and in a control group of 20 healthy volunteers. Nutritional status was determined by the subjective global assessment method and by measuring albumin concentration. Body composition was determined using the parameters of body mass index, and the percentage of body fat and lean body mass (as measured by the near-infrared method). We measured C-reactive protein (CRP) as a marker of inflammatory status. Serum concentrations of 20 AAs were measured by precolumn orthophtalaldehyde derivatization, applying high-performance liquid chromatography (Hitachi-Merck HPLC, Tokyo, Japan) equipped with a C-18 reversed-phase column and a methanol/acetate buffer gradient. Results: Thirteen of 22 (59%) patients were of good nutritional status, and 9/22 (41%) were malnourished, including 1 person with severe malnutrition. In dialyzed patients compared with control subjects, a decreased concentration of essential and nonessential AAs was observed (P , .05). Concentrations of the majority of studied AAs (16 out of 20) were lower in patients dialyzed for a period .2 years, compared with patients dialyzed for a shorter time. The ratio of branched-chain amino acids (BCAAs) to aromatic AAs was lower in the dialyzed group compared with control subjects. This ratio was also lower in patients dialyzed longer compared with patients dialyzed for ,2 years. No correlation between the concentration of some AAs and CRP level or dialysis adequacy was observed. In the malnourished group, an insignificantly lower concentration of some essential AAs (lysine, leucine, isoleucine, valine, and threonine), and a significantly higher (P 5 .04) concentration of CRP, were observed. Conclusion: Despite quite good nutritional status, dialyzed patients present abnormalities in their AA profiles. Moreover, a significant decrease of BCAA concentration is related to calorie-protein malnutrition, inflammation, and a long period of hemodialysis. Ó 2008 by the National Kidney Foundation, Inc.
M
ANY STUDIES INDICATE that patients treated by hemodialysis very often suffer from protein-calorie malnutrition.1–3 The main reasons of this include lack of appetite, inadequate diet, high protein catabolism, and a loss of amino acids (AAs) during dialysis. Losses of about 2 to 3 g of protein and 4 to 8 g of AAs are observed in patients during a standard hemodialysis session.
*Department of Clinical Nutrition and Diagnostic Laboratory, Medical University of Gdansk, Gdansk, Poland. †Department of Nephrology, Transplantation, and Internal Medicine, Medical University of Gdansk, Gdansk, Poland. Address reprint requests to Sylwia Małgorzewicz, MD, PhD, Department of Clinical Nutrition, Medical University of Gdansk, D˛ebinki Str. 7, 80-211 Gdansk, Poland. E-mail: sylwia@
tetra.pl 2008 by the National Kidney Foundation, Inc. 1051-2276/08/1802-0009$34.00/0 doi:10.1053/j.jrn.2007.11.011
Ó
Journal of Renal Nutrition, Vol 18, No 2 (March), 2008: pp 239–247
Among factors that increase the protein catabolism in dialysis patients are hormonal disturbances such as high levels of glucagon, leptine, cortisol, and parathormone, and a decrease in insulin activity, a decrease of growth hormone or testosterone levels, and acidosis.4–6 A strong correlation was observed between calorie-protein malnutrition and increased morbidity and mortality.7 The syndrome of malnutrition, inflammation, and atherosclerosis (MIA) described by Stenvinkel et al,7 which is often recognized in dialysis patients, suggests the role of inflammatory processes as predictors of mortality in this population. Malnutrition and the acute-phase response may contribute to the excessive atherosclerotic cardiovascular mortality in this group of patients.7–9 Variations in serum AA patterns in relation to malnutrition and metabolic disturbances are often 239
240
MAŁGORZEWICZ ET AL
observed in dialysis patients.10,11 A recent study indicated that the concentration of essential and nonessential AAs is a good indicator of protein metabolism abnormalities.12 In dialysis patients, both the plasma essential AA concentration and the ratio of essential AAs to nonessential AAs are decreased. Characteristic of these patients is a serum decrease of tryptophan (TRP), tyrosine (TYR), histidine (HIS), and valine (VAL), as well as an increase of sulfur AAs. A decrease in branched-chain amino acids (BCAAs) such as VAL, leucine (LEU), and isoleucine (ILE) can be attributed to malnutrition.12,13 The aim of the study was to measure the serum concentration of AAs in relation to nutritional status and inflammation in hemodialysis patients.
Patients and Methods The study was performed in 22 maintenance hemodialysis patients dialyzed for 10 to 288 months and in a control group which consisted of 20 healthy volunteers. The characteristics of the study groups are presented in Table 1. Patients were recruited from the Hemodialysis Unit of the Department of Nephrology, Transplantation, and Internal Medicine, Medical University of Gdansk (Gdansk, Poland). All patients were clinically stable, with no signs of infection. They were dialyzed for 4 to 5 hours, three times a week, with polysulfone membrane dialyzers (Fresenius Medical Care AG, Bad Homburg, Germany). Nutritional status was estimated using the subjective global assessment (SGA) method, and by measuring the serum concentration of albumin (bromcresol purple method). Results of the SGA were scored as: A (a score of 6 to 7), good nutrition; B (a score of 3 to 5), mild malnutrition; and C (a score of 1 to 2), severe malnutrition.5,14 Body composition was determined using body mass index (BMI) and the percentages of body fat (%F) and lean body mass (LBM), as measured by the near-infrared method (NIR), using Futrex 2000A (Futrek Inc., Gaithersburg, MD). Serum C-reactive protein (CRP) was measured as a marker of inflammatory status by an immunoturbidimetric method (Roche test, Roche Diagnostics GmbH Mannheim Roche Diagnostics Corp., Indianapolis, IN). Serum concentrations of 20 AAs were measured by precolumn orthophtalaldehyde derivatization, using high-performance
Table 1. Basic Characteristics of Hemodialysis Patients and Control Group
Parameters Age (y) Male/female Body mass index (kg/m2) Kt/V (per dialysis session) Length of dialysis treatment (mo)
Hemodialysis Patients (n 5 22)
Control Group (n 5 20)
56.3 6 12.8 12/10 25.4 6 4.2
48.7 6 19.3 5/15 22.9 6 1.8
1.47 6 0.14 92.3 6 93.6
Data are expressed as means 6 SD.
liquid chromatography (HPLC; Hitachi-Merck equipped with a C-18 reversed-phase column and a methanol/acetate buffer gradient, Hitachi-Merck HPLC, Tokyo, Japan).15 Blood samples were taken from fasted overnight patients before hemodialysis. The adequacy of dialysis treatment was estimated by Kt/V. The mean value (6SD) of Kt/V equaled 1.47 6 0.14, and was considered to be within the reference range.4
Statistical Analysis Data are expressed as means 6 standard deviation (SD). Significant differences were defined at P ,.05. The correlation and significance were evaluated using nonparametric statistics (with Statistica, version 7.0, 2005, StatSoft, Krakow, Poland).
Results Nutritional Status Versus Inflammation and AA Concentration According to SGA, the following grades of nutrition were observed in the study group of dialysis patients: Grade A (6 to 7) (good nutritional status), 13/22 (59.0%); Grade B (3 to 5) (mild malnutrition), 8/22 (36.5%); and Grade C (1 to 2) (severe malnutrition), 1/22 (4.5%). The BMI of study patients was within the proper range (Table 1). The anthropometric data showed a significant increase of %F (P ,.05; Table 2), and an insignificant decrease of LBM in dialyzed patients compared with control subjects. The
241
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS
significantly decreased LBM was observed in malnourished, dialyzed patients (Table 3). A correlation between LBM and SGA score (r 5 0.5; P ,.05) was observed. Dialyzed patients also showed a significantly lower level of albumin compared with control subjects (Table 2). The highest albumin level was observed in well-nourished patients (group A), and this level decreased with the decline in nutritional status (groups B and C) (Table 4). A correlation between SGA parametric score and albumin concentration (r 5 0.7; P , .05) was observed. Almost 8 of 9 (90%) of malnourished patients presented both an elevated CRP level (31.1 6 31.5 mg/L) and a decreased albumin concentration (3.8 6 0.2 g/L). All of these malnourished patients suffered from symptomatic coronary
heart disease. Therefore, in these patients, MIA syndrome was recognized. Only 1 of 9 (11%) malnourished patients at the CRP level was at a reference range (,5 mg/dL). No significant statistical correlation between AA concentrations and adequacy of dialysis was observed. Serum CRP levels were negatively correlated (Spearman R 5 –0.2) with concentrations of lysine (LYS), methionine (MET), arginine (ARG), threonine (THR), and phenylalanine (PHE). The serum albumin concentration was positively but not significantly correlated with the sum of the concentrations of all AAs (Spearman R 5 0.2). The findings of AA concentrations in HD and control groups are presented in Table 2. The decrease of both essential AAs and nonessential AAs was observed in dialyzed patients compared with control subjects. The ratio of
Table 2. Nutritional Status Parameters, C-Reactive Protein, and Amino Acid Concentrations in Hemodialysis Patients and Control Group Parameters Percentage of body fat Lean body mass (kg) Albumin (g/L) C-reactive protein (mg/L) Essential amino acids (mmol/L) Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Valine Tryptophan Total essential amino acids Semi-essential amino acids (mmol/L) Tyrosine Nonessential amino acids (mmol/L) Aminobutyrate Alanine Arginine Asparagine Aspartic acid Glutamine Glutamic acid Glycine Ornithine Serine Nonessential/essential ratio Total nonessential amino acids Total amino acids Data are expressed as means 6 SD. *P , .05, hemodialysis patients vs control subjects.
Hemodialysis Patients (n 5 25)
Control Group (n 5 23)
28.4 6 6.5 47.6 6 14.2 4.16 0.4 16.2 6 23.3
17.5 6 8.2* 54.1 6 11.7 4.6 6 0.1*
40.2 6 20.9 52.6 6 19.7 62.5 6 20.1 102.4 6 53.2 15.5 6 14.0 51.5 6 23.2 65.8 6 38.2 143.6 6 45.3 11.7 6 6.0 545.8 6 240.6
35.7 6 13.9 55.8 6 16.7 84.5 6 28.7* 144.9 6 59.9* 28.7 6 12.1* 39.0 610.7 115.7 6 59.3* 182.7 6 57.8* 35.3 6 16.3* 722.3 6 275.4*
44.1 6 19.1
66.4 6 25.8*
15.4 6 10.7 195.1 6 105.2 79.0 6 43.6 35.9 6 13.2 13.8 6 9.3 454.5 6 162.5 62.6 6 28.8 434.2 6 151.7 85.4 6 34.6 63.0 6 23.9 2.6 1,438.9 6 401.5 2,028.8 6 661.2
21.1 6 10.3 260.0 6 113.3 94.1 6 30.4 46.0 6 4.6* 8.2 6 6.5* 645.2 6 202.4* 32.2 6 16.7* 400.8 6 193.6 69.5 6 26.4 101.2 6 39.7* 2.3 1,678.3 6 653.9* 2,467.0 6 955.1*
242
MAŁGORZEWICZ ET AL
Table 3. Comparison Between Well-Nourished Hemodialysis Patients and Patients Presenting With Malnutrition-Inflammation-Atherosclerosis Syndrome Parameters
Well-Nourished Patients (n 5 15)
MIA Patients (n 5 8)
54.5 6 13.3 25.4 6 4.0 1.45 6 0.1 92.0 6 99.1 7.7 6 11.4 53.2 6 9.7 28.0 6 6.7 4.3 6 0.4
59.5 6 12.0 25.6 6 5.5 1.50 6 0.1 92.8 6 89.7 31.1 6 31.5* 38.0 6 16.3* 29.1 6 6.4 3.8 6 0.2*
44.7 6 30.3 53.8 6 16.5 65.2 6 15.3 107.9 6 57.1 52.5 6 24.9 69.8 6 44.0 149.3 6 42.0 12.1 6 7.0 15.2 615.8 42.4 6 20.9 15.4 6 12.5 192.7 6 76.0 87.7 6 49.3 35.6 6 13.6 14.2 6 6.4 443.1 6 182.0 62.4 6 22.4 417.5 6 146.7 83.7 6 38.8
62.7 6 28.9 50.6 6 25.6 58.0 6 27.3 92.7 6 47.8 49.6 6 21.4 58.7 6 26.3 133.7 6 51.9 11.0 6 3.9 16.1 6 11.1 47.1 616.5 15.5 6 3.2 199.3 6 149.8 64.0 6 27.8 36.5 6 13.3 13.2 6 13.4 474.6 6 130.5 63.0 6 39.4 463.5 6 165.9 88.3 6 29.7
Age (y) Body mass index (kg/m2) Kt/V (per dialysis session) Hemodialysis duration (mo) C-reactive protein (mg/L) Lean body mass (kg) Percentage of body fat Albumin (g/L) Amino acids (mmol/L) Histidine Isoleucine Leucine Lysine Phenylalanine Threonine Valine Tryptophan Methionine Tyrosine Aminobutyrate Alanine Arginine Aspartic acid Asparagine Glutamine Glutamic acid Glycine Ornithine
Data are expressed as means 6 SD. MIA, malnutrition-inflammation-atherosclerosis syndrome. *P , .05, MIA vs well-nourished patients.
nonessential AAs to essential AAs was also lower in these patients compared with the control group (2.6 vs 2.3, respectively). In the serum of dialyzed patients, concentrations of such AAs as LEU, LYS, MET, THR, TRP, TYR, VAL aspartic acid (ASN), glutamine (GLU), and serine (SER) were significantly lower compared with the control group (P ,.05). Serum concentrations of nonessential AAs such as PHE, glutamic acid (GLN), and asparagine (ASP) were
significantly higher in the HD group than in control subjects. The relationship between the study parameters and the nutritional status and inflammation of dialyzed patients is presented in Table 3. In the group of malnourished patients with inflammatory status (group MIA), the tendency to diminish some BCAAs (LEU, LYS, and VAL), and also such AAs as ARG and THR, was observed when compared with the well-nourished patients (group A).
Table 4. Comparison Between Subjective Global Assessment Score and Serum Albumin Level Subjective Global Assessment 7-points score
A 7 (n 5 10) 6 (n 5 5)
B 5 (n 5 6) 4 (n 5 2)
C 3 (n 5 1)
Albumin (g/L)
4.4 6 0.2*
3.8 6 0.2*
3.2 6 0.0*
Data are expressed as means 6 SD. *P , .05, A vs B and B vs C.
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS
Nutritional Status in Relation to Duration of Dialysis Treatment According to duration of dialysis treatment, patients were divided into two groups (#2 years and .2 years). Patients treated for .2 years presented a worse nutritional status compared with patients treated for a shorter period of time. Seven of 14 patients (50%) in the group treated for .2 years, and 2/6 patients (33%) in the group dialyzed for #2 years, were malnourished. A lower content of LBM was observed in the group dialyzed for .2 years (Table 5). In patients dialyzed for .2 years, compared with patients dialyzed for a shorter period, concentrations of the majority of studied AAs (15 of 20) were lower, whereas concentrations of LEU, ARG, SER, ASP, and ASN were significantly lower (P , .05; Table 5). Figure 1 presents the negative correlation between some particular BCAA concentrations and duration of dialysis treatment. The correlations between LEU and ILE and time of dialysis treatment are Spearman R 5 –0.3 and –0.4, respectively. The data presented in Figure 1 show a very weak correlation between VAL and time of dialysis treatment (Spearman R 5 –0.07). Ratio of BCAAs to Aromatic AAs The ratio of BCAAs to aromatic AAs was calculated as a Fisher quotient (FQ 5 ILE 1 LEU 1 VAL/PHE 1 TYR). The Fisher quotient is an indicator of increased protein catabolism, and is usually estimated in patients with hepatic failure.16 In the present study, the FQ was lower (2.7) in dialyzed patients compared with control subjects (3.0). The same finding was observed in patients dialyzed for .2 years compared with those treated for a shorter time, and in well-nourished patients vs malnourished patients (Table 6).
Discussion Regardless of advances in the field of renal-replacement therapy, the problem of malnutrition is persists among maintenance hemodialysis patients. Malnutrition appears in 50% to 60% of this group of patients.2 The estimation by SGA of nutritional status in our study indicated mild malnutrition in 36.5% of patients (9/22), and severe malnutrition in only 4.5% of patients (1/22). Malnutrition was observed more frequently in
243
patients dialyzed for .2 years compared with those dialyzed for a shorter time. The decrease in LBM was characteristic of malnourished patients and patients with a longer history of dialysis. Previous studies indicated an increase of protein breakdown and a decrease of protein synthesis during HD treatment.17 Also, in our study, a low content of LBM was associated with a decreased plasma concentration of BCAA. The positive correlation between plasma concentration of whole BCAAs and LBM is presented in Figure 2. A low content of essential AAs and a decreased ratio of essential AAs to nonessential AAs are recognized as biochemical signs of malnutrition in dialysis patients.11,18 In the study group of HD patients, compared with control subjects, lower serum concentrations of essential AAs, lower levels of nonessential AAs, and lower levels of total AAs were observed. The ratio of essential AAs to nonessential AAs in dialysis patients was also decreased. These results confirm that there was a risk of calorie-protein malnutrition in the study group of dialysis patients. Our results also indicate that malnutrition and AA disturbances occur more often in patients with a longer history of dialysis treatment (.2 years). Arginine is the precursor for nitric oxide synthesis. A decreased availability of ARG in patients with chronic renal failure is attributable to perturbed renal biosynthesis of this AA. A reduction of plasma ARG and nitric oxide production are associated with increased cardiovascular mortality in dialysis patients.19 Rysz et al found that the concentration of ARG was significantly higher in uremic serum, whereas concentrations of such AAs as SER, THR, VAL, and alanine (ALA) were significantly decreased.20 In our group of dialysis patients, we observed decreased concentrations of ASN, GLU, LEU, LYS, MET, SER, THR, TRP, TYR, VAL, and ARG compared with control subjects (79.0 6 43.6 mmol/L vs 94.1 6 30.4 mmol/L, respectively; Table 2). In peritoneal dialysis patients, Brunini et al observed low plasma levels of ARG and an activation of the L-isomer of Arginine transport into erythrocytes.21 The various levels of ARG concentrations in the above studies may be related to the differing metabolic status of study patients. The catabolic state may cause some alterations in ARG metabolism.22 In our study, the ratio of BCAAs to aromatic AAs (FQ) was calculated. This ratio is usually
244
MAŁGORZEWICZ ET AL
Table 5. Comparison Between Hemodialysis Patients in Relation to Duration of Dialysis Treatment (#2 Years Versus .2 Years) and Control Subjects Parameters Age (y) Body mass index (kg/m2) Kt/V (per dialysis session) C-reactive protein (mg/L) Lean body mass (kg) Percentage of body fat Albumin (g/L) Essential amino acids (mmol/L) Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Valine Tryptophan Semi-essential amino acids (mmol/L) Tyrosine Nonessential amino acids (mmol/L) Aminobutyrate Alanine Arginine Aspartic acid Asparagine Glutamine Glutamic acid Glycine Ornithine Serine
#2 Years (n 5 8) 60.3 6 16.2a,b 24.7 6 5.6 b 1.4 6 0.1 16.5 6 29.4 52.1 6 10.7 27.3 6 6.9b 4.1 6 0.4 b
.2 Years (n 5 14)
Control Subjects (n 5 20)
54.0 6 10.3b 25.9 6 3.8 b 1.5 6 0.1 16.1 6 20.3 45.1 6 15.7 29.1 6 6.4b 4.1 6 0.3 b
54.0 6 9.5 19.2 6 6.8 4.5 6 0.1
28.0 6 9.8a 62.3 6 18.1 76.0 6 16.0a 123.2 6 60.8 14.6 6 18.4 52.6 6 21.3 61.7 6 47.5 158.5 6 41.7 14.1 6 7.8
48.0 6 22.6 47.1 619.0b 54.9 618.5 b 90.5 6 46.6 16.0 6 11.6 50.8 6 24.9 68.1 6 33.5 135.2 6 46.5 10.3 6 4.4
35.7 6 13.9 55.8 616.7 84.5 6 28.7 144.9 6 59.9 28.7 612.1 39.0 6 10.7 115.7 6 59.3 182.7 6 57.8 35.3 6 16.3
48.8 6 24.1
41.4 6 16.1b
66.4 6 25.8
11.1 6 4.8b 239.7 6 73.0 106.1 6 59.5a 45.6 6 13.3a 19.2 6 12.2 495.0 6 228.6 72.6 6 36.4 461.5 6 101.1 91.3 6 41.2 79.0 6 25.2a
17.9 6 12.4b 169.7 6 1,114.4b 63.6 6 21.4b 30.4 6 9.7b 10.7 6 5.5b 431.5 6 113.9 56.9 6 23.0 418.6 6 175.9 82.0 6 31.3 53.9 6 18.3b
48.7 6 19.3 22.9 6 1.8
21.1 6 10.3 260.0 6 113.3 94.1 6 30.4 46.0 614.6 8.2 6 6.5 645.2 6 202.4 32.2 6 16.7 400.8 6 193.6 69.5 6 26.4 101.2 6 39.7
Data are expressed as means 6 SD. a P , .05, #2 years vs .2 years. b P , .05, #2 years or .2 years vs control subjects.
applied in the evaluation of AA metabolic disturbances in patients with hepatic failure.16 In patients with malnutrition and in those with a longer history of dialysis treatment, we found a high level of aromatic AAs in comparison to BCAAs. In patients dialyzed #2 years, the calculated FQ was similar to that in the control group (Table 5). Nonetheless, in those dialyzed for #2 years compared with control subjects, there were such AA abnormalities as significantly (P , .05) decreased concentrations of SER, TRP, LEU, LYS, VAL, ARG, and ASN (Table 4). In agreement with previous studies,11,13 our results indicate that AA disturbances may appear before dialysis treatment. These alterations accelerate in the course of maintenance HD. Application of
the FQ may be helpful in estimating BCAA deficiency as well as reaching a decision about which AA should be supplemented. Some authors suggest that AA disturbances can aggravate malnutrition via intensification of a lack of appetite.23–25 High levels of aromatic AAs, decreased levels of BCAAs, and symptoms of inflammation (high levels of CRP, interleukin-1, and tumor necrosis factor-a) may cause a lack of appetite and lead to malnutrition. Previous studies suggest that the level of TRP is correlated with anorexia in uremic patients. Tryptophan is the substrate for serotonin synthesis. High brain serotonin concentrations and a lower serotonin/dopamine ratio cause anorexia.23 In contrast with Lindholm et al,18 who reported high
245
Figure 1. Correlation between valine (VAL), isoleucine (ILE), and leucine (LEU) concentrations and time of hemodialysis (HD) treatment.
The plasma concentration of VAL, ILE, LEU (umol/l)
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS 240 220 200 180 160 140 120 100 80 60 40 20 0 -50
0
50
100
150
200
250
The time of HD treatment (months)
300
VAL R Spearman=-0.07 ILE R Spearman=-0.4 LEU R Spearman=-0.3
During episodes of infection, inflammation, other types of catabolic stress, and malnutrition, glutamine (GLU) is considered a conditionally essential AA. Glutamine also enhances glutathione production, thereby improving antioxidant status. It was demonstrated that GLU-supplemented nutrition was associated with reduced morbidity and mortality of patients in stress and nonstress conditions.12 In our study, in dialyzed patients, the GLU level was lower than in control subjects (Table 2), and it was lower in patients dialyzed for .2 years compared with those dialyzed for a shorter time (Table 4). Glutamine deficiency may be a consequence of chronic inflammation in dialyzed patients. The AA concentrations may depend on the presence of malnutrition and chronic inflammation. The CRP concentration is often elevated in dialyzed patients, and it reflects chronic inflammation and is also recognized as a sensitive and
plasma levels of TRP in uremic patients, we showed that the concentration of TRP was lower in dialysis patients (11.7 6 6.0 mmol/L) compared with control subjects (35.3 6 16.3 mmol/L), and this concentration was similar in well-nourished and malnourished patients. The tendency toward diminishing TRP concentration was observed in patients dialyzed for .2 years. According to the classic definition, there are eight essential AAs (Table 2). Some AAs are essential under particular conditions, eg, in chronic renal failure, the concentration of TYR and its ratio to PHE are consistently low, owing to the reduced oxidation of TYR from phenylalanine (partial inhibition of phenylalanine hydroxylase).12 In our study, the level of TYR in dialyzed patients was significantly lower (44.1 6 19.1 mmol/L) compared with control subjects (70.3 6 22.2 mmol/L). The TYR/PHE ratio was 0.8 in dialyzed patients, compared with 1.7 in control subjects.
Table 6. Fisher Quotient Ratio of Branched-Chain Amino Acids to Aromatic Amino Acids in Hemodialysis Patients in Relation to Duration of Hemodialysis Treatment and Nutritional Status
Branched-chain amino acids (leucine 1 isoleucine 1 valine) Aromatic amino acids (phenylalanine 1 tyrosine) Fisher quotient Data are expressed as means.
Hemodialysis Patients
#2 Years
.2 Years
WellNourished
Malnourished
Control Subjects
258.7
296.8
237.2
268.3
242.3
323.0
95.6
101.4
92.2
94.9
96.7
105.4
2.7
2.9
2.5
2.8
2.5
3.0
246
MAŁGORZEWICZ ET AL
The plasma concentration of whole BCAA (umol/l)
450
400
350
300
250
200
150
100
0
10
20
30
40
50
LBM (kg)
independent marker of malnutrition.7,26 In our study, only 1/9 of malnourished patients had CRP within normal range (,5 mg/L); the rest of them showed an elevated CRP level in spite of their stable clinical condition. Malnourished patients with an elevated CRP concentration presented the most pronounced deficiency of serum BCAAs and a significantly low LBM (Table 3). In accordance with Suliman et al,13 the concentrations of some AAs (SER, THR, ARG, MET, and LYS) were negatively correlated with CRP level. However, this correlation was statistically insignificant. In addition, as far as AA concentration is concerned, there were no statistical differences between the group of patients with MIA syndrome and the group of patients without inflammation (Table 3). However, based on the observed tendency toward reduced concentrations of some AAs in the MIA syndrome group and on the findings of previous studies, the influence of chronic inflammation on AA serum concentration should be taken into consideration. In summary, such AA disturbances as a decreased concentration of essential AAs, a low concentration of GLU, and a high concentration of aromatic AAs occurred more often in the group of dialyzed patients with a longer history of dialysis treatment. The most pronounced correlation in Figure 1 is that between time of dialysis treatment and low ILE and LEU concentrations.
60
70
80
Figure 2. Correlation between plasma concentration of whole branchedchain amino acids (BCAA) and lean body mass (LBM) (Spearman R 5 0.4, P 5 .07).
These disturbances could be partly related the nutritional status and chronic inflammation. However, the decrease of serum concentration in the majority of AAs was observed in malnourished patients and well-nourished dialyzed patients. The results of our investigation suggest that, in spite of good nutritional status according to generally accepted measures (eg, serum albumin and SGA), the study group of dialysis patients presented biochemical signs of malnutrition. The lack of an evident correlation between AA concentrations and albumin, and also between AA concentrations and CRP, suggests a variety of reasons underlying MIA syndrome and malnutrition itself. The present results demonstrate that albumin may not be the only biochemical factor indicating grade of malnutrition, because patients with normal levels of albumin also showed deep AA disturbances. In view of the particularly decreased BCAA concentrations, a nutritional intervention is needed to improve the profile of AAs and support dialysis patients in optimizing their nutritional status, as reported elsewhere.27
Conclusions 1. Despite good nutritional status (as indicated by SGA and an acceptable level of serum albumin), dialysis patients present abnormalities in their AA profiles.
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS
2. A significant decrease in BCAA concentration is related to calorie-protein malnutrition, as well as a long history of maintenance hemodialysis. 3. In cases of acknowledged BCAA deficiency, it seems that supplementation of these AAs may improve both the AA profile and nutritional status in dialysis patients.
References 1. Bistrian BR, McCowen KC, Chan S: Protein-energy malnutrition in dialysis patients. Am J Kidney Dis 33:172-175, 1999 2. Kopple J: Therapeutic approaches to malnutrition in chronic dialysis patients: The different modalities of nutritional support. Am J Kidney Dis 33:180-185, 1999 3. Toigo G, Aparicio M, Attman PO, et al: Expert Working Group report on nutrition in adult patients with renal insufficiency. Clin Nutr 19(197-207):281-291, 2000 4. Rutkowski B (ed.): Dializoterapia. Gdansk: PZWL, 1997 5. Mitch WE, Klahr S: Handbook of Nutrition and the Kidney. Philadelphia: Lippincott Wiliams & Wilkins, 2002 6. Prado de Negreiros Nogueeira Maduro I, Elis NM, Nonino Borges CB: Total nitrogen and free amino acid losses and protein calorie malnutrition of hemodialysis patients: Do they really matter? Nephron Clin Pract 105:9-17, 2007 7. Stenvinkel P, Heimburger O, Lindholm B, et al: Are there two types of malnutrition in chronic renal failure? Evidence for relationships between malnutrition, inflammation and atherosclerosis (MIA syndrome). Nephrol Dial Transplant 15:953-960, 2000 8. Suliman ME, Qureshi AR, Stenvinkel P, et al: Inflammation contributes to low plasma amino acid concentrations in patients with chronic kidney disease. Am J Clin Nutr 82:342-349, 2005 9. Brunini TM, Moss MB, Siqueira MA, et al: Nitric oxide, malnutrition and chronic renal failure. Cardiovasc Hematol Agents Med Chem 5:155-161, 2007 10. Kawakami J, Suzuki Y, Sugino N: Evaluation of amino acid patterns in recipes for kidney disease patients. J Ren Nutr 13: 126-132, 2003 11. Kopple J: Abnormal amino acid and protein metabolism in uremia. Kidney Int 14:340-348, 1978 12. Sobotka L (ed.): Basics in Clinical Nutrition. Prague: Galen, 2004
247
13. Suliman ME, Anderstam B, Lindholm B, Bergstrom J: Total, free, and protein-bound sulphur amino acids in uraemic patients. Nephrol Dial Transplant 12:2332-2338, 1997 14. Enia G, Sicuso C, Alati G, Zoccali C: Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 8:1094-1099, 1993 15. Cooper C (ed.): Amino Acid Analysis Protocols. Totowa, New Jersey: Humana Press, 2000 16. Blonde-Cynober F, Aussel C, Cynober L: Abnormalities in branched-chain amino acid metabolism in cirrhosis: Influence of hormonal and nutritional factors and directions for future research. Clin Nutr 18:5-13, 1999 17. Raj DS, Oladipo A, Lim VS: Amino acid and protein kinetics in renal failure: An integrated approach. Semin Nephrol 26:158-166, 2006 18. Lindholm B, Alverstrand A, Furst P, Bergstrom J: Plasma and muscle amino acids during continuous ambulatory peritoneal dialysis. Kidney Int 35:1219-1226, 1989 19. Baylis C: Arginine, arginine analogs and nitric oxide production in chronic kidney disease. Nat Clin Pract Nephrol 2: 209-220, 2006 20. Rysz J, Guga P, Cia1kowska-Rysz A, et al: Blood serum and neutrophil amino acid concentrations during hemodialysis. Pol Merkuriusz Lek 19:769-773, 2005 21. Brunini TM, Roberts NB, Yaqoob MM, et al: Activation of L-arginine transport in undialysed chronic renal failure and continuous ambulatory peritoneal dialysis patients. Clin Exp Pharmacol Physiol 33:114-118, 2006 22. Reis PF, da Silva CD, Brunini TM: Plasma amino acid profile and L-arginine uptake in red blood cells from malnourished uremic patients. J Ren Nutr16325-331, 2006 23. Aquilera A, Codoceo R, Bajo MA, et al: Eating behaviour disorders in uraemia: A question of balance in appetite regulation. Semin Dial 17:44-50, 2004 24. Aquilera A, Sanchez-Tomero JA, Selgas R: Brain activation in uremic anorexia. J Ren Nutr 1:57-61, 2007 25. Bossola M, Tazzal A, Giungi S, Luciani G: Anorexia in hemodialysis patients: An update. Kidney Int 14:23-25, 2006 26. Nasri H: Serum C-reactive protein (CRP) in association with various nutritional parameters in maintenance hemodialysis patients. Bratisl Lek Listy 106:390-395, 2005 27. Cano NJ, Fouque D, Leverve XM: Application of branched-chain amino acids in human pathological states: Renal failure. J Nutr 136:299-307, 2006
M
ANY STUDIES INDICATE that patients treated by hemodialysis very often suffer from protein-calorie malnutrition.1–3 The main reasons of this include lack of appetite, inadequate diet, high protein catabolism, and a loss of amino acids (AAs) during dialysis. Losses of about 2 to 3 g of protein and 4 to 8 g of AAs are observed in patients during a standard hemodialysis session.
*Department of Clinical Nutrition and Diagnostic Laboratory, Medical University of Gdansk, Gdansk, Poland. †Department of Nephrology, Transplantation, and Internal Medicine, Medical University of Gdansk, Gdansk, Poland. Address reprint requests to Sylwia Małgorzewicz, MD, PhD, Department of Clinical Nutrition, Medical University of Gdansk, D˛ebinki Str. 7, 80-211 Gdansk, Poland. E-mail: sylwia@
tetra.pl 2008 by the National Kidney Foundation, Inc. 1051-2276/08/1802-0009$34.00/0 doi:10.1053/j.jrn.2007.11.011
Ó
Journal of Renal Nutrition, Vol 18, No 2 (March), 2008: pp 239–247
Among factors that increase the protein catabolism in dialysis patients are hormonal disturbances such as high levels of glucagon, leptine, cortisol, and parathormone, and a decrease in insulin activity, a decrease of growth hormone or testosterone levels, and acidosis.4–6 A strong correlation was observed between calorie-protein malnutrition and increased morbidity and mortality.7 The syndrome of malnutrition, inflammation, and atherosclerosis (MIA) described by Stenvinkel et al,7 which is often recognized in dialysis patients, suggests the role of inflammatory processes as predictors of mortality in this population. Malnutrition and the acute-phase response may contribute to the excessive atherosclerotic cardiovascular mortality in this group of patients.7–9 Variations in serum AA patterns in relation to malnutrition and metabolic disturbances are often 239
240
MAŁGORZEWICZ ET AL
observed in dialysis patients.10,11 A recent study indicated that the concentration of essential and nonessential AAs is a good indicator of protein metabolism abnormalities.12 In dialysis patients, both the plasma essential AA concentration and the ratio of essential AAs to nonessential AAs are decreased. Characteristic of these patients is a serum decrease of tryptophan (TRP), tyrosine (TYR), histidine (HIS), and valine (VAL), as well as an increase of sulfur AAs. A decrease in branched-chain amino acids (BCAAs) such as VAL, leucine (LEU), and isoleucine (ILE) can be attributed to malnutrition.12,13 The aim of the study was to measure the serum concentration of AAs in relation to nutritional status and inflammation in hemodialysis patients.
Patients and Methods The study was performed in 22 maintenance hemodialysis patients dialyzed for 10 to 288 months and in a control group which consisted of 20 healthy volunteers. The characteristics of the study groups are presented in Table 1. Patients were recruited from the Hemodialysis Unit of the Department of Nephrology, Transplantation, and Internal Medicine, Medical University of Gdansk (Gdansk, Poland). All patients were clinically stable, with no signs of infection. They were dialyzed for 4 to 5 hours, three times a week, with polysulfone membrane dialyzers (Fresenius Medical Care AG, Bad Homburg, Germany). Nutritional status was estimated using the subjective global assessment (SGA) method, and by measuring the serum concentration of albumin (bromcresol purple method). Results of the SGA were scored as: A (a score of 6 to 7), good nutrition; B (a score of 3 to 5), mild malnutrition; and C (a score of 1 to 2), severe malnutrition.5,14 Body composition was determined using body mass index (BMI) and the percentages of body fat (%F) and lean body mass (LBM), as measured by the near-infrared method (NIR), using Futrex 2000A (Futrek Inc., Gaithersburg, MD). Serum C-reactive protein (CRP) was measured as a marker of inflammatory status by an immunoturbidimetric method (Roche test, Roche Diagnostics GmbH Mannheim Roche Diagnostics Corp., Indianapolis, IN). Serum concentrations of 20 AAs were measured by precolumn orthophtalaldehyde derivatization, using high-performance
Table 1. Basic Characteristics of Hemodialysis Patients and Control Group
Parameters Age (y) Male/female Body mass index (kg/m2) Kt/V (per dialysis session) Length of dialysis treatment (mo)
Hemodialysis Patients (n 5 22)
Control Group (n 5 20)
56.3 6 12.8 12/10 25.4 6 4.2
48.7 6 19.3 5/15 22.9 6 1.8
1.47 6 0.14 92.3 6 93.6
Data are expressed as means 6 SD.
liquid chromatography (HPLC; Hitachi-Merck equipped with a C-18 reversed-phase column and a methanol/acetate buffer gradient, Hitachi-Merck HPLC, Tokyo, Japan).15 Blood samples were taken from fasted overnight patients before hemodialysis. The adequacy of dialysis treatment was estimated by Kt/V. The mean value (6SD) of Kt/V equaled 1.47 6 0.14, and was considered to be within the reference range.4
Statistical Analysis Data are expressed as means 6 standard deviation (SD). Significant differences were defined at P ,.05. The correlation and significance were evaluated using nonparametric statistics (with Statistica, version 7.0, 2005, StatSoft, Krakow, Poland).
Results Nutritional Status Versus Inflammation and AA Concentration According to SGA, the following grades of nutrition were observed in the study group of dialysis patients: Grade A (6 to 7) (good nutritional status), 13/22 (59.0%); Grade B (3 to 5) (mild malnutrition), 8/22 (36.5%); and Grade C (1 to 2) (severe malnutrition), 1/22 (4.5%). The BMI of study patients was within the proper range (Table 1). The anthropometric data showed a significant increase of %F (P ,.05; Table 2), and an insignificant decrease of LBM in dialyzed patients compared with control subjects. The
241
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS
significantly decreased LBM was observed in malnourished, dialyzed patients (Table 3). A correlation between LBM and SGA score (r 5 0.5; P ,.05) was observed. Dialyzed patients also showed a significantly lower level of albumin compared with control subjects (Table 2). The highest albumin level was observed in well-nourished patients (group A), and this level decreased with the decline in nutritional status (groups B and C) (Table 4). A correlation between SGA parametric score and albumin concentration (r 5 0.7; P , .05) was observed. Almost 8 of 9 (90%) of malnourished patients presented both an elevated CRP level (31.1 6 31.5 mg/L) and a decreased albumin concentration (3.8 6 0.2 g/L). All of these malnourished patients suffered from symptomatic coronary
heart disease. Therefore, in these patients, MIA syndrome was recognized. Only 1 of 9 (11%) malnourished patients at the CRP level was at a reference range (,5 mg/dL). No significant statistical correlation between AA concentrations and adequacy of dialysis was observed. Serum CRP levels were negatively correlated (Spearman R 5 –0.2) with concentrations of lysine (LYS), methionine (MET), arginine (ARG), threonine (THR), and phenylalanine (PHE). The serum albumin concentration was positively but not significantly correlated with the sum of the concentrations of all AAs (Spearman R 5 0.2). The findings of AA concentrations in HD and control groups are presented in Table 2. The decrease of both essential AAs and nonessential AAs was observed in dialyzed patients compared with control subjects. The ratio of
Table 2. Nutritional Status Parameters, C-Reactive Protein, and Amino Acid Concentrations in Hemodialysis Patients and Control Group Parameters Percentage of body fat Lean body mass (kg) Albumin (g/L) C-reactive protein (mg/L) Essential amino acids (mmol/L) Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Valine Tryptophan Total essential amino acids Semi-essential amino acids (mmol/L) Tyrosine Nonessential amino acids (mmol/L) Aminobutyrate Alanine Arginine Asparagine Aspartic acid Glutamine Glutamic acid Glycine Ornithine Serine Nonessential/essential ratio Total nonessential amino acids Total amino acids Data are expressed as means 6 SD. *P , .05, hemodialysis patients vs control subjects.
Hemodialysis Patients (n 5 25)
Control Group (n 5 23)
28.4 6 6.5 47.6 6 14.2 4.16 0.4 16.2 6 23.3
17.5 6 8.2* 54.1 6 11.7 4.6 6 0.1*
40.2 6 20.9 52.6 6 19.7 62.5 6 20.1 102.4 6 53.2 15.5 6 14.0 51.5 6 23.2 65.8 6 38.2 143.6 6 45.3 11.7 6 6.0 545.8 6 240.6
35.7 6 13.9 55.8 6 16.7 84.5 6 28.7* 144.9 6 59.9* 28.7 6 12.1* 39.0 610.7 115.7 6 59.3* 182.7 6 57.8* 35.3 6 16.3* 722.3 6 275.4*
44.1 6 19.1
66.4 6 25.8*
15.4 6 10.7 195.1 6 105.2 79.0 6 43.6 35.9 6 13.2 13.8 6 9.3 454.5 6 162.5 62.6 6 28.8 434.2 6 151.7 85.4 6 34.6 63.0 6 23.9 2.6 1,438.9 6 401.5 2,028.8 6 661.2
21.1 6 10.3 260.0 6 113.3 94.1 6 30.4 46.0 6 4.6* 8.2 6 6.5* 645.2 6 202.4* 32.2 6 16.7* 400.8 6 193.6 69.5 6 26.4 101.2 6 39.7* 2.3 1,678.3 6 653.9* 2,467.0 6 955.1*
242
MAŁGORZEWICZ ET AL
Table 3. Comparison Between Well-Nourished Hemodialysis Patients and Patients Presenting With Malnutrition-Inflammation-Atherosclerosis Syndrome Parameters
Well-Nourished Patients (n 5 15)
MIA Patients (n 5 8)
54.5 6 13.3 25.4 6 4.0 1.45 6 0.1 92.0 6 99.1 7.7 6 11.4 53.2 6 9.7 28.0 6 6.7 4.3 6 0.4
59.5 6 12.0 25.6 6 5.5 1.50 6 0.1 92.8 6 89.7 31.1 6 31.5* 38.0 6 16.3* 29.1 6 6.4 3.8 6 0.2*
44.7 6 30.3 53.8 6 16.5 65.2 6 15.3 107.9 6 57.1 52.5 6 24.9 69.8 6 44.0 149.3 6 42.0 12.1 6 7.0 15.2 615.8 42.4 6 20.9 15.4 6 12.5 192.7 6 76.0 87.7 6 49.3 35.6 6 13.6 14.2 6 6.4 443.1 6 182.0 62.4 6 22.4 417.5 6 146.7 83.7 6 38.8
62.7 6 28.9 50.6 6 25.6 58.0 6 27.3 92.7 6 47.8 49.6 6 21.4 58.7 6 26.3 133.7 6 51.9 11.0 6 3.9 16.1 6 11.1 47.1 616.5 15.5 6 3.2 199.3 6 149.8 64.0 6 27.8 36.5 6 13.3 13.2 6 13.4 474.6 6 130.5 63.0 6 39.4 463.5 6 165.9 88.3 6 29.7
Age (y) Body mass index (kg/m2) Kt/V (per dialysis session) Hemodialysis duration (mo) C-reactive protein (mg/L) Lean body mass (kg) Percentage of body fat Albumin (g/L) Amino acids (mmol/L) Histidine Isoleucine Leucine Lysine Phenylalanine Threonine Valine Tryptophan Methionine Tyrosine Aminobutyrate Alanine Arginine Aspartic acid Asparagine Glutamine Glutamic acid Glycine Ornithine
Data are expressed as means 6 SD. MIA, malnutrition-inflammation-atherosclerosis syndrome. *P , .05, MIA vs well-nourished patients.
nonessential AAs to essential AAs was also lower in these patients compared with the control group (2.6 vs 2.3, respectively). In the serum of dialyzed patients, concentrations of such AAs as LEU, LYS, MET, THR, TRP, TYR, VAL aspartic acid (ASN), glutamine (GLU), and serine (SER) were significantly lower compared with the control group (P ,.05). Serum concentrations of nonessential AAs such as PHE, glutamic acid (GLN), and asparagine (ASP) were
significantly higher in the HD group than in control subjects. The relationship between the study parameters and the nutritional status and inflammation of dialyzed patients is presented in Table 3. In the group of malnourished patients with inflammatory status (group MIA), the tendency to diminish some BCAAs (LEU, LYS, and VAL), and also such AAs as ARG and THR, was observed when compared with the well-nourished patients (group A).
Table 4. Comparison Between Subjective Global Assessment Score and Serum Albumin Level Subjective Global Assessment 7-points score
A 7 (n 5 10) 6 (n 5 5)
B 5 (n 5 6) 4 (n 5 2)
C 3 (n 5 1)
Albumin (g/L)
4.4 6 0.2*
3.8 6 0.2*
3.2 6 0.0*
Data are expressed as means 6 SD. *P , .05, A vs B and B vs C.
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS
Nutritional Status in Relation to Duration of Dialysis Treatment According to duration of dialysis treatment, patients were divided into two groups (#2 years and .2 years). Patients treated for .2 years presented a worse nutritional status compared with patients treated for a shorter period of time. Seven of 14 patients (50%) in the group treated for .2 years, and 2/6 patients (33%) in the group dialyzed for #2 years, were malnourished. A lower content of LBM was observed in the group dialyzed for .2 years (Table 5). In patients dialyzed for .2 years, compared with patients dialyzed for a shorter period, concentrations of the majority of studied AAs (15 of 20) were lower, whereas concentrations of LEU, ARG, SER, ASP, and ASN were significantly lower (P , .05; Table 5). Figure 1 presents the negative correlation between some particular BCAA concentrations and duration of dialysis treatment. The correlations between LEU and ILE and time of dialysis treatment are Spearman R 5 –0.3 and –0.4, respectively. The data presented in Figure 1 show a very weak correlation between VAL and time of dialysis treatment (Spearman R 5 –0.07). Ratio of BCAAs to Aromatic AAs The ratio of BCAAs to aromatic AAs was calculated as a Fisher quotient (FQ 5 ILE 1 LEU 1 VAL/PHE 1 TYR). The Fisher quotient is an indicator of increased protein catabolism, and is usually estimated in patients with hepatic failure.16 In the present study, the FQ was lower (2.7) in dialyzed patients compared with control subjects (3.0). The same finding was observed in patients dialyzed for .2 years compared with those treated for a shorter time, and in well-nourished patients vs malnourished patients (Table 6).
Discussion Regardless of advances in the field of renal-replacement therapy, the problem of malnutrition is persists among maintenance hemodialysis patients. Malnutrition appears in 50% to 60% of this group of patients.2 The estimation by SGA of nutritional status in our study indicated mild malnutrition in 36.5% of patients (9/22), and severe malnutrition in only 4.5% of patients (1/22). Malnutrition was observed more frequently in
243
patients dialyzed for .2 years compared with those dialyzed for a shorter time. The decrease in LBM was characteristic of malnourished patients and patients with a longer history of dialysis. Previous studies indicated an increase of protein breakdown and a decrease of protein synthesis during HD treatment.17 Also, in our study, a low content of LBM was associated with a decreased plasma concentration of BCAA. The positive correlation between plasma concentration of whole BCAAs and LBM is presented in Figure 2. A low content of essential AAs and a decreased ratio of essential AAs to nonessential AAs are recognized as biochemical signs of malnutrition in dialysis patients.11,18 In the study group of HD patients, compared with control subjects, lower serum concentrations of essential AAs, lower levels of nonessential AAs, and lower levels of total AAs were observed. The ratio of essential AAs to nonessential AAs in dialysis patients was also decreased. These results confirm that there was a risk of calorie-protein malnutrition in the study group of dialysis patients. Our results also indicate that malnutrition and AA disturbances occur more often in patients with a longer history of dialysis treatment (.2 years). Arginine is the precursor for nitric oxide synthesis. A decreased availability of ARG in patients with chronic renal failure is attributable to perturbed renal biosynthesis of this AA. A reduction of plasma ARG and nitric oxide production are associated with increased cardiovascular mortality in dialysis patients.19 Rysz et al found that the concentration of ARG was significantly higher in uremic serum, whereas concentrations of such AAs as SER, THR, VAL, and alanine (ALA) were significantly decreased.20 In our group of dialysis patients, we observed decreased concentrations of ASN, GLU, LEU, LYS, MET, SER, THR, TRP, TYR, VAL, and ARG compared with control subjects (79.0 6 43.6 mmol/L vs 94.1 6 30.4 mmol/L, respectively; Table 2). In peritoneal dialysis patients, Brunini et al observed low plasma levels of ARG and an activation of the L-isomer of Arginine transport into erythrocytes.21 The various levels of ARG concentrations in the above studies may be related to the differing metabolic status of study patients. The catabolic state may cause some alterations in ARG metabolism.22 In our study, the ratio of BCAAs to aromatic AAs (FQ) was calculated. This ratio is usually
244
MAŁGORZEWICZ ET AL
Table 5. Comparison Between Hemodialysis Patients in Relation to Duration of Dialysis Treatment (#2 Years Versus .2 Years) and Control Subjects Parameters Age (y) Body mass index (kg/m2) Kt/V (per dialysis session) C-reactive protein (mg/L) Lean body mass (kg) Percentage of body fat Albumin (g/L) Essential amino acids (mmol/L) Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Valine Tryptophan Semi-essential amino acids (mmol/L) Tyrosine Nonessential amino acids (mmol/L) Aminobutyrate Alanine Arginine Aspartic acid Asparagine Glutamine Glutamic acid Glycine Ornithine Serine
#2 Years (n 5 8) 60.3 6 16.2a,b 24.7 6 5.6 b 1.4 6 0.1 16.5 6 29.4 52.1 6 10.7 27.3 6 6.9b 4.1 6 0.4 b
.2 Years (n 5 14)
Control Subjects (n 5 20)
54.0 6 10.3b 25.9 6 3.8 b 1.5 6 0.1 16.1 6 20.3 45.1 6 15.7 29.1 6 6.4b 4.1 6 0.3 b
54.0 6 9.5 19.2 6 6.8 4.5 6 0.1
28.0 6 9.8a 62.3 6 18.1 76.0 6 16.0a 123.2 6 60.8 14.6 6 18.4 52.6 6 21.3 61.7 6 47.5 158.5 6 41.7 14.1 6 7.8
48.0 6 22.6 47.1 619.0b 54.9 618.5 b 90.5 6 46.6 16.0 6 11.6 50.8 6 24.9 68.1 6 33.5 135.2 6 46.5 10.3 6 4.4
35.7 6 13.9 55.8 616.7 84.5 6 28.7 144.9 6 59.9 28.7 612.1 39.0 6 10.7 115.7 6 59.3 182.7 6 57.8 35.3 6 16.3
48.8 6 24.1
41.4 6 16.1b
66.4 6 25.8
11.1 6 4.8b 239.7 6 73.0 106.1 6 59.5a 45.6 6 13.3a 19.2 6 12.2 495.0 6 228.6 72.6 6 36.4 461.5 6 101.1 91.3 6 41.2 79.0 6 25.2a
17.9 6 12.4b 169.7 6 1,114.4b 63.6 6 21.4b 30.4 6 9.7b 10.7 6 5.5b 431.5 6 113.9 56.9 6 23.0 418.6 6 175.9 82.0 6 31.3 53.9 6 18.3b
48.7 6 19.3 22.9 6 1.8
21.1 6 10.3 260.0 6 113.3 94.1 6 30.4 46.0 614.6 8.2 6 6.5 645.2 6 202.4 32.2 6 16.7 400.8 6 193.6 69.5 6 26.4 101.2 6 39.7
Data are expressed as means 6 SD. a P , .05, #2 years vs .2 years. b P , .05, #2 years or .2 years vs control subjects.
applied in the evaluation of AA metabolic disturbances in patients with hepatic failure.16 In patients with malnutrition and in those with a longer history of dialysis treatment, we found a high level of aromatic AAs in comparison to BCAAs. In patients dialyzed #2 years, the calculated FQ was similar to that in the control group (Table 5). Nonetheless, in those dialyzed for #2 years compared with control subjects, there were such AA abnormalities as significantly (P , .05) decreased concentrations of SER, TRP, LEU, LYS, VAL, ARG, and ASN (Table 4). In agreement with previous studies,11,13 our results indicate that AA disturbances may appear before dialysis treatment. These alterations accelerate in the course of maintenance HD. Application of
the FQ may be helpful in estimating BCAA deficiency as well as reaching a decision about which AA should be supplemented. Some authors suggest that AA disturbances can aggravate malnutrition via intensification of a lack of appetite.23–25 High levels of aromatic AAs, decreased levels of BCAAs, and symptoms of inflammation (high levels of CRP, interleukin-1, and tumor necrosis factor-a) may cause a lack of appetite and lead to malnutrition. Previous studies suggest that the level of TRP is correlated with anorexia in uremic patients. Tryptophan is the substrate for serotonin synthesis. High brain serotonin concentrations and a lower serotonin/dopamine ratio cause anorexia.23 In contrast with Lindholm et al,18 who reported high
245
Figure 1. Correlation between valine (VAL), isoleucine (ILE), and leucine (LEU) concentrations and time of hemodialysis (HD) treatment.
The plasma concentration of VAL, ILE, LEU (umol/l)
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS 240 220 200 180 160 140 120 100 80 60 40 20 0 -50
0
50
100
150
200
250
The time of HD treatment (months)
300
VAL R Spearman=-0.07 ILE R Spearman=-0.4 LEU R Spearman=-0.3
During episodes of infection, inflammation, other types of catabolic stress, and malnutrition, glutamine (GLU) is considered a conditionally essential AA. Glutamine also enhances glutathione production, thereby improving antioxidant status. It was demonstrated that GLU-supplemented nutrition was associated with reduced morbidity and mortality of patients in stress and nonstress conditions.12 In our study, in dialyzed patients, the GLU level was lower than in control subjects (Table 2), and it was lower in patients dialyzed for .2 years compared with those dialyzed for a shorter time (Table 4). Glutamine deficiency may be a consequence of chronic inflammation in dialyzed patients. The AA concentrations may depend on the presence of malnutrition and chronic inflammation. The CRP concentration is often elevated in dialyzed patients, and it reflects chronic inflammation and is also recognized as a sensitive and
plasma levels of TRP in uremic patients, we showed that the concentration of TRP was lower in dialysis patients (11.7 6 6.0 mmol/L) compared with control subjects (35.3 6 16.3 mmol/L), and this concentration was similar in well-nourished and malnourished patients. The tendency toward diminishing TRP concentration was observed in patients dialyzed for .2 years. According to the classic definition, there are eight essential AAs (Table 2). Some AAs are essential under particular conditions, eg, in chronic renal failure, the concentration of TYR and its ratio to PHE are consistently low, owing to the reduced oxidation of TYR from phenylalanine (partial inhibition of phenylalanine hydroxylase).12 In our study, the level of TYR in dialyzed patients was significantly lower (44.1 6 19.1 mmol/L) compared with control subjects (70.3 6 22.2 mmol/L). The TYR/PHE ratio was 0.8 in dialyzed patients, compared with 1.7 in control subjects.
Table 6. Fisher Quotient Ratio of Branched-Chain Amino Acids to Aromatic Amino Acids in Hemodialysis Patients in Relation to Duration of Hemodialysis Treatment and Nutritional Status
Branched-chain amino acids (leucine 1 isoleucine 1 valine) Aromatic amino acids (phenylalanine 1 tyrosine) Fisher quotient Data are expressed as means.
Hemodialysis Patients
#2 Years
.2 Years
WellNourished
Malnourished
Control Subjects
258.7
296.8
237.2
268.3
242.3
323.0
95.6
101.4
92.2
94.9
96.7
105.4
2.7
2.9
2.5
2.8
2.5
3.0
246
MAŁGORZEWICZ ET AL
The plasma concentration of whole BCAA (umol/l)
450
400
350
300
250
200
150
100
0
10
20
30
40
50
LBM (kg)
independent marker of malnutrition.7,26 In our study, only 1/9 of malnourished patients had CRP within normal range (,5 mg/L); the rest of them showed an elevated CRP level in spite of their stable clinical condition. Malnourished patients with an elevated CRP concentration presented the most pronounced deficiency of serum BCAAs and a significantly low LBM (Table 3). In accordance with Suliman et al,13 the concentrations of some AAs (SER, THR, ARG, MET, and LYS) were negatively correlated with CRP level. However, this correlation was statistically insignificant. In addition, as far as AA concentration is concerned, there were no statistical differences between the group of patients with MIA syndrome and the group of patients without inflammation (Table 3). However, based on the observed tendency toward reduced concentrations of some AAs in the MIA syndrome group and on the findings of previous studies, the influence of chronic inflammation on AA serum concentration should be taken into consideration. In summary, such AA disturbances as a decreased concentration of essential AAs, a low concentration of GLU, and a high concentration of aromatic AAs occurred more often in the group of dialyzed patients with a longer history of dialysis treatment. The most pronounced correlation in Figure 1 is that between time of dialysis treatment and low ILE and LEU concentrations.
60
70
80
Figure 2. Correlation between plasma concentration of whole branchedchain amino acids (BCAA) and lean body mass (LBM) (Spearman R 5 0.4, P 5 .07).
These disturbances could be partly related the nutritional status and chronic inflammation. However, the decrease of serum concentration in the majority of AAs was observed in malnourished patients and well-nourished dialyzed patients. The results of our investigation suggest that, in spite of good nutritional status according to generally accepted measures (eg, serum albumin and SGA), the study group of dialysis patients presented biochemical signs of malnutrition. The lack of an evident correlation between AA concentrations and albumin, and also between AA concentrations and CRP, suggests a variety of reasons underlying MIA syndrome and malnutrition itself. The present results demonstrate that albumin may not be the only biochemical factor indicating grade of malnutrition, because patients with normal levels of albumin also showed deep AA disturbances. In view of the particularly decreased BCAA concentrations, a nutritional intervention is needed to improve the profile of AAs and support dialysis patients in optimizing their nutritional status, as reported elsewhere.27
Conclusions 1. Despite good nutritional status (as indicated by SGA and an acceptable level of serum albumin), dialysis patients present abnormalities in their AA profiles.
AMINO ACIDS VS NUTRITIONAL STATUS IN HD PATIENTS
2. A significant decrease in BCAA concentration is related to calorie-protein malnutrition, as well as a long history of maintenance hemodialysis. 3. In cases of acknowledged BCAA deficiency, it seems that supplementation of these AAs may improve both the AA profile and nutritional status in dialysis patients.
References 1. Bistrian BR, McCowen KC, Chan S: Protein-energy malnutrition in dialysis patients. Am J Kidney Dis 33:172-175, 1999 2. Kopple J: Therapeutic approaches to malnutrition in chronic dialysis patients: The different modalities of nutritional support. Am J Kidney Dis 33:180-185, 1999 3. Toigo G, Aparicio M, Attman PO, et al: Expert Working Group report on nutrition in adult patients with renal insufficiency. Clin Nutr 19(197-207):281-291, 2000 4. Rutkowski B (ed.): Dializoterapia. Gdansk: PZWL, 1997 5. Mitch WE, Klahr S: Handbook of Nutrition and the Kidney. Philadelphia: Lippincott Wiliams & Wilkins, 2002 6. Prado de Negreiros Nogueeira Maduro I, Elis NM, Nonino Borges CB: Total nitrogen and free amino acid losses and protein calorie malnutrition of hemodialysis patients: Do they really matter? Nephron Clin Pract 105:9-17, 2007 7. Stenvinkel P, Heimburger O, Lindholm B, et al: Are there two types of malnutrition in chronic renal failure? Evidence for relationships between malnutrition, inflammation and atherosclerosis (MIA syndrome). Nephrol Dial Transplant 15:953-960, 2000 8. Suliman ME, Qureshi AR, Stenvinkel P, et al: Inflammation contributes to low plasma amino acid concentrations in patients with chronic kidney disease. Am J Clin Nutr 82:342-349, 2005 9. Brunini TM, Moss MB, Siqueira MA, et al: Nitric oxide, malnutrition and chronic renal failure. Cardiovasc Hematol Agents Med Chem 5:155-161, 2007 10. Kawakami J, Suzuki Y, Sugino N: Evaluation of amino acid patterns in recipes for kidney disease patients. J Ren Nutr 13: 126-132, 2003 11. Kopple J: Abnormal amino acid and protein metabolism in uremia. Kidney Int 14:340-348, 1978 12. Sobotka L (ed.): Basics in Clinical Nutrition. Prague: Galen, 2004
247
13. Suliman ME, Anderstam B, Lindholm B, Bergstrom J: Total, free, and protein-bound sulphur amino acids in uraemic patients. Nephrol Dial Transplant 12:2332-2338, 1997 14. Enia G, Sicuso C, Alati G, Zoccali C: Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 8:1094-1099, 1993 15. Cooper C (ed.): Amino Acid Analysis Protocols. Totowa, New Jersey: Humana Press, 2000 16. Blonde-Cynober F, Aussel C, Cynober L: Abnormalities in branched-chain amino acid metabolism in cirrhosis: Influence of hormonal and nutritional factors and directions for future research. Clin Nutr 18:5-13, 1999 17. Raj DS, Oladipo A, Lim VS: Amino acid and protein kinetics in renal failure: An integrated approach. Semin Nephrol 26:158-166, 2006 18. Lindholm B, Alverstrand A, Furst P, Bergstrom J: Plasma and muscle amino acids during continuous ambulatory peritoneal dialysis. Kidney Int 35:1219-1226, 1989 19. Baylis C: Arginine, arginine analogs and nitric oxide production in chronic kidney disease. Nat Clin Pract Nephrol 2: 209-220, 2006 20. Rysz J, Guga P, Cia1kowska-Rysz A, et al: Blood serum and neutrophil amino acid concentrations during hemodialysis. Pol Merkuriusz Lek 19:769-773, 2005 21. Brunini TM, Roberts NB, Yaqoob MM, et al: Activation of L-arginine transport in undialysed chronic renal failure and continuous ambulatory peritoneal dialysis patients. Clin Exp Pharmacol Physiol 33:114-118, 2006 22. Reis PF, da Silva CD, Brunini TM: Plasma amino acid profile and L-arginine uptake in red blood cells from malnourished uremic patients. J Ren Nutr16325-331, 2006 23. Aquilera A, Codoceo R, Bajo MA, et al: Eating behaviour disorders in uraemia: A question of balance in appetite regulation. Semin Dial 17:44-50, 2004 24. Aquilera A, Sanchez-Tomero JA, Selgas R: Brain activation in uremic anorexia. J Ren Nutr 1:57-61, 2007 25. Bossola M, Tazzal A, Giungi S, Luciani G: Anorexia in hemodialysis patients: An update. Kidney Int 14:23-25, 2006 26. Nasri H: Serum C-reactive protein (CRP) in association with various nutritional parameters in maintenance hemodialysis patients. Bratisl Lek Listy 106:390-395, 2005 27. Cano NJ, Fouque D, Leverve XM: Application of branched-chain amino acids in human pathological states: Renal failure. J Nutr 136:299-307, 2006