PROPUESTA BAR RESTAURANTE

Hemodialysis International 2006; 10: S28–S32

Online hemodiafiltration: 4 years of clinical experience ˜ OZ, Isabel GALLARDO, Evandro VALLADARES, Ramo Rosa MUN ´n SARACHO, Isabel MARTI´NEZ, Julen OCHARAN, Jesu ´s MONTENEGRO Nephrology Department, Hospital de Galdakao, Galdakao-Vizcaya, Spain

Abstract Online hemodiafiltration (online HDF) is a new hemodialysis technique combining convection and diffusion and thus also enabling the purification of large molecules. As yet, only a small number of clinical experiences have been published about the effectiveness and safety of online HDF. We present a prospective and observational study conducted on 31 patients treated with online HDF in our center in the last 4 years. The purpose of the study is to compare the evolution of the following aspects before and after starting online HDF: dose of dialysis, purification of medium-sized/large molecules, inflammation, nutrition, Ca-P metabolism, anemia, and intradialytic complications. Online HDF increased Kt/V to 31.0% (p40.001) and reduced postdialysis b2-M to 66.4% (p40.001). The rest of the parameters analyzed did not vary significantly. During online HDF, episodes of symptomatic hypotension fell by 45% in relation to conventional hemodialysis, and no relevant complication occurred. Online HDF is very useful in patients in whom we need to increase replacement therapy, such as patients with a large body surface, those in whom we suspect a residual syndrome or those who have been receiving dialysis for a long time and for whom we wish to prevent amyloidosis. Online HDF is safe and better tolerated than conventional hemodialysis. Key words: Online hemodiafiltration, renal replacement therapy, adequate dialysis, urea elimination, b2 microglobulin removal

INTRODUCTION Online hemodiafiltration (online HDF) is a new hemodialysis technique combining convection and diffusion. The advantage of online HDF over conventional hemodialysis is that it is better capable of purifying not only small but also medium-sized and large molecules.1–3 Online HDF also appears to provide greater hemodynamic stability than conventional hemodialysis.2,4 It has also been described that the application of online HDF is safe in clinical practice.5–7 However, only a small number of reports have been published on the effectiveness and safety of online HDF till now. Correspondence to: Rosa-Ine´s Mun˜oz Gonza´lez, Nephrology Department, Hospital de Galdakao, Barrio Labeaga s.n., 48960 Galdakano, Vizcaya, Spain. E-mail: rmunoz@hgda. osakidetza.net

S28

We have been using online HDF in our dialysis unit for 4 years. The purpose of this study is to compare the evolution of the following aspects before and after starting online HDF: dose of dialysis, purification of mediumsized/large molecules, inflammation, nutrition, Ca-P metabolism, anemia and intradialytic complications.

PATIENTS AND METHOD We have treated 82 patients with hemodialysis and 55 with peritoneal dialysis in the Hospital de Galdakao’s nephrology department. In the last 4 years, we have used online HDF on 31 patients. The characteristics of these patients are given in Table 1. The criteria applied to indicate this technique instead of conventional dialysis are shown on Table 2. The clinical and analytical follow-up of all the patients analyzed in this study was prospective and protocolized.

r 2006 The Authors. Journal compilation r 2006 International Society for Hemodialysis

Online hemodiafiltration

Table 1 Characteristics of the patients (n = 31) Sex (n): 21 men and 10 women Age at the start of online HDF: mean 60.7  16.1; median 64.1 (30.0–87.7) Time in dialysis prior to online HDF (years): mean 5.97  5.73; median 3.12 (0–20.4) Weight (kg): mean 68.8  15.0; median 67.0 (44.0–100.0) Vascular access (n): Radial internal arteriovenous fistula 17 Brachial arteriovenous fistula 10 Femoral arteriovenous fistula 1 Vascular prosthesis 2 Permanent catheter 1

A blood test was performed every 3 months, including determinations related to the following aspects:    



dose of dialysis: Kt/V (second-generation Daugirdas); purification of medium-sized/large molecules: b2-microglobulin (b2-M) before and after dialysis; inflammation: C-reactive protein (CRP) and albumin; nutrition: normalized protein catabolic rate (nPCR), time-average concentration (TAC), albumin, prealbumin, transferrin, lymphocyte count, urea, and creatinine; Ca-P metabolism: Ca, P, and parathyroid hormone (PTH); and anemia: hemoglobin, hematocrit, ferritin, and saturation rate.

All the undesirable effects occurring during dialysis were systematically recorded in the patient’s case history, as were all the medications, either administered or prescribed. To assess the quality of dialysis provided by online HDF, we compared the values of the analytical parameters presented by these 31 patients before starting the online HDF and after 6 months, 1, 2, 3, and 4 years. To assess the safety of online HDF, we have counted all the events occurring during online HDF sessions. We also compared the number of episodes of symptomatic hypotension during dialysis sessions for 6 months before and after starting the online HDF. Episodes of symptomatic

Table 2 Indications of online HDF [n (%)] Need to increase the dose of dialysis On dialysis for a long time Residual syndrome Poor tolerance Ca/P control

Hemodialysis International 2006; 10: S28–S32

13 (41.9) 10 (32.2) 4 (12.9) 3 (9.6) 1 (3.2)

hypotension were defined as episodes requiring the nursing staff to reduce ultrafiltration and/or administer serum. The online HDF regimen used in all the patients was as follows: Fresenius 4008S monitor, Fresenius HF80 high-flux polysulphone 1.8 m2 steam-sterilized dialyzer (Fresenius Medical Care, Bad Homburg, Germany), a mean dialysis time of 208 min, a mean Qb of 343 mL/min, Qd 800 mL/ min, mean Qi 19.8 L/session, always postdilution; and enoxaparin 40 mg in 48% of the patients, 60 mg in 32.3%, 20 mg in 16.1%, and 10 mg in 3.2%. In no case was dialysis time increased from the previous time in conventional hemodialysis. The introduction of online HDF in our dialysis program required neither more staff nor changes in the way work was organized. In the conventional hemodialysis applied to patients before starting to use online HDF, we used a high-flux polysulfone dialyzer, Qd 500 mL/min and Qb, time, and heparin similar to the online HDF. In the statistical analysis, we used either Student’s t test or the Mann–Whitney’s non-parametric test to compare continuous variables for independent samples and the chi square-test or Fisher’s exact test to compare categorical variables.

RESULTS Of the 31 patients included in our online HDF program, 15 are currently continuing with this treatment. The mean time that patients remained with online HDF was 28.6  14.7 months (median 24; 4–51). The reasons for withdrawal from online HDF were renal transplant (n= 9) and death (n= 7), with none of these cases related to the dialysis technique. The evolution of the analytical parameters since the start of online HDF and for the 4 following years is shown in detail in Table 3. The increase in the mean Kt/V of the patients was statistically significant after the introduction of online HDF without increasing dialysis time, and this difference remained throughout the 4 years of follow-up (before online HDF: 1.26; 6 months after online HDF: 1.47, po0.0001; 4 years after online HDF: 1.54, po0.0001). The postdialysis reduction in b2-M was also statistically significant throughout the follow-up period in relation to the patients’ values with conventional hemodialysis (before online HDF: 17.3 mg/L; 6 months after online HDF: 7.41 mg/L, po0.0001; 4 years after online HDF: 9.94 mg/L, po0.0001). Predialysis b2-M did not change significantly with the changeover to online HDF. The predialysis PCR of the patients fell after the start of online HDF, but without reaching statistical significance

S29

Mun˜oz et al.

Table 3 Evolution of analytical determinations Before online HDF (n = 31) Kt/V Predialysis b2microglobulin (mg/L) Postdialysis b2microglobulin (mg/L) PCR (mg/L) Albumin (g/dL) Prealbumin (mg/dL) n-PCR (g/Kg/day) TAC (mg/dL) Transferrin (mg/dL) Lymphocytes (thousands/mm3) Urea (mg/dL) Creatinine (mg/dL) Ca (mg/dL) P (mg/dL) PTH (pg/mL) Hematocrit (%) Hemoglobin (g/dL) Ferritin (ng/mL) Saturation rate (%)

6 months with online HDF (n= 31)

1 year with online HDF (n= 29)

2 years with online HDF (n= 14)

3 years with online HDF (n =12)

4 years with online HDF (n =9)

1.26  0.21 27.2  10.6

1.47  0.19 22.8  7.38

1.58  0.145 22.4  5.36

1.61  0.17 23.1  7.26

1.65  0.235 1.54  0.13 25.6  8.02 30.7  8.91

17.3  7.9

7.41  3.07

7.27  3.07

5.82  2.31

7.38  2.59

9.94  3.66

11.2  9.69 10.3  9 11.6  13 7.83  9.46 8.05  7.9 6.07  10 3.92  0.33 3.85  0.249 3.91  0.298 3.95  0.303 3.94  0.297 4.14  0.255 34.3  7.58 31.2  8.96 30.3  6.68 28.3  5.54 27.3  5.3 32.2  6.65 1.26  0.29 1.3  0.32 1.32  0.29 1.37  0.338 1.18  0.334 1.25  0.215 47.8  11 44.8  12.9 40.7  12.2 45.8  10.8 40.6  9.89 45.6  7.28 186.7  31.8 189  33.2 185  26.3 176.7  27.4 168.4  49.6 185.6  35.4 1594.8  674 1389.3  552.8 1241.3  535.4 1309.5  608.4 1310.2  637.8 1093  414.3 149.9  34.5 147  42.3 143.2  31.8 9.55  1.63 9.29  1.67 9.21  1.41 9.94  0.61 9.85  0.68 9.65  0.507 5.38  1.6 5.23  1.78 5.27  1.22 412  393.7 365.4  321 448.1  326.2 36.1  3.7 35.6  2.78 37.2  2.79 12  1.26 11.8  1.01 12.3  0.977 304.3  184.9 321.2  196.9 288.4  183.1 27.6  9.72 28.6  10.4 31.2  12.1

150.9  36.3 132.8  39.6 145  26.1 9.61  1.47 9.56  1.67 10.8  1.71 9.32  0.611 9.33  0.47 9.48  0.698 5.67  2.27 5.32  1.85 5.54  1.51 551.7  337.7 590.2  484.7 255.3  250.7 37.2  2.65 36.8  2.34 36.8  2.16 12.1  0.669 12.2  0.861 11.8  0.577 350.4  140.3 212.6  63.6 198.4  98.1 35.4  13.3 37  11.8 36.9  14.1

po0.0001 (value before online HDF vs. during online HDF).

(before online HDF: 11.2; 6 months after online HDF: 10.3, po0.513; 4 years after online HDF: 6.07, po0.197). It is worth mentioning the great variability in the PCR value between the different patients and in each patient over time. Time-average concentration and nPCR values remained within normal ranges (nPCR: 41.2 mg/dL; TAC: o50 g/ kg/day) throughout the follow-up period. The rest of the analytical nutritional values remained stable during the follow-up period in relation to the period before starting online HDF. The analytical parameters related to Ca-P metabolism and hematological values did not change significantly after the start of online HDF (Table 3). A parathyroidectomy was performed on 6 patients whereas they were receiving treatment with online HDF; no differences were found between the 2 sub-groups (with or without surgery) in the evolution of Ca-P metabolism parameters related to the change to online HDF.

S30

After introducing online HDF, there was no reduction either in the number or dose of medication required by the patients to treat Ca-P metabolism changes or anemia. The number of episodes of symptomatic hypotension in the 6 months before and after the online HDF fell by 45%, before online HDF, 62 (2.67% of 2,325 dialysis sessions); after online HDF, 28 (1.20% of 2,325 dialysis sessions). The dialyzer coagulated once in 29% of the patients. The repetition of this complication was prevented in all cases by increasing the blood flow and the dose of heparin, and in 4 patients by washing out the hemodialysis circuit once an hour with physiological serum. No other technical complications arose. There was no reaction to pyrogens or other undesirable effects attributable to the online HDF technique.

DISCUSSION The quality of the dialysis provided by online HDF is considerable and unquestionably represents an improve-

Hemodialysis International 2006; 10: S28–S32

Online hemodiafiltration

ment over conventional hemodialysis (p40.0001); in our patients, it increased the Kt/V to 31.0% (p40.0001) and reduced postdialysis b2-M to 66.4% (p40.0001). Furthermore, this improved dialysis quality remained constant during the 4 years of follow-up. The Kt/V values found, from 1.47 to 1.65, are similar to those described in other studies.1,8,3 Our results confirm that online HDF is highly effective for the purification of molecules of all sizes, from small molecules like urea3,8–10 to large molecules such as b2-M.11–13 Not many experiences with online HDF have been published, and only a few series with a number of patients and follow-up period like ours.3,8,14–18 The clinical and analytical follow-up of all the patients analyzed in this study was prospective and protocolized, and identical to the follow-up systematically applied to our other patients receiving hemodialysis. However, our patients were treated under the normal conditions of a hemodialysis unit, and not in an experimental context. This implies bias (of selection, for instance), but confirms the effectiveness shown by online HDF in clinical trials.3,8,14,18 All the patients included under the study were clinically complex cases, and conventional hemodialysis did not provide the dose and quality of dialysis required in most cases. It has been described that, unlike during conventional hemodialysis, the PCR falls during online HDF.18 We only analyzed the PCR before dialysis, which fell discretely after the start of online HDF, although there was considerable interpatient and intrapatient variability and it was never statistically significant. It is important to note that in patients receiving treatment with online HDF, there is no deterioration in the nutritional parameters, although this technique increases albumin loss compared with conventional hemodialysis. In this respect, the improvement in the physical conditions of patients dialysed with online HDF may be related to the loss of major uremic toxins, with albumin taken as an example of their size.19 The tolerable amount of albumin loss is unknown and the response may be the same as in peritoneal dialysis. In our and other authors’ experience,3,8 online HDF does not diminish predialysis serum levels of P, although the opposite is described in other studies.13,17 Likewise, the anemia in our online HDF patients did not improve and we were unable to reduce the erythropoietin doses that they were previously receiving. Something similar was found in some other series,8,18 whereas others described that online HDF reduces anemia.3,15,16 All these discrepancies regarding the effect of online HDF may be because of the limited number of patients included in the

Hemodialysis International 2006; 10: S28–S32

studies and/or their baseline characteristics or limited follow-up times. Online HDF was very well tolerated by all our patients for the 4 years of follow-up. There was no reaction to pyrogens and the online HDF was not suspended for undesirable effects in any of the patients; indeed, symptomatic hypotension episodes fell by 45% compared with conventional hemodialysis. The good tolerance and safety of the technique have been described in all the series in which they have been analyzed.2,4 There is greater procoagulant activity in convective hemodialysis techniques.20 In online HDF, infusion flow is limited by blood flow, transmembrane pressure, ultrafiltration, hematocrit and dialysis time.3 Therefore, in cases in which the UF and hematocrit are very high, the TMP increases, thus favoring circuit coagulation. The circuit coagulated in 9 (29%) of our patients, but the repetition of this complication was easily prevented by increasing the blood flow or dose of heparin or washing out the circuit every hour with physiological serum. We can conclude by confirming that, despite the benefits of online HDF, its introduction into dialysis programs does not involve additional workload or organizational changes. Online HDF is safe, more effective, and better tolerated than conventional hemodialysis. The benefits of online HDF remain over time. Online HDF is especially indicated in patients for whom it is necessary to increase replacement therapy (if they have a large body surface, for instance), a residual syndrome is suspected,21 or to prevent the development of amyloidosis in patients who have been receiving dialysis for a long time.

REFERENCES 1 Canaud B, Bosc JY, Leray H, et al. On-line haemodiafiltration: State of the art. Nephrol Dial Transp. 1998; 13(Suppl 5):3–11. 2 Pizzarelli F, Cerrai T, Dattolo P, Tetta C, Maggiore Q. Convective treatments with on-line production of replacement fluid: A clinical experience lasting 6 years. Nephrol Dial Transp. 1998; 13:363–369. 3 Maduell F, del Pozo C, Garcia H, et al. Change from conventional haemodiafiltration to on-line haemodiafiltration. Nephrol Dial Transp. 1999; 14:1202–1207. 4 Donauer J, Schweiger C, Rumberger B, Krumme B, Bohler J. Reduction of hypotensive effects during online-haemodiafiltration and low temperature haemodialysis. Nephrol Dial Transp. 2003; 18:1616–1622.

S31

Mun˜oz et al.

5 Pizzarelli F, Maggiore Q. Clinical perspectives of on-line haemodiafiltration. Nephrol Dial Trasnplant. 1998; 13(Suppl 5):34–37. 6 Canaud B, Bosc JY, Leray-Moragues H, et al. On-line haemodiafiltration. Safety and efficacy in long-term clinical practice. Nephrol Dial Transp. 2000; 15(Suppl 1): 60–67. 7 Spalding E, Farrington K. Haemodiafiltration: Current status. Nephron Clin Pract. 2003; 93:87–96. 8 Ward R, Schimdt B, Hullin J, Hillebrand G, Samtleben W. A comparison of online hemodiafiltration and high-flux hemodialysis: A prospective clinical study. J Am Soc Nephrol. 2000; 11:2344–2350. 9 Ding F, Ahrenholz P, Winkler RE, et al. Online hemodiafiltration versus acetate-free biofiltration: A prospective crossover study. Artif Organs. 2002; 26:169–180. 10 Wizemann V, Kulz M, Techert F, Nederlof B. Efficacy of haemodiafiltration. Nephrol Dial Transp. 2001; 16(Suppl 4):27–30. 11 Maduell F, Navarro V, Cruz MC, et al. Osteocalcin and myoglobin removal in on-line hemodiafiltration versus low and high flux hemodialysis. Am J Kidney Dis. 2002; 40:582–589. 12 Lornoy W, Because I, Billiouw JM, Sierens L, Van Malderen P, Haenens P. On-line haemodiafiltration. Remarkable removal of beta2-microgrolbulin. Long-term clinical observations. Nephrol Dial Transp. 2000; 15(Suppl 1):49–54. 13 Lornoy W, Because I, Billiouw JM, Sierens L, Malderen P. Remarkable removal of beta2microglobulin by on-line hemodiafiltration. Am J Nephrol. 1998; 18:105–108.

S32

14 Lin CL, Yang CW, Chiang CC, Chang CT, Huang CC. Long-term on-line hemodiafiltration reduces predialysis beta-2-microglobulin levels in chronic hemodialysis patients. Blood Purif. 2001; 19:301–307. 15 Lin CL, Huang CC, Yu CC, et al. Improved iron utilization and reduced erythropoietin resistance by on-line hemodiafiltration. Blood Purif. 2002; 20:349–356. 16 Bonforte G, Grillo P, Zerbi S, Surian M. Improvement of anemia in hemodialysis patients treated by hemodiafiltration with high-volume on-line prepared substitution fluid. Blood Purif. 2002; 20:357–363. 17 Zenhder C, Gutzwiller JP, Renggli K. Hemodiafiltration – a new treatment option for hyperphosphatemia in hemodialysis patients. Clin Nephrol. 1999; 3:152–159. 18 Wizeman V, Lotz C, Techert F, Uthoff S. On-line haemodiafiltration versus low-flux hemodialysis. A prospective randomized study. Nephrol Dial Transplant. 2000; 15(Suppl 1):43–48. 19 Krieter D, Falkenhain S, Chalabi L, Collins G, Lemke H, Canaud B. Clinical cross-over comparison of mid-dilution hemodiafiltration using a novel dialyzer concept and post-dilution hemodiafiltration. Kidney Int. 2005; 67: 349–356. 20 Klingel R, Schaefer M, Schwarting A, et al. Comparative analysis of procoagulatory activity of haemodialysis, haemofiltration and haemodiafiltration with a polysulfone membrane (APS) and with different modes of enoxaparin anticoagulation. Nephrol Dial Transp. 2004; 19:164–170. 21 Depner Uremic toxicity: urea and beyond. Semin Dial. 2001; 14:246–251.

Hemodialysis International 2006; 10: S28–S32