The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
PreviousPrevious
Volume 328:171-175 January 21, 1993 Number 3
NextNext

Effect of Serum Parathyroid Hormone and Bone Marrow Fibrosis on the Response to Erythropoietin in Uremia
D. Sudhaker Rao, Mei-shu Shih, and Ravinder Mohini

 

This Article
-Abstract

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background and Methods Anemia is common in patients with chronic renal insufficiency and secondary hyperparathyroidism. Erythropoietin therapy is effective, but the dose required varies greatly. One possible determinant of the efficacy of erythropoietin therapy is the extent of marrow fibrosis caused by hyperparathyroidism. We examined the relation between the erythropoietic response to erythropoietin and hyperparathyroidism in a cross-sectional study of 18 patients undergoing hemodialysis who had received erythropoietin therapy for one to three years. In 7 patients (the poor-response group) the dose of intravenous erythropoietin needed to maintain a mean (±SD) target hematocrit of 35 ±3 percent was >100 units per kilogram of body weight three times a week, and in 11 patients (the good-response group) it was <= 100 units per kilogram. In all patients, indexes of the adequacy of dialysis and the extent of hyperparathyroidism and aluminum toxicity were determined monthly, and bone histomorphometry was performed.

Results The mean (±SD) dose of erythropoietin required to maintain the target hematocrit was 174 ±33 units per kilogram three times a week in the poor-response group and 56 ±18 units per kilogram in the good-response group. The mean ages, duration and adequacy of dialysis, increment in hematocrit, iron requirements, and serum concentrations of calcium, phosphate, and aluminum were similar in the two groups. The percentages of osteoid volume and surface, the osteoid thickness, and the stainable aluminum content of bone were similar in the two groups. In contrast, the mean serum parathyroid hormone concentration, the percentages of osteoclastic and eroded bone surfaces, and the degree of marrow fibrosis were greater in the poor-response group than in the good-response group (P = 0.03, P = 0.04, P = 0.009, and P = 0.009, respectively).

Conclusions In patients with uremia, the dose of erythropoietin needed to achieve an adequate hematocrit response may depend on the severity of secondary hyperparathyroidism and the extent of bone marrow fibrosis.


Anemia contributes markedly to morbidity in patients with chronic renal insufficiency treated with hemodialysis. Although the pathogenesis of the anemia is multifactorial, a deficiency of erythropoietin secretion is probably the most important cause1,2. Recombinant human erythropoietin corrects anemia in virtually all patients undergoing hemodialysis,3 although some require large doses.

In a recent clinical trial, the median dose of erythropoietin required to maintain the mean (±SD) target hematocrit of 35 ±3 percent was 75 units per kilogram of body weight intravenously three times a week, but the range was wide -- from 12.5 to 500 units per kilogram3. Although complete resistance to erythropoietin therapy has not been reported, approximately 20 to 25 percent of the patients in that trial required more than 100 units per kilogram. Several factors are known to attenuate the erythropoietic response to erythropoietin, both at the inception of therapy and during the course of maintenance therapy4. They include a deficiency of iron or folic acid, infections, inflammatory diseases, myelofibrosis, aluminum-induced microcytosis, and severe hyperparathyroidism. Iron deficiency is the most common5.

Despite the known relation between hyperparathyroidism, marrow fibrosis, and anemia,6 little is known about the effect of parathyroid hormone-induced bone marrow fibrosis on the erythropoietic response to erythropoietin therapy. We undertook this cross-sectional study to examine the relation between hyperparathyroidism and marrow fibrosis and the response to erythropoietin in patients with chronic renal disease treated with hemodialysis.

Methods

Of 90 adult patients with chronic renal insufficiency treated with maintenance hemodialysis, we studied the 18 (20 percent; 9 men and 9 women, ranging in age from 22 to 72 years) who met all the following criteria: more than 1 year of dialysis; more than 1 year of erythropoietin therapy; achievement of the target hematocrit of 35 ±3 percent within 8 to 12 weeks after the initiation of erythropoietin therapy; the maintenance of this hematocrit with a constant dose of erythropoietin; the absence of symptoms of either hyperparathyroidism or aluminum-related bone disease; and serum aluminum concentrations of less than 40 µg per liter at the time of the study. None of the patients had any other known cause of resistance to erythropoietin therapy, such as deficiency of iron or folic acid, infections, inflammatory conditions, or aluminum toxicity,4 and none were receiving immunosuppressive therapy. Five of the 11 patients in the good-response group (see below) and 3 of the 7 in the poor-response group were receiving aluminum hydroxide as the phosphate binder; the remaining 10 patients were receiving calcium carbonate or calcium acetate to treat hyperphosphatemia. Sixteen patients (10 in the good-response group and 6 in the poor-response group) were receiving calcium-channel-blocking drugs as antihypertensive therapy. Fourteen of the patients were participants in a previously reported trial of erythropoietin therapy3. The duration of erythropoietin therapy was one to three years.

For the purpose of this study, we prospectively divided patients into two arbitrarily defined groups: the poor-response group (those who required an intravenous dose of erythropoietin of >100 units per kilogram three times a week to maintain the target hematocrit) and the good-response group (those who required a dose of <= 100 units per kilogram to maintain the target hematocrit). This division was based on the results of the earlier trial, in which 75 to 80 percent of the patients achieved their target hematocrit with less than 100 units of erythropoietin per kilogram three times weekly3. The study was approved by the human rights committee of the institutional review board, and all the patients gave informed consent.

The patients were receiving dialysis with high-flux Polysulfone membranes at the time of the study. The adequacy of dialysis was assessed by measuring serum creatinine and urea nitrogen concentrations. The patients received oral iron or intravenous iron dextran therapy as needed to maintain a serum ferritin concentration of more than 100 ng per milliliter and transferrin saturation in serum of more than 20 percent.

Serum calcium, phosphate, creatinine, urea nitrogen, and alkaline phosphatase were measured by standard methods in the hospital laboratory. Serum aluminum was measured by atomic-absorption spectrophotometry. Serum intact parathyroid hormone was measured by a two-site immunoradiometric assay (Allegro Intact parathyroid hormone, Nichols Institute, San Juan Capistrano, Calif.). Serum 25-hydroxyvitamin D was measured by radioimmunoassay after preliminary ethanol extraction with kits obtained from the Nichols Institute, and 1,25-dihydroxyvitamin D was measured by radioreceptor assay with kits obtained from Incstar (Stillwater, Minn.). All samples were collected immediately before a hemodialysis treatment.

Biopsies of the iliac crest were performed after in vivo double tetracycline labeling with a 3-mm Jamshidi needle in the anterior superior iliac spine7. Quantitative bone histomorphometric analysis was performed as previously described,8 by an investigator who had no clinical, biochemical, or therapeutic information about the study patients. In addition to the routine histomorphometric measurements, areas of marrow fibrosis were measured and expressed as a percentage of bone marrow space. The extent of stainable bone aluminum at the mineralization front, demonstrated by a modified aurine tricarboxylic acid method,9 was measured, and the results were expressed as a fraction of the osteoid surface. The osteoclast surface was expressed as a fraction of the mineralized bone surface (normal range, 0.1 to 2.3 percent). Marrow iron content, demonstrated by Gomori's iron stain, was expressed as a fraction of the osteoid surface.

Statistical analysis was performed with unpaired t-tests for the demographic and biochemical measurements and with Mann-Whitney nonparametric tests for the bone histomorphometric measurements. Simple regression analysis was used for assessing correlations.

Results

The mean (±SD) dose of erythropoietin required to maintain the target hematocrit was 174 ±33 units per kilogram intravenously three times a week (range, 125 to 219) in the poor-response group and 56 ±18 units per kilogram (range, 30 to 85) in the good-response group. There were no significant differences between the two groups in age, duration of dialysis, or adequacy of dialysis (Table 1). Neither the initial hematocrit values nor the increments in hematocrit during erythropoietin therapy differed significantly between the two groups. Serum ferritin concentrations, transferrin saturations in serum, iron dextran requirements, and numbers of blood transfusions were also similar in the two groups (Table 2), as were the serum concentrations of calcium, phosphate, aluminum, 25-hydroxyvitamin D, and 1,25-dihydroxyvitamin D (Table 3). The mean serum parathyroid hormone concentration, however, was significantly higher (P = 0.03) in the poor-response group, although there was considerable overlap between the groups. Serum alkaline phosphatase concentrations were also slightly higher in the poor-response group (P = 0.06) (Table 3). There was a significant correlation between serum parathyroid hormone and alkaline phosphatase concentrations for the entire study group (r = 0.78, P<0.001).

View this table:
[in this window]
[in a new window]
 
Table 1. Characteristics of 18 Patients Undergoing Hemodialysis Who Received Long-Term Erythropoietin Therapy.

 
View this table:
[in this window]
[in a new window]
 
Table 2. Hematologic Values in 18 Patients Undergoing Hemodialysis Who Received Long-Term Erythropoietin Therapy.

 
View this table:
[in this window]
[in a new window]
 
Table 3. Serum Biochemical Values in 18 Patients Undergoing Hemodialysis Who Received Long-Term Erythropoietin Therapy.

 
The percentages of osteoid volume and surface, the osteoid thickness, and the percentage of stainable bone aluminum were similar in the two groups (Table 4). The percentages of osteoclast surface, eroded surface, and marrow fibrosis were significantly higher in the poor-response group (Table 4 and Figure 1). These bone histomorphometric findings, along with the elevated serum parathyroid hormone and alkaline phosphatase concentrations, are consistent with the hypothesis that secondary hyperparathyroidism leads to marrow fibrosis and thus interferes with the erythropoietic response to erythropoietin.

View this table:
[in this window]
[in a new window]
 
Table 4. Bone Histomorphometric Findings in 18 Patients Undergoing Hemodialysis Who Received Long-Term Erythropoietin Therapy.

 

View larger version (87K):
[in this window]
[in a new window]
 
Figure 1. Photomicrographs of Cancellous Bone Tissue from Iliac-Crest Biopsies Stained with Toluidine Blue (pH 6.8) (x50).

Panel A shows extensive marrow fibrosis (MF) in the biopsy specimen of a patient with severe hyperparathyroidism. The mineralized bone (MB) appears as solid dark areas with fine discrete cement lines and is covered by a layer of osteoid (unmineralized matrix). The marrow space is completely obliterated by the fibrous connective tissue. Panel B shows mild paraosseous marrow fibrosis along with hematopoietic cells and fat in the biopsy specimen of a patient with less severe hyperparathyroidism.

 
When the patients were classified on the basis of serum parathyroid hormone concentrations of <150 pg per milliliter or >= 150 pg per milliliter (16 pmol per liter) (a value below which the osseous effects of hyperparathyroidism are absent10), the erythropoietin requirement was significantly higher in the nine patients with serum parathyroid hormone concentrations of >= 150 pg per milliliter (128 ±66 vs. 75 ±26 units per kilogram, P = 0.041). As expected, serum alkaline phosphatase concentrations and the percentage of marrow fibrosis were significantly higher and serum aluminum and stainable bone aluminum were significantly lower (data not shown) in the nine patients with elevated serum parathyroid hormone concentrations.

When the patients were divided according to the presence (13 patients) or absence (5 patients) of stainable bone aluminum, there was no significant difference in the mean doses of erythropoietin required (96 ±64 vs. 116 ±58 units per kilogram). Similarly, there was no significant difference in erythropoietin requirements between the 12 patients in whom the stainable bone aluminum was <10 percent of the osteoid surface (a value at which aluminum-related bone disease is least likely to occur) and the 6 patients in whom the stainable bone aluminum was >= 10 percent (117 ±69 vs. 71 ±28 units per kilogram). However, the percentages of osteoclast surface and marrow fibrosis were significantly higher in the 12 patients with either no stainable bone aluminum (5 patients) or 1 to 10 percent stainable bone aluminum (7 patients). Only three patients (two in the good-response group and one in the poor-response group) had stainable bone aluminum over more than 25 percent of the osteoid surface, a value considered to indicate aluminum toxicity11. Serum aluminum concentrations were less than 40 µg per liter (1500 nmol per liter) in all 3 of these patients, but none had any clinical features of aluminum toxicity, and their mean dose of erythropoietin was not significantly different from that in the remaining 15 patients with stainable bone aluminum values of less than 25 percent (80 ±36 vs. 106 ±66 units per kilogram). Thus, the erythropoietin requirements did not appear to depend on the extent of stainable bone aluminum, and there was no correlation between the dose of erythropoietin and serum aluminum or stainable bone aluminum in the study group as a whole.

In contrast, the 10 patients with marrow fibrosis of >= 1 percent required a significantly higher dose of erythropoietin to maintain the target hematocrit than did the 8 patients with marrow fibrosis of <1 percent (129 ±70 vs. 68 ±28 units per kilogram, P = 0.04), despite a significantly higher value for stainable bone aluminum in the latter (3.9 ±8.5 percent vs. 20.5 ±24.5 percent, P = 0.05). Furthermore, the dose of erythropoietin correlated significantly with the degree of marrow fibrosis (r = 0.47, P = 0.048), as well as with other osseous effects of hyperparathyroidism (Table 5). The biochemical and histomorphometric results thus suggest that in the absence of overt aluminum toxicity or other factors that modify the response, erythropoietin requirements may depend on the extent of bone marrow fibrosis.

View this table:
[in this window]
[in a new window]
 
Table 5. Correlation between the Dose of Erythropoietin and Serum Parathyroid Hormone and Bone Histomorphometric Values in 18 Patients Undergoing Hemodialysis Who Received Long-Term Erythropoietin Therapy, According to Univariate Analysis.

 
Finally, despite similar numbers of blood transfusions in the two groups before erythropoietin therapy and similar requirements for parenteral iron dextran during therapy (Table 2), the percentage of stainable bone iron was significantly higher in the poor-response group (3.5 ±0.8 percent vs. 1.6 ±1.1 percent, P = 0.004). There was also a significant correlation between the dose of erythropoietin and marrow iron content (r = 0.75, P = 0.002) (Table 5).

Discussion

These results confirm previous observations relating anemia to marrow fibrosis in patients with uremia who have secondary hyperparathyroidism,12,13,14 as well as in patients with primary hyperparathyroidism6,15,16,17. Furthermore, improvement in anemia has been reported after parathyroidectomy in both patients with severe renal osteodystrophy13,14,18,19 and those with primary hyperparathyroidism15,17. In addition, a significant correlation between the concentrations of hemoglobulin and serum alkaline phosphatase (a marker of the severity of bone disease) was found in patients with primary hyperparathyroidism with or without mild-to-moderate renal insufficiency17. These observations and the results of our study suggest that excess secretion of parathyroid hormone leads to bone marrow fibrosis and consequent interference with erythropoiesis. A direct inhibitory effect of parathyroid hormone on erythropoiesis was reported in one study20 but was not confirmed by others21,22. Alternatively, parathyroid hormone may inhibit the renal or extrarenal production of erythropoietin, since serum erythropoietin concentrations increase after parathyroidectomy in some patients with uremia14. Other potential effects mediated by parathyroid hormone include alterations in extracellular and intracellular calcium and phosphate levels, the release of cytokines by osteoclasts or resorbed bone, and decreased responsiveness of erythropoietic progenitor cells to exogenous erythropoietin. However, erythropoietin had no effect on bone histomorphometric features in a recent study23.

Anemia due to an increased aluminum burden has been well characterized, and it appears to involve a defect in iron use. Therapy with deferoxamine ameliorates the anemia induced by aluminum24. The response to deferoxamine depends on endogenous erythropoietin secretion,24 however, a finding that is relevant to this study as well as to the treatment of aluminum-related anemia with exogenous erythropoietin. Indeed, aluminum-induced resistance to erythropoietin therapy has been reported, and concurrent therapy with deferoxamine and erythropoietin enhances the erythropoietic response to erythropoietin25. However, we found no differences in serum aluminum and stainable bone aluminum values in the two groups in our study.

Despite the lack of correlation between erythropoietin requirements and serum parathyroid hormone or alkaline phosphatase concentrations, the dose of erythropoietin correlated significantly with the osseous effects of excess parathyroid hormone (Table 5). There was even the suggestion of a dose-response relation between the dose of erythropoietin and the degree of marrow fibrosis, regardless of the serum aluminum or stainable bone aluminum values. The lack of correlation between the dose of erythropoietin and the concentration of serum parathyroid hormone (Table 5) is not surprising, since the effect of excess parathyroid hormone on erythropoiesis is ultimately mediated by marrow fibrosis. The small number of patients studied, the wide variations in serum parathyroid hormone concentrations, and the greater sensitivity of bone histomorphometric analysis in detecting occult parathyroid bone disease26 may also explain the lack of a relation between the serum parathyroid hormone concentration and erythropoietin requirements.

The relation between erythropoietin requirements and stainable marrow iron values (Table 5) needs further study. It is possible that excess iron in bone marrow may simply reflect the inefficient use of iron as a result of marrow fibrosis.

These preliminary results suggest that in the absence of aluminum toxicity or other well-known response-limiting factors, the erythropoietic response to erythropoietin therapy depends largely on the extent and severity of marrow fibrosis due to secondary hyperparathyroidism. In some ways, responsiveness to erythropoietin therapy is analogous to insulin resistance; virtually all patients respond if the dose is large enough. Whether reducing the secretion of parathyroid hormone by treatment with oral or intravenous 1,25-dihydroxyvitamin D or parathyroidectomy improves the erythropoietic response to erythropoietin therapy needs further study27.

We are indebted to Ms. Paulette Wilson for performing the serum parathyroid hormone, 25-hydroxyvitamin D, and 1,25-dihydroxyvitamin D assays; to Drs. M. Kleerekoper and F. Dumler for their continued support during the study; to Drs. A.M. Parfitt and T. Drueke for critical review of the manuscript; and to Ms. Linda Aielo for assistance in the preparation of the manuscript.


Source Information

From the Divisions of Bone and Mineral Metabolism (D.S.R., M.S.) and Nephrology (R.M.), Department of Medicine, Henry Ford Hospital, Detroit.

Address reprint requests to Dr. Rao at the Bone and Mineral Division, Henry Ford Hospital, 2799 W. Grand Blvd., Detroit, MI 48202.

References

  1. Erslev AJ. Erythropoietin. N Engl J Med 1991;324:1339-1344. [Medline]
  2. Nissenson AR, moderator. Recombinant human erythropoietin and renal anemia: molecular biology, clinical efficacy, and nervous system effects. Ann Intern Med 1991;114:402-416.
  3. Eschbach JW, Abdulhadi MH, Browne JK, et al. Recombinant human erythropoietin in anemic patients with end-stage renal disease: results of phase III multicenter clinical trial. Ann Intern Med 1989;111:992-1000.
  4. Stivelman JC. Resistance to recombinant human erythropoietin therapy: a real clinical entity? Semin Nephrol 1989;9:Suppl 2:8-11. [Medline]
  5. Van Wyck DB, Stivelman JC, Ruiz J, Kirlin LF, Katz MA, Ogden DA. Iron status in patients receiving erythropoietin for dialysis-associated anemia. Kidney Int 1989;35:712-716. [Medline]
  6. Albright F, Aub JC, Bauer W. Hyperparathyroidism: a common and polymorphic condition as illustrated by seventeen proved cases from one clinic. JAMA 1934;102:1276-1287. 
  7. Rao DS. Practical approach to bone biopsy. In: Recker RR, ed. Bone histomorphometry: techniques and interpretation. Boca Raton, Fla.: CRC Press, 1983:3-11.
  8. Parfitt AM, Rao DS, Stanciu J, Villaneuva AR, Kleerekoper M, Frame B. Irreversible bone loss in osteomalacia: comparison of radial photon absorptiometry with iliac bone histomorphometry during treatment. J Clin Invest 1985;76:2403-2412.
  9. Maloney NA, Ott SM, Alfrey AC, Miller NL, Coburn JW, Sherrard DJ. Histological quantitation of aluminum in iliac bone from patients with renal failure. J Lab Clin Med 1982;99:206-216. [Medline]
  10. Murphy G, Quarles LD. Intact parathyroid hormone: an index of the presence/severity of osteitis fibrosa. J Am Soc Nephrol 1989;262A. abstract.
  11. Andress DL, Maloney NA, Endres DB, Sherrard DJ. Aluminum-associated bone disease in chronic renal failure: high prevalence in a long-term dialysis population. J Bone Miner Res 1986;1:391-398. [Medline]
  12. Weinberg SG, Lubin A, Wiener SN, Deoras MP, Ghose MK, Kopelman RC. Myelofibrosis and renal osteodystrophy. Am J Med 1977;63:755-764. [Medline]
  13. Zingraff J, Drueke T, Marie P, Man NK, Jungers P, Bordier P. Anemia and secondary hyperparathyroidism. Arch Intern Med 1978;138:1650-1652. [Abstract]
  14. Urena P, Eckardt K-U, Sarfati E, et al. Serum erythropoietin and erythropoiesis in primary and secondary hyperparathyroidism: effect of parathyroidectomy. Nephron 1991;59:384-393. [Medline]
  15. Mallette LE, Bilezikian JP, Heath DA, Aurbach GD. Primary hyperparathyroidism: clinical and biochemical features. Medicine (Baltimore) 1974;53:127-146. [Medline]
  16. Falko JM, Guy JT, Smith RE, Mazzaferri EL. Primary hyperparathyroidism and anemia. Arch Intern Med 1976;136:887-889. [Abstract]
  17. Boxer M, Ellman L, Geller R, Wang CA. Anemia in primary hyperparathyroidism. Arch Intern Med 1977;137:588-593. [Abstract]
  18. Shasha SM, Better OS, Winaver J, Chaimovitz C, Barzilai A, Erlik D. Improvement in the anemia of hemodialyzed patients following subtotal parathyroidectomy: evidence for the role of secondary hyperparathyroidism in the etiology of the anemia of chronic renal failure. Isr J Med Sci 1978;14:328-332. [Medline]
  19. Barbour GL. Effect of parathyroidectomy on anemia in chronic renal failure. Arch Intern Med 1979;139:889-891. [Abstract]
  20. Meytes D, Bogin E, Ma A, Dukes PP, Massry SG. Effect of parathyroid hormone on erythropoiesis. J Clin Invest 1981;67:1263-1269.
  21. Delwiche F, Garrity MJ, Powell JS, Robertson RP, Adamson JW. High levels of the circulating form of parathyroid hormone do not inhibit in vitro erythropoiesis. J Lab Clin Med 1983;102:613-620. [Medline]
  22. McGonigle RJS, Wallin JD, Husserl F, et al. Potential role of parathyroid hormone as an inhibitor of erythropoiesis in the anemia of renal failure. J Lab Clin Med 1984;104:1016-1026. [Medline]
  23. Schober H-C, Winkler R, Schmidt R, Abendroth K, Klinkmann H. Bone histomorphometry in recombinant human erythropoietin-treated patients on chronic haemodialysis. Contrib Nephrol 1991;88:127-135. [Medline]
  24. Bia MJ, Cooper K, Schnall S, et al. Aluminum induced anemia: pathogenesis and treatment in patients on chronic hemodialysis. Kidney Int 1989;36:852-858. [Medline]
  25. Roger SD, Stewart JH, Harris DCH. Desferrioxamine enhances the haemopoietic response to erythropoietin, but adverse events are common. Nephron 1991;58:33-36. [Medline]
  26. Chan YL, Furlong TJ, Cornish CJ, Posen S. Dialysis osteodystrophy: a study involving 94 patients. Medicine (Baltimore) 1985;64:296-309. [Medline]
  27. Rao DS, Mohini R. Erythropoietin dose requirement in patients treated with oral calcitriol or with parathyroidectomy. In: Norman AW, Bouillon R, Thomasset M, eds. Vitamin-D: gene regulation, structure-function analysis and clinical application. Berlin, Germany: Walter de Gruyter, 1991:881-2.

 

This Article
-Abstract

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved.