Advertisement

Effect of sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral ligament suspension on treatment failure in women with uterovaginal prolapse: 5-year results of a randomized clinical trial

Published:March 11, 2021DOI:https://doi.org/10.1016/j.ajog.2021.03.012

      Background

      Vaginal hysterectomy with suture apical suspension is commonly performed for uterovaginal prolapse. Sacrospinous hysteropexy with graft (vaginal mesh hysteropexy) is an alternative, although in 2019 the Food and Drug Administration removed this mesh product from the United States market.

      Objective

      Our objective was to compare the efficacy and adverse events of these 2 procedures.

      Study Design

      At 9 clinical sites in the United States National Institutes of Health and National Institute of Child Health and Human Development Pelvic Floor Disorders Network, 183 postmenopausal women requesting vaginal surgery for symptomatic uterovaginal prolapse were enrolled in a multisite randomized superiority clinical trial, comparing a sacrospinous hysteropexy with graft (hysteropexy) with a vaginal hysterectomy with uterosacral ligament suspension (hysterectomy). Participants consented to remain masked to treatment assignment for the study duration. Study visits were conducted at 6-month intervals through 60 months. The primary treatment failure composite outcome (retreatment of prolapse, prolapse beyond the hymen, or prolapse symptoms) was evaluated with survival modeling. Secondary outcomes included complications or adverse events, individual anatomic measures of the pelvic organ prolapse quantification examination, and presence, severity, and impact and bother of prolapse, urinary, bowel, and pain symptoms as measured by validated questionnaires. The 3-year published results suggested better primary outcomes with sacrospinous hysteropexy with graft, but the differences were not statistically significant (P=.06). This study reports the 5-year outcomes.

      Results

      A total of 183 women with a mean age of 66 years were randomized between April 2013 and February 2015; 93 were randomized to hysteropexy and 90 were randomized to hysterectomy. Notably, 175 were included in the trial, and 156 (89%) completed the 5-year follow-up. The primary outcome showed fewer failures for hysteropexy than hysterectomy through 5 years (adjusted hazard ratio, 0.58; 95% confidence interval, 0.36–0.94; P=.03), with failure rates of 37% vs 54%, respectively, resulting in a difference of −18% (95% confidence interval, −33% to −3%) at 5 years. With the exception of the Urogenital Distress Inventory, no group differences were demonstrated in patient-reported pelvic floor symptoms, prolapse symptoms, bowel function symptoms, general quality of life, body image, or pelvic pain. At their last visit through 5 years, 70% of participants (129 of 183) reported they remained masked to their treatment with no difference in masking between groups. Adverse events for hysteropexy vs hysterectomy included mesh exposure (8% vs 0%), granulation tissue after 12 weeks (1% vs 12%), and suture exposure after 12 weeks (3% vs 21%), respectively.

      Conclusion

      Among women with symptomatic uterovaginal prolapse undergoing vaginal surgery, sacrospinous hysteropexy with graft resulted in a lower composite failure rate than vaginal hysterectomy through 5 years. There were no meaningful differences in patient-reported outcomes between groups. Our results suggest that this vaginal mesh hysteropexy procedure should be made available to patients.

      Key words

      Introduction

      The lifetime risk of undergoing surgery for pelvic organ prolapse (POP) in the United States is 12.6%.
      • Wu J.M.
      • Matthews C.A.
      • Conover M.M.
      • Pate V.
      • Jonsson Funk M.
      Lifetime risk of stress urinary incontinence or pelvic organ prolapse surgery.
      Uterine-preserving prolapse surgeries, known as hysteropexy, lessen operative time and blood loss compared with procedures involving hysterectomy and do not appear to change short-term prolapse outcomes.
      • Meriwether K.V.
      • Antosh D.D.
      • Olivera C.K.
      • et al.
      Uterine preservation vs hysterectomy in pelvic organ prolapse surgery: a systematic review with meta-analysis and clinical practice guidelines.
      Despite the potential for lower morbidity, according to a 2002 to 2012 United States inpatient hospital database, hysteropexy procedures only accounted for 5% of uterovaginal prolapse surgeries, whereas hysterectomies were 8 times more commonly performed.
      • Madsen A.M.
      • Raker C.
      • Sung V.W.
      Trends in hysteropexy and apical support for uterovaginal prolapse in the United States from 2002 to 2012.
      Although suture sacrospinous hysteropexy has been deemed noninferior to vaginal hysterectomy with concomitant uterosacral ligament suspension at 1-year follow-up,
      • Detollenaere R.J.
      • den Boon J.
      • Stekelenburg J.
      • et al.
      Sacrospinous hysteropexy versus vaginal hysterectomy with suspension of the uterosacral ligaments in women with uterine prolapse stage 2 or higher: multicentre randomised non-inferiority trial.
      a 2016 Cochrane review of trials comparing apical suspensions for treatment of uterovaginal prolapse found that no conclusion could be reached regarding the superiority of uterine-preserving surgery vs surgeries including vaginal hysterectomy.
      • Maher C.
      • Feiner B.
      • Baessler K.
      • Christmann-Schmid C.
      • Haya N.
      • Brown J.
      Surgery for women with apical vaginal prolapse.

      Why was this study conducted?

      Vaginal hysterectomy with suture apical suspension is commonly performed for uterovaginal prolapse. Sacrospinous hysteropexy with graft (vaginal mesh hysteropexy) is a controversial alternative.

      Key findings

      At 9 clinical sites in the United States National Institutes of Health and National Institute of Child Health and Human Development Pelvic Floor Disorders Network, 183 postmenopausal women with symptomatic uterovaginal prolapse were enrolled and randomized in a multisite randomized superiority clinical trial. Among women with symptomatic uterovaginal prolapse undergoing vaginal surgery, sacrospinous hysteropexy with graft resulted in a lower composite failure rate than vaginal hysterectomy (adjusted hazard ratio, 0.58; 95% confidence interval [CI], 0.36–0.94; P=.03) resulting in a difference of −18% (95% CI, −33% to −3%) at 5 years.

      What does this add to what is known?

      This study suggests that vaginal mesh hysteropexy procedures are superior to native tissue vaginal hysterectomy procedures and the 2019 Food and Drug Administration ban on these products should be reconsidered.
      Native tissue repairs have higher than desired surgical failure rates,
      • Jelovsek J.E.
      • Barber M.D.
      • 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.
      but the use of transvaginal mesh to augment the treatment of POP has resulted in controversial outcomes.
      • Maher C.
      • Feiner B.
      • Baessler K.
      • Christmann-Schmid C.
      • Haya N.
      • Marjoribanks J.
      Transvaginal mesh or grafts compared with native tissue repair for vaginal prolapse.
      There is a paucity of high-quality and long-term data comparing apical transvaginal mesh with native tissue procedures, and both the United States Food and Drug Administration (FDA) and the United Kingdom National Institute for Health and Care Excellence recommended further research on transvaginal mesh procedures.
      United States Food and Drug Administration
      Urogynecologic surgical mesh Implants.
      To address this, the Study of Uterine Prolapse Procedures—Randomized Trial “SUPeR” was designed to compare effectiveness and adverse events of 2 transvaginal apical suspension strategies for uterovaginal prolapse: sacrospinous hysteropexy with graft (polypropylene mesh) vs vaginal hysterectomy with bilateral uterosacral ligament suture apical suspension.
      • Nager C.W.
      • Zyczynski H.
      • Rogers R.G.
      • et al.
      The design of a randomized trial of vaginal surgery for uterovaginal prolapse: vaginal hysterectomy With native tissue vault suspension Versus mesh hysteropexy suspension (the study of uterine prolapse procedures randomized trial).
      The 3-year published results suggested better outcomes with sacrospinous hysteropexy with graft, but the differences were not statistically significant (P=.06).
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      This study reports the outcomes at 5 years.

      Material and Methods

      Study design

      A 9-center, randomized, superiority trial compared anatomic and functional outcomes in women with uterovaginal prolapse, after a sacrospinous hysteropexy with graft (“hysteropexy”) (Supplemental Figure 1) or a vaginal hysterectomy with uterosacral ligament vault suspension (“hysterectomy”) (Supplemental Figure 2). At 5 years, all participants were unmasked. Those who did not undergo reoperation for prolapse were eligible for a 5-year extended follow-up (ongoing).The study was conducted by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Pelvic Floor Disorders Network. The trial design and the primary 3-year results are published,
      • Nager C.W.
      • Zyczynski H.
      • Rogers R.G.
      • et al.
      The design of a randomized trial of vaginal surgery for uterovaginal prolapse: vaginal hysterectomy With native tissue vault suspension Versus mesh hysteropexy suspension (the study of uterine prolapse procedures randomized trial).
      ,
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      and the 5-year statistical analysis plan can be found in online supplemental materials. The protocol was approved by an independent, NICHD-convened Data and Safety Monitoring Board and individual site institutional review boards. Participants provided written informed consent. This trial is registered with ClinicalTrials.gov: NCT01802281.

      Participants

      Eligible participants included women at the age of ≥21 years requesting vaginal surgery for symptomatic uterovaginal prolapse beyond the hymen. Participants were amenorrheic for at least 1 year (to allow masking), had completed childbearing, and had symptomatic POP that included uterine descent into the lower half of the vagina. Women with previous uterine suspensions, previous synthetic grafts for prolapse repair, or uterine abnormalities were excluded. Detailed inclusion and exclusion criteria were published with the primary 3-year results.
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      For study generalizability and inclusivity, all races and ethnic groups were included; participants self-classified their race or ethnicity with options defined by the investigator.

      Randomization and masking

      Participants consented to remain masked to treatment assignment for the study duration unless a medical indication for unmasking was identified. Participants were randomized 1:1 in the operating room with an automated web-based system using permuted blocks, with block sizes of 2 or 4 stratified by site.
      • Broglio K.
      Randomization in clinical trials: permuted blocks and stratification.
      Concomitant native tissue vaginal prolapse repairs such as anterior or posterior repairs and perineal reconstruction were permitted, as were full-length mesh midurethral slings for stress urinary incontinence.

      Procedures

      Surgeon certification required performance of at least 5 recent procedures of each of the 2 surgeries; surgery standardization is found in the methods publication and the online protocol.
      • Nager C.W.
      • Zyczynski H.
      • Rogers R.G.
      • et al.
      The design of a randomized trial of vaginal surgery for uterovaginal prolapse: vaginal hysterectomy With native tissue vault suspension Versus mesh hysteropexy suspension (the study of uterine prolapse procedures randomized trial).
      ,
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      Hysteropexy was standardized with the Uphold LITE transvaginal mesh support system made out of polypropylene mesh (Boston Scientific, Marlborough, MA), and the ipsilateral uterosacral ligament suspension was performed bilaterally with 1 permanent and 1 delayed absorbable suture on each side.

      Outcomes

      Study visits were conducted at 6-month intervals through 60 months with all participants including those who experienced treatment failure and/or retreatment. Study personnel masked to treatment assignment administered the patient-reported questionnaires and the comprehensive adverse event survey. To minimize the risk of potential bias, the pelvic examination for anatomic prolapse assessment and suture or mesh exposure was performed by a qualified examiner who was not the surgeon. At each visit, participants were queried about their perception of their treatment allocation.
      The primary outcome was treatment failure, a composite measure that included any of the following: (1) retreatment for prolapse (pessary fitting or surgery); (2) anatomic outcomes, defined as any POP-Q
      • Bump R.C.
      • Mattiasson A.
      • Bø K.
      • et al.
      The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction.
      measure beyond the hymen; and (3) symptomatic outcomes, defined as a positive response (and any degree of bother) to the Pelvic Floor Distress Inventory
      • Yalcin I.
      • Bump R.C.
      Validation of two global impression questionnaires for incontinence.
      (PFDI-20) question “Do you usually have a bulge or something falling out that you can see or feel in your vaginal area?” For the primary analysis, this outcome was assessed cumulatively such that once a participant met failure criteria, her outcome was a failure at subsequent time points.
      To further characterize failure over time, the composite surgical failure outcome at discrete time points (6, 12, 18, 24, 30, 36, 42, 48, 54, and 60 months) was evaluated in the intent-to-treat population by randomized intervention. “Persistent Success” was defined as meeting the composite definition of success at all-time points for which the participant had data available, the “Persistent Failure” group was defined as meeting the definition of failure and remaining a failure at all subsequent visits, and the “Intermittent Success/Failure” group was defined as fluctuating between success and failure over time.
      Secondary outcomes included complications or adverse events, individual anatomic measures of the POP-Q examination, and presence, severity, and impact/bother of prolapse, urinary, bowel, and pain symptoms as measured by the Pelvic Floor Distress Inventory,
      • Barber M.D.
      • Kuchibhatla M.N.
      • Pieper C.F.
      • Bump R.C.
      Psychometric evaluation of 2 comprehensive condition-specific quality of life instruments for women with pelvic floor disorders.
      Patient Global Impression of Improvement,
      • Yalcin I.
      • Bump R.C.
      Validation of two global impression questionnaires for incontinence.
      Incontinence Severity Index,
      • Sandvik H.
      • Seim A.
      • Vanvik A.
      • Hunskaar S.
      A severity index for epidemiological surveys of female urinary incontinence: comparison with 48-hour pad-weighing tests.
      Pelvic Floor Impact Questionnaire,
      • Barber M.D.
      • Kuchibhatla M.N.
      • Pieper C.F.
      • Bump R.C.
      Psychometric evaluation of 2 comprehensive condition-specific quality of life instruments for women with pelvic floor disorders.
      Functional Activity Scale,
      • McCarthy Jr., M.
      • Jonasson O.
      • Chang C.H.
      • et al.
      Assessment of patient functional status after surgery.
      and Short Form Health Survey (SF-12).
      • Ware Jr., J.
      • Kosinski M.
      • Keller S.D.
      A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity.
      Sexual function and body image were assessed with the Pelvic Organ Prolapse Incontinence Sexual Function Questionnaire-International Urogynecological Association Revised (PISQ-IR)
      • Rogers R.G.
      • Rockwood T.H.
      • Constantine M.L.
      • et al.
      A new measure of sexual function in women with pelvic floor disorders (PFD): the Pelvic Organ prolapse/Incontinence Sexual Questionnaire, IUGA-revised (PISQ-IR).
      and Body Image Scale,
      • Hopwood P.
      • Fletcher I.
      • Lee A.
      • Al Ghazal S.
      A body image scale for use with cancer patients.
      ,
      • Jelovsek J.E.
      • Barber M.D.
      Women seeking treatment for advanced pelvic organ prolapse have decreased body image and quality of life.
      respectively.

      Statistical analysis

      The planned 180 participants provided a power of 0.86 to detect a risk difference of 12% (8% failure for the hysteropexy group vs 20% failure for the hysterectomy group) at 2 years.
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      Efficacy analyses were performed on all randomized, eligible patients by randomized intervention (intent-to-treat population, n=175); safety analyses were conducted on all participants who received an intervention by actual surgery received (safety population, n=183). Sensitivity efficacy analyses were performed on all treated participants by randomized intervention (full treated population; n=183) and on eligible participants who received the intervention to which they were randomized (per-protocol population, n=173). The analysis included all data through 5 years of follow-up. Because this is a secondary objective, confidence intervals (CIs) and P values are descriptive and are calculated with no adjustments for multiple comparisons. Analyses were performed using SAS, version 9.4 or greater (SAS Institute Inc, Cary, NC).
      Primary efficacy analyses used a proportional hazard survival model to account for interval censoring and an aggregate time-varying hazard, with 5 piecewise exponential constant hazard periods (0–12 months, >12–24 months, >24–36 months, >36–48 months, and >48–60 months), and controlled for site and previous prolapse surgery. The proportional hazards assumption was tested with a treatment by time interaction and plots of the smoothed hazard rate and log cumulative hazard vs log-time. Sensitivity analyses were conducted on the full treated population and per-protocol population by randomized intervention. Additional sensitivity analyses were conducted on the intent-to-treat population without covariate adjustment, using nonparametric maximum likelihood estimation,
      • Wellner J.A.
      • Zhan Y.H.
      A hybrid algorithm for computation of the nonparametric maximum likelihood estimator from censored data.
      and frailty model with random site effect.
      Linear and generalized linear mixed models were used to evaluate continuous and binary outcomes measured longitudinally. Differences between treatment groups for time-specific measures of composite failure were assessed with Mantel-Haenszel chi-squared tests stratified by site. Treatment differences for other categorical measures were analyzed with chi-squared or Fisher exact tests. CIs for incidence of adverse events were estimated via the exact Clopper-Pearson method.

      Results

      Between April 2013 and February 2015, 183 women were randomized (93 in the hysteropexy group and 90 in the hysterectomy group) at 9 sites by 34 Female Pelvic Medicine and Reconstructive Surgery subspecialists, 175 were included in the trial after ineligibility exclusions, and 156 (89%, 81 in hysteropexy group and 75 hysterectomy group) completed the 5-year follow-up (Figure 1). Missed 6-month interval follow-up visits through 60 months ranged from 1.2% to 11.1%. Baseline demographic and clinical characteristics were similar between the 2 groups (Table 1). As previously reported at 3 years follow-up, the only differences noted in operative interventions between groups was that anterior repair was more commonly performed with hysteropexy (79 of 88 [90%]) than with the hysterectomy group (62 of 87 [71%]) (difference, 19%; 95% CI, 7–30; P=.002) and operative time was less in the hysteropexy group compared with the hysterectomy group (111.5 [standard deviation {SD}, 39.7] vs 156.7 [SD, 43.9] mean minutes; difference, −45.2; 95% CI, −57.7 to −32.7; P<.001).
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      Figure thumbnail gr1
      Figure 1Consort diagram of participants
      Colonic Diverticular perforation and leukemia are the causes of death for the 2 participants in the hysterectomy group who died during 0 to 60 months and lung cancer is the cause of death for the 1 participant in the hysteropexy group who died during 0 to 60 months.
      PFDI, Pelvic Floor Distress Inventory; POP-Q, Pelvic Organ Prolapse Quantification.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Table 1Baseline characteristics in the intent-to-treat population by randomized intervention
      CharacteristicHysteropexy (n=88)
      Sample size is 88 for Hysteropexy and 87 for Hysterectomy unless otherwise specified
      Hysterectomy (n=87)
      Sample size is 88 for Hysteropexy and 87 for Hysterectomy unless otherwise specified
      Patient demographics
      Age, mean (SD), y65.5 (7.3)66.2 (7.4)
      Race
      Other race category includes patients who classified themselves as Hispanic
      White73 (83)77 (89)
      Black8 (9)3 (3)
      American Indian or Alaskan Native0 (0)4 (5)
      Asian2 (2)1 (1)
      More than 10 (0)1 (1)
      Other5 (6)1 (1)
      Hispanic or Latina, number/total (%)9/85 (11)7/85 (8)
      Medical history
      Gravidity, median (IQR)3 (2–5)3 (2–4)
      Vaginal parity, median (IQR)3 (2–3)2 (2–3)
      BMI (kg/m2), mean (SD)28.9 (4.0)28.1 (4.4)
      Postmenopausal86 (98)85 (98)
      Previous pelvic organ prolapse surgery5 (6)4 (5)
      Pelvic floor measurements
      POP-Q measurement, mean (SD), cm
      TVL is the total vaginal length from posterior fornix to the hymen when POP-Q point C is reduced to its full normal position. POP-Q point C represents either the most distal edge of the cervix or the leading edge of the vaginal cuff (hysterectomy scar). POP-Q point Bp represents the most distal position of any part of the upper posterior vaginal wall and has a range of −3.0 cm to TVL. POP-Q point Ba represents the most distal position of any part of the upper anterior vaginal wall and has a range of −3.0 cm to TVL. Patients without pelvic organ prolapse have negative values. See Supplemental Figure 3 in Supplement 2 for additional information
      Ba3.3 (2.0)3.1 (2.2)
      Bp0.8 (3.0)1.1 (3.0)
      C0.4 (3.5)0.7 (3.6)
      TVL9.1 (1.1)9.1 (1.1)
      POP-Q stage
      Pelvic Organ Prolapse Quantification (POP-Q) stages: stage 2, the vagina is prolapsed between 1 cm above the hymen and 1 cm below the hymen; stage 3, the vagina is prolapsed more than 1 cm beyond the hymen but is not everted within 2 cm of its length; stage 4, the vagina is everted to within 2 cm of its length
      215 (17)19 (22)
      363 (72)59 (68)
      410 (11)9 (10)
      Patient-reported outcome scores, mean (SD)
      Pelvic floor symptoms
      Pelvic Floor Distress Inventory (PFDI) score
      The Pelvic Floor Distress Inventory (PFDI) score ranges from 0 (least distress) to 300 (greatest distress) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Distress Inventory (POPDI), Urogenital Distress Inventory (UDI), and Colorectal Anal Distress Inventory (CRADI) each ranging from 0 (least distress) to 100 (greatest distress)
      112.9 (56.9)109.5 (50.0)
      Pelvic Floor Impact Questionnaire (PFIQ) score
      The Pelvic Floor Impact Questionnaire (PFIQ) score ranges from 0 (least impact) to 300 (greatest adverse impact) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Impact Questionnaire (POPIQ), Urinary Impact Questionnaire (UIQ), and Colorectal Anal Impact Questionnaire (CRAIQ) each ranging from 0 (least impact) to 100 (greatest adverse impact). A midrange score for these subscales implies bother from prolapse, urinary incontinence, or fecal incontinence
      57.5 (59.6)56.8 (58.5)
      Prolapse symptoms
      Pelvic Organ Prolapse Distress Inventory (POPDI) score
      The Pelvic Floor Distress Inventory (PFDI) score ranges from 0 (least distress) to 300 (greatest distress) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Distress Inventory (POPDI), Urogenital Distress Inventory (UDI), and Colorectal Anal Distress Inventory (CRADI) each ranging from 0 (least distress) to 100 (greatest distress)
      49.1 (23.6)47.8 (21.0)
      Pelvic Organ Prolapse Impact Questionnaire (POPIQ) score
      The Pelvic Floor Impact Questionnaire (PFIQ) score ranges from 0 (least impact) to 300 (greatest adverse impact) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Impact Questionnaire (POPIQ), Urinary Impact Questionnaire (UIQ), and Colorectal Anal Impact Questionnaire (CRAIQ) each ranging from 0 (least impact) to 100 (greatest adverse impact). A midrange score for these subscales implies bother from prolapse, urinary incontinence, or fecal incontinence
      18.7 (24.8)20.9 (22.0)
      Urinary function
      Urogenital Distress Inventory (UDI) score
      The Pelvic Floor Distress Inventory (PFDI) score ranges from 0 (least distress) to 300 (greatest distress) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Distress Inventory (POPDI), Urogenital Distress Inventory (UDI), and Colorectal Anal Distress Inventory (CRADI) each ranging from 0 (least distress) to 100 (greatest distress)
      42.5 (26.0)38.3 (26.5)
      Urinary Impact Questionnaire (UIQ) score
      The Pelvic Floor Impact Questionnaire (PFIQ) score ranges from 0 (least impact) to 300 (greatest adverse impact) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Impact Questionnaire (POPIQ), Urinary Impact Questionnaire (UIQ), and Colorectal Anal Impact Questionnaire (CRAIQ) each ranging from 0 (least impact) to 100 (greatest adverse impact). A midrange score for these subscales implies bother from prolapse, urinary incontinence, or fecal incontinence
      27.6 (27.2)22.8 (27.8)
      Incontinence Severity Index (ISI) score
      The Incontinence Severity Index (ISI) score ranges from 0 (no incontinence) to 12 (severe incontinence). The ISI is a product of the frequency and volume of urine loss; a person losing small splashes of urine a few times per week would score at 6
      4.6 (4.2)3.5 (3.5)
      Bowel function
      Colorectal Anal Distress Inventory (CRADI) score
      The Pelvic Floor Distress Inventory (PFDI) score ranges from 0 (least distress) to 300 (greatest distress) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Distress Inventory (POPDI), Urogenital Distress Inventory (UDI), and Colorectal Anal Distress Inventory (CRADI) each ranging from 0 (least distress) to 100 (greatest distress)
      21.3 (21.1)23.3 (18.3)
      Colorectal Anal Impact Questionnaire (CRAIQ) score
      The Pelvic Floor Impact Questionnaire (PFIQ) score ranges from 0 (least impact) to 300 (greatest adverse impact) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Impact Questionnaire (POPIQ), Urinary Impact Questionnaire (UIQ), and Colorectal Anal Impact Questionnaire (CRAIQ) each ranging from 0 (least impact) to 100 (greatest adverse impact). A midrange score for these subscales implies bother from prolapse, urinary incontinence, or fecal incontinence
      11.2 (20.4)13.2 (20.8)
      Sexual function
      Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR) average score among sexually active women
      The Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR) mean scores ranges from 1 (worse sexual experience) to 5 (better sexual experience). Midrange scores are common in women with pelvic floor disorders23
      (n=30)(n=40)
      Mean (SD)3.4 (0.6)3.4 (0.4)
      General quality of life
      Functional Activity Assessment Scale (FAS) score
      The Functional Activities Scale (FAS) ranges from 0 to 100 with higher values indicating better functional activity. Surgical hernia patients with independent daily living and no comorbidities have mean values of 8016
      88.8 (15.1)87.7 (16.5)
      Body Image Scale (BIS) score
      The Body Image Scale (BIS) score ranges from 0 (no symptoms/distress) to 24 (severe symptoms/distress). A mean score of 8 was typical in cancer patients19
      6.4 (6.4)5.3 (6.0)
      Short Form Health Survey (SF-12)
      The Short Form Health Survey (SF-12) version 1 United States physical and mental component scores are norm-based scores calibrated with 50 as the average score and 10 as the standard deviation.
      (n=87)(n=86)
      Physical component score
      The Short Form Health Survey (SF-12) version 1 United States physical and mental component scores are norm-based scores calibrated with 50 as the average score and 10 as the standard deviation.
      45.0 (10.9)44.4 (10.9)
      Mental component score
      The Short Form Health Survey (SF-12) version 1 United States physical and mental component scores are norm-based scores calibrated with 50 as the average score and 10 as the standard deviation.
      50.9 (9.7)51.8 (9.5)
      BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); GED, general education development; IQR, interquartile range; IUGA, International Urogynecological Association; POP-Q, Pelvic Organ Prolapse Quantification; SD, standard deviation; TVL, total vaginal length.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      a Sample size is 88 for Hysteropexy and 87 for Hysterectomy unless otherwise specified
      b Other race category includes patients who classified themselves as Hispanic
      c TVL is the total vaginal length from posterior fornix to the hymen when POP-Q point C is reduced to its full normal position. POP-Q point C represents either the most distal edge of the cervix or the leading edge of the vaginal cuff (hysterectomy scar). POP-Q point Bp represents the most distal position of any part of the upper posterior vaginal wall and has a range of −3.0 cm to TVL. POP-Q point Ba represents the most distal position of any part of the upper anterior vaginal wall and has a range of −3.0 cm to TVL. Patients without pelvic organ prolapse have negative values. See Supplemental Figure 3 in Supplement 2 for additional information
      d Pelvic Organ Prolapse Quantification (POP-Q) stages: stage 2, the vagina is prolapsed between 1 cm above the hymen and 1 cm below the hymen; stage 3, the vagina is prolapsed more than 1 cm beyond the hymen but is not everted within 2 cm of its length; stage 4, the vagina is everted to within 2 cm of its length
      e The Pelvic Floor Distress Inventory (PFDI) score ranges from 0 (least distress) to 300 (greatest distress) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Distress Inventory (POPDI), Urogenital Distress Inventory (UDI), and Colorectal Anal Distress Inventory (CRADI) each ranging from 0 (least distress) to 100 (greatest distress)
      f The Pelvic Floor Impact Questionnaire (PFIQ) score ranges from 0 (least impact) to 300 (greatest adverse impact) and is a sum of the 3 subscale scores: Pelvic Organ Prolapse Impact Questionnaire (POPIQ), Urinary Impact Questionnaire (UIQ), and Colorectal Anal Impact Questionnaire (CRAIQ) each ranging from 0 (least impact) to 100 (greatest adverse impact). A midrange score for these subscales implies bother from prolapse, urinary incontinence, or fecal incontinence
      g The Incontinence Severity Index (ISI) score ranges from 0 (no incontinence) to 12 (severe incontinence). The ISI is a product of the frequency and volume of urine loss; a person losing small splashes of urine a few times per week would score at 6
      h The Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR) mean scores ranges from 1 (worse sexual experience) to 5 (better sexual experience). Midrange scores are common in women with pelvic floor disorders
      • Rockwood T.H.
      • Constantine M.L.
      • Adegoke O.
      • et al.
      The PISQ-IR: considerations in scale scoring and development.
      i The Functional Activities Scale (FAS) ranges from 0 to 100 with higher values indicating better functional activity. Surgical hernia patients with independent daily living and no comorbidities have mean values of 80
      • McCarthy Jr., M.
      • Jonasson O.
      • Chang C.H.
      • et al.
      Assessment of patient functional status after surgery.
      j The Body Image Scale (BIS) score ranges from 0 (no symptoms/distress) to 24 (severe symptoms/distress). A mean score of 8 was typical in cancer patients
      • Hopwood P.
      • Fletcher I.
      • Lee A.
      • Al Ghazal S.
      A body image scale for use with cancer patients.
      k The Short Form Health Survey (SF-12) version 1 United States physical and mental component scores are norm-based scores calibrated with 50 as the average score and 10 as the standard deviation.
      The primary outcome showed fewer failures for hysteropexy than hysterectomy through 5 years (adjusted hazard ratio, 0.58; 95% CI, 0.36–0.94; P=.03) (Figure 2). The failure was 37% for hysteropexy and 54% for hysterectomy resulting in a common risk difference stratified by site of −18% (95% CI, −33% to −3%) at 5 years gradually increasing over time from a risk difference of −9% (95% CI, −21% to 2%) at 1 year (Table 2). The proportional hazards assumption was not met (P<.001); graphical assessments (Supplemental Figures 4 and 5) indicate a quantitative departure during 18 through 36 months. The hazard ratio is interpretable as an average relative risk across time. Cumulative failures and failure categories based on initial failure type at yearly intervals are found in Table 2. Most failures were either anatomic or anatomic and symptomatic, with few symptom-only failures. Sensitivity analyses with nonparametric Kaplan-Meier curves (Figure 2), without covariate adjustment (Supplemental Figure 6), based on the full treated and per-protocol populations (Supplemental Figures 7 and 8) and frailty model with random site effect (Supplemental Table 1) yielded similar results.
      Figure thumbnail gr2
      Figure 2Failure probability for the composite primary outcome comparing hysteropexy to hysterectomy
      Failure probability from survival analysis in the intent-to-treat population by randomized intervention, excluding the 8 ineligible participants, was conducted using an interval-censored proportional hazard model with an assumed baseline piecewise exponential hazard with 5 constant hazard periods (0–12 months, 12–24 months, 24–36 months, 36–48 months, and 48–60 months) and controlled for site consistent with study randomization and previous prolapse surgery per clinical standards. Available follow-up data were included for all participants through 60 months of follow-up. At the time of analysis, 11 participants were censored before 36 months (7 in the hysteropexy group and 4 in the hysterectomy group), 7 participants were censored between 36 and 60 months (2 in the hysteropexy group and 5 in the hysterectomy group), and 86 were censored at 60 months (50 in the hysteropexy group and 36 in the hysterectomy group). The median (IQR) follow-up time was 60.0 (21.0–60.0) months for the hysteropexy group and 45.0 (9.0–60.0) for the hysterectomy group. The hazard rate in the hysteropexy group (incidence density per person-year) in year 1 was 0.15; year 2, 0.07; year 3, 0.06; year 4, 0.05; and year 5, 0.05. In the hysterectomy group, the hazard rate in year 1 was 0.26; year 2, 0.13; year 3, 0.11; year 4, 0.09; and year 5, 0.09. The solid lines represent the piecewise exponential model and the 95% CI for the piecewise exponential model is shown by the red shaded area for the hysteropexy group and the blue shaded area for the hysterectomy group. The failure probability from the sensitivity nonparametric interval-censored Kaplan-Meier analysis is represented by the dotted lines.
      • Sun J.
      A non-parametric test for interval-censored failure time data with application to AIDS studies.
      CI, confidence interval; IQR, interquartile range.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Table 2Secondary efficacy, safety/tolerability and masking outcomes
      OutcomeHysteropexy
      All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. All analysis for safety or tolerability and masking outcomes are conducted on the safety population (ie, all treated) and as such participants are analyzed by their received surgical intervention. Unless otherwise specified, risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang24 and the P values are from Fisher exact tests. Unless otherwise specified, P values for ordinal categorical measures are from Mantel-Haenszel chi-square tests using standardized midranks. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing the change from baseline of continuous outcomes of the intervention groups are obtained from general linear models adjusted for baseline outcome measure, site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing all other continuous outcomes of the intervention groups are obtained from general linear models adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. For all continuous outcomes, the n shown is for month 60; however, models were fitted using all participants with at least 1 postbaseline visit
      Hysterectomy
      All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. All analysis for safety or tolerability and masking outcomes are conducted on the safety population (ie, all treated) and as such participants are analyzed by their received surgical intervention. Unless otherwise specified, risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang24 and the P values are from Fisher exact tests. Unless otherwise specified, P values for ordinal categorical measures are from Mantel-Haenszel chi-square tests using standardized midranks. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing the change from baseline of continuous outcomes of the intervention groups are obtained from general linear models adjusted for baseline outcome measure, site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing all other continuous outcomes of the intervention groups are obtained from general linear models adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. For all continuous outcomes, the n shown is for month 60; however, models were fitted using all participants with at least 1 postbaseline visit
      Risk or mean difference (95% CI)
      All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. All analysis for safety or tolerability and masking outcomes are conducted on the safety population (ie, all treated) and as such participants are analyzed by their received surgical intervention. Unless otherwise specified, risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang24 and the P values are from Fisher exact tests. Unless otherwise specified, P values for ordinal categorical measures are from Mantel-Haenszel chi-square tests using standardized midranks. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing the change from baseline of continuous outcomes of the intervention groups are obtained from general linear models adjusted for baseline outcome measure, site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing all other continuous outcomes of the intervention groups are obtained from general linear models adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. For all continuous outcomes, the n shown is for month 60; however, models were fitted using all participants with at least 1 postbaseline visit
      P value
      All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. All analysis for safety or tolerability and masking outcomes are conducted on the safety population (ie, all treated) and as such participants are analyzed by their received surgical intervention. Unless otherwise specified, risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang24 and the P values are from Fisher exact tests. Unless otherwise specified, P values for ordinal categorical measures are from Mantel-Haenszel chi-square tests using standardized midranks. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing the change from baseline of continuous outcomes of the intervention groups are obtained from general linear models adjusted for baseline outcome measure, site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing all other continuous outcomes of the intervention groups are obtained from general linear models adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. For all continuous outcomes, the n shown is for month 60; however, models were fitted using all participants with at least 1 postbaseline visit
      Efficacy outcomes in Intent-to-treat populationN=88N=87
      Cumulative failures (Time-to-Event)
      Under the permanent failure state assumption, the number of participants at each visit includes all participants who were still participating in the study or had a failure outcome before withdrawal. The denominator for the cumulative failure at each visit includes all participants who were still participating in the study and attended the visit or had any failure outcome before their withdrawal/missed visit. The failure type corresponding to the first failure time is shown, where the 4 failure types are mutually exclusive, with retreatment failure prioritized above the anatomic and/or symptomatic failure types
      1-y visit(n=88)(n=86)
      Failure, number/total (%)
      All missing data were considered to be missing completely at random and were excluded from both the numerator and denominator in the calculation of estimates and CIs. Point estimates and 95% CIs for risk difference in surgical failures between the treatment groups are based on Mantel-Haenszel estimates for the common risk difference stratified by site with Wald-type CIs with the estimate of the variance of the risk difference based on Sato.25 As these analyses are considered supportive of the primary analysis, P values are not shown
      13/86 (15)21/85 (25)−9 (−21 to 2)
      Initial failure type
      Retreatment0 (0)1 (1)
      Anatomic and symptom failure5 (6)2 (2)
      Anatomic failure only7 (8)16 (19)
      Symptom failure only1 (1)2 (2)
      2-y visit(n=87)(n=85)
      Failure, number/total (%)
      All missing data were considered to be missing completely at random and were excluded from both the numerator and denominator in the calculation of estimates and CIs. Point estimates and 95% CIs for risk difference in surgical failures between the treatment groups are based on Mantel-Haenszel estimates for the common risk difference stratified by site with Wald-type CIs with the estimate of the variance of the risk difference based on Sato.25 As these analyses are considered supportive of the primary analysis, P values are not shown
      19/79 (24)28/83 (34)−10 (−24 to 4)
      Initial failure type
      Retreatment0 (0)1 (1)
      Anatomic and symptom failure7 (9)3 (4)
      Anatomic failure only10 (13)20 (24)
      Symptom failure only2 (3)4 (5)
      3-y visit(n=85)(n=85)
      Failure, number/total (%)
      All missing data were considered to be missing completely at random and were excluded from both the numerator and denominator in the calculation of estimates and CIs. Point estimates and 95% CIs for risk difference in surgical failures between the treatment groups are based on Mantel-Haenszel estimates for the common risk difference stratified by site with Wald-type CIs with the estimate of the variance of the risk difference based on Sato.25 As these analyses are considered supportive of the primary analysis, P values are not shown
      24/78 (31)33/80 (41)−10 (−25 to 4)
      Initial failure type
      Retreatment0 (0)1 (1)
      Anatomic and symptom failure9 (12)6 (8)
      Anatomic failure only12 (15)22 (28)
      Symptom failure only3 (4)4 (5)
      4-y visit(n=83)(n=83)
      Failure, number/total (%)
      All missing data were considered to be missing completely at random and were excluded from both the numerator and denominator in the calculation of estimates and CIs. Point estimates and 95% CIs for risk difference in surgical failures between the treatment groups are based on Mantel-Haenszel estimates for the common risk difference stratified by site with Wald-type CIs with the estimate of the variance of the risk difference based on Sato.25 As these analyses are considered supportive of the primary analysis, P values are not shown
      26/76 (34)39/79 (49)−14 (−29 to 1)
      Initial failure type
      Retreatment0 (0)1 (1)
      Anatomic and symptom failure9 (12)6 (8)
      Anatomic failure only13 (17)27 (34)
      Symptom failure only4 (5)5 (6)
      5-y visit(n=79)(n=78)
      Failure, number/total (%)
      All missing data were considered to be missing completely at random and were excluded from both the numerator and denominator in the calculation of estimates and CIs. Point estimates and 95% CIs for risk difference in surgical failures between the treatment groups are based on Mantel-Haenszel estimates for the common risk difference stratified by site with Wald-type CIs with the estimate of the variance of the risk difference based on Sato.25 As these analyses are considered supportive of the primary analysis, P values are not shown
      29/79 (37)42/78 (54)−18 (−33 to −3)
      Initial failure type
      Retreatment0 (0)1 (1)
      Anatomic and symptom failure9 (11)6 (8)
      Anatomic failure only15 (19)30 (39)
      Symptom failure only5 (6)5 (6)
      Prolapse retreatment through 5 y
      Any retreatment, number/total (%)
      Retreatment for prolapse and treatment for urinary incontinence through 5 years are identified based on detailed treatment information collected at follow-up visits 6 months through 5 years
      8/87 (9)11/86 (13)−4 (−13 to 6).45
      Surgery retreatment, number/total (%)
      Retreatment for prolapse and treatment for urinary incontinence through 5 years are identified based on detailed treatment information collected at follow-up visits 6 months through 5 years
      7/87 (8)9/86 (11)−2 (−11 to 6).58
      Pessary retreatment only, number/total (%)
      Retreatment for prolapse and treatment for urinary incontinence through 5 years are identified based on detailed treatment information collected at follow-up visits 6 months through 5 years
      1/87 (1)2/86 (2)−1 (−7 to 4).62
      Pelvic organ prolapse quantification examination
      For definitions of the Pelvic organ Prolapse Quantification measurements, refer to footnote c in Table 1
      at 5 y
      (n=76)(n=72)
      Ba, Adjusted mean (95% CI), cm−1.3 (−1.6 to −1.0)−0.9 (−1.2 to −0.6)−0.4 (−0.8 to 0.1).10
      Bp, Adjusted mean (95% CI), cm−1.7 (−2.0 to −1.4)−2.0 (−2.3 to −1.7)0.3 (−0.1 to 0.7).14
      C, Adjusted mean (95% CI), cm−5.8 (−6.3 to −5.4)−5.7 (−6.1 to −5.2)−0.2 (−0.8 to 0.5).61
      TVL, Adjusted mean (95% CI), cm8.4 (8.2 to 8.7)7.5 (7.2 to 7.7)0.9 (0.6 to 1.3)<.001
      C ≥−0.5 × TVL, number/total (%)
      Adjusted risk differences, 95% CIs, and P values comparing the proportions of binary outcomes of the intervention groups was planned to be obtained from generalized linear models with identity link adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits as autoregressive order 1. This specified model did not converge, and unadjusted risk difference at 60 months was obtained
      5/76 (7)4/72 (6)1 (−8 to 10)>.99
      Patient-reported functional efficacy outcomes
      The PGI-I score ranges from 1 (very much better) to 7 (very much worse) and has been dichotomized as better (≤2) or worse (>2). Refer to Table 1 footnotes for ranges and clinical interpretation of the other patient-reported measures
      Pelvic floor symptoms at 5 y
      Pelvic Floor Distress Inventory (PFDI) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−78.7 (−86.1 to −71.3)−83.8 (−91.4 to −76.2)5.2 (−5.3 to 15.6).33
      Pelvic Floor Impact Questionnaire (PFIQ) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−41.3 (−47.6 to −34.9)−48.1 (−54.7 to −41.6)6.9 (−2.2 to 15.9).14
      Prolapse symptoms at 5 y
      Patients Global Impression of Improvement (PGI-I, much better or very much better), number/total (%)
      Adjusted risk differences, 95% CIs, and P values comparing the proportions of binary outcomes of the intervention groups was planned to be obtained from generalized linear models with identity link adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits as autoregressive order 1. This specified model did not converge, and unadjusted risk difference at 60 months was obtained
      68/76 (90)69/72 (96)−6 (−15 to 2).14
      Pelvic Organ Prolapse Distress Inventory (POPDI) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−40.6 (−43.4 to −37.7)−41.6 (−44.5 to −38.7)1.1 (−2.9 to 5.1).59
      Pelvic Organ Prolapse Impact Questionnaire (POPIQ) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−17.2 (−19.2 to −15.1)−18.9 (−21.0 to −16.8)1.7 (−1.2 to 4.6).24
      Urinary function at 5 y
      Urogenital Distress Inventory (UDI) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−23.7 (−27.3 to −20.1)−30.2 (−33.9 to −26.5)6.5 (1.4 to 11.7).01
      Urinary Impact Questionnaire (UIQ) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−18.6 (−21.6 to −15.7)−19.8 (−22.8 to −16.8)1.1 (−3.0 to 5.3).59
      Incontinence Severity Index (ISI) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−1.1 (−1.7 to −0.4)−1.6 (−2.2 to −0.9)0.5 (−0.4 to 1.4).30
      New or worsening stress urinary incontinence, number/total (%)
      New or worsening stress urinary incontinence, urgency urinary incontinence, voiding dysfunction, and fecal incontinence are identified based on specific complications collected at follow-up visits 6 weeks through 5 years
      18/88 (21)10/87 (12)9 (−2 to 20).11
      New or worsening urgency urinary incontinence, number/total (%)
      New or worsening stress urinary incontinence, urgency urinary incontinence, voiding dysfunction, and fecal incontinence are identified based on specific complications collected at follow-up visits 6 weeks through 5 years
      22/88 (25)17/87 (20)5 (−7 to 18).39
      New or worsening voiding dysfunction, number/total (%)
      New or worsening stress urinary incontinence, urgency urinary incontinence, voiding dysfunction, and fecal incontinence are identified based on specific complications collected at follow-up visits 6 weeks through 5 years
      7/88 (8)7/87 (8)0 (−8 to 8).98
      Treatment for urinary incontinence through 5 y, number/total (%)
      Retreatment for prolapse and treatment for urinary incontinence through 5 years are identified based on detailed treatment information collected at follow-up visits 6 months through 5 years
      16/87 (18)16/86 (19)0 (−12 to 11).97
      Bowel function at 5 y
      Colorectal Anal Distress Inventory (CRADI) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−14.3 (−17.1 to −11.5)−12.0 (−14.8 to −9.1)−2.3 (−6.2 to 1.6).24
      Colorectal Anal Impact Questionnaire (CRAIQ) score change from baseline(n=76)(n=73)
      Adjusted mean (95% CI)−5.6 (−8.4 to −2.9)−9.3 (−12.2 to −6.5)3.7 (−0.2 to 7.6).06
      New or worsening fecal incontinence, number/total (%)
      New or worsening stress urinary incontinence, urgency urinary incontinence, voiding dysfunction, and fecal incontinence are identified based on specific complications collected at follow-up visits 6 weeks through 5 years
      14/88 (16)7/87 (8)8 (−2 to 17).11
      Sexual function at 5 y
      Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR) average score among sexually active women change from baseline(n=14)(n=23)
      Adjusted mean (95% CI)0.4 (0.2–0.6)0.3 (0.1–0.5)0.1 (−0.2 to 0.3).53
      New or worsening dyspareunia, number/total (%)
      New or worsening dyspareunia identified based on specific complications collected at follow-up visits 6 months through 5 year
      8/87 (9)2/86 (2)7 (0 to 15).10
      General quality of life at 5 y
      Functional Activity Assessment Scale (FAS) score change from baseline(n=76)(n=72)
      Adjusted mean (95% CI)2.6 (0.3–4.8)4.4 (2.1–6.7)−1.8 (−5.0 to 1.3).25
      Body Image Scale (BIS) score(n=76)(n=73)
      Adjusted mean (95% CI)−4.5 (−5.1 to −4.0)−4.0 (−4.6 to −3.4)−0.5 (−1.4 to 0.3).20
      Short Form Health Survey (SF-12) physical component score change from baseline(n=73)(n=72)
      Adjusted mean (95% CI)2.2 (0.3–4.1)3.9 (1.9–5.8)−1.7 (−4.4 to 1.0).22
      Short Form Health Survey (SF-12) mental component score change from baseline(n=73)(n=72)
      Adjusted mean (95% CI)1.4 (−0.4 to 3.2)1.2 (−0.6 to 3.1)0.2 (−2.4 to 2.7).89
      Safety/Tolerability and Masking Outcomes in Safety PopulationN=91N=92
      Mesh Exposure/Erosion Surgery through 5 y
      Mesh exposure and mesh erosion are identified based on specific complications/adverse events collected at follow-up visits 6 weeks through 5 years and the systematic collection of open-ended adverse events from surgery through 5 years. There were no intraoffice, inpatient, or outpatient surgeries for any hysteropexy and/or midurethral sling mesh exposures. 6 hysteropexy mesh exposures were prescribed estrogen cream and 1 was asymptomatic. All 2 midurethral sling mesh exposures were asymptomatic
      Hysteropexy mesh exposure, number/total (%)7/91 (8)
      Binomial (95% CI)
      Binomial 95% CIs for proportions are based on the exact Clopper-Pearson method
      (3–15)
      Hysteropexy mesh erosion, number/total (%)0/91 (0)
      Midurethral sling mesh exposure, number/total (%)0/43 (0)2/47 (4)−4 (−15 to 4).50
      Midurethral sling mesh erosion, number/total (%)0/43 (0)0/47 (0)
      Maximum modified Dindo score surgery through 5 y
      Maximum modified Dindo score is identified across all visits surgery through 5 years based on the reported Dindo score corresponding to the most severe complication at the visit. Participants experiencing no complications from surgery through 5 years are assigned a grade of 0. Grades I through V are assigned as follows:Grade I: Any deviation from the normal intraoperative or postoperative course without the need for pharmacologic treatment or surgical, endoscopic, and radiological interventions. Allowed therapeutic regimens are: drugs as antiemetics, antipyretics, analgesics, diuretics, electrolytes, and physiotherapy. This grade also includes wound infections opened at the bedside.Grade IIa: Oral administration of drugs other than such allowed for grade I, excluding antibiotics.Grade IIabx: Oral administration of drugs other than such allowed for grade I, including antibiotics.Grade IIb: IV administration of drugs other than such allowed for grade I, including antibiotics; blood transfusions and total parenteral nutrition are also included.Grade IIIo: Additional surgical measures required during SUPeR procedure.Grade IIIa: Intervention not under general anesthesia.Grade IIIb: Intervention under general anesthesia.Grade IVa: Single organ dysfunction (including dialysis).Grade IVb: Multiorgan dysfunction.Grade V: Death of a patient.
      Maximum Dindo score, number/total (%).33
      No complication (Grade 0)20/91 (22)15/92 (16)
      Grade I21/91 (23)22/92 (24)
      Grade IIa11/91 (12)8/92 (9)
      Grade IIabx22/91 (24)26/92 (28)
      Grade IIb3/91 (3)5/92 (5)
      Grade IIIo1/91 (1)5/92 (5)
      Grade IIIa4/91 (4)3/92 (3)
      Grade IIIb7/91 (8)6/92 (7)
      Grade IVa2/91 (2)2/92 (2)
      Grade IVb0/91 (0)0/92 (0)
      Grade V0/91 (0)0/92 (0)
      Maximum Dindo score ≥III, number/total (%)14/91 (15)16/92 (17)−2 (−13 to 9).71
      Complications/Adverse Events Surgery through 5 y
      Uterine complications are identified based on the systematic collection of open-ended adverse events from surgery through 5 years. Pelvic infections/abscess, UTI, neuromuscular disorder, and fistula formation are identified based on specific complications or adverse events collected at visits surgery through 5 years and the systematic collection of open-ended adverse events from surgery through 5 years
      Uterine complications, number/total (%)
      The uterine complications identified were cervical polyp (n=1), endometrial polyp (n=1), hematometria requiring a D and C (n=1), and thickened endometrium on imaging with no identified tissue on D and C/ hysteroscopy with subsequent normal endometrial stripe on imaging (n=1)
      4/91 (4)
      Binomial (95% CI)
      Binomial 95% CIs for proportions are based on the exact Clopper-Pearson method
      (1 to 11)
      Pelvic infections/abscess, number/total (%)0/91 (0)0/92 (0)
      Urinary tract infections (UTI), number/total (%)38/91 (42)43/92 (47)−5 (−19 to 9).50
      Neuromuscular disorder, number/total (%)7/91 (8)5/92 (5)2 (−5 to 9).54
      Fistula formation, number/total (%)0/91 (0)0/92 (0)
      Complications/Adverse Events >12 wks through 5 y
      Excessive granulation tissue, suture exposure, suture erosion, pelvic pain, daily pelvic pain, vaginal infections or atypical vaginal discharge, vaginal bleeding, vaginal scarring, and vaginal shortening are identified based on specific complications/adverse events collected at visits 6 months through 5 years and the systematic collection of open-ended adverse events from 12 weeks after surgery through 5 years
      Excessive granulation tissue, number/total (%)1/90 (1)11/91 (12)−11 (−18 to −4).003
      Suture exposure, number/total (%)3/90 (3)19/91 (21)−18 (−27 to −8)<.001
      Suture erosion, number/total (%)0/90 (0)0/91 (0)
      Pelvic pain, number/total (%)6/90 (7)9/91 (10)−3 (−11 to 5).43
      Daily pelvic pain, number/total (%)3/90 (3)4/91 (4)−1 (−8 to 6)>.99
      Vaginal infections or atypical vaginal discharge, number/total (%)4/90 (4)9/91 (10)−5 (−13 to 2).16
      Vaginal bleeding, number/total (%)5/90 (6)2/91 (2)3 (−3 to 11).28
      Vaginal scarring, number/total (%)3/90 (3)0/91 (0)3 (−1 to 9).12
      Vaginal shortening, number/total (%)0/90 (0)0/91 (0)
      Evaluation of masking at 5 y
      Remained masked, number/total (%)66/91 (73)63/92 (69)4 (−9 to 17).55
      Correctly stated what procedure they underwent, number/total (%)
      Of the 25 women in the hysteropexy group and 29 in the hysterectomy group who reported they had become unmasked, 10 and 8 in each group respectively did not provide a guess as to what procedure she underwent.
      13/15 (87)20/21 (95)−9 (−36 to 13).56
      CI, confidence interval; IUGA, International Urogynecological Association.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      a All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. All analysis for safety or tolerability and masking outcomes are conducted on the safety population (ie, all treated) and as such participants are analyzed by their received surgical intervention. Unless otherwise specified, risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang
      • Chan I.S.
      • Zhang Z.
      Test-based exact confidence intervals for the difference of two binomial proportions.
      and the P values are from Fisher exact tests. Unless otherwise specified, P values for ordinal categorical measures are from Mantel-Haenszel chi-square tests using standardized midranks. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing the change from baseline of continuous outcomes of the intervention groups are obtained from general linear models adjusted for baseline outcome measure, site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. Unless otherwise specified, adjusted means, mean difference, and 95% CIs, and P values comparing all other continuous outcomes of the intervention groups are obtained from general linear models adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits with an autoregressive order 1. For all continuous outcomes, the n shown is for month 60; however, models were fitted using all participants with at least 1 postbaseline visit
      b Under the permanent failure state assumption, the number of participants at each visit includes all participants who were still participating in the study or had a failure outcome before withdrawal. The denominator for the cumulative failure at each visit includes all participants who were still participating in the study and attended the visit or had any failure outcome before their withdrawal/missed visit. The failure type corresponding to the first failure time is shown, where the 4 failure types are mutually exclusive, with retreatment failure prioritized above the anatomic and/or symptomatic failure types
      c All missing data were considered to be missing completely at random and were excluded from both the numerator and denominator in the calculation of estimates and CIs. Point estimates and 95% CIs for risk difference in surgical failures between the treatment groups are based on Mantel-Haenszel estimates for the common risk difference stratified by site with Wald-type CIs with the estimate of the variance of the risk difference based on Sato.
      • Sato T.
      On the variance estimator for the Mantel-Haenszel risk difference.
      As these analyses are considered supportive of the primary analysis, P values are not shown
      d Retreatment for prolapse and treatment for urinary incontinence through 5 years are identified based on detailed treatment information collected at follow-up visits 6 months through 5 years
      e For definitions of the Pelvic organ Prolapse Quantification measurements, refer to footnote c in Table 1
      f Adjusted risk differences, 95% CIs, and P values comparing the proportions of binary outcomes of the intervention groups was planned to be obtained from generalized linear models with identity link adjusted for site, intervention, visit, and interaction between intervention and visit, while modeling the within-participant correlations across visits as autoregressive order 1. This specified model did not converge, and unadjusted risk difference at 60 months was obtained
      g The PGI-I score ranges from 1 (very much better) to 7 (very much worse) and has been dichotomized as better (≤2) or worse (>2). Refer to Table 1 footnotes for ranges and clinical interpretation of the other patient-reported measures
      h New or worsening stress urinary incontinence, urgency urinary incontinence, voiding dysfunction, and fecal incontinence are identified based on specific complications collected at follow-up visits 6 weeks through 5 years
      i New or worsening dyspareunia identified based on specific complications collected at follow-up visits 6 months through 5 year
      j Mesh exposure and mesh erosion are identified based on specific complications/adverse events collected at follow-up visits 6 weeks through 5 years and the systematic collection of open-ended adverse events from surgery through 5 years. There were no intraoffice, inpatient, or outpatient surgeries for any hysteropexy and/or midurethral sling mesh exposures. 6 hysteropexy mesh exposures were prescribed estrogen cream and 1 was asymptomatic. All 2 midurethral sling mesh exposures were asymptomatic
      k Binomial 95% CIs for proportions are based on the exact Clopper-Pearson method
      l Maximum modified Dindo score is identified across all visits surgery through 5 years based on the reported Dindo score corresponding to the most severe complication at the visit. Participants experiencing no complications from surgery through 5 years are assigned a grade of 0. Grades I through V are assigned as follows:Grade I: Any deviation from the normal intraoperative or postoperative course without the need for pharmacologic treatment or surgical, endoscopic, and radiological interventions. Allowed therapeutic regimens are: drugs as antiemetics, antipyretics, analgesics, diuretics, electrolytes, and physiotherapy. This grade also includes wound infections opened at the bedside.Grade IIa: Oral administration of drugs other than such allowed for grade I, excluding antibiotics.Grade IIabx: Oral administration of drugs other than such allowed for grade I, including antibiotics.Grade IIb: IV administration of drugs other than such allowed for grade I, including antibiotics; blood transfusions and total parenteral nutrition are also included.Grade IIIo: Additional surgical measures required during SUPeR procedure.Grade IIIa: Intervention not under general anesthesia.Grade IIIb: Intervention under general anesthesia.Grade IVa: Single organ dysfunction (including dialysis).Grade IVb: Multiorgan dysfunction.Grade V: Death of a patient.
      m Uterine complications are identified based on the systematic collection of open-ended adverse events from surgery through 5 years. Pelvic infections/abscess, UTI, neuromuscular disorder, and fistula formation are identified based on specific complications or adverse events collected at visits surgery through 5 years and the systematic collection of open-ended adverse events from surgery through 5 years
      n The uterine complications identified were cervical polyp (n=1), endometrial polyp (n=1), hematometria requiring a D and C (n=1), and thickened endometrium on imaging with no identified tissue on D and C/ hysteroscopy with subsequent normal endometrial stripe on imaging (n=1)
      o Excessive granulation tissue, suture exposure, suture erosion, pelvic pain, daily pelvic pain, vaginal infections or atypical vaginal discharge, vaginal bleeding, vaginal scarring, and vaginal shortening are identified based on specific complications/adverse events collected at visits 6 months through 5 years and the systematic collection of open-ended adverse events from 12 weeks after surgery through 5 years
      p Of the 25 women in the hysteropexy group and 29 in the hysterectomy group who reported they had become unmasked, 10 and 8 in each group respectively did not provide a guess as to what procedure she underwent.
      Longitudinal patterns of success and failure are demonstrated in Figure 3. Most participants (66.7% of the hysteropexy group and 51.2% of the hysterectomy group) met the definition of success at all time points and were classified into the “persistent success” group. Participants that met criteria for failure and stayed in a state of failure, the “persistent failure” group, included 16.1% of the hysteropexy patients and 22.1% of the hysterectomy patients. In each treatment group, intermittent failure/success was more common (17.2% of the hysteropexy patients and 26.7% of the hysterectomy patients) than persistent failure, confirming that success and failure after prolapse surgery are more dynamic than the primary outcome figure shows. More than 60% of participants categorized as intermittent failure/success were terminal successes at their last evaluation (Supplemental Figure 9). The primary outcome may overreport failure rates in both groups based on these results.
      Figure thumbnail gr3
      Figure 3Distribution of the composite surgical failure outcome over time
      The intent-to-treat population in the hysteropexy group consists of 88 participants; however, 1 participant discontinued without attending any visits after 6 weeks and is not included in this figure. The intent-to-treat population in the hysterectomy group consists of 87 participants; however, 1 participant discontinued without attending any visits after 6 weeks and is not included in this figure.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Secondary efficacy, safety, and masking outcomes through 5 years are presented in Table 2. Adjusted mean total vaginal length (cm) was longer in the hysteropexy vs hysterectomy group (8.4 (95% CI, 8.2–8.7) vs 7.5 (95% CI, 7.2–7.7) difference, 0.9; 95% CI, 0.6–1.3; P<.001). In both the hysteropexy and hysterectomy groups, most women reported being much better or very much better on the Patient Global Impression of Improvement at 5 years (hysteropexy, 90%; hysterectomy, 96%; risk difference, −6%; 95% CI, −15% to 2%; P=.14). With the exception of the Urogenital Distress Inventory (UDI), no group differences were demonstrated in patient-reported pelvic floor symptoms, prolapse symptoms, bowel function symptoms, general quality of life, body image, or pelvic pain.
      Notably, 40% of women (70 of 175) were sexually active before surgery (Supplemental Table 2). As measured by the PISQ-IR total score (Table 2), sexual function among the women who remained sexually active at 5 years improved after surgery by an adjusted mean and 95% CI of 0.4 (0.2–0.6) in the hysteropexy group and 0.3 (0.1–0.5) in the hysterectomy group (adjusted mean difference, 0.1; 95% CI, −0.2 to 0.3; P=.53). Dyspareunia prevalence rates among sexually active women decreased from 39% (10 of 26) to 11% (2 of 19) in the hysteropexy group and from 46% (17 of 37) to 17% (4 of 23) in the hysterectomy group (Supplemental Table 2). New or worsening dyspareunia identified based on specific complications collected at follow-up visits 6 months through 5 years was 9% (8 of 87) in the hysteropexy group and 2% (2 of 86) in the hysterectomy group (risk difference, 7%; 95% CI, 0–15; P=.10) (Table 2). De novo dyspareunia as measured by the PISQ-IR was only recorded in 3 hysteropexy and 1 hysterectomy participants (Supplemental Table 2).
      At their last visit through 5 years, 70% of participants (129 of 183) reported they remained masked to their treatment with no difference in masking between groups (risk difference, 4%; 95% CI, −9% to 17%; P=.55) (Table 2). Of participants who stated that they knew their treatment group, 92% (33 of 36) were correct, with no evidence of a group difference (risk difference, −9%; 95% CI, −36% to 13%; P=.56). The most common sources of unmasking were insurance paperwork, imaging studies, and conversations with providers not on the study team.

      Adverse events

      Hysteropexy had an 8% mesh exposure rate which was unchanged from 3 years. Of the 7 patients with hysteropexy mesh exposures, 3 were initially discovered at 6 weeks, 3 discovered at 6 months, and 1 discovered at 18 months. All patients were treated with vaginal estrogen cream and 1 patient underwent office mesh trimming. Only 2 hysteropexy mesh exposures were still present at 60 months. Based on observations starting 12 weeks after surgery, granulation tissue and suture exposure were less common in the hysteropexy group than the hysterectomy group (1% vs 12% and 3% vs 21%, respectively). Of the 12 patients with granulation tissue, 10 were discovered at 6 months and 2 at 12 months. Notably, 4 patients were treated with office silver nitrate and the rest were expectantly managed. Of the 22 suture exposure patients, 12 were discovered at 6 months, 6 discovered at 12 months, 3 at 18 months, and 1 at 54 months; 3 patients required suture removal in the office. No mesh exposures, granulation tissue, or suture exposures required reoperation or management in an operating room. No hysteropexy patient developed any significant uterine pathology; 4 patients required evaluation for uterine bleeding, all with benign findings. The most common other adverse events for all 183 participants during the 5 years of follow-up were urinary tract infections and constipation with no evidence of treatment group differences (Table 2 and Supplemental Tables 3 and 4).

      Comment

      Principal findings

      This study found that women with symptomatic uterovaginal prolapse treated with sacrospinous hysteropexy with graft compared with vaginal hysterectomy with uterosacral ligament suspension had lower surgical failure rates through 5 years. Cumulative composite failure rates indicated that hysteropexy failure was approximately 18% lower than hysterectomy at 5 years. There were no meaningful differences in patient-reported outcomes between groups. Safety outcomes such as modified Dindo scores and most adverse events, including pain-related issues and dyspareunia, were not different between groups. Vaginal mesh exposures, unique to the hysteropexy group, occurred in approximately 8% of those participants; none required surgical intervention.

      Results

      In contrast to the previously reported 3-year analysis that found a nonsignificant (P=.06) difference in failure rates between treatment groups,
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      this 5-year analysis found failure rates were significantly lower following hysteropexy. The previous analysis’ point estimate and survival trends suggested hysteropexy’s superiority, but wide CIs precluded the ability to establish superiority. With additional follow-up, the 5-year cumulative composite failure rates for hysteropexy and hysterectomy more clearly diverged, demonstrating hysteropexy’s lower failure compared with hysterectomy. The current analysis, similar to the previous 3-year analysis,
      • Nager C.W.
      • Visco A.G.
      • Richter H.E.
      • et al.
      Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
      found sustained improvement in most patient-reported outcomes in both groups, including prolapse, bowel and bladder symptoms, and sexual function. The only parameter that differed between groups was the UDI score, favoring hysterectomy. Because secondary outcomes were unadjusted for multiple comparisons, the significance of these UDI differences are uncertain. No other measures, including other urinary parameters, differed between groups. Safety results were also largely unchanged from the 3-year report. Mesh exposures in the hysteropexy group remained stable at 8% as did suture exposure and excessive granulation tissue rates that more commonly occurred after hysterectomy. Pelvic pain events did not differ between groups and occurred in ≤10% of participants.

      Clinical implications

      In February 2019, the FDA convened an advisory committee soliciting expert input on vaginal mesh use. The committee concluded that to support a favorable benefit-to-risk ratio, transvaginal mesh for prolapse repair should be superior to native tissue repair at 36 months, with comparable safety outcomes.
      United States Food and Drug Administration
      Urogynecologic surgical mesh Implants.
      In April 2019, in the absence of superiority data, the FDA ordered manufacturers of transvaginal mesh for prolapse to stop selling and distributing their products.
      US Food and Drug Administration
      FDA’s Activities: urogynecologic Surgical Mesh.
      The demonstrated superiority of the mesh augmented transvaginal prolapse approach demonstrated at 60 months was not evident until after the 36 month primary endpoint, reflecting the rate of prolapse progression of native tissue. In designing the current trial, investigators sought to address the safety and efficacy issues raised in the 2011 FDA Safety Communication.
      Food and Drug Administration
      Urogynecologic surgical mesh: update on the safety and effectiveness of transvaginal placement for pelvic organ prolapse.
      This study found that at 5-year follow-up, the hysteropexy-specific complication of mesh exposure was 8%, similar or slightly lower than the 12% rate reported in a 2016 Cochrane Review.
      • Maher C.
      • Feiner B.
      • Baessler K.
      • Christmann-Schmid C.
      • Haya N.
      • Marjoribanks J.
      Transvaginal mesh or grafts compared with native tissue repair for vaginal prolapse.
      Conversely, the hysterectomy group manifested foreign body reactions in the form of suture exposure and the development of granulation tissue at the vaginal apex. None of the participants who developed granulation tissue or mesh or suture exposures required surgical revision. The 2011 FDA review raised concerns regarding occurrence of pelvic pain and increased adverse events related to vaginal mesh use for prolapse.
      Food and Drug Administration
      Urogynecologic surgical mesh: update on the safety and effectiveness of transvaginal placement for pelvic organ prolapse.
      In this study, vaginal scarring, vaginal shortening, pelvic pain, sexual function, and serious adverse events did not differ between the hysteropexy and hysterectomy groups through 5-year follow-up. The FDA order does not prevent surgeons from using mesh for transvaginal repair, and this study suggests there may be benefit.

      Strengths and limitations

      This study has strengths including the relevance of its efficacy and safety outcomes, the randomized trial design, participant masking, high participant retention rates, and the 5-year follow-up.
      This study does have limitations. First, women willing to participate in a randomized trial using mesh, a controversial intervention, and willing to be masked to treatment through 5 years may constitute a subset of patients with characteristics unrepresentative of all patients. Second, the findings are only applicable to amenorrheic women who have completed childbearing and not applicable to menstruating, premenopausal women. Third, many of these women, largely in their eighth decade of life at 5-year follow-up, were sexually inactive which potentially limits the power to compare sexual function between groups. Fourth, approximately 30% of participants were unmasked, and although most women remained masked (with similar numbers in each group), this could have biased the patient-reported outcomes.

      Conclusions

      Through 5-year follow-up in women with symptomatic uterovaginal prolapse randomized to prolapse repair with sacrospinous hysteropexy with graft or vaginal hysterectomy and uterosacral vaginal suspension, the hysteropexy group had fewer failures. Serious adverse events and patient-reported outcomes did not differ between the hysteropexy and hysterectomy groups through 5-year follow-up.

      Appendix

      Figure thumbnail fx1
      Supplemental Figure 1Sacrospinous hysteropexy with (polypropylene) graft procedure
      This drawing represents the female bony pelvis and relevant surgical anatomy for the sacrospinous hysteropexy with graft procedure using polypropylene mesh. The upper pink structure depicts the uterus and the lower pink structure is the vagina. The blue hatched area depicts the mesh supporting the cervix and upper anterior vagina by its attachment to both sacrospinous ligaments (illustration by Jasmine Tan-Kim, MD).
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx2
      Supplemental Figure 2Vaginal hysterectomy with (bilateral) ipsilateral uterosacral ligament suspension procedure
      This drawing represents the female bony pelvis and relevant surgical anatomy for the vaginal hysterectomy with ipsilateral uterosacral ligament suspension procedure. The uterus has been removed. The lower pink structure is the vagina. The blue suture represents the permanent suture that is placed during the procedure and the green suture depicts the absorbable suture that attached the upper vagina to both uterosacral ligaments. Eventually this absorbable suture dissolves (illustration by Jasmine Tan-Kim, MD).
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx3
      Supplemental Figure 3Pelvic organ prolapse quantification schematic
      The table and schematic depict the locations of the various POP-Q points along the vaginal and perineal topography.
      POP-Q, Pelvic Organ Prolapse Quantification.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx4
      Supplemental Figure 4Smoothed hazard rates for the composite primary outcome comparing hysteropexy to hysterectomy in the intent-to-treat population by randomized intervention
      The Epanechnikov kernel–smoothed hazard rates shown are from the nonparametric interval-censored Kaplan-Meier analysis in . Overall, the hazard rate in the hysterectomy group is higher than the hysteropexy group except in the period of approximately 18 to 30 months where the rates appear similar. The spike at 60 months is caused by the high censoring rate relative to failure rate among the at risk population (50 of 52 censored in the hysteropexy group and 36 of 38 censored in the hysterectomy group).
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx5
      Supplemental Figure 5Estimated log cumulative hazard by log-time for the composite primary outcome comparing hysteropexy with hysterectomy in the intent-to-treat population by randomized intervention
      The estimated log cumulative hazard rates shown are from the nonparametric interval-censored Kaplan-Meier analysis in . Overall, the log cumulative hazard rates are higher in the hysterectomy group than the hysteropexy group with approximately linear (slope of 1) and parallel trajectories in each of the following log periods (log 0–12 months, log 12–24 months, log 36–48 months, and log 48–60 months) supportive of the proportional hazards assumption and the assumed piecewise exponential distribution. Although the linearity and the parallelism of the curves appear deviate from this assumption during log 24–36 months representing a departure from the proportionality assumption, this departure can be characterized as quantitative rather than qualitative because the log cumulative hazard curves do not cross during this period.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx6
      Supplemental Figure 6Sensitivity analysis: unadjusted failure probability for the composite primary outcome comparing hysteropexy to hysterectomy in the intent-to-treat population by randomized intervention
      Unadjusted failure probability from survival analysis excluding the 8 ineligible participants was conducted using an interval-censored proportional hazard model with an assumed baseline piecewise exponential hazard with 5 constant hazard periods: 0–12 months, 12–24 months, 24–36 months, 36–48 months, and 48–60 months. Available follow-up data were included for all participants through 60 months of follow-up. At the time of analysis, 11 participants were censored before 36 months (7 in the hysteropexy group and 4 in the hysterectomy group), 7 participants were censored between 36 and 60 months (2 in the hysteropexy group and 5 in the hysterectomy group), and 86 were censored at 60 months (50 in the hysteropexy group and 36 in the hysterectomy group). The median (IQR) follow-up time was 60.0 (21.0–60.0) months for the hysteropexy group and 45.0 (9.0–60.0) for the hysterectomy group. The hazard rate in the hysteropexy group (incidence density per person-year) in year 1 was 0.17; year 2, 0.08; year 3, 0.07; year 4, 0.06; and year 5, 0.05. In the hysterectomy group, the hazard rate in year 1 was 0.28; year 2, 0.13; year 3, 0.11; year 4, 0.09; and year 5, 0.09. The solid lines represent the piecewise exponential model and the 95% CI for the piecewise exponential model is shown by the red shaded area for the hysteropexy group and the blue shaded area for the hysterectomy group. The failure probability from the sensitivity nonparametric interval-censored Kaplan-Meier analysis is represented by the dotted lines.
      • Sun J.
      A non-parametric test for interval-censored failure time data with application to AIDS studies.
      CI, confidence interval; HR, hazard ratio; IQR, interquartile range.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx7
      Supplemental Figure 7Sensitivity analysis: failure probability for the composite primary outcome comparing hysteropexy to hysterectomy in the treated population by randomized intervention
      Failure probability from survival analysis including the 8 ineligible participants was conducted using an interval-censored proportional hazard model with an assumed baseline piecewise exponential hazard with 5 constant hazard periods (0–12 months, 12–24 months, 24–36 months, 36–48 months, and 48–60 months) and controlled for site consistent with study randomization and previous prolapse surgery per clinical standards. Available follow-up data were included for all participants through 60 months of follow-up. At the time of analysis, 11 participants were censored before 36 months (7 in the hysteropexy group and 4 in the hysterectomy group), 7 participants were censored between 36 and 60 months (2 in the hysteropexy group and 5 in the hysterectomy group), and 90 were censored at 60 months (52 in the hysteropexy group and 38 in the hysterectomy group). The median (IQR) follow-up time was 60.0 (21.0–60.0) months for the hysteropexy group and 46.5 (9.0–60.0) for the hysterectomy group. The hazard rate in the hysteropexy group (incidence density per person-year) in year 1 was 0.15; year 2, 0.08; year 3, 0.06; year 4, 0.05; and year 5, 0.06. In the hysterectomy group, the hazard rate in year 1 was 0.24; year 2, 0.14; year 3, 0.10; year 4, 0.09; and year 5, 0.09. The solid lines represent the piecewise exponential model and the 95% CI for the piecewise exponential model is shown by the red shaded area for the hysteropexy group and the blue shaded area for the hysterectomy group. The failure probability from the sensitivity nonparametric interval-censored Kaplan-Meier analysis is represented by the dotted lines.
      • Sun J.
      A non-parametric test for interval-censored failure time data with application to AIDS studies.
      CI, confidence interval; HR, hazard ratio; IQR, interquartile range.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx8
      Supplemental Figure 8Sensitivity analysis: failure probability for the composite primary outcome comparing hysteropexy to hysterectomy in the per-protocol population by randomized intervention
      Failure probability from survival analysis excluding the 8 ineligible participants and the 2 eligible participants who did not receive surgery as randomized was conducted using an interval-censored proportional hazard model with an assumed baseline piecewise exponential hazard with 5 constant hazard periods (0–12 months, 12–24 months, 24–36 months, 36–48 months, and 48–60 months) and controlled for site consistent with study randomization and previous prolapse surgery per clinical standards. Available follow-up data were included for all participants through 60 months of follow-up. At the time of analysis, 11 participants were censored before 36 months (7 in the hysteropexy group and 4 in the hysterectomy group), 7 participants were censored between 36 and 60 months (2 in the hysteropexy group and 5 in the hysterectomy group), and 85 were censored at 60 months (49 in the hysteropexy group and 36 in the hysterectomy group). The median (IQR) follow-up time was 60.0 (21.0–60.0) months for the hysteropexy group and 45.0 (9.0–60.0) for the hysterectomy group. The hazard rate in the hysteropexy group (incidence density per person-year) in year 1 was 0.14; year 2, 0.07; year 3, 0.06; year 4, 0.05; and year 5, 0.05. In the hysterectomy group, the hazard rate in year 1 was 0.25; year 2, 0.13; year 3, 0.11; year 4, 0.09; and year 5, 0.09. The solid lines represent the piecewise exponential model and the 95% CI for the piecewise exponential model is shown by the red shaded area for the hysteropexy group and the blue shaded area for the hysterectomy group. The failure probability from the sensitivity nonparametric interval-censored Kaplan-Meier analysis is represented by the dotted lines.
      • Sun J.
      A non-parametric test for interval-censored failure time data with application to AIDS studies.
      CI, confidence interval; HR, hazard ratio; IQR, interquartile range.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Figure thumbnail fx9
      Supplemental Figure 9Distribution of composite surgical failure outcome among participants with intermittent failures in the intent-to-treat population by randomized intervention
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Supplemental Table 1Sensitivity analysis: hazard ratio for the composite primary outcome comparing hysteropexy to hysterectomy with random site effect in the intent-to-treat population by randomized intervention
      Sensitivity analysisHazard ratio (95% CI)P value
      Frailty model with random site effect0.61 (0.38–0.98).04
      To assess the sensitivity of the primary outcome results with site as a random effect, the following approach was taken. First, a piecewise exponential model was fitted with fixed site effects but with the interval-censored events fixed at the midpoint of the interval. This model produced essentially the same hazard ratio point and interval estimate and P value as the initial interval-censored model (Figure 2). Then, this same model was fitted as a lognormal frailty model with random site effect. The point estimates of this frailty model were nearly equal to those for the primary model but the confidence interval was slightly wider.
      CI, confidence interval.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      Supplemental Table 2Sexual function and dyspareunia results from Pelvic Organ Prolapse Incontinence Sexual Function Questionnaire-IR a in the intent-to-treat population by randomized intervention
      Outcome
      Some participants responded to the sexually active item on the PISQ-IR, but did not respond to the corresponding dyspareunia item on the PISQ-IR
      Hysteropexy (n=88)Hysterectomy (n=87)Risk difference (95% CI)
      All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. Risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang (Chan, IS, Biometrics 1999;55: 1202–9) and the P values are from Fisher exact tests
      P value
      All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. Risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang (Chan, IS, Biometrics 1999;55: 1202–9) and the P values are from Fisher exact tests
      Baseline, n/N (%)
       Sexually active30/88 (34)40/87 (46)−12 (−26 to 3).11
       Dyspareunia
      The Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR). At the time point of interest, dyspareunia among sexually active is defined as experiencing pain (sometimes, usually, or always have) during sexual intercourse. At the time point of interest, dyspareunia among not sexually active is defined as not engaging in sexual intercourse owing to pain (strongly agree or somewhat agree) during sexual intercourse
      29/84 (35)32/82 (39)−5 (−19 to 10).55
      Among sexually active10/26 (39)17/37 (46)−7 (−32 to 17).56
      Among not sexually active19/58 (33)15/45 (33)−1 (−19 to 18).95
      5 years, n/N (%)
       Sexually active23/75 (31)25/71 (35)−5 (−20 to 11).56
       Dyspareunia
      The Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR). At the time point of interest, dyspareunia among sexually active is defined as experiencing pain (sometimes, usually, or always have) during sexual intercourse. At the time point of interest, dyspareunia among not sexually active is defined as not engaging in sexual intercourse owing to pain (strongly agree or somewhat agree) during sexual intercourse
      13/69 (19)6/62 (10)9 (−3 to 21).14
      Among sexually active2/19 (11)4/23 (17)−7 (−30 to 18).67
      Among not sexually active11/50 (22)2/39 (5)17 (3–30).03
       De novo dyspareunia
      De novo dyspareunia is defined as the absence of dyspareunia at baseline and the prevalence of dyspareunia at follow-up time point of interest. In the hysteropexy group, 1 participant with de novo dyspareunia at 5 years changed sexual activity status from active at baseline to not active at 5 years.
      3/65 (5)1/62 (2)3 (−5 to 11).62
      Among sexually active0/18 (0)1/23 (4)−4 (−23 to 14)>.99
      Among not sexually active3/47 (6)0/39 (0)6 (−3 to 18).25
      CI, confidence interval; IUGA, International Urogynecological Association.
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      a Some participants responded to the sexually active item on the PISQ-IR, but did not respond to the corresponding dyspareunia item on the PISQ-IR
      b All analysis for efficacy outcomes are conducted on the intent-to-treat population and as such participants are analyzed by their randomized surgical intervention. Risk differences, 95% CIs, and P values are unadjusted and based on Mantel-Haenszel estimates for the risk difference with Wald-type CIs for nominal categorical measures with expected cell counts >5, otherwise the exact risk difference and 95% CI limits are obtained by exact methods based on the score statistic based on Chan and Zhang (Chan, IS, Biometrics 1999;55: 1202–9) and the P values are from Fisher exact tests
      c The Pelvic Organ Prolapse Incontinence Sexual Questionnaire-IUGA Revised (PISQ-IR). At the time point of interest, dyspareunia among sexually active is defined as experiencing pain (sometimes, usually, or always have) during sexual intercourse. At the time point of interest, dyspareunia among not sexually active is defined as not engaging in sexual intercourse owing to pain (strongly agree or somewhat agree) during sexual intercourse
      d De novo dyspareunia is defined as the absence of dyspareunia at baseline and the prevalence of dyspareunia at follow-up time point of interest. In the hysteropexy group, 1 participant with de novo dyspareunia at 5 years changed sexual activity status from active at baseline to not active at 5 years.
      Supplemental Table 3Adverse events by system organ class and preferred term
      System Organ Class and Preferred Term are coded using Medical Dictionary for Regulatory Activities (MedDRA)
      in safety population by received surgical intervention
      System organ class
      Systematic collection of reportable open-ended adverse events throughout the course of the trial are presented in descending order of frequency for System Organ Class grouping
      Preferred termHysteropexy (n=91)

      Evt
      For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      n
      For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      (%)
      Hysterectomy (n=92)

      Evt
      For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      n
      For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      (%)
      Total (N=183)

      Evt
      For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      n
      For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      (%)
      AnyAny62 37 (41)99 53 (58)161 90 (49)
      Infections and infestationsAny21 21 (23)29 23 (25)50 44 (24)
      Urinary tract infection18 18 (20)22 22 (24)40 40 (22)
      Vaginal infection0 0 (0)2 2 (2)2 2 (1)
      Wound infection1 1 (1)1 1 (1)2 2 (1)
      Cellulitis0 0 (0)1 1 (1)1 1 (1)
      Clostridium difficile colitis0 0 (0)1 1 (1)1 1 (1)
      Lyme disease1 1 (1)0 0 (0)1 1 (1)
      Pneumonia0 0 (0)1 1 (1)1 1 (1)
      Pyelonephritis1 1 (1)0 0 (0)1 1 (1)
      Vulvovaginal mycotic infection0 0 (0)1 1 (1)1 1 (1)
      Gastrointestinal disordersAny6 6 (7)16 11 (12)22 17 (9)
      Constipation3 3 (3)6 6 (7)9 9 (5)
      Levator syndrome2 2 (2)1 1 (1)3 3 (2)
      Colitis0 0 (0)2 2 (2)2 2 (1)
      Abdominal distension1 1 (1)0 0 (0)1 1 (1)
      Abdominal hernia0 0 (0)1 1 (1)1 1 (1)
      Colitis ischemic0 0 (0)1 1 (1)1 1 (1)
      Diverticular perforation0 0 (0)1 1 (1)1 1 (1)
      Peritoneal hemorrhage0 0 (0)1 1 (1)1 1 (1)
      Peritoneal perforation0 0 (0)1 1 (1)1 1 (1)
      Small intestinal obstruction0 0 (0)1 1 (1)1 1 (1)
      Vomiting0 0 (0)1 1 (1)1 1 (1)
      Reproductive system and breast disordersAny6 5 (5)11 11 (12)17 16 (9)
      Pelvic pain1 1 (1)2 2 (2)3 3 (2)
      Vulvovaginal adhesion1 1 (1)2 2 (2)3 3 (2)
      Vulvovaginal pruritus2 2 (2)1 1 (1)3 3 (2)
      Vulvovaginal pain0 0 (0)2 2 (2)2 2 (1)
      Atrophic vulvovaginitis1 1 (1)0 0 (0)1 1 (1)
      Bartholin's cyst0 0 (0)1 1 (1)1 1 (1)
      Pelvic hematoma0 0 (0)1 1 (1)1 1 (1)
      Perineal pain0 0 (0)1 1 (1)1 1 (1)
      Rectocele0 0 (0)1 1 (1)1 1 (1)
      Vaginal hemorrhage1 1 (1)0 0 (0)1 1 (1)
      General disorders and administration site conditionsAny6 5 (5)5 5 (5)11 10 (5)
      Medical device site reaction5 4 (4)5 5 (5)10 9 (5)
      Local swelling1 1 (1)0 0 (0)1 1 (1)
      Skin and subcutaneous tissue disordersAny5 5 (5)6 6 (7)11 11 (6)
      Excessive granulation tissue4 4 (4)6 6 (7)10 10 (5)
      Lichen sclerosus1 1 (1)0 0 (0)1 1 (1)
      Injury, poisoning and procedural complicationsAny3 3 (3)7 7 (8)10 10 (5)
      Ureteric injury0 0 (0)3 3 (3)3 3 (2)
      Bladder injury1 1 (1)1 1 (1)2 2 (1)
      Femoral nerve injury0 0 (0)1 1 (1)1 1 (1)
      Foreign body0 0 (0)1 1 (1)1 1 (1)
      Postprocedural hemorrhage0 0 (0)1 1 (1)1 1 (1)
      Procedural complication1 1 (1)0 0 (0)1 1 (1)
      Vulvovaginal injury1 1 (1)0 0 (0)1 1 (1)
      Nervous system disordersAny4 4 (4)5 4 (4)9 8 (4)
      Sciatica3 3 (3)3 2 (2)6 5 (3)
      Lacunar infarction0 0 (0)1 1 (1)1 1 (1)
      Spondylitic myelopathy0 0 (0)1 1 (1)1 1 (1)
      Subarachnoid hemorrhage1 1 (1)0 0 (0)1 1 (1)
      Renal and urinary disordersAny2 2 (2)5 5 (5)7 7 (4)
      Bladder spasm0 0 (0)2 2 (2)2 2 (1)
      Cystitis interstitial0 0 (0)1 1 (1)1 1 (1)
      Dysuria1 1 (1)0 0 (0)1 1 (1)
      Hydronephrosis1 1 (1)0 0 (0)1 1 (1)
      Micturition urgency0 0 (0)1 1 (1)1 1 (1)
      Nocturia0 0 (0)1 1 (1)1 1 (1)
      Musculoskeletal and connective tissue disordersAny3 3 (3)3 3 (3)6 6 (3)
      Arthralgia0 0 (0)1 1 (1)1 1 (1)
      Back pain1 1 (1)0 0 (0)1 1 (1)
      Flank pain1 1 (1)0 0 (0)1 1 (1)
      Groin pain1 1 (1)0 0 (0)1 1 (1)
      Musculoskeletal pain0 0 (0)1 1 (1)1 1 (1)
      Spinal column stenosis0 0 (0)1 1 (1)1 1 (1)
      Respiratory, thoracic, and mediastinal disordersAny3 2 (2)3 3 (3)6 5 (3)
      Asthma0 0 (0)1 1 (1)1 1 (1)
      Atelectasis0 0 (0)1 1 (1)1 1 (1)
      Bronchiectasis1 1 (1)0 0 (0)1 1 (1)
      Pulmonary embolism1 1 (1)0 0 (0)1 1 (1)
      Pulmonary edema0 0 (0)1 1 (1)1 1 (1)
      Respiratory failure1 1 (1)0 0 (0)1 1 (1)
      Vascular disordersAny2 1 (1)2 2 (2)4 3 (2)
      Deep vein thrombosis2 1 (1)0 0 (0)2 1 (1)
      Granulomatosis with polyangiitis0 0 (0)1 1 (1)1 1 (1)
      Phlebitis0 0 (0)1 1 (1)1 1 (1)
      Cardiac disordersAny0 0 (0)2 2 (2)2 2 (1)
      Acute myocardial infarction0 0 (0)1 1 (1)1 1 (1)
      Arrhythmia0 0 (0)1 1 (1)1 1 (1)
      Hepatobiliary disordersAny0 0 (0)1 1 (1)1 1 (1)
      Cholecystitis acute0 0 (0)1 1 (1)1 1 (1)
      InvestigationsAny0 0 (0)1 1 (1)1 1 (1)
      Blood culture positive0 0 (0)1 1 (1)1 1 (1)
      Metabolism and nutrition disordersAny0 0 (0)1 1 (1)1 1 (1)
      Dehydration0 0 (0)1 1 (1)1 1 (1)
      Neoplasms benign, malignant, and unspecified (including cysts and polyps)Any0 0 (0)1 1 (1)1 1 (1)
      Colon neoplasm0 0 (0)1 1 (1)1 1 (1)
      Psychiatric disordersAny0 0 (0)1 1 (1)1 1 (1)
      Mental status changes0 0 (0)1 1 (1)1 1 (1)
      Surgical and medical proceduresAny1 1 (1)0 0 (0)1 1 (1)
      Ureteral stent insertion1 1 (1)0 0 (0)1 1 (1)
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      a System Organ Class and Preferred Term are coded using Medical Dictionary for Regulatory Activities (MedDRA)
      b Systematic collection of reportable open-ended adverse events throughout the course of the trial are presented in descending order of frequency for System Organ Class grouping
      c For each adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      Supplemental Table 4Serious adverse events by system organ class and preferred term
      System Organ Class and Preferred Term are coded using Medical Dictionary for Regulatory Activities (MedDRA)
      in safety population by received surgical intervention
      System organ class
      Systematic collection of reportable open-ended adverse events throughout the course of the trial are presented in descending order of frequency for System Organ Class grouping
      Preferred termHysteropexy (n=91)

      Evt
      For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      n
      For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      (%)
      Hysterectomy (n=92)

      Evt
      For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      n
      For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      (%)
      Total (N=183)

      Evt
      For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      n
      For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.
      (%)
      AnyAny7 5 (100)20 16 (100)27 21 (100)
      Gastrointestinal disordersAny0 0 (0)6 5 (31)6 5 (24)
      Colitis0 0 (0)2 2 (13)2 2 (10)
      Colitis ischemic0 0 (0)1 1 (6)1 1 (5)
      Constipation0 0 (0)1 1 (6)1 1 (5)
      Diverticular perforation0 0 (0)1 1 (6)1 1 (5)
      Peritoneal hemorrhage0 0 (0)1 1 (6)1 1 (5)
      Nervous system disordersAny1 1 (20)2 2 (13)3 3 (14)
      Lacunar infarction0 0 (0)1 1 (6)1 1 (5)
      Spondylitic myelopathy0 0 (0)1 1 (6)1 1 (5)
      Subarachnoid hemorrhage1 1 (20)0 0 (0)1 1 (5)
      Reproductive system and breast disordersAny0 0 (0)3 3 (19)3 3 (14)
      Pelvic hematoma0 0 (0)1 1 (6)1 1 (5)
      Rectocele0 0 (0)1 1 (6)1 1 (5)
      Vulvovaginal pain0 0 (0)1 1 (6)1 1 (5)
      Respiratory, thoracic and mediastinal disordersAny2 2 (40)1 1 (6)3 3 (14)
      Asthma0 0 (0)1 1 (6)1 1 (5)
      Pulmonary embolism1 1 (20)0 0 (0)1 1 (5)
      Respiratory failure1 1 (20)0 0 (0)1 1 (5)
      Vascular disordersAny2 1 (20)1 1 (6)3 2 (10)
      Deep vein thrombosis2 1 (20)0 0 (0)2 1 (5)
      Granulomatosis with polyangiitis0 0 (0)1 1 (6)1 1 (5)
      Infections and infestationsAny2 2 (40)0 0 (0)2 2 (10)
      Lyme disease1 1 (20)0 0 (0)1 1 (5)
      Pyelonephritis1 1 (20)0 0 (0)1 1 (5)
      Musculoskeletal and connective tissue disordersAny0 0 (0)2 2 (13)2 2 (10)
      Musculoskeletal pain0 0 (0)1 1 (6)1 1 (5)
      Spinal column stenosis0 0 (0)1 1 (6)1 1 (5)
      Acute myocardial infarction0 0 (0)1 1 (6)1 1 (5)
      Cardiac disordersAny0 0 (0)1 1 (6)1 1 (5)
      Hepatobiliary disordersAny0 0 (0)1 1 (6)1 1 (5)
      Cholecystitis acute0 0 (0)1 1 (6)1 1 (5)
      Injury, poisoning and procedural complicationsAny0 0 (0)1 1 (6)1 1 (5)
      Postprocedural hemorrhage0 0 (0)1 1 (6)1 1 (5)
      Neoplasms benign, malignant and unspecified (including cysts and polyps)Any0 0 (0)1 1 (6)1 1 (5)
      Colon neoplasm0 0 (0)1 1 (6)1 1 (5)
      Renal and urinary disordersAny0 0 (0)1 1 (6)1 1 (5)
      Bladder spasm0 0 (0)1 1 (6)1 1 (5)
      Nager et al. Sacrospinous hysteropexy with graft vs vaginal hysterectomy with uterosacral suspension. Am J Obstet Gynecol 2021.
      a System Organ Class and Preferred Term are coded using Medical Dictionary for Regulatory Activities (MedDRA)
      b Systematic collection of reportable open-ended adverse events throughout the course of the trial are presented in descending order of frequency for System Organ Class grouping
      c For each serious adverse event, the total number of events is presented (Evt) and the distinct number of participants experiencing each adverse event (n) by received surgical intervention.

      References

        • Wu J.M.
        • Matthews C.A.
        • Conover M.M.
        • Pate V.
        • Jonsson Funk M.
        Lifetime risk of stress urinary incontinence or pelvic organ prolapse surgery.
        Obstet Gynecol. 2014; 123: 1201-1206
        • Meriwether K.V.
        • Antosh D.D.
        • Olivera C.K.
        • et al.
        Uterine preservation vs hysterectomy in pelvic organ prolapse surgery: a systematic review with meta-analysis and clinical practice guidelines.
        Am J Obstet Gynecol. 2018; 219: 129-146.e2
        • Madsen A.M.
        • Raker C.
        • Sung V.W.
        Trends in hysteropexy and apical support for uterovaginal prolapse in the United States from 2002 to 2012.
        Female Pelvic Med Reconstr Surg. 2017; 23: 365-371
        • Detollenaere R.J.
        • den Boon J.
        • Stekelenburg J.
        • et al.
        Sacrospinous hysteropexy versus vaginal hysterectomy with suspension of the uterosacral ligaments in women with uterine prolapse stage 2 or higher: multicentre randomised non-inferiority trial.
        BMJ. 2015; 351: h3717
        • Maher C.
        • Feiner B.
        • Baessler K.
        • Christmann-Schmid C.
        • Haya N.
        • Brown J.
        Surgery for women with apical vaginal prolapse.
        Cochrane Database Syst Rev. 2016; 10: CD012376
        • Jelovsek J.E.
        • Barber M.D.
        • 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: 1554-1565
        • Maher C.
        • Feiner B.
        • Baessler K.
        • Christmann-Schmid C.
        • Haya N.
        • Marjoribanks J.
        Transvaginal mesh or grafts compared with native tissue repair for vaginal prolapse.
        Cochrane Database Syst Rev. 2016; 2: CD012079
        • United States Food and Drug Administration
        Urogynecologic surgical mesh Implants.
        (Available at:) (Accessed Mar. 29, 2021)
        • Nager C.W.
        • Zyczynski H.
        • Rogers R.G.
        • et al.
        The design of a randomized trial of vaginal surgery for uterovaginal prolapse: vaginal hysterectomy With native tissue vault suspension Versus mesh hysteropexy suspension (the study of uterine prolapse procedures randomized trial).
        Female Pelvic Med Reconstr Surg. 2016; 22: 182-189
        • Nager C.W.
        • Visco A.G.
        • Richter H.E.
        • et al.
        Effect of vaginal mesh hysteropexy vs vaginal hysterectomy With uterosacral ligament suspension on treatment failure in women With uterovaginal prolapse: a randomized clinical trial.
        JAMA. 2019; 322: 1054-1065
        • Broglio K.
        Randomization in clinical trials: permuted blocks and stratification.
        JAMA. 2018; 319: 2223-2224
        • Bump R.C.
        • Mattiasson A.
        • Bø K.
        • et al.
        The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction.
        Am J Obstet Gynecol. 1996; 175: 10-17
        • Yalcin I.
        • Bump R.C.
        Validation of two global impression questionnaires for incontinence.
        Am J Obstet Gynecol. 2003; 189: 98-101
        • Barber M.D.
        • Kuchibhatla M.N.
        • Pieper C.F.
        • Bump R.C.
        Psychometric evaluation of 2 comprehensive condition-specific quality of life instruments for women with pelvic floor disorders.
        Am J Obstet Gynecol. 2001; 185: 1388-1395
        • Sandvik H.
        • Seim A.
        • Vanvik A.
        • Hunskaar S.
        A severity index for epidemiological surveys of female urinary incontinence: comparison with 48-hour pad-weighing tests.
        Neurourol Urodyn. 2000; 19: 137-145
        • McCarthy Jr., M.
        • Jonasson O.
        • Chang C.H.
        • et al.
        Assessment of patient functional status after surgery.
        J Am Coll Surg. 2005; 201: 171-178
        • Ware Jr., J.
        • Kosinski M.
        • Keller S.D.
        A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity.
        Med Care. 1996; 34: 220-233
        • Rogers R.G.
        • Rockwood T.H.
        • Constantine M.L.
        • et al.
        A new measure of sexual function in women with pelvic floor disorders (PFD): the Pelvic Organ prolapse/Incontinence Sexual Questionnaire, IUGA-revised (PISQ-IR).
        Int Urogynecol J. 2013; 24: 1091-1103
        • Hopwood P.
        • Fletcher I.
        • Lee A.
        • Al Ghazal S.
        A body image scale for use with cancer patients.
        Eur J Cancer. 2001; 37: 189-197
        • Jelovsek J.E.
        • Barber M.D.
        Women seeking treatment for advanced pelvic organ prolapse have decreased body image and quality of life.
        Am J Obstet Gynecol. 2006; 194: 1455-1461
        • Wellner J.A.
        • Zhan Y.H.
        A hybrid algorithm for computation of the nonparametric maximum likelihood estimator from censored data.
        J Am Stat Assoc. 1997; 92: 945-959
        • Sun J.
        A non-parametric test for interval-censored failure time data with application to AIDS studies.
        Stat Med. 1996; 15: 1387-1395
        • Rockwood T.H.
        • Constantine M.L.
        • Adegoke O.
        • et al.
        The PISQ-IR: considerations in scale scoring and development.
        Int Urogynecol J. 2013; 24: 1105-1122
        • Chan I.S.
        • Zhang Z.
        Test-based exact confidence intervals for the difference of two binomial proportions.
        Biometrics. 1999; 55: 1202-1209
        • Sato T.
        On the variance estimator for the Mantel-Haenszel risk difference.
        Biometrics. 1989; 45: 1323-1324
        • US Food and Drug Administration
        FDA’s Activities: urogynecologic Surgical Mesh.
        (Available at:)
        • Food and Drug Administration
        Urogynecologic surgical mesh: update on the safety and effectiveness of transvaginal placement for pelvic organ prolapse.
        (Available at:) (Accessed Mar. 29, 2021)