| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Section of Urogynecology and Reconstructive Pelvic Surgery, Division of Gynecologic Specialties, Department of Obstetrics, Gynecology, and Research, Magee Womens Hospital, Magee Womens Research Institute, University of Pittsburgh, Pittsburgh, Pennsylvania.
Address reprint requests to: Pamela A. Moalli, MD, PhD, Magee Womens Hospital, Department of Obstetrics, Gynecology, and Research, 300 Halket Street, Pittsburgh, PA 15213; E-mail: rsipam{at}mail.magee.edu.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: We conducted a case–control study, with cases selected from all women who had surgery by our urogynecology service from July 1, 1999 to July 1, 2000 and who had a first obstetric delivery at Magee Womens Hospital (n = 80). Controls were patients seen in the general gynecology office over the same time period who had no complaints associated with pelvic floor disorders in the previous 3 years, less than stage I prolapse on pelvic examination, and first obstetric delivery at Magee Womens Hospital (n = 176). Demographic, obstetric, and gynecologic variables were compared between cases and controls.
RESULTS: There were no significant differences in race, current age, gravidity, or parity. Cases were more likely than controls to have a higher body mass index (BMI) (28.6 ± 6.3 versus 26.4 ± 6.1 kg/m2, P = .01), to be younger at first delivery (25.8 ± 5.3 versus 28.4 ± 4.9 years, P < .001), to have undergone a forceps delivery (64% versus 44%, P 
.001), and to have had previous gynecologic surgery (34% versus 16%, P = .003). Using logistic regression modeling, all of these factors were found to be independently associated with pelvic floor disorders. After menopause, use of hormone replacement therapy 5 or more years was protective (P = .001).
CONCLUSION: In our surgical patients, younger age at first delivery, higher BMI, forceps delivery, and history of gynecologic surgery were significantly associated with subsequent development of pelvic floor disorders.
In the United States, disorders of pelvic support, including urinary incontinence and pelvic organ prolapse, are common. Up to 50% of all parous women have some degree of vaginal wall laxity, with 10–20% being symptomatic.1–3 The lifetime risk of undergoing a surgery to correct a dysfunction of the pelvic floor is estimated at 11%.4 The precise etiology of pelvic floor disorders is unknown but likely multifactorial. Previous epidemiologic analyses suggest that obstetric risk factors include operative vaginal delivery, episiotomy, macrosomic infant, epidural analgesia, and increased maternal age at the time of delivery.5,6 Nonobstetric risk factors include prior hysterectomy, aging, and menopause.4,7–10
In this study, we sought to determine the risk factors in our surgical patient population associated with the development of pelvic floor disorders relative to women without pelvic floor dysfunction. To date, most studies have relied on patient recall through questionnaires to determine outcome variables.5,6 We took advantage of the tendency of the women in our local community to stay in the area for decades after childbirth and to continue medical treatment at our institution. In this way, we had the unique opportunity to obtain longitudinal data directly from the medical record on women who underwent surgery for the treatment of pelvic floor disorders. Based on previous studies, we hypothesized that the first obstetric delivery would incur the greatest damage to the pelvic floor.10,11 Our goal was to identify factors that may contribute to the development of pelvic floor disorders, so that ultimately we may formulate preventative strategies designed to protect the pelvic floor.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Controls consisted of patients seen for general gynecology health maintenance consecutively between August 1999 and August 2000 and who had a first obstetric delivery at Magee Womens Hospital. Current gynecology charts of control patients were screened for exclusion criteria, including a complaint associated with pelvic floor dysfunction over the previous 3 years and/or the presence of greater than stage I prolapse or its equivalent on pelvic examination.
All of the outcome variables were selected before chart review and incorporated into a data abstraction form designed by Drs. Moalli and Zyczynski. Medical records were reviewed and data abstracted by trained research nurses. Demographic variables included current age, race, gravidity, parity, and current body mass index (BMI). Variables obtained from first obstetric delivery included age at delivery, mode of delivery, infant birth weight, episiotomy, and vaginal laceration. Variables obtained from gynecologic history included previous gynecologic surgery, menopausal status, and use of hormone replacement therapy (HRT). Previous gynecologic surgery was defined as abdominal or vaginal hysterectomy, salpingo-oophorectomy, and procedures to correct uterine or vaginal prolapse or incontinence (urinary or fecal). Tubal ligation or salpingectomies for tubal pathology were not counted in the category of previous gynecologic surgery. Menopause, whether occurring as a result of surgical removal of both ovaries or through ovarian senescence, was defined as amenorrhea for at least 1 year. Categorization as postmenopausal on HRT required the use of standard doses of estrogen supplementation or estrogen and progesterone for at least 1 year. Among postmenopausal women on HRT, 5 years was the median time of HRT. Therefore, postmenopausal women on HRT were divided into two categories: on HRT for less than 5 years, and on HRT for 5 years or more. Age at first delivery was initially categorized by 5-year increments (younger than 20 years, 20–24 years, 25–29 years, 30–34 years, and 35 years or older). The rates of pelvic floor dysfunction requiring surgical correction were similar for the < 20 years and 20–24 years age groups (60% and 51%, respectively) and for the 25–29 years, 30–34 years, and 35 years or older age groups (21%, 22%, and 25%, respectively). Based on these findings, we then dichotomized age to less than 25 years and 25 years or older for use in logistic regression models. Body mass index was divided into two groups: greater than 26 kg/m2 (overweight) and 26 kg/m2or less (normal and underweight).12
All statistical tests were performed using SPSS statistical software 10.1.4 (SPSS Inc., Chicago, IL). The association of mode of delivery with surgery for pelvic floor disorder(s) was evaluated using the 
2 test for linear trend. The associations of all other categoric variables were evaluated using Fisher exact or 2 tests, where appropriate. Normality was defined by visually assessing the shape of the frequency distribution. The associations of continuous, normally distributed variables were compared using Student t test, whereas the associations of non-normally distributed, ordinal variables were evaluated using the Mann–Whitney U test.13 All statistical tests were evaluated at a significance level of .05. A multivariable logistic regression model was developed to determine which variables were independently associated with the development of pelvic floor dysfunction requiring surgical correction. Variables with a P value of less than 0.25 were considered for inclusion in the regression model. Models were developed using forward stepwise regression based on the likelihood ratio test statistic. Variables were retained in the model if the Wald 2 test statistic had a P value of .05 or less.14
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Cases had a higher BMI than controls (28.6 ± 6.3 versus 26.4 ± 6.1, P = .01) and were of younger age at first delivery (25.8 ± 5.3 vs 28.4 ± 4.9 years, P < .001). Analysis of obstetric and gynecologic risk factors is shown in Table 2. Results of univariate analysis demonstrate that patients were more likely than controls to be younger than 25 years at first delivery and to have a BMI greater than 26 kg/m2 (P < .001). Within both case and control groups, forceps delivery was more common than spontaneous vaginal delivery (cases: 64% versus 31%; controls: 44% versus 39%), and the vast majority of patients had an episiotomy (89%, compared with 78% of controls). There were no differences between cases and controls in having an infant with a birth weight greater than 4000 g or the presence of a third- or fourth-degree obstetric laceration. A linear trend toward an increased likelihood of developing pelvic floor disorder(s) requiring surgical correction was established with cesarean delivery as less of a risk than spontaneous vaginal delivery and forceps delivery associated with the greatest risk (P < .001). As shown in the univariate analysis, forceps delivery carried a 4.7-fold increased risk of pelvic floor surgery.
|
|
, model I), age less than 25 years at first delivery (odds ratio [OR] = 3.6 and 95% confidence interval [CI] 2.0, 6.6) and forceps delivery (OR = 3.2, 95% CI 1.0, 9.9; P = .046) were independently associated with the development of pelvic floor dysfunction requiring surgical correction (Table 3). A second model (model II) was developed to control for both obstetric and gynecologic variables. In this model, forceps delivery (OR = 5.4, 95% CI 1.6, 18.4) and age less than 25 years at first delivery (OR = 4.0, 95% CI 1.9, 8.2) remained significant risk factors. Patients were also more likely than controls to have a BMI greater than 26 kg/m2 (OR = 3.0, 95% CI 1.6, 5.7) and to have had previous gynecologic surgery (OR = 3.9, 95% CI 1.8, 8.8) according to model II. Although hormonal status was not univariately associated with surgery for pelvic floor disorder(s) (P = .20), a negative association was established once the potentially confounding obstetric and gynecologic risk factors were taken into account. As shown in model II, menopausal women currently on HRT for 5 or more years had a 90% decreased risk (OR = 0.1, 95% CI 0.03, 0.3, P < .001) in surgery for pelvic floor disorders.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our data support an association between younger age at first delivery and surgery for pelvic floor disorders. Indeed, a subanalysis of different age groups showed that age less than 25 years at first delivery was associated with the greatest risk. In contrast, a case–control study performed in Canada found that patients with prolapse and incontinence were an average of 4 years older than controls at the time of obstetric delivery.6 Others have also found an association with older age at delivery.15 Because the current age of patients and controls is similar in this study, the women who were younger at first delivery were delivered in the more distant past. Thus, the increased risk in the group of women who delivered at age less than 25 years may reflect a difference in obstetric practices at that time. On the other hand, it is well known that most cases of pelvic floor disorder(s) do not become apparent until decades after the original trauma. Thus, the results may suggest that women who delivered at a younger age had simply passed through a sufficiently long latency period from the time of delivery that allowed an evolving pelvic floor disorder to become manifest.
Our study demonstrated that operative vaginal delivery by the use of forceps increased the risk of developing pelvic floor dysfunction and was common practice at our institution 20–30 years ago. Over the past decade, a relationship between forceps delivery, immediate perineal trauma, and subsequent development of prolapse and incontinence has been developed.10,11,16,17 A recent study by Arya et al18 compared forceps- and vacuum-assisted vaginal deliveries with normal spontaneous vaginal delivery and found that the incidence of persistent urinary incontinence at 1 year following delivery was significantly higher in the forceps group. Forceps delivery has previously been determined to be a risk factor for the development of fecal incontinence.10,11,16,17 Future studies will be needed to determine whether forceps are used as often in current obstetric practices to assist in vaginal delivery as they were 2–4 decades ago. Similarly, as obstetric practices evolve and fewer elective forceps are performed, studies will be needed to determine whether elective forceps pose a different risk than forceps for an indication.
Similar to the findings in this study, previous studies have found that gynecologic surgery, especially hysterectomy and previous prolapse surgery, predispose a woman to develop incontinence and/or prolapse. Karasick and Spettell found that previous hysterectomy was associated with increased likelihood of having fecal incontinence and excessive pelvic floor descent.7 Data obtained from a questionnaire in the Women’s Health Australia project showed an association between the development of urinary incontinence and a history of hysterectomy and/or surgery for prolapse in older women.19 A history of hysterectomy and a history of previous prolapse surgery were found to be among the strongest etiologic predictors of severe pelvic organ prolapse in a retrospective case–control study.20 The data from this analysis do not provide information on the indication for hysterectomy. If performed for prolapse without a concomitant vaginal vault suspension, a hysterectomy would be predictably followed by a recurrence of prolapse and would not be a risk factor for prolapse per se. It is important to note that, to date, no data support that previous gynecologic surgery is a cause for prolapse but rather show that an association exists between surgery and subsequent prolapse. Further prospective analyses of patients undergoing gynecologic surgery that include the indication for surgery will be needed to determine the true incidence of postsurgical pelvic floor disorders.
Multivariable logistic regression modeling showed that, in this study, menopausal women who took standard doses of hormone or estrogen replacement 5 or more years were less likely to develop pelvic floor disorders. Previous studies investigating the use of estrogen as a therapy for prolapse have shown no effect on the stage of prolapse.21 In contrast, estrogen has been found to be therapeutic in the treatment of mild stress urinary incontinence and symptoms associated with urogenital atrophy. Estrogen receptors have been identified throughout the nuclei of the connective tissue and smooth muscle cells of the bladder trigone, urethra, vaginal mucosa, levator ani muscle stroma, the arcus tendineus, and the uterosacral ligament.22–27 If estrogen does positively impact the pelvic floor, it most likely acts in a preventative fashion to optimize the function of the pelvic floor muscle and connective tissue. It is unlikely that once a pelvic floor disorder becomes advanced that estrogen would have any therapeutic effect.
The major limitation of this retrospective study is that it may be affected by selection bias, which occurs commonly in case–control studies. Our cases included only women who had undergone a surgical repair of a pelvic floor disorder and therefore had more advanced disease. By eliminating women with mild disease, the difference between cases and controls may be inflated. For this reason, the results of this study cannot be generalized to all women with pelvic floor disorders but should be limited to those who undergo surgical repair. A second limitation of the study in that it does not provide insight into what types of pelvic floor dysfunction different obstetric variables may predispose to. Finally, it is not clear from the results whether intrinsic factors in women with pelvic floor disorders, such as anatomic or genetic defects, make them more likely to have the risk factors identified in this study to be associated with the development of pelvic floor disorders.
The advantage of performing a case–control study is that it enables us to collect obstetric and gynecologic variables that would be expensive and time consuming to collect prospectively. Indeed, the majority of women who develop incontinence and prolapse do so decades after an obstetric delivery or a gynecologic surgery. The distinct advantage of this particular study over previous retrospective analyses is that the obstetric data from cases and controls were abstracted directly from the labor and delivery record. Thus, we were able to collect longitudinal data from a population of women treated over a period of decades at the same institution. In contrast, previous studies have relied on patient histories to obtain data and consequently are more subject to recall bias.5,6
|
Footnotes |
|---|
Received May 7, 2002. Received in revised form July 23, 2002. Accepted August 22, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Bidmead J, Cardozo LD. Pelvic floor changes in the older woman. Br J Urol 1998;82 Suppl:18–25.
3. Iosif S, Henriksson L, Ulmsten U. The frequency of disorders of the lower urinary tract, urinary incontinence in particular, as evaluated by a questionnaire survey in a gynecological health control population. Acta Obstet Gynecol Scand 1981;60:71–6.[Medline]
4. Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiology of surgically managed pelvic organ propose and urinary incontinence. Obstet Gynecol 1997;89:501–6.[Abstract]
5. Gurel H, Gurel SA. Pelvic relaxation and associated risk factors: The results of logistic regression analysis. Acta Obstet Gynecol Scand 1999;78:290–3.[Medline]
6. Carley ME, Turner RJ, Scott DE, Alexander JM. Obstetric history in women with surgically corrected adult urinary incontinence or pelvic organ prolapse. J Am Assoc Gynecol Laparosc 1999;6:85–9.[Medline]
7. Karasick S, Spettell CM. The role of parity and hysterectomy on the development of pelvic floor abnormalities revealed by defecography. AJR Am J Roentgenol 1997; 169:1555–8.
8. Symmonds R, Williams TJ, Lee RA, Webb MJ. Posthysterectomy enterocele and vaginal vault prolapse. Am J Obstet Gynecol 1981;140:852–9.[Medline]
9. Prior A, Stanley KM, Smith ARB, Read NW. Effect of hysterectomy on anorectal and urethrovesical physiology. Gut 1992;33:264–7.
10. Sultan AH, Kamm MA, Hudson CN, Thomas JM, Bartram CI. Anal sphincter disruption during vaginal delivery [see comments]. N Engl J Med 1993;329:1905–11.
11. Eason E, Labrecque M, Marcoux S, Mondor M. Anal incontinence after childbirth. CMAJ 2002;166:326–30.
12. Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: Prevalence and trends, 1960–1994. Int J Obes Relat Metab Disord 1998;22:39–47.[Medline]
13. Hosmer DW, Lemeshow S. Applied logistic regression. New York: John Wiley & Sons, 1989.
14. Altman DG. Practical statistics for medical research. London: Chapman & Hall, 1991.
15. Zetterström JP, Lopez A, Anzen B, Dolk A, Norman M, Mellgren A. Anal incontinence after vaginal delivery: A prospective study in primiparous women. Br J Obstet Gynaecol 1999;106:324–30.[Medline]
16. Sultan AH, Johanson RB, Carter JE. Occult anal sphincter trauma following randomized forceps and vacuum delivery. Int J Gynaecol Obstet 1998;61:113–9.[Medline]
17. Groutz A, Fait G, Lessing JB, David MP, Wolman I, Jaffa A, et al. Incidence and obstetric risk factors of postpartum and anal incontinence. Scand J Gastroenterol 1999;34: 315–8.[Medline]
18. Arya LA, Jackson ND, Myers DL, Verma A. Risk of new-onset urinary incontinence after forceps and vacuum delivery in primiparous women. Am J Obstet Gynecol 2001;185:1318–23.[Medline]
19. Chiarelli P, Brown W, McElduff P. Leaking urine: Prevalence and associated factors in Australian women. Neurourol Urodyn 1999;18:567–77.[Medline]
20. Swift SE, Pound T, Dias JK. Case-control study of etiologic factors in the development of severe pelvic organ prolapse. Int Urogynecol J Pelvic Floor Dysfunct 2001;12:187–92.[Medline]
21. Lovatsis D, Drutz HP. The role of estrogen in female urinary incontinence and urogenital aging: A review. Ostomy Wound Manage 1998;44:48–53.
22. Mokrzycki ML, Mittal K, Smilen SW, Blechman AN, Porges RF, Demopolous RI. Estrogen and progesterone receptors in the uterosacral ligament. Obstet Gynecol 1997;90:402–4.[Abstract]
23. Hodgins MB, Spike RC, Mackie RM, MacLean AB. An immunohistochemical study of androgen, oestrogen and progesterone receptors in the vulva and vagina. Br J Obstet Gynaecol 1998;105:216–22.[Medline]
24. Smith P, Heimer G, Norgren A, Ulmsten U. Localization of steroid hormone receptors in the pelvic muscles. Eur J Obstet Gynecol Reprod Biol 1993;50:83–5.[Medline]
25. Chen GD, Oliver RH, Leung BS, Lin LY, Yeh J. Estrogen receptor alpha and beta expression in the vaginal walls and uterosacral ligaments of premenopausal and postmenopausal women. Fertil Steril 1999;71:1099–102.[Medline]
26. Moalli PA, Klingensmith WL, Meyn LA, Zyczynski HM. Regulation of matrix metalloproteinase expression by estrogen in fibroblasts that are derived from the pelvic floor. Am J Obstet Gynecol 2002;187:72–9.[Medline]
27. Gebhart JB, Rickard DJ, Barrett TJ, Lesnick TG, Webb MJ, Podratz KC, et al. Expression of estrogen receptor isoforms 
and ß messenger RNA in vaginal tissue of premenopausal and postmenopausal women. Am J Obstet Gynecol 2001;185:1325–31.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
