NDT Advance Access originally published online on July 5, 2006
Nephrology Dialysis Transplantation 2006 21(10):2851-2858; doi:10.1093/ndt/gfl322
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ultrapure dialysate and inflammatory response in haemodialysis evaluated by darbepoetin requirements—a randomized study
1Hospital Meixoeiro, Complejo Hospitalario Universitario de Vigo and 2Centro de Diálisis Os Carballos, Fundación Renal Iñigo Alvarez de Toledo, Nephrology, Vigo, Pontevedra, Spain
Correspondence and offprint requests to: José M. Lamas, Hospital Meixoeiro – CHUVI. unidad de Nefrología. Alto de Puxeiros s/n. 36200. Vigo. Pontevedra. Spain. Email: jose.maria.lamas.barreiro{at}sergas.es
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Background. Dialysate quality has been suggested to influence inflammation status in patients subject to haemodialysis (HD). The aim of this study was to compare ultrapure dialysate (UPD) vs conventional dialysate (CD) with respect to darbepoetin requirements and other inflammation markers.
Methods. A controlled prospective randomized study was carried out on 78 patients from two HD units who were treated with low-flux polyamide dialysers. Patients were assigned to two groups by using different sized blocks per unit and dialysis session. One group received CD treatment while the other was treated with UPD over 12 months. From the groups, 37 patients started treatment with CD and 41 with UPD while 31 patients ended with CD and 30 with UPD. The main variables analysed were haemoglobin (Hb) and darbepoetin dose; other variables studied were C-reactive protein (CRP), albumin, interleukin-6 (IL-6) and interleukin-1 receptor antagonist (IL-1Ra).
Results. No significant differences were observed between the two groups for the variables analysed. At the beginning of the study the following values of CD and UPD were assesed: Hb 11.3 and 11.3 (g/dl); darbepoetin dose: 0.49 and 0.44 (µg/kg/week); CRP: 13 and 24 (mg/l); albumin: 3.8 and 3.7 (g/dl); IL-6: 5.94 and 4.18; and IL-1Ra: 345 and 420 (ng/l), respectively. At the end of the study the values of CD and UPD were: Hb 12 and 11.9 (g/dl); darbepoetin dose: 0.47 and 0.48 (µg/kg/week); CRP: 14 and 14 (mg/l); albumin: 3.8 and 3.7 (g/dl); IL-6: 14.03 and 12.93 and IL-1Ra: 322 and 340 (ng/l).
Conclusions. UPD does not improve the inflammatory status evaluated by darbepoetin requirements in conventional HD patients treated with low-flux polyamide dialyser. Further controlled studies are required to evaluate the clinical influence of UPD in HD with other low- and high-flux membranes.
Keywords: C-reactive protein; darbepoetin; endotoxins; haemodialysis; haemoglobin; inflammation; interleukin; ultrapure dialysate
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Patients treated with haemodialysis (HD) have been observed to maintain an inflammatory chronic condition and may show any of the following relevant clinical and analytical expressions such as: anaemia, reduction of erythropoiesis, epoietin (Epo) resistance, amyloidosis with arthropathy and carpal tunnel syndrome, malnutrition, atherosclerosis, low-albumin and low-cholesterol synthesis, decrease of iron stores and transferrin, and increase in C-reactive protein (CRP) production, protein A and ferritin [1–5]. Such inflammation is mediated by proinflammatory cytokines: interleukin 1ß, interleukin 6 (IL-6), TNF-
and others, which are activated in monocytes and other tissular cells [5]. The inflammatory response may be estimated in the short term through measurement of acute phase reactants such as CRP or albumin by quantifying the aforementioned cytokines or their inhibitors with a longer plasma life—such as the antagonist of IL1-Ra [6,7]—and controlling anaemia and Epo requirements to maintain the desired target haemoglobin (Hb) level.
The literature states that some of the relevant factors responsible for chronic inflammation in HD patients are those derived from foreign components of HD procedures, especially those in direct contact with the bloodstream such as membranes, lines and probably the dialysate. Many studies have discussed the role of membranes in chronic inflammation in HD patients. However, little attention has been paid to the role of the dialysate in such inflammations.
Conventional dialysate (CD) has been shown to contain substances derived from the lining of micro-organisms, namely, lipopolysaccharides (endotoxins), peptidoglycans and muramildipeptides, or delivered from the micro-organism itself (exotoxins), which would pass through the dialyser membrane into the bloodstream, either by diffusion or by back filtration, and stimulate the synthesis of cytokines and consequently the inflammatory response. Furthermore, the varying biocompatibility of membranes could be responsible for their different endotoxin absorption capacity from the dialysate and the quality of the same [8].
Some authors [9–13] have observed a decrease in cytokines and other acute phase reactants with the use of ultrapure dialysate (UPD) in HD as well as significant increases in serum albumin and other nutritional parameters [12,14]. UPD has been defined as dialysate with a culture <0.1 CFU (colony forming units)/ml and undetectable endotoxins <0.03 EU (endotoxins units)/ml [15].
A reduction in Epo requirements from between 26 and 30% with the use of UDP in HD as compared with CD was observed from the third month of initiating treatment in two studies with a small sample size of patients [10,11]. Another study with UPD observed that only 39% of the patients required Epo, with some very low doses (37 ± 28 U/kg/week), and a long-term prevalence reduction of carpal tunnel syndrome [14]. The latter effect had already been related to the use of UDP in a previous report [16].
These data suggest, albeit inconclusively, that the quality of dialysate could have more influence than previously suspected on chronic inflammation development in patients undergoing HD. The aim of this study was to analyse the influence of UPD on inflammation in HD patients, and such influence was evaluated using levels of darbepoetin required to maintain adequate Hb levels and also to observe the evolution of the markers of inflammatory stimuli such as CRP, albumin, IL-6 and IL-1Ra.
|
Subjects and methods |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
An interventional, controlled, prospective and randomized study was designed and developed in two HD units from the same sanitary area in Galicia (Northwest Spain): one from a tertiary public hospital (unit A), and the other from a AHD satellite unit (unit B), managed by a non-profit private foundation.
Patients from these two units were evaluated over a 3-month period prior to start of the study while they were being treated with CD, carried out using water purified through a process of conventional reverse osmosis (RO), where water quality complied with the UNE 111-301-90 standard [17], based on the 1982 recommendations of the Association for the Advancement of Medical Instrumentation (AAMI).
Patient inclusion criteria were: stable Hb level and darbepoetin dose during the three months prior to entry in the study with maximum variation ± 10%, aged over 18 years, voluntary participation and at least three months on HD. Patient exclusion criteria were: chronic infections or active inflammatory diseases, treatment with anti-inflammatory drugs, chronic bleeding or haemolysis, major surgery in the 3 months prior to start of the study and treatment with cytostatics or radiotherapy or active cancer. Acute infections—defined as those requiring <3 weeks of antibiotic treatment—were not considered as criteria for exclusion. The following abandonment criteria were used: voluntary decision by the patient, reception of a renal transplant or death.
After the initial evaluation period and the application of inclusion and exclusion criteria, 78 patients, 32 from unit A and 46 from unit B, were assigned to two groups by using different sized blocks per unit and session. In each unit, one group continued using CD while the other changed over to being treated with UPD for 12 months. No exchange of patients was carried out between the two groups during the study.
Low–flux polyamide dialysers were used in both groups (Poliflux L®, Gambro, with a surface area of between 1.4 and 2.1 m2 to attain the target of single-pooled Kt/V (spKt/V) 
1.3 and ultrafiltration coefficient between 9.5 and 16 ml/mmHg/h) and HD machines: Integra® (Hospal), Monitral® (Hospal) and AK200 Ultra® (Gambro) in group A, and Toray 321 EX® and AK90® (Gambro) in group B.
The following aetiology depicted renal chronic kidney disease (percentage per group: CD/UPD) was: diabetic nephropathy (14/29), vascular nephropathy (16/15), chronic glomerulonephritis (CGN) (27/5), chronic tubulo-interstitial nephropathy (CTIN) (8/7), hereditary nephropathy (16/5), systemic disease (5/7), unknown (6/20) and others (8/12).
UPD was defined as liquid with endotoxin levels <0.05 EU/ml (the lower sensitivity limit in our laboratory) and sterile culture (<0.1 CFU/ml). UPD was therefore obtained by inserting different ultrafilters in the dialysis circuit between the dialysis monitor and dialyser. The following ultrafilters were used: Diapure®, polysulphone (Medica-Izasa) in most of the monitors (Integra®, Monitral®, Toray 321 EX® and AK90®), U8000 S®, polyarylether (Gambro) in a AK200 Ultra® monitor and Arylane® (Hospal) in two Integra® monitors.
The main variables analysed were: Hb and darbepoetin dose, which were measured monthly; other variables related to the inflammation studied were CRP, albumin, measured monthly, and IL-6 and IL-1Ra, measured quarterly. At the beginning of the study, age, sex, time on dialysis (months), underlying disease and comorbidity (expressed as percentage of diabetic patients, ischaemic cardiopathy, peripheral vascular disease, cerebrovascular disease, chronic hepatopathy, chronic obstructive pulmonary disease or tumour) were controlled.
Other variables were analysed: ferritin and transferrin saturation index (every two months), spKt/V, normalized protein nitrogen appearance (nPNA) (monthly), intact parathyroid hormone (iPTH) (every four months), leucocytes and lymphocytes (monthly); total iron dose and number of packed red cell administered during the study. Days of catheter use for HD, number of days admitted to hospital, days of systemic antibiotic treatment for acute infections, and months of treatment per patient with angiotensin converting enzyme inhibitors (ACEIs) and/or angiotensin receptor blocker (ARBs) were also analysed.
A monthly control of cultures and endotoxin levels in RO water, CD and UPD dialysates was carried out in both units, alternating between the different dialysis sites and monitors.
Haematology analysis was performed with a Coulter (Gens), while albumin and the rest of the routine biochemical parameters were measured with an Advia 1650 autoanalyser (Bayer). CRP was analysed by kinetic nephelometry (Beckman®) with an Immage® nephelometer, iPTH was measured by electrochemoluminescence with an Elexis (Roche) and the cytokines IL-6 and IL1-Ra determination was measured by ELISA technique (R&D Systems). The bacteriological analyses of the RO water and dialysate were performed via culture in a substrate of agar soy tripticase extract at 20°C for five days and endotoxins were measured using limulus amebocyte lysate (LAL) assay liquid-phase quantitative photometric method for the first two months and by LAL quantitative chromogenic kinetic (K-QCL) method during the rest of the study due to lack of reliability of the initial colorimetric method.
Follow-up of patients was carried out by two nephrologists in charge of each unit, using the same management treatment criteria for each group. The darbepoetin dose, administered intravenously and weekly, was modified according to the criteria of the nephrologists incharge, to comply with the objective of maintaining Hb levels between 11–13 g/dl and levels of ferritin and transferrin saturation index above 200 mg/dl and 20%, respectively. Intravenous iron was administered in HD if deemed necessary.
The study was evaluated and approved by the Ethics & Clinical Research Committee of Galicia.
Statistical analysis
It was initially estimated that the study should be completed with a minimum of 30 patients from each group in order to detect a 25% difference in darbepoetin consumption at a confidence level of 95% and a power of 80%.
A monthly descriptive analysis was carried out for the variables recorded, where the mean values between both groups at the start and at the end of the study were evaluated using Student's t-test. The results are shown as mean and SD or mean and range for culture and endotoxins in the water and dialysate. The endotoxin results refer to the last 10 months due to lack of reliability of the initial data.
A bilateral hypothesis was assumed and the accepted significance level was taken to be P < 0.05. Multiple regression analysis was performed adjusting for the covariables diabetes and ischaemic cardiopathy—codified as 0 (absent) and 1 (present)—CRP, days of utilization of catheter for HD and days of hospital admission as possible confounders on the effect of the dialysate type on the darbepoetin dose.
Data were recorded using an ACCESS 2000 database, and statistical analysis of the data was carried out using the SPSS 11.5 software package for Windows (Chicago, IL, USA).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
A total of 37 patients continued receiving treatment with CD, while 41 were assigned to the UPD group. Six patients from the CD group abandoned the study (two because of receiving a renal allograft and four due to death) and 11 of UPD group (two because of receiving renal allograft and nine due to death). Thirty-one patients from the CD group and 30 from the UPD group completed the study (Figure 1).
|
No differences in demographic data or in the comorbidity conditions were observed between groups at the onset of the study, except for a higher ratio of diabetic status and ischaemic cardiac disease in the UPD group (Table 1).
|
No significant differences between the two groups were observed at the start or at the end of the study with respect to Hb levels and darbepoetin dose. Differences were not observed either with respect to albumin, CRP, IL-6 and IL-1Ra or the rest of the variables that were analysed (Tables 2 and 3), except for a statistical difference, which was seen to be clinically insignificant, in the initial spKt/V and in the final nPNA. Interleukin levels were compared in the 9th month because of a technical problem with sample management in the 12th month. The monthly evolution of Hb, darbepoetin dose per kilogram body weight, CRP, albumin and spKt/V are shown in Figures 2–6
.
|
|
|
|
|
|
|
Significant differences were not observed between the two groups for the use of arteriovenous fistula at the start and at the end of the study (all patients used arteriovenous fistula or catheter for HD), episodes of arteriovenous fistula failure or days of utilization of catheter for HD. Likewise, no differences were observed in the total amount of iron or red cell package administered, days of hospital admission or systemic antibiotic treatment, the use of ACEIs or ARBs (Table 4) or the number of patients and reasons for abandoning the study. After adjusting by multiple regression analysis, neither diabetic status, ischaemic cardiac disease nor the other variables analysed as possible confounders (CRP, days of utilization of catheter for HD and days of hospital admission) were found to influence the dialysate effect on darbepoetin dose [= 0.018 (95% CI: –0.33 to 0.57)].
|
A base analysis carried out on patients who abandoned the study showed no significant detectable differences between the two groups for the principal variables studied.
In the RO water, the mean of the cultures was 12.30 (0–168) CFU/ml, the cultures were negative in 75% of the samples and the endotoxin mean levels were 0.21 (<0.05–1.16) EU/ml. In CD, the mean of the cultures was observed to be 150 (0–1000) CFU/ml (
200 CFU/ml in 87.50% of the samples) and the mean of endotoxin levels was 353.61 (0.05–2.645) EU/ml (>2 EU/ml in 75% of the samples). In UPD, all cultures were negative, except for a measurement that coincided with an endotoxin level of <0.05 EU/ml and with normofunction of the ultrafilter of the dialysate, which was interpreted as external contamination of the sample, and the endotoxin measurements were observed to be <0.05 EU/ml, except for two samples with values of 0.1 and 0.2 EU/ml.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The present study is the largest controlled study performed to this day on the influence of dialysate on the clinical situation of haemodialysis patients. No significant differences were observed in the inflammatory status, measured by Hb levels and weekly darbepoetin requirements among chronic patients undergoing HD treated with UPD and those treated with CD. This observation, therefore, rules out the initial hypothesis suggested in other studies that UPD would decrease the amount of erythropoietic agents needed for the improvement of the inflammatory state. The consistency of these results is further backed by the uniformity of the monthly evolution of Hb and darbepoetin dose per kilogram body weight. No significant differences were observed between the groups for CRP levels, albumin, IL-6 or IL-1Ra.
The potential clinical advantages of UPD use have already been suggested by two retrospective studies [14,16] in which a decrease in the prevalence of carpal tunnel syndrome was observed at 10 and 13 years, in one study, and at 6 years, in the other, which used UPD and cellulose membranes. Nonetheless, it is possible that other factors may be involved in these differences, as has been suggested in another study [18] in which an 85% reduction in the prevalence of amyloidosis in a population undergoing HD in 1996 was also observed with respect to another population in 1988; both groups were treated with CD, and they were matched with age and time on dialysis. The authors attributed these findings to the improvement in quality of the dialysis water treatment and dialysate in the first group, but they valued other possible causes such as an increasing dialysis dose, use of biocompatible membranes, use of Epo, less transfusion requirements and lower levels of serum aluminium in this group.
Sitter et al. [10], in a 12-month prospective study with 30 male patients undergoing HD, who were treated with high-flux polysulphone membranes and divided into two groups, one using CD and the other using UPD, observed less requirements for Epo, and lower levels of CRP and IL-6 in patients dialysed using UPD with respect to those treated with CD. In another work from the same centre [11], using the same type of dialyser and with 12 patients in each group followed by 6 months with dialysate switch every 3 months, the authors also observed a lesser Epo requirement when using UPD. Both studies pointed out that the initial Kt/V values in both groups did not show significant differences, but do not provide information on the later evolution of this parameter. Furthermore, endotoxins were not measured in the dialysate, which was only defined as UPD by use of an ultrafilter for dialysate and negative cultures.
In two further studies from the same group [12,13], in which they also used high-flux polysulphone membranes, with 48 and 30 patients, and a patient follow-up of 12 and 24 months, respectively, it was observed that patients dialysed with UPD had lower levels of CRP and IL-6, decreased their Epo requirements, showed an improvement in nutritional status [12] and better residual renal function [13] with respect to those patients treated with CD. The authors attributed these benefits to the UPD. Endotoxin levels were determined in the dialysate at the start and end of the first study [12] and at the start, midway and end of the second study [13]. It is remarkable that, in the latter work, the quality of dialysate may have been similar in both groups during the majority of the study, since endotoxin levels were undetectable (<0.03 EU/ml) in 83% of the samples from the patients being treated with CD (15 out of 18), as was also seen in the group being treated with UPD, and the remaining three oscillated between 0.05 and 1.2 EU/ml.
In contrast, Lonneman et al. [8] used a low-flux polysulphone membrane and did not observe significant differences in CRP levels in the two populations of patients, one of which was treated with CD and the other with UPD. This observation agrees with our results. Bommer et al. [19], using high-flux polysulphone membranes, also did not find significant differences between either CRP levels, Hb and albumin in 25 patients dialysed over two years with UPD and their levels in the previous 2.5 years in which they had been using CD or the levels of another 41 patients who continued on HD with CD.
The discrepancies between these studies could be partly due to the endotoxin permeability being different in the various membranes. It has been pointed out that cellulose membranes are less permeable than synthetic membranes [20]. It has also been suggested that there are differences in the permeability to endotoxins between various types of polysulphone membranes from different manufacturers [21]. Likewise, differences have also been observed between the synthetic membranes [8], and those of low flux have been observed to be less permeable to endotoxins than those of high flux [22], which may explain the differences observed between the studies carried out using low-flux synthetic membranes [8] and studies from the group of Schiff and Sitter [10–13] carried out using high-flux polysulphone membranes, although this would not explain the controversial findings between this group and those of Bommer et al. [19].
The findings of the present study, in which CD has high levels of endotoxins, would indicate that the level of these in the dialysate is not as relevant as had been previously thought [23] or that the membrane used, namely, the low-flux polyamide membrane, would have an important capacity to retain these endotoxins, thus minimizing the clinical repercussions [24]. The role of dialyser membrane in the passage of endotoxins from the dialysate to the blood stream has also been suggested by Pereira et al. [25], who observed that mononuclear peripheral blood cells did not show an increased cytokine release upon their being exposed to contaminated dialysate through to low or medium ultrafiltration dialysers, while they show it when being exposed directly to contaminated dialysate. In this regard, it is known that both polyamide and polysulphone membranes are used for the production of ultrapure water due to their capacity for absorbing endotoxins.
One shortcoming of our study was that a strict randomization could not be carried out due to HD centre assignment and HD shift session limitations, which we tried to solve by assigning patients to groups using blocks per unit and session. After making up the groups, we observed a larger proportion of diabetic nephropathy and unknown pathology in the aetiology of IRC for the UPD group, which probably resulted from an expression of unidentified renal vascular pathology, and we further observed a higher proportion of diabetes and ischaemic cardiopathy as morbidity, although these did not influence the results after they were adjusted by multiple regression analysis.
Other shortcoming were that it was impossible to blind the patients assigned to the UPD group and that two HD centres were involved in the study with two different teams of nephrologists, although they used a common treatment and follow-up protocol, which would, on the other hand, increase its external validity.
Another important limitation was the huge dispersion of values of CRP, IL-6 and IL-1Ra observed in our patients, which reduced the comparative analysis statistical potential of these variables and greatly reduced the value of drawing conclusions from the same. The great variation and elevated CPR level in both groups is probably because we studied an unselected HD age group population with a high morbidity. We did not detect any other specific pathology that could explain such CRP levels, save for the intercurrent acute infections, which are not considered as an exclusion criteria of the study, and therefore patients received an average antibiotic treatment of between 5.97 and 4.90 days for the CD and UPD groups, respectively.
The high endotoxin levels detected in CD—above those published by other authors [26], despite their level in RO water remaining within acceptable limits–confirm that the HD monitor is a potential source of dialysate contamination, which warrants more frequent microbiological and endotoxin controls of the dialysate than those already carried out for RO water, especially when high-flux membranes are being used.
Our experience tells us that it is also relevant to select a valid and sensitive method for endotoxin measurements, since our observation in the first two months, using a liquid-phase quantitative photometric method, showed that the levels detected were falsely low—mean value in CD was 0.018(0.01–0.02) EU/ml—as noted with the quantitative chromogenic kinetic method used later.
In light of our results, we agree with Tielemans et al. [27] in that, although it may seem logical to use UPD, as part of a strategy for continuous global improvement in the treatment with HD, there is not enough evidence-based data to recommend the use of UPD in conventional haemodialysis. This becomes more relevant when low flux polyamide and polysulphone membranes are used, in which no differences between UPD and CD have been found to date.
We would recommend that further controlled studies with more homogeneous groups and higher sample size should be carried out to determine the clinical influence of UPD in HD in the medium and long term with other low- and high-flux membranes. In the mean time, and bearing in mind the absence of evidence-based data and the costs of generalizing the use of UPD, any recommended use of the same should be currently limited to HD performed with high-flux membranes, since these could provide a higher risk of passage of endotoxins to the patient.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The authors thank Dr Cigarran and specially Dr Fernandes for their assistance in preparing the English version of this article. This study received a grant from the ‘Fondo de Investigaciones Sanitarias del Ministerio de Sanidad y Consumo de España’, PI 021822, and was partially supported by a research grant from AMGEN Inc.
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
- Yao Q, Axelsson J, Stenvinkel P, Lindholm B. (2004) Chronic systemic inflammation in dialysis patients: an update on causes and consequences. ASAIO J 50:52–57.[Medline]
- Kaysen G. (2003) Serum albumin concentration in dialysis patients: why does it remain resistant to therapy? Kidney Int 64:Suppl 87, S92–S98.
- Drüecke T. (2000) ß2-microglobulin and amyloidosis. Nephrol Dial Transplant 15:Suppl 1, 17–24.
[Abstract/Free Full Text] - Goicoechea M, Martin J, Sequera P, et al. (2000) Role of cytokines in the response to erythropoietin in haemodialysis patients. Kidney Int 54:1337–1343.
- Bologa R, Levine D, Parker T, et al. (1998) Interleukin-6 predicts hypoalbuminemia, hypocholesterolemia and mortality in haemodialysis patients. Am J Kidney Dis 32:107–114.[Web of Science][Medline]
- Pereira B. (1995) Balance between pro-inflammatory cytokines and their specific inhibitors in patients on dialysis. Nephrol Dial Transplant 10:Suppl 7, 27–32.[Abstract]
- Dinarello C. (1995) Interleukin-1 and Interleukin-1 receptor antagonist production during haemodialysis: which cytokine is a surrogate marker for dialysis-related complications? Nephrol Dial Transplant 10:Suppl 3, 25–28.
[Abstract/Free Full Text] - Lonneman G. (2000) The quality of dialysate: an integrated approach. Kidney Int 58:Suppl 76, S112–S119.
- Schindler R, Lonnemann G, Schäffer J, Shaldon S, Koch K, Krautzig S. (1994) The effect of ultrafiltered dialysate on the cellular content of interleukin-1 receptor antagonist in patients on chronic haemodialysis. Nephron 68:229–233.[Web of Science][Medline]
- Sitter T, Bergner A, Schiffl H. (2000) Dialysate related cytokine induction and response to recombinant human erythropoietin in haemodialysis patients. Nephrol Dial Transplant 15:1207–1211.
[Abstract/Free Full Text] - Schiffl H, Lang S, Bergner A. (1999) Ultrapure dialysate reduces dose of recombinant human erythropoietin. Nephron 83:278–279.[CrossRef][Web of Science][Medline]
- Schiffl H, Lang S, Stratakis D, Fischer R. (2001) Effects of ultrapure dialysis fluid on nutritional status and inflammatory parameters. Nephrol Dial Transplant 16:1863–1869.
[Abstract/Free Full Text] - Schiffl H, Lang S, Fischer R. (2002) Ultrapure dialysis fluid slows loss of an residual renal function in new dialysis patients. Nephrol Dial Transplant 17:1814–1818.
[Abstract/Free Full Text] - Kleophas W, Haastert B, Backus G, Hilgers P, Westhoff A, Endert G. (1998) Long-term experience with an ultrapure individual dialysis fluid with a batch type machine. Nephrol Dial Transplant 13:3118–3125.
[Abstract/Free Full Text] - Ledebo I and Nystrand R. (1999) Defining the microbiological quality of dialysis fluid. Artif Organs 23:37–43.[CrossRef][Web of Science][Medline]
- Baz M, Durand C, Ragon A, et al. (1991) Using ultrapure water in haemodialysis delays carpal tunnel syndrome. Int J Artif Organs 14:681–685.[Web of Science][Medline]
- Comité técnico Aenor. Norma UNE 111-301-90. (1991) Características del agua utilizada en hemodiálisis. Nefrología 11:7–8.
- Schwalbe S, Holzhauer M, Schaeffer J, Galanski M, Koch K, Floege J. (1997) ß2-microglobulin associated amyloidosis: a vanishing complication of long-term haemodialysis. Kidney Int 52:1077–1083.[Web of Science][Medline]
- Bommer C, Bommer J, Ritz E. (2004) Do we need ultrapure dialysate for haemodialysis in general? ASN Annual Meeting – St Louis. J Am Soc Nephrol 14:166A.
- Yamagami S, Adachi T, Sugimura T, et al. (1990) Detection of endotoxin antibody in long-term dialysis patients. Int J Artif Organs 13:205–210.[Web of Science][Medline]
- Bommer J, Becker K, Urbaschek R. (1996) Potential transfer of endotoxin across high-flux polysulfone membranes. J Am Soc Nephrol 7:883–888.[Abstract]
- Vanholder R, Van Haeche E, Veys N, Ringoir S. (1992) Endotoxin transfer through dialysis membranes: small versus large pore membranes. Nephrol Dial Transplant 7:333–339.
[Abstract/Free Full Text] - Nubé M and Grooteman M. (2001) Impact of contaminated dialysate on long-term haemodialysis-related complications: is it really that important? Nephrol Dial Transplant 16:1986–1991.
[Free Full Text] - Ertl T, Beck W, Schnell A, et al. (2003) Barrier function of low and high flux synthetic membranes for endotoxins in contaminated dialysis fluid. Blood Purif 21:358.
- Pereira B, Sundaram S, Barrett T, et al. (1996) Transfer of cytokine-inducing bacterial products across haemodialyser membranes in the presence of plasma or whole blood. Clin Nephrol 46:394–401.[Web of Science][Medline]
- Bambauer R, Schauer M, Jung W, Vienken J. (1994) Contamination of dialysis water and dialysate. A survey of 30 centers. ASAIO J 40:1012–1016.[Medline]
- Tielemans C. (2001) Are standards for dialysate purity in haemodialysis insufficiently strict? Semin Dial 14:328–329.[CrossRef][Web of Science][Medline]
Accepted in revised form: 8. 5.06
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Ouseph, S. Jones, N. Dhananjaya, and R. A. Ward Use of ultrafiltered dialysate is associated with improvements in haemodialysis-associated morbidity in patients treated with reused dialysers Nephrol. Dial. Transplant., August 1, 2007; 22(8): 2269 - 2275. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||