Day-to-day variation in iron status indexes is similar for most measures in elderly women with and without rheumatoid arthritis

Chronic disease and iron deficiency are the most common causes of anemia in the elderly (1). Clinicians rely on biochemical laboratory tests for assessing iron status to assist them in [...]
Objective To determine the day-to-day variation in biochemical measures of iron status in a group of elderly women with rheumatoid arthritis compared with a group of healthy elderly women. Design Venous blood samples were collected from each subject on 3 nonconsecutive days during a 2-week study period; subjects had fasted overnight. Variability in hemoglobin level, hematocrit value, serum iron concentration, total iron-binding capacity, transferrin saturation, serum ferritin concentration, and plasma transferrin receptor level was determined. Subjects Two groups of women, one with rheumatoid arthritis (n=10) and another that was apparently healthy (n=10). Statistical analyses Variance component analysis was used to estimate the biological variation ([[Σ.sup.2].sub.day]) and analytic variation ([[Σ.sup.2].sub.rep]) for each iron index. The coefficient of variation (CV) for each variance component was calculated: coefficient of biological variation = [CV.sub.day], coefficient of analytic variation = [CV.sub.rep], and coefficient of a single future determination = [CV.sub.fd]. Results The [CV.sub.rep] for all iron indexes was smaller than the [CV.sub.day] in both groups. The [CV.sub.day] was considerably higher for serum iron concentration and for transferrin saturation than for the other indexes in both groups (16.6% and 16.6% in healthy subjects and 33.6% and 28.2%, respectively, in subjects with rheumatoid arthritis). The higher [CV.sub.day] for serum iron concentration and transferrin saturation translated into a higher [CV.sub.fd] for these indexes. Because of the higher variance for these two indexes, more sampling days were required for reliable estimates. [CV.sub.day] and [CV.sub.fd] for plasma transferrin receptor level were relatively low. Conclusions These findings corroborate our previous finding that variation of serum ferritin concentration in the elderly is lower than that demonstrated in younger populations. This aging effect persists in the presence of rheumatoid arthritis. Fasting appeared to improve reliability in the determinations for serum iron concentration and transferrin saturation. Variability estimates for the indexes other than serum iron concentration and transferrin saturation were not altered by the inflammation of rheumatoid arthritis. Plasma transferrin receptor level is a reliable index for assessing iron status in populations with rheumatoid arthritis. J Am Diet Assoc. 1996; 96:247-251.