Journal Logo


Anabolic Steroid Use and Longitudinal, Radial, and Circumferential Cardiac Motion


Author Information
Medicine & Science in Sports & Exercise: April 2012 - Volume 44 - Issue 4 - p 583-590
doi: 10.1249/MSS.0b013e3182358cb0
  • Free


Anabolic steroids (AS) are synthetic derivatives of the male hormone testosterone. It has long been established that AS increase skeletal muscular size and strength through an up-regulation of protein synthesis (18). These effects have led to the increasing trend of AS use for both performance and image enhancement (4). The exact number of current AS users in the United Kingdom and worldwide is difficult to gauge because of the legal issues surrounding their possession and use. Despite this, reports from needle and syringe programs (NEPs) have pointed to a large and continual growth of AS use in the United Kingdom, with the number of AS-injecting individuals attending the NEPs increasing 2000% from 1991 to 2006 (34).

Because of this trend for increasing AS use, there is a growing concern about the cardiovascular (CV) health consequences of AS use. This is because empirical evidence of increased CV risk is often equivocal. Concerns about CV health have been prompted by substantial case study data that have linked AS use with CV end points such as myocardial infarction, (32) stroke (44), and cardiomyopathies (8). Establishing a causal relationship from case studies is difficult, and thus, several traditional and new CV risk factors, such as blood pressure, blood lipid profiles, as well as homocysteine and C-reactive protein (CRP), have been studied in case–control studies of AS users (20,21,40), again with equivocal results. Other tools have been used to assess CV health including ECG, but data are limited, although a recent study linked AS use to shortened ventricular repolarization (7).

Some studies have suggested that left ventricular (LV) morphology and function may be negatively altered with AS use, but data are not consistent (13,36). With continuing developments in noninvasive imaging, recent studies have assessed LV function at a regional level and measured global function (3,11). Using tissue Doppler and cardiac strain ([Latin Small Letter Open E]) assessment, both D’Andrea et al. (11) and Baggish et al. (3) observed reduced LV function in AS users. D’Andrea et al. (11) reported only longitudinal [Latin Small Letter Open E], and Baggish et al. (3) presented data only for longitudinal and radial [Latin Small Letter Open E] in a small number of AS users. A comprehensive evaluation of cardiac [Latin Small Letter Open E] in all three planes of motion (longitudinal, radial, and circumferential), as well as assessment of rotation and torsion, would provide a significant addition to our understanding of the CV consequences of AS use. The determination of strain rate (SR) to provide peak SR in both systole and diastole provides added subtlety to functional cardiac measures.

The aim of the current study, therefore, was to assess traditional CV risk factors as well as to develop a more comprehensive assessment of LV structure and function using state-of-the-art noninvasive imaging tools. We hypothesized that there would be a significant effect of AS use on traditional risk factors as well as LV systolic and diastolic cardiac function at global and regional levels. We also investigated the short-term effect of a small number of case studies after brief removal from AS.



Strength-training individuals (n = 47, male = 46, female = 1 (AS)) were recruited through local gyms, personal contacts, and local syringe exchange programs. Posters and information sheets were placed in various locations and also distributed by hand to individuals. The posters stated that there was a study investigating the effects of strength/resistance training, either with or without the use of AS, on CV health. Inclusion criteria were age between 18 and 50 yr and resistance training history of minimum of 3 yr with three to four training sessions per week. Exclusion criterion for the study was the presence of known respiratory, CV, or musculoskeletal disease. A specific inclusion criterion for the AS-using group (AS: n = 28, age = 31 ± 7 yr) was a documented self-reported history of AS use for at least 2 yr (including on- and off-cycles). An inclusion criterion for the non-AS (NAS) group (n = 19, age = 28 ± 8 yr) was self-reported history of never taking AS. Before taking part in the study, details of each part of the testing were described to each of the participants. The study was granted ethics approval from the Liverpool John Moores Ethics Committee, and participants provided written informed consent.


A cross-sectional cohort design was used for the study, with participants required to make a single visit to our laboratory. Initially, subjects completed self-report questionnaires related to general health, training status, and history as well as detailed accounts of AS use. This was followed by assessment of body composition, a venous blood sample, and a comprehensive CV evaluation including brachial artery blood pressure, a resting 12-lead ECG, and an echocardiogram. All tests were conducted on the participants after an overnight fast, as well as a 24-h abstention from resistance training.


Participant history and AS use

Training history included data on years of training, the average number and length of sessions per week, as well as self-reported one-repetition maximums for the bench press and squat (Table 1). Those in the AS group provided a detailed history of AS use including names, dosage, and cycling information, and none of the participants self-reported coabuse of any other illicit substances. For simplicity, we provide a list of exemplar AS used but note that all subjects used multiple AS in various stacking procedures with varied periods of abstinence or “off-cycles.” The types of AS currently being used by some of the AS participants included trenbolone (n = 15), testosterone (n = 13), sustanon (n = 9), boldenone (n = 7), nandrolone (n = 10), meathandrostenolone (n = 8), drostanolone propionate (n = 3), oxandrolone (n = 4), and stanozolol (n = 3). Of those in the AS group who provided sufficient information to perform an analysis of their daily usage (n = 14), we found that the mean AS dose was 232 mg·d−1 with a SD of 208 mg (range = 35–707 mg·d−1). To give an indication of an AS regimen, we include a case exemplar of a participants’ AS use: testosterone propionate = 800 mg·wk−1, testosterone cypionate = 500 mg·wk−1, stanozolol = 350 mg·wk−1, and nandrolone decanoate = 600 mg·wk−1. This was broken down into three injections of testosterone propionate (one 400 mg and two 200 mg per injection), two of cypionate (250 mg per injection), and two of nandrolone decanoate (300 mg per injection) per week and 25 mg of oral stanozolol twice a day. This was repeated for 8 wk.

Body composition data and training history in AS and NAS groups (data are mean ± SD).

Body composition

Height and body mass were recorded using standard scales allowing the calculation of body mass index (BMI) and body surface area (BSA) (17). In a subsample from both groups, body composition, including lean mass, fat mass, and body fat percentage, was measured using dual-energy x-ray absorptiometry (QDR Discovery A, Hologic, MA; AS n = 22, NAS n = 15).

Lipid profile

Venous blood samples were taken from the antecubital vein directly into serum gel (serum) and lithium heparin vacutainers (plasma) (BD, Oxford, United Kingdom). Samples were centrifuged for 10 min at 3000 rpm, after which plasma and serum were placed into aliquots for immediate assessment of total cholesterol (AS n = 19, NAS n = 9), HDL (AS n = 20, NAS n = 12), LDL (AS n = 6, NAS n = 4), and triglycerides (AS n = 15, NAS n = 6) using the Daytona RS blood analysis machine (Randox, Co., Antrim, N. Ireland).

Blood pressure and ECG

At the end of a 10-min resting supine period, repeated resting brachial artery blood pressures were recorded from the left arm via an automated blood pressure monitor (AS n = 17, NAS n = 18) (Dinamap, GE Pro 300V2; GE Healthcare Systems, Waukesha, WI). A resting 12-lead ECG was performed (Mac 1200; GE Healthcare Systems) after this supine rest (AS n = 26, NAS n = 19). Key parameters assessed included rate, QRS complex voltages, and QTc duration (Bazett correction).


For the assessment of global and segmental cardiac structure and function, ultrasound echocardiography (Vivid Q; GE Healthcare, Oslo, Norway) was used to gather images of the LV in multiple planes from parasternal and apical acoustic windows (AS n = 37, NAS n = 19). A single experienced echocardiographer performed all imaging with the subject in the left lateral decubitus position. Intraobserver reliability was assessed through intraclass correlations for 2D, Doppler, tissue Doppler imaging, and [Latin Small Letter Open E]/SR data (9,10) with a range of 0.693–0.993 (P < 0.05).

Parasternal long-axis views allowed the collection of M-mode images at the tips of the mitral valve leaflets perpendicular to the septal and posterior walls, taking care to ensure clear endocardial definition. From M-mode traces, septal (IVSd) and LV posterior wall thickness (LVPWd) in diastole as well as LV chamber dimensions at end-diastole and end-systole (LVIDd, LVIDs) were assessed following American Society of Echocardiography guidelines (29). We estimated LV mass using a regression-corrected cube formula (15). An LV mass index was constructed by scaling allometrically for height raised to the power 2.7 because this is the technique most often used in a clinical setting. (14). In a subsample (n = 37, AS = 22, NAS = 15), we scaled LV mass for individual differences in lean body mass derived from dual-energy x-ray absorptiometry.

Apical two- and four-chamber views were digitized to assess LV end-diastolic and end-systolic volumes, which then allowed the estimation of stroke volume and ejection fraction (EF) using the Simpsons biplane method. The four-chamber color Doppler and then pulse-wave Doppler were used to assess peak flow velocities across the mitral valve. Using a 4-mm sample volume in the area of peak flow LV, early (E) and late (A) diastolic in-flow inflow velocities were recorded. In combination with LV outflow spectral Doppler envelopes, we also measured the isovolumetric relaxation time (IVRT) from closure of the aortic valve to opening of the mitral valve. From the same view, tissue Doppler measures of myocardial wall velocities were recorded. Taking care to adjust filters and scale and with the septal wall parallel to the ultrasound beam, we interrogated the mitral annulus at the septal wall, recording peak systolic (S′) as well as early (E′) and atrial (A′) diastolic tissue velocities. This also allowed the production of the E/E′ ratio that has been shown to estimate left atrial pressure (35).

Segmental and global [Latin Small Letter Open E] and SR data were obtained from parasternal short-axis views at the base (just below mitral valve) and apex (1–2 cm above LV cavity obliteration) as well as via a four-chamber apical view. Cine loops of LV motion were captured for offline analysis (Echopac; GE Healthcare, Oslo, Norway). Specific speckle-tracking software that tracks natural acoustic markers or “kernals” facilitated the estimation of [Latin Small Letter Open E] and SR in six wall segments in all views. In short-axis views, the LV was split into septal, anteroseptal, inferior, posterior, anterior, and lateral wall segments. Radial, circumferential, and rotational data from these segments were averaged to provide global [Latin Small Letter Open E], SR, rotation (Rot), and rotation rate (RotR) data. Comparison of basal and apical rotation facilitated the calculation of LV torsion (19,37,38). In the long-axis view, the LV was split into basal, midwall, and apical septal wall segments as well as basal, midwall, and apical lateral wall segments. Again, these were averaged to provide global measures of longitudinal [Latin Small Letter Open E] and SR. SR data were recorded in systole (SSR) and early diastole (ESR). Two-dimensional image optimization was performed including maintaining frame rate between 40 and 90 fps. For all measurements, images were analyzed offline by a single experienced technician with no knowledge of group allocation. Data reflect the average of three to five continuous cardiac cycles.

Case studies of “on”- and “off”-cycle assessments

A subgroup of AS users (n = 4, male = 3, female = 1) were tested during both a period of AS use (“on” cycle) and a period of abstinence (“off” cycle). The same experimental measures were taken at both testing points. To make the study as relevant to “real-world” scenarios, we aimed to test participants at the end of an “on” and an “off” cycle. The “on/off”-cycle durations ranged from 8 to 12 wk.

Statistical Analysis

Statistical analysis of data was performed using statistical software package SPSS Version 17 (SPSS, Inc., Chicago, IL). All data were subjected to tests of normality. Differences between AS and NAS participants were analyzed using paired t-tests if normally distributed or the Mann–Whitney U test if not normally distributed (e.g., E/A ratio data). P < 0.05 was considered significant.


Participant history and AS use

The AS group performed, on average, one more training session per week than the NAS group did, whereas the NAS had a longer (yr) history of training. There was little difference in the average session length between the two groups. Maximal bench press and squat were also significantly higher in the AS group (Table 1).

Body composition

Although height did not differ between groups, participants in the AS group were significantly heavier, which resulted in a significantly elevated BSA and BMI (Table 1). Lean mass and fat mass (81.1 ± 13.3 vs 70.7 ± 6.5 and 14.4 ± 7.7 vs 9.9 ± 7.7 kg, respectively, P < 0.05) were both significantly elevated in the AS users, whereas body fat percentage (14.6% ± 6.5% vs 11.9% ± 3.3%) was not significantly different between groups.

Lipid profile

LDL and HDL were significantly elevated and reduced, respectively, in the AS group, which led to a significantly greater total cholesterol-to-HDL ratio in the AS group (Table 2). Total cholesterol was elevated in the AS group, although this was not statistically significant. There was little difference in triglyceride levels between groups (0.84 ± 0.28 vs 0.82 ± 0.28 mmol/l).

Structural and functional data for the LV, and lipid profiles in AS and NAS groups (data are mean ± SD).

Blood pressure and ECG

Mean systolic pressures were elevated but not significantly higher; although mean diastolic pressures were lower in AS users, they did not reach statistical significance (systolic blood pressure = 133 ± 15 vs 126 ± 12 mm Hg, P = 0.11; diastolic blood pressure = 67 ± 7 vs 73 ± 10 mm Hg, P = 0.07). Resting HR was significantly higher in AS users (79 ± 12 vs 64 ± 13 beats·min−1, P = 0.01). Fourteen AS users and eight NAS met the voltage criteria for LV hypertrophy, but voltage data for R waves in V5 and V6, as well as S waves in V1 and V2, were not different between groups. R-R-corrected QT interval (QTc) was also not significantly different between AS users (409 ± 20 ms) and NAS (401 ± 23 ms). Three members of each group had a QTc marginally below the 380-ms cutoff for short QT.

Cardiac structure and function

Thickness of the IVSd, LVPWd, as well as LV mass were significantly higher in the AS group (Table 2). LV mass index, when scaled for height, remained significantly higher in the AS group; however, when scaled for lean body mass in a subsample, it was no longer significant (P = 0.13). There was no significant difference between groups for LV internal dimensions or volumes in systole or diastole, yet there was significantly reduced EF in the AS group. In diastole, significant reductions in E/A, E′, A′, and E′/A′ were observed in the AS group. A was significantly elevated in the AS group. There was no significant difference between groups in isovolumetric relaxation time (103 ± 14 vs 103 ± 15 ms) and E wave deceleration time (182 ± 37 vs 202 ± 40 ms).

Myocardial [Latin Small Letter Open E] and SR (longitudinal, circumferential, and radial)

In the longitudinal plane, peak [Latin Small Letter Open E] was significantly reduced in the AS group (Table 3). There was no significant difference in peak SSR, although peak ESR was significantly reduced and peak ASR significantly elevated in the AS group. This led to a significantly reduced E/A SR ratio in the AS users.

Longitudinal [Latin Small Letter Open E] and SR data for AS and NAS groups (data are mean ± SD unless otherwise stated).

In the circumferential plane, peak [Latin Small Letter Open E] and SSR did not differ between groups at the basal level (P > 0.05), whereas peak [Latin Small Letter Open E] was significantly reduced in the AS group at the apical level (P < 0.05; Table 4). Peak ESR was significantly reduced in the AS users at the apical level, and E/A SR was significantly reduced in the AS group at both basal and apical levels. At the apical level, peak ERotR was significantly elevated and peak rotation was significantly reduced in the AS group. Peak ASR was significantly elevated in the AS users at the basal level. Peak ARotR was significantly elevated in AS users at the basal level, which also lead to a significantly reduced E/A RotR. All other data were similar between groups (Table 4). Torsion was not different between the two groups.

Circumferential [Latin Small Letter Open E], SR, rotation, rotation rate, and torsion in AS and NAS groups (data are mean ± SD unless otherwise stated).

In the radial plane, there was no difference in peak [Latin Small Letter Open E] and SSR at the basal levels, whereas peak [Latin Small Letter Open E] and SSR were significantly elevated in the AS group at the apical level (Table 5). Peak ESR were similar in AS and NAS groups at both levels, but ASR was significantly elevated in the AS group at the basal level, which led to a significant depression in the basal E/A radial SR ratio.

Radial plane [Latin Small Letter Open E] and SR in AS and NAS groups (data are mean ± SD unless otherwise stated).

Case studies of AS users "on" and "off" cycle.

Moderate changes were seen in mass from “on” to “off” cycle with only a significant change seen in one subject (Table 6). HDL levels increased in all the three subjects when “off” cycle where data were available. Minimal differences in structure were seen at the different points of the AS cycles, with the greatest changes seen in diastolic function with two subjects showing improved E/A values when “off” cycle and all those tested at both time points exhibiting improved E/A SR values.

Cardiac structure, function, and HDL data from AS users when "on" and "off" cycle.


The unique contribution of this study is that it builds on the recent work of D’Andrea et al. (11) and Baggish et al. (3) by providing a comprehensive profile of cardiac [Latin Small Letter Open E] and SR data in multiple planes of LV motion. Specifically, the current study observed decreased longitudinal peak [Latin Small Letter Open E] as well as well as some evidence of decreased diastolic SR in the circumferential (apical) and radial planes (basal). Further, we provide data consistent with previous findings that AS use is associated with significant alterations in lipid profiles (23,28) as well as left ventricular concentric hypertrophy and diastolic impairment (25,36). Taken together, these data suggest that an increased CV risk profile exists in AS users.

Initially, we observed alterations in body composition, lipid parameters, and resting HR in AS users. The significant increases in weight, BSA, and BMI seen in the AS users were somewhat predictable (6) because AS have been shown to upregulate protein synthesis, thereby increasing the amount of lean muscle that can be developed when used in conjunction with intense resistance training (39). It is unlikely that the significant differences in body size and composition can be explained solely by small differences in the training exposure of the two groups.

In addition, we observed an altered lipid profile in AS users. Although mean total cholesterol was higher in AS users, the difference between groups was not statistically significant, which supports data from Sader et al. (43) but contradicts Baldo-Enzi et al. (5). It can be argued that lipid profiles are more important than overall cholesterol when determining CV and atherosclerotic risk (41). The decrease in HDL in AS users in the present study agrees with past research (5,22,28,43). Likewise, an increase in LDL in the current study also supports previous data (22). Supraphysiological doses of AS lead to high hepatic androgen exposure, and high androgen levels can alter levels of lipoprotein (a), which directly affects the formation of HDL (24).

The AS user group had a significantly elevated resting HR compared to NAS users. Although increased HRs in AS users have been shown in previous studies (11), these differences have not been significant. Case studies have observed increased HRs in AS users (45), but the present study is the first to report a significant difference in resting HR in a cohort study. An elevated resting HR in AS users could be due to mechanism(s) including a relative lack of aerobic/endurance training in AS users, an indirect effect of the increased atherosclerotic risk, or a direct effect of steroids on cardiac pace making. Although resting systolic and diastolic blood pressures observed in the present study did not differ significantly between the two groups, systolic blood pressure was elevated, whereas diastolic blood pressure was reduced in the AS group (7 and 6 mm Hg). Blood pressure data were highly variable between individuals, and a lack of between-group significance may be due to low power with respect to these data.

Larger LV wall thicknesses observed in the AS users in the present study confirm data reported in previous studies (3,36,42). It is also important to note that these data contradict other studies. Specifically, D’Andrea et al. (11) and Hartgens et al. (22) reported no increased wall thickness in AS users. Comparisons between AS user studies can be complicated by between-subject differences in training history, AS use history, and others, which may explain variable outcomes. The increase in LV wall thickness and LV mass may be expected because the receptor mechanisms that result in an increase in skeletal muscle protein up-regulation with AS use are present within the myocardium (33). In the current study, the increased LV wall thickness contributed to an elevated LV mass that remained even after scaling for between-subject differences in height (16). When LV mass was scaled for lean mass in a subsample (n = 37, AS = 22, NAS = 15), the difference between groups was no longer statistically significant (P = 0.13). This suggests that some, but likely not all, of the LV hypertrophy could be explained by an increase in whole body lean tissue. Whether there is cardiac tissue hypertrophy in excess of body size, suggesting a possible increased anabolic effect of AS in cardiac tissue over skeletal muscle, requires verification. Although we cannot exclude small between-group differences in training stimulus as a potential mediator of LV morphology, it is unlikely to underpin all of the observed variation between groups.

The increased LV morphology in AS users was also associated with significant between-group differences in LV function. Notably, the AS users had a reduced EF. The EF, however, remained within the normal range, which contradicts a recent report (3) but confirms other studies (22). The lack of difference in S′ between groups also supports previous studies (25,36). Although myocardial [Latin Small Letter Open E] and SSR assessments have become increasingly useful in quantifying regional myocardial function in individuals, from both a clinical (1) and a research perspective (30), their application in AS users has, to this point, been incomplete (3,11). We observed a significant reduction in peak longitudinal [Latin Small Letter Open E] in AS users, similar to that reported previously (3,11), as well as a significant reduction in circumferential [Latin Small Letter Open E] at the apical level. A small nonsignificant reduction in radial [Latin Small Letter Open E] at the basal level along with an increase in radial [Latin Small Letter Open E] at the apical level in AS users was observed in the current study, yet greater, and significant, decreases in AS users were reported by Baggish et al. (3). The absolute values of radial strain were lower in the AS users in the current study compared with the AS users assessed by Baggish et al. (3). Circumferential [Latin Small Letter Open E] as well as SSR in both apical and basal LV levels has not been reported in previous literature. Our observations of these parameters in AS users suggest that there is no consistent change in indices of LV motion/contraction between planes of motion and that these data should be confirmed in ongoing studies. Similarly, peak rotation and peak systolic rotation rates have not been reported previously in AS users and, in the current study, excluding peak rotation at the apical level, were largely unaltered by a history of AS use. Consequently, LV torsion although reduced in AS users was not significantly different between groups.

Whereas evidence of impairment of global diastolic function was observed in AS users, it is interesting to note that E and IVRT were not altered in the AS group. A lengthened IVRT has been reported before in AS users (36). Other alterations observed in the AS users were a decreased E′, suggestive of early relaxation impairment and increased A and A′, suggesting altered atrial contraction or LV compliance. Both E/A and E′/A′ ratios were reduced in AS users, which is suggestive of an overall reduction in “efficiency” of LV motion/filling, and again, these changes parallel published data (25). The lack of significant change in E in the current study is somewhat surprising, but given a reduction in E′, this may reflect pseudonormalization (26), which is supported by an increase in left atrial driving pressure as estimated by E/E′. An increased E/E′ has also been noted in past studies and was present in the present findings (25,27).

Previous studies using [Latin Small Letter Open E] and SR analysis have not assessed diastolic function (3,11). Although not consistent in all planes, levels, and segments, evidence of significant alterations in ESR, ASR, as well as E and A rotation rates provides new evidence for altered diastolic function in AS users. A reduction in ESR likely reflect alterations in early relaxation/elastic recoil of LV wall segments and any alteration in ASR or A rotation rate likely reflect changes in atrial contraction or LV compliance. Overall changes in filling or tissue movement during atrial diastole probably represent an increased reliance on atrial filling in the AS users as the LV adjusts to a reduction in early filling. The diastolic impairment observed in AS users could occur through possible impaired LV relaxation, increased LV diastolic pressure, and/or elevated left atrial pressure (31). However, the specific mechanisms underpinning altered early or atrial diastolic function cannot be directly determined from the current noninvasive study. Altered loading and rate may be considered, but the HR difference is small, and available data suggest that parameters such as E, E′, [Latin Small Letter Open E], and SR would increase with HR, not decline. Altered preload is difficult to assess in a cross-sectional design, and although some differences in blood pressure were apparent, it is problematic to link this to LV function in the current research design. The mechanism(s) by which AS could have a direct effect on LV diastolic relaxation and/or compliance are not fully understood, but some pathways have been postulated. D’Ascenzo et al. (12) speculated that AS negatively alter endothelial cell growth, promote apoptosis, and increase collagen cross-links in LV walls. Human data to support this speculation are currently limited but worth pursuing. Whether the changes in LV function are a direct consequence of long-term AS use or an indirect effect of AS, via altered lipid profiles or blood pressure, is worthy of further study.

Despite only a small number of AS subjects being tested when both “on” and “off” cycles, the results start to give an indication of changes in CV parameters that can occur after a relatively short period, 8–12 wk, of abstinence from AS use. These data also prompt ongoing studies in this area. Our data suggest some improvement in the lipid profiles of those tested, with decreases in overall total cholesterol and LDL and increases in HDL levels, thereby showing an improvement in perceived CV risk. Although strain and SR data from the different cycles were somewhat inconclusive, the increase in E/A SR in the radial plane seen in all users when “off” cycle, in conjunction with an increase in E/A ratios in three of the subjects, gives an indication of a potential improvement in ventricular relaxation after cessation of AS use for a relatively short period. It is important to note, however, that despite an improvement in ventricular filling ratios in three of the subjects, the values seen were still below the average of the NAS group, indicating that some restriction in ventricular relaxation remained.

The implications of the data presented in this study and in other studies are that AS use is associated with multiple changes in CV structure, function, health, and thus risk. This provides some potential links to case studies reporting significant CV events in AS users. Large epidemiological studies, although difficult in this type of cohort, should provide further insights into the “real” risk of AS use. What implications the current data have at an individual level is hard to discern. Overall, the increase in ventricular mass, decreased cardiac function (in particular, diastolic impairment) as well as marginal changes in blood pressure and alterations in blood lipid profiles point to an increased CV disease risk associated with AS use. The small sample of data from “on” and “off” cycles in AS users points to a possible, if limited, reversal of some of the negative CV effects seen in AS users.

As with any study of AS use, there are some inherent limitations that are important to note. Investigating AS use is complicated by the heterogeneity in subject-specific details (age, training status, diet, etc.) that were either not controlled or not assessed in the current study. Obviously, interindividual and intraindividual differences in AS dose (volume, drug type, veracity, or authenticity), self-report details of history, as well as variations in “stacking” and “cycling” approach can occur within the AS-using population. We made no a priori attempt to control these parameters to reflect the reality of AS use. Further to this, the differing types of AS use, which are used for differing outcome, i.e., for bulking or for “cutting” to potentiate lean mass, could have a confounding effect on the results, but we did not account for this in our analysis. We present some case studies of the effect of “cycling” of AS use, but this should be addressed in a larger study.

It is also pertinent to mention that all of the LV structural and functional measures in the present study were obtained through noninvasive techniques, thereby limiting the ability of the researchers to confirm the mechanistic reasons for any changes observed. Although we present what we consider is a comprehensive look at AS and their CV effects, the addition of radial and circumferential data at the mid-LV level may provide additional insight into global systolic functioning. The low number of participants in certain aspects of the data, such as aspects of the lipid profiles we present, does limit their interpretation to a wider group. An important and often overlooked area is the effect of AS on hematocrit levels. There are limited data concerning this area of research, but some results have demonstrated a negative effect (2,28). Further research could help to explain certain CV events and changes to LV geometry and function (e.g., strain and SR) that have been observed in AS users. Specifically, although we have confirmed increased CV risk in AS users by adding new data in relation to cardiac function in AS users at rest, it would make sense to assess cardiac functional responses to exercise stress because this may be an important adjunct to risk and the onset of events seen in case study reports in AS users. Although we document gross structural alterations in the LV of AS users, limited data are available for the right ventricle and the left atria. Further study using magnetic resonance imaging techniques may interrogate the effect of AS use on LV perfusion and/or the potential presence of fibrosis.

In conclusion, the present study has reported a range of negative CV consequence of AS use in conjunction with resistance training that is in agreement with recent findings (3,25). We have developed the available database by indicating an association between AS use and some evidence of impaired [Latin Small Letter Open E] and SR data, largely reflecting altered diastolic cardiac motion. Combined with altered LV mass, lipid profile changes in cardiac function with AS use likely increase the CV risk in AS-using subjects.

Funding for the initial start of the project was provided by a grant from the LJMU Institute of Health Research.

There are no disclosures or conflict of interest declared for each author.

The findings of the present study do not constitute endorsement by the American college of Sports Medicine.


1. Abraham TP, Dimaano VL, Liang HY. Role of tissue Doppler and strain echocardiography in current clinical practice. Circulation. 2007; 116 (22): 2597.
2. Alen M. Androgenic steroid effects on liver and red cells. Br J Sports Med. 1985; 19 (1): 15–20.
3. Baggish AL, Weiner RB, Kanayama G, et al.. Long-term anabolic–androgenic steroid use is associated with left ventricular dysfunction. Circ Heart Fail. 2010; 3 (4): 472–6.
4. Bahrke MS, Yesalis CE. Abuse of anabolic androgenic steroids and related substances in sport and exercise. Curr Opin Pharmacol. 2004; 4 (6): 614–20.
5. Baldo-Enzi G, Giada F, Zuliani G, et al.. Lipid and apoprotein modifications in body builders during and after self-administration of anabolic steroids. Metabolism. 1990; 39 (2): 203–8.
6. Bhasin S, Storer TW, Berman N, et al.. The effects of supraphysiologic doses of testosterone on muscle size and strength in normal men. N Engl J Med. 1996; 335 (1): 1–7.
7. Bigi MA, Aslani A. Short QT interval: a novel predictor of androgen abuse in strength trained athletes. Ann Noninvasive Electrocardiol. 2009; 14 (1): 35–9.
8. Bispo M, Valente A, Maldonado R, et al.. Anabolic steroid–induced cardiomyopathy underlying acute liver failure in a young bodybuilder. World J Gastroenterol. 2009; 15 (23): 2920–2.
9. Chan-Dewar F, Oxborough D, Shave R, et al.. Evidence of increased electro-mechanical delay in the left and right ventricle after prolonged exercise. Eur J Appl Physiol. 2010; 108 (3): 581–7.
10. Chan-Dewar F, Oxborough D, Shave R, Gregson W, Whyte G, George K. Left ventricular myocardial strain and strain rates in sub-endocardial and sub-epicardial layers before and after a marathon. Eur J Appl Physiol. 2010; 109 (6): 1191–6.
11. D’Andrea A, Caso P, Salerno G, et al.. Left ventricular early myocardial dysfunction after chronic misuse of anabolic androgenic steroids: a Doppler myocardial and strain imaging analysis. Br J Sports Med. 2007; 41 (3): 149–55.
12. D’Ascenzo S, Millimaggi D, Di Massimo C, et al.. Detrimental effects of anabolic steroids on human endothelial cells. Toxicol Lett. 2007; 169 (2): 129–36.
13. De Piccoli B, Giada F, Benettin A, Sartori F, Piccolo E. Anabolic steroid use in body builders: an echocardiographic study of left ventricle morphology and function. Int J Sports Med. 1991; 12 (4): 408–12.
14. de Simone G, Devereux RB, Daniels SR, Koren MJ, Meyer RA, Laragh JH. Effect of growth on variability of left ventricular mass: assessment of allometric signals in adults and children and their capacity to predict cardiovascular risk. J Am Coll Cardiol. 1995; 25 (5): 1056–62.
15. Devereux RB, Alonso DR, Lutas EM, et al.. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol. 1986; 57 (6): 450–8.
16. Dewey FE, Rosenthal D, Murphy DJ Jr, Froelicher VF, Ashley EA. Does size matter?: clinical applications of scaling cardiac size and function for body size. Circulation. 2008; 117 (17): 2279.
17. Du Bois D, Du Bois EF. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med. 1916; 17: 863–71.
18. Ferrando AA, Tipton KD, Doyle D, Phillips SM, Cortiella J, Wolfe RR. Testosterone injection stimulates net protein synthesis but not tissue amino acid transport. Am J Physiol. 1998; 275 (5 Pt 1): E864–71.
19. George K, Shave R, Oxborough D, et al.. Left ventricular wall segment motion after ultra-endurance exercise in humans assessed by myocardial speckle tracking. Eur J Echocardiogr. 2009; 10 (2): 238.
20. Grace FM, Davies B. Raised concentrations of C reactive protein in anabolic steroid using bodybuilders. Br J Sports Med. 2004; 38 (1): 97–8.
21. Graham MR, Grace FM, Boobier W, et al.. Homocysteine induced cardiovascular events: a consequence of long term anabolic–androgenic steroid (AAS) abuse. Br J Sports Med. 2006; 40 (7): 644–8.
22. Hartgens F, Cheriex EC, Kuipers H. Prospective echocardiographic assessment of androgenic–anabolic steroids effects on cardiac structure and function in strength athletes. Int J Sports Med. 2003; 24 (5): 344–51.
23. Hartgens F, Kuipers H. Effects of androgenic–anabolic steroids in athletes. Sports Med. 2004; 34 (8): 513–54.
24. Hartgens F, Rietjens G, Keizer HA, Kuipers H, Wolffenbuttel BH. Effects of androgenic–anabolic steroids on apolipoproteins and lipoprotein (a). Br J Sports Med. 2004; 38 (3): 253–9.
25. Kasikcioglu E, Oflaz H, Umman B, Bugra Z. Androgenic anabolic steroids also impair right ventricular function. Int J Cardiol. 2009; 134 (1): 123–5.
26. Khouri SJ, Maly GT, Suh DD, Walsh TE. A practical approach to the echocardiographic evaluation of diastolic function. J Am Soc Echocardiogr. 2004; 17 (3): 290–7.
27. Krieg A, Scharhag J, Albers T, Kindermann W, Urhausen A. Cardiac tissue Doppler in steroid users. Int J Sports Med. 2007; 28 (8): 638–43.
28. Lane HA, Grace F, Smith JC, et al.. Impaired vasoreactivity in bodybuilders using androgenic anabolic steroids. Eur J Clin Invest. 2006; 36 (7): 483–8.
29. Lang RM, Bierig M, Devereux RB, et al.. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005; 18 (12): 1440.
30. Langeland S, D’Hooge J, Wouters PF, et al.. Experimental validation of a new ultrasound method for the simultaneous assessment of radial and longitudinal myocardial deformation independent of insonation angle. Circulation. 2005; 112 (14): 2157.
31. Little WC, Oh JK. Echocardiographic evaluation of diastolic function can be used to guide clinical care. Circulation. 2009; 120 (9): 802.
32. Lunghetti S, Zaca V, Maffei S, et al.. Cardiogenic shock complicating myocardial infarction in a doped athlete. Acute Card Care. 2009; 11 (4): 250–1.
33. Marsh JD, Lehmann MH, Ritchie RH, Gwathmey JK, Green GE, Schiebinger RJ. Androgen receptors mediate hypertrophy in cardiac myocytes. Circulation. 1998; 98 (3): 256–61.
34. McVeigh J, Beynon C, Bellis MA. New challenges for agency based syringe exchange schemes: analysis of 11 years of data (1991–2001) in Merseyside and Cheshire, United Kingdom. Int J Drug Policy. 2003; 14 (5–6): 399–405.
35. Nagueh MD, Sherif F, Middleton RCT, et al.. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol. 1997; 30 (6): 1527–33.
36. Nottin S, Nguyen LD, Terbah M, Obert P. Cardiovascular effects of androgenic anabolic steroids in male bodybuilders determined by tissue Doppler imaging. Am J Cardiol. 2006; 97 (6): 912–5.
37. Oxborough D, Batterham AM, Shave R, et al.. Interpretation of two-dimensional and tissue Doppler-derived strain ([Latin Small Letter Open E]) and strain rate data: is there a need to normalize for individual variability in left ventricular morphology? Eur J Echocardiogr. 2009; 10 (5): 677.
38. Oxborough D, Whyte G, Wilson M, et al.. A depression in left ventricular diastolic filling following prolonged strenuous exercise is associated with changes in left atrial mechanics. J Am Soc Echocardiogr. 2010; 23 (9): 968–76.
39. Payne JR, Kotwinski PJ, Montgomery HE. Cardiac effects of anabolic steroids. Heart. 2004; 90 (5): 473–5.
40. Riebe D, Fernhall B, Thompson PD. The blood pressure response to exercise in anabolic steroid users. Med Sci Sports Exerc. 1992; 24 (6): 633–7.
41. Robinson JG, Wang S, Smith BJ, Jacobson TA. Meta-analysis of the relationship between non–high-density lipoprotein cholesterol reduction and coronary heart disease risk. J Am Coll Cardiol. 2009; 53 (4): 316–22.
42. Sachtleben TR, Berg KE, Elias BA, Cheatham JP, Felix GL, Hofschire PJ. The effects of anabolic steroids on myocardial structure and cardiovascular fitness. Med Sci Sports Exerc. 1993; 25 (11): 1240–5.
43. Sader MA, Griffiths KA, McCredie RJ, Handelsman DJ, Celermajer DS. Androgenic anabolic steroids and arterial structure and function in male bodybuilders. J Am Coll Cardiol. 2001; 37 (1): 224–30.
44. Santamarina RD, Besocke AG, Romano LM, Ioli PL, Gonorazky SE. Ischemic stroke related to anabolic abuse. Clin Neuropharmacol. 2008; 31 (2): 80–5.
45. Sullivan ML, Martinez CM, Gallagher EJ. Atrial fibrillation and anabolic steroids. J Emerg Med. 1999; 17 (5): 851–7.


©2012The American College of Sports Medicine