The efficacy of erythropoietin in human immunodeficiency virus—infected end-stage renal disease patients treated by maintenance hemodialysis

The Efficacy of Erythropoietin in Human Immunodeficiency Virus-Infected End-Stage Renal Disease Patients Treated by Maintenance Hemodialysis D. Shrivastava,

MD, T.K.S. Rao, MD, R. Sinert, MD, E. Khurana, MD, A.P. Lundin, MD, and E.A. Friedman, MD

0 The superimposition of human immunodeficiency virus (HIV) infection, associated opportunistic infections, and anti-retroviral therapy further worsens the severity of anemia in patients also suffering from end-stage renal disease. A major cause of anemia in renal failure is a deficiency of erythropoietin. The causes of anemia in HIV disease include direct and indirect stem cell inhibition by the virus, increased peripheral destruction of red blood cells, and bone marrow suppression by various opportunistic infections and therapeutic drugs, particularly zidovudine. We compared the efficacy of recombinant human erythropoietin (rHuEP0) therapy in improving the anemia in HIV-infected end-stage renal disease patients (group I) with that in nondiabetic (group II) and diabetic (group Ill) hemodialysis patients without HIV infection. All three groups of patients were comparable in dialysis prescription and serum iron studies. Iron supplementation was prescribed to all patients, and none received blood transfusions. After 8 weeks of rHuEP0 therapy (administered intravenously in a dose of 100 U/kg body weight thrice weekly), the mean increase in hematocrit was similar in all responders (5.8% increase in hematocrit in 23 of 30 HIV patients and 8.7% increase in 24 of 30 non-HIV patients). Response in hematocrit was noted in HIV patients despite the presence of opportunistic infections in 15 and zidovudine administration in 11. Seven HIV-positive patients and six non-HIV patients failed to respond to rHuEP0. Irrespective of the HIV status, the baseline serum EPO levels in patients responding to rHuEP0 were significantly lower than those in nonresponders. We conclude that HIVinfected end-stage renal disease patients respond well to rHuEP0 therapy despite further bone marrow suppression by opportunistic infections and zidovudine, and that the responses correlate with low baseline serum EPO levels. 0 1995 by the National Kidney Foundation, Inc. INDEX

WORDS:

Anemia;

end-stage

renal

disease;

erythropoietin;

A

NEMIA is the most common hematologic complication encountered in end-stage renal disease (ESRD), as well as in a majority of patients undergoing either maintenance hemodialysis or peritoneal dialysis. In addition to retained uremic toxins, the accompanying anemia is one of the major factors that contributes to increased morbidity observed in ESRD patients. In patients with chronic renal failure (CRF), the primary cause of anemia is due to either an absolute or a relative deficiency of erythropoietin (EPO), attributable to a decreased production by the diseased kidneys.‘,* Other factors contributing to the development and persistence of anemia include mild hemolysis, blood loss during maintenance hemodialysis, bleeding diathesis from

From the Department of Medicine, Kings County Brooklyn, NY. Received July 19, 1994; accepted in revisedform 14, 1995. Supported by MetPath Laboratories, Teterboro, Address reprint requests to D. Shrivastava, MD, ment of Nephrology, Interfaith Medical Center, 555, Place, Brooklyn, NY 11238. 0 1995 by the National Kidney Foundation, Inc. 0272~6386/95/2506-0012$3.00/O 904

Hospital, February NJ. DepartProspect

American

Journal

human

immunodeficiency

virus.

platelet dysfunction, and inhibition of erythropoiesis by retained uremic toxins.3’4 The introduction of recombinant human EPO (rHuEP0) has revolutionized the management of anemia in patients with CRF and ESRD.5-g Anemia also is the most common hematologic abnormality in patients infected with the human immunodeficiency virus (HIV), the incidence and severity of which increase with the stage of HIV disease. The etiology of HIV-induced anemia is multifactorial, including ineffective red blood cell production, direct and indirect suppression of bone marrow progenitor stem cells by HIV, release of cytokines such as tumor necrosis factor, increased peripheral destruction of red blood cells, marrow suppression by various drugs (antibiotics, zidovudine [AZT], chemotherapeutic agents) and conventional/opportunistic infections, bleeding due to thrombocytopenia, disseminated intravascular clotting, and invasive diagnostic/therapeutic procedures. 1o-‘2 In asymptomatic HIV-seropositive patients and in those with the acquired immunodeficiency syndrome (AIDS) or AIDS-related complex, the serum immunoreactive EPO levels are inappropriately low to the degree of accompanying anemia.‘3-‘5 In a prospective study of symptomatic of Kidney

Diseases,

Vol25,

No 6 (June),

1995:

pp 904-909

EPO IN HIV PATIENTS ON MAINTENANCE

and asymptomatic HIV-infected patients treated with AZT, 30% to 48% developed anemia.” In some patients receiving AZT, the degree of anemia was severe, necessitating blood transfusions.” In AZT-associated anemia, the response to rHuEP0 therapy was predictable by the baseline serum EPO levels.‘6-‘9 Those with low levels responded well to exogenous rHuEP0 administration, while a poor response was seen in those with high baseline serum EPO levels.‘2~‘3 Patients infected with HIV can develop ESRD secondary to either HIV-associated nephropafiY *O,*’or known renal disorders, such as diabetes mellitus, polycystic kidneys, etc. The combination of uremia, HIV infection, and associated comorbid factors, such as the presence of opportunistic infections/malignancies, therapeutic agents used in their management, and the administration of anti-retroviral agents such as AZT, further aggravates the severity of anemia of renal failure. The effects of rHuEP0 in ESRD patients with HIV infection are unknown. We evaluated the efficacy of rHuEP0 therapy in improving the anemia in ESRD patients with HIV disease (receiving AZT and suffering from various opportunistic and other infections) and compared it with that in a group of non-HIV, nondiabetic patients and a group of non-HIV, diabetic ESRD patients. We also correlated the response to rHuEP0 treatment with initial serum EPO levels in HIV patients undergoing maintenance hemodialysis and contrasted it with that in the two groups of nonHIV ESRD patients. MATERIALS

905

HEMODIALYSIS

AND METHODS

Therapy with rHuEP0 is a component of standard management in a majority of patients undergoing maintenance hemodialysis or peritoneal dialysis in’the United States. This analysis was designed to test whether rHuEP0 is efficacious in ESRD patients with HIV infection (group I) as compared with age-matched, non-HIV, nondiabetic subjects (group II) and another group of non-HIV, diabetic maintenance hemodialysis patients (group III). Diabetic patients were chosen as a separate and more appropriate control group because infectious complications, which can suppress response to rHuEP0, also are generally increased in such subjects. Patients older than 18 years with documented ESRD who chose maintenance hemodialysis as renal replacement therapy at Kings County Hospital Center (Brooklyn, NY) were entered into the study. The study period was between November 1990 and December 1991. To exclude patients with acute reversible renal failure, only those patients who received a minimum of 12 weeks of maintenance hemodialysis were included. All blood samples for iron studies, routine

chemistries, and hemoglobin and hematocrit measurements were taken prior to the initiation of hemodialysis. The tests were performed by the certified hospital laboratories. Hemoglobin and hematocrit values were monitored biweekly. Semm samples for EPO levels were obtained in all patients during their first hemodialysis, prior to any administration of rHuEP0, and were frozen at -70°C for subsequent detetmination. Serum EPO assays were performed at MetPath Laboratories (Teterboro, NJ) by radioimmunoassay. All patients received rHuEP0 (Amgen, Thousand Oaks, CA) initially at a starting dose of 100 U/kg body weight intravenously three times a week at the end of each hemodialysis. In responding patients, the subsequent maintenance dose of rHuEP0 was 50 U/kg administered intravenously postdialysis three times a week. Oral iron supplementations (ferrous sulphate 324 mg three times a day) were prescribed to all patients receiving rHuEP0. None of the patients in the study received parenteral iron therapy. Patients were considered to be responders when a consistent increase of more than 5 points in hematocrit was seen at 8 weeks of therapy. The adequacy of hemodialysis was assessed in each patient by monthly measurements of KtN.

Statistical Analysis Data were analyzed by ANOVA or two-tailed Student’s f-test and by chi-squared analysis. Averages are presented as mean values ? SD. Values were considered statistically significant at P < 0.05. Statistical calculations were performed with the SPSS package for Windows (SPSS Inc, Chicago, IL). RESULTS

Sixty patients with ESRD were entered into the analysis, 30 in group I (HIV patients), 15 in group II (non-HIV, nondiabetic patients), and 15 in group III (non-HIV, diabetic patients). The demographics of the patients in each group are detailed in Table 1. There were 36 men and 24 women, with a preponderance of men in the HIV group (23 of 30). In the three groups, 51 patients were African-American and nine were Hispanic; there were no white patients. The patients in group III (non-HIV, diabetic subjects) were significantly older (mean age, 53.2 + 2.7 years as compared with a mean age of 35.7 and 32.7 years in group I and II patients, respectively; P < 0.05). Serum iron concentration and percent iron saturation levels, although slightly lower in HIV patients than in the other two groups of non-HIV patients, were not significantly different. Mean serum ferritin concentration (slightly lower in diabetic patients) also was not different in the three groups. Prestudy hematocrit values were signifi-

906

SHRIVASTAVA Table

1. Patient

Characteristics NON-HIV

HIV Patients

No. of patients

30 5 23l7 26 4 22 t 65.2 + 24.5 % 229.3 f 59.2 r 17 35.7

Age WI Men/women African-Americans Hispanics Pre-hematocrit (%) Serum iron (mg/dL) iron saturation (%) Serum ferritin (mg/dL) Baseline serum EPO (U/ml) No. of patients on AZT therapy No. of infections KW *P < 0.05 between group T P < 0.05 between group $ See text for details.

(Group

1.1

I)

Nondiabetic

32.7

26.3 93.7 32.7 224.1 91.1

1’2 1.10 III and groups I and groups

(Group

15 k 2.3 817 12

II)

Patients Diabetic

53.2

n

0.5t 26 9.7 164.1 44.5

t 0.04

ET AL

1.06

iO.9 2 42 t 18.2 + 120.2 -e 90.6 0 k 0.05

26.2 111.3 35.8 162.6 172.9

1.18

(Group

Ill)

15 + 2.7* 5/l 0 13 2 + 0.8 ? 37 t 16.9 ? 72.1 + 103.7 2 5 0.05

I and II. II and Ill.

cantly lower in group I patients (22% +- 0.5%) than in the patients in the other two groups (mean, 26.3%) (P < 0.05). As a group, serum EPO levels (normal for this laboratory, 0 to 19 mU/mL) were 59.2 + 44.5 mU/mL in HIV-positive patients compared with 91.1 + 90.6 mU/mL and 172.9 + 103.7 mU/mL in group II and III patients, respectively (P > 0.05). Adequacy of dialysis as measured by Kt/V was similar in all three groups. Seventeen of 30 HIV patients were receiving AZT at a dose of 300 to 500 mg/d during the study period. Infectious complications observed during the observation period in 16 HIV patients were bacterial pneumonia in three patients, Pneumocystis carinii pneumonia in two, Mycobacterium sp tuberculosis infection in two, staphylococcal sepsis in three, and fever of unknown origin (presumed to be secondary to tuberculosis) in six. In the diabetic group lII patients, one patient developed an infected foot ulcer and another patient had cell&is in the foot; both required antibiotic therapy. After 8 weeks of rHuEP0 therapy, 23 of 30 (77%) HIV patients and 24 of 30 (80%) non-HIV (both groups II and III) patients responded with an increase of at least five points in the mean hematocrit values. None of the patients were administered any blood transfusions during the study period. The mean increase in hematocrit of 5.6 2 0.7 points in HIV subjects compared with 6.73

2 0.64 points in non-HIV patients was not statistically significant. The increase in hematocrit values in diabetic patients, despite being older, was almost identical to that observed in the non-HIV ESRD control patients. In HIV patients, an increase in hematocrit response was seen despite the presence of severe infections in 11 of 23 responders and in 15 subjects receiving AZT. Seven HIV patients, five with infections (two tuberculosis, one P curinii, and two with fever of unknown origin) and two receiving AZT, did not respond to rHuEP0 therapy. In the non-HIV group, six patients (one with foot infection) were considered nonresponders to rHuEP0 as the increase in hematocrit was less than 3% (Table 2). Prestudy hematocrit levels in both the responders (24.31 + 0.55) and nonresponders (24.35 2 3.17) as a group were not significantly different. The correlation between initial serum EPO levels and response to rHuEP0 therapy can be seen in Fig 1. As shown in Table 2, the mean serum EPO levels were significantly lower in responders (17.1 2 11.8 mU/rnL in HIV patients and 13.3 + 4.6 mU/mL in non-HIV patients) and markedly elevated in nonresponders (416.2 + 184.5 mU/nL in HIV patients and 210.1 + 189.1 mU/mL in non-HIV patients) (P < 0.005). Adequacy of hemodialysis was similar in both responders and nonresponders to rHuEP0 therapy (Kt/V of 1.11 -t- 0.03 and 1.09 f 0.06, respectively).

EPO

IN HIV

PATIENTS

ON

Table

MAINTENANCE

2. Comparison

907

HEMODIALYSIS of Responders

and

Nonresponders

to Etythropoietin

Responders

Nonresponders

HIV No. of patients Pre-hematocrit (%) Post-hematocrit (%) Mean increase in hematocrit Serum EPO (U/ml) No. of patients on AZT

23130 22.16 t 28.88 2 5.81 ? 17.1 2 15123

(%)

lnfectionst

Non-HIV 24130 26.38 -t 32.19 2 6.73 2 13.3 -+ -

0.61 0.67 0.59 11.8

11123

* P < 0.005 between $ See text for details.

responders

and nonresponders

in both

DISCUSSION

Anemia is a major factor contributing significantly to morbidity in ESRD patients undergoing maintenance hemodialysis. The consequences of anemia include weakness, fatigue, reduced appetite, dizziness, headache, tachycardia, reduction in muscular strength and endurance, decreased cardiac performance, and impairment of cognitive

1. Correlation with EPO level.

of

0

0.64 0.67 0.64 4.6

23.01 24.71 2.08 416

1 I24

Complications of rHuEP0 treatment included mild to moderate increases in diastolic blood pressure to more than 100 mm Hg in three nonHIV patients and four diabetic control group patients. Hypertension was easily controlled with antihypertensive medications. Five patients needed declotting of vascular accesses (three patients in the diabetic group and one patient each in the HIV and non-HIV groups).

Fig dHCT

HIV

200

Non-HIV

7130 k 0.62 k 0.58 ~fr 0.29 k 184.5 217

25.9 27.98 1.70 210.1

517

6130 t 1.27 +- 1.49 ? 0.18 k 189.1* 116

groups.

functions. In addition, anemia contributes to a diminution in the quality of life due to amenorrhea, impotence, and impaired sexual function. Erythropoietin produced by the kidneys in response to tissue hypoxia causes an increase in erythrocyte production in the presence of adequate iron stores. The pathogenesis of anemia in CRF is primarily due to decreased EPO production by the diseased kidneys. Generally, in patients with CRF, serum EPO levels are inappropriately low for the degree of anemia, although “normal” or “elevated” levels also have been found.* Recombinant human erythropoietin successfully corrects the anemia in ESRD patients and improves various functional derangements.6-9 The primary target of HIV is the CD4+ T lymphocyte, leading to immune deficiency and development of infections and unusual malignancies. Human immunodeficiency virus infection

400

600

800

EPO

level

1000

1200

1400

908

also results in many systemic complications, including suppression of erythroid progenitor cells and other bone marrow elements. Many infectious and malignant complications occurring in HIV patients, and therapeutic agents (anti-retroviral drugs, antibiotics, and chemotherapy drugs) used in their management, also can lead to severe anemia. Bone marrow biopsy in AIDS patients has shown hypercellular marrow with a relative decrease in erythroid precursors.r6 In a study of 152 HIV-infected patients, anemia was seen in 18% of 22 asymptomatic carriers, in 50% of 26 AIDS-related complex patients, and in 75% of 104 AIDS patients.13 Studies in these patients showed that serum immunoreactive EPO levels were inappropriately lower in anemic AIDS patients, even though their ability to produce endogenous EPO was still intact.“,r3 In a double-blind placebo controlled study, 100 U/Kg of rHuEP0 administered intravenously three times a week to patients with advanced HIV disease who were not receiving AZT resulted in a decrease in the transfusion requirements and a 5% increase in the mean hematocrit.” The response was seen despite the presence of many opportunistic infections and Kaposi’s sarcoma in these treated patients. The mean serum EPO level in rHuEPO-treated patients in this study was 41.4 mU/mL, a value inappropriately low for the observed degree of anemia. Therapy with AZT resulted in the development of anemia in 48% of patients in a study from Baltimore” and in 31.3% of patients in a randomized, controlled study reported by Richman et al. l2 In a randomized, double-blind, placebo-controlled study of rHuEP0 therapy at a dose of 100 U/kg intravenously three times a week in 63 AIDS patients receiving AZT, reduction in transfusion requirements was seen in patients with baseline endogenous EPO levels of less than 500 III/L and not in those with levels greater than 500 IUiL.” In other words, AZT-induced anemia is associated with high serum EPO levels, further substantiating the fact that EPO production is intact in HIV patients, irrespective of the stage of disease. However, even these higher EPO concentrations fail to adequately stimulate erythropoiesis. The combination of HIV infection and ESRD (irrespective of the cause of renal failure) further complicates the severity of anemia and its effects

SHRIVASTAVA

ET AL

on patients. While rHuEP0 is highly useful in correcting anemia in non-HIV ESRD patients, is it equally efficacious in HIV-positive patients as well? In this study, we have shown that the degree of anemia encountered in HIV-infected ESRD patients who sustain multiple infections and receive AZT is much more severe than that seen in non-HIV dialysis patients. The mean hematocrit of HIV patients was 22%, compared with 26% in both HIV-negative nondiabetic patients and in older, sicker, non-HIV diabetic control patients (P < 0.05). Along with supplemental oral iron replacement, rHuEP0 administered at similar dosages resulted in improvements in hematocrit values in all groups. The mean increase in hematocrit response, ranging from 5% to 6%, was similar in both HIV and non-HIV patients, although the posttreatment hematocrit values were still below 30% in HIV patients. Despite similar serum iron values, nonresponse to rHuEP0 was seen equally in HIV-positive and -negative patients in both groups. In this short study, no attempts were made to investigate the causes for failure to respond to rHuEP0 in these patients. Increasing the dose of administered rHuEP0 may have resulted in response in these patients, a regimen that was not attempted. Serum EPO levels in HIV-positive ESRD patients receiving AZT vary from low to high levels. Despite the presence of CRF, HIV infection, and AZT therapy, EPO production is intact, similar to patients with no renal disease. The ability to respond to exogenous rHuEP0 administration seems to depend on the baseline serum EPO levels. Those with low levels responded better than patients with high basal serum EPO concentrations. In HIV patients, even though the mean increase in hematocrit was very similar to control groups, absolute values were still lower, which may indicate that higher doses of rHuEP0 may be needed in these patients. Further studies analyzing the responses to increasing doses of rHuEP0 are needed to answer this question. In summary, HIV-infected ESRD patients undergoing maintenance hemodialysis, despite suffering from life-threatening infections, and receiving the anti-retroviral agent AZT respond well to rHuEP0 and iron supplementation. The increase in hematocrit can be predicted on the basis of baseline serum EPO levels. Whether routine determinations of baseline serum EPO levels

EPO

IN HIV

PATIENTS

ON

MAINTENANCE

HEMODIALYSIS

are necessary prior to rHuEP0 therapy cannot be answered from this limited study. The use of rHuEP0 can facilitate the long-term administration of AZT and other drugs in HIV-positive ESRD patients without the risk of further worsening the anemia of CRF. REFERENCES 1. Fisher JW: Mechanism of anemia of chronic renal failure. Nephron 25:106-l 12, 1980 2. Eschbach JW, Adamson JW: Anemia of end stage renal disease. Kidney Int 28:1-5, 1985 3. McDermott IT, Galbraigh AJ, Corlett RJ: Inhibition of cell proliferation in renal failure and its significance to the uraemic syndrome. Scott Med J 20:317-323, 1975 4. Wallner SF, Vautrin RM: Evidence that inhibition of erythropoiesis is important in the anemia of chronic renal failure. J Lab Clin Med 97:170-175, 1981 5. Eshbach JW, Egrie JC, Downing MR, Adamson JW: Correction of the anemia of end stage renal disease with recombinant human erythropoietin. N Engl J Med 316:7378, 1987 6. Eschbach JW: The anemia of chronic renal failure: Pathophysiology and the effects of recombinant erythropoietin. Kidney Int 35:134-138, 1989 7. Guthrie M, Cardenas D, Eschbach JW, Haley NR, Robertson HT, Evans RW: Effect of erythropoietin on strength and functional status of patients on hemodialysis. Clin Nephrol 39:97-102, 1993 8. Delano BG: Improvements in quality of life following treatment with rHuEP0 in anemic hemodialysis patients. Am J Kidney Dis 14:14-l& 1989 (suppl) 9. Abels R: Review of hematologic effects of erythropoietin. Semin Nephrol lO:l-10, 1990 (suppl 1) 10. Zon LI, Arkin C, Groopman JE: Hematologic manifestations of the human immunodeficiency virus (HIV). Br J Hematol 66:251-256, 1987

909 11. Spivak JL, Bender BS, Quinn TC: Hematologic abnormalities in the acquired immune deficiency syndrome. Am J Med 77:224-228, 1984 12. Richman DD, Fischl MA, Grieco MH, Gottlieb M, Volberding P, La&in 0, Leedom J, Goopman J, Mildvan D, Hirsch M, Jackson G, Durack D, Lehrman SN: The toxicity of azidothymidine (AZT) in the treatment of AIDS and AIDSrelated complex: A double blind placebo controlled trial. N Engl J Med 317:192-197, 1987 13. Spivak JL, Barnes D, Fuchs E, Quinn T: Serum immunoreactive erythropoietin in HIV infected patients. JAMA 261:3104-3107, 1989 14. De Marchi S, Pirisi M, Ferraccioli GF: Erythropoietin and the anemia of chronic diseases. Clin Exp Rheumatol 11:429-444, 1993 15. Reddy MM, Grieco MH: Erythropoietin and granulocyte-macrophage colony-stimulating factor (GM-CSF) levels in sera of patients with HIV infection. Int J STD AIDS 2: 12% 132, 1991 16. Scadden DT, Zon LI, Groopman JE: Pathophysiology and management of HIV-associated hematologic disorders. Blood 74:1455-1463, 1989 17. Fischl M, Galpin JE, Levine JD, Groopman JE, Henry DH, Kennedy P, Miles S, Robbins W, Starrett B, Zalusky R, Abels RI, Tsai HC, Rudnick SA: Recombinant human erythropoietin for patients with AIDS treated with zidovudine. N Engl J Med 322:1488-1493, 1990 18. Groopman JE: Management of the hematologic complications of human immunodeficiency virus infection. Rev Infect Dis 12:931-937, 1990 19. Henry DH, Jemsek JG, Levin AS, Levine JD, Levine RL, Abels RI, Nelson RA, Thompson D, Rudnick SA: Recombinant human erythropoietin and the treatment of anemia in patients with AIDS or advanced ARC not receiving ZDV. J Acquir Immune Defic Syndr 5:847-848, 1992 20. Rao TKS, Friedman EA, Nicastri AD: The types of renal disease in the acquired immunodeficiency syndrome. N Engl J Med 316:1062-1068, 1987 21. Rao TKS: Human immunodeficiency virus (HIV) associated nephropathy. Annu Rev Med 42:391-401, 1991