ORIGINAL RESEARCH The Bone Response to Non–Weight-Bearing Exercise Is Sport-, Site-, and Sex-Specific Faidon Magkos, MS, Stavros A. Kavouras, PhD, Mary Yannakoulia, PhD, Melina Karipidou, MS, Silia Sidossi, MS, and Labros S. Sidossis, PhD Objective: To examine whether skeletal adaptations to chronic non–weight-bearing exercise depend on the type of aquatic exercise (swimming or water polo) as well as on sex (men or women). Design, Setting, and Participants: This was a cross-sectional study at the Laboratory of Nutrition and Clinical Dietetics, Department of Nutrition and Dietetics, Harokopio University, Athens, Greece. A total of 43 water polo players, 26 swimmers, and 30 sedentary individuals, aged 17 to 34 years, were recruited (52 men, 47 women). Main Outcome Measures: Bone mineral content (BMC) and areal bone mineral density (aBMD) of the total body and of various subregions. Results: Compared with controls, swimmers had lower leg and total aBMD (P , 0.05), whereas water polo players had lower leg but higher arm and trunk aBMD (all P , 0.05). Swimmers and water polo athletes differed at the arms (men only), trunk, and total body (all higher in water polo players, at P , 0.05). Bone adaptations to water polo playing were unaffected by sex. Female swimmers, but not male swimmers, had 13% higher arm BMC than controls (P , 0.05), whereas male swimmers, but not female swimmers, had 12% lower leg BMC than controls (P , 0.05). Conclusions: Athletes participating in long-term water polo playing and swimming have substantially different total and regional aBMD. The effect is not mediated by sex in water polo players; however, sex may mediate the differences between swimmers and controls. Whether the observed differences between athlete groups and sexes arise from different bone adaptations to activity or from other factors cannot be answered by the current data. Clinical Relevance: Water polo playing may be preferable over swimming for maintaining bone health; both types of aquatic exercise at the elite level of participation, however, have unfavorable effects on the lower limb bones. Key Words: skeleton, aquatic exercise, swimming, water polo, sex differences (Clin J Sport Med 2007;17:123–128) INTRODUCTION Evidence to date suggests that the bone response to exercise is governed by site specificity and load dependency: only the bones subjected to additional mechanical or gravi- tational forces will increase their mass and structural density, and this will happen as long as externally applied loads exceed normal loading patterns. 1 By virtue of these principles, each skeletal site is expected to respond uniquely to different exercise types. A substantial body of literature, both cross- sectional and longitudinal, supports this notion. 2–6 Sports performed in a weight-supporting environment that primarily involve active loading of the bones because of muscular contractions, such as swimming, 7–19 have not been shown to promote bone mass development. Comparably less is known about the effects of water polo playing. This sport is also practiced in an environment of relative weightlessness, and, despite relevant studies being scarce, 20,21 it is generally believed to have no major impact on bone, insofar as water polo players and swimmers are grouped together by some investigators as ‘‘aquatic’’ athletes. 22,23 Direct evidence to support this assertion, however, is lacking. Another important issue that has not been thoroughly addressed is whether these 2 types of non–weight-bearing exercise confer the same bone adaptations (if any) in men and women. Sex steroid hormones have pleiotropic actions on bone metabolism 24 and modulate the bone response to mechanical loading. 25 Although many investigators have independently examined male and female swimmers, only a few have simultaneously recruited athletes from both sexes. 26,27 Unfortunately, they too treated their data separately, so no inference about the possible interaction between sex and exercise can be made. Yet, evidence was presented in support of a sex-dependent effect of swimming. 26,27 There are no relevant studies with respect to water polo. Against this background, the present study used a cross- sectional design to examine total and regional bone adaptations to long-term water polo playing and swimming in male and female athletes. In particular, we aimed at determining whether the skeletal effects of these 2 types of aquatic exercise are comparable (water polo vs swimming), and whether or not they vary by sex (men vs women). MATERIALS AND METHODS Subjects Ninety-nine male (n = 52) and female (n = 47) subjects aged 17 to 34 years were enrolled for the study. The sample Submitted for publication July 27, 2006; accepted November 10, 2006. From the Laboratory of Nutrition and Clinical Dietetics, Department of Nutrition and Dietetics, Harokopio University, Athens, Greece. Correspondence: Labros S. Sidossis, PhD, Laboratory of Nutrition and Clinical Dietetics, Department of Nutrition and Dietetics, Harokopio University, 70 El. Venizelou Avenue, 17671 Athens, Greece (e-mail: lsidossis@hua.gr). Copyright Ó 2007 by Lippincott Williams & Wilkins Clin J Sport Med  Volume 17, Number 2, March 2007 123