- Email: [email protected]
Recombinant Human Erythropoietin: 18 Months' Experience in Hemodialysis Patients
Recombinant Human Erythropoietin: 18 Months' Experience in Hemodialysis Patients Bernard Canaud, Chantal Polito-Bouloux, Laurie J. Garred, Jean-Pierre Rivory, Patrick Donnadieu, Jean Taib, Pierre Florence, and Charles Mion • It has been shown that the regular administration of erythropoietin (EPO) permits the correction of anemia in end-stage renal failure patients. We analyzed the effect of chronic administration of EPO in 13 stable, regularly dialyzed end-stage renal failure patients over an 18-month period. The effects of EPO were evaluated according to standard criteria including clinical status, blood pressure control, hematology and biochemistry data, protein nutritional status, and dialysis efficiency. Following a 2-week control period, EPO was administered intravenously (IV) after the dialysis session according to a two-phase protocol. The first period (correction phase) consisted of a stepwise EPO dose increment, starting at 3 x 24 IUlkg/wk and doubling the dose every 14 days according to hemoglobin response in order to achieve a target hemoglobin level of approximately 11.0 g/dL (110 gIL). In the second period (maintenance phase) EPO dose was optimized to maintain the hemoglobin level between 100 and 110 gIL (10.0 and 11.0 g/dL), by adjusting either the unit dose or the frequency of injection. Anemia was corrected in all patients within 11 weeks, with EPO dose increasing from 72 to 360 IUlkg/wk. The stabilization of hemoglobin was achieved with an average EPO dose of 275 IUlkg/wk (50 to 4761U1kg/wk). Concomitantly, a subjective and clinical improvement was noted in all patients. The dialysis efficacy remained in an acceptable range throughout the study, falling significantly ("" 10%) through the first 3 months of treatment to stabilize at an effective urea clearance of "" 120 Llwk. The dietary protein intake calculated from urea kinetic modeling ranged between 1.1 and 1.2 g/kg/d. The pre-dialysis arterial pressure remained controlled within an acceptable range (:s; 150190 mm Hg) for all patients. However, in three patients prescription of an antihypertensive drug was necessary, while in the other 10 only dry body weight readjustment was required. Patient blood viscosity increased over the study period by 33%. Other side effects noted were vascular access thrombosis in six cases, flu-like syndrome in one patient, and a transitory pruritus in one patient. Finally, the lack of specific antibody (anti-EPO) and the constant efficacy of EPO over an 18-month period suggest that this molecule is not immunogenic and can be administered chronically in dialyzed patients. © 1990 by the National Kidney Foundation, Inc. INDEX WORDS: Recombinant human erythropoietin; anemia; end-stage chronic renal failure; hemodialysis.
R
ECOMBINANT human erythropoietin (EPO) produced by genetic engineering represents a major advance in the treatment of chronic anemia. Its long-term administration in hemodialysis patients results in essentially complete correction of hemoglobin deficiency, 1.5 and by removing this functional handicap enhances their quality of life. There is no further need for periodic transfusion with the associated risk of viral infection (hepatitis B, human immunodeficiency virus), hemosiderosis, and EPO regulation of anemia may even permit autotransfusion for eventual surgery. Thus, EPO heralds a new era for the anemic chronic renal dialysis patient. 6 Its effectiveness does not appear to diminish with time, which suggests that EPO may represent a longterm treatment for chronic anemia. 7 Prolonged use of EPO, however, does entail a certain number of problems. 8-12 In the short-term it adds both an additional expense and risk to the treatment of uremia. Its impact on the global cost of treatment, while difficult to establish precisely,
it not negligible. A useful parameter for evaluating the efficacy of dialysis treatment will disappear with correction of anemia, a fact that may expose these patients to inadequate dialysis. 12 Finally, the multiplicity of severe secondary effects already reported in the literature, particularly arterial hypertension and convulsive crises, is of concern as EPO comes into wider use with lessened vigilance in its administration. 7 In the long-term, the evaluation in blood viscosity and red blood cell mass coupled with inadequate adaptation of peripheral vascular resistance may lead to cardiovascular complications in patients already at risk. 9 . 13 From the Nephrology Service, Lapeyronie Hospital, Montpellier, France; the Department of Chemical Engineering, Lakehead University, Thunder Bay, Ontario, Canada; the Centre d'Hemodialyse du Languedoc, Montpellier, France; and the Centre Regional de Transfusion Sanguine, Montpellier, France. Address reprint requests to Bernard Canaud, Nephrology Service, Lapeyronie Hospital, 34059 Montpellier, France. © 1990 by the National Kidney Foundation, Inc. 0272-6386190/1 502-0009$3.0010
American Journal of Kidney Diseases, Vol XV, No 2 (February), 1990: pp 169-175
169
170
CANAUD ET AL Table 1.
Patient 1 2 3 4 5 6 7 8 9 10 11 12 13
EPO Patient Clinical Characteristics
Sex
Age (yr)
Dry Weight (kg)
Time on Dialysis (yr)
Vascular Access
Nephropathy Origin
F M F M F F F M F F F M M
58 56 36 52 22 57 55 25 45 35 44 30 40
66.0 60.0 59.0 60.0 54.0 45.5 47.0 51 .0 56.5 48.5 63.5 61.5 57.5
4 20 4 18 2 4 11 3 3 4 3 9 5
AVF AVF AVF AVF AVF AVF ThS AVF AVF AVF AVF AVF AVF
CIN CGN CGN CGN CIN(uropathy) CIN(toxic) PCKD Alport Syndrome CIN Undetermined Undetermined CIN(uropathy) CIN
Anephric
x x x
Abbreviations: AVF, arteriovenous fistula; ThS, Thomas Shunt; CGN, chronic glomerulonephritis; PCKD, polycystic kidney disease; CIN, chronic interstitial nephritis.
Finally, the most preoccupying concern in the long term is the possibility of bone marrow medullary fibrosis resulting from prolonged stimulation at hyperphysiological levels.
hemoglobin at its goal level. At this end of follow-up, the frequency of weekly EPO injection was distributed as follows: nine patients had two injections, three had three injections, one had one injection.
Dialysis Treatment Modes PATIENTS AND METHODS
Patients Following informed consent, 13 (five male, eight female) anemic (hemoglobin <70 giL [7.0 g/dL]) , stable end-stage chronic renal failure patients were included in this study. Table 1 lists the principal clinical parameters of the patient group. They ranged from 22 to 58 years of age (mean ± SEM,42 .7 ± 12.5) and had been dialyzed for an average of 6.9 years (range, 2 to 20). All were anuric, with three being anephric.
EPO Administration Protocol Human recombinant EPO (CILAB Laboratories, Paris, France) was administered intravenously (IV) posthemodialysis according to a three-phase protocol. (1) A 2-week observation period, which served as a control phase. (2) A correction phase in which anemia was corrected by thrice-weekly administration of EPO until a target hemoglobin level of approximately 110 giL (11.0 g/dL) was reached . The initial EPO dose was 24 VI/kg or 72 VI/kg/wk, with the dose being doubled at the end of each 2-week period during which the observed increase in hemoglobin level was less than 20 giL (2.0 g/dL). The maximum EPO dose was limited to 192 VI/kg or 576 VI/kg/wk. (3) A maintenance phase designed to stabilize patient hemoglobin level at 100 to 110 giL (10.0-11.0 g/dL). Weekly EPO administration was customized for each patient either by adjustment of the dose administered at each session or by reducing the number of injections per week in order to maintain
The extracorporeal treatment for each patient, as listed in Table 2, was not altered during the study. Nine patients were treated by hemodialysis (9 to 12 h/wk, 1.2 to 1.6 m2 of cuprophane or AN69, acetate or bicarbonate buffer) , three were on hemodiafiltration (9 to 12 h/wk, 1.9 m2 of polysulfone, bicarbonate buffer, 20 L of convective exchange) and one was on postdilutional hemofiltration (90 L/wk, 1.6 m 2 of polyamide, bicarbonate buffer) . Treatment dose expressed as KTIV index is given for each patient in Table 2. Five patients were included in a dialyzer reuse program. For all patients anticoagulation of the extracorporeal circuit was by continuous heparinization with a 40- to 60-VI/kg loading dose and an infusion rate of 700 to 1,000 IV/h.
Parameters Measured The usual clinical, hematological, and biological parameters were followed to evaluate the effects of EPO administration. For each treatment session , vital signs (weight, temperature, heart rate , arterial blood pressure) were noted before and after dialysis and a hematological control assay (complete blood cell count including reticulocytes) was performed. Each month urea kinetic modeling was performed over three consecutive dialysis cycles using the Gotch method, in order to evaluate the adequacy of the extracorporeal treatment, as well as patient protein nutritional status. Effective urea and creatinine clearance (KDW equivalent to the total body clearances per session were calculated from measured dialyzer clearances corrected for fistula recirculation and urea (or creatinine) generation rate occurring during a session. 10 Effective weekly clearance (L/wk) was calculated for urea (Ku) and creatinine (KC) as KO = KDE x HD session/wk.
171
RECOMBINANT HUMAN ERYTHROPOIETIN Table 2. Patient
Dialysis Modality
1 2 3 4 5 6 7 8 9 10 11 12 13
HD HDF HDF HD HDF HD HF HD HD HD HD HD HD
Dialyzer Type
EPO Patient Dialysis Treatment Program Session Length
Session Frequency
Dialysate Buffer
4.0 3.0 4.0 4.0 3.0 4.0 3.0 4.0 4.0 3.5 4.0 4.0 3.5
3 3 3 3 3 3 3 3 3 3 3 3 3
Bic Bic Bic Bic Bic Bic Bic Acet Acet Acet Acet Acet Acet
Cup AN69 PS AN69 PS PS PAM Cup Cup Cup Cup Cup Cup
KTN
Reuse
1.2 1.3 1.6 1.4 1.7 1.5 1.0 1.1 1.0 1.3 1.0 1.0 1.0
x x x x x
± ± ± ± ± ± ± ± ± ± ± ± ±
0.04 0.04 0.04 0.05 0.04 0.05 0.05 0.03 0.06 0.05 0.05 0.08 0.08
Abbreviations: HD, hemodialysis; HDF, hemodiafiltration; HF, hemofiltration; Cup, Cuprophane; PS, polysulfone; PAM, polyamide; Bic, bicarbonate; Acet, acetate . • Patients who required increased dialysis dose therapy.
Statistical Analysis All results are expressed or presented graphically as mean ± SEM (X ± SEM) . The nonparametric Wilcoxon rank test was used in comparing two groups of paired data using a significance level of P < 0.05. Mean arterial pressure (MAP in mm Hg) was calculated as: MAP [(systolic arterial pressure - diastolic arterial pressure)!3] + diastolic arterial pressure.
f I.5.
300
t
.... I 200
~ 100
~
'"
Il
0
IU
·60
II:
60
120
180
240
300
360
420
480
540
600
240
300
360
420
480
540
600
RESULTS 13
Patient Response to EPO Administration
12
Anemia was corrected and stabilized in all patients to an average hemoglobin level of 108 giL (10.8 g/dL) (range, 97 to 114 giL). The corresponding net improvement in physical condition resulted in an appreciable increase in patient activity.
i
11
Z
10
~
iii
.. 0::
0
....
0
9
i
'"
8
:z:
Effect of EPO on Erythrocyte Parameters The evolution of circulating reticulocyte and hemoglobin levels and of the EPO dose level over the study period is shown in Fig 1. Two distinct sections are apparent for each curve: a linear rise, corresponding to the correction phase, where reticulocyte and hemoglobin levels are roughly proportional to EPO dose and a second horizontal section where all three parameters remain approximately constant, corresponding to the maintenance phase. The proportional relationship observed during the correction phase between hemoglobin level and EPO dose defines a subject's sensitivity to EPO .
t g
6 ·60
t.. "a
g
60
120
180
500
t
I
400 JOO
--",..
!AI
200
~ 100
'"
0 ·60
60
120
180
240
300
360
420
480
540
600
TIllE ON EPO THERAPY (cIIy8)
Fig 1. Evolution of reticulocyte, hemoglobin, and EPO dose levels over the study period.
172
CANAUD ET AL 5 ,------------------------------, petlent. on Iron supplementation (oral or IV) 12113
o
T
011
0/13
o 36
100
164 228 292 356 EPO maintenance doH (Ullkg/Wk)
maintenance ph.
o control phase
420
Fig 2. Histogram of the EPO dose administered in the 13 patients during the maintenance phase.
EPO Requirements
An average of 275 VIIkg/wk was needed to stabilize hemoglobin at its goal level during the maintenance phase, requiring on the average two injections per week. The great disparity in the maintenance dose required from one patient to another (range, 36 to 468 UI/kg/wk) is illustrated in Fig 2. Effect of EPO on Iron and Other Essential Element Reserves and Requirements
In all the patients receiving EPO, reserves were marshalled, resulting in a decrease in iron and ferritin plasma levels (Fig 3). Therefore, oral or IV administration of iron was necessary in 12 of the 13 patients. IV iron administration was used in noncompliant patients or in the case of delayed or insufficient response to oral iron administration. However, no vitamin or essential element deficiency was observed. Effect of EPO on Other Blood Parameters
Leukocyte and platelet counts increased significantly without exceeding the normal range during the anemia correction phase, with a gradual leveling off observed during the maintenance phase (Fig 4). Effect of EPO on Dialysis Efficiency and Patient Protein Nutrition
The effect of EPO on treatment efficacy can be seen in Table 3, where the monthly mean values of effective urea and creatinine clearances and KT IV index are listed at the start and at the 6th and 12th month of EPO treatment. An increase in dialyzer surface area or dialysis duration was required for three patients treated by standard hemodialysis in order to maintain adequate dialysis.
I- I
I
10
0
20
I
30
I 40
I 50
serum Iron (limolll)
355
[]
maintenance ph.
C
643
[J
control ph...
C
C
m
o
m
1000
2000
3000
4000
5000
Ferritin (111110111) normal range
Fig 3. Changes in serum iron, plasma ferritin, and Iron supplementation requirements between the control and maintenance phase of EPO administration.
Neither dietary prescription nor the use of ion exchange resins (Kayexalate) was modified during the study period and no severe episodes of hyperkalemia were observed. Daily protein intake values determined from urea kinetic modeling of protein catabolic rate are reported in Table 3 at the start and at the 6th and 12th month of EPO treatment. Secondary Effects of EPO administration
Correction of anemia results in an increase in blood viscosity. As we reported previously,1O a gradual 33 % increase in blood dynamic viscosity accompanied the elevation in hemoglobin level from 62 to 108 giL (6.2 to 10.8 g/dL). Total heparin needs (loading plus maintenance doses) increased from 6,423 to 8,077 IV (.l25.7 %) per session. There was no significant change observed in predialysis arterial blood pressure over the study period (Table 3). However, in three patients with a previous history of hypertension, its relapse (> 150/90 mm Hg or MAP ~ 110 mm Hg) required prescription of antihypertensive medica-
173
RECOMBINANT HUMAN ERYTHROPOIETIN
290
i IS.
250 230
'" Iii ....
210
!C ....
190
w
a..
t
g §
270
170 150 ·60
60
120
180
240
60
120
180
240
300
360
420
480
540
600
540
600
8
'ilE
Ci co
t
g
7
~
S.
'"...>-w u 0,. :::> Fig 4. Evolution of platelet and leukocyte levels as a function of EPO treatment time.
....
~ 6 5
w
4 ·60
tion, in spite of a reduction in dry weight. Nevertheless, satisfactory blood pressure control was obtained using either a (:1-blocker (Acebutolol, 400 mg/d) or a calcium blocker (Nicardipine Long Acting, 20 mg twice daily). Six episodes of arteriovenous fistula or shunt thrombosis were observed. In three instances this was attributed to a poor development or low blood flow in an old and/or unused vascular access. In three other cases it occurred acutely requiring a surgical declotting intervention. However, in all cases this complication did not interfere with the overall efficacy of the patient's dialysis program. Although remaining in a low and acceptable range, the incidence of thrombotic complications apTable 3.
0
300 360 420 480 TIME ON EPO THERAPY (Cleys)
peared to increase with EPa treatment, since we had observed only two fistula thrombosis episodes in this population over the previous 12 months. An episode of generalized pruritis was observed in 1 patient following EPa administration. It responded to anti-histamine treatment and did not recur despite continuation of the EPa protocol. In one patient a flu-like syndrome was observed, which was effectively treated by analgesics and a decrease in the speed of EPa IV injection. DISCUSSION
EPa administration resulted in a rapid (75 days on average) correction of anemia in each of our study subjects, confirming previously reported ob-
Changes in Dialysis Efficiency and Protein Nutrition in 12 Patients on EPO Therapy for at Least 12 Months Months of EPO Treatment
o Dry weight, kg MAp, mm Hg Effective urea clearance (Ku), Llwk KTN index Effective creatinine clearance (Kc), Llwk Urea time averaged concentration, mmol/L (g/dL) Protein catabolic rate, g/kg/d
NOTE. Values are mean ± SEM. *Three patients on antihypertensive medication. tP < 0.05 compared with start of EPO therapy (month 0).
55.6 93.5 203 1.33 158 17.1 10.2 1.25
± ± ± ± ± ± ± ±
12
6
1.8 5.3 21 0.11 15 1.3 0.7 0.12
56.0 97.5 179 1.25 135 18.2 10.9 1.17
± 2.0 ± 2.6* ± 13t ± 0.08t ± 9t ± 1.4 ± 0.8 ± 0.12
57.4 89.9 189 1.24 138 17.2 10.3 1.11
± 2.2 ± 3.5* ± 12t
± 0.08t
± 13t ± 0.9 ± 0.5
± 0.10t
174
servations. I -4 .7.8 This reversal of chronic anemia was accompanied by a net improvement in the general physical condition of all the patients, renewing their interest in physical activity and life in general due to an improvement in working capacity.15 An improved treatment tolerance was observed in the three patients undergoing acetate hemodialysis. The EPO dose required to achieve hemoglobin level normalization varied greatly from patient to patient. This variation in patient response to EPO prevents its prescription on the conventional anthropological basis (body weight, size, body surface area). EPO administration in progressively increasing doses appears to be the safest method of correcting anemia while minimizing secondary effects. According to our findings, reticulocyte response commences in the first 2 weeks of EPO administration, with anemia correction to a goal hemoglobin level of approximately 110 gIL (11.0 g/dL) being obtained within 11 weeks. The linear relationship between hemoglobin level and EPO dose during the correction phase may be exploited to determine individual patient sensitivity to EPO. Patient EPO requirements appear to remain relatively constant during the maintenance phase; however, a number of factors may necessitate revision of a patient's weekly EPO prescription over time. These include changes in EPO injection frequency, inflammation or infection, consumption of iron reserves and capacity for mobilization, state of protein nutrition, and dialysis treatment mode.? EPO maintenance prescription can only be optimized empirically on an individual basis. The weekly dosage can be adjusted as required by either altering the dose of each injection or the number of injections per week. In our experience, a relative decrease in EPO effectiveness when observed was due to a relative or absolute iron deficiency. Oral or IV iron supplementation restored EPO efficacy in all cases. The relative stability of EPO requirements over the long term and the lack of detectable specific antibodies would appear to confirm the absence of immunogenicity of the synthetic EPO molecule. Treatment effectiveness did not vary over the course of our study despite the marked increase in red blood cell mass and the significant fall in the effective clearances of urea ( :=:: 7 %) and creatinine ( :=:: 12 %). Curiously and despite a general improvement in patient well-being accompanied by a
CANAUD ET AL
subjective increase in appetite, urea kinetic modeling showed no change in patient protein catabolic rate. This observation could reflect a positive dietary nitrogen balance or an anabolic state as has been suggested by others. 16 However, such interpretation remains purely speculative since dietary protein analysis was not performed. We were able to continue the reutilization program for "high-flux" dialyzers throughout the EPO study, realizing an average of nine uses for each device. Nevertheless, the total patient heparin requirements increased by 20 % to 40 %. The incidence of secondary effects of EPO administration were modest in comparison to the benefit gained. Hypertension, which only appeared in patients with a previous history, was readily controlled by reduction in EPO dose level, readjustment of dry weight, and/or prescription of an antihypertensive drug. This volume-resistant hypertension appears to be associated with increases in blood viscosity and total peripheral vascular resistance. 17 However, the possible contribution of EPO itself cannot be excluded due to its immunological parentage with renine substrate, although this remains purely speculative. 18 The increased thrombogenicity observed with EPO treatment may be associated with the elevation in platelet count and blood viscosity and the decrease in cardiac output. However, it is important to note that in most cases fistula thrombosis was associated with a blood access characterized by low flow. CONCLUSION
The recent availability of EPO has opened a new era in the treatment of the anemia often associated with chronic renal failure. It is a remarkably effective agent for elevating blood hemoglobin level and its long-term administration can reverse this functional handicap and the need for periodic transfusions in the anemic dialysis patient. Nevertheless, caution should be exercised, as EPO becomes more widely available, to avoid the multiplicity of associated complications. REFERENCES 1. Winearls CG, Oliver DO, Pippard MI, et al: Effect of human erythropoietin derived from recombinant DNA on the anaemia of patients maintained by chronic haemodialysis. Lancet 2: 1175-1178, 1986 2. Eschbach JW, Egrie IC, Downing MR, et al: Correction
RECOMBINANT HUMAN ERYTHROPOIETIN of the anemia of end-stage renal failure with recombinant human erythropoietin: Results of the phase I and II clinical trial. N EngJ J Med 316:73-78, 1987 3. Bommer J, Alexiou C, Miiller-Biihl U, et al: Recombinant human erythropoietin therapy in haemodialysis patients: Dose determination and clinical experience. NephroJ Dial Transplant 2:238-242, 1987 4 . Polito C, Donadieu P, Rivory Jp, et al : L'erythropoietine recombinante chez J' hemodialy se . Experience Montpellieraine. Euromedecine 87:345-347 , 1987 5. Walczyck MH , Golper TA: Correction of the anemia of end-stage renal failure with recombinant human erythropoietin. N EngJ J Med 317:249, 1987 6. Eschbach Jw, Adamson JW: Recombinant human erythropoietin: Implication for nephrology. Am J Kidney Dis 11 :203-209, 1988 7. Eschbach JW: The anemia of chronic renal failure: Pathophysiology and the effects of recombinant erythropoietin. Kidney Int 35: 134-148, 1989 8. Zehnder C , Gliick Z, Descoueudres C, et al: Human recombinant erythropoietin in anaemic patients on maintenance haemodialysis: Secondary effects of the increase of haemoglobin . Nephrol Dial Transplant 3:657-660, 1988 9. Raine AEG: Hypertension, blood viscosity, and cardiovascular morbidity in renal failure: Implications of erythropoietin therapy. Lancet 7:97-99, 1988 10. Canaud B, Donadieu P, Polito C, et al: Erythropoietin-
175 associated hypertension: What role for blood viscosity changes? Nephron 51:430-431, 1989 II. Shinaberger JH, Miller JH, Gardner PW: Erythropoietin alert: Risks of high hematocrit hemodialysis. ASAIO Trans 34:179-184, 1988 12. Acchiardo S, Moore L, Miles D, et aI : Does erythropoietin treatment change hemodialysis requirements? Kidney Int 35 :237, 1989, (abstr) 13 . Zannad F, Voisin P, Brunott F, et al: Haemorheological abnormalities in arterial hypertension and their relation to cardiac hypertrophy. J Hypertension 6:293-297, 1988 14. Gotch FA: Hemodialysis: Technical and kinetic consideration, in Brenner BM, Rector FC (eds) : The Kidney. Philadelphia, PA, Saunders, 1976, pp 1672-1704 15. Mayer G, Thurn J, Cada EM, et al: Working capacity is increased following recombinant human erythropoietin treatment. Kidney Int 34:525-528, 1988 16. Zehnter E, Pollok M, Ziegenhagen D, et aI: Urea kinetics in patients on regular dialysis treatment before and after treatment with recombinant human erythropoietin. Contrib Nephrol 66: 149-155, 1988 17. Akiba T, Kurihara S, Katoh H, et aI: Hemodynamic changes of hemodialyzed patients by erythropoietin treatment. Kidney Int 35 :237, 1989 (abstr) 18 . Rosenliif K: Immunological comparison of renin substrate with erythropoietin. Scand J Clin Lab Invest 46:497-504, 1986
R
ECOMBINANT human erythropoietin (EPO) produced by genetic engineering represents a major advance in the treatment of chronic anemia. Its long-term administration in hemodialysis patients results in essentially complete correction of hemoglobin deficiency, 1.5 and by removing this functional handicap enhances their quality of life. There is no further need for periodic transfusion with the associated risk of viral infection (hepatitis B, human immunodeficiency virus), hemosiderosis, and EPO regulation of anemia may even permit autotransfusion for eventual surgery. Thus, EPO heralds a new era for the anemic chronic renal dialysis patient. 6 Its effectiveness does not appear to diminish with time, which suggests that EPO may represent a longterm treatment for chronic anemia. 7 Prolonged use of EPO, however, does entail a certain number of problems. 8-12 In the short-term it adds both an additional expense and risk to the treatment of uremia. Its impact on the global cost of treatment, while difficult to establish precisely,
it not negligible. A useful parameter for evaluating the efficacy of dialysis treatment will disappear with correction of anemia, a fact that may expose these patients to inadequate dialysis. 12 Finally, the multiplicity of severe secondary effects already reported in the literature, particularly arterial hypertension and convulsive crises, is of concern as EPO comes into wider use with lessened vigilance in its administration. 7 In the long-term, the evaluation in blood viscosity and red blood cell mass coupled with inadequate adaptation of peripheral vascular resistance may lead to cardiovascular complications in patients already at risk. 9 . 13 From the Nephrology Service, Lapeyronie Hospital, Montpellier, France; the Department of Chemical Engineering, Lakehead University, Thunder Bay, Ontario, Canada; the Centre d'Hemodialyse du Languedoc, Montpellier, France; and the Centre Regional de Transfusion Sanguine, Montpellier, France. Address reprint requests to Bernard Canaud, Nephrology Service, Lapeyronie Hospital, 34059 Montpellier, France. © 1990 by the National Kidney Foundation, Inc. 0272-6386190/1 502-0009$3.0010
American Journal of Kidney Diseases, Vol XV, No 2 (February), 1990: pp 169-175
169
170
CANAUD ET AL Table 1.
Patient 1 2 3 4 5 6 7 8 9 10 11 12 13
EPO Patient Clinical Characteristics
Sex
Age (yr)
Dry Weight (kg)
Time on Dialysis (yr)
Vascular Access
Nephropathy Origin
F M F M F F F M F F F M M
58 56 36 52 22 57 55 25 45 35 44 30 40
66.0 60.0 59.0 60.0 54.0 45.5 47.0 51 .0 56.5 48.5 63.5 61.5 57.5
4 20 4 18 2 4 11 3 3 4 3 9 5
AVF AVF AVF AVF AVF AVF ThS AVF AVF AVF AVF AVF AVF
CIN CGN CGN CGN CIN(uropathy) CIN(toxic) PCKD Alport Syndrome CIN Undetermined Undetermined CIN(uropathy) CIN
Anephric
x x x
Abbreviations: AVF, arteriovenous fistula; ThS, Thomas Shunt; CGN, chronic glomerulonephritis; PCKD, polycystic kidney disease; CIN, chronic interstitial nephritis.
Finally, the most preoccupying concern in the long term is the possibility of bone marrow medullary fibrosis resulting from prolonged stimulation at hyperphysiological levels.
hemoglobin at its goal level. At this end of follow-up, the frequency of weekly EPO injection was distributed as follows: nine patients had two injections, three had three injections, one had one injection.
Dialysis Treatment Modes PATIENTS AND METHODS
Patients Following informed consent, 13 (five male, eight female) anemic (hemoglobin <70 giL [7.0 g/dL]) , stable end-stage chronic renal failure patients were included in this study. Table 1 lists the principal clinical parameters of the patient group. They ranged from 22 to 58 years of age (mean ± SEM,42 .7 ± 12.5) and had been dialyzed for an average of 6.9 years (range, 2 to 20). All were anuric, with three being anephric.
EPO Administration Protocol Human recombinant EPO (CILAB Laboratories, Paris, France) was administered intravenously (IV) posthemodialysis according to a three-phase protocol. (1) A 2-week observation period, which served as a control phase. (2) A correction phase in which anemia was corrected by thrice-weekly administration of EPO until a target hemoglobin level of approximately 110 giL (11.0 g/dL) was reached . The initial EPO dose was 24 VI/kg or 72 VI/kg/wk, with the dose being doubled at the end of each 2-week period during which the observed increase in hemoglobin level was less than 20 giL (2.0 g/dL). The maximum EPO dose was limited to 192 VI/kg or 576 VI/kg/wk. (3) A maintenance phase designed to stabilize patient hemoglobin level at 100 to 110 giL (10.0-11.0 g/dL). Weekly EPO administration was customized for each patient either by adjustment of the dose administered at each session or by reducing the number of injections per week in order to maintain
The extracorporeal treatment for each patient, as listed in Table 2, was not altered during the study. Nine patients were treated by hemodialysis (9 to 12 h/wk, 1.2 to 1.6 m2 of cuprophane or AN69, acetate or bicarbonate buffer) , three were on hemodiafiltration (9 to 12 h/wk, 1.9 m2 of polysulfone, bicarbonate buffer, 20 L of convective exchange) and one was on postdilutional hemofiltration (90 L/wk, 1.6 m 2 of polyamide, bicarbonate buffer) . Treatment dose expressed as KTIV index is given for each patient in Table 2. Five patients were included in a dialyzer reuse program. For all patients anticoagulation of the extracorporeal circuit was by continuous heparinization with a 40- to 60-VI/kg loading dose and an infusion rate of 700 to 1,000 IV/h.
Parameters Measured The usual clinical, hematological, and biological parameters were followed to evaluate the effects of EPO administration. For each treatment session , vital signs (weight, temperature, heart rate , arterial blood pressure) were noted before and after dialysis and a hematological control assay (complete blood cell count including reticulocytes) was performed. Each month urea kinetic modeling was performed over three consecutive dialysis cycles using the Gotch method, in order to evaluate the adequacy of the extracorporeal treatment, as well as patient protein nutritional status. Effective urea and creatinine clearance (KDW equivalent to the total body clearances per session were calculated from measured dialyzer clearances corrected for fistula recirculation and urea (or creatinine) generation rate occurring during a session. 10 Effective weekly clearance (L/wk) was calculated for urea (Ku) and creatinine (KC) as KO = KDE x HD session/wk.
171
RECOMBINANT HUMAN ERYTHROPOIETIN Table 2. Patient
Dialysis Modality
1 2 3 4 5 6 7 8 9 10 11 12 13
HD HDF HDF HD HDF HD HF HD HD HD HD HD HD
Dialyzer Type
EPO Patient Dialysis Treatment Program Session Length
Session Frequency
Dialysate Buffer
4.0 3.0 4.0 4.0 3.0 4.0 3.0 4.0 4.0 3.5 4.0 4.0 3.5
3 3 3 3 3 3 3 3 3 3 3 3 3
Bic Bic Bic Bic Bic Bic Bic Acet Acet Acet Acet Acet Acet
Cup AN69 PS AN69 PS PS PAM Cup Cup Cup Cup Cup Cup
KTN
Reuse
1.2 1.3 1.6 1.4 1.7 1.5 1.0 1.1 1.0 1.3 1.0 1.0 1.0
x x x x x
± ± ± ± ± ± ± ± ± ± ± ± ±
0.04 0.04 0.04 0.05 0.04 0.05 0.05 0.03 0.06 0.05 0.05 0.08 0.08
Abbreviations: HD, hemodialysis; HDF, hemodiafiltration; HF, hemofiltration; Cup, Cuprophane; PS, polysulfone; PAM, polyamide; Bic, bicarbonate; Acet, acetate . • Patients who required increased dialysis dose therapy.
Statistical Analysis All results are expressed or presented graphically as mean ± SEM (X ± SEM) . The nonparametric Wilcoxon rank test was used in comparing two groups of paired data using a significance level of P < 0.05. Mean arterial pressure (MAP in mm Hg) was calculated as: MAP [(systolic arterial pressure - diastolic arterial pressure)!3] + diastolic arterial pressure.
f I.5.
300
t
.... I 200
~ 100
~
'"
Il
0
IU
·60
II:
60
120
180
240
300
360
420
480
540
600
240
300
360
420
480
540
600
RESULTS 13
Patient Response to EPO Administration
12
Anemia was corrected and stabilized in all patients to an average hemoglobin level of 108 giL (10.8 g/dL) (range, 97 to 114 giL). The corresponding net improvement in physical condition resulted in an appreciable increase in patient activity.
i
11
Z
10
~
iii
.. 0::
0
....
0
9
i
'"
8
:z:
Effect of EPO on Erythrocyte Parameters The evolution of circulating reticulocyte and hemoglobin levels and of the EPO dose level over the study period is shown in Fig 1. Two distinct sections are apparent for each curve: a linear rise, corresponding to the correction phase, where reticulocyte and hemoglobin levels are roughly proportional to EPO dose and a second horizontal section where all three parameters remain approximately constant, corresponding to the maintenance phase. The proportional relationship observed during the correction phase between hemoglobin level and EPO dose defines a subject's sensitivity to EPO .
t g
6 ·60
t.. "a
g
60
120
180
500
t
I
400 JOO
--",..
!AI
200
~ 100
'"
0 ·60
60
120
180
240
300
360
420
480
540
600
TIllE ON EPO THERAPY (cIIy8)
Fig 1. Evolution of reticulocyte, hemoglobin, and EPO dose levels over the study period.
172
CANAUD ET AL 5 ,------------------------------, petlent. on Iron supplementation (oral or IV) 12113
o
T
011
0/13
o 36
100
164 228 292 356 EPO maintenance doH (Ullkg/Wk)
maintenance ph.
o control phase
420
Fig 2. Histogram of the EPO dose administered in the 13 patients during the maintenance phase.
EPO Requirements
An average of 275 VIIkg/wk was needed to stabilize hemoglobin at its goal level during the maintenance phase, requiring on the average two injections per week. The great disparity in the maintenance dose required from one patient to another (range, 36 to 468 UI/kg/wk) is illustrated in Fig 2. Effect of EPO on Iron and Other Essential Element Reserves and Requirements
In all the patients receiving EPO, reserves were marshalled, resulting in a decrease in iron and ferritin plasma levels (Fig 3). Therefore, oral or IV administration of iron was necessary in 12 of the 13 patients. IV iron administration was used in noncompliant patients or in the case of delayed or insufficient response to oral iron administration. However, no vitamin or essential element deficiency was observed. Effect of EPO on Other Blood Parameters
Leukocyte and platelet counts increased significantly without exceeding the normal range during the anemia correction phase, with a gradual leveling off observed during the maintenance phase (Fig 4). Effect of EPO on Dialysis Efficiency and Patient Protein Nutrition
The effect of EPO on treatment efficacy can be seen in Table 3, where the monthly mean values of effective urea and creatinine clearances and KT IV index are listed at the start and at the 6th and 12th month of EPO treatment. An increase in dialyzer surface area or dialysis duration was required for three patients treated by standard hemodialysis in order to maintain adequate dialysis.
I- I
I
10
0
20
I
30
I 40
I 50
serum Iron (limolll)
355
[]
maintenance ph.
C
643
[J
control ph...
C
C
m
o
m
1000
2000
3000
4000
5000
Ferritin (111110111) normal range
Fig 3. Changes in serum iron, plasma ferritin, and Iron supplementation requirements between the control and maintenance phase of EPO administration.
Neither dietary prescription nor the use of ion exchange resins (Kayexalate) was modified during the study period and no severe episodes of hyperkalemia were observed. Daily protein intake values determined from urea kinetic modeling of protein catabolic rate are reported in Table 3 at the start and at the 6th and 12th month of EPO treatment. Secondary Effects of EPO administration
Correction of anemia results in an increase in blood viscosity. As we reported previously,1O a gradual 33 % increase in blood dynamic viscosity accompanied the elevation in hemoglobin level from 62 to 108 giL (6.2 to 10.8 g/dL). Total heparin needs (loading plus maintenance doses) increased from 6,423 to 8,077 IV (.l25.7 %) per session. There was no significant change observed in predialysis arterial blood pressure over the study period (Table 3). However, in three patients with a previous history of hypertension, its relapse (> 150/90 mm Hg or MAP ~ 110 mm Hg) required prescription of antihypertensive medica-
173
RECOMBINANT HUMAN ERYTHROPOIETIN
290
i IS.
250 230
'" Iii ....
210
!C ....
190
w
a..
t
g §
270
170 150 ·60
60
120
180
240
60
120
180
240
300
360
420
480
540
600
540
600
8
'ilE
Ci co
t
g
7
~
S.
'"...>-w u 0,. :::> Fig 4. Evolution of platelet and leukocyte levels as a function of EPO treatment time.
....
~ 6 5
w
4 ·60
tion, in spite of a reduction in dry weight. Nevertheless, satisfactory blood pressure control was obtained using either a (:1-blocker (Acebutolol, 400 mg/d) or a calcium blocker (Nicardipine Long Acting, 20 mg twice daily). Six episodes of arteriovenous fistula or shunt thrombosis were observed. In three instances this was attributed to a poor development or low blood flow in an old and/or unused vascular access. In three other cases it occurred acutely requiring a surgical declotting intervention. However, in all cases this complication did not interfere with the overall efficacy of the patient's dialysis program. Although remaining in a low and acceptable range, the incidence of thrombotic complications apTable 3.
0
300 360 420 480 TIME ON EPO THERAPY (Cleys)
peared to increase with EPa treatment, since we had observed only two fistula thrombosis episodes in this population over the previous 12 months. An episode of generalized pruritis was observed in 1 patient following EPa administration. It responded to anti-histamine treatment and did not recur despite continuation of the EPa protocol. In one patient a flu-like syndrome was observed, which was effectively treated by analgesics and a decrease in the speed of EPa IV injection. DISCUSSION
EPa administration resulted in a rapid (75 days on average) correction of anemia in each of our study subjects, confirming previously reported ob-
Changes in Dialysis Efficiency and Protein Nutrition in 12 Patients on EPO Therapy for at Least 12 Months Months of EPO Treatment
o Dry weight, kg MAp, mm Hg Effective urea clearance (Ku), Llwk KTN index Effective creatinine clearance (Kc), Llwk Urea time averaged concentration, mmol/L (g/dL) Protein catabolic rate, g/kg/d
NOTE. Values are mean ± SEM. *Three patients on antihypertensive medication. tP < 0.05 compared with start of EPO therapy (month 0).
55.6 93.5 203 1.33 158 17.1 10.2 1.25
± ± ± ± ± ± ± ±
12
6
1.8 5.3 21 0.11 15 1.3 0.7 0.12
56.0 97.5 179 1.25 135 18.2 10.9 1.17
± 2.0 ± 2.6* ± 13t ± 0.08t ± 9t ± 1.4 ± 0.8 ± 0.12
57.4 89.9 189 1.24 138 17.2 10.3 1.11
± 2.2 ± 3.5* ± 12t
± 0.08t
± 13t ± 0.9 ± 0.5
± 0.10t
174
servations. I -4 .7.8 This reversal of chronic anemia was accompanied by a net improvement in the general physical condition of all the patients, renewing their interest in physical activity and life in general due to an improvement in working capacity.15 An improved treatment tolerance was observed in the three patients undergoing acetate hemodialysis. The EPO dose required to achieve hemoglobin level normalization varied greatly from patient to patient. This variation in patient response to EPO prevents its prescription on the conventional anthropological basis (body weight, size, body surface area). EPO administration in progressively increasing doses appears to be the safest method of correcting anemia while minimizing secondary effects. According to our findings, reticulocyte response commences in the first 2 weeks of EPO administration, with anemia correction to a goal hemoglobin level of approximately 110 gIL (11.0 g/dL) being obtained within 11 weeks. The linear relationship between hemoglobin level and EPO dose during the correction phase may be exploited to determine individual patient sensitivity to EPO. Patient EPO requirements appear to remain relatively constant during the maintenance phase; however, a number of factors may necessitate revision of a patient's weekly EPO prescription over time. These include changes in EPO injection frequency, inflammation or infection, consumption of iron reserves and capacity for mobilization, state of protein nutrition, and dialysis treatment mode.? EPO maintenance prescription can only be optimized empirically on an individual basis. The weekly dosage can be adjusted as required by either altering the dose of each injection or the number of injections per week. In our experience, a relative decrease in EPO effectiveness when observed was due to a relative or absolute iron deficiency. Oral or IV iron supplementation restored EPO efficacy in all cases. The relative stability of EPO requirements over the long term and the lack of detectable specific antibodies would appear to confirm the absence of immunogenicity of the synthetic EPO molecule. Treatment effectiveness did not vary over the course of our study despite the marked increase in red blood cell mass and the significant fall in the effective clearances of urea ( :=:: 7 %) and creatinine ( :=:: 12 %). Curiously and despite a general improvement in patient well-being accompanied by a
CANAUD ET AL
subjective increase in appetite, urea kinetic modeling showed no change in patient protein catabolic rate. This observation could reflect a positive dietary nitrogen balance or an anabolic state as has been suggested by others. 16 However, such interpretation remains purely speculative since dietary protein analysis was not performed. We were able to continue the reutilization program for "high-flux" dialyzers throughout the EPO study, realizing an average of nine uses for each device. Nevertheless, the total patient heparin requirements increased by 20 % to 40 %. The incidence of secondary effects of EPO administration were modest in comparison to the benefit gained. Hypertension, which only appeared in patients with a previous history, was readily controlled by reduction in EPO dose level, readjustment of dry weight, and/or prescription of an antihypertensive drug. This volume-resistant hypertension appears to be associated with increases in blood viscosity and total peripheral vascular resistance. 17 However, the possible contribution of EPO itself cannot be excluded due to its immunological parentage with renine substrate, although this remains purely speculative. 18 The increased thrombogenicity observed with EPO treatment may be associated with the elevation in platelet count and blood viscosity and the decrease in cardiac output. However, it is important to note that in most cases fistula thrombosis was associated with a blood access characterized by low flow. CONCLUSION
The recent availability of EPO has opened a new era in the treatment of the anemia often associated with chronic renal failure. It is a remarkably effective agent for elevating blood hemoglobin level and its long-term administration can reverse this functional handicap and the need for periodic transfusions in the anemic dialysis patient. Nevertheless, caution should be exercised, as EPO becomes more widely available, to avoid the multiplicity of associated complications. REFERENCES 1. Winearls CG, Oliver DO, Pippard MI, et al: Effect of human erythropoietin derived from recombinant DNA on the anaemia of patients maintained by chronic haemodialysis. Lancet 2: 1175-1178, 1986 2. Eschbach JW, Egrie IC, Downing MR, et al: Correction
RECOMBINANT HUMAN ERYTHROPOIETIN of the anemia of end-stage renal failure with recombinant human erythropoietin: Results of the phase I and II clinical trial. N EngJ J Med 316:73-78, 1987 3. Bommer J, Alexiou C, Miiller-Biihl U, et al: Recombinant human erythropoietin therapy in haemodialysis patients: Dose determination and clinical experience. NephroJ Dial Transplant 2:238-242, 1987 4 . Polito C, Donadieu P, Rivory Jp, et al : L'erythropoietine recombinante chez J' hemodialy se . Experience Montpellieraine. Euromedecine 87:345-347 , 1987 5. Walczyck MH , Golper TA: Correction of the anemia of end-stage renal failure with recombinant human erythropoietin. N EngJ J Med 317:249, 1987 6. Eschbach Jw, Adamson JW: Recombinant human erythropoietin: Implication for nephrology. Am J Kidney Dis 11 :203-209, 1988 7. Eschbach JW: The anemia of chronic renal failure: Pathophysiology and the effects of recombinant erythropoietin. Kidney Int 35: 134-148, 1989 8. Zehnder C , Gliick Z, Descoueudres C, et al: Human recombinant erythropoietin in anaemic patients on maintenance haemodialysis: Secondary effects of the increase of haemoglobin . Nephrol Dial Transplant 3:657-660, 1988 9. Raine AEG: Hypertension, blood viscosity, and cardiovascular morbidity in renal failure: Implications of erythropoietin therapy. Lancet 7:97-99, 1988 10. Canaud B, Donadieu P, Polito C, et al: Erythropoietin-
175 associated hypertension: What role for blood viscosity changes? Nephron 51:430-431, 1989 II. Shinaberger JH, Miller JH, Gardner PW: Erythropoietin alert: Risks of high hematocrit hemodialysis. ASAIO Trans 34:179-184, 1988 12. Acchiardo S, Moore L, Miles D, et aI : Does erythropoietin treatment change hemodialysis requirements? Kidney Int 35 :237, 1989, (abstr) 13 . Zannad F, Voisin P, Brunott F, et al: Haemorheological abnormalities in arterial hypertension and their relation to cardiac hypertrophy. J Hypertension 6:293-297, 1988 14. Gotch FA: Hemodialysis: Technical and kinetic consideration, in Brenner BM, Rector FC (eds) : The Kidney. Philadelphia, PA, Saunders, 1976, pp 1672-1704 15. Mayer G, Thurn J, Cada EM, et al: Working capacity is increased following recombinant human erythropoietin treatment. Kidney Int 34:525-528, 1988 16. Zehnter E, Pollok M, Ziegenhagen D, et aI: Urea kinetics in patients on regular dialysis treatment before and after treatment with recombinant human erythropoietin. Contrib Nephrol 66: 149-155, 1988 17. Akiba T, Kurihara S, Katoh H, et aI: Hemodynamic changes of hemodialyzed patients by erythropoietin treatment. Kidney Int 35 :237, 1989 (abstr) 18 . Rosenliif K: Immunological comparison of renin substrate with erythropoietin. Scand J Clin Lab Invest 46:497-504, 1986