Anemia and Iron Deficiencies among Long-Term Renal Transplant Recipients

Materials and Methods

All patients who visited the kidney transplant outpatient service at the University Hospital of Vienna during a period of 4 wk were included in a cross-sectional study. The patients exhibited stable graft function. Blood chemistry values were determined by using standard methods. Transferrin saturation (TSAT) was calculated as iron level/serum transferrin level × 70.9. Whole blood counts and percentages of hypochromic red blood cells (HRBC) were analyzed with a Technicon H*2 hematology analyzer (Bayer Diagnostics, Tarrytown, NY). Creatinine clearance values were calculated with the equation described by Cockcroft and Gault (2).

Independent associations with hemoglobin levels were examined in multiple stepwise regression analyses. The distribution of skewed data was normalized by natural logarithmic transformation. Logistic regression models were constructed to examine the risks (odds ratios and 95% confidence intervals) for anemia (i.e., hemoglobin levels of <12 g/dl for women and <13 g/dl for men) associated with markers of iron metabolism in the highest (HRBC) or lowest (serum ferritin levels, TSAT, serum iron levels, and serum transferrin levels) quartile of the whole patient population. To avoid multicollinearity, anemia was related to five sets of variables, in which one of five iron markers was entered. In these models, the most important predictors of hemoglobin levels (derived from the stepwise regression analyses) were also considered, i.e., creatinine clearance values below or above the median, azathioprine use (yes/no term), and polycystic kidney disease (yes/no term).

Results

The characteristics of all 438 RTR are indicated in Table 1. The proportions of patients with iron status markers within or outside the reference range are presented in Table 2. The prevalence of anemia among RTR was 39.7%. Iron deficiency was indicated by a proportion of HRBC of ≥2.5% for 88 patients (20.1%). The association of all variables with hemoglobin levels is presented in Table 3. Increased serum ferritin levels were associated with decreased hemoglobin concentrations. Serum transferrin levels were elevated in parallel with hemoglobin levels. The risks for anemia conferred by five markers of iron metabolism are presented in Figure 1. Among anemic or severely anemic patients, HRBC elevations were more frequent than hypoferritinemia or decreased TSAT values. The majority of patients with hemoglobin levels of <11 g/dl presented with HRBC values in the highest quartile (Table 4).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Markers of iron status and risk of anemia among 438 renal transplant recipients (the worst versus the other three quartiles, by logistic regression analysis). Hypochromic red blood cells (RBC): odds ratio (OR), 2.35; 95% confidence interval (CI), 1.48 to 3.75; P = 0.00029; serum transferrin level: odds ratio, 1.96; 95% confidence interval, 1.23 to 3.11; P = 0.004395; serum iron level: odds ratio, 1.56; 95% confidence interval, 0.97 to 2.49; P = 0.063608; serum ferritin level: odds ratio, 0.73; 95% confidence interval, 0.46 to 1.18; P = 0.19989; transferrin saturation (TSAT): odds ratio, 1.02; 95% confidence interval, 0.63 to 1.63; P = 0.942653.

Discussion

Our study suggests that we do not attach enough importance to the diagnosis and management of anemia among RTR in stable condition. This is obviously attributable to the prevalence of unrecognized iron deficiencies, which is related to the poor information obtained from ferritin or TSAT measurements. Only 10.1% of severely anemic patients presented with serum ferritin levels of <12 μg/L, and TSAT values of <15% were observed for 29%. These results underestimate by far the number of patients with functional iron deficiencies, because HRBC values of >2.5% were detected for 46.4% and values in the upper quartile were detected for 52.2% of our severely anemic patients, reaching 64.1% among female patients. These findings strongly suggest that iron deficiencies, as diagnosed by serum ferritin or TSAT measurements, remain undetected in a significant proportion of RTR. Factors related to the prevalence of iron deficiencies may include occult blood losses, blood sampling, and adherence to a low-meat dietary regimen.

Ferritin levels may vary and may be elevated independently of iron stores in conditions such as inflammatory or infectious diseases, malignancies, or iron overload (3,4). In contrast to the negative association of serum ferritin levels with hemoglobin concentrations, C-reactive protein levels demonstrated no association with hemoglobin concentrations in our study. This finding may be attributable to immunosuppressive therapy, which was demonstrated to suppress C-reactive protein production (5). There was also a weak negative independent association of C-reactive protein with HRBC in our study population (data not shown), indicating that increases in HRBC values were not related to inflammation anemia.

In this study, a proportion of HRBC in the highest quartile substantially increased the risk for anemia, whereas decreased serum ferritin levels and decreased TSAT conferred no risk (Figure 1). Therefore, our findings support a concept in which continuous decreases in HRBC levels, even to ≤2%, may indicate gradual improvement of the iron supply (6,7).

A potential limitation of our study may be the lack of analysis of angiotensin I-converting enzyme (ACE) inhibition. The effect of ACE inhibition on hemoglobin levels is a matter of debate. However, the largest study reported to date clearly demonstrated that ACE inhibitor therapy did not decrease hemoglobin levels among RTR with normal hemoglobin levels (8). Another point of concern may be the lack of bone marrow examinations as the standard iron supply test. However, invasive studies are not reasonable for a large patient population.

In summary, we demonstrate that (1) anemia and iron deficiencies are under-recognized among RTR in stable condition, (2) the proportion of HRBC is the most important indicator of iron deficiencies among RTR, and (3) measurements of serum ferritin levels and TSAT are not always sufficient for the diagnosis of iron deficiencies among RTR. References.