Clinical outcomes for maintenance hemodialysis patients using a high-flux (FX60) dialyzer

Abstract

Aims: To investigate the clinical outcomes of maintenance hemodialysis (HD) patients using a high-flux (FX60) dialyzer. Method: Thirty patients undergoing dialysis for at least 2 years with a low-flux dialyzer were switched to the FX60 dialyzer for 3 years. Clinical and biochemical analysis was performed monthly for each patient. The parameters monitored included blood pressure, hemoglobin, albumin, intact parathyroid hormone (iPTH), calcium and phosphorus levels, the adequacy of dialysis (Kt/V), beta2-microglobulin (β2-MG) clearance rate, as well as antihypertensive and erythropoietin (EPO) medications. Results: After 3 years of dialysis with an FX60 dialyzer, the mean arterial blood pressure fell, hemoglobin increased, serum phosphate level decreased, iPTH declined and medication doses decreased. Conclusions: Dialysis with the FX60 dialyzer has a better clinical outcome for rectifying renal anemia, controlling hypertension and lowering serum phosphate levels making it a better choice for long-term HD patients.

• Anemia
• FX60 dialyzer
• hemodialysis
• high-flux hemodialysis (HFHD)
• hypertension

Introduction

Theoretically, a successful hemodialysis (HD) procedure should remove accumulated toxic substances, reverse inflammation and reduce oxidative stress. Characteristics of the HD procedure itself, such as the biocompatibility of the dialyzer, have been documented to be important factors in improving the patients’ prognosis.1 With the advancement of dialyzer technology, the use of biocompatible membranes for dialysis not only causes less HD-induced inflammation, but also achieves better clearance of uremic toxins. High-flux dialyzers have been shown to provide a better clinical outcome; however, the clinical results are contradictory.2 It is important to determine whether different HD modalities alter the outcomes in patients undergoing HD. This study was conducted to better understand the effect of high-flux (FX60) dialyzers on the clinical outcomes of patients undergoing maintenance dialysis.

Materials and methods

Subjects

Thirty patients with a Kt/V (as a measure of dialysis effectiveness) ≥1.2 (per dialysis) from the People’s Hospital of Peking University, between 18 and 70 years old, who underwent maintenance HD for ≥24 months with auto arteriovenous fistula as the blood access, were included in this study. The patients had a dialysis schedule of three sessions per week and 4 h per session. Patients with poor residual renal function (residual urine volume ≤100 mL/d) were selected, whereas those with active infection, obvious inflammation, malignancy, impaired liver function, life expectancy ≤2 years or intolerance to dialysis were excluded.

Methods

All patients underwent HD with the same type machine (Fresenius 4008S or 4008H, Bad Homburg, Germany) during the entire study period. The minimal blood flow rate was ≥200 mL/min and the dialysate flow rate was 500 mL/min. A bicarbonate dialysate (35 mmol/L bicarbonate) was used for all treatments. The dialysate calcium concentration was 1.5 mmol/L, except for two patients for whom a dialysate calcium concentration of 1.25 mmol/L was used. Ultrafiltration was dependent on the volume status of each patient. Ultra-pure water was used, and water for dialysate production was produced by standard central pretreatment and reverse osmosis. Water quality was regularly monitored to ensure high bacteriological standards. The prevailing standards for FX60 HD fluid at the time of the study was a total viable count of <0.01 cfu/mL and endotoxin levels of <0.015 EU/mL. Based on the Hemodialysis (HEMO) study, we used 20 mL/h/mmHg as the cutoff point to differentiate low and high flux.3 Our patients began the study by receiving dialysis on a NIPRO 130G (low-flux, Osaka, Japan) dialyzers for at least two years and were then changed to FX60 (high-flux, Fresenius, Bad Homburg, Germany) dialyzers for three years. The performance and technical data of the NIPRO 130G (Nipro, Osaka, Japan) and FX60 (Fresenius, Bad Homburg, Germany) are summarized in Table 1.4

Table 1. Performance and technical data of the NIPRO 130G and FX60 dialyzers.

Performance data	NIPRO 130G	FX60
KUF, mL/h/mmHg	15.3	46
KoA urea, mL/min	252	276
Creatinine, mL/min	175	182
BUN, mL/min	188	193
Phosphate, mL/min	160	177
Vitamin B12, mL/min	108	135
Technical data
Sterilization mode	γ-ray	Inline steam
Volume, mL	75	74
Membrane material	Triacetyl cellulose	Modified polysulfon (helixone, high-flux)
Wall, μm	40	35
Fiber lumen, μm	200	185
Effective surface area, m^2	1.3	1.4

Clearance for NIPRO 130G and FX 60 at QB = 200 mL/min. Data are based on Fresenius Medical Cares instructions and European standards for hemodialyzers, hemodiafilters, hemofilters, hemoconcentrators and associated blood-tubing systems. KUF, the ultrafiltration coefficient of a dialyzer, ml/h/mmHg; KoA urea, calculated dialyzer mass transfer area coefficient for urea, mL/min.

The seated blood pressures were measured before every dialysis section, and the target blood pressure of pre-dialysis was ≤140/90 mmHg. The dry weights were determined on the basis of the empirical methods, after fluid had been drained out, the patient had no pitting edema, no effusion and no obvious pulmonary crackle at the examination, blood pressure was nearly normal. In addition to antihypertensive drugs, all patients were firstly set with dry weights. Most patients were taking one or two antihypertensive medications (e.g., calcium channel blockers, β-blockers, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers). A calcium channel blocker was administrated to 18 patients, an angiotensin-converting enzyme inhibitor to 5 patients, an angiotensin II receptor blocker to 22 patients, β-blocker to 9 patients and α-blocker to 2 patients. Seven patients received no antihypertensive agent. All antihypertensive agents are indicated as defined daily dose (DDD).5 All patients, except for three, were given phosphate binders (calcium carbonate). Seventeen patients were given oral calcitriol and two patients were given statins. Folic acid was prescribed for all patients. The dose of erythropoietin (epoetin alpha) and iron Ferrum treatment for each patient was based on the target hemoglobin level suggested by the National Kidney Foundation KDOQI (see Table 2). All patients have not received cinacalcet (without this drug in China), parathyroidectomy and ethanol injection to control phosphorus and PTH.

Table 2. Demographic, clinical and biochemical data of the 30 patients undergoing MHD.

Variable	Value
Age, years	54.1 ± 11.4
Male/female	23/7
Duration of hemodialysis, years	4.8 ± 1.8
Diabetes (%)	6 (20)
Hypertension (%)	19 (63.3)
Cardiocerebrovascular disease (%)	9 (30)
Systolic blood pressure, mmHg	147.1 ± 21.0
Diastolic blood pressure, mmHg	87.0 ± 4.1
Serum creatinine, μmol/L	1229 ± 111.7
Serum albumin, g/L	38.8 ± 4.6
Triglycerides, mmol/L	1.63 ± 0.95
Total cholesterol, mmol/L	4.12 ± 2.60
High-density lipoprotein, mmol/L	0.94 ± 0.12
Low-density lipoprotein, mmol/L	1.82 ± 0.56
Corrected serum calcium, mg/dL	8.7 ± 0.9
Serum phosphorus, mmol/L	1.80 ± 0.50
Calcium phosphate product, mg^2/dL^2	53.2 ± 15.3
iPTH, pg/mL	450.6 ± 650.2
Kt/V	1.41 ± 0.33
Hemoglobin, g/L	108.6 ± 14.9
EPO supplement, IU/week	8632 ± 1908.3

Figures are either mean ± SD or numbers with percentage in parentheses.

Blood samples

Routine blood counts and biochemistry tests were performed monthly for each patient undergoing regular HD. To calculate patient’s Kt/V values, pre-dialysis and post-dialysis of serum creatinine, blood urea nitrogen (BUN), serum phosphorus, N-terminal pro-brain natriuretic peptide (NT-proBNP) and β2-MG were measured with one FX60 dialysis or one NIPRO 130G dialysis every 3 months. All samples were collected in tubes recommended by the kit manufacturers, centrifuged immediately at 3500 rpm for 10 min at 4°C, and the supernatant was assayed at once in a central laboratory. Every sample was measured in duplicate.

Data collection

Demographic data regarding age, gender, blood pressure, history of diabetes, hypertension and cardio cerebral vascular disease were collected. Predialysis serum albumin, triglycerides, total cholesterol and total serum calcium and phosphorus levels were determined with an automatic biochemical analyzer (Hitachi 7600–110, Hitachi, Tokyo, Japan). Intact parathyroid hormone (iPTH) was determined using a radioimmunoassay method (DiaSorin Inc., Stillwater, MN), and NT-proBNP was quantified using an electrochemiluminescence immunoassay (Elecsys2010, Roche, Indianapolis, IN). The corrected serum calcium formula used was: total calcium (mg/dL) + 0.8 × (4 − serum albumin [g/dL]).

Statistical analyses

All analyses and calculations were performed using the statistical package SPSS, version 10.0 (SPSS, Inc., Chicago, IL). Data were presented as means ± SD for continuous variables and as proportions for categorical variables. Comparison of data between two groups was performed with unpaired Students t-tests, and categorical variable analysis was performed using chi-square test analysis. p < 0.05 was considered statistically significant.

Results

Patients

A total of 30 patients undergoing maintenance hemodialysis, 23 males and 7 females constituted the study population, the patients’ age were 54.1 ± 11.4 years (mean ± SD), and the duration of hemodialysis were 4.8 ± 1.8 years (mean ± SD). Causes of end-stage renal disease (ESRD) included chronic glomerulonephritis (12 cases, 40%), chronic tubular interstitial nephritis (2 cases, 6.7%), benign arteriolar nephrosclerosis (5 cases, 16.7%), diabetic nephropathy (5 cases, 16.7%), other diseases (2 cases, 6.7%) and unknown causes in 4 (13.3%) cases. Six (20%) patients had diabetes, 6 (20%) had cardiovascular disease and 3 (10%) had cerebrovascular disease. Demographic, clinical and biochemical data are shown in Table 2. Performance and technical data of NIPRO 130G and FX60 dialyzer are shown in Table 1.

Comparison of clinical data after one FX60 dialysis and one NIPRO 130G dialysis (n = 30, mean ± SD)

The minimal blood flow rate was 250 mL/min and the dialysate flow rate was 500 mL/min for 4 hours. Data for one NIPRO 130G treatment and one FX60 treatment are shown in Table 3. Serum creatinine, BUN and serum phosphorus were all decreased compared to the pre-dialysis values after one FX60 dialysis or one NIPRO 130G, NT-proBNP and β2-MG decreased after dialysis with the FX60 but was unchanged after dialysis with the NIPRO 130G dialyzer.

Table 3. Comparison of clinical data for one FX60 dialysis and one NIPRO 130G dialysis (n = 30, means ± SD).

Variable	Serum P, mmol/L	Scr, μmol/L	BUN, mmol/L	β2-MG, mg/L	NT-proBNP, pg/mL
Pre-FX60 HD	1.90 ± 0.46	1031.5 ± 164.2	29.8 ± 5.14	40.8 ± 4.2	6934 ± 7821
Post-FX60 HD	0.92 ± 0.20*	411.3 ± 137.2*	10.1 ± 2.23*	26.5 ± 3.5*	5268 ± 4276*
Clearance (%)	49.2 ± 11.0	63.1 ± 6.9	69.5 ± 5.8	35.9 ± 3.8	29.6 ± 3.2
p	0.021	0.007	0.009	0.025	0.045
Pre-NIPRO130G HD	1.81 ± 0.51	1037 ± 157.0	29.5 ± 4.94	42.6 ± 4.5	8135 ± 7351
Post-NIPRO130G HD	0.93 ± 0.38*	418.2 ± 117.1*	10.8 ± 2.12*	40.8 ± 4.8	7853 ± 6613
Clearance (%)	48.1 ± 17.9	63.7 ± 4.5	68.1 ± 4.8	4.8 ± 1.0	2.1 ± 2.2
p	0.031	0.008	0.008	0.363	0.765

*Pre-HD dialysis compared with post-HD dialysis p < 0.05.

P, phosphorus; SCr, serum creatinine; BUN, blood urea nitrogen; β₂-MG, β2-microglobulin; NT-proBNP, amino terminal B-type natriuretic peptide.

Comparison of the clinical data for patients undergoing pre-low-flux HD (LFHD) (NIPRO 130G dialyzer) and post-LFHD after two years (n = 30, mean ± SD)

Table 4 shows there was no difference between the clinical data of patients undergoing dialysis with the pre-low flux HD (LFHD; NIPRO 130G dialyzer) and post-LFHD after two years (p > 0.05).

Table 4. Comparison of clinical data for patients undergoing pre-LFHD (NIPRO 130G dialyzer) and post-LFHD after two years (n = 30, mean ± SD).

Variable	Pre-LFHD	Post-LFHD	p
TP, g/L	69.6 ± 5.4	72.3 ± 6.1	0.115
Alb, g/L	38.8 ± 4.6	38.6 ± 2.2	0.803
Hb, g/L	108.6 ± 14.9	103.9 ± 18.1	0.386
TC, mmol/L	4.12 ± 2.60	3.83 ± 0.80	0.08
TG, mmol/L	1.63 ± 0.95	1.95 ± 0.99	0.404
HDL, mmol/L	0.94 ± 0.12	0.91 ± 0.27	0.752
LDL, mmol/L	1.82 ± 0.56	2.03 ± 0.55	0.523
P, mmol/L	1.80 ± 0.50	1.93 ± 0.48	0.345
SCr, μmol/L	1229 ± 111.7	1027 ± 147.0	0.354
BUN, mmol/L	28.6 ± 6.1	29.4 ± 4.9	0.605
Corrected serum Ca, mg/dL	8.7 ± 0.9	8.8 ± 0.6	0.604
iPTH, pg/mL	450.6 ± 650.2	525.2 ± 695.6	0.713
Kt/V	1.41 ± 0.33	1.42 ± 0.25	0.226
SF, ng/mL	600.3 ± 297.2	573.6 ± 314.2	0.464
TSAT (%)	36.1 ± 11.4	40.3 ± 15.7	0.229
β2-MG, mg/L	42.7 ± 4.6	40.6 ± 4.5	0.363
SBP, mmHg	147.1 ± 21.0	146.0 ± 12.6	0.267
DBP, mmHg	87.0 ± 4.1	86.2 ± 7.3	0.431
MAP, mmHg	105.2 ± 8.6	106.1 ± 7.4	0.462
UF, kg	3.39 ± 0.38	3.30 ± 0.65	0.351
DWT, Kg	70.9 ± 115	71.5 ± 11.8	0.601
Vit D3, μg/w	2.98 ± 1.87	2.22 ± 1.85	0.789
EPO, IU/w	8632 ± 1908.3	7913.5 ± 3531.2	0.659
Iv Fe, mg/w	55.0 ± 33.4	51.0 ± 30.4	0.64
Anti-hypertensive drugs (DDD)	1.76 ± 1.57	1.78 ± 1.67	0.976

*Pre-LFHD compared with post-LFHD p < 0.05.

HD, hemodialysis; UF, ultrafiltration (Kg); DWT, dry weights (Kg); TSAT, transferrin saturation index (iron/TIBC); SF, serum ferritin; SBP, systolic blood pressure; DBP, diastolic blood pressure; CaP, calcium phosphate product; TC, total cholesterol; TG, triglyceride; P, serum phosphorus; Alb, serum albumin; SCr, serum creatinine; BUN, blood urea nitrogen; HDL, high density lipoprotein; LDL, low density lipoprotein; iPTH, intact parathyroid hormone; β₂-MG, β2-microglobulin; EPO, erythropoietin (u/w); Iv Fe intravenous ferrous (mg/w); MAP, mean arterial pressure.

Comparison of the clinical data for patients undergoing post-high-flux HD (HFHD) (FX60 dialyzer) after one or three years (n = 25, means ± SD) with pre-HFHD (NIPRO 130G dialyzer)

After one year of high-flux (HFHD) dialysis, phosphorus, iPTH, β₂-MG, NT-proBNP (pg/mL) and mean arterial pressure (MAP) decreased, and Hb and left ventricular ejection fraction (LVEF) improved compared with the pre-HFHD values. After three years of HFHD, HB, ALB, phosphorus, iPTH, calcium phosphorus product (CaP), β₂-MG, NT-proBNP(pg/mL), LVEF, and MAP all improved compared to the pre-HFHD values, and the doses of antihypertensive drugs and active vitamin D3 doses decreased (p < 0.05, Table 5). During the third year of HFHD, two patients received renal transplantation, one died of cerebral hemorrhage and two patients went to another dialysis center.

Table 5. Comparison of clinical data for patients undergoing pre-HFHD (FX60 dialyzer) and after one-year HFHD or three-years HFHD (mean ± SD).

Variable	Pre-HFHD	Post-HFHD1	Post-HFHD3	P1	P3
Male/Female	23/7	23/7	20/5
TP, g/L	72.3 ± 6.1	71.3 ± 4.4	69.5 ± 3.8	0.239	0.315
Alb, g/L	38.6 ± 2.2	39.3 ± 1.6	40.6 ± 1.7^#	0.522	0.032
Hb, g/L	103.9 ± 18.1	111.8 ± 11.5*	115.8 ± 7.8^#	0.013	0.011
TC, mmol/L	3.83 ± 0.80	3.64 ± 0.93	4.01 ± 0.73	0.111	0.091
TG, mmol/L	1.95 ± 0.99	1.72 ± 0.61	1.88 ± 1.11	0.291	0.304
HDL, mmol/L	0.91 ± 0.27	0.92 ± 0.28	0.95 ± 0.23	0.337	0.352
LDL, mmol/L	2.03 ± 0.55	2.02 ± 0.57	2.01 ± 0.55	0.523	0.533
P, mmol/L	1.93 ± 0.48	1.70 ± 0.21*	1.65 ± 0.39^#	0.047	0.035
SCr, μmol/L	1027 ± 147.0	1025 ± 161.6	1100 ± 227.3	0.905	0.864
BUN, mmol/L	29.4 ± 4.9	29.9 ± 5.1	28.1 ± 4.4	0.395	0.405
Corrected serum Ca, mg/dL	8.8 ± 0.6	8.9 ± 0.4	8.7 ± 0.4	0.829	0.704
CaP, mg^2/dL^2	53.2 ± 15.3	51.2 ± 11.7	45.4 ± 12.2^#	0.271	0.034
iPTH, pg/mL	525.2 ± 695.6	437.3 ± 417.2*	237.8 ± 174.3^#	0.046	0.023
Kt/V	1.42 ± 0.25	1.45 ± 0.39	1.48 ± 0.19	0.215	0.236
SF, ng/mL	573.6 ± 314.2	536.0 ± 290.7	528.4 ± 219.8	0.061	0.060
TSAT (%)	40.3 ± 15.7	35.3 ± 17.4	26.5 ± 6.4^#	0.064	0.039
NT-proBNP, pg/mL	7933 ± 6531	6724 ± 8721*	5274 ± 5612^#	0.035	0.028
β2-MG, mg/L	40.6 ± 4.5	33.5 ± 3.3*	32.6 ± 3.6^#	0.032	0.030
SBP, mmHg	146.0 ± 12.6	142.5 ± 10.1	147.2 ± 14.1	0.067	0.367
DBP, mmHg	86.2 ± 7.3	84.1 ± 6.1*	82.7 ± 6.7^#	0.043	0.031
MAP, mmHg	106.1 ± 7.4	104.5 ± 7.3*	102.4 ± 7.0^#	0.042	0.032
UF, kg	3.30 ± 0.65	3.29 ± 0.69	3.09 ± 0.59	0.667	0.650
DWT, Kg	71.5 ± 11.8	71.5 ± 12.1	70.5 ± 12.4	0.561	0.519
LVEF, %	60.0 ± 7.3	64.2 ± 8.7*	66.2 ± 5.4^#	0.040	0.030
Vit D3, μg/w	2.22 ± 1.85	2.12 ± 1.61	1.96 ± 1.51^#	0.708	0.042
EPO, IU/w	7913.5 ± 3531.2	7490.4 ± 3933.5	7840.9 ± 3132.8	0.332	0.631
Iv Fe, mg/w	51.0 ± 30.4	47.6 ± 29.8	59.1 ± 34.0	0.618	0.340
Anti-hypertensive drugs (DDD)	1.78 ± 1.67	1.58 ± 1.50	1.01 ± 0.86^#	0.381	0.036

* p < 0.05, P1 is post-HFHD1 compared with pre-HFHD, P3 is post-HFHD3 compared with pre-HFHD. Post-HFHD1 is after one year of HFHD, post-HFHD3 is after three years of HFHD.

HD, hemodialysis; UF, ultrafiltration (Kg); DWT, dry weights (Kg); TSAT, transferrin saturation index (iron/TIBC); SF, serum ferritin; SBP, systolic blood pressure; DBP, diastolic blood pressure; CaP, calcium phosphate product; TC, total cholesterol; TG, triglyceride; P, serum phosphorus; Alb, serum albumin; SCr, serum creatinine; BUN, blood urea nitrogen; HDL, high density lipoprotein; LDL, low density lipoprotein; iPTH, intact parathyroid hormone; β₂-MG, β2-microglobulin; EPO, erythropoietin (u/w); Iv Fe intravenous ferrous (mg/w); MAP, mean arterial pressure; LVEF, left ventricular ejection fraction.

Discussion

Successful HD procedures are important to the prognosis of patients with ESRD, but have side effects that contribute to changes in chronic inflammatory status, anemia, calcium-phosphate levels, hyperparathyroidism and endogenous inhibitors of nitric oxide synthesis.6,7 These factors make the ESRD population at higher risk of developing cardiovascular disease (CVD) and other complications. More efficient or more biocompatible dialyzers may attenuate endothelial dysfunction by removing accumulated toxic substances and elicit lower levels of inflammation, which can lead to more favorable endothelial function and a decreased incidence of future CVD.8–10 Technical developments in dialyzer design have improved the efficiency of therapy, particularly regarding large molecule removal, and have improved the reliability and safety of the therapy. Dialysis innovations that have entered routine clinical use, such as hemodiafiltration or high-flux hemodialysis, are demonstrated improved clinical outcomes.11,12 In this research, we used the FX60 dialyzer, and compared it with the NIPRO-130G, to analyze the outcomes of dialysis in the same patient population. During the three-year FX60 dialyzer follow-up period, β2-MG and NT-proBNP levels decreased significantly (p < 0.05), and MAP and LVEF improved compared to the pre-HFHD period. MAP is an indicator of heart health, high pulse pressure can be a sign of a stiffing of the aorta due to high blood pressure or atherosclerosis. Increased pulse pressure can be a dangerous indicator of risks for serious conditions such as heart disease. Diastolic blood pressures and mean arterial pressures decreased after HFHD may lead to a good outcome of heart function. High-flux hemodialysis with FX60 dialyzer induced an increase in LVEF, due to the improvement of anemia as well as reduced plasma levels of homocysteine and inflammatory cytokines. FX60 dialyzer membranes achieve more β2-microglobulin and phosphorus levels clearance rates which may damage the left ventricular structure and function. All these are factors other than lower MAP, contributing to LVEF improvement. And all of these results benefit the patient by decreasing the risk of CVD.13,14

This study also compared the status of hypertension and the adequacy of blood pressure control. 76.7% patients were receiving one or more antihypertensive agents to control their hypertension when enrolled. 140/90 mmHg as the target of pre-dialysis blood pressure, the percentage of patients taking antihypertensive medication decreased during HFHD treatment compared with LFHD (68% vs. 79%). Control of systolic and diastolic hypertension was achieved in a greater percentage of patients when treated with HFHD, compared with the same patients treated with LFHD. 62% of patients undergoing HFHD had a systolic blood pressure less than 150 mmHg compared with 54% of patients undergoing LFHD, and 77% of patients undergoing HFHD and 67% of patients undergoing LFHD had a diastolic blood pressure less than 90 mmHg. However, the differences in the use of antihypertensive medication did not reach statistical significance. The mean average blood pressure of patients differed significantly after three years of HFHD compared with the pre-HFHD values (102.4 ± 7.0 mmHg and 106.1 ± 7.4 mmHg, respectively). We can conclude that the use of the FX60 dialyzer was associated with better blood pressure control in patients undergoing HD.

A high serum phosphate (P) level is an independent determinant of morbidity and mortality in HD patients, with values of <1.78 mmol/L considered to be the ideal target.15,16 However, therapeutic intervention generally fails in about 50% of patients. Standard 4-hours HD allows the removal of about 800 mg P/session, that is, 2400 mg phosphorus in 1 week. This amount is almost half of the quantity of P derived from a normal protein intake. Therefore, standard intermittent HD is not efficacious if not associated with dietary restriction and P binders. Prevention of hyperphosphatemia is based on the restriction of dietary phosphorus intake and reduction of intestinal P absorption. Both therapeutic approaches have major drawbacks. The use of P binders to decrease absorption may produce side effects caused by an excessive load of aluminum or calcium. A more valid alternative is likely represented by the new generation of calcium and aluminum-free agents, however, these are expensive and patient compliance is rarely achieved for these drugs in China. On the basis of this observation, the role of dialytic removal for the achievement of adequate P levels becomes important. Our findings suggest that FX60-HFHD significantly improved the elimination of P in the long-term therapy and should be considered as an additional treatment option for hyperphosphatemia in dialysis patients. This suggests that the type of dialyzer treatment plays a major role in the control of hyperphosphatemia in uremic patients.

Anemia is another important predictor of mortality and morbidity in patients with end-stage renal disease undergoing hemodialysis.17 In addition, anemia persists in 20% to 30% of cases despite erythropoietin (EPO) treatment. Adequate dialysis can contribute to a correction of anemia via many mechanisms,18,19 including the removal of molecules that may inhibit erythropoiesis. One of the aims of this study, which was based on the idea that the clearance of moderate and high molecular weight erythropoiesis inhibitors, such as β2-MG, parathyroid hormone, NT-proBNP and so on, leads to an improvement of anemia, was to investigate the effect of high-flux dialysis on anemia and EPO requirement (number of erythropoietin doses per week) in patients undergoing HD. Although the EPO doses did not change significantly during high-flux dialysis and low-flux dialysis, the HB levels significantly increased (p < 0.05) during HFHD but no increase during LFHD. This suggests that high-flux HD improved the HB levels.

This research demonstrates that high-flux hemodialysis with FX60 dialyzers may effectively improve β2-MG clearance, decrease the levels of P, iPTH and NT-proBNP, and have a better effect on the control of renal anemia and hypertension. HFHD with the FX60 dialyzer effectively reduced complications such as anemia, hypertension and disorders of phosphorus metabolism in patients undergoing maintenance hemodialysis. These beneficial effects are probably mediated by the improved clearance of moderate and high molecular weight toxins. And these results should be confirmed in an appropriately powered, large, prospective and randomized control trial.

Conclusion

The high-flux FX60 dialyzer significantly alleviated HD patients’ renal anemia, hyperphosphatemia and hypertension even within our three-years study periods. This study suggests that high-flux HD might provide better hypertension control, improve anemia and decrease phosphate levels compared to low-flux HD. For patients undergoing long-term maintenance HD therapy, high-flux HD with the FX60 dialyzer might be a better choice.

Limitations of the study

We wish to point out important limitations of our study. Firstly, the very small size of the study, using a self-control design and the non-randomized nature, the fact that this was not blinded and whether any other potential confounders such as better use of phosphate binders or other practice paradigms had changed between the use of the two filters remain drawbacks. Secondly, due to the nature of our study, which aimed at detecting and quantifying the difference between the two dialyzers, we were not able to evaluate all the clinical significance of improved middle molecule elimination. However, for the first time, the FX60 dialyzer performance over three years was measured.

Declaration of interest

The author has no conflicts of interest to disclose.