Thyroid Hormone Abuse in Elite Sports: The Regulatory Challenge.

Abuse of androgens and erythropoietin has led to hormones being the most effective and frequent class of ergogenic substances prohibited in elite sports by the World Anti-Doping Agency (WADA). At present, thyroid hormone (TH) abuse is not prohibited, but its prevalence among elite athletes and nonprohibited status remains controversial. A corollary of prohibiting hormones for elite sports is that endocrinologists must be aware of a professional athlete's risk of disqualification for using prohibited hormones and/or to certify Therapeutic Use Exemptions, which allow individual athletes to use prohibited substances for valid medical indications. This narrative review considers the status of TH within the framework of the WADA Code criteria for prohibiting substances, which requires meeting 2 of 3 equally important criteria of potential performance enhancement, harmfulness to health, and violation of the spirit of sport. In considering the valid clinical uses of TH, the prevalence of TH use among young adults, the reason why some athletes seek to use TH, and the pathophysiology of sought-after and adverse effects of TH abuse, together with the challenges of detecting TH abuse, it can be concluded that, on the basis of present data, prohibition of TH in elite sport is neither justified nor feasible.

The issue of whether exogenous thyroid hormone (TH) use should be prohibited in elite sport remains controversial (1). Several national antidoping agencies believe that use of TH, namely levothyroxine (l-thyroxine, T4) and its active metabolite liothyronine (triiodothyronine, T3) exceeds legitimate medical indications, is risky for athlete’s health, and is contrary to the spirit of sport in seeking drug-induced performance enhancement, and therefore should be prohibited for elite athletes. However, the World Anti-Doping Agencies (WADA) has not included TH on their annually updated Prohibited List for lack of plausible evidence that use of exogenous TH outside medical indications is performance enhancing; indeed available evidence indicates such abuse is likely to be detrimental to health and performance (2).

WADA was established to maintain fairness in elite sports competition primarily by preventing doping. In practice to achieve this, the WADA Code aims to prohibit performance enhancing substances or methods on an annually renewed Prohibited List. For prohibition by WADA, the WADA Code stipulates 3 criteria, of which 2 must be met for that substance or method to be prohibited. These criteria are (a) actual or potential performance enhancing effects, (b) actual or potential harm to athletes, and (c) violates the spirit of sport. Prohibition of any substance does allow for its use under Therapeutic Use Exemptions for valid medical indications (3, 4); in the case of TH, it is exempted for underlying thyroid diseases. Whether the criteria (a) and (b) are met remains a matter of carefully balanced expert medical and scientific judgment based on available evidence of specific effects of the drug or its pharmacological class. Evidence for performance enhancement should ideally involve objective measures of physical or exercise activities likely to enhance sporting performance, rather than subjective beliefs, or testimonials. Yet when implementing these criteria, it must be recognized that robust evidence for individual substances or methods may not always be available or even feasible according to ethical constraints, such as the impossibility of clinical testing for substances not approved for human use. Conversely, for some activities with possible, or even proven, small ergogenic effects, prohibition is not always tenable. For example, dietary supplements such as caffeine, creatine monohydrate, nitrate, beta-alanine, and sodium bicarbonate all have small but significant performance enhancing effects yet are not prohibited (5). For such substances, difficulties in defining clear ergogenic dose thresholds, their widespread use in society, presence in everyday foodstuffs, and inconsistent global regulation of manufacturing, present insurmountable difficulties to being able to fairly differentiate intentional doping from nondoping uses. Similarly, neither living and/or training at high altitude or simulating these conditions using hypoxic sleeping tents are prohibited physical methods, despite their potential to increase hemoglobin and oxygen-carrying capacity to muscles. These dilemmas are challenging, with the ever-increasing plethora of nonprescription nutraceutical and dietary supplements available over the counter and unregistered drugs over the internet. Unlike registered drugs, which must refrain from unproven therapeutic claims, many of these substances are freely advertised over the internet with unsubstantiated promotional claims from sources with vested interests in marketing illicit substances for bodybuilding, the completely unregulated Dorian Gray twinned portrait of elite sports doping.

This review aims to outline the valid clinical indications for TH use, reasons athletes abuse TH as well as the rationale and pathophysiological basis for such claims, the consequences of nonmedical TH use, strategies for detecting TH abuse, and finally, whether prohibition of TH in elite sport is warranted.

Clinical Uses of Thyroid Hormone

The thyroid gland produces 2 active hormones, thyroxine (T4) and triiodothyronine (T3), within the thyroid follicular cell. Following iodine absorption through the gastrointestinal tract, the enzyme thyroid peroxidase organifies the oxidized iodine and generates compounds which couple to form either T4 or T3 (6). T4 is secreted from the thyroid in larger amounts than T3 (90 vs 30 µg/day), with peripheral conversion of T4 to T3 by outer ring deiodination accounting for 70% to 80% of circulating T3 (7-9). T3 has much higher molar potency than T4 based on the higher affinity binding of T3 to TH receptors α (7-fold) and especially β (70-fold) (10), with higher daily turnover rate (60% vs 10%) and shorter half-life (1 vs 7 days) due to its less avid (10- to 15-fold) binding to the circulating thyroxine binding globulin creating a larger volume of distribution (40 vs 10 liters) compared with T4 (9). These features mean that T4 is both an active hormone as well as a prohormone and provide T3 with a faster onset and offset of action. Different affinities of the 2 TH receptor (TR) isoforms (TRα and TRβ) have been exploited with the aim of optimizing tissue-specific TH effects while minimizing side-effects. TRβ selective thyromimetics, such as Sobetirome and Resmetirom, have undergone clinical trials for metabolic or neuroprotective effects (11-13) but none have yet been marketed. Reverse T3 is a biologically inactive T4 metabolite produced by inner ring deiodination of T4. Its concentration rises in physiological states, such as stress or intercurrent illness, due to switching of the peripheral metabolism of T4, diverting more T4 from conversion to the bioactive metabolite T3 to the inactive metabolite reverse T3. Such changes, prominent in the sick euthyroid (or non-thyroidal illness) syndrome (14-17), occur in a wide variety of acute and chronic systemic illnesses, including catabolic states of extreme exercise and/or undernutrition, where it may represent a conservative, reversible adaptive mechanism to environmental adversity geared toward saving energy and/or activation of immune mechanisms during severe illness, injury, or other catabolic states (14). Multiple randomized, placebo-controlled clinical trials of thyroid hormone (T3 and/or T4) administration for this syndrome have shown no consistent benefit. After prolonged controversy, a consensus against combination thyroid hormone administration was reached (14-16, 18, 19).

The actions of TH are critical for development, reproduction, and metabolism. In patients with impaired TH production, T4 is prescribed, usually at a dose of 1.6 µg/kg/day for full replacement (typically, 125 µg for an 80-kg person). There are few medical indications for T3 use alone reflected in its use representing < 1% of T4 prescribing in the UK (20). A subsiding controversy lingers over whether the addition of T3 to usual T4 regimens is beneficial, with some literature exploring mechanisms of physiological limitations of T4 monotherapy (21), which is postulated to explain a suboptimal quality of life on T4 replacement therapy (22). However, clinical trials showing consistently little benefit have prompted major clinical guidelines to discourage T3 combination use (23-25). More recent work has demonstrated little difference in the treatment effect between T4 monotherapy and T4/T3 combination therapy (26). By contrast, T3 is widely advocated in “alternative” medicine practice (27) and especially preferred in illicit uses, including bodybuilding and doping advertised on the internet (see supplementary data (28)). Hence the prevalence of T3 use among elite athletes might be considered a surrogate for the prevalence of TH abuse.

Prescriptions of T4 are mostly treatment for primary hypothyroidism and, less frequently, secondary hypothyroidism, but it is also prescribed for prevention of thyroid cancer recurrence. There are, however, no indications for TH prescription other than for recognized pathological disorders of the thyroid or its hypothalamic-pituitary regulation (eg, hypothalamic disorders, pituitary tumors and their treatment). Iodine deficiency, characteristic of countries with inland mountain ranges that have high rainfall, is the most frequent cause of hypothyroidism worldwide. However, in iodine-replete areas, primary hypothyroidism is mainly due to autoimmune thyroid failure (Hashimoto thyroiditis) and requires T4 treatment to restore euthyroid status. Following surgical or radioiodine ablation for thyroid cancer, full T4 replacement is required and often recommended at a mildly excessive dose for a short to medium period to suppress circulating thyrotropin (thyroid-stimulating hormone; TSH) levels, which reduces risk of cancer recurrence (29). Periodic scanning for recurrence of thyroid cancers requires either withdrawal of T4 treatment, sometimes switching to T3 to allow for shorter periods of symptomatic hypothyroidism, or else recombinant TSH administration (30, 31). Secondary (or central) hypothyroidism is rare (32, 33), having ~1% of the prevalence of primary hypothyroidism (34). It arises from hypothalamic disorders or pituitary tumors and their treatment that reduce pituitary production of sufficient TSH to maintain thyroidal TH production. In patients with untreated secondary hypothyroidism, serum TSH is low or undetectable with low serum T4 and T3, and treatment requires T4 replacement sufficient to restore and maintain circulating T4 levels. Modern third-generation TSH immunoassays have a well-defined lower limit of normal, below which a suppressed TSH indicates a state of hyperthyroidism. Hence, in healthy individuals with normal hypothalamic-pituitary function, a fully suppressed (undetectable) serum TSH is evidence of hyperthyroidism regardless of cause, such as endogenous disease (eg, thyrotoxicosis) or exogenous TH administration.

There has long been concern about overuse of TH (35-37), especially in light of the treatment of subclinical hypothyroidism leading to T4 being among the most frequently prescribed drugs in the USA (see https://www.meps.ahrq.gov/mepstrends/hc_pmed/) and UK (20), exceeding the prevalence of overt hypothyroidism. The prevalence of TH use in the USA is estimated at 4.6% over the whole adult population, based on data from the National Health and Nutrition Examination Survey (NHANES) studies (38). This comprises a prevalence of 9.3% among women and 3.2% among men (39) but fewer than 10% are overt biochemical hypothyroidism, with the remainder subclinical (38, 40). Similarly, the prevalence of TH use in Europe has a pooled estimate of 5.9% among women and 3.5% among men (41), although again only a small proportion is due to overt hypothyroidism, comprising 0.8% and 0.3%, respectively. These findings reflect a temporal trend of “creep” downward in the criteria for treatment of borderline hypothyroidism (elevated serum TSH) for questionable clinical benefit (42), which could also spill over to elite athletes. A global epidemiological survey demonstrated variation among iodine-replete countries with the prevalence of overt hypothyroidism ranging from 0.2% to 5.3% in Europe and between 0.3% and 3.7% in the USA (43). At a global level, it is evident that T4 treatment is used more widely than for treatment of overt hypothyroidism. These disparities in prescribing indications foreshadow complex and burdensome difficulties for requiring Therapeutic Use Exemptions for legitimate medical TH use if TH was prohibited in sport.

TH is among the most widely prescribed medications mainly because prescribing rates in the community rise steeply with age. However, the prevalence of TH use remains much lower among younger individuals of similar age to elite athletes (Table 1). Nevertheless, accurate prevalence estimates of TH use among elite athletes are limited. Disclosure of all substances an athlete has taken (including prescribed and nonprescribed, such as nutritional supplements or over-the-counter medications) in the last 7 days must be declared by the athlete on the Doping Control Forms at the time a sample is taken for a WADA-mandated doping control test. The most comprehensive prevalence data in the elite athlete population was obtained at the Tokyo 2020 Olympic and Paralympic Games held in 2021 (Table 2). Of 15 886 athletes competing across both events, 6364 (40%) athletes were selected for testing. From these tested athletes, there were 95 declarations for T4 and 8 declarations for T3. Female athletes were 3 times more likely (female 77, male 26; sex ratio 3.0; 95% CI, 2.4-4.6) to have a declaration for TH. This finding is consistent with the higher T4 prescription rates for females in the general population. The large majority (75%) of users were aged from 25 to 34 years, but prevalence of declared use was not significantly different between sports or country of origin. However, a limitation of this data is that voluntary declaration of nonprohibited substances, such as TH, may be underreported during the doping control process. Furthermore, the prevalence of TH use among elite athletes identified by Doping Control Form self-disclosure (summary estimate 1.4% from Table 2) is a little higher than most rates of TH use in the general community (Table 1).

Table 1.

Prevalence of thyroid hormone use among young populations

Study	Source	Location	Year	Age	Male (%)	Female (%)	Total (%)
Hunter (2000)	National register	Scotland	1993-5	<22			0.135%
Aoki (2007)	NHANES	USA	1999-2002	12-49			0.2% (overt) 2.2% (subclinical)
Leese (2008)	Regional register	Scotland	2007	<20	0.024	0.1
Virta (2011)	National register	Finland	2007	<29	0.19	0.89
Cerqueira (2011)	National register	Denmark	2008	<39			0.3%
Johansen (2020)	MEPS	USA	2014-16	<30 yr	0.3	3.3
Wouters (2020)	LIFELINES	Netherlands	2020	<29	0.1	0.9

Abbreviations: MEPS Medical Expenditure Panel Survey; NHANES, National Health and Nutritional Examination Survey.

Table 2.

Declarations of thyroid hormone use on Doping Control Forms

EVENT	Nations	Sports	Total athletes	Athletes tested	Use declared	Total	Mean (95% CI)a
Tokyo 2020 Olympics	206	33	11 483	5005
Tokyo 2020 Paralympics	162	22	4403	1359	103 b	6364	1.6 (1.3–2.0)
Minsk 2019 European Games	50	15	4082	999	13 c,d	999	1.3 (0.7–2.3)
PyeongChang 2018 Winter Olympics	92	7	2963	1615	17 c	2060	0.8 (0.5–1.4)
SUMMARY					133	9423	1.4 (1.2 - 1.7)

Doping Control Forms must be completed with every doping control test; athletes must declare all prescribed and nonprescribed (eg, over-the-counter or supplements) drugs taken over the previous 7 days.

aPercentage of tested athletes (95% CI, binomial distribution).

bCombined data for both Tokyo 2020 Olympic and Paralympic events. Declarations were for thyroxine (95) and triiodothyronine (8) by 77 females and 26 males with 75% between ages 25 to 34 years.

cAll declarations for T4, none for T3.

dAthletes (10 female, 3 male) competing in athletics (n = 4), road cycling (n = 3), and 6 other different sports.

Why Do Athletes Use Thyroid Hormone Without Medical Indications?

There is a striking dichotomy of opinion about athletes’ use of TH without medical indications, with expert endocrinologists believing it to be solely detrimental to performance without any likelihood of ergogenic benefit, in contrast to bodybuilder folklore claiming that TH may convey some performance and/or image benefits. Nevertheless, before dismissing nonprofessional opinion, it is salutary to bear in mind the history of androgen abuse, whereby athletes were convinced by anecdotal trial and error of the ergogenic benefits of androgens when such claims were denied by authoritative medical sources (45) before meticulous placebo-controlled randomized clinical trials established the veracity of those beliefs (46, 47). To understand the dichotomy of beliefs about, and motivations for, TH use it is useful to elucidate the motivation of athletes for using TH outside medical indications. Use of TH without prescription among bodybuilders and athletes is usually in cocktails of nonbanned supplements rather than TH alone. Online social media platforms, including Reddit, Instagram, Facebook, and online blogs (including steroidology.com) were surveyed to ascertain the rationales for TH use. These rationales for using nonprescribed TH are reviewed below.

Improving Body Composition

Overwhelmingly, the prime reason cited for TH use is for modulating weight loss to prioritize fat loss over muscle loss. This is most evident for image-based sports, such as body building, but it is also relevant to competitive sports with weight classifications so as to “make weight” (combat sports, weightlifting, rowing) as well as other sports where the power-to-weight ratio is a determinant of energetic efficiency (eg, cycling, rowing, endurance sports). T3, commonly referred to by one trade name, Cytomel (other brand names for T3 include Triostat, Triomet, Tertoxin, Liotyr, and Thybon), is preferred over T4 for its faster onset of action but is considered harder to access and more expensive. The authors were easily able to find Cytomel advocated and/or advertised at numerous supplement websites.

There is no convincing evidence for use of therapeutic doses of TH for weight loss, still less for selective loss of fat over muscle mass. Early studies utilizing T3 in obese euthyroid subjects on low calorie diets showed no efficacy for weight loss (48). Recent meta-analysis of TH use, either as T4 or T3, showed no convincing evidence of overall weight loss (49), although there was insufficient evidence of T3 effects on total body catabolism and skeletal muscle protein breakdown to evaluate selective fat loss (49). Nevertheless, the deliberate use of excessive TH doses causing hyperthyroidism could cause weight loss as a toxic overdosage effect. Consequently, current American Thyroid Association guidelines make a strong recommendation against using T3 for treating obese euthyroid patients due to a lack of sound efficacy and safety data (24). Nevertheless, these data on weight loss, mainly in obesity, may not fully exclude small, short-term, and reversible weight loss induced by deliberate hyperthyroidism that athletes may exploit; whether that is performance enhancing or detrimental requires separate consideration.

T3 is recommended in bodybuilding folklore for optimizing “cutting”, a popular aesthetic technique among bodybuilders and fitness enthusiasts. This aims to become as lean as possible quickly prior to a major competition, and to also create a subjective aesthetic appearance of better muscle “definition.” Such weight loss regimens aiming to maintain muscle while losing fat mass typically combine T3 with clenbuterol, a sympathomimetic, considered in the folklore a good “fat burner,” or a synthetic androgen (“anabolic steroid”). Regimens vary, but most include “cycling” the T3 medication with time on and off therapy (eg, 2 weeks on and off) and “stacking” (combination) with clenbuterol. These regimens mimic characteristics of androgen abuse multidrug regimens with similar rationale—to “restart” thyroid secretion and sensitivity during the off-treatment periods, and to accentuate effects by supposedly synergistic drug combinations, respectively. A typical T3 dose of 20 or 25 µg/day, falsely considered “low” (23), is considered a “top-up” that would cause “shutting down” the thyroid. However, on the contrary, such supraphysiological T3 dosage is likely to cause hyperthyroidism with suppression of serum TSH and endogenous TH secretion.

A variation on TH use for “improving” body composition occurs in drug cocktails mixing TH together with androgens (50). This appears to reflect recognizing that exogenous TH is catabolic when taken at these supraphysiological doses so that it may be a reason for combining with a synthetic androgen (“anabolic steroids” like oxandrolone) aiming to gain balance through “anabolic” (ie, weight gaining) effects of androgens. However, none of these combination regimens has been subjected to any, let alone well-controlled, clinical trials of efficacy or safety.

Boosting Androgen Action

Another rationale for TH use is attempting to surreptitiously enhance endogenous androgen action. One variant of this rationale is based on the observation that hyperthyroidism increases serum sex hormone–binding globulin (SHBG), the main circulating binding protein for testosterone, so that circulating SHBG and therefore testosterone concentrations are increased (51). Evident in TH excess of endogenous or exogenous origin, this increase in hepatic SHBG synthesis and secretion occurs via TH stimulation of hepatocyte nuclear factor-4α gene expression (52) in both men and women (52-54), including at young age (55). However, increased circulating SHBG lowers the whole-body metabolic clearance rates of testosterone (and estradiol), prolonging its circulating residence time and reducing tissue androgen effects (56-58). For example, higher circulating SHBG inhibits rather than enhances net androgen action in bone (59-64) as well as the reduced androgen:estrogen balance explaining the high prevalence of gynecomastia in hyperthyroidism (65, 66).

Another flawed rationale found online to enhance androgen action is to use exogenous TH aiming to increase androgen receptors and thereby amplify androgen action. However, increasing androgen receptors does not necessarily increase androgen action, which depends on triggering by binding of the androgen ligand (67). Similarly, the reported experimental T3-induced increase in an androgen receptor co-regulator, rather than the androgen receptor itself (68), is unlikely to have any effect on overall androgen action, which involves complex mechanisms from androgen binding to the androgen receptor with multiple post-receptor co-regulator mechanisms.

Overcoming Low Circulating T3

Another rationale for using nonprescribed TH includes rectifying the low concentrations (or expected reduction) of serum T3 seen in the energy deficit status of some elite athletes undergoing extremes of training and/or diet (69). Such reversible functional changes in circulating THs are consistent with the concept of Relative Energy Deficit in Sports (RED-S), introduced by the International Olympic Committee in 2014 (70) to broaden the scope of the previously defined (female) athlete’s triad, originally described in females as functional hypothalamic amenorrhea, extending to include men manifesting low serum testosterone (71) as well as to a wider range of exercise performance and nutritional states. Although criticized by defenders of the original concept of the female athlete’s triad (72), this expanded concept is gaining wider recognition (69, 73) when referring to athletes with impaired physical function due to an energy deficit that is not sufficient to maintain normal physiological function (70) and that can lead to secondary amenorrhea and impaired bone health. Analogous findings are reported in other catabolic states due to severe reduction in calorie intake (eg, severe undernutrition) and/or energy drain (eg, excessive exercise intensity) culminating in weight loss as a pathophysiological response to energy deficit states. Energy deficit syndromes including RED-S lead consistently to low serum T3, increased reverse T3, but less consistent, if any, changes in serum T4 as well as serum TSH, which is sometimes mildly lowered but not undetectable (73).

These manifestations of a catabolic state are also consistent with the reversible changes in thyroid function tests seen more broadly in the sick euthyroid (or non-thyroidal illness) syndrome. These may represent an adaptive hypothalamic response to environmental stress during almost any severe or chronic non-thyroidal illness. This syndrome exhibits characteristic, reversible changes in thyroid function tests featuring characteristically reduction in serum T3, increases in reverse T3 but without consistent changes in serum T4 or TSH. While some controversy remains (17-19), it is widely accepted that these biochemical features are not indicative of a TH deficiency state nor is there sound evidence to warrant TH replacement therapy (15, 74-76). Hence while elite athletes may exhibit aberrant thyroid function tests as part of a catabolic state due to intense exercise with or without diet-related nutritional deficits, this functional state does not represent hypothyroidism, so that TH treatment is not required or indicated. Nor is there convincing evidence that TH supplementation improves muscle function, endurance, or weight loss but supplemental TH use in the absence of diagnosed hypothyroidism risks adverse bone and cardiac health.

Biochemical Effects of TH

Another set of rationales for TH use is based on isolated biochemical findings, which may also be (mis)construed as the basis of ergogenic TH effects. However, such biochemical TH effects established in isolated in vitro systems must be considered at best permissive, that is, required for normal cellular function; however, to ascribe ergogenic effects to these findings requires additional physiological verification of boosted muscle or exercise-related performance in humans or, at least, intact animals above and beyond normal physiological function in vivo. Otherwise, isolated biochemical findings may be considered explanations in search of suitable findings to explain.

TH is a critical regulator of whole-body thermogenesis as manifest by the basal metabolic rate, which is elevated in hyperthyroidism and reduced in hypothyroidism (77), both of which would reduce physical performance. Such thermogenic effects appear to be explained by TH-induced stimulation of various futile energetic cycles in skeletal (78) and cardiac (79) muscle as well as white and brown fat (80-82), mostly through effects on mitochondrial energetics and uncoupling proteins (81, 83, 84). These futile cycles involve membrane exchangers (sodium-calcium exchanger, L/T type calcium channels, sarcoplasmic reticulum calcium ATPase [SERCA2], and the ryanodine receptor), all targets of T3 action. However, rather than enhanced organ function, these mechanisms result in deleterious effects of either excess or deficient TH exposure. For example, hyperthyroidism is associated with impaired exercise tolerance, but animal data shows an adaptive response within myocytes and skeletal muscle (85, 86). T3 administration in pigs shows short-term increase in myocardial cross-sectional size and beta-adrenergic receptor density, resulting in tachycardia and increased oxygen consumption (87). Increases in lactic acid production are also associated with T3 administration during increased cardiac output, but how this augments or limits performance is unknown (85). In rats treated with excess T3, blood flow to skeletal muscles during exercise increased up to 3-fold (86); however, in human subjects, acute elevation of TH did not affect skeletal muscle blood flow or muscle oxygenation during dynamic testing (88). Sources of energy are also altered during thyrotoxicosis where fatty acids are a preferential fuel source in the early stages of hyperthyroidism but how this affects energy expenditure is unknown (89). Hence, restoring euthyroid status from hypothyroidism by TH treatment does not imply additional ergogenic benefits of supratherapeutic exogenous TH, especially as the available evidence suggests that hyperthyroidism of any origin would impair exercise capacity.

Overall, none of the bodybuilding street folklore built upon colorful and imaginative elaborations of isolated biological findings provides any objective evidence corroborating alleged performance enhancement due to TH use. Such ornate misinterpretation of evidence represents a credulous numerology based on ill-understood endocrine physiology and pharmacology. Thoughtful and well-informed appraisal of the scant facts and evidence is essential to critically evaluate claims that are little more than science fantasy, driven apparently by opportunistic marketing and/or wishful thinking.

Likelihood of Performance Enhancing Effects

Abuse of hormones for performance enhancement has been and remains the most frequent and effective ergogenic approach to doping (90). The 2 most potent and effective classes of doping hormones are androgens (for power sports) and erythropoietin and blood transfusion (for endurance sports), which have set an exemplary template for assessing evidence to demonstrate unequivocal performance enhancing effects. Ergogenic effects of androgens were established by meticulous randomized placebo-controlled dose-response studies in healthy young men with fully suppressed endogenous testosterone (by depot GnRH analogs) with add-back of testosterone over a 24-fold range of doses from sub- to supra-replacement (46). Testosterone displayed strong dose-dependent effects of increasing muscle mass and strength and blood hemoglobin, effects replicated in older men (47) and by an independent group (91). Similarly, effects of hemoglobin doping by blood transfusion were originally demonstrated in the 1970s by well-controlled studies of blood withdrawal and transfusion effects over a 6-fold range of blood volumes on maximal oxygen consumption in treadmill ergometry (92). Analogous ergogenic findings were reported for administration of erythropoietin, a simpler and more easily concealed hemoglobin doping method compared with blood transfusion (93, 94). Other hormones abused for doping include growth hormone (44, 90, 95), its secretagogues and analogs, synthetic glucocorticoids (96), as well as insulin and its analogs. Although these other hormones are used by athletes, evidence for their performance enhancing effects is equivocal (90, 97).

Compared with the idealized model for proof of performance enhancement set by studies of androgens and of hemoglobin doping, corresponding studies of TH effects on exercise performance, to provide similar robust evidence to inform the status of these drugs, are not available.

In summary, there is a paucity of biochemical or physiological evidence which can be construed as the basis of performance enhancing effect of TH abuse in doping. Indeed, the overall available evidence indicates that the effects are most likely to be detrimental.

Adverse Effects of Thyroid Hormone Overuse

Balancing the unsubstantiated potential ergogenic effects, the use of TH is more likely to be detrimental to both health and performance. At the most extreme, there is a long history of reports describing fatal and serious nonfatal adverse clinical presentations of TH excess from deliberate or accidental TH ingestion (98-104).

An important facet of TH misuse is the narrow therapeutic index of THs (105-107), a term variously defined in drug development to describe the margin of dosage (or more accurately, systemic hormone exposure) between therapeutic and toxic effects (108). The narrow therapeutic index of T4, and even more T3, means that small changes in dosage may precipitate toxic effects. That contrasts markedly with testosterone, which has a high therapeutic index. That arises from its very wide dose-response in ergogenic effects, primarily on muscle, ranging 24-fold in dose from below to above the physiological range. Uniquely among canonical human hormones, testosterone has no naturally occurring hormone excess disorder in men so that even large supraphysiological doses are relatively well tolerated in the short term, with minimal or no consistent adverse physical or symptomatic effects.

The best-known detrimental effects of excessive TH exposure arise from impact of hyperthyroidism, whether due to spontaneous thyrotoxicosis or excess TH doses, resulting in TSH suppression leading to impaired bone health and cardiovascular function. As a result, the American Thyroid Association guidelines on thyroid cancer have modified their recommendation requiring long-term TSH suppression, via mildly supraphysiological T4 administration, to prevent recurrence of thyroid cancer. Rather, they now advocate a dynamic risk stratification with early relaxation of post-thyroidectomy TSH suppression in low- to intermediate-risk cases (29).

Bone health is significantly impaired in those with prolonged suppression of TSH, in both men (109) and women (110). Even among euthyroid postmenopausal women, low-normal TSH levels are associated with reduced bone density and increased risk of osteoporosis and fractures (111). These risks are less studied in younger people so the long-term effects on bone health, including potential reversibility of hyperthyroidism, remain unclear.

Cardiovascular effects are well established in patients with suppressed TSH due to TH excess. Atrial fibrillation, the most common cardiac manifestation of TH excess, occurs at a higher frequency in both overt and subclinical hyperthyroidism (111, 112). Cardiovascular dynamics are altered in hyperthyroidism as TH effects on the heart include increasing cardiac rate, output, contractility, and pulmonary artery pressure; however, these are deleterious effects evident without exertion and likely to severely reduce cardiovascular performance during exercise. Long-standing hyperthyroidism is associated with cardiac failure in patients with thyroid disease although not yet reported with exogenous TH abuse, perhaps because such abuse is usually short term (113). Sudden cardiac death has been reported (98, 99). Even subclinical hyperthyroidism involves cardiac compromise, although to a lesser degree, through increased incidence of atrial fibrillation and coronary heart disease (114).

The specific effects of hyperthyroidism, including excess TH ingestion, on metabolism, exercise capacity, and body composition are less firmly established. One observational study of body composition and metabolic/energetic outcomes associated with long-term TSH suppression due to excess T4 treatment found no significant differences in metabolic parameters (115). However, another short-term (3 days) experimental study of excess T4 treatment (200 µg/day) showed an increase in metabolic rate but reduced energy metabolism without any overall change in exercise efficiency (116). A further experimental study reported that supratherapeutic T3 dose (75 µg/day) over 9 weeks produced loss of both lean and fat mass with increased 24-hour energy expenditure (117). Overall, the available data shows only limited data for deleterious (or advantageous) metabolic or body compositional effects of TH excess due to exogenous TH intake. Any small nonselective weight loss observed is accompanied by adverse metabolic compromise in patients with suppressed TSH.

Taken together, the data suggest that prolonged TSH suppression from either pathological or exogenous TH produces comparable harm on bone, metabolic, and cardiovascular systems as well as possibly increased cancer risk (118). An important caveat is that most adverse findings are reported among older individuals taking TH for therapeutic reasons. How these apparent risks are manifest in younger individuals, especially highly fit elite athletes, with healthy bone and cardiac functions, requires more data to estimate the real risks. An interesting cohort for more analytical investigation is patients with intentionally suppressed serum TSH to reduce recurrence of thyroid cancer. Studies examining metabolic, cognitive, and quality of life outcomes have produced equivocal effects, suggesting reduced quality of life, cognitive function, and impaired glucose tolerance (119). For bone, beyond well-established reduction in bone density, adverse changes in the architecture of trabecular bone are also reported (120); however, extrapolating this data from thyroid cancer patients to athletes requires further validation. Nevertheless, considering the data overall, the effects of using excess TH in otherwise euthyroid individuals risks significant harm to health and performance without countervailing evidence of performance enhancement.

Detection of Thyroid Hormone Overuse

In theory, TH overuse could be readily detected with a completely suppressed, undetectable serum TSH, regardless of whether exogenous T4 or T3 is ingested. In addition, if T3 is being administered, T4 levels may be below the normal range. Hence the biochemical picture of undetectable serum TSH, low serum T4, and high serum T3 would strongly suggest a person is taking exogenous T3. Exogenous T4 ingestion would show a comparable picture except that there may be less reduction of serum T3 due to conversion of exogenous T4 to T3. This suggests that serial screening of serum TSH may detect TH abuse if deployed in an expanded endocrine module of the Athlete Biological Passport (ABP) in a Bayesian implementation of serial adjustment of individual serum TSH results. However, any positive findings of undetectable serum TSH (from a modern third-generation immunoassay (121)) would only indicate hyperthyroidism without distinguishing TH doping from spontaneous thyrotoxicosis due to autoimmune Hashimoto disease, a toxic thyroid adenoma, or thyroiditis or even early pregnancy when circulating human chorionic gonadotropin (hCG) peaks (“gestational thyrotoxicosis”). Such an Atypical Finding would subject the athlete to a specialist endocrine evaluation to exclude endogenous hyperthyroidism by additional blood tests and thyroid uptake scan to distinguish active from suppressed thyroidal secretion. This would be analogous to the results management of a positive urine hCG test in men whereby a medical evaluation is required to exclude a hCG-secreting germ cell tumor before considering a finding as Adverse Analytical Finding, presumed as a possible Antidoping Rule Violation subject to antidoping sanction.

In considering the potential of serial serum TSH monitoring to detect TH abuse in elite athletes, potential influences on serum TSH measurement other than TH excess need to be considered. Within-person serum TSH levels are stable over time, showing less variability than between-person levels, reflecting a high degree of genetic determination of individual set-points (122-124) and circumstances favoring effective integration into an ABP-style Bayesian screening format. In an otherwise healthy population, serum TSH level is mildly affected by ethnicity, body weight, iodine intake, pregnancy, sex, age, and circadian rhythm (24) making determination of population thresholds challenging. Again, this is more compatible with an ABP-style screening that relies on individual-specific rather than population-based thresholds. In patients with hypothyroidism, stabilizing T4 treatment can be slow, with transient low serum TSH during dose optimization due to the long half-life of T4 and the demanding requirements of effective therapeutic use (refrigerated T4 storage, taking on an empty stomach, avoiding concurrent ingestion of calcium, iron and other drugs that inhibit T4 absorption) (107). However, these sources of variation rarely cause undetectable serum TSH levels. Using a modern third-generation TSH immunoassay, undetectable serum TSH together with normal or increased serum T4 and/or T3 indicates significant TH excess due to either thyroid pathology (subclinical or overt thyrotoxicosis from autoimmune thyroid disease, toxic adenoma, or thyroiditis) or exogenous TH intake.

Monitoring serum (or urine) T4 or T3 would be complex to interpret for detection of TH abuse. Direct measurement of T4 and/or T3 is feasible in serum or urine using validated mass spectrometry–based measurements with certified reference materials and quality control available for serum (but not urine). However, exogenous T4 or T3 must be distinguished from their endogenous counterparts, a challenge in common with exogenous forms of other natural hormones (eg, testosterone, growth hormone). Screening for exogenous testosterone use relies on a testosterone-to-epitestosterone ratio (T/E ratio) in urine, which is reasonably effective when deployed in a serial Bayesian format of the ABP; however, confirmation is required by a carbon isotope ratio mass spectrometry test (125, 126). Analogous screening or confirmation tests are not available for THs. Detection of thyromimetics would be simpler, as it is for synthetic androgens, depending on identifying xenobiotic chemical signatures on mass spectrometry, which may provide proof of doping in evidence of an exogenous origin for that substance.

Like serum TSH, these methods would also be best deployed in a Bayesian format of serial self-defined reference limits rather than population-based limits; this would be crucial for international deployment because reference ranges are subject to the variable iodine status of different countries. At present, most serum TH measurements in clinical practice rely on so-called “free” T4 and “free” T3 assays, which lack certified reference standards and quality control and might in practice be difficult to withstand legal challenge to assay validity.

An additional complexity is that use of marketed recombinant TSH (Thyrogen) could create a form of indirect TH doping. In theory, this could be suspected in an otherwise healthy individual displaying a nonsuppressed serum TSH in the presence of elevated serum T4, a pattern to be distinguished from rare alternatives such as TSH-secreting pituitary tumor (127), genetic thyroid hormone resistance due to TH receptor mutations (128, 129) and artifactual heterophile or other antibody interference in TSH immunoassay (130).

Should TH Be Prohibited in Elite Sports?

Prohibition of TH for elite sports competition requires justification of whether T4 and T3 meet at least 2 of the 3 criteria of the World Anti-Doping Code (Code). If prohibition of T4 and T3 was justified, it would extend to TSH and TSH-releasing hormone (TRH), administration of which could increase secretion of endogenous THs. In addition, TH prohibition would have to extend to any of the burgeoning field of tissue-specific thyromimetic T4 or T3 agonistic analogues (12) unless and until they prove to be so tissue-specific as not to enhance exercise capacity.

Meeting 2 of the 3 Code criteria, each accorded equal significance in the Code, is the primary consideration for prohibiting any substance or method. The first criterion is whether THs are actually or potentially performance enhancing. This embodies the primary purpose of the Code of maintaining fairness by eliminating use of substances or methods that enhance sporting performance. There is usually little controversy over whether certain medicinal substances that meet this criterion of performance enhancement are deemed prohibited. However, for unregulated status of nutritional supplements or mixtures, their ubiquity in society together with the difficulty of defining clear ergogenic dose thresholds make them difficult to prohibit, given the difficulty in establishing a reasonable threshold to distinguish innocent use from abuse for doping purposes. As examples of this dilemma, various nutritional supplements with small but proven ergogenic effects are not prohibited (5). In practical terms, a small dynamic range of an ergogenic effect makes it virtually impossible to define any threshold between innocent consumption and doping uses. Hence, the only option is complete or no prohibition. It is worth considering the consequences of relaxing this pivotal first criterion of performance enhancement as advocated by libertarian philosophers (131) but not by practical ethicists (132, 133). Any drug that was consistently effective in performance enhancement would inevitably become an obligatory part of training programs. For example, in power sports this would allow for provision of testosterone, a prescription drug, on demand for nonmedical purposes but without any feasible restriction on dose or age. The harmful consequences of such laissez faire are well understood and rightly feared (134).

For THs, the evidence for performance enhancement is negligible. Despite many rationales used by bodybuilders or athletes for TH use without appropriate medical oversight, none have any scientific or medical credibility and there is no objective evidence of TH effects on athletic performance among the very few relevant studies pressed into service for this objective, and none are well designed or convincing for this purpose. While it is always possible that further critical studies may test and affirm this hypothesis according to the template of studies establishing performance enhancing effects of androgens and hemoglobin, such findings remain unknowable in advance. However, at present, the balance of objective evidence is firmly against there being any performance enhancing effects of TH.

The second and third criteria are based on whether TH is actually or potentially harmful and whether TH use violates the spirit of sport. In terms of the second criterion of harmfulness, taking any medicinal substance that is not therapeutically required may always be potentially harmful. Abundant evidence from multiple sources reviewed herein indicates that TH use in the absence of thyroid diseases and without medical prescription or supervision is likely to cause harm from hyperthyroidism and its metabolic and other tissue consequences. Among drugs of misuse or abuse, this risk is unusually high due to the low therapeutic index of THs. These features clearly meet the second criterion.

In terms of the third criterion of contrary to the spirit of sport, using any prescription medication (like TH) without a valid medical indication aiming to improve athletic performance is clearly against the spirit of sport through an intention to cheat. If such nonmedicinal TH usage was widely employed by athletes, the prevalence of TH use among elite athletes would exceed that of the population norms for age and sex. However, the available evidence, based on voluntary disclosure in Doping Control Forms Table 2), is that the prevalence of TH use among participants in elite competitions may modestly exceed the prescription rates in the age- and sex-matched general population (Table 1). However, this could represent nondisclosure of a drug that is not banned or that TH abuse may occur only among a small minority of elite athletes. On the contrary, it might even be expected that in the current trend of increasing monitoring of athletes’ performance (135), extending naturally to interrogation of the health and performance of elite athletes, might lead to greater use of TH than age- and sex-matched nonathletic norms. In this context, screening with thyroid function tests of elite athletes with nonspecific symptoms and/or reduced performance would create the well-known consequence of unjustified screening leading to spurious false positive test results of uncertain significance that encourage pointless and possibly harmful overtreatment (136), including TSH testing for hypothyroidism (137). These issues are potentially further complicated by the present investigational uses of TH receptor-β specific agonists for treatment of metabolic or neurodegenerative disorders wherein the thyromimetics are intended not to display classical TH effects (11-13). At present, there is little objective evidence of the patterns of TH use among elite athletes. Such data would be helpful in discerning judgment on whether there is widespread unjustified TH usage among elite athletes, which, in any case, is currently not regarded objectively as performance enhancing.

Hence, the available evidence is that TH clearly meets 1 of 3 Code criteria for prohibition—harmfulness—but not the criterion of performance enhancement and there is currently insufficient information to indicate violation of the spirit of sport, the third criterion. That spirit of sport criterion involves complex judgment on the values (138, 139) and intent of an athlete’s drug use. If it is used to gain ergogenic advantage, that would violate that criterion, whereas misguided use in the mistaken (even if genuine) belief of correcting TH deficiency may not. Relevant to these dilemmas, the “creep” downward in elevated serum TSH criteria for treatment of subclinical or borderline hypothyroidism (42) could fuel an athlete’s claims of perceived therapeutic need despite the lack of, or even contrary to, objective supportive evidence. However, if TH was prohibited, athletes who are intent on doping would naturally plead the latter and judgment on the validity of fair usage would require careful expert judgment.

Given that the principal purpose of the Code is to protect the fairness of competition in sport from anticompetitive practices, such as use of doping substances and methods and the lack of performance enhancement, there little argument to prohibit THs. Like other hormone misuses in sport, deterrence is feasible using other means such as educational initiatives and other regulatory discouragement. For example, glucocorticoid abuse in road cycling has been minimized by the International Cycling Union (UCI)’s 2011 introduction of a No Needles Policy, which has effectively eliminated injectable glucocorticoid use in cycling, prior to their recent prohibition across all sports from 2022 (96).

Even if the argument could be made that TH should be prohibited, the feasibility, logistics, and the difficulties of uniformly enforcing such a prohibition must be considered. For example, the regulatory enforcement consequences would be a likely proliferation of therapeutic use exemption applications requiring critical appraisal. While suppressed serial serum TSH in the absence of known thyroid disease can serve as a screening test to detect exogenous TH use, it would require integration into the ABP endocrine module. But just as every positive urine hCG detection requires medical evaluation to exclude an hCG-secreting germ cell tumor, every positive screening test for hyperthyroidism (undetectable serum TSH), would require medical evaluation (including thyroid uptake scan) to exclude undiagnosed thyroid axis disorders before being considered a potential Antidoping Rule Violation subject to sanctions. Screening for exogenous T4 and/or T3 as well as for various thyromimetic analogues would be complex to implement and would require extensive validation of screening and confirmation tests, yet to be devised. This would place an excessive burden on legitimately treated athletes, who are predominantly female, as illustrated by the prevalence detailed in Table 1. Hence the feasibility and resource demands of prohibition of THs are also arguments against prohibition, particularly when the minimal evidence for performance enhancement is unconvincing.

THs currently do not meet the requirement of 2 of the 3 Code criteria, in addition to considerations regarding the feasibility, logistics, and burden on the athlete. On balance, therefore, prohibition of TH in elite sport under the Code is not at present justified. While this remains a challenging area, more convincing evidence on the prevalence of TH use among elite athletes and especially whether TH use has any performance enhancing effects might shift that balance in the future.