In the past 2 decades, there has been a phenomenal increase in the number of prevalent elderly hemodialysis (HD) patients receiving renal replacement therapy (RRT), and this trend is likely to continue. The choices of RRT for elderly patients with renal failure include continuous ambulatory peritoneal dialysis (CAPD) or HD. There is still debate over which is the most appropriate RRT for elderly HD patients. It is assumed that elderly HD patients have more hemodynamic problems than young HD patients, including chronic inflammation, malnutrition, arteriosclerosis, intradialytic hypotension, increased susceptibility to infection, low-turnover bone disease, vascular access failure, disturbance of sleep, and depression.1 Therefore, it could be possible to solve or even slightly attenuate these problems for elderly HD patients by selecting the optimal modality of dialysis and dialysis membranes.
In recent years, the number of maintenance hemodialysis (MHD) patients treated with synthetic polysulfone (PSu) membranes that have a high capacity for removing β2 microglobulins (MGs) has increased. However, it remains to be determined whether it is as advantageous for elderly HD patients to use such a synthetic membrane with a high capacity for solutes removal as it is for younger MHD patients. The hydrogel structure of the acrylonitrile-co-methallyl sulfonate (AN69) membrane has physical characteristics that are unique among non-PSu synthetic membranes. For example, its biocompatibility is derived from both its high hydrophilicity and its negative charge, which allows for adsorptive removal of various uremic substances suspected to be responsible for cardiovascular diseases (CVDs), such as inflammatory cytokines.2 On the other hand, Franssen et al.3 demonstrated that oxalate clearances differ markedly among dialyzers. The authors presumed that lower oxalate clearance by AN69 membranes might be caused by repulsion of the negatively charged oxalate by the similarly charged membrane. Thus, it is possible that the clearance of negatively charged uremic toxins by AN69 membrane is less than that of other dialyzers.
On the basis of the above characteristics using the AN69 membrane for HD, physicians can expect to perform mild and effective HD for elderly HD patients who are likely to be susceptible to unstable hemodynamic conditions, states of malnutrition, and chronic inflammatory conditions.
In this study, we conducted a crossover study for elderly MHD patients aged 75 years or older to compare AN69 with PSu with respect to solute removal during HD, the hemodynamic condition of the patients during HD, and the patients' resulting nutritional status after 3 months of treatment.
Characteristics of Patients
Twenty-eight elderly MHD patients aged 75 years or older (mean age: 78.2 years; 13 men and 15 women; 15 diabetic and 13 nondiabetic; and mean duration of dialysis: 4.2 years) were enrolled in this study (Table 1). Mean time of HD treatment was 3.8 ± 0.5 hours. Frequencies of treatments were 3 times/wk, mean blood flow was 207 ± 18 ml/min, and dialysate flow was 500 ml/min. At the start of the study, the patients with malignant cancer, chronic inflammatory diseases, hematological disorders, or severe liver or lung diseases were excluded from the study (prospective study for analyzing the significant factors affecting prognosis and CVD in chronic kidney disease: University Hospital Medical Information Network [UMIN] Clinical Trial Registry C000000411). The observation of an increased frequency of hypersensitivity reactions (HSRs) (hypotension, nausea, faintness, flushing, wheezing, or bronchospasm) after using the AN69 membrane, particularly in patients with angiotensin converting enzyme inhibitor (ACE-I), led to the hypothesis that these symptoms resulted from activation of the contact-phase system by the negatively charged AN69 surface with enhanced bradykinin generation.4 On the basis of these reports, we excluded patients treated with ACE-I from this study.
All subjects gave their informed consent in accordance with the requirements of the institutional committee on human research, and the committee approved the study protocol (Hyogo College of Medicine No. 319).
APS-SA (PSu) (hollow fiber) (Asahi Kasei Kuraray Medical, Japan), which is composed of a PSu membrane, was compared with the H12-3400 (AN69) (flat sheet) (Gambro, Sweden), which is composed of AN69XS. All the patients who participated in this study were treated with the PSu membrane with ultra-pure dialysates for the first 3 months (first PSu period). The patients were then switched to the AN69 membrane (AN69 period) for the following 3 months, and for the last 3 months, they were returned to the PSu membrane (second PSu period). Thirteen patients were treated with APS-SA 1.0 (membrane area: 1.0 m2) in the first PSu period, they were then switched to H12-2800 in the AN69 period, and subsequently returned to APS-SA 1.0 in the second PSu period. In addition, 15 patients were treated with APS-SA 1.3 (membrane area: 1.3 m2) in the first PSu period, then switched to H12-3400 in the AN69 period, and then returned to APS-SA 1.3 in the second PSu period. With the exception of the dialysis membrane, we did not change any other HD conditions such as the setting of dry weight, blood flow, dialysate flow, dosage or type of anticoagulant used, or the type of dialysate during the entire study period.
Sample Collection and Measurement of Parameters
In the last session of the first PSu period and the AN69 period, pre- and postserum levels of urea nitrogen (UN), β2MG, interleukin (IL)-6, tumor necrosis factor (TNF)-α, para-cresol (P-Cre), and indoxyl sulfate (I-Sul) were measured and calculated for each reduction ratio. At the start of the study and the last session of each membrane period, pre-HD serum total protein, albumin, total cholesterol, triglyceride level, and amino acid fractions were measured to compare the nutritional states among each treatment period. Serum levels of albumin were assayed using bromocresol green (BCG). The BCG dye-binding assay was performed on fresh plasma using a dry chemical method. The serum IL-6 and TNF-α levels were measured with enzyme-linked immunosorbent assay (ELISA) kits (Human IL-6 immunoassay kit; Biosource International, Inc., CA, and Human TNF-α immunoassay kit; Biosource International, Inc.). Serum P-Cre5 and I-Sul6 levels were measured using the high-performance liquid chromatography (HPLC).
All values are presented as the mean ± standard deviation (SD). Differences between two groups were analyzed by the unpaired Student's t-test. A paired t-test was used for the analysis of the changes of reduction of uremic substances, blood pressure, and nutritional indices of each period. p values <0.05 were considered as statistically significant. Statistical analyses were performed with the Statview software program version 5.0 for Macintosh.
Comparison of Dialysis Efficiency
We compared the AN69 membrane with the PSu membrane and found no significant difference in predialysis serum levels of UN, IL-6, and TNF-α from each period (Figure 1). In addition, there was no significant difference in the reduction ratio of UN. On the other hand, higher efficiency for removing β2MG was obtained using the PSu membrane than the AN69 membrane (mean reduction ratio of β2MG, PSu: 62% ± 8% vs. AN69: 28% ± 6%) (Figure 2). After 3 months of using the AN69 membrane period, the predialysis serum β2MG level increased from 24.6 to 27.8 mg/L; however, this was not significant (p = 0.066) (Figure 1).
In the analysis of the single dialysis sessions, the reduction ratios of TNF-α (AN69: 48% ± 4.2% vs. PSu: 47.6% ± 6.2%), I-Sul (AN69: 36.8% ± 3.8% vs. PSu: 35% ± 1.8%), and P-Cre (AN69: 43% ± 3% vs. PSu: 36% ± 4%) with PSu and the AN69 membrane were essentially the same. However, serum IL-6 levels slightly increased with the PSu membrane (mean increase ratio: 6% ± 2%) and significantly decreased with the AN69 membrane (mean reduction ratio: 48% ± 4%) (Figure 2).
Comparison of Hemodynamic Condition
We measured blood pressure before and at 1, 2, 3, and 4 hours during HD and calculated the mean value of blood pressure based on the five measurements in each period. No significant difference was seen in blood pressure (both systolic and diastolic) during the HD sessions between the AN69 period and the PSu period (Figure 3). After switching from the PSu membrane to the AN69 membrane, the frequencies of saline used for hypotension significantly decreased, and after switching back to the PSu membrane, they significantly increased again (Figure 4). Ultrafiltration was subsequently adjusted according to each patient's condition. The mean ultrafiltration per session in the first PSu period was 1,794.2 ± 560.2 ml, 1,943.9 ± 559.6 ml in the AN69 period, and 1,878.3 ± 565.0 ml in the second PSu period. There was no significant difference in total ultrafiltration among the three periods.
Changes in Nutritional Status
All patients had been treated with PSu from August to October 2008 and February to March 2009. All patients were treated with AN69 from November to January 2008–2009. Serum total protein, albumin, total cholesterol, triglyceride levels, and amino acid fractions at the last session of each dialysis membrane period were measured to evaluate the change in nutritional status. The serum concentration of phenylalanine among the essential amino acids and ornithine among the nonessential amino acids significantly decreased when the AN69 membrane was used; however, there were no significant changes in the other amino acid fractions (Table 2). On the other hand, at 3 months after switching from the PSu membrane to the AN69 membrane, indexes of nutritional status, such as serum albumin (3.5 ± 0.1 to 3.7 ± 0.1 g/dl, p = 0.01) and total protein (6.6 ± 0.1 to 6.8 ± 0.1 g/dl, p = 0.01), significantly increased. Furthermore, the total cholesterol level (158 ± 8 mg/dl to 168 ± 12 mg/dl) exhibited an increasing tendency (p = 0.078), although it was not significant. These improvements in nutritional status observed in the AN69 period dissipated in the second PSu period (Figure 5). These data suggest that the nutritional status of elderly HD patients may be better maintained with HD using the AN69 membrane than HD using the PSu membrane.
Removal of Small and Middle Molecular Compounds
Recently, in a subanalysis of the hemodialysis (HEMO) study, it was reported that the risk of death increases significantly in cases where the predialysis serum β2MG concentration exceeds 35 mg/L.7 In addition, Okuno et al.8 reported that MHD patients with predialysis concentrations of serum β2MG concentrations below 32.2 mg/L had more favorable prognoses than those with serum β2MG above 32.2 mg/dl. APS-SA, which we selected as the control membrane in this study, is a high-performance dialysis membrane with high β2MG clearance (the sieving coefficient of β2MG was 0.85) and has been modified to achieve enhanced diffusion performance and increased clearance of small molecule solutes by being equipped with wave-shaped hollow-fiber membrane bundles.9 At 3 months after switching from the PSu dialysis membrane to the AN69 membrane, the predialysis serum β2MG level did not significantly increase among the elderly HD patients enrolled in this study. The slight increase peaked at an average of 27.8 mg/L but did not reach the level reported in the HEMO study,7 which indicated high risk of death. However, the possibility that extended use of the AN69 membrane might cause complications related to high serum β2MG levels should be determined. Furthermore, solute removal rates are highly dependent on blood flow rate. Most HD patients in Japan are treated with a blood flow rate of ≤200 ml/min. Thus, a limitation of this study is that our patients were treated with a lower blood flow rate than other countries, which might hamper the ability to make comparison with previous studies. From the result of this study, the AN69 membrane is thought to have sufficiently removed small and middle molecular compounds, at least in elderly HD patients.
Removal of Inflammatory Cytokines
Interleukin-6, which is one of the inflammatory cytokines, is a well-known predictor of the outcome of MHD patients.10,11 In this study, we observed no significant difference in predialysis IL-6 levels among each period. However, the efficiency of removing IL-6 was significantly higher with the AN69 membrane than with the PSu membrane. Serum IL-6 levels significantly decreased during HD with the AN69 membrane, whereas serum IL-6 levels increased from HD sessions with the PSu membrane. This could be attributed to the AN69 membrane having a negative charge throughout its membrane surface, whereby the negative charge allows for the removal of various inflammatory cytokines such as IL-6, IL-10, and IL-18 by adsorption.2,12,13 In this study, we detected a significantly higher removal of IL-6 after HD sessions with AN69 than with PSu. However, there was no significant difference in pre-HD IL-6 levels between the AN69 period and the PSu period. From these results, we presumed that although the AN69 membrane might not be able to sustainably suppress serum IL-6 levels, it could improve chronic inflammatory conditions during and after the HD sessions.
Even for elderly HD patients, the efficiency of removing IL-6 by adsorption was higher with the AN69 membrane than with the PSu membrane, and this may have affected the nutritional state.
Removal of Protein-Bound Uremic Solutes
In recent years, the relationship between protein-bound uremic substances and CVD risk has been suspected.14,15 Among protein-bound uremic substances, the close relationship between P-Cre16,17 or I-Sul18 and CVD, in particular, has been reported. The molecular weight of P-Cre itself is as small as 108.1 Da. Meijers et al.19 reported that MHD patients with higher serum-free P-Cre levels had higher morbidity of CVD than patients with lower serum-free P-Cre level, thus indicating the possibility that effective removal leads to the prevention of CVD. However, effective removal through conventional HD cannot be expected because most of these molecules are bound with protein. In this study, there were no significant differences in the reduction ratio of I-Sul and P-Cre between both the AN69 and PSu groups. However, the mechanisms by which the removal efficiency of protein-bound uremic toxins (I-sul and P-Cre) by AN69 and PSu, which were substantially similar, are unknown. These results suggest that by using the AN69 membrane, protein-bound uremic substances that are considered difficult to remove through conventional HD may be effectively removed at a level equivalent to HD using the high-performance PSu membrane.
Effect on Hemodynamic Condition during HD
Stabilization of the hemodynamic condition during HD in elderly HD patients is considered an important issue for preventing cardiovascular and cerebrovascular complications and for improving quality of life (QOL) and the outcome of these patients.20 Although decreases in blood pressure at 1 and 3 hours after starting HD were observed in the AN69 period and PSu period (Figure 3), there were no significant differences between the PSu and AN69 period. However, the frequency of saline used for hypotension during HD was significantly decreased with the AN69 membrane compared with first and second PSu periods (Figure 4). These findings suggest that for elderly HD dialysis patients, mild removal combined with adsorption with the AN69 membrane may lead to the stabilization of the hemodynamic condition during HD, compared with rapid removal of solutes using the PSu membrane.
Effect on Nutritional Status
The most notable finding in this crossover study is that in HD by using the AN69 membrane for 3 months, indices for nutritional status, such as serum albumin and total protein, improved significantly after 3 months of HD with AN69 membrane and then returned to baseline levels after switching back to the PSu membrane. Although we selected a dialysis membrane with an albumin sieving coefficient of 0.002, which is assumed to cause a lower loss of albumin, the serum albumin level decreased significantly in the PSu period and then increased again after switching to the AN69 membrane.
Several possible mechanisms for the increase in albumin level after switching the PSu membrane to the AN69 membrane can be postulated. First, loss of amino acids is suspected to be greater in the PSu period than the AN69 period, which would therefore deteriorate the nutritional status. Several reports have stated that HD using a high flux (HF) PSu membrane causes greater loss of amino acids from the dialysis membrane than using a low-flux polymethylmethacrylate (PMMA) membrane.21,22 However, in our study, we observed no significant changes in the predialysis amino acid fractions during each period. Based on these findings, it is not likely that a loss of amino acids caused by the dialysis membrane affected the serum albumin level. Second, it is possible that the chronic inflammatory condition was improved by the adsorption of inflammatory cytokines (IL-6) by the AN69 membrane, which could have an affect on albumin synthesis. Third, stabilization of hemodynamic conditions during HD in the AN69 period may have caused an increase in the dose of HD, decrease in fatigue after HD, and an increase in appetite. We found no evidence or data indicating that stabilized hemodynamic conditions during HD session contribute to improved nutritional status of HD patients. However, we presumed that improvement of inflammatory conditions and stabilization of hemodynamic conditions during HD session with AN69 might be the cause of increased albumin levels in elderly HD patients. It has been reported that seasonal variations in clinical and laboratory parameters, especially in nutritional parameters, are common among HD patients.23 Thus, we should take into consideration that the differences in nutritional conditions or hemodynamic conditions between PSu and AN69 periods might be affected by seasonal factors (e.g., temperature or food intake). Therefore, long-term studies are needed to clarify the effect of the AN69 membrane on nutritional and hemodynamic conditions. Nevertheless, a number of reports have described a higher risk of death in HD cases with hypoalbuminemia (<3.5 g/dl)24 and particularly for elderly HD patients, who are prone to be vulnerable to states of malnutrition. Therefore, it is expected that HD using the AN69 membrane may considerably contribute to both an improved QOL and improved outcome.
In this study, we compared the APS-SA as PSu with the AN69 membrane. However, Hoenich et al.25 compared three types of PSu membranes and showed differences in clearance of β2MG and interaction with proteins and blood cells between membranes. As such, it is possible that the finding of this study do not necessarily apply to all PSu membranes. Furthermore, a limitation of this study is that geometrical differences exist between PSu (hollow fiber) and AN69 (flat sheet). Not only material differences but also structural differences may affect leukocyte activation during the HD session. We cannot rule out the possibility that the difference in serum IL-6 levels during HD sessions with PSu and AN69 membranes may be affected by structural differences.
When trying to select an optimal HD therapy for elderly patients, selecting of dialysis membranes, dialysis modality, and dialysates in a uniform manner may not be an acceptable approach. Thus, it might be necessary to perform HD therapy in accordance with the individual characteristics of each HD patient.
In elderly HD patients, the use of the AN69 membrane is considered to achieve improvements in both malnutrition and chronic inflammatory conditions in accordance with the stabilization of the hemodynamic condition and the high removal of IL-6, when compared with the PSu membrane. Therefore, we suspected that AN69 membrane may be the preferred dialysis membrane for elderly HD patients.
1. Cassidy MJ, Sims RJ: Dialysis in the elderly. New possibilities, new problems. Minerva Urol Nefrol
56: 305–317, 2004.
2. Bouman CS, van Olden RW, Stoutenbeek CP: Cytokine filtration and adsorption during pre- and postdilution hemofiltration in four different membranes. Blood Purif
16: 261–268, 1998.
3. Franssen CF: Oxalate clearance by haemodialysis—A comparison of seven dialysers. Nephrol Dial Transplant
20: 1916–1921, 2005.
4. Coppo R, Amore A, Cirina P, et al
: Bradykinin and nitric oxide generation by dialysis membranes can be blunted by alkaline rinsing solutions. Kidney Int
58: 881–888, 2000.
5. Niwa T: Phenol and p-cresol accumulated in uremic serum measured by HPLC with fluorescence detection. Clin Chem
39: 108–111, 1993.
6. Niwa T, Takeda N, Tatematsu A, Maeda K: Accumulation of indoxyl sulfate, an inhibitor of drug-binding, in uremic serum as demonstrated by internal-surface reversed-phase liquid chromatography. Clin Chem
34: 2264–2267, 1988.
7. Cheung AK, Rocco MV, Yan G, et al
: Serum beta-2 microglobulin levels predict mortality in dialysis patients: Results of the HEMO study. J Am Soc Nephrol
17: 546–555, 2006.
8. Okuno S, Ishimura E, Kohno K, et al
: Serum beta2-microglobulin level is a significant predictor of mortality in maintenance haemodialysis patients. Nephrol Dial Transplant
24: 571–577, 2009.
9. Fukuda M, Miyazaki M, Uezumi S, Yoshida M: Design and assessment of the new APS dialyzer (APS-SA series). J Artif Organs
9: 192–198, 2006.
10. Barreto DV, Barreto FC, Liabeuf S, et al
: Plasma interleukin-6 is independently associated with mortality in both hemodialysis and pre-dialysis patients with chronic kidney disease. Kidney Int
77: 550–556, 2010.
11. Hasuike Y, Nonoguchi H, Ito K, et al
: Interleukin-6 is a predictor of mortality in stable hemodialysis patients. Am J Nephrol
30: 389–398, 2009.
12. Haase M, Silvester W, Uchino S, et al
: A pilot study of high-adsorption in human septic shock. Int J Artif Organs
30: 108–117, 2007.
13. Cole L, Bellomo R, Davenport P, et al
: Cytokine removal during continuous renal replacement therapy: An ex vivo comparison of convection and diffusion. Int J Artif Organs
27: 388–397, 2004.
14. Vanholder R, Baurmeister U, Brunet P, et al
; European Uremic Toxin Work Group. A bench to bedside view of uremic toxins. J Am Soc Nephrol
19: 863–870, 2008.
15. Jourde-Chiche N, Dou L, Cerini C, et al
: Protein-bound toxins—Update 2009. Semin Dial
22: 334–339, 2009.
16. Dou L, Bertrand E, Cerini C, et al
: The uremic solute p-cresol and indoxyl sulfate inhibit endothelial proliferation and wound repair. Kidney Int
65: 442–451, 2004.
17. Meijers BK, Van Kerckhoven S, Verbeke K, et al
: The uremic retention solute p-cresyl sulfate and markers of endothelial damage. Am J Kidney Dis
54: 891–901, 2009.
18. Taki K, Tsuruta Y, Niwa T: Indoxyl sulfate and atherosclerotic risk factors in hemodialysis patients. Am J Nephrol
27: 30–35, 2007.
19. Meijers BK, Bammens B, De Moor B, et al
: Free p-cresol is associated with cardiovascular disease in hemodialysis patients. Kidney Int
73: 1174–1180, 2008.
20. Shoji T, Tsubakihara Y, Fujii M, Imai E: Hemodialysis associated hypotension as an independent risk factor for two-year mortality in hemodialysis patients. Kidney Int
66: 1212–1220, 2006.
21. Ikizler TA, Flakoll PJ, Parker RA, Hakim RM: Amino acid and albumin losses during hemodialysis. Kidney Int
46: 830–837, 1994.
22. Gutierrez A: Protein catabolism in maintenance haemodialysis: The influence of the dialysis membrane. Nephrol Dial Transplant
11(suppl 2): 108–111, 1996.
23. Cheung AK, Yan G, Greene T, et al
; Hemodialysis Study Group: Seasonal variations in clinical and laboratory variables among chronic hemodialysis patients. J Am Soc Nephrol
13: 2345–2352, 2002.
24. Tazza L, Di Napoli A, Bossola M, et al
: Ageing of patients on chronic dialysis: Effects on mortality—A 12-year study. Nephrol Dial Transplant
24: 940–947, 2009.
25. Hoenich NA, Woffindin C, Brennan A, et al
: A comparison of three brands of polysulfone membranes. J Am Soc Nephrol
7: 871–876, 1996.