Materials and Methods

This was a retrospective cohort study of the first 120 patients who underwent placement of Avaulta Solo synthetic vaginal mesh system from January 2006 through April 2008 through the Division of Urogynecology and Reconstructive Pelvic Surgery at Atlantic Health. Atlantic Health is a tertiary care system comprised of 2 hospitals in northern New Jersey. The senior authors (P.J.C. and A.S.) performed all of the surgeries. The study group included all patients who underwent any vaginal prolapse repair incorporating the anterior, posterior, or combined Avaulta Solo systems during the above-mentioned time period. The Atlantic Health Institutional Review Board approved this protocol (#R07-09-016), which was posted on the World Wide Web site www.clinicaltrials.gov (identifier # NCT00774215 ). Prior to their 1 year objective and subjective postoperative assessments, all study patients signed the informed consent document generated by our institutional review board.

Our study group resulted from an obvious selection bias. During the study period at our center we tended to perform the Avaulta Solo procedure for patients who had ≥1 of the following characteristics: (1) were generally in the older age range; (2) had a rather specific isolated support defect; and/or (3) had significant medical comorbidities. At our center, patients who did not fall into ≥1 of these categories tended to be offered robotic assisted laparoscopic sacrocolpopexy. During the study period, we performed very few prolapse repairs that did not involve placement of some type of graft material.

Preoperatively, all patients had a comprehensive gynecologic examination including the pelvic organ prolapse quantification (POP-Q) system. Shortly after the beginning of the study period, we began asking all of our new patients to complete the validated short forms of the Pelvic Floor Distress Inventory, Short Form 20 (PFDI-20) and the Pelvic Floor Impact Questionnaire, Short Form 7 (PFIQ-7). Despite the absence of these preoperative data among the first few patients in the cohort, we decided to use these instruments to assess the subjective outcomes for this study. Both the PFDI-20 and PFIQ-7 are scored from 0-300, with a higher score indicating worse symptoms. The PFDI-20 consists of 3 subscales: the Pelvic Organ Prolapse Distress Inventory-6 (POPDI-6), the Colorectal-anal Distress Inventory-8, and the Urogenital Distress Inventory-6. The PFIQ-7 also consists of 3 subscales: the Urinary Impact Questionnaire-7, the Colorectal-anal Impact Questionnaire-7, and the Pelvic Organ Prolapse Impact Questionnaire-7. The answer to question 3 of the POPDI-6 that asks “Do you have a bulge or something falling out that you can see or feel in the vaginal area?” was also used independently as the subjective part of our definition of “surgical cure.” Regardless of objective POP-Q scores, women who answered this question affirmatively were classified as surgical failures.

At the attending surgeons’ discretion, concomitant vaginal hysterectomies were performed for a portion of the study group. When a vaginal hysterectomy was performed, the vaginal apex was supported via bilateral high uterosacral ligament suspension sutures using a previously described technique. If the correction of urinary stress incontinence was necessary as determined via symptom profiles and multichannel urodynamic testing, a retropubic midurethral sling was performed–always through a vaginal incision made separate from the Avaulta Solo incision. All patients received a single dose of preoperative prophylactic intravenous antibiotics and no routine follow-up antibiotics. The default preoperative antibiotic choice was cefazolin, or combination of gentamicin and clindamycin for those patients who reported penicillin allergies.

All patients received general anesthesia, and were positioned in a modified dorsal lithotomy position. A dilute vasopressin solution was injected beneath the full thickness of the vaginal epithelium to develop either the vesicovaginal or rectovaginal spaces as needed. To place the anterior Avaulta Solo device, a vertical midline incision through the full thickness of the vaginal epithelium was made at the most dependent aspect of the protruding anterior vaginal wall. The endopelvic connective tissue was sharply separated from the vaginal epithelium–first to the pubic ramus and then down to the level of the ischial spines. An anterior colporrhaphy was then performed using a delayed absorbable suture in an interrupted technique.

Bilateral small groin incisions were made at the most superior and inferior aspects of the medial border of the obturator foramen. The Avaulta Solo trocars were then sequentially passed through these incisions, with the superior incisions serving as the site for the distal mesh arms, and the inferior incisions for the proximal arms. Care was taken to place the proximal arms of the mesh through the obturator internus muscles approximately 1 cm above the ischial spines.

The mesh was loosely positioned by gently pulling on each arm, and then tacked down to the anterior colporrhaphy in the midline using delayed absorbable sutures. Cystoscopy was performed to make sure that no lower urinary tract injuries had occurred. At most, a minimal amount of vaginal epithelium was trimmed–with the goal simply being to freshen up these edges. The vaginal epithelium was then closed in a running fashion before the final tensioning of the mesh was performed. The final adjustment of mesh tension was performed such that the apex and anterior vaginal walls were supported in a neutral position and mesh arms were not under any tension.

For placement of the posterior Avaulta Solo devices, a similar hydrodissection was performed followed by a vertical incision through the full thickness of the vaginal epithelium–with care taken to enter the true rectovaginal space. Bilateral stab incisions were made in the buttocks 3 cm lateral to and 3 cm distal from the anus. Trocars were passed through these stab incisions horizontally into the ischiorectal fossa under finger guidance. The trocars were placed through the levator muscles just lateral to the rectum and just medial to the ischial spines, and the superior mesh arms were then set into place. The same buttocks incisions were used for the distal trocar passes. These passes were also performed via finger guidance–allowing the surgeon to pass the trocar from the buttocks to the perineal body bilaterally. The distal landmark for trocar placement was the junction between the bulbocavernosus and transverse perineal muscles. The mesh was then positioned and tacked down in the midline via delayed absorbable sutures. After each posterior needle pass (and before the mesh was actually pulled into place) a digital rectal examination was performed to make sure that no penetration of the rectum had occurred. A final rectal examination was performed to verify that no mesh was palpable in the rectum and that no tension could be felt along the mesh arms. For patients in whom both the anterior and posterior devices were placed, the worst compartment was repaired first. In other words, the leading edge of the given prolapse was dealt with as the first step of the combined anterior/posterior cases. If a suburethral sling was required, it was performed through a separate vaginal incision. A Foley catheter and vaginal pack were left in place until the morning after surgery. All patients underwent a voiding trial the day after surgery. In general, patients who received only the anterior procedure had no measurable posterior compartment prolapse symptoms or relaxation.

During the study period (July 2007) the product manufacturer introduced a second-generation Avaulta Solo that was very similar to the first-generation product except for an improvement in the trocar design. Because we believed that the new trocar design was intuitively superior to that of the first-generation product, we exclusively used the second-generation Avaulta Solo once it was available. It just so happened that this second-generation product was used on exactly one half of the patients in our study group. It should also be noted that our group never tried the sister product, called the Avaulta Plus Biosynthetic Support System, which was released at the same time. That product differs from the Avaulta Solo in that it has a porous acellular sheet of cross-linked porcine collagen affixed to the polypropylene mesh.

Each patient was evaluated ≥1 year postoperatively. At that time, the PFDI-20 and PFIQ-7 were completed for all patients. We also administered a validated surgical satisfaction questionnaire. All patients also underwent a gynecologic examination including the POP-Q assessment–performed by a physician other than the original operating surgeon. In addition to POP-Q assessments, these examinations also focused on identifying mesh erosion and mesh-related vaginal/pelvic pain.

Surgical cure at ≥12 months was defined by considering subjective and objective findings simultaneously. A patient with any POP-Q point >0 or any reports of a vaginal bulge on the PFDI-20 was considered a surgical failure. We classified the rest of the patients as having had surgical cure. Secondary outcomes included the rates of mesh erosion and new-onset vaginal or pelvic pain. Lastly, anatomic success and complication rates between the 2 generations of the Avaulta Solo were compared. A given patient was classified as having a mesh erosion if any amount of any visible or palpable mesh material was discovered by physical examination at any postoperative point.

Statistical analysis was performed using software (SAS 9.1; SAS Institute Inc, Cary, NC). Preoperative and postoperative POP-Q points and questionnaire values were compared using paired t tests. Comparisons between proportions were done using χ ² tests. Values between independent groups were compared using 2-sample independent t tests.