Anemia in the elderly: clinical implications and new therapeutic concepts

Anemia in the elderly (defined as people aged > 65 years) is common and increasing as the population ages. In older patients, anemia of any degree contributes significantly to morbidity and mortality and has a significant effect on the quality of life. Despite its clinical importance, anemia in the elderly is under-recognized and evidence-based guidelines on its management are lacking.

Part of the problem here relates to its definition, which is based on WHO-criteria established in 1968.1 The WHO definition of anemia is hemoglobin (Hb) less than 130 g/L in men, Hb less than 120 g/L in non-pregnant women, and less than 110 g/L in pregnant women. Hemoglobin levels decline with age, and there has been a debate as to whether these values are applicable to older people, although there is no accepted alternative definition of anemia in this age group. Most clinicians, however, accept this definition and are of the opinion that the normal hemoglobin range should not be lowered for older people because of its association with morbidity, mortality and hospitalization. The challenge of defining a normal hemoglobin range lies in part in finding a cohort of ‘healthy’ elderly subjects confounded by the high prevalence of comorbidities and impairments in parallel with advancing age. In the analysis of Cheng et al.,2 an important proportion (60%) of the older adults were excluded due to frequent diseases including obesity, arterial hypertension, diabetes, recent treatment for anemia, or recent surgery or hospitalization. Thus, the introduction of selection bias limits the practical applicability of this approach. Another approach is based on the definition of Hb concentrations that are optimal for the clinical outcome of elderly subjects. Based on the distribution of Hb levels, the elderly can be grouped into quartiles or quintiles, revealing inverse J-shaped correlations with unfavorable outcome. An increased mortality was found in the lower quintile (<137 g/L for men; <126 g/L for women) as defined in the Cardiovascular Health Study cohort.3 Similarly, anemia correlated with increased hospitalization4 and mortality.54 Thus, a suggested optimal Hb value to avoid hospitalization and mortality was 130–150 g/L for women and 140–170 g/L for men, suggesting a redefinition of cut-off points for anemia.

Nonetheless, based on the WHO definition, studies have estimated that, in people over 65 years, the prevalence of anemia is 12% in those living in the community, 40% in those admitted to the hospital, and as high as 47% in nursing home residents. All in all, an estimated 17% of those over 65 has been found to be anemic (Table 1).146 Based on this proportion, the current number of anemic elderly persons in the European Union is estimated to be as many as 15 million. This number is likely to increase dramatically in the coming years due to an aging population in Western societies.138

Table 1.Prevalence and subtypes of anemia in the elderly in selected studies.

Anemia in the elderly is particularly relevant as it has a number of serious consequences. Anemia has been associated with a higher incidence of cardiovascular disease,4 cognitive impairment,15 decreased physical performance and quality of life,1816 and increased risk of falls and fractures.16 Furthermore, presence of anemia is significantly associated with longer hospital stays194 and with an increased risk of mortality, in particular, mortality related to cardiovascular disease.4 More importantly, anemia might be an early sign of a previously undiagnosed malignant disease.20

Causes of anemia in the elderly are divided into three broad groups: nutritional deficiency, anemia of chronic disease (ACD) and unexplained anemia (UA). These groups are not, however, mutually exclusive. In any given patient, several causes may co-exist and may each contribute independently to the anemia. Nutritional deficiencies represent a treatable subgroup and include lack of iron, vitamin B₁₂ or folate. The most frequent nutritional anemia is due to iron deficiency, which is characterized by low serum ferritin levels and transferrin saturation (Table 1). However, normal/high serum ferritin levels do not rule out iron deficiency, as ferritin represents an acute phase protein, which might be elevated in inflammatory processes and with advanced age. Thus, the diagnosis should be mainly based on decreased transferrin saturation. Diagnosis of iron deficiency should not be an end in itself but should rather be the initiation of a search for its cause, including looking for a possible site of blood loss and for possible underlying malignancy. The pathophysiology of ACD is multifactorial and relates to a reduced efficiency of iron recycling from red blood cells resulting in a functional iron deficiency. There is enhanced apoptosis of erythroid progenitor cells in the marrow, an inadequate production of erythropoietin (EPO) and impaired response to EPO. It has been proposed that elevated pro-inflammatory cytokines such as TNFα, IL-6, IL-1 and macrophage migration inhibitory factor (MIF) underlie ACD and a key mediator is the induction of hepcidin synthesis by IL-6. Hepcidin inhibits iron absorption in the intestine and the release of recycled iron from the macrophages, resulting in an iron-restrictive anemia (reviewed by Weiss and Goodnough21). Unexplained anemia (UA) accounts for approximately one-third of all anemias in the elderly and represents primarily a diagnosis of exclusion, unclassifiable by currently available methods. The pathophysiology is complex and poorly understood. Although undiagnosed malignancy including myelodysplasia,13 previously unrecognized chronic kidney disease, and other uncommon causes may explain a proportion of the UAs, their combined contribution is relatively small. In populations where thalassemia is prevalent, thalassemia trait may account for another proportion of the UAs.108 Dissecting the causes of UA is confounded by the high frequency of co-morbidities in the elderly, age-associated increases in levels of pro-inflammatory cytokines such as IL-6 that may reduce sensitivity of stem cells and progenitors to growth factors and induce hepcidin synthesis in an environment of reduced pluripotent hemopoietic stem cell reserve. Elevated hepcidin levels have been detected in UA, suggesting that inflammatory processes might contribute to anemia in the elderly, involving mechanisms similar to those encountered in ACD.12 Thus, a cause for a large proportion of the UAs remains unclear despite comprehensive hematologic evaluation (Table 1).109

Our great challenge now is to refine the pathological classification of anemia based on the integration of currently routinely available parameters, e.g. ferritin, transferrin saturation (TSAT), reticulocyte hemoglobin content (CHr), pre-inflammation markers and new parameters including plasma hepcidin, erythroferron and soluble hemojuvelin. This should lead to a better understanding of its pathophysiology and the role of hepcidin-targeted therapeutics.

Considering the central role of hepcidin in the pathophysiology of anemia in the elderly, it is not surprising that a plethora of drugs manipulating the hepcidin pathway for therapeutic purposes have been developed22 with some of these agents already being applied in clinical studies.

It is envisaged that a combination of these biochemical and genetic tests could refine classification of anemia in the elderly, leading to the development of individualized treatment algorithms that would facilitate the appropriation of therapeutic options, including ESAs with a target Hb 100–120 g/L, intravenous iron and novel oral iron formulations, as well as drugs directed at hepcidin or ferroportin.22 Blood transfusions should be kept to a minimum.

Anemia of the elderly represents a challenge and a burden for the individual, the community and health care providers. All healthcare providers, including hematologists, should be aware that anemia impacts a significant group within our societies. It is an entity that lies within our ability to diagnose and treat.

Acknowledgments

We thank Claire Steward for help in preparation of the manuscript.

Footnotes

• Reinhard Stauder is Associate Professor of the Department of Internal Medicine V (Haematology and Oncology) at the Innsbruck Medical University, Austria. His main clinical and scientific focus lies in myelodysplastic syndromes, in geriatric oncology, and in anemia in the elderly. His main goal is the development of individualized treatment algorithms in elderly patients. Swee Lay Thein is a Professor of Molecular Haematology/Consultant Haematologist at King’s College London and King’s College Hospital, London, UK. Her main field of interest is inherited and acquired anemias, with a special focus on the hemoglobin disorders.
• Financial and other disclosures provided by the author using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are available with the full text of this paper at www.haematologica.org.