It is estimated that one American woman in nine will undergo surgical repair for pelvic organ prolapse (POP) or urinary incontinence (UI) or both in her lifetime.1 Some degree of minor POP is found in nearly all parous women of all ages,2,3 but we know very little about the natural history of prolapse. Thus, we are unable to answer important clinical questions, such as how often women develop clinically significant prolapse, what their risk is of minor prolapse worsening, or whether prolapse may even regress.
The pelvic organ prolapse quantification (POP-Q) system, which has demonstrated good intraexaminer and interexaminer reliability, is a tool that quantitatively describes findings on vaginal examination.4,5 The POP-Q examination has substantially improved our ability to measure vaginal descent, but a clinically relevant definition for POP within the POP-Q staging system has not yet been identified. Because many women have some vaginal descent, there is debate about the clinical importance of such mild prolapse. An improved understanding about what happens to vaginal support over time will help us understand the relevance of these anatomic findings.
The broad objective of this longitudinal 4-year study was to describe the natural history of POP in a group of postmenopausal women not seeking care for pelvic floor disorders. Our primary aim was to estimate annual and 3-year prolapse incidence and resolution risks, and our secondary aim was to describe factors associated with the progression and regression of vaginal descent.
PARTICIPANTS AND METHODS
We approached women enrolled in the Women’s Health Initiative Hormone Therapy clinical trial at one midwestern site about enrolling in this ancillary study. Every woman was postmenopausal and had a uterus. Enrolled subjects attended annually for a research visit, and subjects that had POP-Q measurements taken on at least two occasions were included in this longitudinal study. Year 4 data could not be collected in most women because the overall Women’s Health Initiative trial was halted prematurely.6 This study was approved by the University of Iowa Institutional Review Board, and each subject completed an informed consent document.
At each visit, a POP-Q examination was performed by one of two experienced urogynecology research nurses, in addition to the pelvic examination required by the Women’s Health Initiative trial protocol. Examination times were set at the participant’s convenience. Each participant was asked to empty her bladder before the examination, but bladder volumes were not otherwise standardized. The examiners were blinded to previous results at subsequent examinations. Before beginning the study, we tested the interrater reliability for the POP-Q staging system between one investigator (I.E.N.) and the primary research nurse. The weighted kappa for stage was 0.63 (95% confidence interval [CI] 0.31–0.94), suggesting substantial agreement.7
The POP-Q examinations were performed with women in the dorsal lithotomy position. Negative POP-Q values denote support above, and positive values denote support below the hymeneal ring. A “0” value indicates that the area of vagina being measured rests at the level of the hymen. All measurements, with the exception of the total vaginal length, were taken with a marking stick marked at 1-cm intervals when the subject was performing a maximal Valsalva maneuver. If the women had difficulty performing a Valsalva maneuver, they were asked to cough forcefully, and measurements were taken at maximal cough.
Demographic and background data were abstracted from the Women’s Health Initiative database, and subjects provided additional information about obstetric, medical, and surgical history in an annual written questionnaire. We studied change in vaginal descent in two ways: 1) as the incidence or resolution of prolapse which we defined, based on previous work, as present if the point of maximal vaginal descent (the most distal POP-Q point) was at or beyond the hymen (that is, 0 cm or greater)2,8 and 2) as the progression and regression of the point of maximal vaginal descent. We examined progression and regression of the point of maximal vaginal descent by intervals of both 1 cm or greater and 2 cm or greater.
Continuous variables were summarized by mean and standard deviation or by median and range for variables with nonnormal distributions (as determined by the Shapiro-Wilk test). Differences between groups were tested with the Kruskal-Wallis test. Annual prolapse incidence and resolution risks (with 95% CIs) were calculated; 3-year prolapse incidence and resolution risks (with 95% CIs) were calculated for those women who had examination data at both years 1 and 4. Because the dependent variables were categorical and repeatedly measured over time, longitudinal data analysis based on generalized linear (logistic) models were used to estimate rates of vaginal descent progression and regression over time, controlling for the baseline maximal vaginal descent. The parameters in the model were estimated via the Generalized Estimation Equation method.9
Time-to-event analyses were not used because we found that women might meet criteria for prolapse at one examination and then not at follow-up exams. Because this condition of prolapse was not final, survival analysis (for instance) would not be appropriate or useful to best describe our observed results.
Similar models were used to study the relationship between potential risk factors for the progression and regression of vaginal descent. Any (1 cm or greater) increase or decrease in the maximal vaginal descent was included as the dependent variable in each model. The independent variables included baseline maximal vaginal descent, time (1, 2, or 3 years), and risk factors. Risk factors studied included age, body mass index (BMI), waist circumference, education level, vaginal parity, current smoking, and the randomization assignment to estrogen-progesterone treatment compared with placebo. Risk factors associated with the dependent variable (P<.2) were included in the multivariable models. In addition, an interaction between time and the baseline maximal vaginal descent was studied and considered for inclusion in the models. A backwards model selection process was performed. Categorization of risk factors was examined during univariable analysis, and the best function forms were applied to the fitted multivariable model. For continuous variables, categorization methods based on quartiles were first examined. Adjacent levels having similar effects on the dependent variable were then combined into one level. Odds ratios (ORs) with 95% CIs for progression and regression of vaginal descent for each of the significant risk factors were computed from the fitted model.
Statistical analyses were performed with SAS 9.1 (SAS Institute Inc, Cary, NC). Associations were considered significant at a P<.05 level. Our sample size was predetermined by the number of women in this ancillary study. Our primary aims were descriptive, and we conducted post hoc power analyses for our secondary aim.
Of the 270 women who participated in the baseline examination, 259 (95.9%) had at least one additional examination, and these women constitute the study population. These 259 women were followed over a period of 1–3 years, for a total of 559 woman-years. The mean age of this group was 68.1 years (standard deviation [SD] 5.5 years), mean BMI 30.1 kg/m2 (SD 6.2 kg/m2), mean waist circumference 89.9 cm (SD 13.0 cm), and the median vaginal parity was 4 (range 0–12). At enrollment, 21 (8.1%) women were current smokers. One hundred eighteen (45.6%) had been randomized to the placebo group in the larger Women’s Health Initiative trial and 141 (54.4%) to the estrogen-progesterone group. Other characteristics of the study sample at baseline have been previously described.2 Five women (1.9%) reported prior surgery for POP on enrollment (occurring from 3 to 31 years earlier). One woman underwent surgery for prolapse (hysterectomy and rectocele repair) during the course of the study. Four women reported pessary use for POP or UI at one or more visits during the study.
Seventy three women (28%) had four, 128 (49%) had three, and 58 (22%) had two examinations. Women with different numbers of examinations were similar in age, BMI, parity, waist circumference, and smoking (P values all >.05). While the median baseline maximal vaginal descent in women with different numbers of examinations was the same (–1 cm), the range was greater in those with three examinations (maximal vaginal descent range –3 to +3 cm compared with –3 to +2 cm for those with four exams and –3 to +1 for those with two exams; P=.04).
In every year, the median maximal vaginal descent was –1 cm (range –3 to +3 cm). In each of years 1 through 4, 23.6% (61 of 259), 39.0% (94 of 241), 47.1% (98 of 208), and 49.4% (42 of 85) of women had prolapse (as defined dichotomously above), but this condition of prolapse waxed and waned from year to year in individual women (Fig. 1). Overall, 1-year and 3-year prolapse incidence risks were 26% (95% CI 20–33%) and 40% (95% CI 26–56%); and 1-year and 3-year prolapse resolution risks were 21% (95% CI 11–33%) and 19% (95% CI 7–39%). Of 199 women with no prolapse at baseline, 115 (58.1%) had no prolapse at any subsequent evaluation, whereas 32 (16.2%) had prolapse at all subsequent evaluations.
The probabilities of any increase and decrease in maximal vaginal descent over time varied depending on the baseline maximal vaginal descent. The estimated probabilities of vaginal descent progression and regression of at least 1 cm for 1- and 3-year time intervals are presented in Table 1 and are presented graphically for 1-year intervals in Figure 2. One-year progression (increase in maximal vaginal descent by 1 cm or greater) rates exceeded regression (decrease in maximal vaginal descent by 1 cm or greater) rates when the baseline maximal vaginal descent was –1 cm or less, and 1-year regression rates exceeded progression rates when the baseline maximal vaginal descent was 0 cm or greater. A similar pattern was seen for 3-year rates, except that 3-year progression rates were equal to regression rates when the baseline maximal vaginal descent was 0 cm.
The likelihood of vaginal descent progression and regression of at least 2 cm similarly depended on the baseline level of vaginal descent (data not shown). Overall, a 1-year progression (increase) in maximal vaginal descent by 2 cm or greater occurred in 5.8% (95% CI 3.2–9.6%) of the women and a 1-year regression (decrease) of 2 cm or greater occurred in 1.2% (95% CI 0.26–3.6%). Over 3 years, the maximal vaginal descent increased by 2 cm or greater in 11.0% (95% CI 4.9–20.5%) of the women and decreased by 2 cm or greater in 2.7% (95% CI 0.3–9.5%).
Risk factors associated with any increase and any decrease in the maximal vaginal descent are presented in Table 2. Increasing BMI and grand multiparity (five or more vaginal deliveries) increased the risk for vaginal descent progression. Current smokers had an independent decreased risk for vaginal descent progression in the model, but smokers had a lower mean BMI than nonsmokers (26.2 versus 30.5 kg/m2; P=.002).
In a separate model, increasing BMI and parity were inversely associated with vaginal descent regression. Randomization assignment to estrogen-progesterone treatment compared with placebo was not associated with changes in vaginal descent. A post hoc power calculation using our actual sample size and the observed 1-year vaginal descent progression rate in the women assigned to the placebo group (34.5%) suggests that our sample size provided 80% power at a 5% significance level to identify an increase or decrease of at least 19 percentage points in this rate in the estrogen-progesterone group compared with the placebo group.
Waist circumference correlated strongly with BMI (r=0.83, P<.001), and thus could not be included in the same multivariable models. When models were created that included waist circumference instead of BMI, waist circumference was also found to be independently associated with progression and regression of the maximal vaginal descent. A waist circumference of more than 80 cm, compared with 80 cm or less, increased the risk of vaginal descent progression (OR 2.77, 95% CI 1.58–4.86) and decreased the risk of vaginal descent regression (OR 0.39, 95% CI 0.21–0.72).
Based on this study’s findings, mild POP is a fluid state. On balance, older parous women are more likely to develop new or progressive prolapse than to regress, but the “net” annual change in vaginal descent (progression rate – regression rate), is much lower than the overall progression rate. Risk factors for maximal vaginal descent progression include the baseline level of vaginal descent, as well as BMI, waist circumference, and vaginal parity. In this group of women, smoking was protective against increasing vaginal descent. However, given the association between smoking and BMI, smoking may be a proxy for lower BMI as opposed to having a true protective effect.
When initially designing this study, we did not consider the possibility that prolapse could resolve. During the course of the study, Handa et al10 reported intriguing results from women enrolled at a different Women’s Health Initiative site: annual regression rates for prolapse exceeded progression rates. However, prolapse was not assessed using a validated or quantitative tool, but rather using the larger Women’s Health Initiative trial examination technique in which the vagina was visualized during straining without a speculum in place. Given this less rigorous method of assessing prolapse, some considered these findings a result of measurement error. Our results, using a standardized, quantitative, and reproducible method of assessing prolapse confirm that early POP does indeed wax and wane. However, contrary to Handa’s findings, progression of prolapse was more likely in our study than regression.
On the other hand, alternative conclusions from our results must also be considered. The vaginal descent “progression” and “regression” observed in these women could represent variability in vaginal descent. The time of examination, exact bladder volume, and Valsalva efforts during examination were not standardized and might impact observable levels of vaginal descent or prolapse. In addition, although the POP-Q examination tool has proven reproducibility, no tool is perfect, and our results might also result in part from variability inherent in POP-Q measurements. To partially offset such limitations, we studied progression and regression of descent in terms of 2-cm or greater changes in addition to 1-cm changes.
We evaluated changes in pelvic support by describing both how vaginal descent changed from year to year using the continuous variable of maximal vaginal descent and how women changed back and forth from having prolapse (defined dichotomously as present when the maximal vaginal descent was at or beyond the hymen). Although the initial POP-Q report defined prolapse as present when the POP-Q stage was I or greater, this definition would classify 98% of our subjects as having prolapse at baseline, precluding a meaningful comparison of women with and without prolapse.4 We chose to use the hymeneal ring as the cut point for a dichotomous definition of prolapse because women with descent at or below this level are more likely to report seeing or feeling a vaginal bulge than are women with less vaginal descent, and researchers increasingly suggest this demarcation to define prolapse.8,11,12
Women in this study were of a similar age to women most likely to undergo surgery for prolapse, and therefore we believed this to be an ideal age at which to study the natural history of POP.13 Each of our participants had a uterus. We chose to study this group to understand the natural history of pelvic support in women without detachment of the apical vaginal support structures, and women who have had hysterectomy are more likely to have had prior prolapse surgery. Our findings cannot be generalized to younger women or to those without a uterus.
In conclusion, we suggest that vaginal descent occurring within the vagina is common in older, parous women, and that the point of maximal vaginal descent (as measured by the POP-Q system) may decrease, as well as increase, over time. In fact, we found that women with minimal vaginal descent at baseline are more likely to progress in future examinations and women with more descent, to regress, over time. This finding is perhaps intuitive because each has more potential movement in the opposite direction, but this has not been previously reported. This finding is also suggestive of the well-known phenomenon of regression to the mean. This principle states that, of related measurements, where first measurement is higher or lower relative to the mean, the expected value of the second measurement will be closer to the mean than to the observed value of the first measurement.14
Overall, clinically significant progression of vaginal descent was uncommon in this group over a 3-year period in that only one woman underwent surgery for this reason and four reported pessary use. One in 10 women demonstrated 2-cm or greater progression in maximal vaginal descent, a potentially significant anatomic change. The number of women who regressed by this amount was much smaller (2.7%). Even longer follow-up and larger numbers of women will be needed to identify women as they transition from having early, largely asymptomatic prolapse to having bothersome bulges. Our data suggest that clinicians can reassure women with vaginal descent within the vagina that their chance of significant progression over a 1- to 3-year period is low.
1. Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 1997;89:501–6.
2. Nygaard I, Bradley C, Brandt D. Pelvic organ prolapse in older women: prevalence and risk factors. Obstet Gynecol 2004;104:489–97.
3. Swift SE. The distribution of pelvic organ support in a population of female subjects seen for routine gynecologic health care. Am J Obstet Gynecol 2000;183:277–85.
4. Bump RC, Mattiasson A, Bo K, Brubaker LP, DeLancey JO, Klarskov P, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7.
5. Hall AF, Theofrastous JP, Cundiff GW, Harris RL, Hamilton LF, Swift SE, et al. Interobserver and intraobserver reliability of the proposed International Continence Society, Society of Gynecologic Surgeons, and American Urogynecologic Society pelvic organ prolapse classification system. Am J Obstet Gynecol 1996;175:1467–71.
6. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–33.
7. Landis J, Koch G. The measurement of observer agreement of categorical data. Biometrics 1977;33:159–74.
8. Swift S, Woodman P, O’Boyle A, Kahn M, Valley M, Bland D, et al. Pelvic Organ Support Study (POSST): The distribution, clinical definition, and epidemiologic condition of pelvic organ support defects. Am J Obstet Gynecol 2005;192:795–806.
9. Diggle P, Liang K, Zeger S. Analysis of longitudinal data. New York (NY): Oxford University Press; 1994. p. 143–5.
10. Handa VL, Garrett E, Hendrix S, Gold E, Robbins J. Progression and remission of pelvic organ prolapse: a longitudinal study of menopausal women. Am J Obstet Gynecol 2004;190:27–32.
11. Bradley CS, Nygaard IE. Vaginal wall descensus and pelvic floor symptoms in older women. Obstet Gynecol 2005;106:759–66.
12. Swift S. Pelvic organ prolapse: is it time to define it? Int Urogynecol J Pelvic Floor Dysfunct 2005;16:425–7.
13. Boyles SH, Weber AM, Meyn L. Procedures for pelvic organ prolapse in the United States, 1979–1997. Am J Obstet Gynecol 2003;188:108–15.
14. Freedman D, Pisani R, Purves R, Adhikari A. Statistics. 2nd ed. New York (NY): W. W. Norton & Company; 1991.
APPENDIX: SHORT LIST OF WOMEN’S HEALTH INITIATIVE INVESTIGATORS
Program Office: (National Heart, Lung, and Blood Institute, Bethesda, MD) Barbara Alving, Jacques Rossouw, Shari Ludlam, Linda Pottern, Joan McGowan, Leslie Ford, and Nancy Geller.
Clinical Coordinating Center: (Fred Hutchinson Cancer Research Center, Seattle, WA) Ross Prentice, Garnet Anderson, Andrea LaCroix, Charles L. Kooperberg, Ruth E. Patterson, Anne McTiernan; (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker; (Medical Research Labs, Highland Heights, KY) Evan Stein; (University of California at San Francisco, San Francisco, CA) Steven Cummings.
Clinical Centers: (Albert Einstein College of Medicine, Bronx, NY) Sylvia Wassertheil-Smoller; (Baylor College of Medicine, Houston, TX) Jennifer Hays; (Brigham and Women’s Hospital, Harvard Medical School, Boston, MA) JoAnn Manson; (Brown University, Providence, RI) Annlouise R. Assaf; (Emory University, Atlanta, GA) Lawrence Phillips; (Fred Hutchinson Cancer Research Center, Seattle, WA) Shirley Beresford; (George Washington University Medical Center, WA, DC) Judith Hsia; (Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA) Rowan Chlebowski; (Kaiser Permanente Center for Health Research, Portland, OR) Evelyn Whitlock; (Kaiser Permanente Division of Research, Oakland, CA) Bette Caan; (Medical College of Wisconsin, Milwaukee, WI) Jane Morley Kotchen; (MedStar Research Institute/Howard University, WA, DC) Barbara V. Howard; (Northwestern University, Chicago/Evanston, IL) Linda Van Horn; (Rush Medical Center, Chicago, IL) Henry Black; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (State University of New York at Stony Brook, Stony Brook, NY) Dorothy Lane; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Alabama at Birmingham, Birmingham, AL) Cora E. Lewis; (University of Arizona, Tucson/Phoenix, AZ) Tamsen Bassford; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of California at Davis, Sacramento, CA) John Robbins; (University of California at Irvine, CA) F. Allan Hubbell; (University of California at Los Angeles, Los Angeles, CA) Howard Judd; (University of California at San Diego, LaJolla/Chula Vista, CA) Robert D. Langer; (University of Cincinnati, Cincinnati, OH) Margery Gass; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Hawaii, Honolulu, HI) David Curb; (University of Iowa, IA/Davenport, IA) Robert Wallace; (University of Massachusetts/Fallon Clinic, Worcester, MA) Judith Ockene; (University of Medicine and Dentistry of New Jersey, Newark, NJ) Norman Lasser; (University of Miami, Miami, FL) Mary Jo O’Sullivan; (University of Minnesota, Minneapolis, MN) Karen Margolis; (University of Nevada, Reno, NV) Robert Brunner; (University of North Carolina, Chapel Hill, NC) Gerardo Heiss; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; (University of Tennessee, Memphis, TN) Karen C. Johnson; (University of Texas Health Science Center, San Antonio, TX) Robert Brzyski; (University of Wisconsin, Madison, WI) Gloria E. Sarto; (Wake Forest University School of Medicine, Winston-Salem, NC) Denise Bonds; (Wayne State University School of Medicine/Hutzel Hospital, Detroit, MI) Susan Hendrix.