Regional Performance of Apical Support Procedures at Time of Hysterectomy for Benign Indications: What Is the Role of Surgeon Training?

A total of 12,345 hysterectomies for benign indications were performed during the study period: 11,165 (90.4%) without the diagnosis of prolapse and 1,180 (9.6%) with prolapse ( Table 1 ). Baseline characteristics were analyzed between groups. The majority of patients were white (no prolapse 69.3%; prolapse 90.0%), privately insured (no prolapse 69.9%; prolapse 54.9%), and received care at a teaching hospital (no prolapse 56.5%; prolapse 55.5%). Of the 11,165 hysterectomies performed without the diagnosis of prolapse during the study period, only 119 (1.1%) had concomitant APS performed. Most patients were ≤50 years old, white, with private insurance, and underwent care at teaching institutions ( Table 1 ). ASPs were more likely to be performed with increasing age and subsequent year of analysis. The majority of hysterectomies were via the total laparoscopic (44.5%) or total abdominal (27.3%) route when an ASP was not performed. In contrast, in cases that included a concomitant ASP, hysterectomies were more often performed via the laparoscopic-assisted vaginal (23.5%) or total vaginal (31.1%) route. General Obstetrician Gynecologists (Ob/Gyn) or Gynecologic Oncologists (Gyn Onc) performed the majority of hysterectomies in both groups (Ob/Gyn: no ASP 66.9%; ASP 52.9%; Gyn Onc: no ASP 24.1%; ASP 37.0%). The most common indications for hysterectomy were leiomyoma (no ASP 31.6%; ASP 18.3%) and abnormal uterine bleeding (no ASP 28.1%; ASP 25.2%). Multivariable logistic regression analysis was performed to determine factors associated with performance of concomitant ASP in patients undergoing hysterectomy without a diagnosis of prolapse, and results are displayed in Table 2 . Factors associated with performance of concomitant ASP at time of benign hysterectomy were hysterectomy route and surgeon type. ASP were less likely to be performed at non-teaching hospitals (OR 0.64, CI 0.45–0.93). Patients were almost 4 times more likely to undergo ASP with laparoscopic assisted vaginal hysterectomy and supracervical hysterectomy and 21 times more likely with total vaginal hysterectomy when compared to total laparoscopic hysterectomy. Gynecologic Oncologists were almost 3 times more likely to perform ASP and Urogynecologists 74 times more likely to perform concurrent ASP when compared to general Obstetrician Gynecologists. Of the 1,180 hysterectomies performed with the diagnosis of prolapse during the study period, the majority (n=613, 51.9%) did not include an ASP ( Table 1 ). Most women were white with private insurance. Concomitant ASP were more often performed at teaching institutions (73.7%) and were more likely to be performed with increasing age and subsequent year of analysis. Most hysterectomies were total vaginal hysterectomies (no ASP 50.2%; ASP 40.2%) and the majority were performed by general Obstetrician Gynecologists (no ASP 86.3%; ASP 57.5%). The most common indication for hysterectomy was UVP (76.8%). On multivariable logistic regression analysis, performance of hysterectomy with concomitant ASP was associated with age, hysterectomy route, and surgeon type in patients with a diagnosis of prolapse ( Table 2 ). ASP were 2 to 5 times more likely to be performed with older age (61+) compared to ages 18–40. ASP were 5 times more likely to be performed with supracervical hysterectomies compared to total laparoscopic hysterectomy. Minimally invasive gynecologic surgeons were 6 times more likely to perform concomitant ASP and Urogynecologists were almost 11 times more likely to perform concomitant ASP when compared to general Gynecologists. Secondary analyses were performed excluding all supracervical hysterectomies as this route of hysterectomy does not compromise level 1 support. Results from this analysis were similar to findings with inclusion of supracervical hysterectomy and thus not fully described.

All hysterectomies performed for benign indications at all institutions in the Barnes Jewish Corporation (BJC) Healthcare System from 2011 to 2017 were identified by electronic search of all hospital discharges that included a code for hysterectomy occurring within the hospitalization. BJC Healthcare is a large, nonprofit, healthcare organization comprised of 15 hospitals in Missouri and Illinois and includes teaching and non-teaching facilities. International Classification of Diseases (ICD), Ninth and Tenth Revisions, Clinical Modification (ICD-9-CM, ICD-10-PCS) procedure codes and Current Procedural Terminology (CPT) codes were used to identify eligible cases based on surgical indication and procedure. Due to changes in coding during the timeframe of interest, ICD-9-CM procedure codes were used to identify eligible cases from 2011 to 2015, after which time ICD-10-PCS and CPT codes were utilized. Cases with an associated diagnosis or admission code for a urogenital (ICD-9-CM: 179–184, 188–189, 236; ICD-10-CM: C53-C58, C67-C68), bowel (ICD-9-CM: 152–159; ICD-10-CM: C17-C26), or other intraabdominal malignancy including neuroendocrine tumors (ICD-9-CM: 209; ICD-10-CM: C7A-C7B) were excluded. Cases were divided based on whether uterovaginal prolapse was included in the admission or procedure diagnosis (ICD-9-CM: 618, 618.01, 618.02, 618.04, 618.09, 618.1, 618.2, 618.3, 618.4, 618.5, 618.6, 618.89, 618.9; ICD-10-CM: N81.10, N81.11, N81.12, N81.2, N81.3, N81.4, N81.6, N81.89, N81.9) and analyzed separately. Uterovaginal prolapse was chosen as a proxy for apical descent due to limitations associated with using ICD and CPT codes. The primary outcome was performance of an apical support procedure (ICD-9-CM: 69.22, 70.4, 70.8, 70.64, 70.77, 70.78, 70.92; ICD-10-PCS: OUS44ZZ, OUS90ZZ, OUTG7ZZ, OUQF7ZZ, OJQC3ZZ, OJQC0ZZ; CPT: 57268, 57270, 57282, 57283, 57425, 57280, 58263, 58270, 58280, 58292, 58294) at time of hysterectomy. Surgeon subspecialty was analyzed as a secondary outcome and included in the multivariable logistic regression analysis. Surgeons who self-identify as Female Pelvic Medicine and Reconstructive Surgeons (FPMRS) subspecialists were classified in this way, regardless of whether they had completed an FPMRS fellowship. Categorical variables were described using frequencies with percentages and compared using chi-squared tests. Multivariable logistic regression was performed to determine factors associated with performance of an apical support procedure. All analyses were performed using SAS version 9.4 (Cary, NC). This study was reviewed and approved by the Washington University School of Medicine Human Research Protection Office, approval #201710076, approved August 16, 2018.

Despite increasing evidence to support the apex at the time of hysterectomy for prolapse, the majority of hysterectomies performed in this regional medical system from 2011 to 2017 did not include an ASP. In the 2007 practice bulletin on pelvic organ prolapse, the American College of Obstetricians and Gynecologists stated that hysterectomy alone is not an adequate treatment for UVP and this was reaffirmed in updated versions. 6 , 10 A 2013 study using Medicare data reported that 10-year reoperation rates for pelvic organ prolapse were significantly lower after concomitant ASP was performed at the time of hysterectomy. 11 In spite of data to support ASP at the time of hysterectomy for UVP, 8 , 11 – 13 ASP were not routinely performed with hysterectomy in our study, even in patients with a preoperative diagnosis of UVP. Guidelines have been developed by the American Association of Gynecologic Laparoscopists (AAGL) and Royal Congress of Obstetricians and Gynaecologists in conjunction with the British Society of Urogynaecology for prevention of VVP after hysterectomy, including recommendations for performance of ASP at the time of hysterectomy for benign indications aside from UVP. These guidelines specifically recommend uterosacral ligament suspension during abdominal hysterectomy and McCall culdoplasty during vaginal hysterectomy, with uterosacral ligament suspension being an acceptable alternative to McCall culdoplasty during vaginal hysterectomy. 4 , 14 AAGL additionally recommends uterosacral ligament suspension at the time of laparoscopic hysterectomy. 13 The findings of our study reveal that an ASP is more likely to be performed by surgeons with fellowship training. Gynecology Oncologists and Minimally Invasive Gynecologic Surgeons were more likely to perform ASP at the time of hysterectomy for patients without and with a diagnosis of UVP, respectively. As expected, FPMRS subspecialists were more likely to perform ASP with hysterectomy both for UVP and other benign indications as FPMRS specializes in the clinical and surgical care of women with complex pelvic floor disorders including UVP. However, the rate of ASP performed by FPMRS at the time of hysterectomy in patients with a diagnosis of prolapse was not 100%. This is likely due to the inclusion of surgeons in this regional healthcare system that are declared FPMRS subtype, which include both physicians that were fellowship trained and board-certified and those that specialize in pelvic floor disorders but did not undergo a specific FPMRS fellowship training program and may or may not be board-certified. Similar outcomes have been demonstrated in other studies, including a study in ovarian cancer care showing that deviations from standard of care are more likely to occur in patients treated by non-specialty trained surgeons. 15 A study in 2017 found that fellowship-trained FPMRS surgeons were more likely to perform proposed quality measures in patients who had hysterectomy for pelvic organ prolapse when compared to grandfathered FPMRS surgeons and generalists. 16 Generalists still performed the most benign hysterectomies for patients with and without prolapse in this study, highlighting the importance of adequate recognition and treatment of apical prolapse in this group. In addition, this highlights an opportunity for education focusing on identifying apical prolapse, the role of apical prolapse in anterior vaginal wall prolapse, and treatment options that specifically address the apex. For most Obstetrics and Gynecology trainees, their only exposure to apical support procedures occurs during their FPMRS rotations and may not be emphasized during hysterectomies performed for non-prolapse indications. It is imperative that trainees are educated on the importance of apical support in overall vaginal support. Collaboration with FPMRS colleagues in a co-surgeon case model may be one way to address this, particularly for recent residency graduates once out in practice. Additionally, current practicing generalists may prefer to close the cuff vaginally after laparoscopic/robotic hysterectomies due to physician comfort with vaginal suturing versus laparoscopic/robotic suturing, which may require more laparoscopic technical skill. Previous studies have demonstrated evidence that vaginal cuff dehiscence rates are lower with vaginal cuff closure compared to laparoscopic or robotic cuff closure, 17 , 18 however newer data suggest laparoscopic/robotic closure may be superior. 19 If surgeons decide to close the cuff vaginally with incorporation of the uterosacral ligament for apical support after laparoscopic or robotic hysterectomy and adequate descent is present for vaginal cuff closure, they should consider a concomitant ASP or involving FPMRS colleagues who possess a higher level of training and comfort with these procedures as uterosacral stitch placement at the level of the vaginal cuff likely will not provide adequate support; typically more proximal stitch placement is needed. Often a modified McCall stitch as described by Shull et al, a high uterosacral ligament suspension, or a sacrospinous ligament suspension may be needed to provide adequate apical support. 20 – 23 This can and should be determined prior to surgery by systematically assessing for apical support loss utilizing the Pelvic Organ Prolapse-Quantification (POP-Q) examination and assessing the apex in both the supine and standing positions. At times, patients may need to be brought back for re-examination later in the day if they report prolapse symptoms but no significant prolapse is observed during the initial examination. If significant apical support loss is present, further discussion of treatment options may need to be expanded to include a sacrocolpopexy. These findings also support collaboration between general gynecologists and FPMRS in pre-operative evaluation and intra-operative management. For example, a general gynecologist who is planning for a vaginal hysterectomy for UVP may perform the vaginal hysterectomy and then have a FPMRS colleague perform the apical suspension procedure. Route of hysterectomy was additionally found to be associated with performance of ASP. In patients without a diagnosis of prolapse, ASP was almost 4 times more likely to be performed via laparoscopic assisted vaginal hysterectomy, 21 times more likely with vaginal hysterectomy, and just over 3 times more likely with supracervical hysterectomy. In patients with a diagnosis of prolapse, ASP was 5 times more likely to be performed with supracervical hysterectomy. Supracervical hysterectomy is associated with increased rates of ASP in both groups with and without prolapse as support procedures including abdominal modified McCall’s, other culdoplasty, or sacrocolpopexy are often performed concomitantly with supracervical hysterectomy. 4 Patients without a diagnosis of prolapse undergoing both laparoscopic-assisted vaginal hysterectomy or traditional vaginal hysterectomy are also more likely to have an ASP, which is likely due to the inherent uterine descent needed to perform these procedures. 24 – 26 ASP may not be performed in these patients if surgeons fail to recognize the role of apical descent in uterovaginal descent. Further education on recognition and diagnosis of prolapse is paramount for successful treatment. This study has many strengths, including a large volume of hysterectomies performed in a regional area which enables generalization to other metropolitan areas across the United States. The data set also includes surgeons of varying training and years of practice, which is also generalizable to patterns in other urban areas. Moreover, this is the first study to examine the trend of ASP performance at the time of benign hysterectomy that includes data on surgeon training. Limitations of this study include its retrospective design and data collected based on hospital-supplied diagnosis and procedure codes. While based on operative records, diagnosis and procedure codes are generated by billing staff, not the surgeons caring for the patients. This can lead to transcription errors or inaccurate classification of diagnoses and procedures. It is possible that misclassification occurred in patients with a diagnosis of prolapse, meaning they were analyzed within the incorrect group. Additionally, patients may have had ASP performed that were not coded appropriately but described within the operative note, or had miscoded ASP where the surgeon dictated a procedure that was assigned a code that does not technically qualify as a true ASP. Unfortunately, due to the use of coded data, cases were not able to be identified for individual review. The use of coded data does, however, allow the ability to capture considerably more data than with the standard chart review method. The procedures and diagnoses included in this study only represent the encounter for the surgery, and not follow up postoperatively. Further research is needed in long term follow-up of these patients to determine the true rate of incident and recurrent VVP in patients with versus without ASP performed at the time of hysterectomy for or not for prolapse. Our regional study evaluating hysterectomies performed for benign indications including with and without UVP demonstrates that an overall low number of patients had concomitant ASP performed, despite level 1 evidence recommending re-supporting the vaginal apex. Only 48% of patients with a diagnosis of prolapse had concomitant ASP. Ongoing educational efforts during training and postgraduate at the national and regional level on the importance of reestablishing apical vaginal support at time of hysterectomy is needed to prevent incident and recurrent post-hysterectomy vaginal vault prolapse. In addition, collaborative management between general gynecologists and FPMRS specialists may be a successful solution to address this issue.

Hysterectomy is the second most common surgical procedure performed for women in the United States. By the age of 60, more than one-third of women have had a hysterectomy. 1 The majority of hysterectomies are performed for benign gynecologic disease, with 17% of hysterectomies performed for prolapse. 2 Uterovaginal prolapse (UVP) is a multifactorial disorder caused by weakening or damage of uterine and vaginal support resulting in the descent of vaginal compartments in response to intraabdominal pressure. Total hysterectomy compromises level 1 support of the vaginal apex with detachment of the uterosacral ligaments from the vaginal apex. 3 Reattachment of the vaginal cuff to the uterosacral complex is necessary to re-establish level 1 support in these patients. 4 Lack of performance of apical support procedures (ASP) potentially predisposes women to future and recurrent vaginal vault prolapse (VVP). 5 The rates of VVP and surgical intervention for prolapse are expected to increase by almost 50% by 2050. 6 Despite level 1 evidence recommending apical support reestablishment, 5 previous studies have demonstrated that very few apical support procedures are performed concurrently. 7 , 8 A national study investigating rates of ASP performed at the time of benign hysterectomy concluded that a low number of concurrent ASPs were performed at the time of benign hysterectomy, regardless of whether there was an associated diagnosis of prolapse. 7 This study did not include surgeon training in the analysis. Additionally, regional studies have demonstrated only around half of cases with the diagnosis of uterovaginal prolapse and less than 5% of those without the diagnosis of prolapse had concomitant ASP performed at the time of hysterectomy. 8 , 9 The aim of this study was to determine factors associated with performance of apical support procedures at time of benign hysterectomy at a regional medical system, where surgeon-level data was available for analysis.