Sildenafil Citrate (Viagra®) in the Treatment of Men with Erectile Dysfunction in Southern Latin America: a Double-Blind, Rando

Sildenaﬁl citrate (Viagra1) in the treatment of men with erectile dysfunction in southern Latin America: a double-blind, randomized, placebo-controlled, parallel-group, multicenter, ﬂexible-dose escalation study

E Becher1*, A Tejada Noriega2, R Gomez3 and R Decia4 on behalf of the Southern Latin America Sildenaﬁl Study Group, Buenos Aires, Argentina

1Centro de Diagno´stico Urolo´gico, Buenos Aires, Argentina; 2Servicio de Urologıá, ANDROMED, Instituto de Andrologıá, Urologıá y Sexologıá, Lima, Peru; 3Servicio de Urologıá, INTEGRAMEDICA, Santiago de Chile, Chile; and 4Servicio de Urologıá, Hospital Italiano de Montevideo, Montevideo, Uruguay

Our objectives were: (1) to determine the efficacy, safety, and tolerability of sildenafil citrate (Viagra1) administered to men with broad-spectrum erectile dysfunction (ED) in southern Latin America; and (2) to correlate Rigiscan1 measurements assessing ED etiology with the investigator’s assessment. A total of 141 men with broad-spectrum ED (mean age 57) were enrolled in a randomized, 12-week, double-blind, placebo-controlled, flexible-dose escalation study of sildenafil. After the 12-week treatment period, the mean score for the primary efficacy variables had risen significantly: for the sildenafil group, 66.2% from baseline for question 3 of the International Index of Erectile Function and 77.6% for question 4, vs 15.1% and 21.2% for the placebo group, respectively (P < 0.0001). Rigiscan1 data confirmed investigator assessments of etiology. Headache and flushing, usually mild and transient, were the most common adverse events. Sildenafil was an effective, well-tolerated treatment for men in southern Latin America with broad-spectrum ED. International Journal of Impotence Research (2002) 14, Suppl 2, S33–S41. doi:10.1038=sj.ijir.3900896

Keywords: sildenaﬁl; erectile dysfunction; Latin American; safety; efﬁcacy; intercourse success rates

Introduction the opening of potassium channels and hyperpolar- ization of the muscle cell membrane, sequestration of intracellular calcium by the endoplasmic reticu- Erectile dysfunction (ED) is defined as the persistent lum, and blocking of calcium influx by the inhibi- inability to achieve or maintain an erection suffi- tion of calcium channels.4 The consequence is a 1 cient to permit satisfactory sexual intercourse. It drop in cytosolic calcium concentrations and has been estimated to affect up to 30 million men in relaxation of the smooth muscle. During the return 2,3 the USA. ED may result from psychological, to the flaccid state, cGMP is hydrolyzed to GMP by neurologic, hormonal, vascular, or cavernosal phosphodiesterase type 5 (PDE5).4 Other phospho- impairment or from a combination of these fac- diesterases are also found in the corpus cavernosum, 4,5 2–5 tors. The disorder is age-associated, with but they do not appear to have an important role in estimated prevalence rates of roughly 39% among the erection process. Sildenafil citrate (Viagra1, men 40-y-old and 67% among those 70-y-old. Pfizer Inc, New York, NY, USA) is a selective Normal penile erection depends on the relaxation inhibitor of cGMP-specific PDE5.6–8 of smooth muscles in the corpora cavernosa. In By selectively inhibiting cGMP catabolism in response to sexual stimuli, cavernous nerves and cavernosal smooth muscle cells, sildenafil enables endothelial cells release nitric oxide (NO), which the natural erection response to sexual stimulation, stimulates the formation of cyclic guanosine mono- but does not cause erections in the absence of such 4 phosphate (cGMP) by guanylate cyclase. Cyclic stimulation. GMP in turn activates a specific protein kinase, In a number of randomized, placebo-controlled, which phosphorylates certain proteins, resulting in double-blind, and open-label studies, sildenafil has been shown to be an effective oral treatment for male ED and to have an adequate safety profile.9–14 *Correspondence: E Becher, Centro de Diagno´stico Urolo´gico (CDU), Av. Cordoba 2424, Ciudad de Buenos Aires, Whether the results of these initial double-blind Argentina 1120. and open-label randomized trials can be extrapo- E-mail: [email protected] lated to a population-based sample of men with ED Sildenafil citrate in the treatment of ED in Latin America E Becher et al S34 in Latin America remains to be determined. There- prior to sexual activity, as alcohol may impair the fore, the objectives of the present study were to ability to have erections, and thus potentially cause determine the efficacy, safety, and tolerability of the treatment to be less effective. A total of six study sildenafil over a 12-week period, taken as required, visits were conducted to provide assessments of prior to anticipated sexual activity in male out- efficacy, safety, and tolerability. Following rando- patients with objectively determined broad- mization, patients returned to the clinic for follow- spectrum etiology. The patients were all from up visits after 2, 4, 8, and 12 weeks of treatment. southern Latin America (Argentina, Chile, Peru, Other concomitant medications, which could and Uruguay). This was also the first study to have an effect on erectile function, remained objectively classify ED etiology through Rigiscan1 constant during the study, unless changes were (Timm Medical Technologies, Eden Prairie, MN) required for patient safety. measurements and to correlate it with the investi- Patients who experienced no adverse events (AEs) gator’s clinical assessment. with the 50-mg dose of sildenafil (or corresponding placebo), but whose ED was insufficiently improved at this dose, could increase their dose to 100 mg at Methods the subsequent visit. Patients who responded well at a particular dose were not allowed to receive higher doses. Patients who were receiving the 50- or 100-mg This was a double-blind, randomized, placebo- dose were allowed to decrease the dose to the next controlled, multicenter, parallel-group, flexible-dose lowest level only if they were experiencing intoler- escalation study. Each man had to be in a stable able or severe AEs. Any dose adjustments between relationship with a female partner of at least 6 regularly scheduled visits were done for safety months’ duration. The cause and etiology of the ED purposes only. Sildenafil was provided as 25-, 50-, were determined from the medical history, physical and 100-mg tablets in bottles (14 tablets=bottle). examination, and other diagnostic procedures, in- At weeks 0 (baseline) and 12, each participant cluding a Rigiscan test (nocturnal penile tumescence completed the International Index of Erectile Func- [NPT] testing). The penile tumescence and rigidity tion (IIEF), a 15-question validated, multidimen- measure was performed at the base and=or tip of the sional, self-administered questionnaire used for the penis at the prescreening evaluation. This monitor- clinical assessment of ED and treatment outcomes in ing of nocturnal erections was performed at home for clinical studies.15,16 The global efficacy assessment two nights, with a monitoring period of at least 5 h question (GEAQ) (‘Did treatment improve your per night, in order to distinguish between psycho- erections?’) was asked at week 12. Event logs (patient genic and organic ED, and to enable the investigators diaries) were completed after each attempt at sexual to perform a subanalysis according to the established intercourse and reviewed at each study visit. etiology. The results of this NPT testing were The primary efficacy measures in this study were expressed as the mean average rigidity reached by the responses to questions 3 (Q3, ability to achieve the patients during the monitoring period. an erection) and 4 (Q4, ability to maintain an Men were excluded if they had penile anatomical erection) of the IIEF. IIEF responses were rated on defects, a primary diagnosis of another sexual a scale of 1 (‘almost never=never’) to 5 (‘almost disorder (for example, premature ejaculation), spinal always=always’). A score of 0 indicated no attempt cord injury, any major psychiatric disorder not well at sexual intercourse. The secondary efficacy mea- controlled with treatment, poorly controlled diabetes sures in the study were the responses to the GEAQ mellitus, a history of alcohol or substance abuse, and the responses on the event log, including the major hematological, renal, or hepatic abnormalities, proportion of successful attempts at intercourse. hypotension (sitting blood pressure < 90=50 mm Hg) In addition, the responses to the IIEF were used to or malignant hypertension, a recent (occurring with- evaluate the effect of sildenafil or placebo on five in the previous 6 months) stroke or myocardial domains of male sexual erectile function (questions infarction, or if they were receiving nitrates. 1 – 5, and question 15 of the IIEF; possible total score The institutional review board at each center 1 – 30), orgasmic function (questions 9 and 10 of the approved the protocol, and all the men gave written IIEF; possible total score 0 – 10), sexual desire informed consent to participate. (questions 11 and 12 of the IIEF, possible total score To confirm the diagnosis of male ED after the 2 – 10), intercourse satisfaction (questions 6, 7, and 8 preliminary evaluation and following a 4-week run- of the IIEF; possible total score 0 – 15), and overall in period with no treatment, during which baseline satisfaction (questions 13 and 14 of the IIEF; data on sexual function were collected, patients possible total score 2 – 10). The domain scores were entered a 12-week double-blind treatment period. computed by adding the scores for the individual The men were instructed to take a dose of the study questions in each domain. drug approximately 1 h before planned sexual All observed or volunteered AEs, regardless of activity but no more than once daily. Patients were treatment group or suspected causal relationship to only allowed to drink up to two units of alcohol study drug, were recorded by the investigators.

International Journal of Impotence Research Sildenafil citrate in the treatment of ED in Latin America E Becher et al S35 Statistical analysis from Peru, 15 from Chile, and 11 from Uruguay). Seventy-two patients were assigned to receive active treatment with sildenafil and 71 to the placebo The treatment groups were checked for similarity group. Two patients from the placebo group were with respect to demographic and event log variables lost to follow-up after visit two. The Study Project using descriptive statistics. Continuous variables are Team assumed they took one dose of study drug. presented as mean standard deviation or median Therefore, they were included in the final safety and total ranges. Discrete variables are given as datasets and excluded from the efficacy analysis. In frequencies and percentages. Univariate statistical summary, 141 patients were considered for the testing was performed with the likelihood ratio chi- efficacy analysis and 143 patients were included in square test with Yates’ correction or Fisher’s exact the safety analysis. test, as appropriate, for discrete variables, and with two-sample t-tests or Mann – Whitney tests as appropriate for continuous variables. The primary time point for the primary efficacy and secondary efficacy analyses was the end of double-blind therapy (week 12). Each of the two questions defining the primary Baseline characteristics efficacy measure (Q3 and Q4) were analyzed separately, using analysis of covariance (ANCOVA) Baseline characteristics of the patients are summar- methods specified below. Significant results at ized in Table 1. No significant differences were the 5% level in both the analyses were required observed between both treatment groups in terms of to demonstrate efficacy over placebo. An ANCOVA age, race, mean body weight (kg), average height method was fitted for each question, which included (cm), time elapsed since the first diagnosis of ED, main-effect terms for investigational center and and smoking status. treatment effect (as ordered categorical variables), The more frequent and=or relevant concomitant with baseline score, patient age, smoking, and illnesses present in each patient group at the time duration and cause of ED as covariates. of enrolment are listed in Table 2. In the The answers to the GEAQ (yes or no) were overall population sample, 25 patients had diabetes analyzed using logistic regression analysis, account- mellitus (17.5%). No significant differences were ing for the same covariates as those listed for the observed between the sildenafil and the placebo ANCOVA models. The proportion of successful groups in the prevalence of any concomitant attempts at intercourse derived from the event illness (for example, 31.9% had hypertension in log data were analyzed using a general linear the sildenafil vs 43.6% in the placebo groups, model. All significance tests comprised all subjects respectively). The most frequent=relevant used receiving active medication, regardless of the drugs are summarized in Table 3 (some patients dose actually taken, into a single group, and all received more than one concomitant drug). No patients receiving placebo, regardless of the number significant difference was observed in the use of tablets actually taken, into another single group. of concomitant medications between treatment All reported probability values are 2-tailed, groups. and P-values < 0.05 were accepted as statistically significant. All statistical analyses were performed using SAS System for Windows version 6.12 software (Cary, NC). Table 1 Demographic characteristics

Sildenafil Placebo P Results Gender Male Male Number of subjects 72. 71. — Age (mean s.d.) 57.2 11.5 56.7 10.9 NS Ethnicity, n (%) Demographics White 6.(8.3) 7.(9.8) NS Hispanic 66.(91.7) 64.(90.2) NS Weight (kg) (mean s.d.) 84.2 11.8 83.4 14.2 NS A total of 156 patients underwent the prescreening; Height (cm) (Mean s.d.) 173.9 7.8 172.9 7.7 NS Duration of erectile dysfunction 146 were randomized after giving written informed since first diagnosis (y) 3.5 2.6 consent. Three screened patients were excluded (mean range) (0.5 – 22.4) (0.5 – 20.5) NS from the study because they did not take the study Smoking Status, n (%) medication and were lost to follow-up. (They did Current smoker 32.(44.4) 27.(38.0) NS not enter in the active treatment phase.) The Former smoker 28.(38.9) 26.(36.6) NS Nonsmoker 12.(16.7) 18.(25.4) NS remaining 143 patients entered the active treatment phase of the study (93 patients from Argentina, 24 NS ¼ not significant.

International Journal of Impotence Research Sildenaﬁl citrate in the treatment of ED in Latin America E Becher et al S36 Table 2 Concomitant illnesses

Sildenaﬁl Placebo n (%) n (%) P

Number of subjects 72 71 — Diabetes mellitus Type I 4 1 Type II 8 12 Total 12.(16.6) 13.(18.3) NS Hypertension 23.(31.9) 28.(43.6) NS Hypercholesterolemia 3.(4.1) 3.(4.2) NS Myocardial infarction (old) 5.(6.9) 1.(1.4) NS Angina pectoris 1.(1.38) 1.(1.4) NS Figure 1 Severity of erectile dysfunction at baseline by IIEF Coronary artery disease 2.(2.77) 1.(1.4) NS domain 1. IIEF ¼ International Index of Erectile Function. Peripheral vascular disease 2.(2.77) 1.(1.4) NS Benign prostate hyperplasia 9.(12.5) 11.(15.4) NS Carcinoma of prostate 2.(2.77) 5.(7.0) NS Radical prostatectomy 2.(2.77) 5.(7.0) NS The investigator judged etiology of the ED to be Transurethral resection of 6.(8.3) 6.(8.4) NS psychogenic in 44.3% of the patients, organic in prostate 39.3%, and mixed in 16.4%. A baseline Rigiscan Impaired renal function 1.(1.38) 1.(1.4) NS Gastritis 6.(8.3) 9.(12.6) NS measurement was also performed on each patient to Gastroesophageal reﬂux 2.(2.77) 2.(2.8) NS evaluate the causality of the ED. The Rigiscan Peptic ulcer 8.(11.1) 3.(4.2) NS measurements conﬁrmed the investigator’s assess- Visual abnormalities 5.(6.9) 3.(4.2) NS ment of organicity. The average rigidity (mean Neuropathy 2.(2.8) 0.(0) NS percentage s.d.) according to etiology was 85.4 NS ¼ not signifcant. 9.8% (psychogenic), 54.4 26.9% (organic), and 73.9 18.7% (mixed). Table 3 Concomitant medications

Sildenaﬁl Placebo n (%) n (%) P Duration of therapy=number of doses taken

Number of subjects 72. 71. — Number of subjects with any 51.(70.8) 51.(71.8) NS The median number of doses taken by the patients concomitant medication was 33 (range 1 – 80) in the sildenafil group and 31 Antihypertensive drugs 22.(30.6) 25.(35.2) NS Alpha-adrenoreceptor blocking 5.(6.9) 2.(2.8) NS (range 1 – 83) in the placebo group. The average drugs number of doses taken per month were 15.2 in the Angiotensin-converting enzyme 12.(16.6) 22.(30.98) 0.069 sildenafil group and 14.4 in the placebo group. inhibitors After 12 weeks of treatment, the proportions of Beta-blocking drugs—single agents 7.(9.7) 6.(8.5) NS Calcium-channel blocking drugs 7.(9.7) 9.(12.7) NS men taking 25-mg, 50-mg, and 100-mg sildenafil Diuretics 0.(0) 3.(4.2) NS were 5.5% (four men), 34.7% (25 men), and 50% (36 Cardiac glycosides 0.(0) 2.(2.8) NS men), respectively. For the group taking placebo, the Drugs used in diabetes 9.(12.5) 10.(14.1) NS corresponding proportions were 0% (0 men), 11.6% Insulins 4.(5.5) 2.(2.8) NS (8 men), and 82.6% (57 men). Oral antidiabetic drugs 7.(9.7) 8.(11.2) NS Anti-inflammatory analgesics 9.(12.5) 8.(11.2) NS Antidepressant drugs 3.(4.2) 2.(2.8) NS Cytotoxic drugs 0.(0) 1.(1.4) NS Drugs for hyperlipidemia 6.(8.3) 6.(8.5) NS Efficacy Hypnotics, sedatives, anxiolytics 6.(8.3) 9.(12.7) NS NS ¼ not significant. The primary efficacy measures in this study were the responses to Q3 and Q4 of the IIEF at baseline and at the end of the 12 weeks in an intent-to-treat Data on baseline erectile dysfunction analysis. The mean scores for Q3 and Q4 were significantly higher after treatment for the sildenafil group than for the placebo group (P < 0.0001) Figure 1 shows the observed severity (%) of ED (Table 4, Figure 2). The percentage increase from according to the erectile function domain of the baseline was 66.2% for Q3 and 77.6% for Q4 for the IIEF16 at the baseline visit. Fifty-one percent of the men taking sildenafil, whereas it was 15.1% and study population sample presented moderate ED in 21.2% for those taking placebo, respectively. the baseline assessment. The remaining patients The analysis of the primary efficacy endpoint were distributed as follows: 7.8% had mild ED, using only patients who completed the 12-week 24.3% had mild to moderate ED, and 17.1% had treatment period yield a similar result. After a 12- severe ED. week treatment period, the mean score for Q3 was

International Journal of Impotence Research Sildenaﬁl citrate in the treatment of ED in Latin America E Becher et al S37 Table 4 Results of primary efﬁcacy analysis (IIEF Q3 and Q4) (intent-to-treat analysis)

Overall Last Observation Baseline (Week 12)

Least squared Percent change Treatment P Parameter Treatment group n Mean s.d. mean s.e. from baseline* value{

Q3: Frequency of Sildenadﬁl 66 2.313 1.359 3.844 0.171 66.2 < 0.0001 penetration Placebo 65 2.663 0.176 15.1 Q4: Frequency of Sildenaﬁl 66 2.031 1.227 3.607 0.176 77.6 < 0.0001 maintained erection Placebo 65 2.462 0.182 21.2

*Percent differences are between the last observation (end of treatment) mean scores and the baseline mean scores. {Intergroup P value from analysis of covariance model which includes terms for baseline, treatment group, site, and treatment group by site interaction. IIEF ¼ International Index of Erectile Function.

After 12 weeks of treatment, improved erections were reported by 77.3% (CI 95%, 65.6 – 85.8) of the patients taking sildenafil as compared with 33.8% (CI 95%, 23.4 – 46.1) of those taking placebo (P ¼ 0.0001) (Figure 4). Nearly 63% of all attempts at sexual intercourse by the men receiving sildenafil were successful as compared with 26.5% for those receiving placebo (P ¼ 0.0051) (Figure 5). The analysis of the responses to Q3 and Q4 of the IIEF according to baseline rigidity measurements reached results similar to those of the entire patient cohort. The mean scores for Q3 and Q4 achieved by sildenafil were higher than those for placebo for patients with Rigiscan rigidity measurements 70% Figure 2 Primary efficacy results, answers to IIEF Q3 and Q4 (Q3 sildenafil, 3.6 vs placebo, 2.8; Q4 sildenafil, 3.2 (ITT). ITT ¼ intent-to-treat; IIEF ¼ International Index of Erectile > Function. vs placebo, 2.4) or 70% (Q3 sildenafil, 4.0 vs placebo, 2.7; Q4 sildenafil, 3.9 vs placebo, 2.4), as well as for those patients with Rigiscan rigidity measurements 50% (Q3 sildenafil, 3.6 vs placebo, > 3.9 0.2 (s.e.) for the patients taking sildenafil vs 2.8; Q4 sildenafil, 3.3 vs placebo, 2.2) or 50% 2.8 0.2 (s.e.) for those taking placebo (P < 0.0001). (Q3 sildenafil, 3.9 vs placebo, 2.7; Q4 sildenafil, 3.7 The mean score for Q4 was 3.8 0.2 (s.e.) for the vs placebo, 2.5). patients taking sildenafil vs 2.4 0.2 (s.e.) for those taking placebo (P < 0.0001). The secondary efficacy measures in the study Safety included the responses to the GEAQ, the separate responses to the five domains of male sexual function of the IIEF (erectile function, orgasmic All 143 patients enrolled in this study were function, sexual desire, intercourse satisfaction, considered for safety analysis. Discontinuation rates overall satisfaction), and the proportion of success- were low. During active treatment, seven of the 72 ful attempts at intercourse. The mean score for the patients receiving sildenafil (9.7%) and six of the 71 erectile function domain was significantly higher for patients taking placebo (8.5%) discontinued treat- the men taking sildenafil (20.5 0.6) than for those ment. None of the discontinuations were related to taking placebo (15.9 0.7; P < 0.0001) (Figure 3A). the study drug in the opinion of the investigator. The mean scores for the orgasmic function, sexual Two men (2.8%) in the sildenafil group stopped desire, intercourse satisfaction, and overall satisfac- taking the drug due to nontreatment-related AEs tion were also significantly higher in the sildenafil (diabetic coma and motor vehicle accident, respec- group (Figure 3B – E). These results were consistent tively), and one patient (1.4%) in the placebo group across all questions constituting the five domains. died. The separate statistical analyses of each question Forty-three patients (59.7%) in the sildenafil yielded similar results, showing that the sildenafil group reported AEs compared with 21 patients group achieved significantly better scores than the (29.6%) in the placebo group (P ¼ 0.079, not sig- placebo group throughout the entire questionnaire nificant) (Table 6). Two patients in the sildenafil (Table 5). group and one in the placebo group experienced

International Journal of Impotence Research Sildenaﬁl citrate in the treatment of ED in Latin America E Becher et al S38

Figure 3 (Continued)

Figure 3 (A – E) Secondary efﬁcacy results, IIEF by domains (intent-to-treat group). IIEF ¼ International Index of Erectile Function. Figure 4 Global efﬁcacy assessment question.

serious AEs. None of the serious AEs were con- placebo group experienced a myocardial infarction sidered to be related to the study drug in the that ended in the patient’s death. None of the men investigator’s judgment. The most frequently taking sildenafil experienced a myocardial infarc- reported AEs are summarized in Table 6. Patients tion or an acute ischemic syndrome. No significant enrolled in the sildenafil treatment arm experienced differences were observed between the sildenafil significantly more flushing (22.2%) and headache and placebo groups in the rates of the remaining AEs (23.6%) than those receiving placebo (flushing (Table 6). 4.2%, P ¼ 0.003; headache 8.4%, P ¼ 0.05). None of The summary of AEs according to their severity these events were serious, and they were frequently and corresponding treatment group showed that mild and transient in nature. One patient in the almost 89% of the AEs in the sildenafil group were

International Journal of Impotence Research Sildenafil citrate in the treatment of ED in Latin America E Becher et al S39 patients with ED in Latin America was based on a double-blind, randomized, placebo-controlled, multicenter, flexible-dose escalation trial design. The administration of the study drug ‘as required’ in a flexible-dose escalation design closely replicates the conditions found in a community-based medical practice.2 The study was performed in a natural env- ironment, which meant that we had to rely on the men’s own report of efficacy. However, the self- administered questionnaire of the IIEF has a high degree of sensitivity and specificity for detecting Figure 5 Proportions of attempts at successful sexual intercourse treatment-related changes in men with ED.16 This (intent-to-treat analysis). questionnaire, together with the GEAQ, provided a comprehensive assessment of erectile function, and mild, 7.7% moderate, and only 3.5% severe. None of the event log provided information about individual the severe AEs were considered related to study drug administrations. drug according to the investigator’s judgment. In the Recognizing the fact that in most men ED is a placebo group, 63.4% of the events were categorized multifactorial problem, we enrolled patients whose as mild by the investigator, 31.7% as moderate, and ED stemmed from a broad variety of causes. Overall, 4.9% as severe. the results of the efficacy assessments demonstrated that sildenafil significantly improved erectile function and the success of intercourse. These efficacy Discussion results are consistent with other recently published sildenafil double-blind studies.9–14 The therapeutic response to sildenafil has been reported to be similar This 12-week study of the efficacy, safety, and in men both with different etiologies and different tolerability of sildenafil in the treatment of male degrees of severity of ED.17 – 24 Although the present

Table 5 Results for secondary efﬁcacy analysis (IIEF Q1, 2, 5 – 15) (intent-to-treat analysis)

Overall Baseline Last Observation (Week 12)

Least squared Treatment P Parameter Treatment group n Mean s.d. mean s.e. value*

Q1: Able to get an Sildenafil 66 2.221 1.303 3.837 0.171 < 0.0001 erection Placebo 65 2.659 0.176 Q2: Erection hard Sildenafil 66 2.168 1.278 3.748 0.18 < 0.0001 enough Placebo 65 2.612 0.186 Q5: Maintain erection Sildenafil 66 3.027 1.389 3.481 0.173 < 0.0001 to completion Placebo 65 2.0 0.168 Q6: Attempted sexual Sildenafil 66 2.634 1.254 3.793 0.14 0.0097 intercourse Placebo 65 3.260 0.145 Q7: Satisfaction of Sildenafil 66 2.443 1.365 3.863 0.175 < 0.0001 sexual intercourse Placebo 65 2.635 0.181 Q8: Enjoyment of Sildenafil 66 2.237 1.214 3.387 0.147 < 0.0001 sexual intercourse Placebo 65 2.50 0.152 Q9: Frequency of Sildenafil 66 3.176 1.643 4.094 0.182 0.0284 ejacualtion Placebo 65 3.510 0.189 Q10: Frequency of Sildenafil 66 3.115 1.562 4.116 0.185 0.0128 orgasm or climax Placebo 65 3.436 0.193 Q11: Frequency of Sildenafil 66 3.595 1.135 4.175 0.106 0.0192 sexual desire Placebo 65 3.812 0.109 Q12: Rating of sexual Sildenafil 66 3.046 0.867 3.517 0.093 0.0327 desire Placebo 65 3.225 0.096 Q13: Satisfaction with Sildenafil 66 2.656 1.207 3.911 0.149 < 0.0001 sex life Placebo 65 2.811 0.154 Q14: Satisfaction with Sildenafil 66 2.901 1.270 4.173 0.141 < 0.0001 sexual relationship Placebo 65 3.058 0.146 Q15: Confidence to Sildenafil 66 2.305 0.894 3.321 0.122 < 0.0001 get and keep an erection Placebo 65 2.484 0.123

*Intergroup P value from analysis of covariance model which includes terms for baseline, treatment group, site, and treatment group by site interaction. IIEF ¼ International Index of Erectile Function.

International Journal of Impotence Research Sildenafil citrate in the treatment of ED in Latin America E Becher et al S40 Table 6 Summary of the most frequent adverse events between the sildenafil and the placebo group in a pooled analysis of double-blind and open-label Sildenafil Placebo 14 n (%) n (%) P studies. We believe this is the first study to demonstrate a Number of subjects 72 71 — positive correlation between the investigator’s clin- Number of subjects with 43.(59.7) 21.(29.6) 0.079.(NS) ical assessment of Rigiscan determinations regard- adverse events Myocardial infarction 0.(0) 1.(1.4) NS ing ED etiology (organic vs psychogenic) in a clinical Flushing 16.(22.2) 3.(4.2) 0.003 drug trial. Headache 17.(23.6) 6.(8.4) 0.05 Hypertension 1.(1.38) 3.(4.2) NS Nasal congestion 2.(2.77) 1.(1.4) NS Visual disturbance 4.(5.5) 0.(0) NS Conclusions Tachycardia 3.(4.1) 0.(0) NS

NS ¼ not significant. This 12-week study of the efficacy, safety, and tolerability of sildenafil in the treatment of male study was not designed to address either of these patients with ED in Latin America demonstrated questions in this population of men with ED from that sildenafil significantly improved erectile func- southern Latin America, a retrospective analysis of tion and the percent of successful intercourse the present study (unpublished data) suggests that attempts. These efficacy results are consistent with similar results can be expected across the different other recently published sildenafil double-blind and etiologies of ED. In addition, a specific analysis of open-label studies. the study efficacy data involving different degrees Sildenafil was well tolerated. The most frequent of organic ED shown in the present study (analysis AEs reported were headache and flushing, and they of efficacy in patients presenting Rigiscan measure- were usually mild and transient. The proportion of ments above or below 50%, and above and below discontinuations was similar between the sildenafil 70%), suggests that similar results can be expected and the placebo group, suggesting a relatively high across the different degrees of severity of organic ED. level of drug tolerability and acceptance. None of However, these conclusions must be confirmed by the men enrolled in the sildenafil group experienced other prospective studies due to the small number of acute ischemic syndromes or sudden death as AEs. patients in the current one. A high degree of correlation was found regarding In earlier studies, the mean scores on the IIEF the etiologic diagnosis (ie, organic, psychogenic, or after sildenafil therapy approached those of age- mixed) of ED in the investigator’s clinical assess- matched men without ED.4,9 Although sildenafil did ment and objective Rigiscan measurements. not affect the level of sexual desire in previous studies, the present trial found a significant increase in the mean score for sexual desire in patients taking Acknowledgements sildenafil. This finding might be explained by the positive emotional influence exerted by the achieved improvement on penile erection capabil- We are indebted to all investigators who partici- ity, but we must acknowledge that it has no specific pated in this trial, and to Dr Ernesto Ferreiro´s, or definite explanation that can be attributed to the Clinical Data Manager of Pfizer SoLA for data study drug.6–8 management, data analysis, preparation, and revi- Sildenafil was well tolerated in our study. The sion of the manuscript. main AEs were headache, flushing, and visual This study was funded by Pfizer South Latin disturbances, and they were usually mild and America Region. transient. These AEs have been previously reported, with a similar degree of severity, in several sildenafil studies.9 – 14,17 – 24 They reflect the pharmacologic nature of sildenafil as a PDE5 inhibitor, with modest References vasodilator and weak PDE6 inhibitory properties.6–8 Few men discontinued sildenafil treatment, and the 1 NIH Consensus Development Panel on Impotence. Impotence. proportion of discontinuations was similar between NIH Consensus Conference. JAMA 1993; 270: 83 – 90. the sildenafil and the placebo group, suggesting a 2 Feldman HA et al. Impotence and its medical and psychoso- cial correlates: results of the Massachusetts Male Aging Study. relatively high level of drug tolerability and accep- J Urol 1994; 151: 54 – 61. tance. None of the men enrolled in the sildenafil 3 Laumann EO, Paik A, Rosen RC. Sexual dysfunction in the group experienced acute ischemic syndromes or United States: prevalence and predictors. JAMA 1999; 281: sudden death as AEs. These results are in con- 537 – 544. cordance with a recently presented report from 4 Lue TF. Erectile dysfunction. New Engl J Med 2000; 342: 1802 – 1813. Mittleman and colleagues that showed a similar 5 Morgentaler A. Male impotence. Lancet 1999; 354: incidence rate of death and myocardial infarction 1713 – 1718.

International Journal of Impotence Research Sildenafil citrate in the treatment of ED in Latin America E Becher et al S41 6 Moreland RB, Goldstein I, Traish A. Sildenafil, a novel 22 OlssonAM, PerssonC-A.Efficacyandsafetyof Viagra(sildenafil inhibitor of phosphodiesterase type 5 in human corpus citrate) in men with cardiovascular disease and erectile cavernosum smooth muscle cells. Life Sci 1998; 62: dysfunction. J Am Coll Cardiol 2000; 35(Suppl A): 329-A. PL-309 – PL-318. 23 Conti RC, Pepine CJ, Sweeney M. Efficacy and safety of 7 Boolell M et al. Sildenafil: an orally active type 5 cyclic sildenafil citrate in the treatment of erectile dysfunction in GMP-specific phosphodiesterase inhibitor for the treatment of patients with ischemic heart disease. Am J Cardiol 1999; 83: penile erectile dysfunction. Int J Impot Res 1996; 8: 47 – 52. 29C – 34C. 8 Ballard SA et al. Effects of sildenafil on the relaxation of the 24 Morales A et al. Clinical safety of oral sildenafil citrate human corpus cavernosum tissue in vitro and on the activities (Viagra) in the treatment of erectile dysfunction. Int J Impot of cyclic nucleotide phosphodiesterase isozyme. J Urol 1998; Res 1998; 10: 69 – 73. 159: 2164 – 2171. 9 Goldstein I et al. Oral sildenafil in the treatment of erectile dysfunction. New Engl J Med 1998; 338: 1397 – 1404. Appendix 10 Feldman R, Meuleman EJH, Steers W. Sildenafil (Viagra) in the treatment of erectile dysfunction: analysis of two flexible- dose-escalation studies. Int J Impot Res 1998; 10(Suppl 3): S33. The members of the Southern Latin America Abstract. Sildenafil Study Group are: Jose´ Arias Delgado 11 Steers WD. Meta-analysis of the efficacy of sildenafil (Viagra) (ANDROMED: Instituto de Andrologıá, Urologıáy in the treatment of severe erectile dysfunction. J Urol 1998; 159: 910. Sexologıá, Lima, Peru); Amado Bechara (Instituto 12 Langtry HD, Markham A. Sildenafil. A review of its use in Me´dico Especializado [IME], Buenos Aires, erectile dysfunction. Drugs 1999; 57: 967 – 989. Argentina); Edgardo Becher (Centro de Diagno´stico 13 Steers WD. Viagra — after one year. Urology 1999; 54: 12 – 17. Urolo´gico, Buenos Aires, Argentina); Adolfo Casabe´ 14 Mittleman MA, Glasser DB, Orazem J, Collins M. Incidence of myocardial infarction and death in 53 clinical trials of (Instituto Me´dico Especializado [IME], Buenos Viagra (sildenafil citrate). J Am Coll Cardiol 2000; Aires, Argentina); Isidoro Davidzon (Clinica 35(Suppl A): 302A. Suizo-Argentina, Buenos Aires, Argentina); Ricardo 15 Rosen RC et al. The International Index of Erectile Function Decia (Hospital Italiano de Montevideo, Montevideo, (IIEF): a multidimensional scale for assessment of erectile Uruguay); Reynaldo Gomez (INTEGRAMEDICA, dysfunction. Urology 1997; 49: 822 – 830. 16 Cappelleri JC et al. Diagnostic evaluation of the erectile Santiago de Chile, Chile); Guillermo Gueglio function domain of the International Index of Erectile (Hospital Italiano de Buenos Aires, Buenos Aires, Function. Urology 1999; 54: 346 – 351. Argentina); Osvaldo Mazza (Buenos Aires, 17 Rendell MS, Rajfer J, Wicker PA, Smith MD. Sildenafil for Argentina); Rau´ l Olmedo (Fundacioń Urolo´gica de treatment of erectile dysfunction in men with diabetes: a randomized controlled trial. JAMA 1999; 281: 421 – 426. Cordoba para la Docencia e InvestigaciońMe´dica 18 Derry FA et al. Efficacy and safety of oral sildenafil (Viagra) in [FUCDIM], Cordoba, Argentina); Miguel A, Rivero men with erectile dysfunction caused by spinal cord injury. (Buenos Aires, Argentina); Freddy Romanelli (Bahıá Neurology 1998; 51: 1629 – 1633. Blanca, Argentina); Alberto Tejada Noriega 19 Zippe CD et al. Treatment of erectile dysfunction after radical (ANDROMED: Instituto de Andrologıá, Urologıáy prostatectomy with sildenafil citrate (Viagra). Urology 1998; 52: 963 – 966. Sexologıá, Lima, Peru); Juan Andre´s Venegas 20 Osterloh I, Boolell M, Prisant M. Sildenafil citrate is a well- (Hospital de la Seguridad, Valparaı´so, Chile); Lelio tolerated treatment for patients with erectile dysfunction Zeno (Sanatario Parque de Rosario, Rosario, taking concomitant antihypertensive therapy. Eur Heart J Argentina). 1999; 20(Suppl): 78. Abstract P-554. 21 Zusman R, Collins M. Cardiovascular safety of sildenafil in the Pfizer Steering Committee: Honorio Silva, Julio treatment of erectile dysfunction. J Am Coll Cardiol 1999; Camps, Claire Wohlhuter, Richard Siegel, Enrique 33(Suppl A): 309-A. Comesan˜ aDıáz, Jack Mardekian.

International Journal of Impotence Research