ATLAS: Ambulatory Treatments for Leakage Associated with Stress Incontinence
ATLAS was a nine-site, assessor-blinded randomized trial in women seeking nonsurgical care for stress-predominant urinary incontinence that compared a fitted continence pessary, behavioral pelvic-floor therapy, and their combination, finding at three months that behavioral therapy improved the no-bothersome-symptoms endpoint versus pessary while the global-improvement comparison was not significant and combination therapy failed its prespecified superiority rule.[1]
- Design and population
- Nine-site, assessor-blinded, three-arm randomized superiority trial. Women aged at least 18 years with stress-only or stress-predominant mixed urinary incontinence seeking nonsurgical care.
- Arms
- Behavioral therapy: pelvic-floor muscle training and continence strategies versus physician- or nurse-fitted continence ring or dish pessary versus continence pessary plus behavioral therapy
- Report and follow-up
- 2010 · Primary outcomes at 3 months; repeated assessments at 6 and 12 months
| Endpoint | Results | Effect | Finding |
|---|---|---|---|
| 3-month much or very much better; behavioral minus pessary risk difference in percentage points | 72/146 (49.3%) vs 59/149 (39.6%) | difference 9.7; P 0.09 | Not met |
| 3-month PFDI UDI-stress success; behavioral minus pessary risk difference in percentage points | 71/146 (48.6%) vs 49/149 (32.9%) | difference 15.7; P 0.006 | Met |
| 3-month PGI-I; combined therapy had to exceed both single treatments | 80/150 (53.3%) vs 72/146 (49.3%) vs 59/149 (39.6%) | — | Not met |
| 3-month PFDI UDI-stress success; combined therapy had to exceed both single treatments | 66/150 (44.0%) vs 71/146 (48.6%) vs 49/149 (32.9%) | — | Not met |
| 12-month much or very much better; intention to treat | 48/146 (32.9%) vs 47/149 (31.5%) vs 49/150 (32.7%) | — | Descriptive |
| 12-month PFDI UDI-stress success; intention to treat | 59/146 (40.4%) vs 52/149 (34.9%) vs 49/150 (32.7%) | — | Descriptive |
Harms: Vaginal discharge by 3 months: 6% vs 16% vs 9%; Unexpected urologic adverse events: 2/146 vs 3/149 vs 5/150
Design
Eligible women had stress-only or stress-predominant mixed incontinence. Randomization was stratified by incontinence type and the frequency of episodes on a seven-day diary. Participants and interventionists knew the assignment; staff performing examinations and collecting telephone outcomes were blinded. Withdrawal or use of another incontinence treatment counted as failure in the intention-to-treat analysis.[1]
The coprimary outcomes at three months were a Patient Global Impression of Improvement (PGI-I) rating of “much better” or “very much better” and absence of bothersome stress symptoms on the Pelvic Floor Distress Inventory (PFDI) stress subscale. Combination therapy qualified as superior only if it significantly exceeded both single treatments. The final PGI-I comparison used a 0.0492 significance level after a planned interim analysis; the other comparisons used 0.05.[1]
Results
On PGI-I, 72/146 women (49.3%) assigned to behavioral therapy and 59/149 (39.6%) assigned to pessary reported being much or very much better at three months, a 9.7-percentage-point difference (P = 0.09). The PFDI stress endpoint was met by 71/146 (48.6%) and 49/149 (32.9%), respectively, a 15.7-percentage-point difference (P = 0.006).[1]
In the combination arm, 80/150 (53.3%) met the PGI-I endpoint and 66/150 (44.0%) had no bothersome stress symptoms. Combination exceeded pessary on those measures (P = 0.02 and P = 0.05) but did not exceed behavioral therapy (P = 0.49 and P = 0.42), so neither comparison met the trial's combination-superiority rule. Treatment satisfaction at three months was 110/146 (75.3%) with behavioral therapy, 94/149 (63.1%) with pessary, and 118/150 (78.7%) with combination; the behavioral-versus-pessary comparison had P = 0.03 in the Results and Table 3 footnote.[1]
At 12 months, global improvement was reported by 48/146 (32.9%), 47/149 (31.5%), and 49/150 (32.7%), respectively. No bothersome stress symptoms were reported by 59/146 (40.4%), 52/149 (34.9%), and 49/150 (32.7%). Neither endpoint differed significantly among groups at that assessment.[1]
A secondary analysis of 295 women in the pessary and behavioral arms found improvement in pelvic symptom distress and condition-specific quality of life in both arms at three months, without a significant between-group difference in the analyzed scores.[2] An adjusted secondary analysis among 445 women across all three arms associated no prior incontinence surgery with greater odds of stress-symptom success (OR 3.15, 95% CI 1.04–9.53) and fewer than 14 weekly episodes at baseline with greater odds of treatment satisfaction (OR 1.97, 95% CI 1.21–3.19).[3] The investigators found no significant interaction between these predictors and treatment assignment.[3]
Surgery and treatment delivery
There was no surgical intervention. The main trial allowed up to three visits at one- to two-week intervals to obtain a usable continence ring or dish fit. Trained nurses, nurse practitioners, and physical therapists delivered behavioral treatment in four standardized visits about two weeks apart. They prescribed daily pelvic-floor exercises and taught active muscle strategies for stress and urge leakage, followed by an individualized home program after the eight-week supervised period.[1]
In the prespecified fitting analysis of the pessary-containing arms, 235/255 women (92%) with a recorded final outcome were fitted successfully. Trained nurses, nurse practitioners, or physical therapists selected a ring or dish; a usable fit allowed a finger between device and vaginal wall and retained the device while the woman stood, coughed, and strained. Of the successful fittings, 122 used a ring and 113 a dish. POP-Q total vaginal length, point C, and point D did not predict final device diameter.[4]
By three months, withdrawal had occurred in 39/149 (26.2%) assigned to pessary, 22/146 (15.1%) assigned to behavioral therapy, and 18/150 (12.0%) assigned to combination. Twelve pessary-arm withdrawals were attributed to inadequate fit. At one year, 45% reported continued pessary use and 57% reported continued pelvic-floor exercises.[1]
Adverse events
Vaginal discharge was reported at three months by 16% in the pessary arm, 6% in the behavioral arm, and 9% in the combination arm. The report also recorded vaginal yeast infection in 7% overall and no vaginal erosions. Ten unexpected urologic events comprised seven urinary tract infections, two possible infections, and one possible urinary retention event; the counts were 3/149 with pessary, 2/146 with behavioral therapy, and 5/150 with combination.[1]
Limitations
The interventions could not be blinded to participants, and early withdrawal was greatest with pessary. Missing outcomes and withdrawals were generally treated as failures, making continued treatment use important when interpreting the 12-month estimates. Combination-arm participants could remain in follow-up while using only one component. The behavioral regimen comprised four supervised visits, so its results describe that treatment intensity.[1]
In conversation
Guidelines and approvals
The European Association of Urology 2026 female lower urinary tract symptoms guideline summarizes evidence comparing pelvic-floor muscle training with continence pessaries, including the ATLAS comparison through its cited evidence review. It recommends supervised intensive pelvic-floor muscle training for at least three months as first-line therapy for women with stress or mixed urinary incontinence .[5]
Report history
| Published | Report | Cutoff | Headline results | Citation |
|---|---|---|---|---|
| 2009-06-16 | Continence pessary fitting analysis · secondary | — | Final recorded fitting outcome in pessary-assigned arms: 235/255 women (92%) with a recorded final fit outcome were fitted; the paper describes 266 assigned to the pessary-containing arms.; — Descriptive Association of POP-Q measurements with final pessary diameter: Total vaginal length, point C, and point D did not predict final diameter; mean vaginal length was 9.6 cm with successful fit versus 8.8 cm with unsuccessful fit (P<0.01).; — Descriptive | Nager CW, Richter HE, Nygaard I, et al. "Incontinence pessaries: size, POPQ measures, and successful fitting." *Int Urogynecol J.* 2009;20(9):1023-1028. doi:10.1007/s00192-009-0866-1 DOI |
| 2010 | Primary trial and 12-month outcomes · primary | — | 3-month much or very much better; behavioral minus pessary risk difference in percentage points: 72/146 (49.3%) vs 59/149 (39.6%); difference 9.7; P 0.09 Not met 3-month PFDI UDI-stress success; behavioral minus pessary risk difference in percentage points: 71/146 (48.6%) vs 49/149 (32.9%); difference 15.7; P 0.006 Met 3-month PGI-I; combined therapy had to exceed both single treatments: 80/150 (53.3%) vs 72/146 (49.3%) vs 59/149 (39.6%); — Not met 3-month PFDI UDI-stress success; combined therapy had to exceed both single treatments: 66/150 (44.0%) vs 71/146 (48.6%) vs 49/149 (32.9%); — Not met 3-month satisfied or somewhat satisfied; behavioral minus pessary risk difference in percentage points: 110/146 (75.3%) vs 94/149 (63.1%) vs 118/150 (78.7%); difference 12.2; P 0.03 3-month 7-day diary response: 68/146 (46.6%) vs 69/149 (46.3%) vs 80/150 (53.3%); — 12-month much or very much better; intention to treat: 48/146 (32.9%) vs 47/149 (31.5%) vs 49/150 (32.7%); — Descriptive 12-month PFDI UDI-stress success; intention to treat: 59/146 (40.4%) vs 52/149 (34.9%) vs 49/150 (32.7%); — Descriptive | Richter HE, Burgio KL, Brubaker L, et al. "Continence pessary compared with behavioral therapy or combined therapy for stress incontinence: a randomized controlled trial." *Obstet Gynecol.* 2010;115(3):609-617. doi:10.1097/AOG.0b013e3181d055d4 DOI |
| 2012 | Pelvic symptom and quality-of-life analysis · secondary | — | 3-month change, pessary versus behavioral therapy: improved vs improved; — Descriptive | Kenton K, Barber M, Wang L, et al. "Pelvic floor symptoms improve similarly after pessary and behavioral treatment for stress incontinence." *Female Pelvic Med Reconstr Surg.* 2012;18(2):118-121. doi:10.1097/SPV.0b013e31824a021d DOI |
| 2012 | Predictors of success and satisfaction · secondary | — | Adjusted odds of 3-month UDI-stress success, no prior surgery versus prior surgery: Adjusted association among 445 analyzed women across all three treatment arms; no significant treatment interaction.; OR 3.15; CI 1.04–9.53; P 0.04 Adjusted odds of 3-month satisfaction, fewer than 14 versus at least 14 episodes per week: Adjusted association among 445 analyzed women across all three treatment arms; no significant treatment interaction.; OR 1.97; CI 1.21–3.19; P 0.01 | Schaffer J, Nager CW, Xiang F, et al. "Predictors of success and satisfaction of nonsurgical therapy for stress urinary incontinence." *Obstet Gynecol.* 2012;120(1):91-97. doi:10.1097/AOG.0b013e31825a6de7 DOI |
References
1. Richter HE, Burgio KL, Brubaker L, et al. "Continence pessary compared with behavioral therapy or combined therapy for stress incontinence: a randomized controlled trial." Obstet Gynecol. 2010;115(3):609-617. doi:10.1097/AOG.0b013e3181d055d4
2. Kenton K, Barber M, Wang L, et al. "Pelvic floor symptoms improve similarly after pessary and behavioral treatment for stress incontinence." Female Pelvic Med Reconstr Surg. 2012;18(2):118-121. doi:10.1097/SPV.0b013e31824a021d
3. Schaffer J, Nager CW, Xiang F, et al. "Predictors of success and satisfaction of nonsurgical therapy for stress urinary incontinence." Obstet Gynecol. 2012;120(1):91-97. doi:10.1097/AOG.0b013e31825a6de7
4. Nager CW, Richter HE, Nygaard I, et al. "Incontinence pessaries: size, POPQ measures, and successful fitting." Int Urogynecol J. 2009;20(9):1023-1028. doi:10.1007/s00192-009-0866-1
5. European Association of Urology. EAU Guidelines on Management of Non-neurogenic Female Lower Urinary Tract Symptoms. Limited update March 2026; section 4.2.4.a.3.g.