RESPONSE

Abstract

Dear Editor-in-Chief: Each of the three components of the female athlete triad (the Triad) can alone, or in combination, negatively affect health and, for athletes, impair athletic performance (4). Our study has confirmed that the Triad exists both in elite athletes and in active controls, as suggested by Otis et al. in the 1997 American College of Sports Medicine (ACSM) Position Stand (4). The fact that it exists also among active controls does not and should not negate our concern for athletes. It seems to us that the letter writers are criticizing our paper because it does not meet their standards for a laboratory study, which it was never meant to be. The comments in both letters seem to be directed at what we should have done if we had infinite time, money, laboratory resources, and did not enroll elite athletes who are traveling about 230 d·yr−1. In our project we conducted a questionnaire study (6,7) of all subjects, a clinical interview for those athletes and controls participating in the clinical part of the project, took DXA measurements, and drew blood samples to analyze hormones. In the two papers mentioned by the writers, the following variables were used to investigate the prevalence of the Triad: Disordered eating/eating disorders (self-reported use of one or more pathogenic weight control methods, a high score on the drive for thinness (DT) or body dissatisfaction (BD) subscale of the Eating Disorder Inventory, self-reported past or clinically verified eating disorders). Menstrual dysfunction (present primary amenorrhea, secondary amenorrhea, oligoamenorrhea and short luteal phase, or a lifetime prevalence of primary or secondary amenorrhea). Low bone mineral density (BMD) (a Z-score below −1.0). Hormone values were not included in these papers. The 1997 ACSM Position Stand (4) included "inadvertently failing to balance energy expenditure with adequate energy intake" as a form of disordered eating. Also, in the International Olympic Committee Position Stand (2), low energy availability is included as part of the component related to disordered eating/eating disorders. This is in our opinion very important in terms of prevention or early recognition of the Triad. However, neither position stand suggests how and when energy balance should be measured. Evaluating energy status during the clinical evaluation could result in a positive or negative balance; four hours later you could have the opposite depending on the athletes' situation. Therefore, before using energy balance as a criterion we need to discuss when and how to measure it. We could have decided our own definition as we did with menstrual dysfunction and low BMD, but due to the questions related to when and how and the fact that doing energy expenditure and intake calculations are time consuming and would interfere even more with the athletes schedule, we did not include this measurement. We have discussed our use of the term "short luteal phase," criticized by Dr. Williams and Dr. De Souza, in our manuscript. Our definition may be debatable. Daily hormonal measurements of LH and progesterone in the blood or urine for multiple consecutive menstrual cycles have been referred to as the best way to diagnose short luteal phase (1). This was not possible in this study and has been taken into consideration when discussing the results. As a consequence, the prevalence of menstrual dysfunction may be even higher than what is reported in our study and underlines the need for prevention, and treatment of the Triad. We do not understand Dr. Louck's statement that data on secondary risk factors are lacking in order to diagnose osteoporosis in these women. Major risk factors are hypoestrogenemia as well as eating disorders. A total of 10.2% of the athletes in our study were diagnosed both with low BMD and eating disorders/disordered eating and 5.4% were diagnosed with both low BMD and menstrual dysfunction (7). In addition, athletes with eating disorders had lower total body and lumbar spine BMD than athletes without eating disorders (5). The International Society of Clinical Densitometry (ISCD) has taken the position that osteoporosis should not be diagnosed in premenopausal women and children on the basis of bone density alone (3). Instead, the ISCD recommends that bone density be evaluated with reference to possible clinical problems affecting bone metabolism. No DXA Z-score is provided as a guideline. Is the diagnosis left up to each physician? In addition, we have not seen any medical group requiring two scans to diagnose a woman as osteoporotic as suggested by Dr. Loucks. Jorunn Sundgot-Borgen Monica Klungland Torstveit The Norwegian University of Sport and Physical Education Oslo, Norway