Introduction

Urinary tract infections (UTIs) are a common type of bacterial infections in women. It is estimated that up to 50% of women develop at least one UTI in their lifetimes, with as many as 25% having at least one recurrence [1]. Recurrent UTIs (rUTIs) are typically defined as symptomatic episodes occurring at least twice in a 6-month period or three times in a 1-year period [2]. Recurrent UTIs have been postulated to be caused by two main mechanisms, either repeated ascending infections or chronic infections in the bladder [3]. Recent research advances have also identified bacteria unique to the urinary microbiome in rUTI pathobiology. Therefore, further research is needed to understand these interactions [4]. A review from Glover et al. [3] cites animal models to demonstrate the persistence of infection through quiescent intracellular bacterial reservoirs that are protected by biofilms. Additionally, recent data from De Nisco et al. indicated the presence of resident bacteria in the bladder wall biopsies of post-menopausal women with rUTIs, confirming the presence of deep-seated niches of infection in some patients [5]. Taken together, these data suggest that the etiology of infection in the bladder might be important for the management of uncomplicated rUTIs.

Management options for rUTIs can be divided into non-antimicrobial and anti-microbial therapies. Estrogen supplementation, increased fluid intake, probiotics, cranberry, D-mannose, and methenamine hippurate have shown limited success in preventing UTIs [6]. Prophylactic antibiotic therapy has also been recommended, but this approach has raised concerns owing to cost, allergy, resistance, and side effects. Therefore, direct administration of antimicrobial agents in the bladder has been considered. Although the intravesical antimicrobial (IVA) approach has been part of clinical practice since the 1960s, IVAs have typically been reserved as end-of-line management strategies, and thus infrequently used.

With the rise in antimicrobial resistance, there is a need to understand all effective treatment modalities for rUTIs [7]. The aim of this systematic review is to recapitulate all available data on the efficacy of IVAs in the management of uncomplicated RUTIs.

Materials and methods

This systematic review was performed in accordance with Cochrane and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards [8]. A comprehensive review was performed on articles published from inception to April 2021 in the following databases: PubMed MEDLINE, OVID Embase, Biomed Central, and Scopus. Additionally, the references of the articles were searched by hand individually by the reviewers to identify any additional articles that may have been otherwise missed by search engines.

The following are the criteria for including studies in this review.

Inclusion criteria:

  • All full text, English-language prospective cohort, retrospective cohort, and randomized control trial studies of female patients over the age of 18

  • Studies explicitly focused on the intravesical administration of antimicrobial treatment for uncomplicated recurrent UTIs

  • “Antimicrobial” treatment defined as both traditional antibiotic and non-antibiotic therapies targeting bacteria

Exclusion criteria:

  • Individual case reports, review articles, nonhuman studies, abstracts-only texts

  • Studies with male-only or pediatric-only populations; note that studies including both men and women were not excluded to optimize the limited dataset

  • Studies reporting IVA treatments for complicated rUTIs with underlying secondary pathology, including but not limited to neurogenic bladder, catheterization, or spinal cord injury

  • Studies reporting non-antimicrobial intravesical treatments, including Botox injections

The search was conducted using the following key words: “Urinary tract infection,” “lower urinary tract dysfunction,” “recurrent urinary tract infection,” “intravesical drug administration,” “antibiotic treatment,” and “anti-microbial treatment.” Alternative spellings, names, and abbreviations of these keywords were searched. Keywords appeared in the title, abstract, or both. The following medical subject headings were used: “urinary tract infection” AND “intravesical administration” AND “anti-infective agents.”

A total of 476 titles were initially collected. For the first round of extractions, the titles of the articles were reviewed and were excluded if they explicitly did not meet study criteria or were duplicates. Sixty-eight abstracts were further reviewed, from which 15 articles were selected for full-text review. Two independent trained and experienced reviewers (MR and PZ) evaluated all selected articles with a similar methodology.

Following PRISMA guidelines, risk of bias was assessed for each full-text study included. The Cochrane Risk of Bias Tool for Randomized Control Trials was used for critical appraisal. The tool utilizes 7 main domains to assess bias on a scale of judgment (high, low, or unclear). The Joanna Briggs Institute critical appraisal checklist for non-RCTs was used for the remaining studies in this review [9].

Results

Thirteen studies were included in the final review (Fig. 1, flowchart). This included 2 RCTs, 4 prospective studies, 6 retrospective studies, and 1 study that used both retrospective and prospective analysis [10,11,12,13,14,15,16,17,18,19,20,21,22,23]. The countries of origin included Italy (n = 6), United Kingdom (n = 3), United States (n = 1), Greece (n = 1), Austria (n = 1), and the Netherlands (n = 1).

Fig. 1
figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow chart

Full size image

Definition and diagnosis of UTI

The definitions of recurrent UTI are included for each study in Table 1. Eleven out of 13 studies defined recurrent UTI with a culture confirmed method, with the lowest threshold at 103 colony-forming units/ml. A few studies did not specify culture confirmation as a diagnostic tool and based their UTI diagnosis on symptomatology alone.

Table 1 Baseline characteristics of all the studies included
Full size table

Baseline population characteristics

With these studies combined, a total of 761 participants received IVA treatment. Specifically, 3 studies administered gentamicin IVA and the remaining 10 studies administered hyaluronic acid (HA) and chondroitin sulfate (CS). The mean age range of the participants in the studies included was 27–80 (median 53.1). Eleven studies had female participants only and the remaining 2 studies had a male to female ratio of 1:4 (74–77%). Most IVA participants had uncomplicated recurrent UTIs and failed oral antibiotic therapies (Table 1). The number of failed oral antibiotic treatments needed to qualify for IVA differed across studies or were not specified in most of the studies. Only Abrams et al. specified that women in whom double oral antibiotic treatment failed qualified for IVA [21].

Outcomes

Details of the IVA interventions and outcomes are presented in Table 2. All three studies including IVA gentamicin mentioned doses of 80 mg per instillation. There was variability in the amount of solution in which the 80 mg gentamicin was dissolved, ranging from 20 to 60 ml. The frequency of the instillations during the treatment duration varied from 6 to 73. For example, the Chernyak and Salamon study prescribed instillations to participants twice a week for 3 weeks, whereas the Stalenhoef et al. study prescribed instillations to participants once a day for 2 weeks, following which participants received instillations every other day for 10 weeks and then twice weekly for 12 weeks. There was a long follow-up time after these instillations in each of these studies of at least 6 months. Within the 10 studies that included IVA of HA and CS, 6 studies dosed each instillation at 800 mg (1.6%) of HA and 1 g (2%) of CS in 50 ml solution. The remaining 4 studies dosed the HA at 40 mg in 50 ml solution. Additionally, the follow-up time ranged in these studies from 6 weeks to 20 months. The number of HA + CS instillations ranged from 6 to 10 across the studies.

Table 2 Intervention characteristics and primary outcomes of the studies included
Full size table

RCT outcome descriptions

In the RCT by De Vita and Giordano, the primary outcome was the number of UTI episodes, which reported a post-hoc statistical power of 0.023. The treatment group, which received a commercially available IVA of HA + CS, was compared with the control group, which received only oral antibiotic prophylaxis. In the RCT by Damiano et al., the primary outcome was the mean rate of UTI episodes per patient per year. In the study a power of 30 subjects per group was calculated to be necessary to detect an induced difference in a 70% decrease of the UTI rates between groups. The treatment group, which received a commercially available IVA of HA + CS was compared with a control group that received placebo instillations.

Effect of IVA

Most participants who underwent IVA instillations experienced positive outcomes. All three gentamicin studies (87 participants) reported decreases in UTI recurrence after completion of the IVA instillations compared with before IVA. In the study by Stalenhoef et al., the mean number of UTIs decreased from 4.8 to 1.2 (p < 0.001). In the study by Chernyak and Salamon, the mean UTI frequency was reduced from 2.5 to 1.5 UTIs (p = 0.025). Additionally, 9 out of 10 studies with IVA of HA and CS (674 participants) reported a decrease in UTI recurrence. Of these studies, 8 showed a statistically significant decrease in UTI recurrence. Of note, the study by Ciani et al. reported a greater number of UTI recurrences in the HA + CS group compared with the standard care group; however, the authors stated that when adjusted for the incidence risk ratio, there were nonsignificant differences in the treatment and control groups. There was an overall mean decrease of 2.3 UTIs/year in the gentamicin studies and 4.03 UTIs/year across the HA + CS studies.

The discontinuation rates ranged from 0% to 22% in the gentamicin studies. A total of 5 participants (7.42%) dropped out of the treatment course across all HA + CS studies. The most frequently cited reason was loss to follow-up. Adverse events were minimal across all studies. Of the gentamicin studies, the study by Stalenhoef et al. reported IVA side effects of abdominal discomfort (5%), hearing loss (3%), and vaginal discomfort (20%). In the HA and CS studies, only minor side effects were cited. Resistance to treatment rates were reported in 2 gentamicin studies. The studies by Abrams et al. and Stalenhoef et al. reported 11 patients with gentamicin-resistant infections after instillation treatment. None of the HA + CS studies reported resistance rates after treatment.

Risk of bias

Of the 13 studies included in this review, 2 studies were identified as RCTs and the Cochrane Risk of Bias Tool for Randomized Control Trials was used for these 2 studies. The Joanna Briggs Institute critical appraisal checklist for non-RCTs was used for the remaining 11 studies in this review.

A summary of the risk of bias assessment for the studies is provided in Tables 3 and 4. The majority of the non-RCT studies studied outcomes in the same group of participants before and after IVA treatment, leading to minimal differences in exposures between the two groups for comparison. We found that most of the studies did not explicitly identify strategies in dealing with confounding variables. Despite high levels of bias in select categories, we found the quality of the 13 studies to be high enough for inclusion.

Table 3 Risk of bias assessment of the randomized control trials included in this review. The Cochrane risk of bias tool for randomized control trials was used
Full size table
Table 4 Risk of bias assessment of the nonrandomized controlled trials included in this review. Joanna Briggs Institute critical appraisal checklist for non-RCTs was used
Full size table

Discussion

Echoing recent American Urological Association (AUA) guidelines on rUTIs, this review focused on uncomplicated rUTI treatments and explored the efficacy of intravesical antimicrobial therapy for uncomplicated recurrent UTIs in a predominantly female population [24]. The search yielded two frequently used antimicrobials: gentamycin and hyaluronic acid with chondroitin sulfate therapy. Both therapies yielded a successful decrease in recurrent UTI rates, with an overall mean decrease of 2.3 UTIs/year in the gentamicin studies and 4.03 UTIs/year across the HA + CS studies.

Intravesical drug administration provides a localized target for delivery. The urothelium lining provides a strong barrier that allows administered drugs to achieve a localized luminal effect on the planktonic bacteria while minimizing reabsorption, thus leading to fewer systemic adverse effects [25]. Even at considerably high dosages, absorption of IVA has been minimal when tested with blood serum levels. Intravesical gentamicin has been able to combat multi-resistant E. coli in the bladder [26]. The efficacy of HA and CS agents is related to their ability to replenish the glycosaminoglycan (GAG) layer at the bladder surface [25]. Evidence surrounding this GAG layer theory has been supported by various animal models, such as in porcine or rat cells [27, 28].

This review underscores the high level of efficacy of IVAs in women with uncomplicated rUTIs failing repeated oral antibiotic courses. Gentamicin provided a significant decrease in UTIs at least up to 6 months after IVA treatment. Because the frequency of instillations differed across studies, ranging from 6 to 73, it was not possible to determine the optimal number of instillations to reach a favorable outcome. Of note, all three studies used the same dosing of 80 mg of gentamicin per instillation, demonstrating the efficacy of the dosage. Although IVAs are generally seen as less likely to cause antimicrobial resistance than oral antibiotics owing to its localized effect, two of the gentamicin studies included reported gentamicin-resistant infections in patients after at least 6 months of treatment exposure. Evidence is limited in the resistance to gentamicin after several months of bladder instillation exposure.

The European market offers two different formulations of HA and CS IVA, either 0.08% HA (40 mg) or 1.6% HA (800 mg) [25]. Most of the studies included utilized the 800-mg dosage, but both dosages were found to be efficacious for UTI reduction. A 2017 review highlighted HA and CS therapy as a novel strategy to reduce the risk of potential drug resistance that could be found when using other IVAs, such as with gentamicin in limited cases [29]. Although there is preliminarily research touting the efficacy of HA IVA, this therapy is not available in the United States for therapeutic use owing to the lack of FDA approval [30].

Cost of treatment is an important aspect of rUTI management. According to the Medicare fee service program, the national average cost of one intravesical instillation of gentamicin is $82.90 [31]. The cost of a commercially available HA + CS instillation, named iAluRil, is £105.60 ($146.81) in the United Kingdom [32]. Comparing cost efficacy across treatment regimens is difficult owing to unstandardized protocols; however, all studies included in this review used at least six instillations per patient to find significant utility in IVA. Using a minimum of 6 instillations per course would bring the total cost of gentamicin IVA to $497.4 and for HA + CS IVA to $880.86 per treatment course. Even though IVA treatments tend to be costlier than oral antibiotics and other conservative therapies, they are less expensive than IV antibiotics, which would be similarly used for more recurrent infections [33].

Limitations and areas of future research

Our review was limited by the nature of the studies (observational or cohort), with only two RCTs. Owing to the high level of heterogeneity across study designs and types, it was not possible to compare the outcomes across these studies through a meta-analysis. Additionally, two studies had a limited number of male participants in their cohort that were not separately analyzed [14, 21]. These two studies were not excluded from this review but explains why our conclusions stipulate that the reported data were “mostly for women.”

There is a greater need for RCTs on IVA therapy in female participants with rUTIs. Particularly, greater efforts in exploring other antimicrobial therapies could be beneficial in expanding the field of intravesical treatments. Other IVA therapies have been reported but are limited to case reports or are not large-scale studies [34]. For example, in 2004 Wood et al. described a case of tobramycin IVA used to treat UTI in a 69-year-old woman. This patient had a protracted hospital course during which she developed Enterobacter cloacae, which was only susceptible to cefotetan and tobramycin [35]. Additionally, the study by Dutta and Lane found promising results in which 39 women were treated weekly with heparin IVA at a dose of 10,000 units mixed with 250 mg solumedrol in each instillation over the course of 6 weeks [36]. In the study by Stalenhoef et al., 3 patients were treated with a different aminoglycoside owing to culture resistance to gentamicin at baseline (amikacin (1) and tobramycin (2)) without comments on their effectiveness. Further exploration of the efficacy of these IVA therapies through larger cohort or randomized controlled trials may broaden the scope of these treatments, especially when IVA gentamicin or HA + CS cannot be utilized.

Additionally, almost all of the studies included enrolled women in whom antibiotic therapy courses originally failed. However, most studies did not specify the number of failed courses before utilizing IVA therapy. Further research on defining the ideal candidates for IVA after oral antibiotics would help clinicians interested by this treatment modality in real life practice. Furthermore, although the majority of studies followed a weekly instillation schedule, the Stalenhoef et al. study reported a daily instillation schedule for the first 2 weeks. Further research exploring different administration intervals is therefore needed.

Recurrent UTIs are associated with creating a clinical, personal, and economic burden of illness [37]. Therefore, there is a need to explore the benefits of IVA therapy beyond reduction in rUTI rates. IVA therapy has been able to improve sexual dysfunction and other quality-of-life symptoms related to rUTIs [38]. A systematic review by Goddard and Janssen included three studies that used the pelvic pain and urgency/frequency (PUF) assessment tools to report on symptom change. The mean change in PUF score was −6.55, with the lower and upper scoring limits being 0 to 36 [39]. Therefore, there is a need to explore the benefits of IVA therapy beyond a reduction in rUTI rates.

Conclusions

In this review with a limited number of IVA studies on uncomplicated rUTIs, predominantly in women, instillations of gentamicin and hyaluronic acid with chondroitin sulfate had efficacy in the treatment of rUTIs when oral antibiotics had failed. Dosages of 80 mg of gentamicin per instillation and both 40 mg and 800 mg of hyaluronic acid were used effectively with limited adverse effects and short-term benefits. These encouraging findings should stimulate further research into the optimal modality of IVA instillation and into better defining ideal candidates for this bladder-targeted therapy.