Clinical Question: Does iron therapy help with fatigue in nonanemic women?

Setting: Outpatient (primary care)

Study Design: Randomized controlled trial (double-blinded)

Synopsis: Researchers enrolled 144 women with fatigue from primary care centers in Switzerland. Women with anemia, chronic fatigue syndrome, or what was broadly defined as “obvious physical or psychiatric cause for fatigue” were excluded from the study. Most of the women (85 percent) had serum ferritin levels of less than 50 ng per mL, despite having normal hemoglobin levels (greater than 11.7 g per 100 mL), and 51 percent of the women had serum ferritin levels less than 20 ng per mL. Allocation was concealed from the enrolling investigators, although they were unable to enroll consecutive patients who presented.

After the women indicated their degree of fatigue using a 10-point visual analog scale (with 10 being the worst), they were randomized to receive placebo or 80 mg per day of elemental iron as ferrous sulfate for four weeks. Although the tablets were identical in appearance, no attempt was made to completely blind patients by adding anything to the placebo to cause stool discoloration, and patients were not asked to guess whether they received iron or placebo. As a result, some patients may have known they were receiving iron, making them more inclined to report that it worked.

In the women treated with iron, fatigue decreased from a baseline score of 6.4 to 4.5, and in the women treated with placebo to 5.6 (difference: 0.97 of 10; 95 percent confidence interval, 0.32 to 1.62). Although statistically significant, the improvement was small and the women still reported fatigue on this simple scale. Also, the response occurred only in women with a serum ferritin level of less than 50 ng per L.

Bottom Line: Women with unexplained fatigue, especially those with low-normal levels of serum ferritin, report a small clinical improvement after receiving iron supplementation for four weeks. Iron is not a panacea; fatigue levels were still reported as 4.5 out of 10 after one month of treatment. (Level of Evidence: 1c)