Characteristics and outcome of patients with heart failure due to anabolic-androgenic steroids

Abstract

Objectives. The objective of the study was to analyse the outcome of patients with advanced heart failure due to abuse of anabolic–androgenic steroids. Design. A retrospective chart review of patients admitted or referred for advanced heart failure, due to anabolic–androgenic steroid abuse, in the period 2009–2013 was performed. Results. In 6 of 9 patients (median age: 31, all males) referred in the study period, some potential for recovery of left ventricular (LV) function was seen (P < 0.002), with a maximal improvement in LV ejection fraction reached within 6 months of treatment with angiotensin-converting enzyme inhibitors and beta-blockers. The remaining 3 patients required implantation of a LV assist device (LVAD) and were listed for heart transplantation. No recovery of LV function in the patients treated with assist device was seen. Conclusion. Anabolic–androgenic steroid-induced advanced heart failure is generally not a reversible condition. If diagnosed in the early stages some recovery of ventricular function is possible, but the long-term prognosis is uncertain. Likely, a substantial proportion of patients will eventually require LVADs or cardiac transplantation.

Key words::

• AAS
• anabolic
• androgenic
• heart failure
• steroids

Anabolic–androgenic steroids (AAS) include the male hormone testosterone and its synthetic derivatives. Since the 1950s, AAS have been abused by athletes in competitive sports because of their performance-enhancing and strength-increasing abilities (1). In the 1980s, the abuse spread to the general public, when amateur and recreational athletes in regular gyms began to abuse AAS (2). Access to the illegal drugs is now easier than ever: stacking regimes are generously provided by sellers and users of AAS on the internet. Among the substances used are nandrolone, stanozolol, boldenone, drostanolone, trenbolone, decanoate and androstane. Illicit use of AAS often results in blood concentrations of testosterone (or equivalent) more than 100 times the physiological blood levels. When AAS are used in supraphysiological doses, the number of side effects is abundant and involves most organ systems. Endocrinological side effects include gynaecomastia, testicular atrophy, acne, infertility, suppression of testicular function and male baldness. AAS abuse has been associated with hepatocellular adenomas. Other hepatic effects are general hepatotoxicity resulting in steatosis or peliosis. Central nervous system side effects include increased aggressiveness (3). In a web-based survey among bodybuilders, self-reported side effects included acne, testicular atrophy, loss of hair, gynaecomastia and fluid retention (4).

The most severe side effects are seen in the cardiovascular system. It is now generally accepted that AAS abuse may cause hypertension and it is suspected to be a cause of sudden cardiac death. Several case reports link AAS abuse to intracoronary thrombi, and atrial and ventricular arrhythmias. It now also seems established that AAS abuse, maybe only in prone individuals, can cause cardiomyopathy (CM), either in the form of hypertrophic or dilated CM (DCM) (5–8). A recent study shows that the long-term effects of AAS abuse may affect the systolic and diastolic function of the heart to a greater extent than previously reported (9). However, very little information is available on the presentation and outcome of AAS abusers referred for management of advanced heart failure (HF).

Given the increasing abuse of AAS in countries worldwide, a better understanding of the natural history of patients with AAS-induced HF seems important. We present a consecutive series of patients referred for management in a tertiary care and heart transplant centre.

Methods

We conducted a retrospective chart review of patients referred for management of advanced HF in the Department of Cardiology, Rigshospitalet, Copenhagen, Denmark from 2009 to 2013. This series includes patients referred from outside hospitals (catchment area of approximately 1.5 million inhabitants for advanced HF management and 3 million for heart transplantation [HTX]) as well as few patients admitted directly from the hospitals’ emergency department. Patients with severe HF in whom current or previous AAS use was identified were included. AAS users who might have been admitted due to other cardiac complications such as myocardial ischaemia or arrhythmias, but did not have severe HF, were not identified or included.

Follow-up was conducted by chart review by assessing the status at the last clinic visit to the department or by death and was 100% complete.

The study is strictly descriptive, and conventional descriptive statistics have been used. Data are presented as medians (range). Paired data were compared using Wilcoxon test. A P value of < 0.05 was considered significant.

Results

A total of 9 patients with current or previous AAS abuse were referred to our hospital for treatment of advanced HF (Table I). Of these patients, 8 presented as acute de-novo HF, whereas one had chronic HF and was referred for transplant evaluation. Of the 8 patients presenting with de-novo HF, 7 were initially admitted to the cardiology intensive care unit (ICU) before being transferred to the HF ward.

Table I. Baseline characteristics.

Baseline characteristics	Median (range)
Age, years	31 (20–29)
Height, cm	180 (130–183)
Weight, kg	92 (64–119)
Body mass index, kg/m^2	29.6 (20.4–38.9)
Diabetes mellitus type 2 (yes/no)	1/8
Hypertension (yes/no)	0/9
Atrial fibrillation (yes/no)	2/7
Abused substance	Within 3 months/former use/none/unknown
AAS injections	4/5/0/0
Alcohol	0/1/8/0
Illicit drugs	3/2/2/2
Smoking	4/1/4/0

AAS, Anabolic–androgenic steroids; Illicit drugs, Cocaine, cannabis, amphetamine, ecstasy, speed and ephedrine

The patients were all male, aged from 20 to 49 years (Table I). None had prior cardiac disease or familiar disposition to DCM or hypertrophic CM.

Five patients were recreational athletes, one former elite boxer, one professional weight lifter, one bodybuilder, and one former semi-professional ice hockey player. For several patients, the AAS abuse had not been revealed at the time of referral, but during the hospital stay all patients admitted abuse of AAS. In all patients, abuse included injected drugs. In most cases it was not possible to extract a complete history of individual drugs and doses used, but most commonly used drugs were testosterone, trenbolone, Decadron, oxymetholone, nandrolone and dehydroepiandrosterone. Several patients reported use of tamoxifen to prevent gynaecomastia. The duration of abuse ranged from two months to fifteen years. On admission, 4 patients were still active users (abuse within one year) whereas 5 had ended the abuse previously (1 1/2 to 10 years ago). More than half of the patients reported previous or present abuse of other illicit drugs. The drugs most frequently used were cocaine, cannabis and/or amphetamine.

Symptoms and signs

Symptom duration prior to admission ranged from three weeks to three months, dyspnoea being the predominant symptom, and five patients had haemoptysis prior to admission. Six patients had pulmonary oedema at admission and two complained about lower extremity oedema. One patient complained of angina, but there was no evidence of acute myocardial infarction and the coronary angiogram was normal.

All patients presented with sinus tachycardia, except one who had atrial fibrillation with high ventricular rate. Echocardiography revealed a left ventricular ejection fraction (LVEF) between 10 and 27% at time of admission. All patients presented with a DCM phenotype with left ventricular internal dimension in diastole (LVIDd) ranging from 6.1 to 8.5 cm. Indeed, increased wall thickness was seen only in one patient at the time of referral (Table II).

Table II. Echocardiographic measures at presentation and the last follow-up.

Patient	Day	LVEF	LVIDd	IVSd	LVPWd	LA size	E/A	E/é
1	0	19	6.1	0.9	1	85	2.2	7.6
	7	20	6.3	0.8	1.4	131	2.0	5.3
2	0	19	8.5	0.9	1.0	128	2.4	12.2
	694	13	93	0.9	0.8	198	−	25.4
3	0	19	7	1	0.9	175	−	−
	1	21	7.2	0.6	0.7	155	1.3	5.2
4	2	13	7.5	0.7	0.8	96	2.3	12.7
	386	33	6.1	0.9	0.9	52	0.9	5.2
5	0	10	7.1	0.6	1.1	127	1.3	−
	418	48	5.9	1	0.8	65	−	−
6	5	20	7	0.9	1.0	105	0.9	4.3
	762	43	6	1.0	0.9	68	1	4.7
7	0	27	5.7	0.8	0.9	52	1.1	−
	152	49	5.2	1.1	0.9	39	0.7	3.8
8	0	22	7.4	1.2	0.9	210	−	−
	331	55	5.8	0.9	0.5	89	3.5	4.7
9	0	20	7.6	1.1	1	103	1.3	20
	351	53	5.3	1	1	43	1.1	4.2

LVEF, Left ventricular ejection fraction; LVIDd, Left ventricular internal dimension in diastole; IVSd, Interventricular septal thickness at diastole; LVPWd, Left ventricular posterior wall dimension; LA, Left atrium

In-hospital course

For the seven patients admitted to the ICU, length of stay ranged from one to seven days. One patient required intubation and four required treatment with inotropes. One patient was treated with an intra-aortic balloon pump. Two patients required long-term mechanical circulatory support and had a left ventricular assist device (LVAD) (Heartmate II [HMII]) implanted 7 and 12 days from initial admission. They subsequently underwent HTX.

Atrial arrhythmias occurred during admission in three patients and ventricular arrhythmias in one.

All patients were discharged alive from hospital.

One patient referred for HTX evaluation had an LVAD implanted as a bridge to transplant 696 days after initial evaluation. The patient later died due to pump thrombosis of the LVAD.

Long-term outcome and recovery of LV function

After discharge, patients were seen in the outpatient clinic for adjustment of medical treatment (uptitration of angiotensin-converting enzyme [ACE] inhibitors and beta-blockers, and adjustment of diuretics). Of the 6 patients not treated with an LVAD, 4 were titrated to target doses of ACE inhibitors within four months and two reached target doses for beta-blockers. Patients treated with LVADs were also uptitrated in ACE inhibitors and beta-blockers to promote LV recovery.

Table II shows the change in LVEF over time for all patients. Among the patients who did not require an LVAD, some recovery of LV function was evident [from 20 (10–27)% to 49 (33–55)%, P = 0.027]. Recovery seems to have reached maximum within 6–12 months of treatment. No further improvement was seen in the patients who had two years of follow-up.

Three patients had a HMII implanted as a bridge to transplantation, two of whom underwent HTX after 14 and 18 months. There was no recovery of LVEF when supported with an LVAD in any of the 3 patients. The remaining 6 patients have a median follow-up of 16 (range: 7–41) months, and are all alive.

Discussion

The current study demonstrates that advanced HF associated with AAS abuse has a grave prognosis, although improvement in LV function is seen in some patients upon cessation of the abuse and with medical management of HF. To our knowledge, this series is the first to report on outcomes in AAS- induced advanced HF.

The pathophysiology of AAS-induced CM in humans is poorly understood, but animal studies have shed some light on the mechanisms involved. In rats treated with anabolic steroids, cardiac muscle shows decreased compliance resulting in a lower stroke volume (10). Hassan et al. found apoptotic changes and hypertrophy of cardiomyocytes in rats treated with AAS (11). Zaugg et al. found that rat cardiomyocytes are prone to apoptotic cell death in a dose-dependent manner and hypertrophy when exposed to AAS (12).

An autopsy study of 4 sudden cardiac deaths related to AAS showed left ventricular (LV) hypertrophy associated with fibrosis and myocytolysis as well as small vessel disease with intima hyperplasia in all (13). In a cohort of 87 deceased males who tested positive for AAS, there was a significantly higher heart mass compared with that in the control group (14). In a study of 12 long-term AAS abusers, Baggish et al. (2012) found severe systolic and diastolic dysfunction severe enough to increase the risk of HF. They proposed a toxin-mediated myocardial impairment as an explanation and suggested no dose–response effect, but a possible individual cardiotoxicity (9). In a study of AAS abusers, Urhausen et al. found LV hypertrophy even after discontinuing AAS abuse (15).

Based on rat studies and pathological studies, the effects of AAS on human cardiac muscle is thought to be LV hypertrophy causing impaired diastolic function and eventually leading to systolic dysfunction and progression to DCM.

Acknowledging the absence of precise information on doses of AAS in our series, there was no obvious direct dose-dependent effect on the severity of the HF or recovery potential. One patient with massive abuse over several years was able to demonstrate recovery of LVEF from 27 to 49% with medical treatment, while a patient with only 6 months of weekly injections was not able to sustain adequate heart function and required a HTX. In between these extremes were patients who reported only a few months of AAS abuse, or even a few months of abuse years earlier, who were able to obtain a partial recovery of LVEF, from 20 to 43% and 13 to 33%. Individual susceptibility to the toxic effect of AAS might explain the different response in the patients.

AAS-induced HF is probably a slowly progressing disease. However, it is impossible to predict if all or only some of our patients will eventually need an LVAD or HTX. We believe that the 6 patients with partial recovery of LV function represent the early stages of AAS-induced HF. In two of our patients, the AAS-induced HF was so advanced that they would have died without an LVAD as a bridge- to-transplant. So far, AAS abuse has been a minor indication for transplantation in our heart transplant programme. To date, only 3 out of 402 transplanted patients had an underlying diagnosis of AAS abuse. There are numerous limitations of this series.

First of all it included only 9 patients. We have no reliable information regarding the substances abused or doses administered by our patients. This is further complicated by the fact that AAS are bought from an uncontrolled market and the contents of the ampullas are not pharmacologically verified.

Conclusion

AAS may lead to advanced HF requiring treatment with LVAD or HTX. Some recovery is seen in most patients, but it is difficult to predict. In patients presenting with unexplained HF, a detailed medical history including specific questions on current or previous AAS use is essential.

Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.