ANALYSIS

bmj.com/podcast ЖЖListen to a podcast interview with the authors of this analysis article Sex, health, and athletes Recent policy introduced by the International Olympic Committee to regulate hyperandrogenism in female athletes could lead to unnecessary treatment and may be unethical, argue Rebecca Jordan-Young, Peter Sönksen, and Katrina Karkazis

he International Olympic Committee nations, Semenya said, “I have been subjected to Caster Semenya: questions about gender (IOC) and international sports federa- unwarranted and invasive scrutiny of the most tions have recently introduced policies intimate and private details of my being.”7 to have significantly higher testosterone levels requiring medical investigation of Intended to improve the handling of such cases, than either non-elite athletes (as measured by women athletes known or suspected these policies have nevertheless generated con- saliva13) or non-athletes.14 The only large scale Tto have hyperandrogenism. Women who are troversy.3 8-10 Most of the debate, however, has study of testosterone in elite athletes showed that found to have naturally high testosterone levels focused on questions of fairness, such as the logic 11 of 234 (5%) of elite female athletes sampled and tissue sensitivity are banned from competi- of using testosterone levels as grounds for exclu- immediately after competition had testosterone tion unless they have surgical or pharmaceutical sion while allowing all other natural variations values >10 nmol/L, and 32 (14%) had a testos- interventions to lower their testosterone levels.1 2 among athletes that affect performance, rather terone level >2.7 nmol/L, the upper limit of the Sports authorities have argued that women than the medical justification. normal reference range.14 Using these data, up to with naturally high testosterone have an unfair 14% of women athletes could be investigated for advantage over women with lower levels, and thus How many athletes are affected? “hyperandrogenism.” the primary aim of the policies is to address this The hyperandrogenism policies now extend perceived advantage. However, sports bodies have beyond elite athletes (international level compet- Defining and detecting hyperandrogenism also claimed that the investigations are for the itors). The IOC mandates that national Olympic Medically, the question of when raised androgen medical benefit of athletes with hyperandrogen- committees “actively investigate any perceived levels become a problem is complex and is not ism.3-5 We consider this claim in the light of a new deviation in sex characteristics” before registering simply answered by laboratory measurement study of four young athletes (aged 18-21) from athletes,2 and at least one government, India, has of testosterone or other androgens. Most cases developing countries who had gonadectomy and already complied, creating a policy that applies of hyperandrogenism are linked to polycystic “partial clitoridectomy” after being identified as to women athletes at every level.11 Some inter- ovary syndrome (affecting 5-10% of women), hyperandrogenic under these policies.6 The report national federations governing particular sports and medical management in these and other notes that these procedures were not required for (such as track and field, rowing, and football) cases is focused on dealing with patients’ symp- health reasons. These interventions are invasive have adopted similar rules, and these affect ath- toms and identifying any underlying health and irreversible, and given the potential number letes competing below international level. risks. Although hyperandrogenism may precipi- of female athletes affected the report prompts an The number of women affected will depend, tate disease in various organ systems,15 it does important question: do the new policies under- in part, on the threshold for testosterone that is not invariably cause morbidity.16 17 Authorita- mine ethical care? set. Some policies, including those of the IOC and tive guidelines such as those of the American Fédération Internationale de Football Association A­ssociation of Clinical Endocrinologists do not Why were policies on hyperandrogenism (FIFA), set no specific limit for testosterone,2 12 set specific testosterone levels that indicate introduced? and those that do set a threshold use different disease­ and warrant intervention.15 The new rules were made in response to inter- values. For example, the International Association By contrast, the definition and detection of national outrage over the investigation of Caster of Athletics Federations (IAAF) sets the threshold hyperandrogenism in the sports setting consid- Semenya, a South African middle distance run- at 10 nmol/L,1 whereas the Indian policy sets ers only high testosterone and tissue sensitiv- ner, after fellow athletes questioned her sex at the it at 6.9 nmol/L.11 Although few women in the ity.1 2 11 12 Cases seem to be mainly identified 2009 Berlin world championships. After undergo- general population will have testosterone levels through systematic hormonal screening as part of ing intensive medical and psychological exami- this high, elite female athletes have been shown the “athlete biological passport” (an anti-doping programme),1 6 though current protocols distin- KEY MESSAGES guish between “doping” and endogenous testos- New policies require women athletes known or suspected to have hyperandrogenism to lower terone. Not all international bodies have specified testosterone in order to compete the full testing process, but the IAAF procedure, The regulations’ aim of lowering testosterone regardless of athletes’ health, symptoms, and fertility for example, begins with a clinical examina- goals conflicts with the medical approach to hyperandrogenism tion and an endocrine assessment to determine Four young women athletes have had medically unnecessary gonadectomy to comply with the policies if there “are grounds to indicate an athlete with Sports policies regulating hyperandrogenism in women athletes are not in keeping with best ethical hyperandrogenism”; if so, a full examination and practice diagnostic process ensues, consisting at minimum

20 BMJ | 10 MAY 2014 | VOLUME 348 ANALYSIS

st­eroid production, and occasionally cause seri- have no role in interventions undertaken for ath- ous cortisol deficiency.19 letes’ eligibility or health. Balancing side effects with efficacy in lowering Poor and rural women from developing testosterone can also be difficult. Medical care countries seem to be most affected,6 22 amplify- requires that physicians weigh patients’ discom- ing concerns about threats to autonomy and the fort from symptoms and concern about metabolic possibility for coercion. When pharmacological indicators (such as insulin resistance and choles- intervention or gonadectomy is a precondition terol levels) against the presence and future risk for eligibility to compete, an athlete has to make of side effects. a profound life and health altering decision for non-medical reasons. These are not merely Gonadectomy individual decisions: athletes are embedded in Women with 46,XY karyotype are especially likely families, teams, organisations, and even nations to be affected by the sports bodies’ policies because that depend on them to compete. Athletes can be JOHN GILES/PA JOHN they often have very high testosterone levels. Gona- “regarded as vulnerable to undue, even extreme of full physical and gynaecological examinations, dectomy was until recently the standard care for situational pressures arising from the decision- endocrine assessments of blood and urine, medi- these women because of the risk of a germ cell making environment,”23 especially when a com- cal and family histories, and psychological assess- tumour of the gonads. However, because tumour petitive career is also a path to economic mobility ment.1 The four athletes with hyperandrogenism risk varies with specific diagnoses, together with and stability. described in the recent report also had karyotyping the serious health consequences of gonadectomy, The lifelong burden of these interventions will and genetic analysis, abdominal-pelvic magnetic a recent review in BJU International concluded that sit especially heavily on poor women. Long term resonance imaging, and radiography to determine the tumour risk is low enough in most cases that pharmaceutical management of testosterone lev- bone mineral density and composition.6 gonadectomy is not warranted.20 els is costly, and most drug combinations require The sport policies are likely to disproportion- Gonadectomy will cause hypogonadism, com- multiple doses a day. Management of side effects ately affect women with intersex conditions in promising bone and muscle strength and risking and lifelong care by medical specialists add to which testosterone is especially high, such as chronic weakness, depression, sleep disturbance, the cost. None of the sport governing bodies indi- congenital adrenal poor libido, adverse cates that they will pay for these interventions; the hyperplasia, androgen When testosterone is high because effects on lipid profile, IAAF, for example, explicitly states that it will not insensitivity syndrome, of natural physiological variations, diabetes, and fatigue. It pay. Finally, apart from the cost, specialist long and 5α reductase defi- sports authorities should not require will necessitate lifetime term follow-up may be made necessary by the ciencies. These women medical interventions to lower it hormonal replacement, interventions. Such care may not be readily avail- may have no health which imposes a poten- able in rural and remote regions. complaints stemming from the androgen excess tial financial burden. Gonadectomy also makes Women athletes will continue to be identified as itself. Women with intersex conditions may also women sterile . having high testosterone levels through universal be investigated if they have not had genital “femi- The decision about whether, and when, to anti-doping screening. When testosterone is high nisation” surgery: doping officers observe athletes remove gonads has very high stakes, especially for because of natural physiological variations, sports during urination, and they may report a woman young people. Notably, gonadectomy is not men- authorities should not require medical interven- perceived to have an enlarged clitoris.6 Thus, tioned as a treatment option in the guidelines of tions to lower it. The interventions are too serious, under these policies, women without symptoms the American Association of Clinical Endocrinolo- especially given that sports officials have said, or health complaints may have medically unnec- gists.15 Nevertheless, this was the treatment used “Women with some DSDs [disorders of sex devel- essary intervention with long term consequences for the four athletes studied in the recent report opment] have no more competitive advantage in order to continue their careers. despite the authors’ acknowledgment that leav- than other elite athletes with favourable genetic ing the women’s gonads in place posed “no health characteristics.”24 The testosterone based eligibil- Pharmacologically lowering testosterone risk.”6 All four have 5α reductase deficiencies. ity policies turn standard medical decision mak- The policies’ requirement to lower testosterone ing about hyperandrogenism on its head. Rather regardless of the presence of symptoms raises Do the new policies undermine ethical care? than pegging treatment decisions to women’s multiple concerns. The process is not simple—it is Approaching hyperandrogenism as a sports prob- overall wellbeing, symptoms, fertility goals, self not always possible to lower testosterone pharma- lem raises ethical concerns about designating pos- image, physical functioning, and risks (if any) cologically—and can take months. Among a large sibly benign physical variation as “unhealthy,” associated with a specific diagnosis, intervention cohort of women receiving suppressive therapies, resulting in potentially unnecessary medicosur- is mandated when officials decide that naturally roughly two thirds experienced important side gical intervention and possible neglect of the long occurring testosterone confers unfair advantage. effects.16 These side effects can be serious for an term consequences of interventions. This seems to undermine ethical care. athlete, including diuretic effects that cause exces- The most striking and troubling illustration Rebecca M Jordan-Young associate professor, Barnard sive thirst, urination, and electrolyte imbalances; of these concerns is that all of the athletes in College, Columbia University, New York, USA 6 Peter H Sönksen emeritus professor of endocrinology, disruption of carbohydrate metabolism (such as the recent report had “partial clitoridectomy.” St Thomas’ Hospital and King’s College, London and glucose intolerance or insulin resistance); head- Clitoridectomy­ is not medically indicated, does Southampton University, Southampton, UK ache; fatigue; nausea; hot flushes; and liver toxic- not relate to real or perceived athletic “advan- Katrina Karkazis senior research scholar, Stanford ity.16-18 Liver function and blood chemistry must tage,” and is beyond the policies’ mandate. More- University, 1215 Welch Road, Stanford, CA 93405, USA Correspondence to: K Karkazis [email protected] be carefully monitored during and after treatment over, this technique is long eschewed because it References and supplementary information can be found because some anti-androgens have the poten- has poor cosmetic outcomes and damages sexual on bmj.com. 21 tial to damage the liver, disrupt other ne­cessary sensation and function. Clitoral surgery should Cite this as: BMJ 2014;348:g2926

BMJ | 10 MAY 2014 | VOLUME 348 21