The presence of trigonal non-keratinising squamous metaplasia is not associated with menopausal or estrogen status but may be associated with urinary tract infection in women

Introduction

Infection of the lower urinary tract is commonly seen in both community and hospital settings. Although affecting all genders and ages worldwide, women have a 50% lifetime incidence of urinary tract infection (UTI) ¹ with approximately half having recurrence in the 6–12 month period following a primary infection. ² Recurrent or persistent infection of the urinary tract is seen commonly in the post-menopausal female population and has been shown to negatively impact quality of life. ³ It is therefore not surprising that UTIs carry a large economic burden with costs estimated at more than $2 billion dollars every year in the US alone. ²

In women with symptoms suggestive of recurrent UTIs, identification of a causative uropathogen can be challenging as they often have negative urine culture results. ⁴ This impairs the ability to target treatment and increases the prescription of multiple courses of antibiotics, risking microbial resistance, unwanted side effects, and suboptimal bacterial elimination. Cystoscopy is often indicated to investigate women with lower urinary tract symptoms such as haematuria. However, most international guidelines do not recommend the routine use of cystoscopy to investigate recurrent UTIs as cystoscopic evaluation has not been shown to aid in diagnosis or management. ⁵

A common finding on cystoscopy in women, is the appearance of white plaques at the trigone of the bladder known as non-keratinising squamous metaplasia which we will call squamous metaplasia (SM). ⁶ In SM there is replacement of urothelium to stratified squamous epithelium, which has been described as a normal variant in women, ⁷ associated with the influence of estrogen. ⁸ Metaplasia in other epithelial tissue is also activated in response to chronic irritation or infection. It has previously been shown that women with symptoms of dysuria have more severe squamous metaplasia and submucosal fibrosis at the bladder trigone on histological examination. ⁹ It has also been suggested that the trigone may be the site of bacterial reservoirs in women with recurrent urinary tract infections (UTI), which is supported by the success of trigonal fulguration in patients not responding to antibiotic treatment. ¹⁰

In this study we hypothesize that trigonal SM seen in women during cystoscopy is indicative of urinary tract infection in both the pre- and post-menopausal population, and that diagnostic cystoscopy may be an additional useful tool in the evaluation of women with symptoms suggestive of recurrent UTI.

Methods

Medical notes were reviewed retrospectively from pre- and post-menopausal women in a busy tertiary referral centre who had undergone cystoscopic investigation for symptoms of bladder pain, recurrent urinary tract infection, or refractory overactive bladder symptoms from January 2022 to January 2023. Ethical approval for this study was obtained as a urogynaecology tissue subcollection under the trial approval number: URO_VK_16_041 under the Imperial College Healthcare NHS Trust Tissue and Biobank.

All women were investigated with 70 degree rigid cystoscopies performed under a general anaesthetic with the bladder filled to capacity with sterile water for 3 min with the fluid bag being positioned a minimum of 1 m above the patient. The bladder was then emptied and refilled to allow for full evaluation of the dome and trigone. Two cold cup bladder biopsies were taken 1 cm above the trigonal/dome junction of the bladder during each procedure, with one being sent for tissue microscopy, culture, and sensitivities and the second for histological examination. This is usual practice for the investigation of chronic bladder pain as is recommended by the ESSIC (International Society for the Study of Bladder Pain Syndrome), where cystoscopy under general anaesthetic allows for filling of the bladder to capacity emptying and refilling as well as recognition of a pain response without the patient experiencing discomfort. ¹¹ At the time of cystoscopy, a clean catheter sample of urine was collected aseptically following cleansing of the vagina and urethra and sent for microscopy, culture, and sensitivities. Documented findings in the surgical notes were reviewed for a description of the bladder trigone at cystoscopy. Squamous metaplasia was reported as: absent, mild, moderate, or severe.

Microbiological assessment was carried out using both direct and enrichment culture methods on all bladder biopsy samples received. The enrichment culture method was performed to identify more fastidious organisms which often evade conventional culture techniques. The Ballotini bead approach was chosen to liberate organisms through pulverisation without damaging them. The beads provided the added advantage of being made from neutral glass that is free from contamination to ensure maximum recovery of cells from homogenates with minimal risk of cytotoxicity. The steps involved are broadly outlined as follows; aseptically, one piece of pulverised sample was added to 10 mL of Brain Heart Infusion (BHI) broth. The BHI broth is a highly nutritious and versatile liquid infusion medium recommended for the cultivation of streptococci, Neisseria, and other fastidious organisms. It creates conditions of varying oxygen tension which favour the growth and primary isolation of aerobes and anaerobes ¹² whilst even easily cultivated organisms show improved growth in it. ¹³ Once the sample is in the broth it is sub-cultured as soon as it turns turbid (checked daily) or at 5 days if there is no growth or scanty growth, onto plates. All samples were kept incubated at 37°C for at least 48 h and read daily. If no growth occurred, then a terminal subculture was performed again at 5 days from the date of receipt. The direct culture method involved conventional and standard microbiological assessment techniques and if bacteria were isolated from the enrichment culture only, then this was reported as ‘isolated from enrichment’.

We defined proven urinary tract infection as either a positive urine culture or a positive tissue culture. Statistics were carried out using Excel XLSTAT, Microsoft, 2021 with Chi-squared tests performed for independence for ordinal variables was used.

Results

In total ninety-seven women were included in this study with medical notes being reviewed retrospectively. Ages ranged from 18 to 83 with a mean age of 47. There were 39 post-menopausal women (40%) of which 18 (46% of post-menopausal women) were taking either topical, systemic, or a combination of both types of estrogen prior and during the time of cystoscopy. The remaining 58 women were pre-menopausal (60%). In the 97 notes reviewed there were no episodes of return to theatre for bleeding, no episodes of bladder perforation, and no re-admissions for treatment of urine infection requiring intra-venous antibiotics. We were unable to capture data for the number of women who may have presented for treatment of urine infection with oral antibiotics as they were likely to attend their community general practitioner.

There were ten positive urine cultures (10.3%) and fifty-five positive tissue cultures from cold cup tissue biopsy (56.7%). Of the fifty-five positive tissue cultures, there were a variety of organisms grown including bacteria considered to likely be normal microbiome of the bladder. Forty-two of the tissue biopsies taken, cultured known uropathogenic bacteria (43.3%) (see Figure 1). Overall forty-two (43.3%) of the participants had a proven urinary tract infection – defined as either a positive urine culture, a positive bladder tissue culture, or both. We found that bladder tissue culture grew a larger variety of uropathogens (21) when compared with urine culture (4) (see Figure 1).OPEN IN VIEWER

Eighty-two women had squamous metaplasia (SM) reported at time of cystoscopy (84.5%). Twenty-seven were defined as having mild SM, forty-three as moderate SM, and twelve as severe SM. All the histopathology reports described chronic inflammatory change in the bladder biopsies taken from women with squamous metaplasia. We found that the presence of SM had a 62% positive predictive value for UTI, and the absence of SM had a negative predictive value of 73% for UTI, with a sensitivity of 93% and specificity of 76%. Performing a Chi-squared test showed that there is a statistically significant relationship between SM at the trigone and urinary tract infection p = .015 (where <p = .05 is taken to be statistically significant).

We found that with increasing severity of SM, positive urine culture was less likely in this cohort, but positive culture of a uropathogen in bladder tissue biopsy was more likely (see Table 1). There were three patients with a positive urine culture that had a negative tissue culture result, and these were all women with SM reported as mild (n = 2) or moderate (n = 1). The women with the most severe appearance of SM on cystoscopy all had negative urine cultures with five showing positive tissue cultures. Therefore, with increasing severity or grade of SM, urine culture was less likely to identify a possible causative uropathogen and tissue culture was more likely to identify a uropathogen.

OPEN IN VIEWER

Table 1.

Table showing grade of severity of non-keratinising squamous metaplasia seen at the bladder base at cystoscopy and proportions of positive culture results in urine and tissue expressed as a percentage.

	mild SM	moderate SM	severe SM
Sample size	27	43	12
% positive tissue culture with negative urine culture	83.33	95.45	100
% negative tissue culture with positive urine culture	16.67	4.55	0

We also looked at the relationship between estrogen status and the presence of SM, as SM has previously been described as a normal variant in women under the influence of estrogen. Women were grouped together who were pre-menopausal or post-menopausal and taking topical and/or systemic estrogen and compared with post-menopausal women not taking any estrogen.

Seventy-six women in our cohort were under the influence of estrogen (78%), and 21 were post-menopausal and not taking estrogen (22%). Chi-squared analysis showed that there was no statistically significant relationship between the presence of SM and estrogen status in out cohort, p = .866, where statistical significance was taken as p < 0.05.

Discussion

We have found that the presence of non-keratinising squamous metaplasia (SM) correlates to urinary tract infection (UTI) in women with lower urinary tract symptoms undergoing cystoscopy. This study has shown that the presence of SM has a positive predictive value of 62%, and that a negative finding, that is, the absence of SM predicts the absence of UTI in 73%. We have also found that this relationship is statistically significant.

The presence of trigonal SM in women during cystoscopy has been described previously as a common occurrence and has been thought to be a normal variant observed even in those without any lower urinary tract symptoms (LUTS). ⁶ One theory for the cause of trigonal metaplastic change without LUTS is the influence of estrogen. It is thought that the embryological origin of the trigone may be from the common nephric duct which is mesodermal (like the ureters and vagina), rather than being from endodermal origin like the rest of the bladder.^14,15 This may mean that urothelium of the trigone behaves differently in response to exposure to estrogen. Additionally, several groups have found that estrogen receptors are present and distributed in a similar fashion to vaginal epithelium in the bladder trigone of women.^16,17 More recently Seitz et al have found that expression of estrogen receptors is greater in women with trigonal SM than those without. The authors of this paper postulate that the reduced estrogen levels experienced post-menopause may trigger trigonal metaplastic change and stimulate estrogen receptor expression. The resulting SM may then make the bladder more vulnerable to infection. ⁸ This theory is supported by the increasing evidence that topical vaginal estrogens are an important tool in the management of post-menopausal women with recurrent UTIs. ⁵ In our small cohort we have not found a statistically significant relationship between estrogen status and the present of SM (p = 0.866).

Infection alone may be reason enough to induce metaplastic change at the bladder trigone. Many authors have reported on the correlation between recurrent urinary tract infection and the presence of trigonal SM.^10,18–20 Resulting SM may then make bacterial adherence, invasion, and replication more likely, predisposing women with trigonal SM to persistent or recurrent urine infection.^10,21 Our results show that with increasing degree or severity of SM described at cystoscopy, urine culture was less likely to yield a positive result. This seems counter-intuitive as you might expect to see greater metaplastic change of the trigone in more severe infection and therefore have a greater bacterial load with more positive urine culture results. We hypothesise that in our women with moderate and severe metaplasia we may be capturing women with recurrence or persistent pathogenic colonisation. More episodes of infection could trigger more metaplastic change at the trigone. These women are likely to have already received multiple courses of oral antibiotics or even be on long-term prophylactic antibiotics, making urine culture results less informative. Mouse models ²² and human bladder tissue studies ²³ have shown that uropathogens can invade and replicate in intracellular colonies inside cells of the bladder itself. Intracellular bacterial colonies offer a mechanism by which bacteria can evade detection in the urine, treatment from antibiotics, and hide from host immune defences. This may be one reason why we found that culture of bladder tissue was more likely to detect a causative uropathogen in women with more severe SM in our study.

One of the limitations of our study is that some women with bacteriuria will not have urinary symptoms and this does not amount to a UTI that requires treatment. In our study we defined a UTI as either a positive urine or tissue culture result. The reason this definition was chosen was to have a strict objective description defined by culture results. Particularly as the group of women investigated all had lower urinary tract symptoms warranting cystoscopy, this is a cohort with a higher likelihood of having UTI. It is likely that we were discounting some women who may have had UTI with negative cultures of both urine and tissue, but there is small possibility that we have included a small number of women with asymptomatic bacteriuria.

Our results give evidence for the use of cystoscopy as an additional adjunctive tool in women with recurrent or persistent symptoms of UTI with negative urine culture results. Routine cystoscopy is currently not recommended in any international guidelines for recurrent UTI ⁵ ; however, this study supports the use of cystoscopy for aiding the detection of ‘hidden’ infection in women with overlapping symptoms and pathologies. Cystoscopy in such patients may lead to detection of overlying infection and appropriate treatment with antibiotics.

Conclusion

We have found that the appearance of SM at the bladder base is associated with infection, and that this finding is independent of menopausal status or of topical or systemic estrogen use in post-menopausal women. This finding may be particularly useful in women with persistent symptoms of infection clinically but negative urine cultures, or women with overlapping symptoms and pathologies. We conclude that cystoscopy may be a useful adjunctive tool in such selected patients.