Author + information
- Received May 15, 2002
- Revision received July 12, 2002
- Accepted July 24, 2002
- Published online December 4, 2002.
- Udho Thadani, MD, FACC*,* (, )
- William Smith, MD, FACC†,
- Stephen Nash, MD, FACC‡,
- Neville Bittar, MD, FACC§,
- Stephen Glasser, MD, FACC∥,
- Puneet Narayan, MD, FACC¶,
- Richard A Stein, MD, FACC#,
- Sharon Larkin, RN**,
- Arthur Mazzu, PhD**,
- Robert Tota, MD, FACC††,
- Kenneth Pomerantz, PhD‡‡ and
- Pavur Sundaresan, MD, PhD**
- ↵*Reprint requests and correspondence:
Dr. Udho Thadani, OUHSC, Cardiology, 920 SL Young, WP 3120, Oklahoma City, OK 73104, USA.
Objectives The effect of vardenafil, a potent and highly selective phosphodiesterase-5 (PDE5) inhibitor, on symptom-limited exercise time, time to first awareness of angina, and time to ischemic threshold (ST-segment depression ≥1 mm from baseline) during exercise tolerance testing (ETT) was examined in patients with stable coronary artery disease (CAD).
Background Erectile dysfunction (ED) is common among men with CAD. PDE5 inhibition is increasingly the preferred treatment option for ED. However, the effect of PDE5 inhibition on exercise-induced ischemia in CAD patients has received limited prospective evaluation.
Methods In this double-blind, crossover, single-dose multicenter study, 41 men with reproducible stable exertional angina due to ischemic CAD received vardenafil 10 mg or placebo, followed by ETT (5 to 10 metabolic equivalents [METS], Bruce protocol) 1 h postdose. Sublingual nitrate use was prohibited for ≥24 h pre- and postexercise study days. End points included symptom-limited treadmill exercise time, time to first awareness of angina, time to ischemic threshold, and safety.
Results Relative to placebo, vardenafil 10 mg did not alter exercise treadmill time (427 ± 105 s vs. 433 ± 109 s, p = 0.39), or time to first awareness of angina (292 ± 110 s vs. 291 ± 123 s, p = 0.59), but significantly prolonged time to ischemic threshold (334 ± 108 s vs. 381 ± 108, p = 0.0004). At peak exercise, vardenafil 10 mg did not alter blood pressure, heart rate, or rate-pressure product relative to placebo. The most common adverse events (facial flushing and headache) were of mild or moderate intensity, and short-lived.
Conclusions Vardenafil 10 mg did not impair the ability of patients with stable CAD to exercise at levels equivalent or greater than that attained during sexual intercourse (average of 2.5 to 3.3 METS).
Erectile dysfunction (ED) is often observed in men with cardiovascular disease, and some investigators have suggested that ED may serve as a predictor for this condition (1). In epidemiologic studies, hypertension, ischemic heart disease, and peripheral vascular disease are frequently associated with ED (2,3). The association between ED and coronary artery disease (CAD) is supported by results from several studies that correlate CAD with difficulty in achieving erections (4–9). In some, the onset of ED predated the symptoms and diagnosis of CAD (9).
Sildenafil was the first oral phosphodiesterase-5 (PDE5) inhibitor for the treatment of men with ED. However, significant concerns about the possible systemic hemodynamic effects of this drug, particularly in combination with other vasoactive cardiovascular medications, prompted the American College of Cardiology/American Heart Association (ACC/AHA) to provide guidelines regarding its use, especially in patients with ischemic heart disease (10).
Vardenafil is a highly selective PDE5 inhibitor (11,12). The efficacy and safety of vardenafil (dosed at 5, 10, or 20 mg) has been demonstrated in Phase II (13)and Phase III trials involving men with ED of various etiologies (14), including type 2 diabetes mellitus (15)and postprostatectomy (16). The present study examines the effect of vardenafil on symptom-limited treadmill exercise time, first awareness of development of angina, time to ischemic threshold (time to ST-segment depression ≥1 mm change from baseline) during treadmill exercise using the Bruce protocol, and effects on heart rate and blood pressure at rest and during exercise in patients with stable CAD.
The trial was performed using a randomized, double-blind, placebo-controlled, crossover, single-dose multicenter study design (Fig. 1). Following a placebo run-in and screening, patients with reproducible exertional angina or angina-equivalent performed an exercise tolerance test (ETT) after administration of 10 mg vardenafil or placebo. Following a 5- to 21-day washout period, patients were crossed over to the alternate treatment. The study protocol and amendments were prepared in accordance with the Declaration of Helsinki and with the Investigational New Drug and Bioresearch Regulations of the United States Food and Drug Administration. The study protocol and amendments were approved by the Institutional Review Board/Independent Ethics Committee at each investigator’s institution, and written informed consent was obtained from all patients.
The primary objective of this study was to determine the influence of vardenafil compared to placebo on total exercise time in patients with exertional angina of moderate severity and/or cardiac ischemia during exercise (termination of exercise due to angina or appearance of ST-segment depression ≥1 mm change from baseline on electrocardiogram [ECG]) between 3 to 10 min of exercise. Secondary measures included the time to first awareness of angina, and time to ischemic threshold (time to onset of ST-segment depression change from baseline of ≥1 mm).
Men between 40 and 80 years of age with stable CAD were eligible for inclusion in the study. Erectile dysfunction was not a specified inclusion or exclusion criteria for study eligibility. Subjects were excluded if they had evidence of major organ system failure that might impair their ability to complete the study; body mass index >35; vasospastic or unstable angina; resting 12-lead ST-segment depression >1.0 mm; clinically relevant chronotropic incompetence; symptomatic postural hypotension (supine systolic <90 mm Hg); sinus bradycardia (<45 beats/min) or resting heart rate >110 beats/min; significant exercise-induced hypotension (in the investigator’s opinion from prior history); ST-segment elevations on treadmill exercise testing in non–Q-wave leads; New York Heart Association functional class III or IV congestive heart failure; significant pulmonary disease; significant cardiac rhythm disorder; severe aortic stenosis; left bundle branch block; significant left main disease (>50% occlusion) that had not been surgically corrected; significant peripheral vascular disease; malignancy other than squamous or basal cell carcinoma in situ; acute or significant psychiatric or mental illness; epilepsy or adult seizure disorders; hypersensitivity to PDE5 inhibitors; or allergy to ≥2 drug classes.
Also excluded were patients unable to discontinue sublingual nitrates 24 h before and after an ETT and those unable to discontinue slow-release and or long-acting nitrates one week before visit 1 and throughout the remainder of the study. Concurrent use of corticosteroids, azole antimycotics (except fluconazole), macrolides (except azithromycin), nefazodone, rifampin, androgens, immunosuppressants, and any drug known to interfere with cytochrome P-450 was grounds for exclusion as was concurrent sildenafil use within five days of any ETT visit, or consumption of grapefruit juice or products containing grapefruit juice within 24 h before each dosing. Aspartate aminotransferase, alanine aminotransferase, or creatine kinase >3 times the upper limit of normal, hematocrit <32% at screening and optional follow-up, serum creatinine >2.0 mg/dl at screening, and any other laboratory abnormalities judged to be clinically significant were grounds for exclusion, in addition to a history of drug or alcohol abuse, or blood donation or investigational drug use within 30 days of visit 1. Patients were screened by complete physical examination, medical history, chest X-ray, ECG, and laboratory tests up to three weeks before study commencement. Patients were required to discontinue slow-release and long-acting nitrates one week before visit 1 and could not resume taking them until after study completion. Patients must have been withdrawn from sublingual nitrates 24 h before ETT and could not resume taking them for 24 h after completing the test. Patients taking beta-blockers, diltiazem, and/or verapamil postponed their dose of these medications on the mornings of each ETT until after it was completed. The continued use of aspirin, dihydropyridine calcium channel blockers, ACE inhibitors, and diuretics was allowed.
Exercise tolerance testing
The ETTs were conducted 1 h after study drug administration in the morning, using the standard Bruce protocol. Blood pressure and heart rate were recorded before ETTs (supine after 5 min, sitting after 2 min, and standing after 2 min at baseline), at the last minute of exercise during each Bruce Stage (standing), when and if angina occurred, at termination of exercise, and every 2 min postexercise (sitting) until patients were stable. At completion, patients rated their perceived degree of exertion on a modified Borg scale (0 to 10) (17). The ECGs were monitored continuously during the test by the investigator through at least three leads (two chest and one limb lead). A 12-lead ECG was recorded every minute during the test by the investigator. The ECGs were examined continuously during the test for ST-segment changes. If present, the time to ST-segment depression ≥1 mm change from baseline, the maximum ST-segment depression from baseline, and the time to resolution of significant ST-segment change were all recorded.
Patients were qualified for the study based on the results of two ETTs separated by 5 to 21 days (visits 1 and 2). At both visits, patients had to be able to exercise for ≥3 min using a standard Bruce ETT protocol. They must also have terminated the ETT within 10 min because of symptoms of angina pectoris or angina equivalents (intensity of at least 7 on a scale of 1 to 10) and/or significant ST-segment depression, defined as a ≥1-mm horizontal or downsloping change from baseline or ≥2-mm slowly upsloping change from baseline. In addition, symptom-limited total exercise time on treadmill at visits 1 and 2 could not differ from each other by more than 15%. If the first two ETTs did not meet this criterion, patients were allowed to return one week later for a third qualifying test. Results of this test were required to be within 15% of those for the second test in order for the patient to enter double-blind treatment. If the patient terminated exercise because of anginal equivalent of shortness of breath, there must have been additional evidence of myocardial ischemia (i.e., positive prior imaging study with similar symptoms).
During the double-blind phase of the study, patients meeting the above reproducibility criteria received either vardenafil 10 mg orally or placebo 1 h before symptom-limited ETT. Each treatment was administered after an overnight fast. After a 5- to 21-day washout period, patients were crossed over to the alternate treatment.
Blood samples were drawn ∼90 min after dosing for determination of plasma vardenafil concentrations. The vardenafil assay used reverse-phase high performance liquid chromatography with mass spectroscopy/mass spectroscopy detection (18). Calibration curves ranged from 0.5 to 50 μg/l for vardenafil. Precision ranged from 2.06% to 5.87%, and accuracy ranged from 89.1% to 102.8%.
All patients who received vardenafil were evaluated for safety. Observations pertinent to safety included results from clinical laboratory tests, physical examinations, vital signs, ECGs, and adverse events.
The natural logarithm of the time to termination of exercise was analyzed using analysis of variance with terms for sequence, patient within sequence, period, and treatment. Because the study population was small, logarithmic transformation of the data was performed consistent with standard biometric protocols and the prospectively designed statistical plan. A one-sided test procedure with noninferiority limit of 0.85 and significance level of 5% was used to determine noninferiority. A one-sided test procedure was chosen because the measure of interest was the potential for vardenafil to impair the ability of patients to endure exercise. The influence of vardenafil versus placebo was considered to be noninferior if least square (LS) mean differences (LS, means corrected for sequence, patient within sequence, period, and treatment) were <15%. Retrospective analysis examined the influence of vardenafil in the subpopulation of patients who exhibited both angina and ST-segment changes ≥1 mm from baseline. Retrospective analysis was also performed to determine the influence of vardenafil, relative to placebo, on heart rate, systolic and diastolic pressures, and the heart rate-pressure product measured at baseline (1 h postdose) and at peak exercise. Frequencies of adverse events, summary statistics for vital signs and ECGs, and changes from baseline in vital signs and ECG were determined for each treatment period.
A total of 53 patients were screened, and 41 were randomized to double-blind treatment. Of the 12 patients not randomized, 5 failed to meet inclusion criteria (ETT requirements), 3 failed to adhere to the protocol screening timeline, and 1 each violated protocols due to concomitant medication, laboratory abnormality, work-schedule conflict, or a positive test for illicit drug use. All 41 randomized patients completed the study. Of these, 39 patients were considered valid per protocol. One patient was excluded from analysis due to disallowed concomitant medication usage (erythromycin). Another patient was excluded from analysis as a consequence of inadvertently exceeding the maximal treadmill time. Most of the patients completing double-blind treatment had ≥1 cardiovascular risk factor and were taking one or more cardiovascular medications (Table 1). In 17% of patients, a diagnosis of ED was present. Overall, the patients who entered double-blind treatment were representative of a population with well-controlled cardiac disease.
Effects of treatment on exercise tolerance testing
Relative to placebo, vardenafil 10 mg did not significantly alter mean total exercise treadmill time (427 ± 105 s for placebo vs. 433 ± 109 s for vardenafil, mean ± SD, p = 0.394) or time to first awareness of angina pectoris (292 ± 110 s for placebo vs. 291 ± 123 s for vardenafil, p = 0.594; Fig. 2). However, vardenafil significantly prolonged the time to ST-segment depression ≥1 mm change from baseline (334 ± 108 s for placebo vs. 381 ± 108 s for vardenafil, p = 0.0004; Fig. 2).
One patient was excluded from the analysis after exceeding the defined time interval for total exercise duration in the qualifying ETT (10 min) due to staff error. Following double-blind treatment with vardenafil or placebo, the total exercise duration was 12 min 7 s (727 s) and 11 min 45 s (705 s), respectively. This represented a 3% improvement under active drug conditions, consistent with the overall mean change. The time to first awareness of angina improved by ∼40 s without evidence of ST-segment depression ≥1 mm.
Similar results were obtained in a subgroup of patients (n = 27) who exhibited angina and ST-segment depression ≤1.0 mm (Table 2). Vardenafil 10 mg did not significantly alter mean exercise treadmill time, or time to first awareness of angina pectoris, but significantly prolonged the time to ST-segment depression ≥1 mm change from baseline (331 ± 104 s for placebo vs. 377 ± 109 s for vardenafil vs. p = 0.002). Uncorrected mean data demonstrated similar findings compared to logarithmically transformed data described above.
The concentration of vardenafil in plasma from blood samples taken approximately 90 min after dosing was 7.9 ± 5.1 ng/ml (mean ± SD; range 1.2–23.0 ng/ml), a concentration near maximal plasma concentration of vardenafil observed after a 10-mg dose in previous pharmacokinetic studies (18,19).
There were no treatment-emergent, clinically relevant changes in blood pressure (BP), heart rate (HR), respiration rate, or body temperature. Changes in BP, HR, and rate-pressure product that occurred during exercise testing were consistent with exertion levels achieved during treadmill exercise. At baseline (1 h postdose), resting standing systolic blood pressure (SBP) and diastolic blood pressure (DBP) in the vardenafil 10-mg group was 6 mm Hg and 5 mm Hg less (p < 0.05) compared to placebo. The resting standing HR was also higher in the vardenafil 10-mg group compared to placebo (3 beats/min, p < 0.05), with no clinically relevant change in rate-pressure product. At peak exercise, both SBP and DBP were lower following vardenafil 10 mg by 8 and 7 mm Hg, respectively, compared to placebo (p < 0.05). There was no difference in HR following treatment with vardenafil 10 mg. When corrected for preexercise resting values, no significant differences existed in HR or BP changes with exercise with vardenafil 10 mg relative to placebo. At peak exercise, indirect measure of myocardial oxygen demand (rate-pressure product) was not altered by vardenafil 10 mg relative to placebo (Table 3). All exercise-induced ECG changes noted during testing were consistent with each patient’s underlying CAD.
Overall, vardenafil was well tolerated; 24% (n = 10) of the patients experienced adverse events during treatment with vardenafil versus 2% (n = 1) for placebo. All adverse events are consistent with those that might be expected to occur as a result of vasodilation with PDE5 inhibition (Table 4). Adverse events were typically mild in intensity and were self-limiting. No treatment-emergent, clinically significant abnormalities were identified on physical examination, ECG, or laboratory measurements. No patient was discontinued prematurely because of adverse events. One serious adverse event, sepsis associated with an E. coliinfection, was reported four days after vardenafil treatment. The patient was hospitalized, treated with antibiotics, and made a complete recovery. No deaths occurred during this study.
The results of this study in symptomatic patients with stable CAD who tolerated at least 3 min on ETT using the Bruce protocol showed that, relative to placebo, vardenafil did not alter total treadmill exercise time or time to the onset of angina. Interestingly, vardenafil significantly increased ischemic threshold (as occurred by ST-segment depression) during exercise by approximately 15% relative to placebo. Furthermore, there were no untoward hemodynamic effects of exercise as changes in HR, BP, and rate-pressure product were similar in both placebo- and vardenafil-treated patients. These results support the conclusion that administration of vardenafil 10 mg to patients with stable symptomatic CAD does not exacerbate an ischemic response to exercise relative to placebo at an exertional level of approximately 5 to 10 metabolic equivalents (METS). This level is similar to or greater than energy levels typically required to complete sexual intercourse (average of 2.5 to 3.3 METS, range of 2 to 5.4 METS) (20).
This is the first study to evaluate the effects of vardenafil during exercise in patients with stable CAD with or without angina. In a recent study, single-dose sildenafil did not adversely influence total exercise duration in patients with exercise-induced angina and myocardial ischemia (21). In that study, total exercise duration increased, but ischemic threshold did not change after sildenafil compared to placebo. Similar results were reported in patients with known or probable CAD and ED, where sildenafil altered resting, but not exercise-induced cardiovascular dynamics following supine bicycle exercise (22). Interestingly, the latter study only included patients with ED, compared to the present study, in which 17% of patients volunteered their ED history. Collectively, these findings reinforce the concept that the negligible systemic hemodynamic effect of PDE-5 inhibition in patients with chronic stable angina during exercise is largely unrelated to the specific effects of PDE5 inhibition on regional (corporal cavernosal) blood flow.
The finding that vardenafil increased ischemic threshold implies that patients were able to undergo exercise for a longer period without manifesting electrocardiographic evidence of cardiac ischemia. An increase in ischemic threshold has been reported with beta-blockers and with sublingual nitrate regimens at dosages that do not produce tolerance (23). These antianginal drugs, however, also prolong exercise duration in patients with stable angina. The majority of patients in the present study were taking beta-blockers, although they did not take the beta-blocker on the morning of the exercise test. Whether vardenafil will exert anti-ischemic effects when given concomitantly with beta-blockers before exercise needs further study. Collectively, these findings are consistent with the view that PDE5 inhibitors do not adversely affect performance during stress in patients with stable angina.
The cardiac stress associated with sexual activity has the potential to increase the risk for cardiac events in patients with CAD. However, for most patients, this risk is relatively slight; the relative risk of a myocardial infarction (MI) occurring within 2 h after sexual activity was 2.5 in patients versus other times, and that the relative risks of triggering onset of MI during this period among patients with a history of prior angina or prior MI were 2.1 and 2.9, respectively (24). Thus, sexual activity can trigger the onset of MI, but the absolute risk is low.
Although sporadic reports of cardiac events in patients taking sildenafil have been published (25,26), review of the sildenafil experience supports its cardiovascular safety in a wide range of patients, with no excess in the incidence rates for death or MI among patients receiving sildenafil (27). In addition, during the first 4.9 months of sildenafil use by 5,601 patients with ED, there was no evidence of increased risk for MI or ischemic heart disease (28). Finally, sildenafil caused no treatment-related cardiovascular adverse events other than flushing in patients with both ED and cardiovascular disease (29). Based on these findings, the ACC/AHA Expert Consensus Document recommended that sildenafil be used with caution in certain men at risk for cardiovascular events. More recently, the Princeton guidelines recommended that patients at low risk (including those with mild stable angina) can be considered candidates for sildenafil use, whereas patients at high risk be restricted from its use (30). The ACC/AHA and Princeton recommendations suggest there is a degree of cardiac risk associated with sexual activity and that, before treating ED, physicians should consider patients’ cardiovascular status and/or the impact of their resuming sexual activity and counsel them appropriately (20,30).
In the present study, patients did not take long-acting nitrates throughout the study and they did not take sublingual nitroglycerin 24 h prior to or after vardenafil administration, consistent with the recommendations of the ACC/AHA guidelines against concomitant use of nitrates and sildenafil. Future studies need to be performed to develop guidelines for the judicious use of anti-ischemic medications, including vasodilators, when patients with symptom-limited angina may require additional support after PDE5 use in conjunction with sexual (or physical) activity.
In this limited study of 41 patients, vardenafil 10 mg was well tolerated. Adverse events were either mild or moderate and short-lived. The most common drug-related events (facial flushing and headache) were consistent with the vasodilating properties of the drug class. There were no treatment-emergent, clinically significant changes on physical examination or in laboratory parameters. All changes in the ECGs recorded during exercise testing were consistent with patients’ underlying CAD. Larger studies are needed to verify these findings.
In conclusion, vardenafil 10 mg did not significantly alter patients’ response to exercise testing, and delayed the onset of ST-segment changes relative to placebo in 41 patients with stable CAD with or without exertional angina pectoris. Thus, vardenafil 10 mg did not impair CAD patients’ ability to exercise at a level considered equivalent to or greater than sexual intercourse. When considering vardenafil for the treatment of cardiac patients exhibiting ED, physicians should consider the cardiovascular status of their patients as recommended by the ACC/AHA guidelines (20)(Appendix).
We thank Jonathan Reuning-Scherer, PhD, for performing the statistical analyses.
Investigators and institutions
Udho Thadani, MD, Oklahoma University, HSC, and Veterans Affairs Medical Center, Cardiology Section, Oklahoma City, Oklahoma (Study Coordinator, Michelle Thresher, RN).
William Smith, MD, New Orleans Center for Clinical Research, New Orleans, Louisiana (Study Coordinator, Clarese Noblesse).
Stephen Nash, MD, Syracuse Preventive Cardiology, Syracuse, New York (Study Coordinator, Kristine Westpfal).
Neville Bittar, MD, FACC, Gemini Scientific, Inc., Madison, Wisconsin (Study Coordinator, Margaret Spatola).
Stephen Glasser, MD, University of Minnesota, Minneapolis, Minnesota (Study Coordinator, MaryAnne Knipp, RN).
Puneet Narayan, MD, Veterans Affairs Medical Center, Washington, D.C. (Study Coordinator, Madeline Metcalf, RN).
Richard A. Stein, MD, Department of Cardiology, Brooklyn Hospital Center, Brooklyn, New York.
Bayer Corporation: Arthur Mazzu, PhD; Pavur Sundaresan, MD, PhD, Robert Tota, MD (Clinical Monitors, Sarah Ferry, BSN; Martha Gallentine, Cindy Lane, RN, MS; Sharon Larkin, BSN, RN; Teresa Lopes, MS; Charmelle Casella, MS).
☆ This study was funded by the Bayer Corporation, Pharmaceutical Division, West Haven, Connecticut. The University of Oklahoma received funding for conducting the study; Dr. Thadani did not receive any funds. Dr. Thadani acts as a consultant to the Bayer Corporation and other pharmaceutical companies for designing studies in angina, and he has received honoraria for speaking engagements from Bayer and other pharmaceutical companies. Dr. Thadani does not own any pharmaceutical company stock. Dr. Nash has served as consultant for Bayer, Merck, Bristol Myers Squibb, Sankyo, and Koss Pharmaceuticals. He is conducting research with Bristol Myers Squibb, Merck, Pfizer, Astra Zeneca, Aventis, and Glaxo Smith Kline. He reports no stock ownership. S. Larkin and Drs. Mazzu, Tota, Pomerantz, and Sundaresen are Bayer Corporation employees.
The results of this study were presented in part at the 2002 Annual Meeting of the American College of Cardiology.
- coronary artery disease
- diastolic blood pressure
- erectile dysfunction
- exercise tolerance test
- metabolic equivalents
- myocardial infarction
- systolic blood pressure
- Received May 15, 2002.
- Revision received July 12, 2002.
- Accepted July 24, 2002.
- American College of Cardiology Foundation
- Technology and Practice Executive Committee,
- Cheitlin M.D.,
- Hutter A.M. Jr.,
- Brindis R.G.,
- et al.
- Angulo J.,
- Cuevas P.,
- Fernandez A.,
- et al.
- Bischoff E.,
- Pages Palau E.,
- Haning H.,
- Es Sayed M.,
- Schenke T.,
- Schlemmer K.H.
- Hellstrom W.,
- Gittleman M.,
- Shapiro J.,
- Thibonnier M.,
- Segerson T.
- Investigators of the Vardenafil Diabetes Phase III Study,
- Goldstein I.
- Vardenafil PROSPECT Group,
- Brock G.,
- Taylor T.,
- Seger M.
- Sachse R.,
- Rohde G.,
- Stark S.,
- Klotz T.
- Porst H.,
- Steidle C.,
- Hollister A.,
- Schmidt A.C.,
- Segerson T.
- American College of Cardiology/American Heart Association,
- Cheitlin M.D.,
- Hutter A.M. Jr.,
- Brindis R.G.,
- et al.
- Fox K.M.,
- Thadani U.,
- Ma P.T.,
- Nash S.D.
- Shakir S.A.,
- Wilton L.V.,
- Boshier A.,
- Layton D.,
- Heeley E.