[Diabetic nephropathy and heart surgery--prevention of perioperative acute renal failure]

The diabetic micro- and macroangiopathy leads to retinopathy, nephropathy, peripheral arteriosclerosis and coronary heart disease. Diabetic patients with identified comorbidity, such as nephropathy and peripheral arteriosclerosis, have the highest mortality after heart surgery. Long-term survival is better after coronary surgery compared to catheter angioplasty without or with stent implantation. Compared to "on-pump" surgery using a cardiopulmonary bypass, "off-pump surgery is associated with a lower incidence of acute renal failure. In order to prevent acute renal failure in the course of heart surgery in diabetic patients, the following recommendations should be followed: i) the degree of renal damage (as indicated by the parameters: creatinine-clearance, albuminuria and blood pressure) has to be known before start of surgery; ii) the volume status (central venous pressure, central venous oxygen saturation) should be controlled tightly starting 12 hours before surgery; iii) if the volume status gets out of control post surgery, intensive care treatment using dopamine or loop diuretics should be stopped after 12-24 hours in case of treatment failure; iv) reduce the dose of or better avoid nephrotoxic substances (radio contrast media, antibiotics, non-steroidal pain killers; v) start effective renal replacement therapy early (daily intermittent or continuous hemodialysis, hemofiltration or hemodiafiltration).

Low dialysance of asymmetric dimethylarginine (ADMA)--in vivo and in vitro evidence of significant protein binding

BACKGROUND

Increased blood levels of the endogenous nitric oxide synthase inhibitor asymmetric dimethylarginine (ADMA) predict cardiovascular mortality in patients with end-stage renal disease. Despite its low molecular weight, available information on the impact of hemodialysis (HD) on ADMA plasma levels is controversial.

METHODS

We assessed plasma concentrations, dialyzer clearance and total amount of ADMA removed in 30 patients with end-stage renal disease during regular HD. In addition, plasma ADMA levels were assessed in 10 patients with acute renal failure treated with extended HD.

RESULTS

Regular HD decreased plasma creatinine (from 774 +/- 42 to 312 +/- 17 micromol/l) and urea (from 24.5 +/- 1.5 to 8.4 +/- 0.5 mmol/l) concentrations significantly (both p < 0.001), whereas plasma ADMA remained unchanged (4.35 +/- 0.19 vs. 4.76 +/- 0.24 micromol/l). ADMA clearance was 92 +/- 6 ml/min, and the total amount removed in the spent dialysate was 37 +/- 4 micromol. The clearances of creatinine (161 +/- 3 ml/min) and of urea (173 +/- 3 ml/min) were significantly higher. Furthermore, even during extended HD, plasma ADMA concentrations did not decrease significantly (1.73 +/- 0.22 vs. 1.63 +/- 0.18 micromol/l).

CONCLUSION

In conclusion, dialysance of ADMA is markedly lower than expected from its molecular weight because of significant protein binding of the substance. Since markedly increased ADMA blood concentrations have been linked to cardiovascular complications due to atherosclerosis in patients with ESRD, new strategies should be evaluated to remove this putative uremic toxin.

Low dialysance of asymmetric dimethylarginine (ADMA) - in vivo and in vitro evidence of significant protein binding

BACKGROUND

Increased blood levels of the endogenous nitric oxide synthase inhibitor asymmetric dimethylarginine (ADMA) predict cardiovascular mortality in patients with end-stage renal disease. Despite its low molecular weight, available information on the impact of hemodialysis (HD) on ADMA plasma levels is controversial.

METHODS

We assessed plasma concentrations, dialyzer clearance and total amount of ADMA removed in 30 patients with end-stage renal disease during regular HD. In addition, plasma ADMA levels were assessed in 10 patients with acute renal failure treated with extended HD.

RESULTS

Regular HD decreased plasma creatinine (from 774 +/- 42 to 312 +/- 17 micromol/l) and urea (from 24.5 +/- 1.5 to 8.4 +/- 0.5 mmol/l) concentrations significantly (both p < 0.001), whereas plasma ADMA remained unchanged (4.35 +/- 0.19 vs. 4.76 +/- 0.24 micromol/l). ADMA clearance was 92 +/- 6 ml/min, and the total amount removed in the spent dialysate was 37 +/- 4 micromol. The clearances of creatinine (161 +/- 3 ml/min) and of urea (173 +/- 3 ml/min) were significantly higher. Furthermore, even during extended HD, plasma ADMA concentrations did not decrease significantly (1.73 +/- 0.22 vs. 1.63 +/- 0.18 micromol/l).

CONCLUSION

In conclusion, dialysance of ADMA is markedly lower than expected from its molecular weight because of significant protein binding of the substance. Since markedly increased ADMA blood concentrations have been linked to cardiovascular complications due to atherosclerosis in patients with ESRD, new strategies should be evaluated to remove this putative uremic toxin.

Interleukin-18, interleukin-18 binding protein and impaired production of interferon-gamma in chronic renal failure

Uremia is associated with suppressed cellular immune responses, manifested, in part, by impaired interferon-gamma (IFNgamma) production. We investigated the influence of kidney function on plasma levels of interleukin-18 binding protein (IL-18BP), the naturally occurring inhibitor of IL-18.

METHODS

Plasma levels of IL-18, IL-18BP and IFNgamma were measured by specific immunoassays in patients with normal kidney function (NKF, n = 29), in patients with chronic renal insufficiency (CRI, n = 29), and in patients on hemodialysis (HD, n = 40). In addition, Staphylococcus epidermidis-induced production of IFNgamma and IL-18 in whole blood cultures was determined in 12 patients on HD and compared to production in 9 controls with NKF.

RESULTS

Plasma IL-18 (mean +/- SEM) in NKF was 17.9 +/- 3.6 pg/ml, in CR142.6 +/- 7.0 pg/ml (p < 0.01), and in HD 93.5 +/- 13.6 pg/ml (p < 0.001). The level of IL-18BP in NKF was 3.4 +/- 0.4 ng/ml, in CRI 7.5 +/- 0.7 ng/ml (p < 0.001), and in HD 13.1 +/- 0.8 ng/ml (p < 0.001). Plasma IL-18BP was inversely correlated with creatinine clearance (correlation coefficient: -0.7479). The level of free IL-18 was calculated in NKF to be 13.8 +/- 3.3 pg/ml, in CRI 23.6 +/- 3.9 pg/ml (not significant), and in HD 39.6 +/- 5.9 pg/ml (p < 0.001). Stimulated whole blood production of IFNgamma in NKF was 185 +/- 74 pg/10(6) mononuclear cells (PBMC), but suppressed in HD to 27.3 +/- 16 pg/10(6) PBMC (p < 0.05).

CONCLUSION

In uremia, retention of IL-18BP does not suffice to neutralize most of the concomitantly raised levels of total IL-18 resulting in elevated levels of free IL-18. Nevertheless, IFNgamma production in whole blood is reduced in patients on HD. Therefore, suppression of IFNgamma production in uremia may be due to inhibitors of IFNgamma production other than IL-18BP.

Interleukin 18 and interleukin 18 binding protein: possible role in immunosuppression of chronic renal failure

Although interleukin (IL)-18 is a member of the IL-1 family of ligands, IL-18 appears to have unique characteristics, particularly in the regulation of the T helper type 1 (Th1) response. Th1 responses are required for tumor surveillance, killing intracellular organisms, and to provide help for antibody production. In patients with chronic renal failure, the well-known immunosuppression contributes to a failure to respond to infectious challenges and vaccinations. The most salient biological property of IL-18, linking this cytokine to the Th1 response, is its ability to induce interferon gamma (IFN-gamma). In fact, IL-18 was originally identified as an IFN-gamma-inducing factor, and IFN-gamma production is the hallmark of the Th1 response. Dysregulation of IFN-gamma production resulting from reduced activity of IL-18 would explain one of the mechanisms of immunosuppression in patients with chronic renal failure. The activity of IL-18 can be regulated by the IL-18-binding protein (IL-18BP), a glycoprotein of 40,000 daltons, which is constitutively expressed and appears to be the natural inhibitor of IL-18 activity. Unlike soluble receptors for IL-18, IL-18BP does not have a transmembrane domain; IL-18BP is a secreted protein possessing a high-affinity binding and ability to neutralize IL-18. IL-18BP was discovered in human urine and is excreted in health following glomerular filtration. With decreasing renal function, the concentrations of IL-18BP in the circulation are elevated as compared with subjects with a normal renal function, and these elevated levels may result in a decreased IL-18 activity. Because of the importance of IL-18 and IFN-gamma in the Th1 response, the biology of IL-18 and IL-18BP is reviewed here in the context of the immunosuppression of chronic renal failure.

A switch to high-flux helixone membranes reverses suppressed interferon-gamma production in patients on low-flux dialysis

Long-term hemodialysis (HD) induces an inflammatory response and is associated with a suppressed cellular immune response manifested, in part, by impaired interferon (IFN-gamma) production. We investigated the effect of high-flux HD using the synthetic Helixone membrane and ultrafiltered dialysate on plasma levels of inflammatory mediators and on the whole blood production of IFN-gamma.

METHODS

Twelve ESRD patients were dialyzed under low-flux HD (polysulfone F6) and again after 6 weeks of high-flux HD (Helixone FX100). Ultrafiltered bicarbonate dialysate without bacterial growth and no detectable endotoxin was used throughout the study. Plasma levels of urea, albumin, beta(2)-microglobulin (beta(2)-m), interleukin (IL)-6, C-reactive protein (CRP), IL-1 receptor antagonist (IL-1Ra), IL-18, and IL-18-binding protein (IL-18BP) were measured. In addition, the Staphylococcus epidermidis-induced production of IFN-gamma and IL-18 was assessed in whole blood cultures of HD patients as well as in 9 healthy subjects.

RESULTS

Plasma levels of urea, albumin, IL-6, IL-1Ra and CRP were not significantly different between high-flux and low-flux HD. In contrast, beta(2)-m levels decreased significantly by 31% with high-flux Helixone (p < 0.002). Stimulated whole blood production of IFN-gamma was reduced in low-flux HD but increased to near normal levels after 6 weeks of high-flux HD. Plasma levels of free IL-18 and its specific inhibitor IL-18BP were not different between the two dialyzer membranes.

CONCLUSION

Compared to low-flux polysulfone HD with ultrafiltered dialysate, high-flux HD with the synthetic Helixone membrane did not result in a significant change in plasma levels of proinflammatory (IL-6, CRP, IL-18) and anti-inflammatory (IL-1Ra, IL-18BP) cytokines. However, high-flux HD restores whole blood IFN-gamma production without significant changes in free IL-18. Therefore, the immune modulation in high-flux HD is likely due to removal of inhibitors of IFN-gamma production other than IL-18BP.

When Good Water Goes Bad: How It Happens, Clinical Consequences and Possible Solutions

Dialysis fluid produced by state-of-the-art water preparation and distribution is contaminated with gram-negative bacteria and cytokine-inducing substances (CIS) derived from these microorganisms. The presence of a biofilm increases the risk of continuous contamination of dialysis fluid. Depending on the type of dialyzer membrane (cellulosic vs. synthetic) and the mode of dialysis (low flux vs. high flux with backfiltration), CIS may penetrate intact dialyzer membranes, induce cytokine production in the patient's blood and contribute to chronic inflammation associated with long-term hemodialysis therapy. Measures to improve the microbiological quality of dialysis fluid are: (1) the awareness of the problem and regular testing of dialysate samples using adequate methods; (2) disinfection of the entire water preparation and distribution system on a regular basis, replacement of biofilm-containing tubings, and (3) installation of ultrafilters in the dialysate circuit in particular when high-flux hemodialysis modalities are performed.

A pilot study of coupled plasma filtration with adsorption in septic shock

OBJECTIVE

To test the hypothesis that nonselective plasma adsorption by a hydrophobic resin (coupled plasmafiltration and adsorption) could improve hemodynamics and restore leukocyte responsiveness in patients with septic shock.

DESIGN

Prospective, pilot, crossover clinical trial.

SETTING

General intensive care unit in a teaching hospital.

SUBJECTS

Ten patients with hyperdynamic septic shock.

INTERVENTIONS

Patients were randomly allocated to 10 hrs of either coupled plasma filtration adsorption plus hemodialysis (treatment A) or continuous venovenous hemodiafiltration (treatment B) in random order. We measured the change in mean arterial pressure, norepinephrine requirements, and leukocyte tumor necrosis factor-alpha (TNF-alpha) production (both spontaneous and lipopolysaccharide-stimulated) after 10 hrs of each treatment. We also tested TNF-alpha production from normal human adherent monocytes incubated with patients' plasma obtained before and after the resin, both with or without incubation with an anti-interleukin-10 monoclonal antibody.

RESULTS

Mean arterial pressure increased after 10 hr by 11.8 mm Hg with treatment A and by 5.5 mm Hg with treatment B (p =.001). There was an average decrease of norepinephrine requirement of 0.08 microg/kg/min with treatment A and 0.0049 microg/kg/min with treatment B (p =.003). All patients but one survived. Spontaneous and lipopolysaccharide-induced TNF-alpha production from patients' whole blood increased over time with treatment A. This increase was more marked in blood drawn after the device (plasmafiltrate-sorbent plus hemodialyzer) (p =.009). Preresin plasma suppressed lipopolysaccharide-stimulated production of TNF-alpha by 1 x 10(6)cultured adherent monocytes from healthy donors. This suppressive effect was significantly reduced after passage of plasma through the resin (p =.019) and after incubation with anti-interleukin-10 monoclonal antibodies (p =.028).

CONCLUSIONS

In patients with septic shock, coupled plasmafiltration-adsorption combined with hemodialysis was associated with improved hemodynamics compared with continuous venovenous hemodiafiltration. This result might be related to its ability to restore leukocyte responsiveness to lipopolysaccharide. These findings suggest a potential role for blood purification in the treatment of septic shock.

Beta(2)-microglobulin amyloidosis: effects of ultrapure dialysate and type of dialyzer membrane

The available data on the pathophysiology of beta(2)-microglobulin amyloidosis (beta(2)mA) suggest that this progressive disease associated with end-stage renal failure develops in several consecutive phases. First, declining kidney function leads to retention of beta(2) microglobulin (beta(2)m) and its deposition preferentially in the synovial tissue of bigger joints such as wrists, shoulders, and hips. Second, at the site of deposition, formation of unique amyloid fibrils, whose major component is beta(2)m, takes place. Deposition and fibril formation occur in the absence of modification of beta(2)mA by advanced glycoxidation end products and also in the absence of a local inflammatory response. It is later, in the third phase, that advanced glycoxidation end product modification of beta(2)m induces a local inflammatory response by attracting macrophages chemotactically and by stimulating these cells to produce and release proinflammatory cytokines. In addition, unmodified beta(2)m itself induces inflammatory activities such as upregulation of cyclooxygenase-2 and metalloproteinase-1. The severity of the local inflammation seems to determine the degree of the destructive processes in tissue and bone accompanying beta(2)mA.

β2-Microglobulin Amyloidosis: Effects of Ultrapure Dialysate and Type of Dialyzer Membrane

ABSTRACT. The available data on the pathophysiology of β2-microglobulin amyloidosis (β2mA) suggest that this progressive disease associated with end-stage renal failure develops in several consecutive phases. First, declining kidney function leads to retention of β2 microglobulin (β2m) and its deposition preferentially in the synovial tissue of bigger joints such as wrists, shoulders, and hips. Second, at the site of deposition, formation of unique amyloid fibrils, whose major component is β2m, takes place. Deposition and fibril formation occur in the absence of modification of β2mA by advanced glycoxidation end products and also in the absence of a local inflammatory response. It is later, in the third phase, that advanced glycoxidation end product modification of β2m induces a local inflammatory response by attracting macrophages chemotactically and by stimulating these cells to produce and release proinflammatory cytokines. In addition, unmodified β2m itself induces inflammatory activities such as upregulation of cyclooxygenase-2 and metalloproteinase-1. The severity of the local inflammation seems to determine the degree of the destructive processes in tissue and bone accompanying β2mA.

Pyrogen retention by highly permeable synthetic membranes during in vitro dialysis

Pyrogen permeability of the new highly permeable synthetic membrane polyethersulfone (DIAPES) was compared to polysulfone in vitro dialysis experiments with heparinized human donor blood in the blood compartment. After sterile dialysis for 5 min, dialysate was contaminated with a culture filtrate of Pseudomonas aeruginosa using high and moderate challenge doses (Limulus assay reactivity 20,000 EU/ml and 50 EU/ml, respectively). Whole blood samples were separated from the blood compartment during the sterile (5 min) and contaminated (60 min) phases of dialysis and incubated for 6 h at 37 degrees C. Blood samples were lysed, and interleukin-1beta and tumor necrosis factor alpha were measured by specific ELISAs. Moderate dialysate contamination (50 EU/ml) did not induce detectable cytokine production in whole blood. High challenge dose (20,000 EU/ml) induced whole blood interleukin-1beta and tumor necrosis factor alpha production in the blood compartment, which was higher with DIAPES than with polysulfone after 30 min. After 60 min, membrane-dependent differences were no longer detectable. Pyrogen concentrations in the dialysate decreased with time indicating adsorption of cytokine-inducing substances to the dialyzer membrane. Pyrogens adsorbed to dialyzer membranes were resuspended during recirculation of sterile phosphate-buffered saline in the dialysate compartment and retained cytokine-inducing activity as seen from whole blood incubation experiments. DIAPES and polysulfone adsorbed pyrogens in the presence of whole blood. Pyrogen adsorption to the membrane polymer and/or to the protein coat on the membrane prevented the passage of pyrogens in the presence of moderately contaminated dialysate. High grade dialysate contamination caused breakthrough of pyrogens into the blood with DIAPES and polysulfone. In order to reduce the risk of a dialysis-dependent inflammatory response, dialysate of high bacteriological quality (ultrapure dialysate) should be mandatory.

Effects of dialyzer membrane on interleukin-1beta (IL-1beta) and IL-1beta-converting enzyme in mononuclear cells

BACKGROUND

In vitro stimulation of mononuclear cells (peripheral blood mononuclear cells; PBMCs) with an endotoxin (lipopolysaccharide; LPS) revealed elevated cell-associated levels of interleukin-1beta (IL-1beta) in end-stage renal disease (ESRD) patients on Cuprophan hemodialysis (HD), suggesting a defect in the process of IL-1beta's release from activated cells. IL-1beta is initially synthesized as an inactive precursor called proIL-1beta. ProIL-1beta is processed into the biologically active mature form of IL-1beta (mIL-1beta) requiring the specific IL-1beta-converting enzyme (ICE).

METHODS

Using specific immunoassays (enzyme-linked immunosorbent assays), we measured cell-associated and extracellular proIL-1beta as well as mIL-1beta in LPS-stimulated PBMCs to test whether ICE-dependent processing of proIL-1beta and/or secretion of mIL-1beta was impaired in ESRD patients compared with healthy controls. PBMCs of healthy controls (N = 9), of ESRD patients on peritoneal dialysis (PD, N = 10), and of those patients on intermittent HD (N = 8) were studied. The same HD patients were studied three times with low-flux Cuprophan (GFS 12), low-flux polysulfon (F6 HPS), and high-flux polysulfon (F60S) in randomized order. PBMCs were separated from whole blood by Ficoll-Hypaque centrifugation and incubated in vitro for 18 hours in the presence LPS (10 ng/mL). Extracellular (PBMC culture supernatants) and cell-associated (cell lysates) levels of proIL-1beta and mIL-1beta were measured.

RESULTS

The total production (cell-associated plus extracellular) of LPS-induced IL-1beta (proIL-1beta plus mIL-1beta) was similar in healthy controls (25.96 +/- 0.84 ng/2.5 x 10(6) PBMC), PD patients (29.53 +/- 1.31 ng/2.5 x 106 PBMC), and in Cuprophan-treated HD patients (23.28 +/- 1.24 ng/2.5 x 10(6) PBMC). In normal controls, 43.6% of the total IL-1beta was processed into mIL-1beta, which was significantly more than that in PD patients (27.3%, P < 0.02) but was similar to that in Cuprophan-treated HD patients (37.1%). Comparing cell-associated and extracellular concentrations of mIL-1beta, PBMCs of normal controls secreted 82.2% of mIL-1beta; this was significantly more than that in PD patients (59.4%, P < 0.01) and that in Cuprophan HD patients (54.2%, P < 0.01). When HD patients were switched from Cuprophan to F6 HPS or F60S, neither total IL-1beta production nor processing of IL-1beta changed. However, secretion of mIL-1beta increased significantly with F6 HPS (80.6%, P < 0.01) as well as with F60S (76.6%, P < 0.02) compared with Cuprophan.

CONCLUSION

We conclude that the ability of PBMCs to produce IL-1beta in response to LPS is normal in PD patients as well as in HD patients. ICE-dependent processing of inactive proIL-1beta into biologically active mIL-1beta is reduced in PD patients, but not in HD patients. Secretion of mIL-1beta is impaired in PD and HD patients treated with Cuprophan. This impaired ability to secrete active mIL-1beta seems to be independent of ICE activity and is normalized when HD-patients are switched from Cuprophan to low- or high-flux polysulfon. Increased cell-associated levels of biologically active mIL-1beta in circulating PBMCs represent a state of inflammation that may contribute to chronic inflammatory diseases such as beta2-microglobulin amyloidosis. Replacement of Cuprophan by synthetic membranes normalizes PBMC function and reduces the state of inflammation in ESRD patients.

Chronic inflammation in hemodialysis: the role of contaminated dialysate

Routine sodium bicarbonate-buffered dialysate is contaminated with predominantly gram-negative micro-organisms. These bacteria release pyrogenic substances such as endotoxins, peptidoglycans, exotoxins and fragments thereof. Pyrogens derived from contaminated dialysate either alone or in costimulation with activated complement components are the most important activators of circulating mononuclear cells in patients on chronic intermittent hemodialysis. Activated mononuclear cells release proinflammatory cytokines which are key mediators in acute and chronic inflammatory diseases associated with long-term hemodialysis therapy. Recent experimental and clinical data suggest that the use of pyrogen-free dialysate prevents activation of mononuclear cells and improves the state of chronic inflammation, as indicated by decreased plasma levels of C-reactive protein in chronic hemodialysis patients. Future clinical studies have to prove whether the use of pyrogen-free dialysate in combination with biocompatible dialyzer membranes and tubings reduces the incidence and severity of chronic inflammatory diseases (beta(2)-microglobulin amyloidosis, muscle protein wasting, atherosclerosis) in long-term hemodialysis patients.

Extended daily veno-venous high-flux haemodialysis in patients with acute renal failure and multiple organ dysfunction syndrome using a single path batch dialysis system

BACKGROUND

In the treatment of acute renal failure in patients with multiple organ dysfunction syndrome (MODS), continuous renal replacement therapies (CRRT) are increasingly used because of excellent volume control in the presence of improved cardiovascular stability. Patients with MODS, however, are frequently catabolic and have a high urea generation rate requiring either cost-intensive high-volume CRRT or additional intermittent haemodialysis to provide adequate clearance of small-molecular waste products. We tested the closed-loop batch haemodialysis system (called Genius((R))) for the treatment of acute renal failure in patients with MODS in the intensive care unit.

METHODS

Blood flow and countercurrent dialysate flow were reduced to 70 ml/min. Thus the 75 l dialysate tank of the Genius((R)) system lasts for 18 h of extended single-path high-flux haemodialysis (18 h-HFD) using polysulphous F60 S((R)) dialysers. Blood pressure, body temperature, and venous blood temperature in the extracorporeal circuit (no heating of the dialysate), ultrafiltration rate, serum urea levels, dialyser urea clearance, and total urea removal were monitored. In addition we tested the bacteriological quality of the spent dialysate at the end of 18-h treatments.

RESULTS

Twenty patients with acute renal failure and MODS were investigated. Averaged dialyser urea clearance was 59.8 ml/min (equal to 3.6 l/h or 64.8 l/day). Total removal of urea was 14.1+/-6.5 g/day keeping serum levels of urea below 13 mmol/l. Mean arterial pressure remained stable during the 18-h treatments with a mean ultrafiltration rate of 120 ml/h. The temperature in the venous blood tubing dropped by 5+/-0.5 degrees C during the 18-h treatment (0.28 degrees C/h) in the presence of unchanged core temperature in the patients. There was no bacterial growth in 2.5 l of spent dialysate (<0.0004 colony forming units/ml).

CONCLUSIONS

Extended high-flux dialysis using the Genius((R)) system combines the benefits of CRRT (good cardiovascular stability, sterile dialysate) with the advantages of intermittent dialysis (high urea clearance, low treatment costs). High efficiency, simplicity and flexibility of the system offers the unique opportunity to use the same dialysis machine for extended time periods (18 h) as well as for shorter intermittent renal replacement therapy in critically ill patients.

The quality of dialysate: an integrated approach

The role of bacterial contamination of dialysis water with respect to chronic inflammatory diseases associated with long-term hemodialysis therapy has been greatly underestimated in the last two decades. In the present article, recent multicenter studies assessing the bacteriological quality of water and dialysate are discussed. In addition, we describe that pyrogenic substances of bacterial origin derived from contaminated dialysate penetrate intact dialyzer membranes with the consequence of the induction of an inflammatory response in the patients. The influence of dialyzer membrane characteristics on the passage of bacterial substances from dialysate into blood are discussed. Reaching the patients blood, bacteria-derived substances activate circulating mononuclear cells to produce proinflammatory cytokines. Cytokines such as interleukin-1 beta and tumor necrosis factor-alpha are mediators of the acute phase response resulting in elevated levels of acute phase proteins (for example, C-reactive protein). The consequence is a state of microinflammation that may contribute to progressive inflammatory diseases in chronic renal failure such as beta2-microglobulin amyloidosis, protein catabolism, and atherosclerosis. The use of sterile dialysate reduces cytokine production and plasma levels of acute phase proteins, and may positively influence progressive inflammatory diseases in patients with end-stage renal failure.

Pyrogen retention by the Asahi APS-650 polysulfone dialyzer during in vitro dialysis with whole human donor blood

The purpose of this study was to test the pyrogen permeability of the new Asahi polysulfone APS 650 (APS) dialyzer membrane with a high permeability for middle molecules (up to 40 kDa) in comparison with the high-flux Fresenius polysulfone F60S (F60S) membrane. Dialyzers were tested in parallel in vitro dialysis experiments with whole human donor blood in the blood compartment and contaminated bicarbonate dialysate in the dialysate compartment. Dialysate was contaminated by a filtrate (0.45 microm) of a Pseudomonas aeruginosa culture in bicarbonate dialysate. The production of interleukin-1beta (IL-1beta) and tumor necrosis factor alpha (TNFalpha) in whole blood samples taken from the in vitro dialysis system was used to detect the passage of cytokine inducing bacterial substances derived from P. aeruginosa across the two highflux polysulfone membranes. Compared with a sterile control period at the beginning of each experiment (n = 5), the TNFalpha inducing activity in the dialysate increased from (mean +/- SEM) 42 +/- 12 pg/ml to 1,288 +/- 356 pg/ml with F60S dialyzers and from 37 +/- 10 pg/ml to 928 +/- 249 pg/ml with APS dialyzers 30 minutes after the dialysate was contaminated. The IL-1beta inducing activity in the dialysate increased similarly. In the presence of this significant contamination in the dialysate, whole blood circulating in the blood compartments for 60 minutes was not stimulated to produce increased amounts of TNFalpha or IL-1beta with neither of the two tested membranes. We conclude that F60S and APS membranes are equal in their ability to prevent the passage of cytokine inducing bacterial substances from highly contaminated dialysate into the patients' blood during hemodialysis.