Clinical Evaluation of a New Dialyzer, FLX-12 GW, with a Polyester-Polymer Alloy Membrane

Artific i r / / U r g ~ r . c 17(5):339-345. Blackwell Scientific PublicationL, Inc., Boston G 1993 lntcrnational Soclety for Artificial Organs

Clinical Evaluation of a New Dialyzer, FLX-12 GW, with a Polyester-Polymer Alloy Membrane

G. Stein, K. Gunther, H. Sperschneider, H. Carlsohn, M. Huller, K. Schubert, and R. Schaller

Depriv~inrnts of Internal Medicine rind Rmdiology, Institrite of Clinical Chemistry, Uniuersity of Jencr, Jenu, Germany

Abstract: The performance of a membrane in renal failure therapy is determined by its structure, its overall mass transfer properties, and its blood compatibility. In this regard. removal of P2-microglobulin (P2M) has become a major objective of dialysis therapy. In the present study, a newly developed high-flux membrane composed of a polyester-polymer alloy (PEPA) with the components of polyarylate and polyethersulfone (dialyzer FLX- 12 GW; Nikkiw Co.. Japan) has been evaluated with regard to both biocompatibility and elimination capacity for P2M during hemodialysis of 8 stable chronic uremic patients. The clearance values of low molecular weight solutes were in the came range as those reported for high-flux dialyzers of comparable surface area. There was no drop in leukocyte counts and only a minimal fall in platelet counts

nearly in the same range as has been observed by other investigators using polyamide membrane. C3a Des Arg generation was low, and C5a Des Arg formation was not significantly influenced. There was a sharp drop in the serum P2M level ( -35%) during dialysis with a clearance between 59.7 t 5.6 mlimin ((2, 200 ml/min) and 70.1 t 9.7 mlimin (Qe 300 mlimin), respectively. Accordingly, the sieving coefficient was calculated to be 0.2 at 30 min after start of dialysis and 0.6 I h later. The membrane was able to remove 184.0 ? 22.3 mg/4 h due t o an apparent rate of adsorption during the first hour of treatment in combination with high transmembrane transfer in the following time. Key Words: High-flux dialysis membrane-Polyester-polymer alloy-Biocompatibility-& Microglobulin removal.

During the last years, progress in membrane development has contributed to the continuous improvement of the extracorporeal treatment of renal failure. The performance of a membrane in renal failure therapy is determined by its structure, its overall mass transfer properties, and its blood compatibility. Synthetic structures offer a wide range of possibilities and have the potential for further tailoring of the membrane transport properties and blood compatibility to be optimized for the different processes in renal failure therapy (1-4). In the late 1980s, the removal of higher molecular weight substances such as P,-microglobulin (P2M), the amyloidogenic precursor of AB-amyloidosis, became clinically desirable (5-7).

~

Received October 1992; revired November 1992 Addre\\ correspondence and reprint requests to Prof Dr G .

Stein at Klinik fur Innere Medizin, Erldnger Allee 101, 0-6902 Jena, Germany

In uremic patients, plasma concentration in- creases progressively with decreasing renal function (8- 10); in patients undergoing intermittent hemodialysis treatment, plasma P2M levels are elevated up to 60-fold (9,lI-13). Dialysis-related amyloidosis appears to be largely derived from circulating P2M. This suggests that amyloid formation and deposition may be in part a function of the duration of the markedly increased serum and tissue levels of P2M. The successful removal of P2M may prevent the formation and deposition of new amyloid fibrils.

High-flux membranes such as hydrophilic polyamide, polysulfone, and acrylonitrile have P2M sieving coefficients of up to 0.6 and thereby allow greater removal compared with standard low-flux membranes (7,14-16). The present investigation was conducted to evaluate the performance, efficacy, biocompatibility, and P2M removal capacity of a recently made new synthetic membrane of polyester-polymer alloy (PEPA).

33 9

340 G. STEIN ET A L .

MATERIALS AND METHODS

Patients Eight stable patients with chronic uremia, mean

age 53 years (range 36-66 years) undergoing long- term hemodialyri5 for 27.5 months (range 12-94 months), were studied. The urinary output was less than 500 m1/24 h. All gave informed consent before inclusion in the study.

Dialysis Hemodialysis was normally performed 3 times a

week for 4 to 5 h using low-flux Cuprophan membranes. For study purposes, the dialyzer FLX-12 GW (Nikkiso Co., Japan) was used. The membrane was composed of a polyester-polymer alloy (PEPA) with the components of polyarylate and polyethersulfone. The membrane thickness was 30 pm, fiber inner diameter 210 pm, effective surface area I .2 I$. Bicarbonate was alw;iys used as the dialysis fluid. At the beginning of dialysis, clearance values were measured with blood flow rates of 200, 250, and 300 mlimin, respectively. Then the blood flow was kept constant at 300 mlimin during the following treatment time.

Sampling procedures Whole-blood samples were drawn from the pa-

tients’ artcriovenous fistulae after heparinization before and during hemodialysis. All samples were anticoagulated with EDTA and centrifuged im- mediately. I’la5ma samples were stored at -20°C until examination.

The ultrafiltration rate was calculated according to the weight gain between dialysis and the general condition and blood pressure at the beginning of treatment.

Dialysate samples were collected at hourly inter- vals starting 1 h after treatment was begun. The sam- plcs were also stored at -20°C until processing.

Analytical procedures Serum urea. crcatininc, uric acid, and phosphate

lcvcls were determined by Autoanalyzer (Beckman Instrurnents, Germany). Hemoglobin, hematocrit (Hct), leukocytes, and platelets were counted in whole-blood samples, anticoagulated with EDTA, by an electronic counter (Baker System 9000, Ser- ono. Germany). Plasma levcls of the complement fragments C3a Des Arg and C5a Des Arg were measured by means of radioimmunoassays (Amersham, U.K.); serum proteins, albumin, retinol binding protein (RBP), C-reactive protein (CRP), a,-microglobulin (a1 M), and &-microglobulin (P2M) levels were

determined in plasma and dialysate samples by applying an enzyme-linked immunoassay (Boehring, Germany). All values were corrected for extracellu- lar volume concentration by the following equation:

Hct outlet - Hct inlet Hct inlet

Correction factor = 1 -

The dialyzer clearance (CI) was calculated on the “blood side” of the dialyzer using the formula

C Pi - C Po C Pi X Q B ? Cl, =

where C = concentration, P = plasma, Qj = blood flow, i = dialyzer inlet, o = dialyzer outlet.

Statistics Results are expressed as mean * standard error

of the means. The significance of the differences was assessed by the paired t test. Significancc was defined as a p value < 0.05.

RESULTS

Dialyzer clearance values I5 min after beginning of dialysis, determined from plasma concentrations for urea, creatinine, uric acid, and phosphate with blood flow rates ( Q B ) between 200 and 300 ml/niin, Q,500 ml/min, and mean QF (filtration rate) of 34.2 -C 2.6 ml/(h . mm Hg) are shown in Table 1 . There was a significant blood-flow-related increase of these clearance values. N o difference exists between the beginning and after 4 h of dialysis (Fig. I ) . The total eliminated amount in the dialysate was for urea 61 1 * 42 nimo1/4 h, creatinine 19 -C 1 mmol/ 4 h, uric acid 7 ? 0.5 mmoli4 h, and phosphate 47 2 5 mmol/4 h hemodialysis. The serum reduction rate after 4 h of hemodialysis was 59% for urea, S3% for creatinine, 64% for uric acid, and 32%for phosphate.

The mean predialysis plasma P2M concentration was 32.0 2 5.2 mg/L; the acute effect of a4-h dialysis session led to a postdialysis plasma P2M value of 22.2 2 2.9 mg/L. The sieving coefficient of P2M (isolated ultrafiltration) was 0.15 after 30 min and increased to approximately 0.6 at 2 h . At 15 min after the beginning of hemodialysis, the P2M clearance estimated from plasma values was 59.7 +- 5.6 ml/ min (QA 200 mllmin, Qu SO0 ml/min) and 70.1 t 9.7 mlimin (Q8 300 ml/min, or, 500 ml/min), respectively, but only 2.0 2 0.54 mlimin (QU 300 ml/min) was estimated from dialysate values. At the end of the 4-h hemodialysis session, there was a fall of blood-side P2M clearance to 20.0 -+ 8.4 rnlirnin ((II,

HIGH-FLUX DIALYSIS MEMBRANE 34 I

TABLE 1. Cleurance (mlltnin) of' iirea, creatinine, uric acid, and phospkate

200

QR (mlimin)

250 300

Urea 165.8 ? 0.5 193.0 ? 3.2" 203.3 ? 12.6" Creatinine 134.9 5 2.1 156.6 c 4.7" 180.8 _t 15.1" Uric acid 140.0 ? 2.1 163.8 5 3.7" 170.5 5 10.7' Phosphate 129.8 t 6.4 161.7 ? 6.2" 163.8 ? 14.5

I' p < 0.005 I1 vs.1. " p > 0.025 111 vz .1 . QB 200-300 mlimin; Q,, 500 mlimin; or (filtration rate) 34.2 ? 2.6 ml Ih mrn Hg); meiln i SEM.

I Dialyzer clearance (plasma)- beginning and end of HD

250 FIG. 1. Dialyzer clearance of urea, creatinine, uric acid, and phosphate at the beginning and end of dialysis treatment estimated from plasma values (mean ? SEM).

- ' 4 2 too

50

0

\

I

Urea

300 mlimin, Q, 500 mlimin) whereas dialysate-side clearance increased to 22.2 ? 2.1 mlimin.

The serum reduction rate of P2M after 4 h of hemodialysis was 35.4 ? 2.6%. The amount of P2M in the hourly collected dialysate was 22.3 2 4.2 mg in the first hour, 31.3 f 5.2 mg in the second hour, 31.1 -1: 3.2 in the third hour, and 34.6 f 3.0 in the fourth hour for a total of 126.1 f 4.3. This shows clearly that the P2M removal during the first hour of dialysis was due to a high adsorption capacity whereas the removal in the following time was due to a high permeability of the membrane. The total amount of P2M adsorbed by the membrane, determined by integration of the mass balance error over the treatment time, was 77.3 ? 15.9 mg. Therefore, the total amount of P2M adsorbed by the membrane and the total amount filtered were added, and the overall removal with this membrane amounted to 184.0 -+ 22.3 mg. This means that 37.4 ?C 4.0% ha5 been adsorbed.

At 15 min after beginning of dialysis, no leukope- nia WBS found (Fig. 2), but a slight fall in platelet cell count of 17% was observed (Fig. 3). As for C3a Des Arg, a slight increase (Fig. 4) was found, but no significant change in C5aDes Arg formation could be observed (Fig. 5 ) .

Meawrements at 15 min in the arterial and venous

Begin of HD I 0 After 4hours of HD

Phosphate Uric acid Creatinine

lines showed no significant difference of hematocrit, erythrocyte, and platelet counts. There was a slight generation of C3a Des Arg and C5a Des Arg. After 4 h of hemodialysis, no significant difference in serum level of vitamin B,, (1.5 f 1.3%) and CRP (1.6 * 3.2%) was seen. There was an increase in mean RBP plasma level (15.1 -+ 7.2%) and a decrease in a l M (10.4 * 2.4%) and albumin (6.8 f 3.4%) level as compared with predialysis values.

DISCUSSION

The present study was conducted to evaluate both the performance and the biocompatibility profile including the capacity to remove P2M of a newly developed high-flux membrane composed of polyarylate and polyethersulfone. The results concerning clearance values and percentage reduction in the serum clearly show that the performance of the FLX-12 GW dialyzer used in the high-flux mode is within the same range of clearance of low molecular weight solutes as that reported for high-flux and standard dialyzers of comparable surface area (17,18).

The drop in leukocyte counts that can be observed with cellulosic membranes is not found with this PEPA membrane as has also been described with polyamide and polysulfone membranes (1 9-21).

342

7 -

6 -. c

.... 5..

c Q

+%

A

4 V)

&

$ 3 - .

2 z - -

0 s

J I

0 7

G . STEIN ET AL.

* t +

..

-. I

* I i 1

o l : : : . : - : : : : . : ~ : : : ~ : : ~ : - : : : ~ ~ ~ . : ~ : " 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240

Time on dialysis [rnin] (* Sign. p< 0.05 vs t ime 0) FIG. 3. Platelet cell count during hemodialysis with PEPA (n = 8) (mean +- SEM).

-0 m 150 f c?

I 100

0 I 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240

Time on dialysis (min] ( * Sign. p< 0,05 vs time 0)

FIG. 4. Plasma levels of C3a Des Arg during hemodialysis. Values were obtained from the arterial side

HIGH-FLUX DIALYSIS MEMBRANE 343

0 4 L I

Time on dialysis [min] 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240

FIG. 5. Plasma levels of C5a Des Arg during hemodialysis. Values were obtained from the arterial side

There was a minimal fall in platelet counts nearly in the same range as has been observed by other investigators using polyamide membrane (19,21).

Complement activation was low as assessed by measuring complement anaphylatoxin C3a Des Arg and CSa Des Arg derivatives in plasma and determin- ing their generation by estimating the arteriovenous differences. The same results have been reported by other groups using polyamide and polysulfone membranes (2,15,22) including measurement of ter- minal complement activation during in vivo and in vitro contact of blood with different membranes.

The development of osteoarticular amyloidosis in long-term hemodialysis patients has been recog- nized as one of the major complications of long-term renal replacement therapy (5,6,12,23). Although P2M plasma levels do not correlate with the pres- ence or absence of this disease (8,24) and it is unclear whether the reduction of serum P2M has any influence o n the regression or possible prevention of P2M amyloidosis, it would seem reasonable to attempt to lower the precursor amyloidogenic protein to as low a level as possible (25). Therefore, effective therapeutic or preventive strategies are needed for dialysis-related amyloidosis.

Until now, only a successful renal transplantation has been shown to result in rapid improvement of amyloid-associated symptoms like humeroscapular periarthritis due to suppression of the inflammatory reaction rather than due to a regression of the depos- its per se (26). A change of treatment from using a

low-flux to high-flux membrane is unlikely to have a substantial role in the therapy of an established amyloidosis. High-flux membranes such as hydrophilic polyamide, polysulfone, and acrylonitrile have P2M sieving coefficients of up to 0.6 and thereby allow agreater removal compared with standard low-flux membranes (7,27,28). Hemodialysis with high-flux membranes will result in a P2M removal ranging from 100 to 200 mg with polyacryloni- trile dialyzers (5) and up to 260 mg for polysulfone membranes with different surface areas of I .25 m7 (15,20). The polyamide membrane Polyflux 130 was able to remove 235 2 I I mg P2Mi4 h (15).

Besides the membrane permeability, removal of P2M largely depends on convective fluxes because of its molecular weight ( 1 1.800 kilodaltons) (14,29-31). Diffusive clearance of P2M will be less important provided a high blood flow and conse- quently high ultrafiltration rates are achieved. Re- cent results clearly demonstrated that hemofiltration using a polyamide hemofilter removed significantly more P2M than Cuprophan hemodialysis (3). The serum concentration of P2M was reduced by more than 70% with a negative mass balance of almost 300 mg of P2M during a single hemofiltration treatment with hydrophitic polyamide membrane.

Removal of P2M may be due to transmembrane transfer, membrane adsorption, or both (15,32,33). The overall removal rate of P2M using the new PEPA membrane was estimated from the P2M con- tent of the dialysate collected during the 4 h dialysis

Avtf Orgcitia. Vol. 17, N o . 5 , 1993

344 G. STEIN ET AL.

session. The P2M concentration in the dialysate was much lower than expected, which suggests a marked binding to the dialysis membrane. The P2M amount ofthe dialysate may underestimate the total removal of /32M by 2s-30% as estimated from the clearance valucs and arteriovenous difference of plasma concentration. Therefore, we calculatcd that at least 80 mg of ki2M has been adsorbed during the treatment time. The apparent rate of adsorption was highest during the first hour of treatment and decreased by virtually zero during the last pcriod of treatment. This correlated very well with the sieving coefficient estimated 30 rnin after start of dialysis as 0.2 and after I h a s 0.6. Similar results have been obtained using acrylonitrile membrane (14,291.

Dcpending on the membrane uscd, Charnard et al. were able to remove 350-350 mg of P2M during hernofiltration with 20 L of exchanged volume (6). The average total net mcmbrane adsorption of the AN 69 has been roughly estimated as 124 rng; for F 60, Sh mg; for Duoflux, 39 rng; and for FH 77, 34 mg; the amounts recovered in the dialysate were in a higher range (7,27). Sufficient removal may bc defined as an amount of P2M that will exceed daily or weekly synthcsis, which has been estimated in normal control subjects as about 1,300 to 2,100 rng per wcck (34). This has not yet been achieved either by membrane used as a high-flux dialyzcr or by ii

hemofiltration system. Thus, it is clear that prospec- tive studies and more reliable clinical amyloid mark- ers will be necessary to evaluate the issue of prevcn- t ion of dial y s i s- re I at ed am y I oi dos i s with high- fl u x membranes or high-flux dialyzers (27,3S).

In conclusion, i t has becn demonstrated that the ncw PEPA membrane provides high ultrafiltration and diffusive permeability for low molecular weight substances. Its use allows the effectivc elimination of a wide range of uremic toxins. This membrane has remarkably good biocompatibility charactcristics a s documented by the low rate of production ofcomple- mcnt split products, C3a Des Arg and CSa Des Arg, and stable blood cell counts in the course of treatment. Since the membrane charactcristics that de- termine 8 2 M removal are also important for biocompatibility properties, this PEPA membrane may become a sccond objective.

REFERENCES

I . Engbei-s GH. Fcijen J . Current techniques to improve the hloocl compalihility of hiomatci-ial surlaces. /rit./Arrtf'Orgrrrr.\ l9Yl:14:199-2l5.

2 . Deppisch K, Bet/ M . Hiinach GM, Rauterberg E-W, RitL E. Bioconipatihility of the polyamidc membranes. Coiitrih Nrplir( 1 / I 9 92 : 96: 26-46.

3. Kaiser J P . Hagemann 1 . von Herrath. D. Schaefer K . Differ-

ent handling of p?-rnicroglobulin during hernodialysis and he- mofilt ration, Ncdirori I988 :48: 132-5.

4.

5 .

6.

7.

8.

9.

10.

1 1 .

12.

13.

14.

15.

16.

17.

18.

19.

2 0 .

21.

22.

23.

24,

Stein G , Schaucr S , Suss J , Muller A, Huller M. Schaefer K , Falkenhagen D. I h s s W. Influence of membranes on generation of p2M and release of leukocyte lysosomal en- zymes. In/ J Art!f 'Org~rns 1990:13:359-64. Bardin T, Zingraff J , KuntL D, Drueke T. Dialysis-related amyloidosis. N q h d D i d Trmnsp/trri/ 1986; 1 : 151-54. Charnard .I, Rindie P, Lavaud S, 'l'oupance 0, Maheut t i , Lacour F. Carpal tunnel syndrome and type of dialysis rneni- brane. Br Mcd .I 1989:298:867-8. Floege J , Granolleras C. Deschodt G. Heck M , Baudin G , Branger B. Tournier 0, Reinhard B, Eisenbach GM, Smeby I,C. Shaldon S. High-flux synthetic versus cellulosic membrane for /i'~-microglobulin removal during heniodialysis, hemodiafiltration and hemofiltration. Nephrol Uitrl 7rori.s- p / N / i t 1989:4:653-7. Gqjyo F, Homma N, Suzuko Y, Arakawa M. Serum levels of P,-microglobulin as a new form of amyloid protein in patients undergoing long-term hemodialysis. N E ~ g l J Mrt/ 1987: 314:585-6. Vincent C. Rivilard J P , Galland M, Trager J . Serum &-microglobulin in hernodialyyed patients. Nrp/ri.orr 1978:2 1 :260-8. Stein Ci. Schneider A. l 'hoss H , Linke K, Schaefer K , RitL E. Serumspiegel und 0rganabk.y-ungen von pl-Mikroglobtilin bci Dialysepatienten. Kliri Woc~hrrr,\c~/rr 1990:68: 1150-5. Fischbach M. Haniel G. Koehl C . Geisert J . P-2-microglobti- lin in hernodialillrated children: Long-term efficcncy follow- up. Ncplirotr 198953: 110-4. Miura Y . Ishiymia -1. Inomata A, Tukeda T. Senma S. Neuo- kauyama K, Suyuki Y. Radiolucent bone cysts and the type of dialysis membrane used in patients undergoing long-term hemodialysis. Nopliroti 1992:60:268-73. Ritz E. Deppisch R . Betz M , Bomnier J , Stein Ci. p-2-microglobulin-derived amyloid in osleoarthrapathy in dialysis pa- ticnls. Kitlric,\~ Mr,ttrh Rorir 0 i . c 1989;2:7-17. Floege J. Granolleras C, Bingel M, Deschodt G . Rrange B. Oules K , Koch KM. Shaldon S. /&-Microglobulin kinetics during hemodialysis and hernofiltration. Nephrol / X c r / Trtrric- p l t r r r t 1987: 1:223-8. Schaeccr KM. Ciilge U , Goehl H , Heidland A. Evaluation of ii new polyaniid membrane (Polyflux 130) in high-flux dialysis.

Simon P. Cavarle YY, Ang KS. Cam G, Catheline H . Long- term variations of serum /&microglobulin levels in hemodia- lyscd urernics according to permeability of dialysis mem-

Jacobs C, Petitclerc T. High-flux hernodialysis with a poly- amid dialyzer Contrih Nrplirol 1992:96: 106- 10. Haas T. Hillion D, Dongrady G. Phosphate kinetics in dialysis

von Herrath D. Hijfler M , Hartenstein ti. Kutschera D, Schaefer K , Dati F. Good biocornpatibility of the polyaniid hemotilter. Blood P~rrif' 1988;6: 106-10. Schaefer RM, Hubei- L, Cilge U . Bausewein K , Vienken J . Heidland A. Clinical evaluation of a new high-flux cellulose acetale membrane. f r i t J Ar/;f 'Orgtoi .s 1989: 12:85-90. Fournier A , Moriniere P. A new capillary hemodiafiltcr based on polyamide membrane: Polyflux 130. Clinical report HCE 7497. Lund: Gambro, 1989. Bingel M. Arndt W. Schulze M. Floege J . Shaldon L, Koch KM. Gtilze 0. Comparative study ofC5a plasma levels with different heniodialysis membranes using an enzyme-linked inimunosorbent-assay. Nrpliroti 1989:s 1:320-4. RitL E, Deppisch R, Stein G. P!-Microglobulin-derived amyloid in dialysis patients. In: Hiirl W H , Schollmeyer JP, eds. Ncw pi,r.spcc,/ivc.s i r i h e r i / o t / i t r / y ~ i.\ , pi~r i /o / i i , t i l di~rlv.\ is , (irit>- riour~mrrs ~ i ~ ~ t t r o ~ l / r ~ r / i r i ~ i , trrirl p lr i .s i t icrpl ir i~~~si .s . New York: Plenum Press, lY89;lI-8. van Ypersele de Strihou C, Honhon €3, Vandenbroucke JM, Htiaux JP, Noel H, Maldague B. Dialysis amyloidosis. Arlu Neplrrol 1988;17:40 1-2 I .

Blood Prrrif' 1990;8:23-3 I.

brane\. R/(JOd P//r( f 1988;6: 1 1 1-6.

patlellts. NPphI'(J/ I);[// ~ / ' ( / ! / . \ / ~ ! ~ / l f / 1991 :sUppl 2: 108-13.

HIGH-FLUX D t A L YSIS MEMBRANE 345

25. van Ypersele de Strihou C, Jadoul M. Malghem J . Maldague B. Jamart J . The Working Party on Dialysi5 Amyloidosis. Effect of dialysis membrane and patient's age on signs of dialysis-related amyloidosis. K i d t i q In / 1991 3 9 : 1012-9.

26. Jadoul M. Maldague M. Pirson Y . van Ypersele de Trihou c'. Renal transplantation for dialysis arthropathy [ L-ctter]. I . t r~ ic ,c . l 1988:2:849.

27. k'loege J . Smeby L. Dialysis related amyloidosi3 and high- flux membranes. Contrih Ncplirol 1992;96: 124-37.

28. Peterson J , Schoy D, Yeh J . Removal of j$-microglobulin during high-Hux dialy.

29. Mineshima M. Hoshino T. Era K. Diffusive and convective mass transport characteristics of pl-microglohulin removal.

30. Jorstad S. Smeby LC. Balstad T. Wiederoe T-E. Removal. generation and adsorption ofa-2-niicroglobulin during hemofiltration with five different membranes. Blood Prrrif 1988:6:96-105.

k ' id i iq / t i / 1988;33:324.

Z'r(lii,\ A M SOC. A r / ( f ' O / . g ~ / i . ~ 1987:33: 103-6.

31. Kerr PG, Argiles A , C m a u d B , Flavier J-L, Mion Ch. The effects o f reprocessing high-flux prolysulfone dialyrer with peroxyacetic acid on /3?-microglobulin removal in hemodiafiltration. Ani J Kicltrry I>;.\ 1992:19:433-8.

37. Goldman M. Lagniiche M. Dhoene M. Amraoui 2. Thayse C, Vanherweghem J-L. Adsorption of /&-microglobulin in dialyLers membranes: Comparison of different dialyzers and effects of reuse procedure>. /ufJAr/if'Orgci/i.s 1989: 12:373-8.

33. Driieke T, Urena P, Mann NK. Zingraff J . Membrane transfer. membrane adsorption and possible membrane-induced generation of P-2-microglobulin. Cotrtrih Neplrrol 1983;74: 113-9.

34. Floege J . Bartsch A. Schulze M . Shaldon S. Koch KM, Smeby LC. Clearance and synthesis rates ofp,-microglobulin in patients undergoing hemodialysis and in the normal subjects. J Loh Cliii M i 4 1991;118:l53-65,

35. Zingraff J . Driieke T. Can the nephrologiat prevent dialysis- related amyloidosis'? At,/ J Kidtic,\. Dis 1991:18:1-1 I .

Documents

Clinical Evaluation of a New Dialyzer, FLX-12 GW, with a Polyester-Polymer Alloy Membrane