Erythropoietin

EPO, a glycoprotein secreted by interstitial cells in the kidney (3), has a major role in regulating erythropoiesis in humans. EPO is primarily regulated by blood oxygen content and secretion is stimulated by hypoxia (4, 5). EPO stimulates hematopoiesis by increasing proliferation (6) and preventing apoptosis (3, 7) of erythroid progenitor cells. The primary site of action of EPO in the bone marrow is the late stage colony-forming unit erythroid (CFU-E). EPO induces these cells to both proliferate and mature as well as to become resistant to apoptosis (3, 7).

Erythropoietin and Aging

Under normal conditions (i.e., without disease) erythropoietin serum levels increase with advancing age. In this regard, the data from cross-sectional studies are both limited and conflicting (8-14), but that from longitudinal study is quite revealing. For example, samples obtained from the National Institute on Aging (NIA) Baltimore Longitudinal Study of Aging (BLSA) clearly demonstrated a gradual and sustained rise in serum EPO among healthy individuals as they age (Fig. 9.1) (15). Importantly, for those in the BLSA analysis who were to develop hypertension and/or diabetes during their tenure as study participants, the slope of the incline was significantly less pronounced (Fig. 9.2). Thus, it was speculated that renal erythropoietin production or secretion was negatively influenced by the same disease processes that impair renal excretory function (i.e., diabetes and hypertension). While there is evidence that erythro-poietin serum levels tend to increase with advancing age, for some the rise is of insufficient magnitude to maintain a hemoglobin concentration in the normal range (2). This observation has also been reported for patients with diabetes, even without associated measurable renal insufficiency (16).

Erythropoietin Insufficiency in Renal Failure

The prevalence of anemia increases incrementally with declines in creatinine clearance (17, 18). This has been attributed to several mechanisms including: (1) decreased capacity for EPO production or secretion;

0 10 20 30

Time (years)

Fig. 9.1. Serum erythropoietin levels over time in normal adults. Erythropoietin levels measured at two year intervals for a minimum of four determinations, each separated by two years. Each of the thin lines indicates the predicted erythropoietin level for each individual. The bold line represents the population average change over time (15)

0 10 20 30

Time (years)

Fig. 9.1. Serum erythropoietin levels over time in normal adults. Erythropoietin levels measured at two year intervals for a minimum of four determinations, each separated by two years. Each of the thin lines indicates the predicted erythropoietin level for each individual. The bold line represents the population average change over time (15)

Paper Coating Blade

Fig. 9.2. Predicted erythropoietin levels using a linear mixed-effects model as a function of hemoglobin level (g/dL) and time (years). The top panel projects results for those who remained healthy free of hypertension or diabetes (Group 1, n = 84), some of whom, however, developed anemia (n = 15), whereas the remainder did not (n = 69).The bottom panel projects results for those individuals who during the course of their participation on this study developed diabetes, hypertension, or both conditions (n = 59). Some of these individuals developed anemia (n = 15), whereas others did not (n = 44) From Ershler et al. (15)

Fig. 9.2. Predicted erythropoietin levels using a linear mixed-effects model as a function of hemoglobin level (g/dL) and time (years). The top panel projects results for those who remained healthy free of hypertension or diabetes (Group 1, n = 84), some of whom, however, developed anemia (n = 15), whereas the remainder did not (n = 69).The bottom panel projects results for those individuals who during the course of their participation on this study developed diabetes, hypertension, or both conditions (n = 59). Some of these individuals developed anemia (n = 15), whereas others did not (n = 44) From Ershler et al. (15)

(2) inadequate bone marrow response to EPO; and, (3) presence of EPO inhibitors (19). Most currently agree the predominant mechanism is EPO deficiency (20, 21). It has been recognized for several decades that EPO levels were low in anemic hemodialysis patients (22, 23). Additionally, work by Zucker and colleagues demonstrated decreased EPO levels in patients with anemia and renal failure but no measurable inhibitors or inadequate marrow response to exogenous EPO when anemic kidney failure patients were compared with iron deficiency or nonanemic individuals (24). Furthermore, in a sheep model of chronic kidney failure and uremia, excellent erythropoietic response and correction of anemia was observed by treatment with erythropoietin-rich plasma (25). More recently, epidemiological data capitalizing on the development of more sensitive serum EPO assays have further clarified the tight negative correlation of EPO and kidney function. For example, the InChianti Study, a prospective population based analysis of 436 men and 569 women (65 years or older), found that low-hemoglobin levels and lower erythropoietin levels were clearly demonstrable in people with creatinine clearance of 30 mL/min and below (26). Finally EPO replacement has been shown to correct anemia in almost all iron-replete anemia patients with CKD (27). These data, and much more indicate the importance of EPO deficiency in pathogenesis of anemia associated with kidney failure.

Erythropoietin Response in Patients with Iron Deficiency

EPO levels are usually higher in anemic patients with iron deficiency than when anemia is caused by other processes, such as inflammation or renal insufficiency (28-31). Pregnant rats fed with an iron-deficient diet were found to have significantly increased EPO levels (32) and, similarly elevated levels were found in iron-deficient pregnant women (33). Serum EPO levels in rheumatoid arthritis (RA) patients with coexisting iron deficiency and anemia were higher than those with RA and anemia but normal iron stores (34). Thus, to the extent that iron deficiency contributes to anemia, one would expect to observe appropriately elevated EPO levels. And, as a corollary, if such is not observed, contributing factors, such as inflammation or renal insufficiency should be considered. Similarly, EPO levels in iron-deficient elderly anemic patients, although higher than those without iron deficiency, were found to be lower than expected for the degree of anemia by several investigators (2, 10, 11, 13). Thus, we speculate that even for those elderly patients with iron deficiency, the late-life factors that, in composite, result in UA (discussed below) are also contributing to the observed anemia.

Erythropoietin Insufficiency in HIV Anemia

Anemia is an extremely common finding in patients infected with human immunodeficiency virus (HIV). For example, Sullivan and colleagues reported a 1-year incidence of anemia as 37% among patients with clinical AIDS, 12% among patients with CD4+ cell count <200 cells/mm3 in the absence of an AIDS-defining clinical condition, and 3% among HIV-infected individuals with neither clinical nor immunologic AIDS (35). This was particularly remarkable because the criterion for 'anemia' in this survey was a hemoglobin level below 10 g/dL. With HIV infection, the cause of anemia is considered multifactorial and may involve direct bone marrow toxicity by the virus, nutrient deficiencies, and the myelosuppressive effects of certain antiviral drugs (36, 37). Low levels of EPO have been demonstrated consistently in HIV patients and these correlate significantly with the presence of anemia. The adequacy of endogenous EPO response in anemic HIV patients was assessed in 42 subjects and compared to the response observed in patients with anemia of chronic disease or iron deficiency. With comparable degrees of anemia, EPO levels in AIDS patients and those with chronic disease were lower than those with iron deficiency (31). Similar results were also reported by Spivak and colleagues who demonstrated the mean incremental increase in serum immunoreactive EPO levels for a given decline in hemoglobin was significantly less in AIDS patients when compared to those with iron deficiency (38, 39). It is also notable that the use of recombinant human EPO to treat anemia in HIV infected patients, particularly those with low-EPO levels, has been shown to correct anemia and improve quality of life (40, 41).

Erythropoietin Deficiency in Diabetes Mellitus

Anemia is a common complication of both type I and II diabetes mel-litus. In a cross sectional survey of 820 patients with diabetes in long-term follow up, the prevalence of anemia defined as a hemoglobin concentration of <12 g/dL in women and <13 g/dL in men was two to three times greater in diabetic patients with comparable renal impairment and iron stores, as observed in the general population (42). Anemia occurs in patients with only minor derangement of renal excretory function and at any level of glomerular filtration. Furthermore, anemia is generally more severe in diabetic patients than nondiabetics (43, 44). The pathogenesis of anemia in diabetes has been attributed predominantly to reduced levels of EPO (42, 45, 46). In a study conducted by Bosman et al., serum EPO levels in anemic diabetic patients failed to increase and were much lower than the levels in nondiabetic anemic patients and in patients with microcytic anemia (44). In a cross-sectional study of 604 diabetic patients, more than 71% of those with anemia demonstrated functional EPO deficiency and this was independent of the severity of renal impairment (46). Similar results were also reported in type I diabetic patients without overt nephropathy by Cotroneo et al. (47) and Thomas et al. (45). In patients with normal serum creatinine, EPO level was predictive of a more rapid deterioration of renal function (43).

Numerous mechanisms have been proposed to explain the low EPO levels in diabetic patients including presence of structural renal abnormalities and a possible inhibitory effect of advanced glycosylation end products on EPO production (48). Curiously, the presence of anemia in patients with DM has been associated with neuropathy (47, 49, 50) and it has even been proposed that reduced EPO levels are causally related to diabetic auto-nomic polyneuropathy (51) with efferent sympathetic denervation (44, 52). Regardless of the mechanism, EPO insufficiency is an important cause of anemia in patients with diabetes and studies are underway to assess the value of treatment with recombinant EPO in terms of quality of life, physical function and the progression of diabetic complications.

The Role of EPO Deficiency in the Late-Life Occurrence of UA

Investigation into the etiology of anemia in individuals older than 65 years of age revealed a high percentage of cases (15-45%) where no cause was definable (1,2, 9). Currently termed "Unexplained Anemia," the likelihood is that this condition is the result of not one, but several contributing factors which, in composite, result in the anemic condition. Several of these factors are listed in Table 9.1 and discussed in the paragraphs below.

Kidney Function and Aging. Of the many physiological changes that occur with age, there is a gradual decline in renal excretory function (53). In the absence of disease, however, renal production of EPO appears adequate even with the added demands of diminished stem cell proliferative capacity necessitating compensatory levels as evident by the continued rise in serum EPO with age in healthy subjects (15). However, it is possible that this reserve in EPO production capacity runs out in very late-life, or earlier in patients with renal damage secondary to diabetes, hypertension or other disease processes. Thus, it is likely that renal insufficiency, either on the basis of age alone, or in combination with underlying disease contributes to UA by its associated decline in EPO production.

Inflammation and Age-Associated Cytokine Dysregulation. In most elderly, particularly frail elderly, there exist at least one, and frequently several comorbid conditions which have, at their root, inflammation. One common

Table 9.1. Factors contributing to unexplained anemia in the elderly

Factor

Renal insufficiency

Inflammation

Age-associated cytokine dysregulation

Stem cell Myelodysplasia

Mechanism

Certain age-associated diseases including diabetes and hypertension impaired kidney function. Also, aging itself is associated with a gradual decline in GFR.

Cytokine-induced hepcidin inhibits iron absorption and mobilization. Cytokines may also inhibit EPO synthesis or ligand-binding.

As above (Inflammation)

Regenerative capacity diminished

Stem cell proliferative impairment

EPO deficiency involved? Directly

Indirectly and directly

Indirectly and directly

Indirectly

Indirectly

Androgen deficiency Age-associated decline Indirectly feature of chronic inflammatory disease is anemia and to the extent that underlying inflammatory disease is unrecognized, coexisting anemia might be miscategorized as UA. Artz and colleagues (2) recognized this and in their nursing home series of UA patients (mentioned above), excluded from analysis those with an elevated erythrocyte sedimentation rate (ESR) or C-reactive protein (CRP). However, even with this screen, it is likely undiag-nosed inflammatory processes contribute to some extent to the composite picture of UA. Compounding this, it is now generally accepted that the serum levels of certain proinflammatory cytokines rise with age, even in the absence of inflammatory disease. For example, interleukin-6, now considered a biochemical marker of frailty, is typically measurable only in sub-picogram quantities in the absence of acute inflammation in young adults but rises gradually after menopause (or andropause) and its level correlates with several features of the frail phenotype, including sarcopenia, osteope-nia, functional decline, and anemia (54-62). Thus, to the extent that cytokine dysregulation occurs with advancing age, even in the absence of overt inflammatory disease, these same cytokines may contribute to the development of anemia. It is noteworthy that the cytokine most frequently associated with aging and frailty, IL-6, is also the one most commonly associated with inflammation-associated hepcidin upregulation, reduced iron availability, and anemia (63-66).

Stem Cells and Aging. Studies conducted in mice demonstrated reduced proliferative capacity of erythroid progenitor cells with age (67, 68). However, even with this decline, old mice do not become anemic; suggesting in the absence of disease, the acquired intrinsic stem cell defect is compensated by an increase in other factors, including an increase in erythropoietin level. In fact, this may explain the aforementioned rise in serum EPO observed in the BLSA study (15). However, if there is impairment in EPO production, such as in advanced age or with kidney disease, diabetes or inflammatory disease (as described above), insufficiently compensated augmentation of the age-associated decline in stem cell proliferative capacity may be an important component of the pathogenesis of UA.

Myelodysplasia and UA. Myelodysplasia (MDS) occurs with increasing frequency in late-life (69). Although it usually can be diagnosed on the peripheral blood smear by dysplastic white blood cell features associated with macrocytic erythrocytes, it may present as anemia alone, particularly in the elderly (70). Examination of a bone marrow aspirate and biopsy provide more diagnostic accuracy, but if the anemia is mild, this may not be clinically warranted outside a research setting. Thus, some patients considered to have UA may actually have MDS although it is likely this would comprise a small fraction of the total UA pool.

Circulating EPO levels in myelodysplastic syndromes are typically elevated (71, 72) likely representing a compensatory effect to overcome intrinsic defects within the progenitor cell compartments. Bone marrow CFU-E and BFU-E in patients with MDS grew poorly in vitro despite high levels of added EPO (72). The inhibited proliferative response was shown to be associated with the absence of STAT5 (Signal Transducer and Activator of Transription-5) suggesting a defect in the Epo-receptor (EpoR) signal transduction pathway contributes to anemia in myelodysplasia (73).

Androgen Deficiency and UA. Epidemiological studies have demonstrated a negative correlation between circulating androgen levels and hemoglobin concentration in an elderly population (74) raising the speculation that an age-associated decline in androgens is yet another contributing factor to UA. One mechanism that testosterone enhances erythropoiesis is by enhancing renal EPO secretion (75). In older, compared with younger rats, orchiectomy reduced EPO release from kidney cells in response to hypoxia to much greater extent. Furthermore, replacement of testosterone restored the production of EPO to normal levels indicating that in aging rats hypoxia induced release of EPO may be diminished by androgen deficiency (76). In anemic human subjects replacement with testosterone was shown to increase levels of EPO (77). Hence indirectly androgen deficiency could cause anemia by reducing EPO levels.

Although there remains no direct evidence that EPO levels are decreased in patients with androgen deficiency, studies have shown a synergistic action of increasing concentrations of testosterone and EPO in stimulating erythropoiesis in vitro. These effects were completely blocked by pretreat-ment with the androgen antagonists cyproterone and flutamide (78).

Was this article helpful?

0 0
Arthritis Relief Now

Arthritis Relief Now

When you hear the word arthritis, images of painful hands and joints comes into play. Few people fully understand arthritis and this guide is dedicated to anyone suffering with this chronic condition and wants relief now.

Get My Free Ebook


Post a comment