Indications and early results of sildenafil (Viagra) in erectile dysfunction

ADULT UROLOGY

INDICATIONS AND EARLY RESULTS OF SILDENAFIL (VIAGRA) IN ERECTILE DYSFUNCTION RONALD VIRAG

ABSTRACT Objectives. To assess the acceptability, feasibility, and early results of sildenafil (Viagra) in a nonselected cohort of 316 patients (median age 58 years) who sought treatment for male sexual dysfunction during a 10-week period. Methods. Erectile status and activity were evaluated by questionnaire; erectile function was assessed by pharmacologic testing and visual sexual stimulation. Cardiovascular contraindications were assessed. Patients selected for the trial received treatment for 2 months. Results based on the possibility of penetration and individual satisfaction (scale from 0 to 10) were classified as good, fair, or bad. Multifactorial analysis was performed to define factors influencing the response to sildenafil. Results. Twenty-five percent of the patients from the initial cohort refused or did not meet the criteria for oral treatment; 25% of the remaining had a cardiovascular contraindication. At the end of the trial, 157 patients (88.7%) had completed the study; the efficacy of and satisfaction with sildenafil were considered good for 50 (31.84%), fair for 46 (29.29%), and bad for 61 (38.85%). Spontaneous nocturnal erections, organic etiologies, especially cavernovenous impotence, and previous treatment with self-intracavernous injections were significant factors influencing responses to oral treatment. Finally, 32% of the patients after completing the trial (17.2% of the initial cohort) were using sildenafil as their sole treatment, 34% chose self-intracavernous injections, and 25% decided to alternate between oral and local therapy. Conclusions. In the present study, sildenafil had a 60% efficacy rate and was chosen as the sole treatment by only 30% of the patients tested. We propose pretreatment tests to help to predict the response to this medication. UROLOGY 54: 1073–1077, 1999. © 1999, Elsevier Science Inc.

L

ocal treatments using intracavernous or intraurethral medications or vascular or implant surgery have been used with success to treat erectile dysfunction (ED) for more than 15 years.1– 4 Their acceptability and efficacy varies from 25% to 80%. Thus, the possibility of a successful oral agent has provoked wide expectations. Sildenafil is a selective type 5 phosphodiesterase inhibitor that helps erections occur when the subject is sexually stimulated. Much of the available data on efficacy for Viagra have been collected under studies funded by the manufacturer.5,6 They lack data on the etiology and severity of the erectile impairment. Thus, little is known of the efficacy of the medication in nonselected patients. Moreover, exFrom the Centre d’Exploration et Traitement de l’Impuissance, Paris, France Reprint requests: Ronald Virag, M.D., Centre d’Exploration et Traitement de l’Impuissance, 108 Avenue du Maine, 75014 Paris, France Submitted: April 5, 1999, accepted (with revisions): June 15, 1999 © 1999, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED

tensive use of this therapeutic agent is challenged by the questions raised about its potential cardiovascular risks.7–9 The present study was designed to identify the acceptance, feasibility indications, and efficacy of sildenafil in 316 consecutive male patients seeking treatment at our institution between September 15, 1998 and December 30, 1998. MATERIAL AND METHODS The overall population (n ⫽ 316) seeking treatment for male sexual dysfunction between September and December 1998 were included in the study. Initially, 120 (37.97%) had undergone no treatment, 26 (8.22%) were unsatisfied with previous treatments (15 had had an unspecified oral therapy, 4 had already tried sildenafil, 1 had had a penile implant, 4 had had vascular surgery, and 2 had been practicing sexual therapy), and 170 (53.79%) were treated successfully by self-intracavernous injections (SICI). All patients were informed about sildenafil. The medical history was recorded, especially cardiovascular risk factors (CVRFs), cardiovascular history, and other diseases with the potential to cause ED. Pretreatment spontaneous erectile activity (ie, morning erections and soli0090-4295/99/$20.00 PII S0090-4295(99)00310-6 1073

TABLE I. Etiology, contributing factors, and cardiovascular diseases Condition or Disease Coronary heart disease (%) Other cardiac conditions (%) Cardiomyopathy (n) Wolf-Parkinson-White (n) Arrhythmia (n) Arteritis of lower limbs (%) Diabetes (%) Hypertension* (%) Smoking† (%) High cholesterol‡ (%) Pelvic cancer (%) Prostate (n) Bladder (n) Rectum (n) Neurologic disorder (%)

Overall (n ⴝ 316) 8.2 1.9 1 1 4 0.6 4.4 16.4 9.8 10.4 4.5 12 1 1 4.7

Tested (n ⴝ 157)

Good Result (%)

2.54 1.8

25 33

0.8 1.9 15.3 9.5 11 5.1

100 33 17.3 26.7 30 0

4.5

0

* Thirty patients were treated with one drug, 17 with two, and 4 with three. † Heavy smoking when 20 cigarettes per day or more. ‡ High-density lipoprotein ⬍35 mg/mL.

tary masturbation) was assessed using the erectile activity index (IAE),10 a quantified data assessment from 0 to 10 (see Appendix), for information on baseline erectile status. An echo-Doppler ultrasound of the penis after pharmacologic and visual sexual stimulation (VSS) was performed on 275 patients (87.02%).11 Responses were classified into three categories: fully rigid (erection lasting more than 10 minutes after stopping the VSS); unstable erection (rigidity lasting less than 5 minutes); and tumescence without rigidity. Candidates for sildenafil were divided into four groups: absolute contraindication (ie, coronary heart disease or potential use of nitrates), relative contraindication (ie, patients with cardiac arrhythmia or hypertension taking two or more antihypertensive drugs), potential contraindication (ie, patients with penile arterial disease and/or two or more CVRFs), and no contraindications. When prescribed, four tablets of 50 mg sildenafil were dispensed, with the advice to take the pill on an empty stomach and without alcohol. If two trials with the initial dose were ineffective, patients were allowed to try 100 mg. Patients were reassessed after 8 weeks. Results were evaluated on the basis of the possibility of having intercourse and by recording satisfaction on an analogue scale from 0 to 10 (sexual satisfaction index [ISS]).10 Responses were received in writing and/or verbally during telephone interviews or face to face. They were classified as good (when penetration was possible and the ISS was 7 or greater), fair (when penetration was possible and the ISS was between 4 and 6), and bad (when no penetration was achieved and/or the ISS was less than 4). Side effects were also reported. According to these results and patient preference, a routine treatment was established. Age, previous treatment, erectile status quantified by the IAE, intracavernous drug use, and global and specific responses to the pharmacologic test were analyzed to compare sildenafil responders and nonresponders. For 113 of the 157 patients treated, the capacity to maintain an erection was evaluated according to the results of pharmacologic testing and duplex ultrasound scanning.11,12 The results were classified as 0 (no cavernovenous leak), 1 (medium leak), and 2 (severe and major leak), and the results were compared with the responses to sildenafil. 1074

AGE AND ORIGIN

The mean age of the patients was 56.48 years (SD ⫽ 12.57). The median age was 58 years (range 22 to 85). The patients using SICI (age 58.83 ⫾ 11.68) were significantly older than the previously untreated patients (age 53.50 ⫾ 12) (P ⬍0.001). Except for 1.9% who were African, the population was white (98.1%).

STATISTICAL ANALYSIS Data were analyzed with the Statistica software using the t test, chi-square test, analysis of variance, and Whitney-McMann test.

RESULTS CLINICAL STATUS AND CVRFS One hundred fifty-six patients (49.2%) were free of CVRFs and any other general disease. Table I shows the incidence (50.8%) of the main diseases and CVRFs in the entire study population and how they interfered in the population treated with sildenafil (57%). Of the 275 patients who underwent the pharmacologic testing and VSS, the responses were as follows: fully rigid in 80%, rigid but nonlasting in 12.72%, and tumescent in 7.27%. CHOICE AND INDICATION FOR TREATMENT Of the original 316 patients, 26 (8.22%) with nonerectile sexual dysfunction were excluded. Of the 290 remaining, 52 (16.45%) refused to use sildenafil, giving an acceptability rate (ie, number of patients accepting of the total number meeting the indications for sildenafil treatment) of 82.07%. Of the 238 patients (75.31% of the initial cohort) agreeing to participate in this trial, 10% (n ⫽ 24) had an absolute and 11.8% (n ⫽ 28) a relative UROLOGY 54 (6), 1999

TABLE II. Adverse events related to sildenafil (n ⴝ 157) Adverse Event Flush Headache Dizziness Dyspepsia Blue vision Nasal obstruction Decrease of efficacy*

No. of Patients (%) 14 13 14 1 5 1 3

(9) (8) (9) (0.64) (3.18) (0.64) (1.91)

* Two patients already treated with sildenafil when they entered the study; the third one had used the medication more than 10 times during the study.

contraindication. Alternative treatments were proposed to them. In addition, 35 patients (14.70%) who had a potential contraindication were referred to a cardiologist for evaluation, including 17 who underwent a stress test, before prescription of sildenafil. After this investigation, 26 of the 35 patients were included in the study, 3 underwent coronary surgery, and 6 refused treatment. One hundred seventy-seven patients (56% of the initial cohort) were selected, giving a feasibility rate (patients medically suitable for treatment and/or patients volunteering for treatment) of 74.36%. Patients never treated had all treatment options, including sildenafil, fully explained. Patients already using SICI were asked if they wanted to switch to sildenafil. Both groups were informed of the safety profile and possible side effects. SILDENAFIL RESULTS Post-treatment data were available for 157 patients (83.51% of those accepted for the trial); 8 (5.09%) were lost to follow-up and 12 (7.64%) had not started the treatment. Penetration during sexual intercourse was achieved in at least half of the attempts for 73.2% of the patients. The results were considered good for 50 (31.84%), fair for 46 (29.29%), and bad for 61 (38.85%). The final dosage was 25 mg for 1.91%, 50 mg for 28.02%, 75 mg for 0.63%, and 100 mg for 69.42%. Table II indicates the adverse events observed in 42 patients (26.75%). Three patients (1.92%) experienced severe hypotension and/or flush while taking the medication and stopped the treatment. Patient age and previous treatment did not influence the results with sildenafil use. The IAE was significantly higher for patients with good results (5.18 versus 4.19 for patients with bad results; P ⫽ 0.047). Considering the results of pharmacologic testing, bad results (77%) were significantly higher in the groups of patients who reached incomplete or nonlasting erections than in the group of patients with fully rigid erections (33.64%). The mean values of the cavernovenous leak index revealed a signifiUROLOGY 54 (6), 1999

cant increase (P ⫽ 0.0011) in the good (0.18), fair (0.52), and bad (1.2) groups of responses to sildenafil. In the major leak group (n ⫽ 24), no patient had a good result and 83% reported a bad response. The frequency of the contributing factors was much higher for the groups with fair and bad results than for the group with good results. None of the 8 patients who had undergone previous radical prostatectomy (n ⫽ 7) or rectal amputation (n ⫽ 1) had a good or fair result. FINAL TREATMENT CHOICE In the group of patients who tried sildenafil, 32% chose to use it exclusively, 34% chose SICI, and 25% chose to alternate both treatments (two require both oral and local treatment in association). The proportion of patients choosing sildenafil was significantly different for patients who had never been treated (42.6%) than for those who had been using SICI (18%). Considering the overall 290 patients initially accepted into the trial, 50 (17.24%) were using sildenafil exclusively and 155 (53.44%) SICI. For patients who had never had any treatment (n ⫽ 120), 32 (26.66%) were using sildenafil alone, 46 (38.33%) SICI, and 9 (7.5%) both treatments. COMMENT When designing this study I had two main concerns: how safety would affect prescription and whether there were elements to predict the efficacy of sildenafil. The answer to these two questions would help decide how appropriate it is to prescribe sildenafil as a first line treatment. The data of the study clearly demonstrate that safety precautions will eliminate as many as 25% of the potential users. Clinical (IAE) and hemodynamic (venous leak) studies may select those patients unsuitable for oral treatment. In addition, most patients satisfied with SICI return to their former treatment after testing sildenafil. Also, the cardiovascular risks of sildenafil are not yet clearly evaluated. Although the potential effects of the type 5 phosphodiesterase inhibitors are as yet poorly understood, they seem real.7,8 Finally, in addition to the known cardiovascular risks of coitus,13 it has been shown that 16% of the impotent patients presenting with an arterial etiology for their ED are found to have silent coronary heart disease.14 Thus, we have decided to follow the recommendations of both the American College of Cardiology and the American Heart Association (ACC/AHA expert consensus document),15 and add to them the requirement of a cardiac stress test in patients free of cardiac disease but presenting with arterial insufficiency impeding their erection. Observing these modalities, 25% of the patients volunteering for the treatment were 1075

excluded from the trial, and SICI (which does not carry the same risk) was suggested to them. In addition, the use of SICI in patients with coronary heart disease allows the use of nitrates to prevent or treat any angina during coitus. As far as efficacy is concerned, this study revealed two apparently contradictory features: two thirds of the patients treated reported successful intercourse with the oral medication, but only 32% of previously untreated patients chose it as their sole treatment and only 18% of patients who had previously been using SICI chose it. Most patients observed easier, longer, and stiffer erections with SICI. Local treatments using papaverine or prostaglandin E1 with or without alpha-blocking agents intervene mainly by enhancing the cyclic adenosine monophosphate route to produce an erection without interfering with sexual and neural stimulation; sildenafil requires the local production of nitric oxide resulting from sexual stimulation to interfere in the cyclic guanosine monophosphate route. The latter is more natural but less effective in the case of decreased desire and high anxiety; the former is less natural and more effective in all circumstances causing ED. Patients suffering from premature ejaculation, severe performance anxiety, and difficulties in maintaining an erection are often reported to respond poorly to sildenafil, which seems unable to sustain a firm erection for long in such cases. It can be hypothesized that an excess of circulating adrenaline might impede or decrease the proper action of sildenafil. The results of the present study clearly demonstrate that the quality of the responses to sildenafil is linked to the capacity of the penis to sustain an erection provoked in the laboratory by pharmacologic and sexual stimulation. In the group with good results, 80% had no cavernous leak; in the group with bad results, 80% had a severe or major cavernovenous leak. Thus, responses to pharmacologic testing with VSS, which provides comprehensive data to evaluate the level of erectile impairment,11 might be proposed as a criterion to select patients with a better chance of responding to sildenafil. Previous studies relating the results of sildenafil lack this information. Also, although the efficacy of sildenafil has been demonstrated versus placebo, the severity of the ED in those studies is generally unknown.5,6 In one of the prominent studies, only 20% of the patients had undergone a pharmacologic test or nocturnal penile tumescence study.6 Currently, classification of organic or psychogenic impotence without a comprehensive evaluation of the contributing factors is mostly artificial because ED is a symptom almost always caused by both components. It is obvious that the response to any treatment is more dependent on the quality of the erection than on the proper etiology of ED. I de1076

cided to evaluate the IAE, rather than the International Index for Erectile Function (IIEF) (which is unable to grade the severity of ED), as an easy-touse clinical test that might discriminate responders from nonresponders. The higher the score was on the IAE, the better was the response to sildenafil. In conjunction with the pharmacologic testing with VSS (only 10.5% of the patients who did not achieve a full rigid lasting erection reported good results) and the evaluation of the cavernovenous leak index (no good results reported when the score was greater than 1), the IAE, when less than 4, was highly predictive of a sildenafil failure. These results reflect the activity of a highly specialized center. Nevertheless, they challenge the idea that any initial treatment of ED should start with a prescription of sildenafil. The present study suggests that a careful evaluation with the use of indexes such as the IAE and pharmacologic testing with VSS matched with duplex scan echocardiography is mandatory to identify respective responders to oral and local therapy. CONCLUSIONS After a 2-month trial with sildenafil, 42.6% of previously untreated treated patients and 18% of the patients already treated with SICI chose the oral drug as their sole treatment. The overall satisfaction rate with sildenafil shown in this study is lower than in previously published studies. This raises the question of when it is relevant to prescribe sildenafil as the first line treatment for ED. I propose that it will be better to determine potential users by using the IAE and pharmacologic testing with VSS. REFERENCES 1. Bennett AH: Impotence: Diagnosis and Management of Erectile Dysfunction. Philadelphia, WB Saunders, 1994. 2. NIH Consensus Panel Development on Impotence. JAMA 270: 83–90, 1993. 3. Feldman HA, Goldstein I, Hatzichristou DG, et al: Impotence and its medical and psychological correlates: results of the Massachusetts Male Ageing Study. J Urol 151: 54 – 61, 1994. 4. Virag R: Intracavernous injection of papaverine for erectile failure. Lancet 2: 938, 1982. 5. Boolell M, Allen MJ, Allard SA, et al: Sildenafil: an orally active type 5 cyclic GMP-specific phosphodiesterase inhibitor for the treatment of penile erectile dysfunction. Int J Impot Res 8: 47–52, 1996. 6. Goldstein I, Lue TF, Padma-Nathan H, et al: Oral sildenafil in the treatment of erectile dysfunction. N Engl J Med 338: 1399 –1404, 1998. 7. Budenholzer B, Marshal KG, Meikl AW, et al: Sildenafil in the treatment of erectile dysfunction. N Engl J Med 339: 700 –701, 1998. 8. Feenstra J, van Drie-Pierik FJ, Lacle´ CF, et al: Acute myocardial infarction associated with sildenafil. Lancet 19: 957–958, 1998. 9. Federal Drug Administration: Reports of the Side EfUROLOGY 54 (6), 1999

fects of Sildenafil. Available at: http://www.fda.gov. Accessed August, 1998. 10. Virag R, and Beck-Ardilly L: Nosology, epidemiology, and clinical quantification of erectile dysfunctions (in French). Rev Med Interne 18(suppl 1): 10s–13s, 1997. 11. Sussman H, and Virag R: Vascular and functional evaluation of erectile dysfunctions using erection provocation tests (in French). Rev Med Interne 18(suppl 1): 17s–21s, 1997. 12. Virag R, and Sussman H: Exploration of the deep dorsal vein of the penis using pulsed Doppler ultrasonography. Preliminary study (in French). J Mal Vasc 23: 195–198, 1998. 13. Muller J: Cardiac risk of coitus. JAMA 275: 1045–1047, 1997. 14. Andersson M, Nicholson B, Lovie E, et al: An analysis of vasculogenic erectile dysfunction as a predictor of occult cardiac disease (abstract). J Urol 159(suppl 5): 30, 1998. 15. Cheitlin M, Hutter AM Jr, Brindis RG, et al, for the American College of Cardiology/American Heart Association: ACC/AHA expert consensus document. Use of sildenafil (Viagra) in patients with cardiovascular disease. J Am Coll Cardiol 33: 273–282, 1999.

UROLOGY 54 (6), 1999

APPENDIX: ERECTILE ACTIVITY INDEX The IAE measures the frequency, quality, and duration of erections independent of the sexual activity of the couple. It is calculated by determining (a) the frequency of nocturnal and/or morning erections, with a score of 3 if the erections occur more than twice a week, 2 if they occur once to twice a week, 1 if they occur less than once a week, and 0 if they never occur; (b) the quality of the erection in the morning (b1) and/or at solitary masturbation (b2) by asking the patient to record on a drawing the position and strength of his penis when erect (see example below); and (c) the duration of the morning erection (if well sustained add 1 to the score, if it falls quickly substract 1). Thus, IAE ⫽ a ⫹ (b1 ⫹ b2) ⫹ c. If there is no masturbation, record 2b1. The normal values are 8.96 ⫾ 0.66.10

1077