We reported previously that a gender-specific defect of acquisition of lumbar bone mass plays an important role in the pathogenesis of male idiopathic osteoporosis (IO) and that there is a strong heritability of this maturational defect, which is particularly manifest in sons of men with IO. A hypothetical role of an altered sex steroid status and/or of a (TTTA)(n)- repeat polymorphism of the aromatase (CYP19) gene in male IO remains to be established. We evaluated bone mineral density (BMD) at the lumbar spine and femoral neck in 64 male IO probands (selected on the basis of a z-score of -2 or less), 21 of their sons, 41 of their brothers, and 126 healthy, age-matched controls. Serum testosterone (T), estradiol (E(2)), and SHBG levels were measured by immunoassays. Free T (FT) and free E(2) (FE(2)) levels were calculated from total T, E(2), SHBG, and albumin concentrations using a previously validated equation. Probands, sons, and brothers had lower body weight than age-matched controls, with mean differences of 5.0, 4.6, and 4.0 kg, respectively. In probands, sons, and brothers, SHBG levels were higher compared with controls. Significantly lower FE(2) levels were observed in probands and sons compared with their respective controls (P < 0.05 and P < 0.01, respectively). The brothers had nonsignificantly lower FE(2) levels compared with their controls. In the total group of sons with significantly lower FE(2) levels, tertile analysis according to lumbar spine BMD showed that only in the subgroup of sons belonging to the lowest tertile both FE(2) and FT were decreased compared with their controls. The differences in FE(2) levels in IO probands and their sons remained significant after adjustment for body mass index (BMI), even though in multivariate analyses BMI was a major determinant of BMD. The frequency distribution of the CYP19 gene (TTTA)(n)- repeat length (determined by fragment analysis, GeneScan) was not different between men with IO and their controls. In conclusion, the finding of a relative FE(2) deficit in both men with IO as well as their affected sons, even after adjustment for BMI, suggests that estrogen-related perturbances may be involved in the pathogenesis of the deficient acquisition of peak bone mass in male IO.