Treatment of Erectile Dysfunction with Sildenafil

Leonard S. Marks, M.D, Catherine Duda, Frederick J. Dorey, Ph.D, Maria Luz Macairan, M.D., Paul Bryan Santos

Departments of Urology and Biostatistics, UCLA School of Medicine, Los Angeles, CA and Urological Sciences Research Foundation, Culver City, CA

Urology, 53(1): 19-24, 2001
Click here for the full-text article in PDF Format

Publication costs were supported in part by an unrestricted educational grant from Pfizer, Inc.

Reprint requests and correspondence to Leonard S. Marks, M.D., Urological Sciences Research Foundation, 3831 Hughes Avenue, Suite 501, Culver City, CA 90232. Phone: 310-559-9800; Fax: 310-559-7821; e-mail:

Editor's Note---

This report is noteworthy because:
1. It is the first peer-reviewed test of sildenafil (Viagra) to appear since FDA approval of the drug.
2. It is an independent "real life" evaluation, performed in a clinical practice setting, outside of the Pfizer-sponsored research trials.
3. It shows that severity of the erectile dysfunction can be used as a predictor of response to the drug, ie, the worse the dysfunction, the worse the erectile response to Viagra (Figure 2), and vice versa.
4. It incorporates the response of the partners, which has not been previously reported (Figure 3).
5. It confirms that, in general, Viagra "is an effective treatment of ED (erectile dysfunction)."
6. It is the first online article of our developing website! Alan W. Partin, M.D., Editor


For more information about Urology visit the journal's home page at:


Objectives. To determine the efficacy of sildenafil for the treatment of erectile dysfunction (ED) in a clinical practice setting; to evaluate the correlation between patient and partner perceptions of treatment outcomes; and to assess the relationship between the severity of ED and response to treatment.

Methods. Among the first 100 men to receive sildenafil in a urology practice setting, 74 (mean age = 64+11 years) completed a validated sexual function questionnaire (International Index of Erectile Function, IIEF) before and after a 4 – 6 week treatment period. A modified version of the same questionnaire was independently completed by partners. ED was categorized into a Severity Class of I - IV.

Results. Sildenafil treatment improved erections by 71% - 95%, according to responses in key IIEF questions 3 and 4. Overall, 77% of patients (57/74) desired to continue treatment after the test period. Patient score on the IIEF was correlated with partner score on the modified questionnaire, before and after treatment (r=0.67 – 0.81, p<0.01). IIEF scores were reflected in a simple Severity Classification system. Men with the best preservation of erections (Severity Class I) exhibited the best responses to sildenafil, whereas men with no erections (Severity Class IV) were much less likely to respond to the drug and desire continuation of treatment (p<0.01). Radical prostatectomy patients were relatively refractory to sildenafil, except in two of five who had undergone a nerve-sparing operation.

Conclusion. In clinical practice, sildenafil is an effective treatment of ED, according to partner-validated questionnaire responses; and the results of treatment are predictable with an ED Severity Classification system.


Sildenafil citrate is the first oral agent approved by the U.S. Food and Drug Administration for the treatment of erectile dysfunction (ED). From approval on March 27 through August 7, 1998, more than 3.8 million prescriptions were written for more than 2.5 million men totaling more than 31 million doses (source: Pfizer Marketing Division). The efficacy and safety of sildenafil has been reported in a large-scale, placebo-controlled clinical trial in which selection of study participants was subject to rigorous inclusion and exclusion criteria (1). However, scientific reports of usage outside of clinical trials, i.e., in the general population of men seeking treatment for ED, are not yet available. We therefore conducted a study of sildenafil usage among the first men to seek this treatment following drug approval.


This was a prospective cohort study. The study included the first 100 men who elected to receive sildenafil therapy for the treatment of ED at a metropolitan Los Angeles urology practice, who met the study inclusion criteria, and who agreed to participate in a short-term, open-label, non-comparative trial (4-6 weeks’ duration). Inclusion criteria were: 1) a history of ED, as defined by the NIH Consensus Development Panel on Impotence (2) of at least one years’ duration; 2) at least a 10th grade education; 3) participation in a discussion of therapeutic options, including the known therapeutic effects and side effects of sildenafil as detailed in the package insert; 4) the ability and willingness of patients and wherever possible, partners to complete questionnaires; and 5) a willingness to return for a follow-up visit after four to six weeks of treatment. Exclusion criteria were 1) concomitant use of organic nitrates or nitric oxide donors, 2) alcohol or drug dependence or 3) major, concurrent, uncontrolled medical illness.

Prior to treatment, all study participants underwent a detailed medical exam that included information about the severity of the ED and prior treatments for ED. Prior ED therapies used by these patients included intraurethral alprostadil (45%), intracavernosal injections (39%), and vacuum devices (14%); 35% had received no prior ED treatment. Patients also received a routine physical examination, including urinalysis, complete blood count, and multiphasic serum testing prior to receiving sildenafil.

The present report focuses upon the 74 men who completed questionnaires (see Evaluations, below) both before and after treatment. Of the 74 men with complete questionnaires, partner questionnaires were available before treatment in 34 and after treatment in 31. No attempt was made to interview the partners.


Patients meeting the entry criteria were given a prescription for eight 50-mg tablets of sildenafil (n=47) or received eight 50-mg sample tablets (n=27), if cost were a potential barrier to study participation. Each patient was instructed to take a 50 mg tablet approximately one hour prior to planned sexual activity, and, if displeased with resulting efficacy, to take two tablets at next activity. Patients were instructed to take sildenafil only once within a 24-hour period. The mechanism of action and potential side effects of sildenafil were explained and a package insert was given to each patient. A follow-up visit was scheduled for four to six weeks later. Additional drug was made available to each patient as needed.


Efficacy was assessed using the International Index of Erectile Function (IIEF), a 15-item, self-administered questionnaire (3). The maximum total IIEF score is 75 and the minimum IIEF score is 5, with higher scores indicating greater erectile function. The 15 questions of the IIEF can be grouped into five response domains that address key elements of sexual dysfunction: erectile function, orgasmic function, sexual desire, intercourse satisfaction, and overall satisfaction. Patients completed the IIEF at the start of the study (baseline visit) and again at the follow-up visit. Modified versions of the IIEF questionnaire, along with a stamped, self-addressed envelope, were given to the patient for his partner to complete independently, i.e., without patient consultation, at baseline and at the end of the study. Modifications of the questionnaire for the partner were minor; each of the original 15 questions was simply re-phrased, asking the partner to evaluate the patient for that item, e.g., Question 1, "How often was your partner able to get an erection during sexual activity?".

Patients were queried at follow-up regarding their usage of the drug, side effects, and desire to continue treatment.


Scores for the IIEF before and after treatment were compared using the paired t-test and the non-parametric sign rank test. Multivariate linear regression analysis was used to determine the relationship between the pre-treatment variables and the total outcome and domain scores. Statistical significance was chosen a priori to be p<0.05.


Of the first 100 patients electing to receive sildenafil, 74 completed the IIEF questionnaire both at baseline and at the follow-up visit. A summary of their baseline characteristics is shown in Table 1. The mean age was 64±11 years (range 29 to 80 years). The majority of patients (82%) had a history of concomitant medical conditions often associated with erectile dysfunction. Over the four to six week period, the mean number of tablets taken per patient was 4.5±2.6. At the end of the study, 50 patients (68%) were still taking the recommended 50mg starting dose and 24 patients (32%) had increased the dosage to 100mg.

Table 1. Characteristics of 74 Men at Baseline.

Age (yrs)
Mean ±SD 64±11
Range 29-80

White 50
African American 18
Hispanic 4
Asian 2

Marital Status
Married 45
Single 29

Concomitant Conditions (%*)
Hypertension 35%
Coronary Artery Disease 30%
Radical Prostatectomy** 14%
Diabetes Mellitus 10%
None 18%

Severity Class
I (Satisfactory penetration possible) 14
II (Marginal penetration possible) 30
III (Partial erections, no penetration) 19
IV (No erections) 11
*Percentages do not total 100% because of co-morbidity. Only the more common conditions are shown.

**Average interval of 35±29 (S.D.) months between operation and start of sildenafil (range, 6 to 109 months)

Mean responses to questions of the IIEF are shown in Table 2. Scores on key questions 3 and 4--which address aspects of the NIH definition of ED and are relevant to the ability to achieve and maintain an erection sufficient for sexual activity--were at the follow-up visit 71% and 95% greater than baseline scores (p<0.001). When items of the IIEF were grouped into sexual domains, mean IIEF scores at the follow-up visit were significantly greater than mean IIEF scores at the baseline visit for all domains (all p<0.001). The greatest improvement from baseline was demonstrated in the domain comprising erectile function (66%) and the least improvement was in sexual desire (13%).

Table 2. Patient Responses from IIEF (International Index of Erectile Function) Questionnaire.
Domain* Baseline Score (mean±SD) Follow-up Score (mean±SD) Change (%) p Value
Erectile Function 12.5 +7.2 20.7 +9.6 66 <0.01
Question 3† 2.0 +1.5 3.4 +1.8 71 <0.01
Question 4‡ 1.7 +1.4 3.4 +1.9 95 <0.01
Orgasmic Function 5.7 +3.3 7.3 +3.4 27 <0.01
Sexual Desire 6.5 +1.9 7.3 +2.2 13 <0.02
Intercourse Satisfaction 6.8 +3.7 9.8 +4.1 44 <0.01
Overall Satisfaction 5.4 +2.7 7.5 +2.7 39 <0.01
*Domains are: Erectile function (possible total score 1 to 30), orgasmic function (possible total score 0 to 10), sexual desire (possible total score 2 to 10), intercourse satisfaction (possible total score 0 to 15), and overall satisfaction (possible total score 2-10).
†Question 3 of IIEF is "When you attempted sexual intercourse, how often were you able to penetrate (enter) your partner?". Rated on a scale of 1 (almost never or never) to 5 (almost always or always); score of zero = no attempts made.
‡Question 4 of IIEF is "During sexual intercourse, how often were you able to maintain your erection after you had penetrated (entered) your partner?". Rated on a scale of 1 (almost never or never) to 5 (almost always or always); score of zero = no attempts made.

Specifically, the ability to resume intercourse was substantial for the entire group. However, this ability was inversely related to ED severity before treatment. On the key questions, penetration/maintenance was improved with sildenafil to a response level of 4 or 5 (‘success most times’) in 35% – 44% of men who before treatment were at level O or 1 (‘no attempts or almost never successful’) and in 78% – 80% of men who were at level 2 or 3 (‘success < half the times’). Of 30 men, who were unable to make penetration, 13 were able to do so with the drug. Regarding resumption of successful intercourse with sildenafil, the less impaired men significantly outperformed the more severely impaired men (p<0.01).

In Figure 1, the inverse relationship between total IIEF score and Severity Class at baseline is shown for the 74 men completing all questionnaires. The higher the Severity Class, the lower the IIEF score. For example, men in Severity Class I, i.e., capable of satisfactory penetration, reported an IIEF score averaging 85% of maximum, which is 75. At the other extreme, men with Class IV severity, i.e., no erections, reported total IIEF scores that were on average only 53% of the maximum possible IIEF score. The relationship between the IIEF total score and Severity Class is statistically significant (p<0.01).

In Table 3, the relationship is shown between Severity Class and patient responses on the key IIEF questions 3 and 4, for the men with the lowest questionnaire scores (0-1). As seen in the table, the higher the Severity Class, the greater the percentage of men with very low scores on the questionnaires. In a corresponding analysis, men with the highest questionnaire scores were found to have a low Severity Class. The two measures--one based on a patient’s written response to a question and the other based on a physician’s impression during an interview--are correlated (p<0.001, r= -0.53 to –0.48).

Table 3. Percent (%) of men with a score of 0 or 1 (severe impairment) on key IIEF questions grouped by Severity Class*.

Severity Class

IIEF Question**






14 (2/14)

30 (9/30)

68 (13/19)

91 (10/11)


21 (3/14)

40 (12/30)

74 (14/19)

91 (10/11)

Numbers shown are percentage (%) (IIEF / Severity)
*Severity Classes are defined in Table 1.
**IIEF Questions 3 and 4 are given in legend of Table 2.

In Figure 2, the outcome variable, "Do you wish to continue sildenafil?" (vertical axis), is related to the patient’s Severity Class at baseline (horizontal axis) for the 74 men with complete questionnaires. The higher the Severity Class, the less likely the patient would be to desire to continue treatment: 93% of Class I patients, but only 55% of Class IV patients wanted to continue sildenafil. Stated differently, men capable of some penetration before treatment were almost twice as likely to desire continuation of sildenafil as men not capable of any erection before treatment. Overall, 57 of the 74 men (77%) elected to continue sildenafil following their initial trial. Of the 17 men electing not to continue, only 2 discontinued because of bothersome side effects (see below).

Modified IIEF questionnaires were completed by 34 partners at baseline and 31 after treatment. As shown in Figure 3, for both baseline and post-treatment evaluations, patient and partner assessment were correlated (r= 0.67 – 0.81, p<0.001). In other words, judging by IIEF total scores, a statistically significant association was found between patient and partner perceptions of the patient’s sexual function. When key questions 3 or 4 were analyzed separately, similar correlations were found.

Among the 14 men who had undergone radical prostatectomy, results of sildenafil therapy were substantially worse than in the other men. Most of these men were in Severity Class IV, i.e., having no erections, and none of these postoperative Class IV patients were satisfied with sildenafil treatment. Nerve-sparing surgery had been performed in five of the 14, and two of them, both capable of penetration without therapy, exhibited further symptomatic improvement and elected to stay on treatment. The other 12 postoperative patients failed and chose not to remain on sildenafil as a long-term treatment option.

A principle, bothersome side effect of sildenafil was identified as follows by 16 of the 74 patients (22%): headache (5); flushing (5); nasal congestion (3); dyspepsia (1); pain in left arm (1) and blue vision (1). Two patients discontinued treatment because of bothersome side effects, one with headache and one with flushing. Otherwise, reported side effects were mild and transitory and did not interfere with treatment. No priapism and no deaths were encountered.

Whether the patient received free samples (n=27) or not (n=47), outcomes were essentially the same. Baseline IIEF scores were approximately the same in the two groups (total score 51-53). Upon completion of the study, ‘desire to continue sildenafil’ was 73% in the free-sample group and 82% in the others (p=NS).


The advent of sildenafil has dramatically changed treatment of male erectile dysfunction (ED). "Not surprisingly, patients are clamouring for the drug…. (4)" and, as noted in the introduction, physicians are responding positively to the demand. The safety and efficacy of sildenafil have been convincingly demonstrated in a large, rigorously controlled clinical trial (1). However, with other treatments, major differences have been noted between results obtained in a clinical trial and those seen in a clinical practice setting. For example, in clinical trials, transurethral alprostadil was found to allow intercourse in 70% of men with ED (5), but in a clinical practice setting, the success rate was found to be only 37% (6).

In the present study, differing considerably from a controlled trial, we found that sildenafil usage enhanced erectile function in a practice setting. Patients studied here were on average approximately 5 years older, had more concomitant disease, were less inclined to take the 100 mg dose, and were more often lacking a stable partner than men in the trial. In addition, the present group had a great deal more information about the product: the package insert--not available during the trial-- was reviewed with each patient and a copy given to him upon drug administration. Despite the above differences, the practice results approximated the trial outcomes. Using key questions 3 and 4 from the IIEF, erectile function was improved by 71% to 95% in the present group, and overall. Overall, 77% of the men elected to continue using sildenafil after completion of the study period; while ‘desire to continue with sildenafil’ does not prove drug efficacy, it must approximate such an outcome, especially since all continuing men in this study would then be paying for the product.

An erectile dysfunction Severity Class was found to reflect the more complicated IIEF total score and appeared helpful in predicting treatment response. Although not validated, the 4-category Severity Class appeared meaningfully related to an important clinical outcome, the patient’s desire to continue sildenafil usage after the test period and to key questions 3 and 4 on the IIEF. The more severe the dysfunction, the less likely is the patient to have a satisfactory response to the drug, irrespective of the traditional classification of organic, psychogenic or mixed etiology. The potential utility of such a classification system was suggested by the National Institutes of Health Consensus Development Panel On Impotence (2). Use of a severity classification, which is simple and easy to apply, appears worthy of further exploration.

A potential source of bias was that 26% of the study group (26/100) failed to complete questionnaires. To resolve this potential bias, we contacted these men by telephone after completion of the study. Eleven men never used the drug, the most common explanation being fear of an adverse event, especially death, which was being reported in the lay press at the time. Two men were lost to follow-up. Of those who had used the drug, 10 reported improved erections and a desire to continue, and 3 reported no benefit and did not plan to continue. Thus, the telephone reports yielded approximately the same proportion of positive and negative responses as in the men who completed questionnaires. The data do not appear to be biased by the men who failed to complete questionnaires.

Prior to this report, scant attention has been given to the partner’s appraisal of therapeutic response in men with ED, despite a clear call for such involvement (2, 7). We modified the IIEF and used it in available partners to assess results of treatment and to substantiate the patient’s responses. Both before and after treatment, the partner’s responses confirmed the patient’s (r=0.67 – 0.81, p<0.01). This finding must be interpreted cautiously because of the limited number of partners completing the questionnaire (n=31-34) and lack of validation of the modified questionnaire methodology. Nevertheless, the partner confirmations were statistically significant and support a non-quantified, similar finding in the clinical trial (1). Future study of partner involvement in the evaluation and treatment process would appear warranted.

Sildenafil responses in the 14 men who had ED after radical prostatectomy were disappointing. Only two of these men responded favorably to the drug, and in them, erectile dysfunction was minimal before sildenafil administration (Class I).Should this finding be confirmed in larger studies, the information would be important in future counseling of such men. The relationship between surgical technique and response to sildenafil is the subject of ongoing investigation.


Sildenafil is an effective treatment of erectile dysfunction in a clinical practice setting. Previous data obtained via clinical trials are confirmed. The present findings, based largely on IIEF questionnaire responses, were substantiated by the responses of partners, who independently completed a modified version of the same questionnaire. Classifying the dysfunction according to severity appears to be a simple way to predict treatment response. Men who report some erectile function (Severity Class I or II) respond to sildenafil much more efficaciously than men with no erections (Severity Class IV).


Table 1. Characteristics of 74 Men at Baseline.

Table 2. Patient Responses from IIEF (International Index of Erectile Function) Questionnaire.

Figure 1. Relationship between total IIEF, denoted as % of maximum score (75) (vertical axis) and Severity Class (horizontal axis) is shown for the 74 men at baseline.

Table 3. Percent (%) of men with a score of 0 or 1 (severe impairment) on key IIEF questions grouped by Severity Class*.

Figure 2. Relationship between the patient's desire to continue sildenafil (vertical axis) and baseline Severity Class (horizontal axis). The more severe the dysfunction, the less likely the patient's desire to continue sildenafil.

Figure 3. Figure 3. Scattergrams showing relationship between responses of patient (vertical axis) and partner (horizontal axis)on IIEF total score before (a) and after (b) sildenafil treatment. Patient and partner assessments of erectile function are directly correlated.


1. Goldstein I, Lue TF, Padma-Nathan H, Rosen RC, Steers WD, and Wicker PA: Oral sildenafil in the treatment of erectile dysfunction. N Engl J Med 338: 1397-1404, 1998.

2. NIH Consensus Development Panel on Impotence. JAMA 270: 83-90, 1993.

3. Rosen RC, Riley A, Wagner G, Osterloh IH, Kirkpatrick J, and Mishra A: The International Index of Erectile Function (IIEF): A multidimensional scale for assessment of erectile dysfunction. Urology 49: 822-830, 1997.

4. Rosen RC: Sildenafil: medical advance or media event? Lancet 351:1599-1600, 1998.

5. Padma-Nathan H, Hellstrom WJ, Kaiser FE, Labasky RF, Lue TF, Nolten WE, Norwood PC, Peterson CA, Shabsigh R, and Tam PY: Treatment of men with erectile dysfunction with transurethral alprostadil. N Engl J Med 336: 1-7, 1997.

6. Werthman P and Rajfer J: MUSE therapy: Preliminary clinical observations. Urology 50: 809-811, 1997.

7. Jackson SE and Lue TF: Erectile dysfunction: Therapy health outcomes. Urology 51: 874-882, 1998.