Introduction

Due to an ever-increasing population of elderly individuals, clinicians will be called on to manage the hypoestrogenic patient more frequently. The effect of hormonal loss on lower urinary tract symptoms (LUTS) and voiding dysfunction is often underappreciated by both patients and clinicians. The collection of symptoms that includes vasomotor symptoms, vaginal dryness, and urinary urgency, dysuria, and recurrent urinary tract infections has recently been renamed genitourinary syndrome of menopause (GSM) [1]. Stress urinary incontinence (SUI) worsens in the early postmenopausal period, while overactive bladder (OAB), with or without urge incontinence, becomes more prevalent as women progress further beyond menopause. Elderly women are also prone to chronic, recurrent urinary tract infections (UTIs). Treatment of all these conditions, be it anticholinergic medications for OAB, repeated antibiotics for UTI, or surgery for SUI, carries potential side effects and complications.

Hormone replacement therapy (HRT), specifically estrogen replacement therapy (ERT), offers an alternative or adjunctive treatment option. Identifying who is an appropriate candidate for ERT, how to administer it, and how to counsel and monitor the patient on ERT for treating LUTS is important. The focus of this review will be on the pathophysiology of hypoestrogenism and its relationship to lower urinary tract symptoms as well as differences between topical and systemic ERT.

Pathophysiology of Hormonal Deficiency on the Lower Urinary Tract

Despite their discovery in the 1950s, estrogen receptors (ER) in the proximal and distal urethra, vagina, and trigone of the bladder were not identified until the 1980s [2]. In bladder mapping these receptors, Blakeman et al. demonstrated that estrogen receptors were not expressed by normal transitional cell epithelium; they were, however, expressed by all tissues containing squamous epithelium including transitional cell epithelium in the trigone and proximal urethra that had undergone squamous metaplastic changes. In addition, estrogen receptors have also been identified in the musculature of the pelvic floor with the exception of the levator ani muscles [3, 4], and Chen et al. found ER in the vaginal walls and uterosacral ligaments of premenopausal women despite being absent in postmenopausal women [5].

Given the location of these receptors, it is easy to understand why hormonal changes can manifest clinically as LUTS. Robinson et al. demonstrated that cyclical variations in the levels of both estrogen and progesterone during the menstrual cycle led to changes in urodynamic variables and LUTS, with 37 % of women noticing stronger symptoms before menstruation [6]. After menopause, these changes are no longer cyclical and remain permanent, or progress. The loss of estrogen in menopause has also been shown to lead to subsequent thinning of the urethral mucosa impacting not only the integrity of the pelvic floor but also to a more sensitive and trauma-prone bladder [7].

On a cellular level, Liang et al. looked at the exogenous effects of estrogen on cellular proliferation and apoptosis of the detrusor muscle after ovariectomy in mice. Increased apoptosis was observed in the bladders of ovariectomized rats without intramuscular estrogen replacement suggesting the structural change was related to estrogen deprivation. In the same study, there was also a significant decrease in apoptosis observed in the estrogen replacement group, indicating that cell apoptosis in the detrusor muscle was at least in part mediated by estrogen. Proliferative indices also supported the role of estrogen as a stimulator of cell proliferation [8].

Progesterone receptors are also expressed in the lower urinary tract. Although their role is less clear, as they are inconsistently reported in the bladder, trigone, and vagina [9], progestogens have been associated with an increase in irritative bladder symptoms and urge incontinence in women taking combined HRT [9]. A murine study showed an increase in detrusor overactivity (DO) during the luteal phase of the rat menstrual cycle. This finding was thought to be due to increased plasma progesterone levels after ovulation given the antagonizing effect it has on estrogen which has been shown to inhibit rat detrusor contractions [10].

In addition to estrogen, recent studies have investigated testosterone and its effect on LUTS as androgen receptors have been identified in the vagina, bladder, and urethra [9]. Compared to the acute decline of estrogen at menopause, testosterone also decreases with age, but in a more gradual pattern. The decline in testosterone (T), dehydroepiandrosterone (DHEAS), and androstenedione is steepest in the early reproductive years, flattening out in midlife with a tendency for a small increase in the later years [11]. Of note, a 40-year-old woman has half the mean plasma total testosterone as a 21-year-old woman [12]. Testosterone is also the major substrate for estrogen in both men and women and thus has an indirect effect via the estrogen receptor [13]. A recent Korean study examined which hormones contributed to LUTS and sought to determine if changes in hormone levels during the menopausal transition significantly affected LUTS using standardized voiding symptom questionnaires. Their results suggested that serum testosterone levels in women undergoing the menopausal transition negatively correlated with LUTS [14•]. Lin et al. studied contractility changes in the bladders of female rabbits that were given an aromatase inhibitor to a control group. Their results identified testosterone as an important regulator for the female rabbit bladder in the absence of estrogen [15].

Estrogen and Overactive Bladder

The worldwide prevalence of overactive bladder (OAB) was estimated to exceed 500 million in 2013, and the number of patients affected by urgency urinary incontinence (UUI) is expected to increase from 49 million in 2008 to 60 million in 2018 (22.5 % increase), which corresponds to an increase in prevalence from 1.15 to 1.21 % [16]. OAB with urge incontinence in women has also been shown to increase with age more than ninefold, from 2.0 % in those 18–24 years of age to 19.1 % among those 65–74 years of age [17]. The average worldwide annual direct cost of OAB was €1.2 trillion to €2.7 trillion in 2008 and can be expected to increase to €1.4 trillion to €3.2 trillion by 2018 [16], which makes it imperative to find effective treatments for overactive bladder and urge urinary incontinence.

Several studies conducted on the use of ERT in the setting of OAB and UUI have shown estrogen to increase the sensory threshold of the bladder in some women [18], increase urethral closure pressure and alpha-adrenergic tone, improve urethral blood flow, and help cellular maturation in the lower urinary tract [18]. A 2003 Cochrane systematic review found that systemic estrogen treatment could improve or cure urge incontinence in about 50 % of patients, compared to 25 % with placebo [19]. In later studies, however, the Women’s Health Initiative (WHI) trial [20] and the Heart Estrogen/progestin Replacement Study (HERS) [21] found increases in incontinence among women taking systemic equine estrogen, with or without progestin. Notably, these results were not primary outcome measures of either study; both studies were designed to assess cardiovascular risk with HRT. Other problems with these studies include uncontrolled confounders such as body mass index, tobacco use, and caffeine use. In addition, no other types, routes, or dosages of estrogen were evaluated [22].

More recently, a study conducted by Northington et al., using standardized interviews of community-based women initially reporting no urinary incontinence (UI), found postmenopausal women with a history of estrogen use for 5 or more years were likely to report newly incident UI after 10 years of follow-up [23]. An updated 2012 Cochrane review determined UI might be improved with the use of local estrogen treatment. There was little evidence from the trials on the long-term effects of the therapy. Conversely, systemic hormone replacement therapy using conjugated equine estrogen was found to worsen incontinence, but there was not enough data to address estrogen type and dose, and no direct evidence comparing routes of administration [24].

The basic science of HRT and its relationship with OAB and UUI has been elucidated in several studies. In a placebo-controlled, randomized clinical trial, women receiving 2 mg of oral estradiol valerate over 6 months showed significant decreases in total periurethral collagen. The study observed profound effects of estrogen on collagen metabolism, with both increases in synthesis and degradation via increased matrix metalloproteinase-2 activity. They surmised replacing aged collagen with new tissue increased the overall cross-link concentration and may therefore result in increased tissue strength, despite overall collagen loss. They suggested a longer treatment interval would likely show total collagen content increase [25]. Conversely, an in vivo study demonstrated that ovariectomized rats showed a decrease in voided volume and an increase in 24-h frequency, and estrogen replacement reversed these changes [9].

Similarly, soy isoflavone replacement was shown to reduce urodynamic DO. Soy food is a rich source of phytoestrogens, namely isoflavones, known to have both estrogenic and anti-estrogenic effects. In an estrogen-deprived environment, isoflavones may exert estrogen-like effects, serving as estrogen agonists [26]. In this study, connexin-43 (which aids in intercellular communication) expression was significantly greater in both the suburothelial and detrusor layers of the urinary bladder of ovariectomized rats, which induced DO. Isoflavone replacement reduced this increased expression [27]. In other studies, estradiol has been found to reduce the amplitude and frequency of spontaneous rhythmic detrusor contractions and may increase the sensory threshold of the bladder in some women [28, 29].

Other promising studies have focused on testosterone and dehydroepiandrosterone (DHEA) replacement. Cayan et al. found that bladder dysfunction is related to estrogen and androgen deficiency, and combination therapy may improve bladder functions and histology much better than estrogen therapy alone [30•]. Another study found that after ovariectomy in rabbits, testosterone infusion completely restored vaginal smooth muscle structure and contractility [31]. Casson et al. also argued for comprehensive replacement therapy because they discovered postmenopausal ERT alone decreased DHEAS and T by 23 and 42 %, respectively. ERT was actually found to exacerbate the endogenous adrenal and ovarian androgen deficiency state seen in postmenopausal women [32]. If androgens do in fact improve bladder functions and restore vaginal smooth muscle structure, a comprehensive replacement therapy may lead to better symptomatic improvement. While androgen use in the postmenopausal woman has been used for sexual dysfunction, its role in the management of OAB remains uncertain at this point.

Given such promising findings, researchers have attempted to identify if ERT in lieu of, or in addition to, standard pharmacotherapy, specifically antimuscarinics for OAB, is beneficial. Nelkin et al. found that ultralow-dose estradiol-releasing vaginal ring replacement and oral oxybutynin were similarly effective in decreasing the number of daily voids in postmenopausal women with an overactive bladder [33]. Another study by Serati et al. found no synergistic effect of local estrogens and antimuscarinics in the treatment of OAB [34].

Looking at a completely different pathway, Mishra and colleagues reported that urge urinary incontinence is not related to estrogen deficiency, but rather childhood enuresis. They reason that UUI in some adults represents an inadequacy in the neurological reflex mechanisms [35]. This relationship was independent of psychological symptoms before the onset of menopause, suggesting that UUI, childhood enuresis, and psychological symptoms may have a common causal factor [36]. Given these findings, it is likely that the etiology for OAB in the elderly is multifactorial, and HRT alone may not be an effective therapeutic option.

Hormonal Changes and Stress Urinary Incontinence

Unlike OAB and UUI, ERT for stress urinary incontinence (SUI), another bothersome LUT condition in postmenopausal women, has been a different story. Jackson et al. randomized 66 women to 2 mg estradiol valerate or placebo but found no subjective or objective clinical benefit at 6 months [37]. Subsequently, the Women’s Health Initiative (WHI) report in 2002 [38], a landmark study enrolling more than 27,000 healthy postmenopausal women (approximately 16,000 with an intact uterus and 10,000 post hysterectomy), demonstrated both an increased incidence of UI and worsening of baseline UI in women on oral conjugated equine estrogens (CEE) with or without medroxyprogesterone acetate (MPA). The authors used SUI as an outcome measure based on a single question, “When do you usually leak urine?” with “When I cough, laugh, sneeze, lift, or exercise.”

Baseline SUI prevalence was similar at 24–26 %, but at 1 year, the relative risk of daily SUI in either the CEE or MPA arm vs. placebo was 2.48 and 2.39, respectively [21]. Recently, the WHI published a long-term follow-up study and demonstrated that the risk of UI continued to remain higher in the oral ERT group compared to placebo at 13-year follow-up; the risk of UI dissipated after ceasing therapy [39•]. The mechanism of action has been purported to be due to breakdown of total periurethral collagen, which in turn can affect urethral closure pressure [25].

The Nurses’ Health Study II, another well-designed trial studying systemic ERT, used mailed questionnaire responses to over 100,000 women in 1989 and with follow-up questionnaires in 70,000 women in 2001, asking about UI in women on systemic ERT, oral or transdermal, with or without progestin. The authors reported incident UI, occurring at least monthly, in 1026 of 7341 women who responded and concluded that the odds ratio for any urinary incontinence was 1.39; this was similar for women on oral estrogen alone or with progestin. The risk of new SUI, however, was much lower with an odds ratio of 1.09 [40].

Topical estrogen, unlike systemic therapy, has been poorly studied in the clinical treatment of SUI but could have an indirect benefit on the pathophysiologic mechanisms of SUI. The Cochrane database review [24] reported on 17 studies involving estrogen therapy for SUI, but the most common setting was in studying topical estrogen in relationship to the use of perioperative topical hormones [41, 42]. Beisland et al. in a small, but well-designed trial showed that both phenylpropanolamine and estriol increased the maximal urethral closure pressure, with combined therapy proving even more effective [43]. Topical estrogen therapy has also been shown, using Doppler flow, to increase periurethral vasculature in women on three different types of topical estrogen therapy [44]. Similarly, Long et al. identified that both topical and oral estrogen replacement decreased the pulsatility index at the bladder neck and mid-urethra, which was indicative of improved blood flow to the area [45]. It should be noted that none of these studies involved a placebo arm.

Estrogen Replacement as Prevention for Recurrent Urinary Tract Infections

Recurrent urinary tract infections (UTIs) are another aspect of GSM. Risk factors for recurrent UTIs include urinary incontinence, incomplete bladder emptying, and changes in the lower urinary tract environment. Estrogen is effective in preventing infections by strengthening the epithelial integrity of the urothelium [46] as well as maintaining an environment of healthy microbes including lactobacilli. Lactobacilli in turn maintain a low pH environment which prevents other uropathogenic bacteria from thriving [47]. These are the two mechanisms by which estrogen is thought to be effective. The adjunctive use of ERT in reducing OAB and incontinence, as already discussed, may indirectly contribute to reducing the number of recurrent UTIs in the elderly female patient.

The clinical utility of ERT for recurrent UTIs is well established in a number of placebo-controlled trials, perhaps more so than its relationship to OAB. A randomized control trial (RCT) comparing Estring, an estradiol secreting vaginal ring, to no estrogen treatment found a 45 % chance of being infection free after 36 weeks of treatment compared to 20 % in the non-treated group (p = 0.008). Estring lowered the vaginal pH and increased the time to first recurrence [48]. In a recent Cochrane review examining nonantibiotic therapies for recurrent UTIs, they identified two RCTs with vaginal therapy and found that both studies significantly reduced the proportion of women with UTIs. Interestingly when examining trials with oral estrogens, the proportion of women with UTIs did not decrease, and in fact, an increase in adverse events was noted [49]. In the RCT by Raz and Stamm, 93 women were randomized to receive topical intravaginal estriol cream or placebo. At 8 months of follow-up, they found that the estriol group had significantly reduced UTIs (0.5 vs 5.9 episodes per patient year, p < 0.001). Lactobacilli reappeared in 61 % of the women treated with estriol compared to none of the placebo patients (p < 0.001). The vaginal pH decreased from 5.5 to 2.8 in the estriol arm with a subsequent decrease in the vaginal colonization with Enterobacteriaceae (67 to 31 % vs. 67 to 63 %, p < 0.005) [50]. In a meta-analysis of estrogen therapy for UTIs, data from 334 subjects were aggregated and estrogen was found to be superior to placebo (odds ratio = 2.51, 95 % confidence interval = 1.48–4.25). However, in the 10 studies included, different routes of estrogen administration were included making it difficult for comparison. The data from the studies with vaginal estrogen administration showed the best efficacy [51].

Based on the results from several studies, vaginal estrogen is an effective adjunctive therapy for patients with recurrent UTIs, particularly in the era of increasing antibiotic resistance. A typical regimen is the daily use of vaginal cream for 2 weeks followed by a twice-weekly maintenance dose [50]. We will discuss clinical use of HRT later in this review.

Newer Hormonal Replacement Therapy Agents

The valid concerns over increased risks of cardiovascular complications and malignancy with unopposed systemic estrogen replacement have led to the introduction of alternative hormonal manipulation treatments. Ospemifene (Osphena®, Shionogi) is a selective estrogen receptor modulator (SERM) that acts as a tissue selective estrogen agonist/antagonist. Recently, two phase III randomized trials comparing ospemifene at 60 mg/daily to placebo demonstrated a significant improvement in vulvovaginal dryness and reduction in dyspareunia [52, 53]. A 52-week extension to monitor for adverse events noted no increase in pelvic organ prolapse or incontinence; however, the definition of incontinence other than as a self-reported AE was not reported in this study [54]. The incidence of UTI was slightly higher in the ospemifene arm, but overall incidence was low [55•].

In contrast to ospemifene, prior SERMs were found to have deleterious effects on women’s lower urinary tract symptoms. Earlier SERMs, including idoxifene and levormeloxifene, were aborted during phase III clinical trials due to a fourfold higher rate of incontinence, as well as double the rate of worsening pelvic organ prolapse compared to placebo [56]. Subsequently, another second-generation SERM, most notably raloxifene, was compared to placebo and showed no a significant increase in the rate of new incontinence, with an odds ratio of 1.05 after 3 years [57]. Lasofoxifene, a newer SERM prior to ospemifene, was studied in 8556 women, using the King’s Health Questionnaire as an outcome measure for UI. The authors reported a similar incidence in UI and pelvic organ prolapse between 0.25 mg, 0.50 mg, and placebo; however, the incidence for new UI or prolapse surgery was higher in the 0.25 mg group compared to placebo. In addition, there was a higher incidence of endometrial polyp development in those women followed with transvaginal ultrasound [58].

The use of “bioidentical” hormones has increased in popularity in light of the well-publicized risks of systemic ERT. These include estrogens (i.e., a combination of 17-beta-estradiol, estrone, and estriol), progesterone, testosterone, and perhaps DHEAS that have been derived from plant extracts and modified chemically to “mimic” endogenous hormones [59]. While there has been evidence to show benefit in treating the vasomotor symptoms of menopause, to date, there are no studies to assess whether bioidentical hormones have any positive or negative effect on lower urinary tract symptoms.

HRT Use: a Practical Approach

The type of local replacement therapies includes the more commonly prescribed, commercially available CEE creams (Premarin®, Estrace®), 17B estradiol vaginal suppositories (Vagifem®), and estriol pessaries and rings (e.g., Estring®). A Cochrane meta-analysis concluded that, while all forms of local estrogen are efficacious, estradiol tablets were better tolerated than creams, with patient preference also favoring estrogen rings over creams in head-to-head studies [60, 61]. Overall, there was much heterogeneity between studies, which makes direct comparison of the various forms of topical estrogen difficult to interpret [62].

It has been our experience that women who show objective evidence of postmenopausal vaginal atrophy (increased parabasal cells on wet mount, pH 6.0–7.5, visible epithelial changes) and who have signs and symptoms of vaginal dryness, dyspareunia, and/or a history of culture-proven recurrent bacterial UTIs should be offered topical HRT. The choice of agent is based on tolerance, cost, and effectiveness during initial use. In general, we use the lowest allowable dose (0.5 g twice weekly) but will advise a “loading dose” of 0.5–1.0 g daily for the initial 7 days in women with severe atrophy and marked local irritation, tapering down to thrice weekly, then to twice weekly thereafter. If cost or tolerance of topical treatments is an issue, we refer to a local compounding pharmacy that can provide the same product in a hypoallergenic, alcohol-free base. In this case, we use estradiol emollient cream at a low concentration of 0.2 mg/g, increasing to a maximum strength of 0.8 mg/g if necessary and as tolerated.

The evidence we have shown suggests that topical HRT has sufficient benefit for women with vulvovaginal atrophy as well as OAB refractory to dietary and behavioral therapy, or with limited response to a single anticholinergic agent. However, women who do not show any improvement after a 3-month trial of local HRT would be advised, if not encouraged, to cease therapy as risk may outweigh benefit in this cohort, particularly if they have an intact uterus. In general, we avoid the use of any HRT in women with a history of estrogen receptor-positive breast cancer or deep venous thrombosis. Women who have undergone treatment for endometrial cancer would be considered high risk for any type of HRT; however, a recent meta-analysis showed no increased risk for endometrial cancer recurrence with HRT [63]. It would also be prudent to advise patients to follow-up with an endometrial ultrasound on a semi-regular basis if they choose to continue use of any type of local HRT for an extended (>1 year) period of time.

The use of local ERT at the time of vaginal surgery for pelvic organ prolapse (POP) is based on maximizing vaginal epithelial thickness to optimize surgical repair [64, 65]. In a randomized trial by Rahn et al., vaginal biopsies were performed at the time of hysterectomy in women who received estrogen therapy compared to placebo. They noted a significant increase in epithelial and muscularis thickness as well as in the synthesis of mature collagen [66]. Since loss of vaginal epithelial thickness increases the risk of mesh extrusion, a complication unique to vaginal mesh, this is also relevant in light of the current legal environment regarding vaginal mesh surgery.

Conclusion

The use of estrogen replacement for hypoestrogenism in postmenopausal women remains controversial, with level 1 evidence demonstrating that the risk of systemic unopposed estrogen use outweighs the benefit. Judicious use of local HRT is open to debate and counseling with individual patients but benefits to the lower urinary tract—particularly for symptoms of recurrent UTI and OAB—likely outweigh risks of endometrial hyperplasia, malignancy, cardiovascular risks, and breast cancer. Whether or not newer agents, such as SERMs and/or bioidentical hormones, are beneficial for voiding dysfunction has yet to be elucidated.