Effects of hemodialysis and longterm erythropoietin treatment on protein C, and on free and total protein S

Thrombosis

Pergamon

Research, Vol. 80, No. 2, pp. 161-168, 1995 Copyright 0 1995 Elsevier Science Ltd Printed in the USA. All rights reserved 0049-3848/95 $9.50 + .OO

0049-3848(95)00162-X

LONGTERM

EFFECTS OF HEMODIALYSIS AND ERYTHROPOIETIN TREATMENT ON PROTEIN AND ON FREE AND TOTAL PROTEIN S

C,

Naomi Clynel , Nils Egberg2, Lam-Eric Lins 1 Department of Internal Medicine1 and Department of Clinical Chemistry and Blood Coagulation2, Karolinska Hospital, Stockholm, Sweden. (Received

6 February

1995 by Editor V Rasi; reviseal/accepted

4 August

1995)

ABSTRACT We studied 37 hemodialysis patients during hemodialysis in order to assess the effects of dialysis on endogenous plasma coagulation inhibitors (antithrombin III, protein C, free and total protein S). The patients were examined prior to erythropoietin (EPO) treatment, upon reaching target hemoglobin (Hb) and after 3 months at steady state Hb levels. The levels of protein C increased significantly during dialysis. However, EPO treatment did not affect the levels of any of the endogenous coagulation inhibitors, either upon reaching target Hb or after 3 months of steady state Hb. The sequence of change during dialysis for protein C, free and total protein S was constant when comparing respective patterns prior to EPO treatment to those at target Hb and steady state Hb respectively. In conclusion, hemodialysis seems to activate synthesis of endogenous coagulation while partial correction of anaemia with EPO does not affect the levels of these inhibitors.

Uremic patients have a haemorrhagic tendency mainly due to a defective platelet function (1) and a decreased number of red blood cells (2,3). This defective coagulation has been regarded as a useful adjunct to anticoagulation during hemodialysis. There are several physiological systems regulating the activity of the coagulation system; among others the antithrombin-heparin system and the protein C/protein S/thrombomodulin system. Previous studies have shown that contact with the dialysis membrane activates plasma coagulation as well as increasing tibrinolytic activity (43).

Key words: hemodialysis, coagulation, erythropoietin, protein C, protein S Corresponding author: Naomi Clyne, MD, Division of Nephrology, Department of Internal Medicine, Karolinska Hospital, S- 171 76 Stockholm, Sweden. Tel no: 46-8-590 03 124, FAX no: 46-8-590 03 066

161

162

INHIBITORS

DURING HEMODIALYSIS

Vol. 80, No. 2

Adequate heparinisation minimizes generation of fibrinopeptide A and formation of thrombinantithrombin complexes (4). Previous studies have given somewhat conflicting evidence on the effects of hemodialysis on protein C and protein S (6,7,8,9). Erythropoeitin (EPO) is now routinely used to treat renal anaemia. The partial correction of anaemia in patients with chronic renal failure often results in a normalised or shorter bleeding time (IO,1 1) and improved platelet function (12,13,14). Several studies have shown an increase in heparin requirement during hemodialysis after partial correction of anaemia with EPO (15,16,17). The possible effects of EPO on the protein C/protein S/thrombomodulin system are as yet not entirely clear. The question of possible prothrombotic effects of EPO has been raised, with some early reports indicating an increased risk of occlusion of the vascular accesses after EPO treatment (18). Moreover, successful hemodialysis depends on adequate anticoagulation to counteract the activation of the coagulation system which occurs when blood comes into contact with the dialyzer membranes. The protein C system has an important role in the prevention of thrombosis (19). The aim of the present study was therefore to examine possible changes in the protein C system during EPO treatment and hemodialysis by studying the effects of hemodialysis on levels of protein C, free and total protein S prior to and after partial correction of anaemia with EPO as well as after three months at steady state hemoglobin (I&).

PATIENTS The study comprised 37 consecutive hemodialysis patients (15 women, 22 men; mean age 59 f 16, range 25 - 86 years) who all had been on maintenance hemodialysis with a functioning arterio-venous fistula or graft for at least 3 months before entering the study. All patients were studied prior to and upon reaching target Hb. Of these 37 patients 31 were studied after 3 months at steady state Hb levels. Of the 6 patients unable to complete the study, 3 underwent renal transplantation, 2 died and one patient switched to continuous ambulatory peritoneal dialysis (CAPD). The initial Hb prior to treatment was 82 f 9 g/L. The primary renal diseases were glomerulonephritis (9 patients), polycystic kidney disease (5 patients), chronic pyelonephritis (5 patients), hypertensive renal disease (3 patients), papillitis necroticans (1 patient), sarcoidosis (1 patient), renal dysplasia (1 patient) and cause unknown (12 patients). No patient was treated with aspirin or nonsteroid antiinflammatory drugs during the study. Informed consent was obtained, and the study was approved by the Committee of Ethics at the Karohnska Hospital

EPO treatment The initial dose of EPO (RecormonQ Boehringer Mannheim or Eprex@, Cilag) was 40 to 50 U/kg body weight administered i v 3 times a week after dialysis. The dose was increased by 50 % every 4 weeks in order to achieve an increment in Hb of at least 8 g/L per month. The target Hb was between 105 and 120 g/L. After target Hb had been reached the EPO dose was unchanged or reduced by 50 % every 4 weeks, when necessary, in order to maintain the Hb level. Time to reach target Hb averaged 4.8 f 2.7, range 1.5 to 12 months. After reaching target Hb, this level of Hb was maintained at steady state and the patients were re-examined after 3 months.

Vol. 80, No. 2

INHIBITORS

DURING

HEMODIALYSIS

163

METHODS Dialysis regime Twenty-eight patients had arterio-venous fistulae and 9 patients had arterio-venous grafts. The dialyzers used were polysulphon hollow fibre (Fresenius AG, Bad Homburg, Germany) in 26 patients and cuprophan hollow fibre (Fresenius AG, Bad Homburg, Germany) in 11 patients. The dialyzer area ranged from 0.7 to 1.3 m2. The type of dialyzer used was unchanged throughout the study period. Single-pass dialysis systems were used with bicarbonate as the dialysis buffer in 30 patients and acetate in 7 patients. PVC blood lines (Fresenius AG, Bad Homburg, Germany) were used for all treatments. The blood flow rate was approximately 225 mb’min and the dialysate flow rate was 500 ml/min. The number of hours of dialysis per week was on average 9 hours, ranging from 7.5 to 10.5 hours and was unchanged throughout the study period. The ultrafiltration rate was regulated to achieve the dry weight of each patient. The same nurse performed the standardized dialysis procedure at baseline, at target Hb and after 3 months at steady state Hb in every patient. Blood sampling Blood samples were drawn prior to dialysis from the fistula or graft, at the end of dialysis Tom the arterial and venous lines respectively. Samples were drawn in tubes containing l/10 volume of sodium citrate solution, 0.129 mol/L (Vacutainer tubes, Becton Dickinson). The samples were centrifuged for 20 minutes at 3000 g:s and the plasma was separated immediately. All samples were stored at -700 C until analysed. Heparin treatment Prior to entering the study the heparin dose infused during dialysis treatment was individually titrated for each patient to increase the whole blood activated coagulation time (WBACT) by 25 - 50% and the heparin dose was maintained at the same level at target Hb, after which a new heparin dose was individually titrated. The standardized dialysis procedure after 3 months at steady state was thus performed with a new heparin dose when necessary. Heparin (Kabi, Sweden) was administered as a bolus followed by a continuous infusion that stopped 30 min prior to the end of the dialysis session. Laboratory methods Hemoglobin concentration, hematocrit and serum concentrations of sodium, potassium, creatinine and urea were measured by routine methods. Whole blood activated coagulation time (WBACT) was measured with a Hemochron@ coagulometer (normal range: 125-148 set; Technidyne Carp, New Jersey). Protein C was measured with a photometric method (normal range: 0.68-1.20 U/r&; Coatest protein C, Chromogenix, Molndal, Sweden). Free and total protein S were measured with a ELISA kit (normal range free protein S: 0.20-0.50 U/mL; normal range total protein S: 0.56-1.34 U/r&; Stago Diagnostica, Asnieres, France). Antithrombin III (AT III) was determined by a chromogenic substrate procedure (normal range: 0.86-1.30 U/mL; Coatest antithrombin, Chromogenix, Molndal, Sweden). All results were corrected for changes in hematocrit (Hct) during the hemodialysis procedure. The correction factor F was calculated from the formula (20) (1 -Hctpost) x Hctpre F= __________________-_______ (1 -Hctpre) x Hct post

pre = predialysis post = postdialysis

,164

INHlBlTORS

DURING HEMODIALYSIS

Vol. 80, No. 2

Statistical analyses The data were analysed for statistical significance using ANOVA for repeated measures design with stages and levels as the factors. Assumption of normality was checked and transformations carried out when necessary. Interaction between stages and levels was also evaluated. Overall tests were followed by specified linear contrasts (pairwise comparisons) when factors were found significant.

RESULTS Prior to EPO treatment the Hb averaged 82 f 9, increased to 111 f 12 at target Hb (p < 0.0001) and was 108 f 15 g/L afler 3 months at steady state (p < 0.0001). The reticulocyte count was 1.1 f 0.6 prior to EPO, 1.3 f 0.7 at target Hh (NS) and 1.3 f 0.6 after 3 months at steady state @IS). The serum Na, K, creatinine were unchanged throughout the study period. Serum urea showed a statistically significant increase at target Hb, but after 3 months at steady state the values no longer differed from baseline. The levels of protein C (table I) increased significantly postdialysis in both the arterial and venous lines also showing a significant increase in the venous line compared with the arterial line postdialysis. The levels of free and total protein S (table I) were unchanged throughout the whole dialysis procedure. None of the levels of protein C, free or total protein S were affected by EPO treatment. The sequence of change during dialysis for all three proteins was constant when comparing the patterns during dialysis prior to EPO treatment to target Hb and steady state Hb respectively (table I). Antithrombin III levels were not changed at target Hb nor after 3 months at steady state (baseline: 0.97 f 0.14, target Hb: 0.93 h 0.15, steady state Hb: 0.98 f 0.13 U/ml). The total heparin dose increased significantly at target Hb and this increase persisted after 3 months at steady state (baseline: 5 541 f 1 073, target Hb: 5 838 f 1 079 U, p < 0.0005; steady state: 5 758 f 1 134 U, p < 0.005). The percentual increase in WBACT at the end of dialysis compared to predialysis was significantly lower at target Hb and this reduction persisted after 3 months at steady state (baseline: 33 f 12%, target Hb: 26 f 16%, p < 0.05; steady state Hb: 27 f 12%, p < 0.05). DISCUSSION In the present study we found levels within the normal range for protein C and free and total protein S immediately before starting dialysis. Protein C increased significantly by the end of the dialysis procedure and also showed a significant increase after contact with the dialyzer, as levels in the venous lines were significantly greater than in the arterial lines at the end of dialysis. This is in line with some previous studies where hemodialysis was found to induce an increase in protein C (8,9), but in contradiction to others who either registered no change (6) or a decrease in protein C levels during dialysis (7). One possible reason for this discrepancy could be the fact that we, and both studies registering an increase in protein C during dialysis, had corrected for changes in Hct during dialysis, while the investigators with unchanged or decreased values had not.

Vol. 80, No. 2

INHIBITORS

DURING HEMODIALYSIS

165

TABLE 1 Protein C (PC), Free (FPS) and Total Protein S (TPS) Pre- and Postdialysis, During EPO Treatment (mean&SD). PreHDa U/ml PCBaseline

PCTarSet

Hb

PCSteady state Hb

POSt HDS U/ml

pre-posta

Post HDv U/ml

pre-postv

Prior to and

POSta-POStv

1.15 kO.29

1.23 kO.28

0.001

1.31 kO.27

0.001

0.001

1.15 20.26

1.20

0.001

1.26 20.28

0.001

0.001

to.30

1.12 LO.24

0.001

1.25 to.29

0.001

0.001

20.23

1.16

FPSBaseline 0.40

0.39 +0.09

NS

0.42 20.11

NS

NS

20.15

0.36 +o.os -

0.34 +o.os

NS

0.36 kO.10

NS

NS

0.34 20.06

0.35 _ +0.06

NS

0.34 kO.06

NS

NS

TPSBaseline

1.28 +0.32

1.27 kO.25

NS

1.45 to.50

NS

NS

TPSTar,et m

1.19 F0.24

1.18 iO.26

NS

1.24 20.20

NS

NS

1.31 20.3 1

1.28 +0.27

NS

1.33 50.25

NS

NS

FPSTarget

HIJ

F=steady stateHb

T=steady state Hb

PreHDa = predialysis arterial line; Post HDa = postdialysis arterial line; pre-posta= predialysis compared to postdialysis arterial line, Post HI& = postdialysis venous line; pre-postv= predialysis compared to postdialysis venous line, posta-paste postdialysis arterial line compared to postdialysis venous line.

INHIBITORS

DURING HEMODIALYSIS

Vol. 80, No. 2

The dialyzer initiates an inflammatory reaction with a release of cytokines during dialysis (21). However, while levels of active free protein S may be reduced during an acute phase reaction, there are no reports on the effects of an acute phase reaction on protein C synthesis. Levels of free and total protein S during dialysis were unchanged, which is in accordance with a previous study (8). There might be changes in as yet unidentified plasma factors that stimulate synthesis of protein C during dialysis. As protein C has a relatively short regeneration time with a half life of about 7 hours, mechanisms triggered during dialysis would affect the plasma levels within the 3 to 4 hours duration of dialysis treatment. Thus, the increased levels of protein C that we found during hemodialysis would rather indicate a preventive effect towards thrombosis in the dialysis patient rather than point to a thrombotic tendency. There were no changes in levels of protein C, free or total protein S or AT111 during EPO treatment. The pattern of change observed during dialysis for protein C, free and total protein S was unaffected by EPO treatment. Thus EPO treatment does not seem to affect the levels of endogenous anticoagulants during hemodialysis, a finding in accordance with recent studies on hemodialysis patients (23,24). Another investigator reports unchanged levels of protein C and free and total protein S after EPO treatment of CAPD patients (25). In one study a significant decrease in protein C and protein S was reported after EPO treatment of hemodialysis patients, albeit within the normal range (26), while an earlier study showed a decrease in protein C and protein S after four months of EPO treatment down to levels known to predispose for thrombosis, although these levels became normal after 8 months of treatment (17). The protein C levels in this latter study were already prior to EPO treatment in the lower part of the reference interval, thus the decrease indicating a risk for thrombosis may partly be due to the low initial values. None of these studies had standardized for heparin treatment during dialysis, a necessary procedure, as most studies show an increase in heparin requirement by about 10% after partial correction of anemia with EPO (15,16). Thus the obeservations in contradiction to our findings could reflect an inadequate degree of anticoagulation during dialysis. All previous investigators measured either predialysis levels of protein C (17,23,24) or took the tests 24 hours after dialysis (26), thereby not taking the effects of hemodialysis into account. Moreover most studies only report changes during the initial phase of correction and lack the long-term observation period during steady state. In conclusion, levels of protein C increase significantly towards the end of the dialysis session, while levels of free and total protein S are unaffected. This pattern of increase is unaffected by partial correction of anemia with EPO. Moreover, the levels of protein C, free and total protein S and AT111 are unaffected by long-term EPO treatment. Thus, the protein C system may have a protective antithrombotic influence during and after hemodialysis and counteract the prothrombotic stimulus that is exerted by contact with the dialyzer. Moreover, possible prothrombotic effects of EPO treatment do not seem to be mediated through the protein C system. Acknowledgements The authors would like to thank MS Anette Bygden for her excellent and meticulous technical assistance.

Vol. 80. No. 2

INHIBITORS

DURING

HEMODIALYSIS

167

REFERENCES G, LIVIO M, MERINO J, DE GATEANO G. Bleeding in renal failure. Altered platelet function in chronic uraemia only partially corrected by haemodialysis. Nephron 22, 347 - 353,1978. 2. LIVIO M, MARCHES1 D, REMUZZI G, GOTTI E, MECCA G, DE GAETANO. Uraemic bleeding: Role of anaemia and beneficial effect of red cell transfusions. Lancet 2, 1013 - 1015,1982. FERNANDEZ F, GOUDABLE C, SIE P, TON-THAT H, DURAND D, SUC JM, 3. BONEIJ B. Low haematocrit and prolonged bleeding time in uraemic patients: Effect of red cell transfusions.Br J Haematol 59, 139 - 148,1985. 4. IRELAND HA, BOISCLAIR MD, LANE DA, THOMPSON E, CURTIS JR. Hemodialysis and heparin. Alternative methodsof measuring heparin and of detecting activation of coagulation. Clin Nephrol 35 (l), 26 - 34,199 1. NAKAMURA Y, TOMURA S, TACHIBANA K, CHIDA Y, MARUMO. Enhanced 5. fibrinolytic activity during the course of hemodialysis. Clin Nephrol 38 (2), 9096,1992. 6. KNUDSEN F, S0RENSEN PJ, NIELSEN AH, DYERBERG J. Functional impairment of protein C activity during hemodialysis - a new mechanism of extracorporeal thrombogenesis.Blood Purif 7, 230-232, 1989. 7. ALEGRE A, VICENTE V, GONZALEZ R, ALBERCA I. Effect of hemodialysison protein C levels. Nephron 46, 386-387,1987. 8. LA1 KN, YIN JA, YUEN PMP, LI PKT. Protein C, protein S and antithrombin III levels in patients on continuous ambulatory peritoneal dialysis and hemodialysis. Nephron 56, 271-276,199O. 9. ARIK N, ~ZDEMIR 0, AKPOLAT T, ARINSOY T, ~ZCEBE 0, DUNDAR s, YASAVUL U, TURGAN C, CAGLAR S. Protein C and its inhibitors during hemodialysis.Nephron 60, 488,1992. 10. MOIA M, MANUCCI PM, VIZZOTTO L, CASATI S, CAITANEO M, PONTICELLI C. Improvement in the haemostaticdefect of uraemia after treatment with recombinant erythropoietin. Lancet 2, 1227 - 1229,1987. 11. VIGANO G, BENIGN1 A, MENDOGNI D, MINGARDI G, MECCA G, REMUZZI G. Recombinanthumanerythropoietin to correct uremic bleeding. Am J Kidn Dis 18 (1) 44 - 49,1991. 12. FABRIS F, CORDIANO I, RAND1 ML, CASONATO A, MONTINI G, ZACCHELLO G, GIROLAMI A. Effect of humanrecombinanterythropoetin on bleeding time, platelet 14number and function in children with end-stage renal disease maintained by haemodialysis.Pediatr Nephrol 5, 225 - 228,1991, 13. HURAIB S, AL MOMEN AK, GADER AMA, MITWALLI A, SULMANI F, ABUAISHA H. Effect of recombinant human erythropoietin (rHueopo) on the hemostatic systemin chronic hemodialysispatients,Clin Nephrol 36 (5) 252 - 257,1991. 14. CASES A, ESCOLAR G, REVERTER JC, ORDINAS A, LOPEZ PEDRET J, IEwR REVERT L, CASTILLO IX. Recombinanthumanerythropoietin treatment imporoves platelet t%nctionin uremic patients.Kidney Int 42 66%672,1992. 15. SCHAEFER RM, KUERNER B, ZECH M, DEhJNINGER G, B0RNEFF C, HEIDLAND A. Treatment of the anemiaof hemodialysispatientswith recombinant humanerythropoietin. Int J artif Organs II (4) 249 - 254,1988. 1.

REMUZZI

168

INHIBITORS

DURING

HEMODIALYSIS

Vol. 80, No. 2

16.

BUUR T, LUNDBERG M. Secondary effects of erythropoietin treatment metabolism and dialysis efficiency in stable hemodialysis patients. Clin Nephrol

17.

MACDOUGALL IC, DAVIES ME, HALLETI I, COCHLIN DL, HUTTON RD, COLES GA, WILLIAM JD. Coagulation studies and fistula blood flow during erythropoietin therapy in hemodialysis patients. Nephr Dial Transpl 6, 862-867,199l. DAHLBACK B. Physiological anticoagulation. Resistance to activated Protein C and venous thromboembolism. J Clin Invest 94, 923-927, 1994. SUNDAL E & KAESER U. Correction of anaemia of chronic renal failure with recombinant human erythropoietin: Safety and efficacy of one year’s treatment in a European multicentre study of 150 haemodialysis dependent patients. Nephrol Dial Transplant 4, 979-987,1989. PETAJA J. Fibrinolytic response to venous occlusion for 10 and 20 minutes in healthy subjects and in patients with deep venous thrombosis. Thromb Res 56, 251263, 1989. ZAOUI P & HAKIM RM. The effects of the dialysis membrane on cytokine release. J Am Sot Nephrol 4, 1711-1718,1994. GUNNARSSON B, ASABA H, DAWIDSON S, WILHELMSSON S, BERGSTROM J. The effects of three different heparin regimes on heparin concentrations in plasma and fibrin formation in dialyzers. Clin Nephrol 15 (3), 135-142, 1981. AUNSHOLT NA, STEFFENSEN G, AHLBOM. Influence of recombinant human erythropoietin on platelet function and coagulation factor. Blood Purif 10, 248 253,1992. TAYLOR JE, BELCH JJF, MCLAREN M, HENDERSON IS, STEWART WK. Effect of erythropoietin therapy and withdrawal on blood coagulation and fibrinolysis in hemodialysis patients. Kidn Int 44, 182-190,1993. LA1 KN, YIN JA, LI PKT, YUEN PMP, LUI SF. Effect of subcutaneous administration of recombinant human erythropoietin on plasma protein C, protein S, and antithrombin III. Int J Artif Organs 1.5(5), 264-268, 1992. WIRTZ JJJM, VAN ESSER JWJ, HAMULYAK K, LEUNISSEN KML, VAN HOOF JP. The effects of recombinant human erythropoietin on hemostasis and fibrinolysis in hemodialysis patients. Clin Nephrol 38 (5), 277-282,1992.

on 34

(5), 230 - 235,199O.

18. 19.

20. 21.

22. 23. 24. 25. 26.