Substitution of citrate with tissue plasminogen activator (rt-PA) for catheter lock does not improve patency of tunnelled haemodialysis catheters in a randomised trial.

BACKROUND: The study aim was to establish if substitution of citrate with rt-PA for catheter lock once weekly can reduce the incidence of catheter-related blood stream infections (CR-BSI) or improve patency of tunneled haemodialysis catheters. METHODS: All incident patients undergoing insertion of a tunneled haemodialysis catheter were screened and included except those suffering infection or using oral anticoagulation. Study participants were randomized into two arms according to the solution applied as catheter lock: receiving either trisodium citrate (Citra-LockTM 4%) only or rt-PA (Actilyse® 1 mg/ml) on the middle session each week with citrate used on the first and third sessions. The incidence of CR-BSI (confirmed by positive blood culture), catheter non-function (complete obstruction), and malfunction (blood flow < 250 ml/min) was recorded. Statistical significance was tested with ANOVA, post hoc analysis was performed by means of multiple linear regression. RESULTS: Totally, 18 patients were included and followed during 655 haemodialysis sessions. No episode of CR-BSI was detected while 6 catheter non-functions (0.9% sessions) and 101 malfunctions (15.4% sessions) were recorded. The incidence of both events was equal between the study arms: 4 non-functions and 55 malfunctions in the rt-PA arm and 2 non-functions and 46 malfunctions in the citrate arm (p = 0.47 and p = 0.24, respectively). Additionally, the mean blood flow achieved did not differ significantly between the arms: 326 ± 1,8 and 326 ± 1,9 ml/min (p = 0.95) in rt-PA and citrate arms, respectively. Post hoc analysis identified time elapsed since previous session (ß = 0.12, p = 0.005) and malfunction on previous session (ß = 0.25, p < 0.001) as significant factors affecting the occurrence of malfunction. By contrast, the study arm, rt-PA application on previous session, and catheter vintage did not enter the model. CONCLUSION: Substitution of citrate with rt-PA for catheter lock does not reduce the incidence of catheter malfunction neither does it affect the blood flow achieved during haemodialysis. Catheter patency is related rather to the time interval between sessions and to previous malfunction (thus probably reflecting undefined individual factors). The incidence of CR-BSI within pre-selected haemodialysis population is sporadic (less than 1 per 4.3 patient years in our sample). TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry, ACTRN12612000152820 . Retrospectively registered 03/02/2012.

Citrate-Buffered Dialysis Solution (Citrasate) Allows Avoidance of Anticoagulation During Intermittent Hemodiafiltration-At the Cost of Decreased Performance and Systemic Biocompatibility.

Reportedly, citrate-based dialysis solution enables heparin dose tapering or even complete exclusion, particularly in postdilution hemodiafiltration (HDF). The aim of the study was to verify this strategy in predilution setting and to assess its short-term safety, efficacy, and biocompatibility. Ten regular hemodialysis patients were assigned to predilution HDF on acetate- and citrate-based dialysis solutions (0.8 mmol/l trisodium citrate) at random order. Acetate HDF was performed using routine dose of heparin while citrate HDF was heparin free. Plasma calcium, thrombin-antithrombin complexes (TAT), and citrate levels were measured at 0, 30, 60, 120, and 240 min. Following each session, a semiquantitative dialyzer clotting score (DCT 1-5) was assessed and HDF adequacy was determined as spKt/V. Statistical relevance was tested by ANOVA with pP < 0.05 held significant, data are given as means ± standard deviations. All sessions were accomplished successfully, premature termination or circuit re-setting was not necessary. However, DCT was significantly higher in citrate-HDF compared to acetate-HDF regimen (3.4 ± 0.65 and 1.8 ± 0.79, respectively, P = 0.002) as well as TAT generation rate (increase per session by factor 11.0 ± 8.43 and 2.1 ± 1.26, respectively, P = 0.004 between regimens). Ionized calcium declined only by the end of citrate-HDF (from 1.09 ± 0.086 to 0.99 ± 0.030 mmol/L, P = 0.002) yet without accompanying clinical symptoms. Systemic citrate levels increased along the citrate-HDF session but stayed an order of magnitude below concentrations needed to establish citrate anticoagulation (peak at 0.276 ± 0.112 mmol/L). Dialysis adequacy estimated by spKt/V was found lower in citrate-HDF vs. acetate-HDF (1.48 ± 0.163 and 1.58 ± 0.165, respectively, P = 0.006). Although predilution HDF using citrate-based dialysate is feasible without heparin, both dialysis adequacy and biocompatibility is significantly compromised. Therefore, this approach can be adopted for a single procedure but is not acceptable on a regular basis.

Previous Vaccination and Age are More Important Predictors of Immune Response to Influenza Vaccine than Inflammation and Iron Status in Dialysis Patients.

BACKGROUND/AIMS: The immune response to influenza vaccine may be influenced by many factors, e.g. age, comorbidities or inflammation, and iron status. METHODS: We studied the vaccine-induced production of hemagglutination-inhibition antibodies (HI) in 133 hemodialysis patients (HD) and 40 controls. To identify variables associated with the immune response, uni- and multivariate regression analyses were performed with seroconversion in HI titers as a dependent variable, with demographics, comorbidities, previous vaccination, inflammation, and iron status as independent variables. RESULTS: Seroconversion rates were lower in HD than in controls [43% versus 73% (H1N1 strain; p < 0.05); 43% versus 53% (H3N2; P=NS); 36% versus 62% (B; p < 0.05)]. In both HD and control groups, the predictors of the inferior HI production were pre-vaccination seroprotection, vaccination in the previous season, and old age. We did not find associations between seroconversion rates and inflammation and iron status in the studied populations. This was also true for a subanalysis of patients without pre-vaccination seroprotection. CONCLUSION: The influenza vaccine-induced antibody production was lower in HD than in controls and was independent of inflammation and iron status in both groups. Besides dependence on dialysis, the variables associated with inferior seroconversion rates included pre-vaccination seroprotection, previous vaccination, and old age.

Asymmetric dimethylarginine and progression of chronic kidney disease: a one-year follow-up study.

BACKGROUND/AIMS: Asymmetric dimethylarginine (ADMA) is a prognostic factor in patients with chronic kidney disease (CKD). However, the relationships among factors influencing the metabolism of ADMA and the CKD progression are not fully understood. METHODS: Serum ADMA, and variables related to the metabolism of ADMA were measured in 181 non-dialysis patients (CKD stages 3-5) and in 46 controls. Patients were assessed at baseline, and 6 and 12 months after the initiation of the study. RESULTS: Patients had increased baseline ADMA, advanced glycation end products (AGE), and advanced oxidation protein products (AOPP) compared with controls (P<0.001). In a total of 164 patients who completed a one-year study, the estimated GFR (eGFR) declined from 23.5 (17.7-36) mL/min/1.73m(2) to 21 (14.7-31.5) (P=0.018), AGE rose from 1.58 (1.38-1.90) µmol/L to 1.76 (1.52-2.21) (P<0.001), while ADMA, AOPP, tubular function, and proteinuria remained stable. In a multiple regression model (adjusted R(2) = 0.49, P<0.0001), the interaction of relatively higher baseline eGFR, i.e. > 25 mL/min/1.73m(2), with higher ADMA (P=0.02) and higher AOPP (P=0.04) predicted the severest decrease in eGFR per year. Other predictors of progression were higher baseline AGE (P<0.001), proteinuria (P=0.003), hypertension (P=0.01), and higher baseline eGFR (P=0.03). CONCLUSION: Elevated ADMA and markers of oxidative stress were strong predictors of progression in patients with eGFR between 25-40 mL/min/1.73m(2) , i.e. at the borderline of CKD stages 3-4.

High ferritin, but not hepcidin, is associated with a poor immune response to an influenza vaccine in hemodialysis patients.

BACKGROUND: The immune response to vaccination in hemodialysis (HD) patients can be influenced by disorders of iron metabolism, iron overload or chronic inflammatory state. Elevated levels of hepcidin are considered a new marker of iron metabolism impairment and anemia of inflammation in HD patients. METHODS: We studied the effects of hepcidin, other markers of iron status and intravenous iron (Fe(iv)) on the response to an influenza vaccine (Influvac(R) subunit 2008/2009) in 40 HD patients. The immune response of HD patients was compared with that of 46 controls without renal disease according to serum antihemagglutinin antibody titer (anti-HA). RESULTS: A total of 31 HD patients (responders) attained seroconversion (at least a 4-fold increase in anti-HA) to at least 1 of 3 vaccine strains; 9 patients (nonresponders) did not respond to any strain. Responders did not differ from nonresponders in hepcidin [99 microg/l (36-200) vs. 97 microg/l (23-216), p = 0.97]. Responders had lower ferritin (571 +/- 291 vs. 821 +/- 309 microg/l, p = 0.031) and were administered higher doses of Fe(iv) within the last 12 weeks prior to vaccination [625 mg (312-625) vs. 312 mg (0-625), p = 0.029]. The seroconversion to A(H1N1), A(H3N2) and B strains was noted in 20, 52 and 40% of HD and in 11, 39 and 48% of controls, respectively (HD vs. controls, p = nonsignificant). The rates of seroprotection (anti-HA > or =40) to vaccine strains in HD (27, 85 and 95%) and controls (24, 96 and 98%) were also comparable. CONCLUSION: Antibody production following influenza vaccination in HD patients may be suppressed by very high ferritin levels. Hepcidin does not correlate with immune response and high levels of hepcidin may reflect its retention in HD patients. Fe(iv) administration was not associated with a poorer immune response. The immunogenicity of the A(H1N1) strain was inadequate in HD patients and controls alike.

Asymmetric dimethylarginine in hemodialysis, hemodiafiltration, and peritoneal dialysis.

Asymmetric dimethylarginine (ADMA) is a mediator of endothelial dysfunction. Production and elimination of ADMA may be affected by the type of renal replacement therapy used and oxidative stress. Plasma ADMA, advanced glycation end products (AGE), and homocysteine were assessed in 59 subjects: 20 hemodialysis (HD) patients, 19 patients undergoing peritoneal dialysis (PD), and 20 controls. Results were compared between the groups. The effect of 8 weeks of HD and high-volume predilution hemodiafiltration (HDF) was compared in a randomized study. HD patients showed higher ADMA (1.20 [0.90-1.39 micromol/L]) compared to controls (0.89 [0.77-0.98], P < 0.01), while ADMA in PD did not differ from controls (0.96 [0.88-1.28]). AGE and homocysteine were highest in HD, lower in PD (P < 0.01 vs. HD), and lowest in controls (P < 0.001 vs. HD and PD). PD patients had higher residual renal function than HD (P < 0.01). The decrease in ADMA at the end of HD (from 1.25 [0.97-1.33] to 0.66 [0.57-0.73], P < 0.001) was comparable to that of HDF. Switching from HD to HDF led to a decrease in predialysis homocysteine level in 8 weeks (P < 0.05), while ADMA and AGE did not change. Increased ADMA levels in patients undergoing HD, as compared to PD, may be caused by higher oxidative stress and lower residual renal function in HD. Other factors, such as diabetes and statin therapy, may also be at play. The decrease in ADMA at the end of HD and HDF is comparable. Switching from HD to HDF decreases in 8 weeks the predialysis levels of homocysteine without affecting ADMA.

Brain natriuretic peptide and N-terminal proBNP in chronic haemodialysis patients.

BACKGROUND: Brain natriuretic peptide (BNP) and N-terminal proBNP (NT-proBNP) are released into circulation as a result of congestive heart failure (HF). As HF and water overload are frequent complications in haemodialysis (HD) patients, we decided to study the levels of BNP and NT-proBNP and their changes during HD. METHODS: BNP and NT-proBNP levels were determined in 94 HD patients before and after a regular 4-h HD. We followed changes in these peptides during HD depending on age, sex, HF (NYHA classification and left ventricular ejection fraction [LVEF]), duration on HD, presence of hypertension, coronary artery disease, type of membrane used for HD [low-flux (LFx) or high-flux (HFx)] and body mass change during HD. Furthermore, patients basic medication and creatinine levels and presence of diabetes mellitus were monitored. RESULTS: Respectively,94% and 100% of the patients had pre-dialysis concentrations of BNP and NT-proBNP above the cut-off values for HF. The marker levels correlated significantly both before and after HD (r = 0.903 and 0.888, respectively, p < 0.001). BNP levels significantly decreased (p < 0.0001), whereas NT-proBNP significantly increased (p < 0.0001) during HD on LFx membranes. HD on HFx membranes caused greater decrease of BNP (compared to LFx membranes, p < 0.001), but also a decrease of NT-proBNP (p < 0.001).We did not find any significant differences in marker levels for HF and non-HF patients (NYHA classification). However, both peptides reached higher levels in the group with LVEF < or = 50% (p < 0.001 for both peptides). Body mass change during HD negatively correlated only with the change of NT-proBNP (r = -0.27, p < 0.05). In the multiple regression model, the change of both peptides during HD was significantly influenced by membrane type (p = 0.003 for BNP and p = 0.001 for NT-proBNP). NT-proBNP change during HD was further significantly influenced by LVEF (p = 0.012), sex (p = 0.002) and duration on HD (p = 0.006). CONCLUSIONS: Both BNP and NT-proBNP levels were significantly increased in HD patients prior to dialysis. The change in concentrations of both peptides during HD is influenced by membrane type. HD probably triggers increased production of both peptides and this increase is emphasized by impaired LVEF. This fact can be clinically observed only on NT-proBNP levels, because BNP levels are biased by significant removal of this protein during HD.

Relationship between hepcidin and ferritin in haemodialysed patients.

BACKGROUND: Hepcidin is a key regulator of iron metabolism. It binds to ferroportin and causes the trapping of iron in cells, rendering it unavailable for erythropoiesis. The synthesis of hepcidin is upregulated by high iron stores and inflammation. Haemodialysed patients suffer from anaemia and impaired iron management, the cause of which is multifactorial. Our aim was to describe the relationship between hepcidin and other parameters of iron metabolism, erythropoiesis, and inflammation. PATIENTS, MATERIALS AND METHODS: Complete blood cell counts, hepcidin, parameters of iron metabolism, and inflammation were measured in samples from 164 dialysed patients and 37 control healthy volunteers. Patients were subdivided according to the time of dialysis session. RESULTS: According to the time of haemodialysis, iron levels showed an insignificant tendency for diurnal variability, whereas hepcidin levels were markedly different. Non-parametric correlations showed a weak, but statistically significant correlation between parameters of iron metabolism and inflammation in the entire group of patients. No correlation was found between hepcidin and other biochemical parameters in controls. Non-parametric correlations were also performed in the time subgroups of patients. CONCLUSION: It seems that the influence of inflammation on hepcidin levels in haemodialysed patients is not crucial and other factors (e.g. hepcidin retention) are involved.

Cell-free plasma DNA during peritoneal dialysis and hemodialysis and in patients with chronic kidney disease.

The mechanisms of clearance of circulating plasma DNA are not fully understood, and so we aimed to examine it in patients with impaired renal function compared with healthy individuals. We also assessed the effect of peritoneal dialysis and hemodialysis on circulating plasma cell-free DNA (cfDNA) in our treated patients. Overall, 20 healthy volunteers, 20 patients with chronic kidney disease (CKD), 18 patients undergoing peritoneal dialysis (PD), and 17 patients on hemodialysis (HD; high-flux polysulfone membrane) were examined. Cell-free DNA levels were determined using real-time GADPH gene sequence amplification. The levels of cfDNA in all groups of our patients did not differ significantly from those of healthy volunteers. In HD patients, cfDNA levels were significantly increased compared with those of CKD patients (P < 0.05) and PD-treated patients (P < 0.01). In PD-treated patients, cfDNA was detectable in overnight effluent, with its levels correlating inversely with the duration of PD treatment (r=-0.619, Spearman's coefficient, P= 0.008). Factors contributing to these differences may include changes in the quality and quantity of the cell population of the peritoneum, highlighting the need for additional studies clarifying the dynamics of cfDNA during PD. The plasma levels of cfDNA do not seem to be dramatically altered even in CKD patients or those on PD or HD (as long as they are measured prior to the procedure in the latter two). Our data suggest renal elimination is not the main mechanism of circulating cfDNA clearance.

The effect of intravenous iron on oxidative stress in hemodialysis patients at various levels of vitamin C.

BACKGROUND/AIMS: Vitamin C levels decrease during hemodialysis (HD), which deteriorates antioxidant defense. Vitamin C may also act pro-oxidatively, via reduction in Fe(III). We sought to determine whether intravenous iron (Fe(iv))-induced oxidative stress differs in HD patients with low and physiological vitamin C levels and whether intravenous vitamin C (C(iv)) administration during HD would change the response to Fe(iv). PATIENTS AND METHODS: Twenty patients with vitamin C deficiency (median 15.7 micromol/l, range 8.0-22.7) received Fe(iv) (100 mg iron sucrose between 150 and 180 min of HD). After 4 weeks of oral supplementation, the levels of vitamin C were comparable with those of controls (60.1 micromol/l, range 47.4-70.9). Patients were subsequently treated with (1) Fe(iv), (2) Fe(iv) and continuous 2 mg/min C(iv) throughout HD, (3) saline (S), and (4) S+C(iv). Plasma thiobarbituric acid reacting substances (TBARS) and vitamin C were assessed before, during and after FE(iv)(S), and 15, 30 and 60 min after infusion. RESULTS: Fe(iv) induced a comparable rise in TBARS in patients with vitamin C deficiency (before Fe(iv), 1.9 micromol/l, range 1.4-1.9; after Fe(iv), 2.6 micromol/l, range 2.3-2.9; p < 0.01) and in those with normal vitamin C (before Fe(iv), 1.9 micromol/l, range 1.7-2.1; after Fe(iv), 2.6 micromol/l, range 2.5-2.9; p < 0.01). Fe(iv)+C(iv) resulted in a greater increase in TBARS (after Fe(iv), 3.1 micromol/l, range 2.8-3.2) compared with Fe(iv) (p < 0.01). CONCLUSION: Iron sucrose-induced oxidative stress is comparable in HD patients with vitamin C deficiency and in those with normal vitamin C. We documented a pro-oxidative effect of vitamin C during Fe(iv)+C(iv) administration.

Asymmetric dimethylarginine, homocysteine and renal function--is there a relation?

The adverse effect of hyperhomocysteinemia on the vascular wall can be partially explained by increasing plasma concentration of asymmetric dimethylarginine (ADMA), a potent inhibitor of nitric oxide synthase. The aim of the study was to compare ADMA and homocysteine levels in three groups of subjects: blood donors with normal homocysteine concentration (group A), patients with hyperhomocysteinemia and normal kidney function (group B) and hemodialysis patients who are known to be hyperhomocysteinemic (group C). Concentrations of homocysteine (enzymatic method), ADMA (enzyme-linked immunoassay) and creatinine (Jaffe method) in EDTA plasma were measured. Plasma ADMA levels were significantly higher in both groups with hyperhomocysteinemia (1.60+/-0.56 micromol/L in group B, 1.81+/-0.57 micromol/L in group C) when compared with those in blood donors (0.82+/-0.29 micromol/L, p<0.001 in both cases). Significant positive correlations were found between concentrations of ADMA and homocysteine (r=0.42, p<0.0001), ADMA and creatinine (r=0.39 p<0.001), homocysteine and creatinine (r=0.69, p<0.0001), age and homocysteine (r=0.47, p<0.001), age and ADMA (r=0.57, p<0.001) and age and creatinine (r=0.37, p<0.001). Increased ADMA concentrations in hyperhomocysteinemic patients were confirmed, but multiple linear regression analysis showed that this significant correlation is only apparent due the dependence of both parameters on age.

Does treatment modality have an impact on anemia in patients with chronic renal failure? Effect of low- and high-flux biocompatible dialysis.

Renal anemia significantly affects the morbidity and mortality of dialysis patients. The aim of the present study was to establish whether the severity of anemia and success of its treatment differs when using high-flux (HF) or low-flux (LF) hemodialysis. Twenty-five patients on long-term hemodialysis with a mean hematocrit (Hct) of 33%, S alb of 36 g/L, and Kt/V urea of 1.5 were divided into Groups X (n = 13) and Y (n = 12) in a prospective randomized crossover study. Group X was initially treated with LF hemodialysis to be followed by HF hemodialysis for 8 weeks each time. Group Y started with 8 week HF hemodialysis and continued, after crossover, with 8 weeks of LF hemodialysis. HF and LF hemodialysis were performed with polysulfone dialyzers F 7HPS and F60S (Fresenius Medical Care, Bad Homburg, Germany). Hct values, examined at 2 week intervals, did not differ significantly between Groups X and Y during 16 weeks of the study at any measuring interval. In another analysis, all results of HF hemodialysis (first 8 weeks of Y + second 8 weeks of X) were pooled as were all data of LF hemodialysis (first 8 weeks of X + second 8 weeks of Y). No significant relationships were noted between duration of treatment with HF hemodialysis and Hct (n = 72, rS = 0.11, p = 0.36) and between duration of LF hemodialysis and Hct (n = 74, rS = 0.02, p = 0.87) in the newly formed groups. The Hct measured during HF hemodialysis and LF hemodialysis did not differ significantly. Analysis of variance did not reveal a relationship between Hct and the HF or LF membrane. As HF and LF hemodialysis differed significantly in Kt/V urea, the variables were adjusted to identical Kt/V urea using analysis of covariance. No relationship between Hct and dialysis membrane permeability was demonstrated even in this case. Use of biocompatible LF and HF membranes in standard hemodialysis modes under conditions of adequate dose of dialysis and the time period studied did not result in different effects on anemia.

Fibrinolysis defect in long-term hemodialysis patients with type 2 diabetes mellitus and its relation to metabolic disorders.

BACKGROUND/AIMS: Patients with chronic renal failure (CRF) secondary to diabetes mellitus show a high incidence of atherosclerosis with its thrombotic complications. Both CRF and type 2 diabetes mellitus (DM2) results in fibrinolysis defects causally related to atherogenesis and thrombogenesis. It is not well known whether or not and, if so, how fibrinolysis is altered in patients with both CRF and DM2. Our study was designed (1) to identify the fibrinolysis defect present in patients with DM2-mediated CRF and treated by long-term hemodialysis (DM2HD), and (2) to establish whether the fibrinolysis defect is related to the metabolic abnormalities observed in CRF or DM2. METHODS: Sixteen DM2HD patients and 23 healthy individuals (HI) had their euglobulin clot lysis time (ECLT), and tissue-type plasminogen activator (tPA) and plasminogen activator inhibitor-1 (PAI-1) activities (act) and concentrations (ag) assessed before and after standard fibrinolytic stimulus (i.v. administration of 0.4 microg/kg BW 1-deamino-8-D-arginine vasopressin, DDAVP) along with metabolic status markers. RESULTS: DDAVP caused a significant shortening of ECLT, rises in tPA act and ag, and a significant decrease in PAI-1 act. PAI-1 ag declined significantly in HI, but not in DM2HD. A comparison of responses to DDAVP revealed the groups differed significantly in the change in PAI-1 ag. Whereas, in HI, PAI-1 ag decreased by 11.8 ng/ml, no decrease was seen in DM2HD (0.0 ng/ml) (p < 0.0001; medians given; unpaired Wilcoxon's test). Stepwise regression analysis showed the change in PAI-1 ag was highly group-specific (DM2HD vs. HI, regression coefficient 21.22; partial correlation 0.58; p < 0.0001) and, also dependent on the serum concentrations of apolipoprotein A-I (-32.41; -0.46; p < 0.01) and homocysteine (0.35; 0.36; p < 0.05). CONCLUSIONS: Patients with type 2 DM and CRF on long-term hemodialysis have a fibrinolysis defect manifesting itself after standard fibrinolytic stimulus by an insufficient decrease in PAI-1 concentrations. The defect is related to decreased serum levels of apolipoprotein A-I and increased serum levels of homocysteine. The defect might be a factor contributing to accelerated atherosclerosis and thrombotic complications in these patients.