A Large Proportion of Elderly Patients with Anemia Seen in the Outpatient Setting Have Unexplained Anemia, Which Is Characterized as a Hypoproliferative, Normocytic Anemia.

Anemia in the elderly, even when seemingly trivial, is associated with substantial morbidity and even mortality. Approximately 10% of elderly women and 11% of elderly men have anemia, but the cause of the profound production defect is only partly understood. Moreover, one third of these patients have anemia with no discernable etiology, that is, unexplained anemia (UA). The characteristics of, pathogenesis, clinical course, and optimal therapeutic options for patients with UA remain poorly described. Accordingly, we undertook a study designed to evaluate the several causes of anemia in this population, anticipating that a full hematologic evaluation would explain the majority of cases hitherto described as UA. We enrolled independently living outpatient men and women 65 and older with an ECOG performance status of 2 or better with anemia as defined by World health Organization (WHO) criteria. Each patient underwent a comprehensive hematologic evaluation. Each peripheral smear was reviewed by the study investigators. Further evaluation was dependent on the initial findings. All those with macrocytosis (mean corpuscular volume (MCV) ≥ 100) or dysplasia on the peripheral smear were recommended to have a bone marrow evaluation to rule out myelodysplastic syndrome (MDS). Iron deficiency was defined as either a response to iron supplementation or standard laboratory criteria. Anemia due to renal insufficiency was defined as an estimated glomerular filtration rate (GFR) of less than 30 ml/min/1.73 m². Patients were categorized as having UA if no cause of the anemia was found. Patients had ongoing follow-up in hematology clinic. Forty-six patients were enrolled and 36 have completed their evaluation. The median age was 78 (range 66–91). Sixty-seven percent of enrolled patients were white, and 87% were men. The mean hemoglobin (hgb) was 11.5 g/dL in men and 11.2 in women. Eight of 36 (22%) had anemia due to iron deficiency. Five of those with iron deficiency had an MCV > 81 fL at the time of diagnosis of iron deficiency. Two of 36 (6%) had findings suspicious for MDS but declined a bone marrow evaluation, and 7 of 36 (19%) had anemia due to multiple causes, categorized as multifactorial anemia. One patient each was found to have chronic myelomonocytic leukemia, anemia due to renal insufficiency, and large granular lymphocyte disease. Despite our full evaluation, 15 of 36 (42%) were categorized as having UA. In those with UA, the mean hgb was 11.7 in men and 11.2 in women, the mean MCV was 93 (SD 4.9), the mean erythropoietin level was 17.1 mU/ml (SD 8.5), the mean absolute reticulocyte count was 46 K/uL (SD 17.7) and the mean GFR was 66 (SD 13.4). Five of the 15 had a GFR between 30 and 60. Despite a full hematologic evaluation, 42% of this cohort had UA. Those with UA had a normocytic, hypoproliferative anemia. The impact of a GFR between 30 and 60 in this population requires further investigation. Further analyses of bone marrow samples in those with UA is planned. Also of clinical importance, 6% of patients appeared to have MDS (not confirmed thus far with a bone marrow evaluation), and it should be emphasized that a normal MCV does not preclude iron deficiency anemia.