Hullfish, K L. MD1,3; Bovbjerg, V E. PhD2; Steers, W D. MD3
According to the 2003 U.S. Census and 2003 National Hospital Discharge Survery, nearly 200,000 women had surgery for pelvic organ prolapse (POP).1 This fact, coupled with the rapid increase in the aging U.S. population, has broad implications for those who provide health care to women. As in most countries of the world, there are more older women than older men in the United States, and the proportion of the population that is female increases with age. In 2000, women are estimated to account for 58% of the population age 65 and older and 70% of the population age 85 and older.2 Over the next 30 years, Luber et al3 estimated a 45% increase in demand for pelvic floor disorder services, which outpaces the population growth by twofold.
In the properly counseled patient who does not desire the ability to have vaginal intercourse, obliterative surgery is a viable operative treatment for pelvic organ prolapse. Our current research in outcomes highlights the need to address patient goals and expectations in the treatment of pelvic floor disorders.4,5 The main objectives of this study were to 1) describe attainment of preoperative goals in a cohort of women undergoing vaginal closure, 2) determine the association of goal attainment with disease-specific quality-of-life instruments, and 3) report postoperative patient satisfaction, regret, and complications.
MATERIALS AND METHODS
After obtaining institutional review board approval from the University of Virginia Human Investigation Committee, 94 women who underwent obliterative vaginal surgery to treat their POP were offered the opportunity to complete preoperative patient-described goals, pelvic floor dysfunction–specific quality-of-life scales (Urogenital Distress Inventory and Incontinence Impact Questionnaire), and demographic information. This information was obtained in a preprinted questionnaire, before each patient’s original consultation in our urogynecology clinic at the University of Virginia Health Center. Goals were solicited with the following question: “In an effort to better understand your current gynecologic or urologic problem, would you please list your GOALS that you have for treating your condition, along with what you consider to be the ‘best’ and ‘worst’ outcome for each goal. Please use your own words and be as specific as possible.” Patients were allowed to provide up to 5 individual treatment goals. Between March 2000 and December 2005, follow-up postoperative questionnaires were mailed to each patient on an annual basis, and they were asked to describe each goal’s attainment on a scale of strongly disagree (–)2 to strongly agree (+)2.
Follow-up also included the Incontinence Impact Questionnaire and Urogenital Distress Inventory, a single-item outcome satisfaction assessment (very satisfied to very unsatisfied), and three dichotomous (yes/no) questions asking whether the patient would “recommend the procedure to a friend,” would elect to “have the same procedure again” in similar circumstances, or had “any regrets about undertaking the procedure.” A second mailing was conducted at 3 months, and a follow-up telephone call was made at 6 months for initial nonresponders. These analyses use data from women who completed both the presurgery and postsurgery survey and provided complete data on the surveys. Women were allowed to respond “not applicable” if a given goal category did not apply to them.
To minimize responder bias, pretreatment goals were kept blinded from the operative surgeon, and posttreatment surveys were collected and analyzed by the nonclinical team.
Preprocedure and postprocedure Urogenital Distress Inventory and Incontinence Impact Questionnaire scores were calculated as continuous measures (range 0–100) using published algorithms.6 Frequency distributions were calculated for goal attainment and patient satisfaction. The association of Urogenital Distress Inventory and Incontinence Impact Questionnaire scores to goal attainment, calculated as the mean across all goal categories to which a response was given, was estimated by Pearson correlation coefficients. The association between postprocedure Urogenital Distress Inventory and Incontinence Impact Questionnaire scores and mean goal attainment, adjusting for baseline values, was assessed using partial correlation coefficients. Presurgery to postsurgery changes in Urogenital Distress Inventory and Incontinence Impact Questionnaire scores were assessed using paired t tests.
Of 94 women eligible to participate, 46 (48.9%) completed the surveys, and 40 (42.6%) provided complete data on the study variables. Participants included in the analyses (ie, those with complete data) did not differ significantly from those excluded in age, weight, vaginal parity, family history of pelvic floor dysfunction, personal history of previous pelvic floor dysfunction surgery or hysterectomy, or race (Table 1). Although smoking was uncommon in both groups, those included were more likely (15.0%) than others (3.7%) to use tobacco (χ2=3.8, P=.051).
Among those with complete follow-up data (n=40), mean follow-up was 2.75 years (±1.90). Participants were primarily white (90.0%), older (mean age 75.4±6.8), and vaginally parous (85.0%, 35% with parity 4 or greater). Seventeen (42.5%) had a prior hysterectomy, 12 patients (30%) had at least one prior surgery for pelvic floor dysfunction, and 9 (22.5%) reported a first-degree female family member having had a pelvic floor dysfunction history. Most women had advanced prolapse, with 25 having stage IV, 14 stage III, and, one stage II. Among these women, 20 patients had other pelvic floor symptoms in addition to prolapse. Urodynamic diagnoses for urinary symptoms among women with complete data included voiding dysfunction/incomplete bladder emptying (7), genuine stress incontinence (7), urge incontinence (4), and mixed incontinence (2).
At baseline, two thirds of patients (67.7%, of total n=94) reported that they were not sexually active for lack of a partner (widowed or single status). Twenty-five (26.9%) women listed themselves as married but not currently having sexual intercourse. The remaining five women listed themselves as sexually active but not having vaginal intercourse.
We categorized goals by symptom relief (for example, “get rid of the vaginal pressure”), restoration of normal anatomy (“put the bulge back”), urinary incontinence/frequency (“fewer trips to the bathroom”), bladder emptying, physical activity (“work in my garden again”) and self-image (“look like a normal woman”).
Colpocleisis was either performed or directly supervised by a single surgeon using multiple mattress layered sutures of 0-polyglactin or 2–0 polyglactin. A wide perineorraphy was included to narrow the genital hiatus. Of the women with complete data, additional anti-incontinence procedures were performed in 14 cases: five fascial slings, four minimally invasive midurethral slings (tension-free vaginal tape), and five suburethral plications. These procedures were done when women had a urodynamic diagnosis of stress urinary incontinence.
Table 2 lists postsurgical self-reported goal attainment by goal category. A great majority of patients reported successful attainment of prolapse goals (symptom relief and restoration of normal anatomy).
In a continuous range from –2 to +2, mean goal attainment (1.4±0.6) was not significantly associated with either the presurgery Urogenital Distress Inventory score (r=–0.28, P=.09) or Incontinence Impact Questionnaire score (r=–0.21, P=.19). However, mean goal attainment was associated with postsurgery Urogenital Distress Inventory (r=–0.45, P=.003) but not Incontinence Impact Questionnaire scores (r=–0.19, P=.23). (Spearman correlation coefficients were nearly identical to Pearson [eg, goal attainment and postsurgery Urogenital Distress Inventory r=–0.46, P=.003] and did not change any results qualitatively.) Partial correlations of postsurgery goal attainment with postsurgery quality of life, adjusted for baseline quality of life, were significant for Urogenital Distress Inventory (partial r=–0.38, P=.02) but not Incontinence Impact Questionnaire scores (partial r=–0.03, P=.86). While correlations of presurgery to postsurgery Urogenital Distress Inventory (r=0.73, P<.01) and Incontinence Impact Questionnaire (r=0.86, P<.01) scores were strong, in paired t tests, mean scores improved presurgery to postsurgery for both Urogenital Distress Inventory (39.9±24.9 versus 21.0±20.3, t=7.0, P<.01) and Incontinence Impact Questionnaire (35.4±29.3 versus 17.3±24.6, t=7.5, P<.01). Variation in follow-up duration was considerable but was not significantly associated with either mean goal attainment (r=0.08, P=.64) or any of the Urogenital Distress Inventory or Incontinence Impact Questionnaire measures, and thus unlikely to confound relationships. For instance, the correlation of mean goal attainment to postsurgery Urogenital Distress Inventory was nearly identical with (r=–0.44, P=.005) and without (r=–0.45, P=.003) adjustment for duration of follow-up.
There were 18 perioperative complications and no perioperative deaths. Specific complications obtained from the medical record included urinary tract infection (4), de novo urge urinary incontinence (4), stress incontinence (2), recurrent pelvic organ prolapse (2), lower extremity deep vein thrombosis (1), postoperative bleeding due to over-anticoagulation (1), atypical chest pain (1), atrial fibrillation (1), rectal prolapse (1), and preoperative postmenopausal bleeding with pathology-proven grade I endometrial cancer (1). (This incidence of cancer was found on hysterectomy specimen. This was the only patient who had concomitant total vaginal hysterectomy during closure surgery in this series. She had a preoperative diagnosis of postmenopausal bleeding and a preoperative endometrial biopsy showing a benign polyp.) Eleven complications occurred in the nonresponder group, and seven in the responder group.
One woman expressed dissatisfaction with her outcome, and one woman was not sure, while 14 (35%) were satisfied and 24 (60%) were very satisfied. Women who were very satisfied had higher mean goal attainment compared with women who reported being satisfied (1.76±0.41 versus 0.85±0.31; t=7.2, P<.001). Four respondents answered “yes” to the query “Do you have any regrets about having your surgery?” One of these respondents found a new partner, one reported “feeling less like a woman but OK with surgery,” one described it as “my choice, of the two, is a little regrettable, sometimes,” and one “wanted my bladder to empty better.” One person indicated that she would not recommend the procedure to a friend or would not have the procedure again.
Surgeons and patients have an ever-expanding variety of procedures from which to choose in addressing pelvic organ prolapse. Limited prospective data suggest that abdominal reconstructive approaches have superior durability over vaginal approaches but not without a concomitant increase in operative morbidity.7–9 Nonetheless, increasing numbers of elderly women and their surgeons are electing to proceed with obliterative vaginal surgeries, which in theory, should offer high levels of durability with lower rates of morbidity. In the absence of comparative, prospective data on colpocleisis versus other pelvic organ prolapse surgeries, this investigation shows results similar to other case series. In a recent comprehensive review of the colpocleisis literature, FitzGerald et al10 reported success rates of 91–100%, variable rates of major and/or minor postoperative complications, and regret rates of 3–9%. Similar findings were noted by Wheeler et al11 in their retrospective series. However, outcome variables in case series remain nonstandardized and often difficult to interpret, especially with respect to urinary function.
More than 75% of women in this series described themselves as not sexually active. This observation is similar to cross-sectional, community derived, epidemiological survey data that indicate 78% of married women aged 70–79 years are not sexually active.12 Thus, colpocleisis is a potentially viable option in a large percentage of older women.
In this investigation, goal attainment was uniformly high for patients expressing a preoperative desire to address prolapse symptoms. Perhaps this is an easier goal for patients to measure, as bulging symptoms tend to be present or absent, bothersome or not. Goal achievement of improved voiding function and urinary incontinence was attained more than 75% of the time in this series. Due to our small sample size, we were unable to characterize any potential reasons for the relatively lower level of goal achievement for urinary symptoms. Urinary symptom relief may be achieved from the colpocleisis itself or the anti-incontinence procedures done at the time of closure surgery or both. It may also be explained by the presence of diverse urinary symptoms, compared with prolapse symptoms, which may be inherently more difficult to describe and measure from the patient’s perspective. These are obvious speculations which merit further study.
Our questionnaire response rate of approximately 50% must be recognized as a limitation which brings potential bias to this study. While respondents and nonrespondents did not differ significantly in several key respects, the potential exists for unmeasured differences (eg, in satisfaction) between responders and nonrespondents to bias our findings, and future efforts should focus on more complete response rates. Our project lacked objective/simultaneous parameters of long-term physical examination, voiding diaries, etc, in validating patients’ subjective replies. While future efforts should measure associations with other outcomes, the focus of our work is on patient-centered outcomes that are inherently subjective. The disease-specific quality-of-life instruments we used did not specifically measure colorectal or sexual function. Nonetheless, the association between a change (ie, improvement) in postoperative Urogenital Distress Inventory and Incontinence Impact Questionnaire scores and goal attainment supports the concept of utilizing patient-centered outcomes in assessing successful colpocleisis surgery. Finally, our patient population consists of referrals to a tertiary care, academic center which may not be generalizable to other settings.
In summary, we believe that the results of this study can be utilized by pelvic surgeons during patient selection and preoperative counseling about outcomes after vaginal closure procedures. Future prospective studies in a multi-centered setting will provide more rigorous insights into colpocleisis outcomes and aid in identifying what influences goal selection, along with what potential interventions can be utilized to improve patients’ long-term goal attainment.
1. Shah AD, Rajan SS, Kohli N, Hoyte L. The age distribution of pelvic organ prolapse surgery in the United States, Int Urogynecol J. 2006;17:S409–78.
2. Federal Interagency Forum on Aging-Related Statistics. Older Americans 2000: Key Indicators of Well-Being. Federal Interagency Forum on Aging-Related Statistics. Washington, DC: U.S. Government Printing Office; August 2000.
3. Luber KM, Boero S, Choe JY. The demographics of pelvic floor disorders: current observations and future projections. Am J Obstet Gynecol 2001;184:1496–503.
4. Hullfish KL, Bovbjerg V, Gibson J, Steers WD. Patient-centered goals for pelvic floor dysfunction surgery: what is success and is it achieved? Am J Obstet Gynecol 2002:187:88–92.
5. Hullfish KL, Bovbjerg VE, Steers WD. Patient-centered goals for pelvic floor dysfunction surgery: one- and two-year follow-up, Am J Obstet Gynecol 2004;191:201–5.
6. Uebersax J, Wyuman F, Shumaker S, McClish K, Fantl J. Continence program for women research group. Short forms to assess life quality and symptoms distress for urinary incontinence in women: The Incontinence Impact Questionnaire and the Urogenital Distress Inventory. Neurourol Urodyn 1995;14:131–9.
7. Benson JT, Lucente V, McClellan E. Vaginal versus abdominal reconstructive surgery for the treatment of pelvic support defects: a prospective randomized study with long-term outcome evaluation. Am J Obstet Gynecol 1996;175:1418–21.
8. Maher CF, Qatawneh AM, Dwyer PL, Carey MP, Cornish a, Schluter PJ. Abdominal sacral colpopexy for vaginal vault prolapse: a prospective randomized trial. Am J Obstet Gynecol 2004;190:20–6.
9. Nygaard IE, McCreery R, Brubaker L, Connolly A, Cundiff G, Weber AM. Abdominal sacrocolpopexy: a comprehensive review. Obstet Gynecol 2004;104:805–23.
10. FitzGerald MP, Richter HE, Siddique S, Thompson P, Thompson P, Zyczynski H. Colpocleisis: a review. Int Urogyencol J 2006;17:261–71.
11. Wheeler TL, Richter HE, Burgio KL, Redden DT, Chen CC, Goode PS, et al. Regret, satisfaction, and symptom improvement: analysis of the impact of partial colpocleisis for the management of severe pelvic organ prolapse. Am J Obstet Gynecol 2005;193:2067–70.
12. Patel D, Gillespie B, Foxman B. Sexual behavior of older women: results of a random-digit-dialing survey of 2,000 women in the United States. Sex Trans Ds 2003;30:216–20.
Figure. No caption available.
© 2007 The American College of Obstetricians and Gynecologists