No Association between Serum Ferritin Levels 1 10 g/l and Hair Loss Activity in Women

the lower limits of normal based on reference sample groups. In women of childbearing age, using a cutoff of 10–15 g/l yields a sensitivity of 75% and specificity of 98% [19] . A cutoff of 30 g/l yields a sensitivity of 92% and a specificity of 98% [18] . The reference laboratory cited in this study uses 10  g/l as lower limit of normal for menstruating women, and 30  g/l for children, men and nonmenstruating women. Increased shedding of telogen hair results either from synchronous transition of hair follicles from the growing (anagen) to the resting (telogen) stage of the hair cycle, or from progressive shortening of duration of anagen [20, 21] . The former mechanism underlies TE, the latter FPHL. Different methods have been adapted for quantification of hair loss. Of these, the trichogram is the first technique standardized for this purpose, and has probably been the most widely used in dermatologic practice. Initially developed in 1957 by van Scott’s group, subsequently other authors modified and standardized the technique defining uniform and strict criteria for reliably assessing the different morphological hair root structures to ensure comparable results [22, 23] . A telogen rate of 1 15% in scalp hair is considered pathologic in women. In TE, the telogen rate is elevated in all regions of the scalp, including the occipital area; in FPHL, the telogen rate is elevated in the frontal and centroparietal scalp region and spares the occipital scalp. Hair loss is a common complaint in women and affects over 25% of women in developed countries [1] . Female pattern hair loss (FPHL) and diffuse telogen effluvium (TE) account for the majority of cases [2–7] . In both conditions, low iron stores have been considered a possible contributing factor. Therefore, assessment of serum ferritin levels is generally recommended as part of the routine investigation, and dermatologists commonly prescribe iron supplementation in women under the assumption that low iron stores may cause hair loss. Contradictory observational data have so far failed to support this practice, however. Although nonanemic iron deficiency as an etiologic factor for diffuse hair loss in women was first postulated by Hard in 1963 [8] , it is not until recently that the significance of iron stores as assessed by serum ferritin levels in women with hair loss has been systematically studied [9] . Various observational studies have evaluated the association between decreased ferritin levels and hair loss and resulted in opposing conclusions [10–16] . The controversy starts with a debate over what is the normal serum ferritin level for women [17] , and is further complicated by the use of different reference ranges by different laboratories, based on individual interpretations of the literature on this subject. A cutoff point of 10–15 g/l is considered to yield a sensitivity of 59% and a specificity of 99% for diagnosing iron deficiency [18] and is used by many laboratories as Published online: February 29, 2008