Over the past several decades anabolic steroid use in athletes has increased. Athletes perceive benefits of steroid use as gains in strength and rapid recovery from stressful workouts. Because the use of anabolic steroids remains illicit, detailed knowledge of their side effects is unknown. Reported side effects emanate from case reports, retrospective studies, and comparisons drawn from prospective studies of steroid therapy in patients with a variety of wasting syndromes. Further confusing the specific side effects of a given anabolic steroid is the common practice that many athletes take multiple drugs simultaneously and in multiple administration routes [1••]. Additionally, physiologic differences between patients with wasting syndromes and athletes limit our ability to extrapolate the results of controlled studies to anabolic steroid use in sports . One other compounding factor in determining which side effects arise from anabolic steroid use is that users also ingest other illicit drugs including cocaine, marijuana, alcohol, and tobacco [1••]. Doctors and training staff must be aware of the potential harms in order to both recognize and more importantly educate patients about these possibly life-threatening adverse effects.
Testosterone is the prototype for androgenic anabolic steroids (AAS) and in its unmodified form has a very short biologic half-life [3••]. Rapid metabolism by the cytochrome P450 system in the liver leads to poor bioavailability with oral administration [1••]. Designer AAS are altered so as to increase the bioavailability and prolong the desired effects. For example, testosterone's half-life is measured in minutes, whereas fluoxymesterone, a synthetic AAS, has a half-life of 9.2 hours.
The two major effects of testosterone are androgenic and anabolic. Androgenic effects stimulate development of secondary sexual characteristics and growth of the male reproductive system, and anabolic effects increase protein synthesis and nitrogen fixation . The goal of many of the designer steroids is to minimize the androgenic side effects while maximizing the anabolic potential. Because no synthetic molecule is completely anabolic, the results these molecules produce remain mixed with both androgenic (masculinizing) and anabolic (tissue-building) changes seen in the body.
Two modifications made to the testosterone molecule alter the androgenic/anabolic profile, and thus the type of side effects seen from usage of the various AAS [3••]. Class A modifications are formulated by esterification of the 17-α-hydroxyl group, which increases the lipophilic properties, allowing a slow, delayed absorption as an injectable form [1••]. This is the most common form of modification because it allows the injection to be administered as infrequently as once every 2 to 6 weeks [3••]. This pharmacokinetic alteration also results in increased and unwanted androgenic effects [3••]. Class B modifications result from alkylation of the 17-α portion of the molecule, which decreases hepatic metabolism (Table 1) [3••]. This allows increased oral absorption and slower hepatic degradation. Slower clearance from the liver results in greater hepatic toxicity [1••].
Athletes often alternate or combine multiple forms of AAS in an effort to maximize benefits while minimizing side effects [3••]. Users often take an average of five different drugs to achieve supraphysiologic anabolic levels exceeding 40 to 100 times the normal physiologic hormonal effects [3••,5]. This use of a combination of products is commonly referred to as stacking, and it is this stacking that makes it difficult to pinpoint the direct effect of different agents [3••,5]. The major side effects of trying to achieve supraphysiologic anabolic levels are directly attributable to the associated increased androgen exposure [6••]. Because AAS are suspected of being harmful, determining side effects through prospective, randomized control studies would be unethical and clearly no one would be likely to study the effects of stacking in a human population. Therefore, much of our evidence is based upon animal studies, case reports, or long-term data in patients with chronic disease treated with androgen therapy.
The effects of AAS tend to follow a linear dose-response curve for both anabolic and androgenic actions, although at high doses there appears to be a plateau to their physiologic effects. The supraphysiologic doses used by athletes far exceed the saturation point of the androgenic receptors. Therefore, there must be additional mechanisms to explain why supraphysiologic doses of steroids seem to enhance strength. Following are three different physiologic mechanisms through which AAS tend to exert their effects.
First, AAS improve the body's utilization of ingested protein, which favorably alters nitrogen balance. They mainly stimulate protein synthesis by turning on gene transcription after binding to androgenic receptors at the cellular level. Androgens exert their effects on a number of varied tissues including bone, adipose tissue, skeletal muscle, brain, prostate, liver, and kidney, as well as reproductive tissues. Therefore, a more complex understanding at the cellular level is needed to explain the variety of effects, knowing that their actions are mediated by only one androgen receptor. . These receptors appear to be identical in muscle and various other organs, but their absolute number and affinity for various types of AAS potentially explain the variety of different effects in multiple organ systems and from various AAS products . Some effects of testosterone are mediated through conversion to other bioactive compounds including dihydrotestosterone and estradiol .
A second effect of steroids is to displace glucocorticoids from binding to their receptors, thus exerting an anticatabolic effect. Because glucocorticoids usually depress protein synthesis, this antagonistic effect explains muscle mass gain and also the utility of these drugs in wasting syndromes .
A third postulated effect of steroids is that they confer a psychologic benefit to athletes wishing to gain strength. AAS users describe a euphoria or high following ingestion that allows them to work harder during workouts and recover more rapidly, which allows them both to intensify their training and become more aggressive in competition .
An association between premature cardiovascular events and the misuse of AAS in athletes has been observed [10••]. This is believed to be primarily mediated through changes in the lipid profile. Although endurance exercise favorably affects the lipid profile primarily through an antiatherogenic effect of raising high-density lipoprotein (HDL) cholesterol and lowering triglycerides, heavy resistance training alone fails to show a similar effect . This implies that the effect of AAS on lipid profile may confounded by the type of training the athlete pursues. Multiple prospective studies of AAS effects on cholesterol have yielded varied results. The majority of results (studies ranging from 3–26 weeks) show no overall change in levels of total serum cholesterol; however, some individual studies show increases and others show a decrease in total cholesterol . Despite the varied results on total cholesterol, the effects on HDL seem more consistent. Reductions in HDL range from 39% to 70% depending on the type of AAS and also appear to be dose dependent [5,11]. Several studies show reductions of HDL down into the teens, which, based on Framingham data, places these patients at a three times greater risk for coronary artery disease compared with men with HDL above 50 mg/dL . There even appears to be some variation in the dose-effect based on sex . In one study of hemodialysis patients, weekly administration of nandrolone resulted in a reduction of HDL-2 and apolipoprotein A-1 levels, complemented by a corresponding increase in apolipoprotein B and triglycerides . The oral 17-αalkylated steroids, as opposed to parenteral nandrolone, seem to exert the greatest effects on lipids and lipoproteins, which can be seen as early as the first few days of administration [5,10••,11]. This reduction often reaches a plateau effect after 8 weeks of use . Although the direct mechanisms of action and impact on cardiovascular disease  remain unproven this negative effect on HDL suggests a higher risk of atherogenesis. The alteration in lipid profile seems completely reversible upon discontinuation, but may take at least 4 to 12 weeks, often depending on dose and duration of steroid use .
In addition to HDL effects, an increase in low-density lipoprotein (LDL) appears to parallel the HDL reduction . Significant LDL increases were appreciated in just 8 weeks of anabolic steroid use in one study, and often had not returned to baseline 6 weeks after cessation of AAS use . Because HDL acts as a primary scavenger of LDL particles, LDL changes possibly reflect a secondary rather than primary effect. These alterations in HDL and LDL cholesterol are more profound in athletes engaged in heavy resistance sports taking AAS as compared with endurance athletes, possibly reflecting the influence of the athlete's training regimen . In one arm (self-administered, prospective, nonblinded portion) of the study from the Netherlands, reductions in lipoprotein (a), which is an independent risk factor for vascular disease, seem to provide a slightly beneficial effect on the lipid profile [10••]. Reductions of as much as 50% of lipoprotein (a) were observed in as little as 8 weeks of AAS use, and remained decreased at 6 weeks postcessation [10••]. Longer duration of AAS did not correlate directly with further serum reductions, but did demonstrate a more prolonged return to baseline of lipoproteins [10••]. In the second phase (randomized controlled trial, double-blinded portion) of the Dutch study, both placebo groups and nandrolone decanoate both demonstrated reductions (19% and 40%, respectively) in lipoprotein (a) that was nonsignificant [10••]. One explanation for the difference is that the oral 17-α alkylated steroids, taken in the first portion, seem to exert the greatest effects on lipids and lipoproteins as opposed to the parenterally administered nandrolone used in the second arm. This is mediated by the first-pass metabolism of the orally administered drugs through the liver [10••]. These effects can be seen as early as the first few days of administration, and seem to be more dependant on the type of steroid as opposed to the duration, although no long-term studies exist currently [10••,11]. Concentrations of lipoprotein (a) have been shown to have a close correlation with deposition in vascular walls, are often genetically determined, and seem resistant to current lipid therapies [10••]. In other studies of postmenopausal women and hemodialysis patients treated with nandrolone decanoate, a similar reduction in lipoprotein (a) was observed, but effects on overall mortality and cardiovascular events were not documented, and at this point can only be extrapolated based on serum changes [12,13]. The Dutch study (a randomized, double-blind placebo-controlled trial) and many others confirm no significant changes in serum triglycerides and total cholesterol [5,10••].
Various studies have demonstrated transient elevations of liver function tests (elevated plasma alkaline phosphates, aminotransferases, conjugated bilirubin, and plasma proteins) with and without significant hepatic injury . The orally 17-α alkylated steroids have a higher incidence of hepatotoxicity than other preparations [1••]. The mechanism of action is most likely from a direct toxic effect due to the brief period of time between exposure and liver damage and a dose-related effect. The most common used measures of hepatotoxicity are aspartate aminotransferase (AST), alanine aminotransferase (ALT), and lactate dehydrogenase (LDH). Values are usually in the range of two to three times normal. These changes often mimic the effects seen with oral contraceptives. Elevations in AST, ALT, and γ-glutamyl-transferase (GGT) tend to peak within 2 to 3 weeks of consumption even at relatively low doses, and will usually return to baseline within several weeks upon discontinuation . GGT was the most sensitive enzyme to detect hepatic dysfunction . Physicians must be careful in evaluating serum elevations of these enzymes, because strenuous exercise alone can cause muscle breakdown, leading to transaminase elevations . In addition, with the exception of LDH, the enzymes can be found in other body tissues confounding the picture even more .
Other hepatic disturbances secondary to AAS in animal studies have given rise to great concern in athletes. Steroids induce a wide range of hepatic disorders ranging from impaired excretion, cellular hepatocyte changes, cholestasis, peliosis hepatis, and hepatocellular hyperplasia to carcinomas . Androgen-related cholestasis has been observed in varying frequency from a few cases to 17.3% in some studies [14,15]. The cholestasis results from the reduced bile transport and disruption of intrahepatic microfilaments. This jaundice appears to be transient in nature and is secondary to biliary stasis in the biliary canonicals without any structural hepatic injury. This is in contrast to the associated inflammation and necrosis seen with other forms of hepatitis. There may also be a relationship between cholestasis and hypercholesteremia .
Other rare hepatic lesions include some potentially life-threatening lesions. Peliosis hepatis is a blood-filled cyst seen in many case reports in patients taking oral androgens, and is often correlated with more prolonged use . In the majority of these cases, the lesions were identified incidentally, most on autopsy, and the patients were completely symptom free. Several case reports of patients show direct mortality from internal hemorrhage or hepatic failure secondary to these lesions . Both cholestasis and peliosis hepatis are believed to be explained by similar processes. Oral 17-α alkylated androgens produce hepatocyte hyperplasia, with enlarged hepatocytes occluding both hepatic venous return and sinusoids . Sinusoidal dilatation at the peripheral zone of the hepatic lobule is a common finding with anabolic steroid use .
Medical literature documents isolated cases of androgen-associated hepatic tumors. Many of these reports were in patients with known hereditary anemias being treated with long-term androgens. Most of these tumors are benign adenomas, but early detection may lead to prevention of life-threatening complications associated with these tumors. Hepatic adenomas are usually found in young women taking oral contraceptives, and are still relatively uncommon. These adenomas are hypervascular tumors with relatively thin-walled capsules [18••]. They rarely transform to malignant tumors, but have risks of sudden rupture and bleeding, leading to hemoperitoneum [18••] which is a life-threatening condition. A particular problem is differentiating adenomas from hepatocellular carcinomas by ultrasound [18••]. If the differentiation is not obvious by clinical and histologic findings, a surgical resection is recommended [18••]. Treatment protocols for these lesions are difficult to formulate secondary to their varied prognosis, their difficult resection, and their unknown potential for malignancy [18••]. However, regression of the majority of these lesions after discontinuation of the steroids raises the question if there is truly malignant potential attributable to AAS use . Most of these lesions are linked to usage of the orally administered 17-α alkylated AAS [18••]. Sources recommend repeat ultrasound every 6 months with consideration of excision if late diagnosis is made [18••]. Nonsurgical options should be the optimal approach because many of the tumors will regress after discontinuation of the AAS, especially if detected early. Prompt detection of these lesions prevents important, potentially life-threatening sequelae and possible malignant deterioration [18••].
One case report describes an asymptomatic 35-year-old body builder with a 15-year history of AAS use at high doses in cycles of a combination of oral stanozolol and nandrolone for 8 weeks, who presented without jaundice and no past history of alcohol or tobacco abuse [18••]. Clinical and laboratory examination revealed severe hepatomegaly and mild elevations of ALT (75 IU/L), AST (53 IU/L), and alkaline phosphatase (403 IU/L) [18••]. Serum α-fetoprotein and coagulation test results were completely normal, bilirubin was elevated to 1.6 mg/mL, and viral hepatitis markers were negative [18••]. Abdominal ultrasound revealed several large hyperechogenic (∼ 12 cm) lesions with peripheral areas of blood flow that were later identified as hepatic adenomas by cytology, with no signs of malignancy [18••]. Further MRI showed a heterogeneous signal with mixed patterns consistent with areas of necrosis and hemorrhage within the adenomas [18••]. One year after the diagnosis, the patient still had lower yet persistent elevations of liver enzymes, and the ultrasound demonstrated no changes. Repeat biopsy revealed only mild atypia without any signs of malignancy [18••]. However, 4 years after termination of the AAS, and despite normalization of the liver enzymes, repeat ultrasound showed only a slight decrease in size of the adenomas [18••]. The patient was referred for liver transplantation secondary to the significant hepatomegaly and potential malignant risk of the adenomas [18••].
Exogenous steroids exert their affect on adrenal hormones by negative feedback on the hypothalamic-pituitary-gonadal axis . Large doses of exogenous AAS lead to a decrease in both follicle-stimulating hormone and luteinizing hormone serum concentrations. There is some controversial evidence to show that high doses of exogenous steroids lead to spontaneous increases in growth hormone (GH) production . In one study of men with hypogonadotropic hypogonadism, testosterone enanthate administration resulted in marked increases in spontaneous GH secretion . This elevation of GH provides the negative feedback that leads to decreased endogenous testosterone and gonadotropin release. Suppression of gonadotropin release leads to oligospermia and hypogonadism. This is reflected by decreased levels of testicular precursors of testosterone, namely serum pregnenolone, progesterone, 17-hydroxyprogesterone, and 17-hydroxypregnenolone . Supraphysiologic doses of AAS may lead to infertility due to decrease in both quantity and quality of sperm after several months of use . The return of normal levels of FSH and LH concentrations after discontinuation of AAS can take 6 to 12 weeks, whereas normalization of endogenous serum testosterone levels requires several additional weeks even after the return of normal gonadotropin levels . The amount of time needed for full recovery of normal reproductive function varies both on dose and duration of steroid use and can range from 4 to 5 months to several years . Many athletes turn to human chorionic gonadotrophin or clomiphene to reverse or even maintain spermatogenesis following or during courses of AAS .
Testosterone is broken down to form estradiol in the peripheral tissues of the body. In supratherapeutic doses of testosterone and its metabolites, the peripheral aromatization will lead to a major increase in estradiol. Serum levels of estradiol are often seven times the levels prior to onset of AAS use and approach medium levels of hormones found in normal women. In male athletes these levels of estradiol will often lead to gynecomastia and a heightened voice [1••]. Gynecomastia may cause breast pain and undesirable cosmetic affects for athletes. Some athletes even resort to taking estrogen-blocking medications such as tamoxifen to prevent these effects, although no scientific data support this practice . In most cases, gynecomastia will remit upon discontinuation of the steroids. However, in more prolonged use of AAS, gynecomastia can be permanent and thus require surgical resection for cosmesis .
Women who take AAS can experience androgenic effects including changes in libido, male-pattern baldness, deepening of the voice, acne vulgaris, and other masculinizing effects in early use. Longer-term use causes clitoromegaly, changes in pubic hair growth, menstrual irregularities, and even breast reduction . Adolescents can experience premature closure of their epiphyses, resulting in short stature [1••].
Renal side effects of AAS are uncommon and have been documented in a few isolated case reports. These reports noted a minimal effect of AAS use on renal function with a mild elevation in serum creatinine . The combination of AAS and creatine supplementation, commonly abused by weight lifters, may increase renal damage . One case of membranoproliferative glomerulonephritis has been cited in an athlete with prolonged AAS abuse . An unusual cancer in adults, Wilms' tumor, has been seen in a few athletes who were self-administering AAS over several years .
In one case report, a 23-year-old bodybuilder on cycles of 8 weeks of oral stanozolol and oxymetholone and parenteral nandrolone, testosterone, and boldenone, along with a high-protein diet, developed profound symptoms after only 6 months of use [18••]. Of note, the patient was also limiting sodium and water intake and taking the diuretic torasemide. The patient presented with acute confusion, asthenia, and anorexia of 1 month's duration [18••]. The patient had acute renal failure with a creatinine of 10.2 [18••]. The patient was also found to have signs of rhabdomyolysis and muscle damage with elevations of AST/ALT (130/178 IU/L) and LDH (716 IU/L), myoglobinuria and creatine phosphokinase measuring 5499 IU/L that may have been exacerbated by the renal failure and hypokalemia [18••]. AAS are known to cause Na+ retention, which leads to increased K+ and H+ excretion, leading to metabolic alkalosis and hypokalemia [18••]. This patient was found to have a pH of 7.62 with Na of 147 and K of 2.1 [18••]. Confounding interpretation of these side effects are the concomitant ingestion of a loop diuretic, limited Na+ and water, and AAS, which makes it difficult to know which drug caused which findings in this patient . Further evaluation revealed hepatomegaly with a few hepatic adenomas without signs of malignancy after biopsy was performed [18••]. After undergoing three hemodialysis treatments, this patient showed complete clearing of symptoms and normalization of laboratory values within 20 days [18••].
From reviews of a combination of case studies and retrospective studies, expert medical opinion suggests that both temporary and more permanent side effects accompany anabolic steroid use. Much of our knowledge is confounded by athletes who use unknown doses of AAS products in multiple combinations and through multiple routes of administration. Controlled randomized trials seem unlikely to occur secondary to ethical concerns. Even though the medical literature offers inconsistent evidence, AAS products appear to affect lipid levels, hepatic function and structure, adrenal glands, and kidney function. Worrisome reports of hepatocellular carcinoma and Wilms' tumors raise the more serious specter of AAS association with malignancy. In spite of publicized reports of fatal complications of AAS use and widespread knowledge within the athletic community of the negative effects, potential performance enhancement still entices many athletes to use AAS. Physicians should continue to target efforts at counseling adolescents and other athletes about the potential harms of AAS and the legal options to improve strength and performance. Knowledge of the multiple potential serious side effects of these drugs will help physicians earlier identify those individuals with complications from use.
References and Recommended Reading
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
1.•• Reents S: Androgenic-anabolic steroids
. In Sport and Exercise Pharmacology. Champaign: Human Kinetics; 2000:161–181.
Recent good review of the pharmacology of anabolic steroids and their biochemical properties. Looks at how chemical substances affect exercise performance in athletes.
2. Hickson RC, Ball KL, Falduto MT: Adverse effects of anabolic steroids
. Med Toxicol Adverse Drug Exp
3.•• Hall R: Abuse of supraphysiologic doses of anabolic steroids
. South Med J
Most recent review article regarding specific side effects of high doses of AAS. Article reviews some of the history of AAS and the mechanisms for utilizing supraphysiologic doses of AAS. The article then reviews the major side effects broken down by organ system.
4. Shahidi NT: A review of the chemistry, biological action, and clinical applications of anabolic-androgenic steroids
. Clin Ther
5. National Institute on Drug Abuse Research Report–Steroid Abuse and Addiction. National Institutes of Health Publication No. 00-3721. Bethesda: National Institutes of Health; 2000.
6.•• Sturmi JE, Diorio DJ: Anabolic agents
. Clin Sports Med
A comprehensive review of four different types of anabolic agents used by athletes. The article reviews physiology and chemistry as well as performance and medical side effects.
7. Haupt HA, Rovere GD: Anabolic steroids: a review of the literature
. Am J Sports Med
8. Snyder PJ: Androgens
. In Goodman and Gilman's The Pharmacological Basis of Therapeutics, edn 10. Edited by Hardman JG, LE, Limbird, Gilman AG. New York: McGraw Hill; 2001:1635–1648.
9. Wu FCW: Endocrine aspects of anabolic steroids
. Clin Chem
10.•• Hartgens F, Rietjens G, Keizer HA, et al.
: Effects of androgenic-anabolic steroids on apolipoproteins and lipoprotein (a)
. Br J Sports Med
A recent study evaluating the effects of anabolic agents on apolipoproteins and overall lipid panel in strength athletes. The study is nonblinded, but looks at the relationship between changes in lipids profiles as they may be further indicators of cardiovascular risk. Focuses predominantly on HDL subfractions and apolipoprotein(a) and lipoprotein.
11. Hartgens F, Kuipers H: Effects of androgenic-anabolic steroids in athletes
. Sports Med
12. Lippi G, Guidi G, Ruzzenente O, et al.
: Effects of nandrolone decanoate on serum Lp(a), lipids, and lipoproteins in women with postmenopausal osteoporosis
. Scand J Clin Lab Invest
13. Teruel JL, Lasuncion MA, Rivera M, et al.
: Nandrolone decanoate reduces serum lipoprotein(a) concentrations in hemodialysis patients
. Am J Kidney Dis
14. Cicardi M, Bergamaschini L, Tucci A, et al.
: Morphologic evaluation of the liver in hereditary angioedema patients on long-term treatment with androgen derivatives
. J Allergy Clin Immunol
15. Pecking A, Lejolly JM, Najean Y: Hepatic toxicity of androgen therapy in aplastic anemia
. Nouv Rev Fr Hematol
16. Cohen JC, Noakes TD, Benade AJS: Hypercholesteremia in male power lifters using anabolic-androgenic steroids
. Physician Sportsmed
17. Bagheri SA, Bayer JL: Peliosis hepatis associated with androgenic-anabolic steroid therapy- a severe form of hepatic injury
. Ann Intern Med
18.•• Socas L, Zumbado M, Perez-Luzardo O, et al.
: Hepatocellular adenomas associated with anabolic androgenic steroid abuse in bodybuilders: a report of two cases and a review of the literature
. Br J Sports Med
A case report of two male body builders using high doses of AAS who developed serious liver disease, including adenocarcinomas. One patient was asymptomatic; the other was hospitalized for acute renal failure as well. The cases follow the patients and shows the long-term reversibility of the disease and the need for monitoring of patients taking AAS.
19. Peters RL: Pathology of hepatocellular carcinoma
. In Hepatocellular Carcinoma. Edited by Okuda K, Peters RL. New York: John Wiley & Sons; 1976:107–168.
20. Friedl KE: Effects of anabolic steroids on physical health
. In Anabolic Steroids in Sport and Exercise. Edited by Yesalis CE. Champaign: Human Kinetics; 1993:107–150.
21. Vihko R: Endocrine consequences of androgenic-anabolic steroids in male athletes
. In Official Proceedings from International Athletic Foundation World Symposium on Doping in Sport. Edited by Bellotti P, Benzi G, Ljungqvist A. Florence, Italy. 1987:141–147.
22. Juhn M: Popular sports supplements and ergogenic aids
. Sports Med
23. Maravelias C, Dona A, Stefanidou M, Spiliopoulou C: Adverse effects of anabolic steroids in athletes: a constant threat
. Toxicol Lett
24. VanHelder WP, Kofman E, Tremblay MS: Anabolic steroids in sport
. Can J Sports Sci