Inverse Correlation between Testosterone and Ventricle Ejection Fraction , Hemodynamics and Exercise Capacity in Heart Failure Patients with Erectile Dysfunction

Background: Neurohormonal activation and abnormalities in growth hormone and testosterone concentrations have been reported in heart failure (HF). Erectile dysfunction(ED) is common in these patients and contributes to a low quality of life. No data are known regarding the correlation between testosterone and hemodynamics, exercise capacity and cardiac function in HF patients with ED, a marker of endothelial dysfunction. The aim of this study was to correlate testosterone levels with cardiac function, hemodynamic and exercise capacity in HF patients with ED. Materials and Methods: Fifteen HF patients underwent a six-minute treadmill cardiopulmonary walking test (6’CWT) and, ten minutes later, a maximum cardiopulmonary exercise test. Also, testosterone and other hormones were determined at rest. Results: Among hemodynamic variables only diastolic blood pressure on 6’CWT was correlated with testosterone levels(r =0.66, p = 0.007). The variables on exercise tests, VE/VCO2 slope and oxygen consumption did not show any correlation, except the distance at 6’CWT (r = 0.50, p = 0,047). Right and left ventricle ejection fraction showed inverse correlation with testosterone (r =0.55, p = 0.03 and r =0.69, p = 0.004 respectively). Conclusion: Testosterone levels correlated directly with distance at six-minute cardiopulmonary walk test and inversely with diastolic blood pressure, right and left ventricle ejection fraction in heart failure patients with erectile dysfunction. Further elucidation of mechanisms as regards testosterone action in these patients is warranted.


INTRODUCTION
Heart failure (HF) can be considered as the of mortality and morbidity worldwide (1).The left ventricular systolic dysfunction and limited exercise capacity manifested by breathlessness and fatigue are determinants of mortality and clinical events in the follow-up of HF patients.(2,3).Multiple mechanisms have been reported to be related to exercise capacity including diastolic and systolic cardiac function, (3).Sexual satisfaction is an important component dysfunction (ED) affects 60 to 70% of HF clinic outpatients (4).Symptoms, hormonal abnormalities, hemodynamic status, medication side effects and psychological factors are the major contributors to this sexual disorder.
In the physiopathology of chronic HF, the neurohormonal hypothesis for progression of heart failure has been considered of greater interest than the original hemodynamic mechanisms (5).Activation of sympathetic nervous system, renin-angiotensin-aldosterone system, arginine vasopressin, and endothelins are considered as neurohormonal targets in the treatment of HF.However, other hormonal abnormalities in growth hormones, insulin-like growth factor I and testosterone concentrations (6).The hormonal and cytokine activation contributes to peripheral muscle tissue wasting as well as anabolic/catabolic imbalance (7).In men, testosterone seems to play a role and vasodilator processes.No data has been reported as regards testosterone's effect on hemodynamics, cardiac function and exercise capacity in HF patients with ED.
The aim of this study was to evaluate the correlation between serum testosterone levels and cardiac function, hemodynamics, and exercise capacity in HF patients with ED.In addition, we determined a HF izing hormone, follicle-stimulating hormone, resting norepinephrine, and thyroid hormones.

Studied Population
Fifteen randomized chronic male HF patients, 50 ± 10 years with ED (Table-1) for at least four months, in a steady relationship and presenting inability to achieve or maintain a durable erection to permit satisfactory sexual intercourse (the 5th-item of the International Index of Erectile Dysfunction) (8,9).Patients were in stable clinical condition for three months without testosterone replacement or drugs teride, opiates, glucocorticoids and anticonvulsants.All patients were sedentary and did not have heavy alcohol consumption, nephrotic syndrome or liver cirrhosis history.All patients were evaluated for ED syndrome.The protocol was approved by the Ethical Committee of the Heart Institute.Subjects provided written informed consent before participation.Excluprevious ED therapy, recent use of phosphodiesterase inhibitors, psychiatric or psychological disorders, unstable angina or recent myocardial infarction, syncope, high-risk arrhythmias, disease with limitation for exercise except HF, and symptomatic hypotension or systolic blood pressure (SBP) < 85 mm Hg.

Exercise Protocol
We measured systolic blood pressure (SBP) and diastolic blood pressure (DBP) with the patient in an upright position immediately before each exercise test, at the last minute of the six-minute walking test (6'WT), at maximum exercise peak, and at 1-minute recovery (10).Electrocardiography was continuously monitored.Pulmonary ventilation and gas exchange data were determined on a breathby-breath basis with a computerized system (model Vmax 229 Sensormedics).The six-minute walking test (6'WT) was performed using a programmable treadmill without inclination and with patient-controlled speed (Series 2000, Marquette Electronics) at least 2 hours after a light meal and with controlled room temperature (21°C to 23°C).The patients were oriented to walk according to Borg's scale, with exertion level ranging from light to somewhat hard, from 11 to 13.After return of heart rate (HR), SBP, DBP, and symptoms to basal condition, patients underwent a progressive exercise test using a modiperform maximum exercise until exhaustion or the onset of non tolerated symptoms occurred and the respiratory exchange ratio exceeded 1.0.The peak oxygen consumption (peak VO 2 ) was considered the maximum reached VO 2 value.

Cardiac Function and Hormonal Determinations
The right ventricular and left ventricular ejection fraction (RVEF and LVEF, respectively), as a percentage, were determined by echocardiogram.Hormonal dosage at rest before the exercise test included determination of serum total testosterone hormone, norepinephrine and follicle stimulating hormone.The normal value of testosterone in this study total cholesterol, HDL cholesterol, LDL cholesterol, Triglycerides.Due to hormonal circadian rhythms, all these tests were performed in the same period (early morning).

Statistical Analysis
The descriptive analysis was presented as mean and standard deviation.The variables studied underwent the non-parametric Spearman test for correlation, considering p < 0.05 (SPSS Statistical Software for Windows version 11.5.Inc., Chicago, IL, USA).

RESULTS
All patients had total testosterone serum levels in the range for normal subjects (200-950 ng/dL).Four patients had prolactin serum levels below the normal range (2.5-11.5 ng/mL).One patient had hypothyroidism and the other hyperthyroidism.-4)

Correlation between Hormonal Levels and Exercise Capacity Data
Exercise capacity variables on 6´CWT, VE/VCO 2 slope at six minutes, oxygen consumption at six minutes did not any show correlation with testosterone levels or hormonal serum levels.Only maximum distance at 6´CWT showed correlation (r = 0.50, p = 0.047) with testosterone levels.Also during the MCT, maximum VE/VCO 2 slope, maximum oxygen consumption and time of maximum test did not show correlation either (Table -

Figure 1 -A) Regression linear plots in all patients between total testosterone and right ventricular ejection fraction (%, by echography). B) Regression linear plots in all patients between total testosterone serum levels and left ventricular ejection fraction (%, by echography).
Figure 2 -Regression linear plots in all patients between total testosterone serum levels systemic diastolic blood pressure at the last minute of the 6 minutes walking test.
levels, except diastolic blood pressure and distance on 6'CWT.Other hormones abnormalities can be found in heart failure patients with erectile dysfunction.
Low testosterone levels were reported in severe stroke and acute myocardial infarction (11).In HF, the testosterone serum levels may be low or in the normal range depending of the severity of the disease (11).Also, it has been reported that plasma levels of dehydroepiandrosterone sulfate decreased in patients with chronic HF in proportion to the severity evaluated by marker of cardiac function (12).However, correlation was demonstrated between left and right systolic cardiac function and total testosterone serum levels in selected patients with HF and ED (6,13,14).It is also partially discordant that, after testosterone administration, serum total levels remained in the were found in BNP, TNF-or LVEF.Also, there was also a positive correlation between testosterone and cardiac output (15).However, it was concordant with worsening of left ventricular remodeling with testosterone administration.The mechanisms to explain the heart are controversial and alternative hypotheses could be proposed.
terone serum blood levels could play a role in pathophysiology of HF, worsening cardiac function.This is concordant with the concept that anabolic steroids were considered as having cardiac toxicity with alterations of cellular pathology and organ physiology similar to those seen with heart failure and cardiomyopathy (16).In addition, testosterone treatment in very high supra-physiological doses causes myocardial hypertrophy and stiffening (17).Investigators administrated physiologic doses of testosterone and found an increase in the left ventricular diameters, but did The second hypothesis, in contrast is that fects on HF, and a resistance for testosterone could be proposed in selected HF patients (6).In rats, androgen therapy has been reported to improve coronary blood improving cardiac function.In addition, animal stud-

COMMENTS
Ours results demonstrated that there was an inverse correlation between right and left ventricular function and total testosterone serum levels in patients with ED.Other hemodynamic and exercise variables did not show a correlation with testosterone serum renin-angiotensin system (10,28).Also, vascular effects could be dose-dependent manner, with opposite effects according to the dose.Despite previous reported improvement in New York Heart Association functional class and exercise capacity, the potential prescription of testosterone for HF should be evaluated.This potential prescription should consider our results and previous study that showed the increment of left diameters in HF patients after testosterone administration (17).Further investigations should be performed concerning elucidation of its action in the heart and to determine if it is safe.Acute hemodynamics, exercise and primary endpoints for heart failure treatment (17,27).
Although this study is limited by the number of patients and did not include other markers of HF, nevertheless cardiac function could be considered as one of the main surrogate endpoints in heart failure.A better elucidation of testosterone mechanisms and action is warranted in a larger patient population.

CONCLUSION
Testosterone levels correlated directly with distance at six-minute cardiopulmonary walk test and inversely with diastolic blood pressure, right and left ventricle ejection fraction in heart failure patients with erectile dysfunction.

CONFLICT OF INTEREST
None declared.ies demonstrated vasodilator effects of androgens with (19), increment in IGF-1 levels with reduction in hyperinsulinemia and insulin resistance (20).In addition, resistance to growth hormone (GH) was also proposed in severe HF patients based on higher GH concentrations in proportion with low IGF-1 concentration (21).However, the hypothesis of testosterone resistance needs to be proved in future investigations including mediators, nongenomic and genomic androgen action mechanisms (10).
Our method to include these patients in this and higher severity of our patients in comparison with other studies (18).Patients with HF may experience erectile dysfunction for similar reasons to the general population, however, there are social, psychological, physiological, and drug-related consequences spepatients should be considered if the concept that this symptom is a marker of severity of HF and endothelial dysfunction is accepted (23).Endothelial dysfunction appears to affect all cardiac and peripheral circulation.
and functional class and six-minute walk test has been reported (24).Moreover, as suggested for GH resistance in more severe patients, our results could The correlation between total testosterone and exercise capacity is concordant with previous randomized, double blind, placebo-controlled parallel trial of testosterone replacement therapy.Exercise capacity compared with placebo (18).There are evidences in animal studies that anabolic androgens attenuate muscle fatigue in response to exercise, although the androgens may have different effects on heart and peripheral muscle (15).
Our correlation between testosterone and exercise, systemic DBP is in concordance with vasodilator effect of testosterone, but previous reports have speculated that for a vasodilatation effect of testosterone supra physiological concentrations are required (26,27).However, beyond the vessel relaxation testosterone effect, this may activate the

EDITORIAL COMMENT
Decreased testosterone level was reported throughout a 4-year follow-up in elderly patients with erectile dysfunction (ED) in addition to the association heart failure (CHF) and may underpin features of advanced disease, including reduced skeletal muscle mass and fatigue (2).Testosterone is known to act as a vasodilator in systemic, coronary and pulmonary vascular beds, as well as having anabolic properties (3).This effect could potentially lead to increased cardiac output and improved cardiovascular function administration of testosterone to men with chronic congestive heart failure may lead to hemodynamic positively correlated with cardiac output and exercise improvement in both these parameters has been observed following testosterone replacement therapy.Testosterone therapy has also been shown to reduce Although the mechanisms are poorly understood, the improvement in exercise capacity was shown to be positively correlated with the increase in serum testosterone level and was accompanied by a small increase in internal left ventricular length (2).As testosterone is a vasodilator, this could explain its anti-ischemic effects on cardiac function during exercise.However, it is currently unknown whether the fatigability of skeletal muscle in a similar fashion.Therefore, adjunctive testosterone therapy might augment the positive effects of exercise rehabilitation on these clinical outcomes in hypogonadal males with stable CHF.
Currently, the main drawbacks in the design of clinical protocols and the inclusion of patients for the study of hormonal alteration lie in establishing baseline T cut off level.Although some limitations of the current study have been mentioned by the authors, the important feature of this study is that it addresses one timely and important issue, which is the correlation between serum testosterone levels and cardiac function, hemodynamic, and exercise capacity in HF patients with ED. but also to the cardiologist and general physicians.However, 61% of men included in this study were taking spironolactone, which is an anti-androgen Understanding the correlation between TT and cardiac patients with heart failure.
Further studies using cardiac speckle technique and tissue Doppler imaging can provide more accurate assessment of cardiac function and perhaps demonstrate a correlation between TT levels and cardiac function.

Cardiology Consultant National Heart Institute
Cairo, Egypt

Dr. Wael Zohdy
Department of Andrology Cairo University Cairo, Egypt E-mail: wzohdy62@hotmail.com Left ventricle ejection fraction (r =-0.69, p = 0.004) (Figure-1) and right ventricle ejection fraction (r =-0.55, p = 0.03) (Figure-1) showed inverse correlation with testosterone levels.SBP, DBP and heart rate during MCT (maximum cardiopulmonary test) did not show any correlation with testosterone levels or any hormonal serum levels except for DBP during the 6´WT that correlated with testosterone (r =-0.66, p = 0.007) (Figure-2).Left ventricular end diastolic diameter and left ventricular end systolic diameter did not show any correlation with testosterone levels either (Table

2 .
Olsson LG, Swedberg K, Ducharme A, Granger heart failure with and without left ventricular tan in Heart failure-Assessment of Reduction in Mortality and morbidity (CHARM) program.J 3. Harrington D, Anker SD, Chua TP, Webb-Peploe Skeletal muscle function and its relation to exercise tolerance in chronic heart failure.J Am Coll Sexual function in patients with advanced heart 5. R Ferrara, F Mastrorilli, G Pasanisi, S Censi, N tion in chronic heart failure.Eur Heart J. 2002; 4 6.Kontoleon PE, Anastasiou-Nana MI, Papapetrou 9. Rosen RC, Cappelleri JC, Smith MD, Lipsky abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction.Int J Impot Res.1999; 10.Bocchi EA, Guimarães G, Mocelin A, Bacal exercise, neurohormonal activation, and erectile blind, placebo-controlled, randomized study followed by a prospective treatment for erectile and cardiovascular disease.Endocr Rev. 2003; 12. Moriyama Y, Yasue H, Yoshimura M, Mizuno Y, of dehydroepiandrosterone sulfate are decreased in patients with chronic heart failure in proportion to the severity.J Clin Endocrinol Metab.2000; 13.Woolf PD, Hamill RW, McDonald JV, Lee LA, nadism caused by critical illness.J Clin Endo-14.Jeppesen LL, Jørgensen HS, Nakayama H, Raastestosterone in men with acute ischemic stroke.54. in low-output cardiac failure.Clin Endocrinol 16. Sullivan ML, Martinez CM, Gennis P, Gallagher 17. Karila TA, Karjalainen JE, Mäntysaari MJ, Viisteroids produce dose-dependant increase in left ventricular mass in power atheletes, and this effect is potentiated by concomitant use of growth 18.Malkin CJ, Pugh PJ, West JN, van Beek EJ, Jones blind randomized placebo controlled trial.Eur Testosterone administration increases insulinlike growth factor-I levels in normal men.J Clin 21.Bocchi E, Moura L, Guimarães G, Conceição doses of growth hormone in the introduction and optimization of medical treatment in decompensated congestive heart failure.Int J Cardiol.2006; Why do patients with heart failure suffer from gestions on how to approach this problem.Int J 23. Maguire SM, Nugent AG, McGurk C, Johnston in human chronic cardiac failure are both endothelium dependent and endothelium independent.Sexual function in patients with advanced heart 25.Tamaki T, Uchiyama S, Uchiyama Y, Akatsuka A, exercise tolerance.Am J Physiol Endocrinol beta-estradiol, progesterone, and testosterone on Ca2+ entry mechanisms of coronary vasoconstric-1034-40. namic effects of testosterone in men with chronic of vascular function in hypertension.Hyperten-Accepted after revision: April 20, 2008 Correspondence address: Dr. Vitor Oliveira Carvalho Laboratório de ICC-Transplante INCOR, HCFMUSP Av.Dr. Enéas de Carvalho Aguiar 05403-000, Sao Paulo, SP, Brazil The cardiac function, left ventricular diameters, and hormonal values of patients are reported in Table-2.Table-3 shows the hemodynamic and exercise variables.

Table 1 -
Characteristics of patients with erectile dysfunction.

Characteristics of Heart Failure Patients N of Patients (%) or Value
Values are mean ± SD or n (%); ACE = angiotensin-converting enzyme inhibitor, CHF-symptoms = congestive heart failure; ED = erectile dysfunction. 4) Values are mean ± SD or n (%); LV= left ventricular; RV= right ventricular.

Table 2 -
Cardiac function, left ventricular diameters and hormonal values in patients with erectile dysfunction.

Table 3 -
Hemodynamic and exercise variables.
2 on 6m cardiopulmonary walking test; VO 2 6min = Oxygen consumption at six minutes of 6m cardiopulmonary walking test; SlopeMax = Slope VE/VCO 2 at maximum cardiopulmonary test; VO 2 Max = VO 2 peak at maximum cardiopulmonary test; Time Max = Total time at maximum cardiopulmonary test.

Table 4 -
Linear regression results between total testosterone serum levels and cardiac function, and exercise variables.