Duloxetine: A Summary of Published Clinical Experience

URINARY INCONTINENCE IN WOMEN Duloxetine: A Summary of Published Clinical Experience Roger R. Dmochowski, MD Department of Urologic Surgery, Vanderbilt University, Nashville, TN Effective oral therapy for genuine stress urinary incontinence (SUI) in women has, to date, been an unattainable goal. Although oral pharmacologic agents have been used for this condition, none has ultimately been successful, because of side effects, lack of efficacy, or problematic compliance with drug ingestion. The availability of an effective oral agent for SUI would increase the range of therapeutic options for symptom management and possibly make treatment accessible to more women who otherwise feel that surgical therapy is not an option because of social, personal, or medical reasons. Duloxetine is a selective serotonin (5-HT) and norepinephrine reuptake inhibitor that has been shown to increase rhabdosphincter activity. Rhabdosphincter contractility changes are thought to occur as the result of increased stimulation of 1-adrenergic and 5-HT2 receptors in the sacral spinal cord, resulting in increased efferent pudendal nerve activity, producing increased pelvic floor tonus. Two large-scale studies have been completed employing subjective and objective outcomes to assess the therapeutic index of duloxetine as a therapy for SUI. [Rev Urol. 2004;6(suppl 3):S56-S63] © 2004 MedReviews, LLC Key words: Duloxetine • Incontinence episode frequency • Incontinence Quality of Life scale • Stress urinary incontinence tress urinary incontinence (SUI) is an extremely bothersome condition that afflicts millions of women, many of whom experience a substantive disruption in their quality of life as a result of the disorder. Bother associated with SUI is exacerbated by the presence of a mixed symptomatology (urgency and frequency). Of the estimated 5 million to 29 million women in the United States who experience SUI, between 29% and 60% have coexistent urinary urgency and frequency.1 S S56 VOL. 6 SUPPL. 3 2004 REVIEWS IN UROLOGY Clinical Experience With Duloxetine In the past, sympathomimetic agents, such as ephedrine and phenylpropanolamine, as well as agents possessing a component of sympathomimetic activity, such as tricyclic antidepressants, have been used for the pharmacologic management of SUI based on their known effect on the internal sphincteric mechanism. However, little evidence exists to support these agents as efficacious for the treatment of SUI. Pelvic floor muscle training (PFMT) is often employed as a noninterven- affecting pudendal neural activity.3-5 The overall effect of the 5-HT/NE agonist, at least in feline and rodent models, is to exert suppression on parasympathetic activity (increasing bladder storage functionality) and to increase sympathetic and somatic neural activity, thus enhancing pelvic floor tonus. The stimulatory activity of duloxetine on the pudendal motor neurons and the resulting increase in striated urethral sphincter contractility are postulated to be the basis for the effect Given the large room for improvement in the management of SUI, an effective new treatment option with a reasonable tolerability profile has been eagerly anticipated. tional management technique for women with SUI, especially following parturition. However, long-term compliance in the ambulatory setting is imperfect. In one study, the majority (61%) of women prescribed PFMT for SUI trained only once weekly (15%), once monthly (3%), or not at all (43%).2 Despite this relatively poor compliance, PFMT has, to date, been the front-line nonsurgical intervention for SUI. Given the large room for improvement in the management of SUI, an effective new treatment option with a reasonable tolerability profile has been eagerly anticipated. Oral pharmacologic therapy for SUI is currently lacking. Because oral therapy is generally well accepted and easy to use, the development of an oral agent for SUI has the potential to substantively improve the treatment armamentarium for this condition. Duloxetine hydrochloride, a reuptake inhibitor of serotonin (5-HT) and norepinephrine (NE), has been shown to stimulate urethral sphincteric activity in a feline model, probably through activity within the sacral spinal cord of this agent in women with SUI. This article reviews the published phase 2/3 clinical trial data on duloxetine as an oral therapy for the treatment of SUI in women. Trial Design and Outcomes Reporting Various outcomes for SUI have been used to assess surgical and behavioral interventions. Variables used to evaluate SUI improvement (or cure, if attained) have included history/physical examination, stress testing in the outpatient setting with a fixed bladder volume (with or without standardized pad testing), and urodynamics, as well as general and disease-specific quality-of-life assessment. In many cases, a single outcome measure has been used to report the efficacy (or lack thereof) of SUI therapy. Recently, however, with the growing emphasis on more global outcomes reporting, combinations of subjective and objective measures have been used to more fully express incontinence outcomes. As recently demonstrated by Ward and Hilton,6 strikingly different magnitudes of success of SUI surgical interventions can be seen, dependent on methods used to assess overall impact of the surgery. In a randomized, controlled trial, the investigators compared Burch colposuspension with tension-free vaginal tape (TVT) for the treatment of SUI. Results demonstrated substantial differences among cure rates associated with colposuspension versus TVT as measured by urodynamic outcomes (67% vs 81%, respectively), a combination of pad testing and urodynamic outcomes (57% vs 66%, respectively), and the objective absence of any SUI (53% vs 59%, respectively). However, despite relatively low cure rates for both treatment groups, overall satisfaction was high: 82% with colposuspension and 85% with TVT. Therefore, the impact that improvement of SUI has on affected patients cannot be downplayed. However, the method used to estimate this improvement needs to be better defined. Preferably, a combination of subjective and objective outcomes should be used so as to best express overall global outcomes. The assessment of SUI outcomes for pharmacologic intervention should employ similar criteria, that is, criteria that are global in scope as well as reproducible over multiple time points during which drug ingestion occurs. In addition, to determine efficacy, it is crucial to obtain estimates not only of change in the frequency of episodes of urinary dysfunction (in this case, SUI events) but also of any effect on the magnitude of each individual event (ie, amount of urinary loss), because both of these parameters contribute to life disruption and bother experienced by the patient as a result of the disorder. Traditionally, patient diary reporting over a standard time frame—most often 1 week (although shorter periods may be used)—is employed to obtain the number of incontinence episodes. Magnitude of urinary loss is usually determined by VOL. 6 SUPPL. 3 2004 REVIEWS IN UROLOGY S57 Clinical Experience With Duloxetine continued standardized pad testing. Results are typically expressed as change in pad weight compared with baseline measurements, occurring with some level of predetermined activity over a defined time period (eg, 1-hour stress pad test or 24-hour ambulatory pad testing). Another variable to be considered when evaluating urinary loss is human behavior. Behavioral responses often are superimposed on urinary conditions as conscious or subconscious management methods used to decrease the bother associated with voiding dysfunction. Therefore, a crucial counterpoise evaluation to change in incontinence episode frequency and associated decrease in volume of urinary loss is any change in urinary habit associated with these decreases. This can be measured by number of volitional urinations or change in time between urinations, which represents a surrogate measure for change in frequency; obviously, if the time between urinations increases, the number of urinations in a 24-hour period will decrease. Outcomes for SUI pharmacologic interventions can be established based on these easily measurable criteria. Change in incontinence episode frequency (IEF) is a reproducible and overt manifestation of change in urinary incontinence. This parameter is utilized as a primary outcome measure in trials of all types of voiding dysfunction, including SUI. Another Table 1 Baseline Clinical Characteristics of Illness Severity* Characteristic Duloxetine Randomized, n 344 339 Caucasian, %† 89 91 Age, y† 52.3 (10.4) 53.3 (11.5) BMI, kg/m 29.5 (6.6) 29.2 (6.6) Prior continence surgery, including injection, %† 12.2 13.1 Perform PFMT, % 16.9 18.0 BDI-II score‡ 4.0 (5.4) 3.5 (5.1) 18.2 (14.3) 19.0 (14.6) 2† † IEF per week ‡ Voiding interval, min 145 (43) 148 (43) I-QOL score‡ 62.0 (20.2) 64.3 (17.7) Moderate or severely abnormal urinary tract function by PGI-S, %† 68.4 66.8 ‡ *Data are expressed as mean (SD) unless otherwise indicated. †All randomized patients. ‡Assessable patients (those with at least 1 postrandomization measurement). BMI, body mass index; PFMT, pelvic floor muscle training; BDI-II, Beck Depression Inventory–II; IEF, incontinence episode frequency; I-QOL, Incontinence Quality of Life; PGI-S, Patient Global Impression of Severity. Data from Dmochowski RR et al. J Urol. 2003;170:1259-1263.12 and are useful assessments of SUI. Unfortunately, because of the variability in volume of urinary loss associated with incontinent episodes, pad testing demonstrated extreme variability and did not correlate with any other outcome measure.7 Symptom improvement as subjectively perceived by the patient is also a Symptom improvement as subjectively perceived by the patient is a crucial treatment outcome and should roughly parallel the objective outcome criteria. objective measure of SUI treatment response is a reciprocal of IEF—mean time between voids (MTBV), which represents the actual time that elapses between voiding events. As was seen in the early-stage trials of duloxetine, IEF and MTBV have reproducibility S58 VOL. 6 SUPPL. 3 2004 Placebo crucial treatment outcome and should roughly parallel the objective outcome criteria. Multiple validated symptom scales exist for SUI, one of which is the Incontinence Quality of Life (I-QOL) scale. The I-QOL scale is a globally accepted, validated, disease-specific, REVIEWS IN UROLOGY 22-item instrument that evaluates the effects of urinary incontinence in 3 domains: avoidance and limiting behavior, social embarrassment, and psychosocial impact. Scores are calculated so that 100 represents the best and 0 represents the worst conditionspecific quality of life.8 This scale is considered by the World Health Organization to be the primary quality-of-life measure for SUI.9 I-QOL score, which was selected to be the main outcome variable for quality-oflife assessment in the duloxetine trials, correlated well with other objective responses seen in analysis of duloxetine therapy and is considered to be an independent outcome factor.1 The Patient Global Impression of Improvement (PGI-I) scale was also used to assess quality-of-life response to treatment in the duloxetine trials. Clinical Experience With Duloxetine Phase 2 Analysis The first large-scale, controlled, randomized trial comparing duloxetine with placebo for the treatment of SUI was conducted by Norton and colleagues.7 In this trial, several important general questions regarding pharmacologic therapy for SUI were addressed. In addition, the effect of duloxetine at several dosing levels was compared with placebo to establish a therapeutic index. Study inclusion criteria included women aged 18 to 65 years with a clinical diagnosis of bothersome SUI, 4 or more stress incontinent episodes per week, daytime voiding frequency of fewer than 8 voids per day, nocturnal frequency of fewer than 3 voids per day, absence of predominant urge incontinence symptoms, and a positive cough stress test and positive stress pad test after supine bladder filling to 400 mL. Women who could not tolerate bladder filling to 400 mL or who experienced early bladder sensation (first sensation of bladder filling <100 mL) were excluded from the study. Subjects who had previously undergone incontinence surgery or were receiving treatment for depression were also excluded. The study schema included a 2-week screening period followed by a 2-week no-drug lead-in period and a 2-week single-blind placebo lead-in period. Subsequently, subjects were randomized in a double-blind manner to 12 weeks of treatment with duloxetine, 20 mg/d, 40 mg/d, or 80 mg/d or placebo. For all study cohorts, treatment was administered as 2 identical capsules given twice daily. Subjects were seen at 4-week intervals through- Table 2 Median Percent Changes in Incontinence Episode Frequency (IEF) Median Percent IEF Change Subjects Duloxetine Placebo P Value ITT 50.0 27.5 <.001 Completers 50.0 29.3 <.001 51.8 25.0 <.001 All IEF ≥14 per week at baseline ITT ITT, intention-to-treat analysis of all randomized subjects with any postrandomization outcomes data; Completers, analysis of subjects who completed the 12-week blinded study period only. Data from Dmochowski RR et al. J Urol. 2003;170:1259-1263.12 out the treatment period. The primary efficacy variable was IEF as determined by real-time paper diaries, which were also used to record voiding intervals in minutes. Two 7-day diaries were completed before randomization. Three diaries were completed during the active treatment period, at 1 week before each monthly visit (4, 8, and 12 weeks). Secondary outcomes included the I-QOL and PGI-I scores, which were obtained at each postrandomization visit. The validated Patient Global Impression of Severity rating was completed only at baseline. Drug tolerability was determined Figure 1. Median percent decrease in incontinence episode frequency (IEF) for duloxetine and placebo at each of the 3 visits during the randomized, double-blind, 12-week, phase 3 duloxetine trial: the separation of duloxetine response from placebo response was significant at each visit, and the onset of effect was evident at the first postrandomization visit. Reproduced, with permission, from Dmochowski RR et al. J Urol. 2003;170:1259-1263.12 by evaluation of treatment-emergent adverse events, discontinuations due to adverse events, and clinical laboratory tests. Reports of adverse events were elicited as opposed to spontaneously reported. An event was considered treatment-emergent (ie, related to treatment) if it occurred for the first time or worsened during the double-blind treatment period. Variables were analyzed according to the intent to treat (ITT), which was the basis for all principal efficacy conclusions. For subjects who prematurely discontinued treatment, the last outcome measure was carried forward in the ITT analysis only. Van 0 4 weeks Percent Change in IEF This tool provides a simple singleitem measure of patient approbation of intervention. It has been validated in women with SUI and shows a similar magnitude of effect as the I-QOL scale.10 8 weeks 12 weeks –10 –20 –30 Placebo Duloxetine –40 –50 P < .001 VOL. 6 SUPPL. 3 2004 REVIEWS IN UROLOGY S59 Clinical Experience With Duloxetine continued Figure 2. Mean improvement in Incontinence Quality of Life (I-QOL) score for duloxetine and placebo at each of the 3 visits during the randomized, double-blind, 12-week, phase 3 duloxetine trial: the separation of duloxetine response from placebo response was significant at each visit, and the onset of effect was evident at the first postrandomization visit. Reproduced, with permission, from Dmochowski RR et al. J Urol. 2003;170: 1259-1263.12 Increase in I-QOL Score (Points) 14 12 10 P < .001 8 6 Placebo Duloxetine 4 2 0 4 weeks 8 weeks Elteren’s test11 was used to assess IEF median percent changes between the 2 groups stratified by severity at last visit compared with baseline. Analysis of covariance (ANCOVA) modeling was used to assess mean changes in I-QOL scores. Similar analysis was used to interpret changes in mean voiding intervals. IEF decreased from baseline to end point by 40% in the placebo group. Last visit analyses showed reductions in IEF of 44%, 59%, and 58% for subjects in the duloxetine, 20 mg/d, 40 mg/d, and 80 mg/d groups, respectively. In addition, subjects who received duloxetine experienced significant, dose-dependent improvements in quality of life. Duloxetine was generally well tolerated. Nausea was the most common reason for discontinuation and appeared to be dose-related.7 An interesting sidebar to this study was the development of a clinical algorithm for identifying SUI with condition-specific accuracy without the need for invasive urodynamics. Using a clinical diagnostic paradigm for SUI that included history and physical examination, cough stress testing, and stress pad testing, the authors observed a 92% diagnostic concordance with results of multi- S60 VOL. 6 SUPPL. 3 2004 12 weeks channel urodynamic studies.7 Based on these results, this diagnostic methodology was used to determine inclusion in subsequent trials evaluating duloxetine as a pharmacologic intervention for SUI. Phase 3: North American Trial Data The phase 3 North American duloxetine trial was similar to the phase 2 study in its design and inclusion/exclusion criteria. The phase 3 study did, however, include women older than 65 years and those who had previously undergone incontinence surgery. In addition, subjects must have reported 7 or more SUI episodes weekly.12 Several additional issues were assessed in the phase 3 trial. Patients were stratified by severity for outcomes reporting using IEF; condition severity was described as an IEF of less than 14 or 14 or more episodes per week. In addition, subjects completed the Beck Depression InventoryII (BDI-II) at randomization and during the last treatment visit. Van Elteren’s test11 was used to assess the percent changes in IEF between the 2 groups (active drug vs placebo) at last visit compared with baseline. Because of extreme outliers in both treatment arms (placebo, +700%; duloxetine, +550%), median data were used to report results. ANCOVA modeling was used to assess mean changes in I-QOL scores. Similar analysis was used to interpret changes in mean voiding intervals. In this double-blind, placebocontrolled trial, 683 women aged 22 to 84 years were randomized to receive duloxetine (n = 344) or placebo (n = 339). Eighty-seven percent of Table 3 Incontinence Quality of Life (I-QOL) Score Changes I-QOL Improvement, Mean (SD) Duloxetine Placebo P Value ITT 11.1 (14.8) 6.8 (13.8) <.001 Completers 12.1 (15.3) 7.2 (14.2) <.001 Avoidance/limiting behavior 11.1 (15.8) 7.1 (14.8) <.001 Psychosocial 10.2 (15.5) 5.7 (14.6) <.001 Social embarrassment 12.4 (19.8) 8.4 (18.6) .02 I-QOL total score I-QOL subscales (ITT) Completers, analysis of subjects who completed the 12-week blinded study period only; ITT, intention-to-treat analysis of all randomized subjects with any postrandomization outcomes data. Data from Dmochowski RR et al. J Urol. 2003;170:1259-1263.12 REVIEWS IN UROLOGY Clinical Experience With Duloxetine Table 4 Treatment-Emergent Adverse Events Occurring in at Least 5% of Subjects Randomized to Duloxetine Event Rate, % Event Duloxetine Placebo P Value Nausea 22.7 2.1 <.001 Fatigue 14.8 3.8 <.001 Insomnia 14.2 2.4 <.001 Dry mouth 12.2 0.9 <.001 Constipation 9.6 2.1 <.001 Somnolence 8.7 0.3 <.001 Dizziness 7.6 2.4 .002 Headache 7.3 3.5 .04 Diarrhea 6.1 2.7 .04 Table 5 Discontinuation for Treatment-Emergent Adverse Events in More Than 1% of Subjects Randomized to Duloxetine Discontinuation Rate, % Event Duloxetine Placebo P Value Nausea 6.4 0 Fatigue 2.6 0.3 Insomnia 2.0 0.3 .07 Somnolence 2.0 0.3 .07 Dizziness 1.5 0.3 .22 Blurred vision 1.2 0 .12 subjects who received placebo completed the trial, compared with 69% of those who received duloxetine—a significant difference attributable to side effects experienced by subjects in the duloxetine group. Approximately two thirds of the study subjects considered their urinary tract function to be moderately or severely affected by their condition, and these women averaged 2 or more incontinence episodes daily. Mean baseline BDI-II scores were well below the level associated with symptomatic depression (≥17) (Table 1). <.001 .02 At study completion, the duloxetine group had a median absolute decrease in IEF of 7 episodes per week, compared with a decrease of 3 episodes per week in the placebo group (P < .001). Baseline incontinence magnitude had no effect on duloxetine treatment response (Table 2). The IEF changes with active treatment were observed at the first postrandomization assessment visit (4 weeks) and noted throughout the entire 12-week treatment period (Figure 1). These changes occurred despite an increase in average voiding interval compared with the placebo group (20.0 vs 1.7 minutes; P < .001). Furthermore, 51.4% of subjects in the duloxetine group had a decrease of 50% to 100% in IEF, compared with only 33.5% of subjects in the placebo group (P < .001). On the last diary evaluation, 10.5% of subjects in the duloxetine group had no incontinence episodes, compared with 5.9% of those in the placebo group (P < .05). Quality-of-life changes were also apparent by the first postrandomization visit and sustained throughout the trial (Figure 2); these changes were statistically superior in the subjects who received duloxetine compared with placebo. Significant improvements were observed in the 3 subscales of the I-QOL correlated with condition-specific response (Table 3). Results of the PGI-I scale demonstrated that 62.0% of subjects who received duloxetine rated their condition improved, compared with 39.6% of those who received placebo (P < .001). In this phase 3 analysis, the outcome indicators of IEF, MTBV, I-QOL score, and PGI-I were closely correlated and similar in magnitude of effect. The expanded evaluation of these outcome factors in the 683 participants in this trial demonstrated the reproducibility of these metrics for the symptom-specific evaluation of duloxetine for the treatment of SUI. Tolerability-related adverse events occurred in significantly more women in the duloxetine group than in the placebo group (74% vs 50%; P < .001), resulting in a discontinuation rate that was significantly higher for the duloxetine group (24.1% vs 4.1%; P < .001). Nausea was the most common side effect experienced by the women who received duloxetine, occurring in 22.7% of subjects in this group, compared with 2.1% of those in the placebo group (P < .001) (Table 4). Nausea resulted in a 6.4% discontinuation rate for patients in the VOL. 6 SUPPL. 3 2004 REVIEWS IN UROLOGY S61 Clinical Experience With Duloxetine continued duloxetine group (Table 5). Of the women who experienced nausea, 91% reported the symptom within 4 weeks of initiation of therapy and most within the first 2 days. In the majority of these subjects (87%), the nausea was mild to moderate in severity; in 81%, the symptom resolved within 1 month. Of the subjects who developed nausea while receiving duloxetine, 74% completed the study. Anticholinergic-like adverse events among subjects in the duloxetine group included dry mouth (n = 31) and constipation (n = 30). However, no patient who experienced dry mouth discontinued as a result of this adverse effect, and only 1 patient who experienced constipation consequently discontinued. Other serious adverse events did not occur with significantly increased frequency in the duloxetine group compared with the placebo group. Discussion The results of duloxetine trials indicate that this agent represents an efficacious option for the pharmacologic treatment of SUI. The disparity between the number of women who have symptomatic SUI and the number who ultimately receive surgical therapy indicates that the vast majority do not feel that their condition warrants surgical intervention. Oral therapy for SUI expands the options available for these patients and may result in more women presenting for treatment in earlier stages of the disorder. Nonsurgical interventions, such as PFMT, are currently used to manage bothersome SUI. However, in general practice, long-term compliance with PFMT regimens is poor.2 Ideally, PFMT could be utilized to provide a basis for subsequent intervention and to forestall more invasive treatments. However, motivation and consistent inclusion of apropos regimens appear to be possible for only a minority of women. The central mode of action of duloxetine—that is, sacral cord pudendal motor nucleus stimulation of the striated urethral sphincter, rather than a peripheral mode of action to stimulate smooth muscle of the urethra— provides a novel mechanism for stimulation of pelvic floor muscle activity. In clinical trials, duloxetine has been shown to be significantly more effective for the treatment of SUI than placebo. With both ITT and completers analyses, a reduction in incontinence episodes of 50% to 100% was achieved in more than half of the women who received duloxetine, with 10.5% of women experiencing no incontinence episodes. This effect was noted early in treatment, persisted over the 3-month study duration, and was independent of initial incontinence severity. The IEF response with duloxetine treatment was observed despite an increase in the average voiding interval by 20 minutes, excluding any behavioral contribution to the salubrious results associated with drug ingestion.12 Similarly, quality-of-life improvements occurred within the first month of therapy and persisted throughout the trial. The significant improvement seen in all I-QOL subscales allows some characterization of the type of improvements experienced with duloxetine therapy. Improvements in the psychosocial impact subscale reflect an increased ability to leave home and/or perform activities previously restricted by urine leakage, Main Points • Although pelvic floor muscle training is the front-line nonsurgical therapy for stress urinary incontinence (SUI) in women, compliance is poor. Therefore, another option for SUI that imparts efficacy with a reasonable tolerability profile has been eagerly anticipated. • Duloxetine, which is currently under review by the Food and Drug Administration, is a dual reuptake inhibitor of serotonin and norepinephrine. Its stimulatory activity on the pudendal motor neurons and the resulting increase in striated urethral sphincter contractility are postulated to be the basis for drug effect in women with SUI. • Incontinence episode frequency (IEF) represents a reproducible and overt manifestation of urinary incontinence change. This parameter is utilized as a primary outcome measure in trials of all types of voiding dysfunction, including SUI. • The Incontinence Quality of Life (I-QOL) scale is a validated, disease-specific, 22-item instrument that evaluates the effects of urinary incontinence in 3 domains: avoidance and limiting behavior, social embarrassment, and psychosocial impact. • Published phase 2/3 clinical trial data demonstrate that, compared with placebo, duloxetine significantly reduces IEF and improves I-QOL score in women with SUI. These effects were evident early after initiation of treatment and were maintained throughout the trial duration. • The disparity between the number of women with symptomatic SUI and the number who ultimately receive surgical therapy indicates that the vast majority of patients do not feel that their situation warrants surgical intervention. Effective oral therapy for SUI expands the options available for these patients and may result in more women presenting for therapy in earlier stages in their condition. S62 VOL. 6 SUPPL. 3 2004 REVIEWS IN UROLOGY Clinical Experience With Duloxetine as well as an ability to enjoy life more, feel less helpless, and worry less about wearing certain types of clothing or participating in sexual activity. Improvements in the avoidance and limiting behavior subscale reflect an increased independence from the need to plan every activity around the availability of a toilet, as well as an increased ability to perform physical activities and avoid fluid restrictions. Improvements in the social embarrassment subscale reflect less worry over odor, embarrassment, or the possibility that incontinence will worsen over time. The small percentage of subjects who had appreciable symptoms of depression had no significant change in I-QOL score, demonstrating that an improvement in unrecognized depression as a consequence of the known antidepressant effect of duloxetine was not an important factor in the overall quality-of-life improvements achieved with duloxetine therapy. The lack of a significant change in mean BDI-II scores indicates that duloxetine did not exert a significant effect on mood in the nondepressed subjects, who comprised 96% of women in the duloxetine group.12 The PGI-I scale adds perspective to the IEF and I-QOL improvements. Yalcin and Bump10 have previously reported that women with SUI who rate themselves as improved using the PGI-I scale have decreases in their IEF of 43.4% or greater and improvements in their I-QOL scores of 6.4 points or more. In contrast to results seen with placebo, duloxetine-associated median reduction in IEF (50%) and mean improvement in I-QOL score (11.1 points) are well above these threshold values.12 Significantly more patients in the duloxetine group experienced treatment-emergent adverse events than in the placebo group, and 24% of subjects who received duloxetine discontinued therapy because of side effects. Nausea was the most frequent adverse event, with 6.4% of all duloxetine subjects discontinuing because of nausea. Nausea tended to begin early after the initiation of duloxetine; most cases were mild or moderate, did not worsen, and resolved within 1 week to 1 month. Seventy-four percent of women who experienced treatment-emergent nausea completed the study. Dry mouth and constipation were not uncommon but together resulted in only a single discontinuation. Tolerability issues result in early discontinuation of duloxetine in some patients. The proven resolution of most problematic effects by 4 weeks of drug ingestion implies that adverse events tend to occur early, are not progressive and, in the majority of patients, resolve. Given the fact that drug efficacy is seen within a relatively short time after drug ingestion, adverse events can be weighed against this rapid onset of effect, providing bothered women with a reasonable risk/benefit ratio. Further experience with duloxetine has now been accrued in additional large-scale trials, and initial results indicate experiences similar to those described in this article. To date, the data support the overall efficacy of duloxetine therapy for bothersome SUI. Duloxetine, which is currently under review by the Food and Drug Administration, appears to provide a unique treatment option for women with problematic SUI. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Patrick DL, Martin ML, Bushnell DM, et al. Quality of life of women with urinary incontinence: further development of the incontinence quality of life instrument (I-QOL). Urology. 1999;53:71-76. Lagro-Janssen T, Van Weel C. Long-term effect of treatment of female incontinence in general practice. Br J Gen Prac. 1998;48:1735-1738. Espey MJ, Du H-J, Downie JW. Serotonergic modulation of spinal ascending activity and sacral reflex activity evoked by pelvic nerve stimulation in cats. Brain Res. 1998;798:101108. Thor KB, Hisamitsu T, de Groat WC. Unmasking of a neonatal somatovesical reflex in adult cats by the serotonin autoreceptor agonist 5methoxy-N,N-dimethyltryptamine. Brain Res Dev Brain Res. 1990;54:35-42. Danuser H, Thor KB. Spinal 5-HT2 receptormediated facilitation of pudendal nerve reflexes in the anaesthetized cat. Br J Pharmacol. 1996;118:150-154. Ward K, Hilton P, on behalf of the United Kingdom and Ireland Tension-free Vaginal Tape Trial Group. Prospective multicentre randomized trial of tension-free vaginal tape and colposuspension as primary treatment for stress incontinence. BMJ. 2002;325:67-70. Norton PA, Zinner NR, Yalcin I, Bump RC. Duloxetine versus placebo in the treatment of stress urinary incontinence. Am J Obstet Gynecol. 2002;187:40-48. Wagner TH, Patrick DL, Bavendam TG, et al. Quality of life of persons with urinary incontinence: development of a new measure. Urology. 1996;47:67-72. Donovan JL, Badia X, Corcos J, et al. Symptom and quality of life assessment. In: Abrams P, Cardozo L, Khoury S, Wein A, eds. Incontinence. Plymouth, UK: Plymbridge Distributors Inc; 2002:267-316. Yalcin I, Bump RC. Validation of two global impression questionnaires for incontinence. Am J Obstet Gynecol. 2003;189:98-101. van Elteren PH. On the combination of independent two-sample tests of Wilcoxon. Bull Int Stat Inst. 1960;37:1-13. Dmochowski RR, Miklos JR, Norton PA, for the Duloxetine Urinary Incontinence Study Group. Duloxetine versus placebo for the treatment of North American women with stress urinary incontinence. J Urol. 2003;170:1259-1263. VOL. 6 SUPPL. 3 2004 REVIEWS IN UROLOGY S63