29 September, 2014: 

Urinary incontinence (UI) is a common and frequently overlooked problem in aging women [1,2]. Leaking urine limits daily and working activities, social interaction and sexual intimacy, and therefore severely disrupts quality of life [3]. The relationship between menopause, aging and hysterectomy is complex and still unresolved. While it is clinically appreciated that all these factors have some impact on the frequency and severity of urinary incontinence, epidemiologic evidence is contradictory and a clear demonstration of how these factors interact in favoring incontinence is not available.

Removal of the uterus is one of the most frequent surgical procedures performed in women, and appropriate counseling related to future risks, particularly that of UI or pelvic organ prolapse (POP), is currently based on feeble evidence.

The Women’s Health Initiative Observational Study (WHI OS) has explored this issue, assessing prevalence, at baseline and after a 3-year time interval, of different forms of UI in 92,093 postmenopausal women between 50 and 80 years (53,569 with the uterus in place, 38,524 hysterectomized) [4]. This is the largest cohort of women in which information on UI has ever been gathered.

The main finding of the study is that prior removal of the uterus is associated with a slightly higher prevalence of UI at baseline (odds ratio (OR) range 1.09–1.69, p < 0.0001) and over the 3-year study period (OR 1.18–1.20, p = 0.0002). All types of UI (stress UI, urge UI or mixed UI) were similarly related to previous hysterectomy. In parallel, no statistical association was found with removal or conservation of the ovaries or with use of menopausal hormone therapy.

Comment

An inherent bias in all studies looking at urinary incontinence lies in the definition of the problem itself. The WHI OS assessed UI using very strict definitions. Women who had ever involuntarily leaked even a very small amount of urine were categorized as having UI, whereas most scientific societies define UI as a condition where urinary leakage determines a clinical complaint [5]. This explains the incredibly high prevalence of UI found in the study (66.5% of the population was categorized as incontinent) that does not correspond to clinical experience. Nonetheless, the study correctly identifies the expected association between known risk factors for UI, such as multiparity and obesity, with actual prevalence of UI, which supports the validity of the experimental setting.

Prior removal of the uterus is associated with an increased prevalence of UI that is similar to that linked to having had one or more children (OR 1.19–1.32, p < 0.0001) or to being overweight and obese (OR 1.24–1.57, p < 0001). In all cases, hysterectomized women differed from women with a conserved uterus only for the prevalence of medium or severe forms of UI, while only subtle differences were seen for very mild forms of UI, which makes sense in many ways.

The finding of a null effect of bilateral oophorectomy or of postmenopausal hormonal supplementation on UI is consistent with clinical practice and the discrepancy with previous studies, including the interventional part of the WHI trial [6] or the available meta-analyses [7] should not surprise, since the relationship between hormonal status and UI has so far only been assessed as a secondary outcome in settings that were not technically appropriate to diagnose it correctly. Thus these results have been considered with caution by urogynecologists.

The large population enrolled in the trial allowed the picking up of small differences in the prevalence of UI based on the presence or absence of the uterus. However, important limitations in data collection do not allow the use of this study as evidence that hysterectomy increases the risk of UI.

There are different hypothetical reasons why surgical removal of the uterus could facilitate UI [3]. Lesions to the hypogastric nerve bundles during dissection of the ureters or the section of the uterosacral ligaments may result in pelvic floor muscle dysfunction. Equally, interruption of the continuity between the fascial support of the anterior and apical compartment may destabilize the pelvic floor. Vascular compromise due to uterine and/or superior vescical artery occlusion may also contribute. All these potentially harmful maneuvers are avoided by supracervical hysterectomy, which many gynecologists prefer for this reason. Information on possible differences in UI prevalence based on removal or conservation of the uterine cervix would have provided stronger evidence to dissect the relationship between surgical pelvic anatomic disruption and urinary incontinence. Unfortunately, this information is not available in the WHI OS.

Another major limitation of the study lies in the absence of any information on the route of hysterectomy or on the prevalence of prior UI or POP surgery. This is a major pitfall, since it is estimated that around 30% of postmenopausal women have a clinically evident pelvic organ prolapse [8], and a significant part of these women also have clinical or latent UI. In many (if not most) cases, these women receive a hysterectomy at the time of pelvic floor reconstruction. Thus, the number of women with pelvic floor defects that preceded (and not followed) hysterectomy should be assessed to be able to infer any conclusion on the role of hysterectomy on UI.

While the limitations of the study do not allow any information to be drawn on the causal role of hysterectomy in the development of UI, I agree with the authors’ conclusion that the suggestion of an increased risk of UI in women who undergo hysterectomy may be helpful as clinicians counsel their patients about the potential risks of this procedure. Too many uteri are still being removed world-wide for questionable indications. Every gynecologist should be aware that hysterectomy might have adverse consequences, so that optimal clinical judgement in surgery is taken. On the other side, the absence of any effect of hormone therapy on UI in this study clarifies how some of the available trials were simply not suitable to address the impact of hormonal interventions on urinary incontinence, and that they have been largely over-interpreted.

Tommaso Simoncini Department of Clinical and Experimental Medicine, University of Pisa, Italy

References

1. Melville JL, Katon W, Delaney K, Newton K. Urinary incontinence in US women: a population-based study. Arch Intern Med 2005;165:537-42 http://www.ncbi.nlm.nih.gov/pubmed/15767530 

2. Irwin DE, Milsom I, Hunskaar S, et al. Population-based survey of urinary incontinence, overactive bladder, and other lower urinary tract symptoms in five countries: results of the EPIC study. Eur Urol 2006;50:1306-14 http://www.ncbi.nlm.nih.gov/pubmed/17049716 

3. Mannella P, Palla G, Bellini M, Simoncini T. The female pelvic floor through midlife and aging. Maturitas 2013;76:230-4 http://www.ncbi.nlm.nih.gov/pubmed/24055286 

4. Kudish BI, Shveiky D, Gutman RE, et al. Hysterectomy and urinary incontinence in postmenopausal women. Int Urogynecol J 2014 Jun 26. Epub ahead of print http://www.ncbi.nlm.nih.gov/pubmed/24964761 

5. Haylen BT, de Ridder D, Freeman RM, et al. An International Urogynecological Association (IUGA)/International Continence Society (ICS) joint report on the terminology for female pelvic floor dysfunction. Neurourol Urodyn 2010;29:4-20 http://www.ncbi.nlm.nih.gov/pubmed/19941278

6. Hendrix SL, Cochrane BB, Nygaard IE, et al. Effects of estrogen with and without progestin on urinary incontinence. JAMA 2005;293:935–48 http://www.ncbi.nlm.nih.gov/pubmed/15728164 

7. Ismail SI, Bain C, Hagen S. Oestrogens for treatment or prevention of pelvic organ prolapse in postmenopausal women. Cochrane Database Syst Rev 2010(9):CD007063 

http://www.ncbi.nlm.nih.gov/pubmed/20824855 8. 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 http://www.ncbi.nlm.nih.gov/pubmed/14749630