Growth hormone (GH) replacement in adulthood results in variable bone responses as a function of the gonadic hormonal milieu. We performed a retrospective analysis of a large cohort of adult males and females with confirmed GH deficiency (GHD) prior to treatment and during 3 years of replacement therapy. Potential confounders and effect modifiers were taken into account. Sixty-four adult patients with GHD (20 females and 44 males; mean age 34 years, range 18-64) were included in the analysis. GH replacement induced a different effect on bone in males compared to females. Bone mineral content increased in males and decreased in females at the lumbar spine, total femur, and femoral neck; bone mineral density showed a similar trend at the lumbar spine and femoral neck. There was no significant gender difference in bone area at any measured bone site. In both sexes we observed a similar trend for serum markers of bone remodeling. Sex predicted bone outcome on multivariate analysis, as did age, onset of GHD (childhood/adulthood), pretreatment bone mass, baseline body mass index (BMI), and BMI change during GH replacement. Serum IGF-I levels during treatment did not show any relationship with bone outcome at any measured site. This study confirms that bone responsiveness to GH replacement in adult GHD varies as a function of sex even after controlling for potential confounders and highlights the importance of other cofactors that may affect the interaction between GH replacement therapy and bone remodeling.