OPTIMAL: Operations and Pelvic Muscle Training in the Management of Apical Support Loss

Bottom line Negative

In women having vaginal apical prolapse repair with a midurethral sling, bilateral uterosacral ligament suspension and unilateral sacrospinous ligament fixation had similar 2-year composite surgical success (64.5% versus 63.1%) and serious adverse events; perioperative behavioral therapy with pelvic floor muscle training did not improve outcomes.

Neither suspension was superior at 2 years; behavioral therapy added no benefit; 5-year failure estimates did not differ significantly.

OPTIMAL was a multicenter, masked, 2×2 factorial randomized trial in 374 women having vaginal surgery with a retropubic midurethral sling for apical prolapse and stress urinary incontinence at 9 US centers. It compared bilateral uterosacral ligament suspension (ULS) with unilateral sacrospinous ligament fixation (SSLF) and, in a second randomization, perioperative behavioral therapy with pelvic floor muscle training (BPMT) with usual care. At 2 years composite surgical success was 100/155 (64.5%) after ULS and 94/149 (63.1%) after SSLF (adjusted OR 1.1, 95% CI 0.7–1.7), and BPMT did not improve urinary symptoms at 6 months or prolapse symptoms or anatomic failure at 2 years.^[1][2]

Design and population

Multicenter, masked, 2×2 factorial randomized superiority trial. 374 women with stage 2–4 apical or uterine prolapse, bulge symptoms and objectively demonstrated stress urinary incontinence who were having vaginal surgery with a retropubic midurethral sling at 9 US centers.

Arms

Bilateral uterosacral ligament suspension versus unilateral sacrospinous ligament fixation

Report and follow-up

2014 · 24 months; surgical-success data for 304 of 374 randomized women (155 uterosacral, 149 sacrospinous)

Endpoint	Results	Effect	Finding
2-year composite surgical success (apical descent no more than one-third of vaginal length, no anterior or posterior wall beyond the hymen, no bothersome bulge, no retreatment); corrected analysis	100/155 (64.5%) vs 94/149 (63.1%)	OR 1.1; CI 0.7–1.7; P 0.84	Not met
Adjusted mean change from baseline in Urinary Distress Inventory score (range 0–300) at 6 months: −94.6 with BPMT versus −87.9 with usual care	Mean difference, BPMT minus usual care. Minimum important difference for the score was 11 points.	difference -6.7; CI -19.7–6.2; P 0.31	Not met
Adjusted mean change from baseline in Pelvic Organ Prolapse Distress Inventory score (range 0–300) at 24 months: −73.3 with BPMT versus −65.2 with usual care	Mean difference, BPMT minus usual care.	difference -8; CI -22.1–6.1; P 0.26	Not met
24-month anatomic failure (apical descent beyond one-third of vaginal length, anterior or posterior wall beyond the hymen, or retreatment): 37/153 (24.2%) with BPMT versus 42/164 (25.6%) with usual care; corrected analysis	The table row appears under the surgical columns in the online text; the counts are the behavioral comparison.	OR 0.9; CI 0.6–1.6; P 0.82	Not met

Harms: Serious adverse events: 31/188 (16.5%) vs 31/186 (16.7%); Neurologic pain requiring additional treatment: 13/188 (6.9%) vs 23/186 (12.4%); Intraoperative ureteral obstruction: 6/188 (3.2%) vs 0/186

NCT00597935 · Report DOI

Design

Women were 18 years or older and having vaginal surgery for stage 2–4 prolapse (uterine or vaginal descent 1 cm proximal to the hymen or beyond). Eligibility required bulge symptoms, descent of the uterus or vaginal apex at least halfway into the vagina, stress incontinence symptoms and objective demonstration of stress incontinence by office or urodynamic testing in the previous 12 months. Mean age was 57 years, and 100 of 374 women (48/188 ULS, 52/186 SSLF) had a prior hysterectomy.^[1]

Each woman had two randomizations by sealed, opaque envelopes. BPMT versus usual care was assigned before surgery and stratified by site; ULS versus SSLF was assigned in the operating room and stratified by surgeon (31 surgeons) and concomitant hysterectomy. Participants were masked to the surgical assignment, surgeons to the behavioral assignment, and outcome assessors, including the examiners who performed Pelvic Organ Prolapse Quantification (POP-Q) examinations and the central telephone interviewers, to both.^[1]

The primary surgical outcome was composite success at 2 years, defined as the absence of all of the following: apical descent more than one-third of vaginal length, anterior or posterior wall descent beyond the hymen, bothersome vaginal bulge symptoms, and retreatment by surgery or pessary. The primary behavioral outcomes were Urinary Distress Inventory score at 6 months and Pelvic Organ Prolapse Distress Inventory score and anatomic failure at 24 months. The investigators judged in advance that a difference in surgical success under 15 percentage points would not change practice; enrollment stopped at 418 women (374 randomized) because early loss to follow-up was lower than expected. The trial was not powered for interactions between the two interventions.^[1]

The extension (E-OPTIMAL) followed participants who completed the 2-year visit through 5 years with annual examinations and telephone interviews. Its primary surgical outcome was time to surgical failure: apical descent more than one-third of vaginal length, anterior or posterior wall beyond the hymen, bothersome bulge symptoms, or retreatment, with a participant who met any criterion remaining a failure at later time points.^[3]

Results

At 2 years composite success was 100/155 (64.5%) with ULS and 94/149 (63.1%) with SSLF (adjusted OR 1.1, 95% CI 0.7–1.7; unadjusted difference 1.4 percentage points, 95% CI −9.4 to 12.2). The superiority comparison was not significant, and the interval excludes a difference of 15 percentage points in either direction. Bothersome bulge symptoms were present in 25/151 (16.6%) after ULS and 30/154 (19.5%) after SSLF, and retreatment by surgery or pessary occurred in 8/161 (5.0%) and 8/155 (5.2%). Overall, apical descent beyond one-third of vaginal length was found in 16/155 (10.3%) and 21/152 (13.8%).^[1][2]

In an exploratory interaction analysis, apical descent among women assigned to usual care was 4.9% after ULS versus 15.6% after SSLF (adjusted OR 0.3, 95% CI 0.1–0.8); among women assigned to BPMT it was 16.2% versus 12.0% (adjusted OR 1.4, 95% CI 0.6–3.7).^[2]

A 2015 correction followed data-entry errors in POP-Q values at one of the nine sites, where positive values had been entered for negative ones and some successes were classified as failures. The primary surgical outcome changed for 16 of 307 participants, and the authors reported that the conclusions were unaffected. The figures above are the corrected ones; the author manuscript in PubMed Central retains the earlier values (93/157, 59.2% versus 92/152, 60.5%).^[2]

BPMT did not improve the behavioral outcomes. The adjusted mean change in Urinary Distress Inventory score at 6 months was −94.6 with BPMT and −87.9 with usual care (difference −6.7, 95% CI −19.7 to 6.2; the score's minimum important difference is 11 points). At 24 months the change in Pelvic Organ Prolapse Distress Inventory score was −73.3 and −65.2 (difference −8.0, 95% CI −22.1 to 6.1), and anatomic failure was 37/153 (24.2%) with BPMT versus 42/164 (25.6%) with usual care (adjusted OR 0.9, 95% CI 0.6–1.6).^[1][2]

At 5 years the estimated probability of surgical failure was 61.5% after ULS and 70.3% after SSLF (adjusted difference −8.8 percentage points, 95% CI −24.2 to 6.6); observed first failures were 94/145 (64.8%) and 104/146 (71.2%), with a median time to failure of 2.6 and 1.8 years. Estimated anatomic failure was 47.5% versus 61.8% (difference −14.3 points, 95% CI −32.1 to 3.5), bothersome bulge 37.4% versus 41.8% (difference −4.4 points, 95% CI −21.3 to 12.5) and retreatment for prolapse 11.9% versus 8.1% (difference 3.9 points, 95% CI −3.8 to 11.5). In the behavioral comparison, estimated anatomic failure was 45.6% with BPMT and 47.2% with usual care (difference −1.6 points, 95% CI −21.2 to 17.9), and prolapse symptom scores improved by 59.4 and 61.8 points (mean difference 2.4, 95% CI −13.7 to 18.4).^[3]

Symptoms stayed improved while failure accumulated. Prolapse distress scores fell by 67.6 points after ULS and 74.2 after SSLF at 5 years (mean difference 6.6, 95% CI −9.5 to 22.7), both beyond the 11-point minimum important difference. The extension's modeled 2-year failure (45.7% after ULS, 52.0% after SSLF) is higher than the 35.5% and 36.9% that follow from the 2-year success rates; the extension carried a first failure forward, and 65 of 198 surgical failures (32.8%) met the composite success definition at their last visit without further treatment.^[1][3]

Surgery and treatment delivery

SSLF was unilateral, a modification of the Michigan four-wall technique; ULS was bilateral, a modification of the Shull technique. Both used four sutures (two permanent and two delayed absorbable). Women with uterine prolapse had a vaginal hysterectomy, so the trial did not test uterine preservation. Biologic or synthetic graft for the prolapse repair was not allowed. All but 3 of the 374 women had a retropubic tension-free vaginal tape sling, and other concomitant procedures were at the surgeon's discretion. Operative variables such as estimated blood loss, operative time and length of stay did not differ significantly between the surgical groups.^[1]

The BPMT program consisted of one visit 2–4 weeks before surgery and four after it (2, 4–6, 8 and 12 weeks), each with individualized pelvic floor muscle training and education on behavioral strategies for urinary and colorectal symptoms, delivered by centrally trained interventionists. Self-reported pelvic floor exercise was 93.4% at 6 months and 81.4% at 24 months in the BPMT group, and 8/188 women (4.3%) in the usual-care group had supervised therapy outside the trial.^[1]

Adverse events

Serious adverse events occurred in 31/188 (16.5%) after ULS and 31/186 (16.7%) after SSLF (adjusted OR 0.9, 95% CI 0.5–1.6). Modified Dindo grade III events accounted for 15 and 17 of them; one death after ULS was judged unrelated to surgery.^[1]

The two procedures differed in their characteristic complications. Intraoperative ureteral obstruction occurred in 6/188 women (3.2%) after ULS and none after SSLF; it was relieved by suture removal in five and a stent in one, and one further ureteral injury after ULS was recognized postoperatively. Neurologic pain requiring additional treatment (acute buttock, groin or leg pain on the side of the suspension sutures) occurred in 13/188 (6.9%) after ULS and 23/186 (12.4%) after SSLF (adjusted OR 0.5, 95% CI 0.2–1.0; P = 0.049). It persisted to the 4–6 week visit in 1 versus 8 women, and 3 women after SSLF returned to the operating room for suture removal. Bladder injury occurred in 22 (11.7%) and 18 (9.7%), most during sling placement, and transfusion in 7 (3.7%) and 4 (2.2%).^[1]

Suture-related findings emerged later. Vaginal granulation tissue between 6 and 24 months was recorded in 36/188 (19.1%) after ULS and 26/186 (14.0%) after SSLF, suture exposure in 29 (15.4%) and 32 (17.2%), and sling erosion or exposure in 3 (1.6%) and 1 (0.5%). By 5 years the cumulative figures were granulation tissue 35/121 (28.9%) versus 21/112 (18.8%) (difference 10.5 points, 95% CI −0.2 to 21.2), suture exposure 31/120 (25.8%) versus 29/113 (25.7%), sling erosion or exposure 2/113 (1.8%) versus 4/108 (3.7%) and repeat prolapse surgery 10/118 (8.5%) versus 5/110 (4.5%). Most adverse events occurred in the first 2 years.^[1][3]

Limitations

• Every woman had a stress incontinence indication and all but 3 received a retropubic sling, and no graft or mesh was used in the prolapse repair.
• Composite success required anatomic support, no bothersome bulge and no retreatment, so it is not isolated apical success. The 2-year success rates sat below the 70–90% reported in earlier series, which the authors attributed in part to the composite definition, masked examiners and strict anatomic criteria.
• Primary outcome data were available for 304 of 374 randomized women at 2 years.
• At baseline, posterior prolapse beyond the hymen was more frequent in the SSLF group (25.3% versus 13.8%) and women in the ULS group had more vaginal deliveries.
• Secondary and interaction analyses were exploratory, and the trial was not powered for interactions.
• E-OPTIMAL enrolled 285 of 309 eligible women. Participants with better outcomes were more likely to enroll, which could bias absolute estimates; the authors judged that it probably affected both groups similarly.
• The 5-year interval for ULS minus SSLF (−24.2 to 6.6 percentage points) is wide enough to include a substantial difference in either direction. Failure was irreversible in the time-to-event model although about a third of failures later met the success definition.^[1][3]

In conversation

• Built on by IMPROVE (2023). Used OPTIMAL's uterosacral and sacrospinous suspension framework and a similar composite failure definition.
• Built on by SUPeR (2019). Its uterosacral suspension modified the Shull technique, which the same Pelvic Floor Disorders Network had used in OPTIMAL.
• Parallel to OPUS (2012). Same Pelvic Floor Disorders Network; OPUS randomized a sling in women without stress incontinence having vaginal prolapse surgery; all OPTIMAL women had stress incontinence.
• Parallel to ASPIRe (2024). ASPIRe's native-tissue arm allowed the two operations OPTIMAL compared: sacrospinous fixation (82% of the arm) or uterosacral suspension.
• Parallel to ESTEEM (2019). Both Pelvic Floor Disorders Network trials randomized perioperative behavioral therapy alongside pelvic surgery, in different patient populations.

Same setting VUE (2020)

OPUS, from the same network, randomized a midurethral sling or sham incisions in women without stress incontinence having vaginal prolapse surgery;^[4] OPTIMAL's women all had stress incontinence and a sling, so the two trials enrolled different continence populations. OPTIMAL's 2-year composite is a status at one time point, whereas E-OPTIMAL estimates time to first failure.

Report history

OPTIMAL report history

Published	Report	Cutoff	Headline results	Citation
2014-03-12	2-year randomized report (corrected 2015) · primary	—	2-year composite surgical success (apical descent no more than one-third of vaginal length, no anterior or posterior wall beyond the hymen, no bothersome bulge, no retreatment); corrected analysis: 100/155 (64.5%) vs 94/149 (63.1%); OR 1.1; CI 0.7–1.7; P 0.84 Not met Adjusted mean change from baseline in Urinary Distress Inventory score (range 0–300) at 6 months: −94.6 with BPMT versus −87.9 with usual care: Mean difference, BPMT minus usual care. Minimum important difference for the score was 11 points.; difference -6.7; CI -19.7–6.2; P 0.31 Not met Adjusted mean change from baseline in Pelvic Organ Prolapse Distress Inventory score (range 0–300) at 24 months: −73.3 with BPMT versus −65.2 with usual care: Mean difference, BPMT minus usual care.; difference -8; CI -22.1–6.1; P 0.26 Not met 24-month anatomic failure (apical descent beyond one-third of vaginal length, anterior or posterior wall beyond the hymen, or retreatment): 37/153 (24.2%) with BPMT versus 42/164 (25.6%) with usual care; corrected analysis: The table row appears under the surgical columns in the online text; the counts are the behavioral comparison.; OR 0.9; CI 0.6–1.6; P 0.82 Not met Bothersome vaginal bulge symptoms at 24 months, uterosacral versus sacrospinous: 25/151 (16.6%) vs 30/154 (19.5%); OR 0.8; CI 0.5–1.5; P 0.49 Retreatment for prolapse by surgery or pessary at 24 months, uterosacral versus sacrospinous: 8/161 (5.0%) vs 8/155 (5.2%); OR 0.9; CI 0.3–2.6; P 0.81 Apical descent greater than one-third of vaginal length at 24 months, uterosacral versus sacrospinous, within usual care: 4.9% vs 15.6%; OR 0.3; CI 0.1–0.8	Barber MD, Brubaker L, Burgio KL, et al. Comparison of 2 transvaginal surgical approaches and perioperative behavioral therapy for apical vaginal prolapse: the OPTIMAL randomized trial. *JAMA.* 2014;311(10):1023-1034. DOI
2015-06-09	Correction of outcome data · correction	—	—	Incorrect outcomes data. *JAMA.* 2015;313(22):2287. DOI
2018-04-17	5-year extension (E-OPTIMAL) · updated	—	Estimated surgical failure by year 5 (anatomic failure, bothersome bulge or retreatment; time to event), uterosacral versus sacrospinous: 61.5% vs 70.3%; difference -8.8; CI -24.2–6.6; P 0.25 Not met Estimated anatomic failure by year 5: 45.6% with BPMT versus 47.2% with usual care: Adjusted risk difference in percentage points, BPMT minus usual care.; difference -1.6; CI -21.2–17.9 Not met Change from baseline in Pelvic Organ Prolapse Distress Inventory score by year 5: −59.4 with BPMT versus −61.8 with usual care: Adjusted mean difference, BPMT minus usual care.; difference 2.4; CI -13.7–18.4 Not met Estimated anatomic failure by year 5, uterosacral versus sacrospinous: 47.5% vs 61.8%; difference -14.3; CI -32.1–3.5; P 0.11 Estimated bothersome bulge symptoms by year 5, uterosacral versus sacrospinous: 37.4% vs 41.8%; difference -4.4; CI -21.3–12.5; P 0.60 Retreatment for prolapse by year 5, uterosacral versus sacrospinous: 11.9% vs 8.1%; difference 3.9; CI -3.8–11.5 Change from baseline in Pelvic Organ Prolapse Distress Inventory score at 5 years, uterosacral versus sacrospinous: −67.6 vs −74.2; difference 6.6; CI -9.5–22.7 Descriptive	Jelovsek JE, Barber MD, Brubaker L, et al. Effect of uterosacral ligament suspension vs sacrospinous ligament fixation with or without perioperative behavioral therapy for pelvic organ vaginal prolapse on surgical outcomes and prolapse symptoms at 5 years in the OPTIMAL randomized clinical trial. *JAMA.* 2018;319(15):1554-1565. DOI

---

References

1. Barber MD, Brubaker L, Burgio KL, et al. "Comparison of 2 Transvaginal Surgical Approaches and Perioperative Behavioral Therapy for Apical Vaginal Prolapse: The OPTIMAL Randomized Trial." JAMA. 2014;311(10):1023-1034. doi:10.1001/jama.2014.1719

2. "Incorrect Outcomes Data." JAMA. 2015;313(22):2287. doi:10.1001/jama.2015.4817

3. Jelovsek JE, Barber MD, Brubaker L, et al. "Effect of Uterosacral Ligament Suspension vs Sacrospinous Ligament Fixation With or Without Perioperative Behavioral Therapy for Pelvic Organ Vaginal Prolapse on Surgical Outcomes and Prolapse Symptoms at 5 Years in the OPTIMAL Randomized Clinical Trial." JAMA. 2018;319(15):1554-1565. doi:10.1001/jama.2018.2827

4. Wei JT, Nygaard I, Richter HE, et al. "A Midurethral Sling to Reduce Incontinence after Vaginal Prolapse Repair." N Engl J Med. 2012;366(25):2358-2367. doi:10.1056/NEJMoa1111967