Introduction
The cornerstone of catheter associated urinary tract infection (CAUTI) treatment is antibiotic therapy. One significant knowledge gap pertains to whether it is necessary to replace the catheter as part of the treatment for CAUTI. In case of obstruction or malposition of the catheter, there is consensus — the catheter needs to be replaced. However, for a well-functioning catheter, the debate is more contentious. Current guidelines recommend to change the catheter if it has been in place for longer than two weeks [
1,
2]. This is believed to result in faster recovery, a shorter duration of symptoms, and a lower risk of recurrent CAUTI. However, this recommendation is based on limited evidence.
The effectiveness of catheter replacement in mitigating symptoms or preventing recurrent CAUTIs is a topic of ongoing debate. While guidelines suggest that routine catheter replacement may reduce the risk of re-infection, others argue that this practice may be unnecessary, costly, and potentially harmful to patients due to the risks associated with catheter removal and insertion.
The aim of this systematic review was to assess the available evidence regarding catheter replacement for CAUTI on recurrence of CAUTI and clinical outcome.
Methods
This systematic review was reported in accordance with the
Preferred Reporting Items for Systematic reviews and Meta-analyses (PRISMA) 2020 guidelines [
3]. Details of the protocol for this systematic review were registered at PROSPERO (CRD42023467230).
We included studies investigating the effect of catheter replacement in patients with CAUTI on duration of symptoms and/or recurrence of urinary tract infection (UTI). Eligible studies encompassed various catheter types, including transurethral, suprapubic, nephrostomy, and JJ-catheters. Inclusion criteria extended to all adult patients, including those who were immunocompromised. Additionally, all time points for catheter replacement—whether before the initiation of antibiotic therapy or during antibiotic therapy—were considered for inclusion.
To be included, a study had to report on either recurrence of UTI, duration of symptoms, clinical cure rate, length of hospital stay and/or mortality. Studies that were limited to asymptomatic bacteriuria were excluded. To avoid language bias, studies published in non-English language journals were eligible for inclusion if one of the investigators could read the foreign language (French, Italian, Spanish, German and Dutch). All study settings (community, outpatient and inpatient) were allowed.
We searched multiple electronic databases: PubMed, Embase, Web of Science, the Cochrane Library, and Academic Search Premier, clinicaltrials.gov from inception to October 15th 2023. Our search strategy, constructed by an experienced librarian and based on a PICO-style approach, is provided in Supplement
1. Next, we carried out a ‘snowball’ search to identify additional studies by searching reference lists of study reports included in this systematic review. We did not apply any filters regarding date of publication. Before submission of this review, a search update was conducted to identify recently published studies.
References were imported into Covidence software [
4]. Title and abstract screening, as well as full-text screening, was performed independently by two reviewers (ML, AW). In case of disagreement, consensus was reached by discussion between the two researchers. In case of persisting disagreement, a third researcher was consulted (JvU). For each study that was selected for data extraction, the following information was collected: study design, eligibility criteria, population characteristics, number of participants, type of catheter, definition of CAUTI, recurrence of UTI (and definition), duration of symptoms (and definition), clinical cure rate (and definition), mortality rates, length of hospital stay, ICU-admittance, complications of catheter replacement, and duration of follow-up. For each paper, data extraction was performed independently by two reviewers (LP and ML). In case of disagreement a third reviewer was consulted (AW).
The Newcastle Ottawa Scale was used for assessing risk of bias in cohort studies and the RoB tool for assessing bias in randomized trials [
5,
6].
Discussion
This systematic literature review identified four articles that investigated the effect of catheter replacement in patients with CAUTI. The two observational studies showed a neutral effect of catheter replacement [
9,
10]. The two RCT’s had contradicting results [
7,
8]. One RCT favoured catheter replacement, demonstrating higher rates of clinical cure, and lower recurrence rates in the replacement group [
7]. The other RCT favoured retainment, with a lower recurrence rate in the retainment group [
8].
Notably, out of the four identified studies only one study endorses catheter replacement as a component of CAUTI therapy [
7]. This study provided the basis for the current guideline recommendations for catheter replacement in patients with CAUTI. The study suggests a substantial benefit of catheter replacement on both time to symptom resolution and recurrence rate. The difference with the other three studies may be explained by the generally low sample sizes and the heterogeneity in patient populations, definitions of CAUTI, antimicrobial treatment strategies and outcome assessments.
The overarching limitation lies in the modest scale of these studies. Both RCT’s have small sample sizes and the observational studies exhibit various methodological challenges and potential biases, of which (non-measured) confounding is the most important one. For example, doctors may be more inclined to change the catheter if the patient is severely ill or has had recurrent infections in the past. In all studies, misclassification of CAUTI is a potential risk both in the inclusion criteria and in defining recurrences. Diagnosing CAUTI is notoriously difficult, which is reflected by the different definitions used in the four studies. Asymptomatic bacteriuria may be erroneously diagnosed as CAUTI. For example, in the study by Kumazawa, the diagnosis was based on bacteriuria and pyuria, and symptoms were not specified. In the study of Raz et al., systemic symptoms were required for diagnosis, but no other investigations were performed to exclude alternative causes. In five patients of this study, urine cultures were already negative before start of treatment, which would not be expected in CAUTI. In two studies, either dosing or length of antibiotic therapy is a confounding factor. However, in the study by Kumazawa, levofloxacin dosage was lower in the retainment group, and therefore is not expected to impact the finding that catheter replacement did not improve clinical outcome. This is different for the RCT by Darouiche, where patients that were randomized to retainment of the catheter, received a longer duration of antibiotics, which may mask an effect of catheter replacement.
There are limitations to the current review. Firstly, there is the inherent risk of not capturing all relevant literature, despite efforts to conduct a comprehensive search. Additionally, reliance on the information provided in the selected studies introduces the potential for information bias. While the review benefitted from the expertise of a librarian and involved a duplicate screening process to enhance reliability, the human element in interpretation remains.
On theoretical grounds, catheter replacement is rational. It aligns with the general principle of source control in infectious disease management, aiming to eliminate or reduce the source of infection to optimize patient outcomes. Removing the source, i.e., the catheter, could theoretically accelerate symptom resolution. Furthermore, biofilms form on catheters, and pathogens can persists in these biofilms despite treatment [
12]. Therefore, removing the catheter – and with it the biofilm – may prevent recurrences. Furthermore, biofilms can act as a reservoir for antibiotic resistance genes, and retaining the catheter combined with the selection pressure of antimicrobial therapy, may lead to the development of multidrug-resistant pathogens. Despite the theoretical basis for catheter replacement in CAUTI, empirical evidence from studies has not yet satisfactorily substantiated its efficacy. Of note, in case of catheter obstruction or malposition a replacement is always indicated, but in case of a adequate drainage, it remains unclear whether or not the catheter should be replaced.
Catheter replacement is not without burden [
13]. It causes discomfort in patients, has a risk of complications and an impact on healthcare resources [
14]. This is the case for transurethral catheters and suprapubic catheters, and the burden is even higher in JJ-catheters and nephrostomy catheters. For the latter category of patients there is not one study to assess the effect of a catheter change on symptom resolution or recurrence. Beyond the immediate health implications, catheter replacement contributes to the consumption of medical materials and has an environmental impact within the context of sustainable development goals (SDGs).
Future trials
Based on this review, it is clear that a new trial is necessary, incorporating the lessons learned from the included studies. A randomized design is required because even with propensity score matching, the risk of bias due to unmeasured confounding persists. Secondly, the patient population should be well-defined. No definition is perfect, but using the IDSA definition of CAUTI will limit misclassification and enhance comparability across studies in the CAUTI field [
1]. Patients for whom catheter replacement is non-debatable, such as those with catheter malfunction, should be excluded. The intervention should focus solely on the retention or replacement of the catheter, with antimicrobial therapy being consistent between both groups.
Outcome parameters should include both the clinical course of the initial CAUTI (for example time to resolution of symptoms) and relapse of CAUTI, as these are relevant on theoretical grounds and have shown effects in the study by Raz. Additionally, patient-related outcome measures, such as quality of life metrics, should be included. Outcome parameters should be established in collaboration with both patients and healthcare professionals. Sample size should be based on the ability to detect relevant clinical differences, as established by involving different stakeholders, including patients. Given that opinions on catheter replacement differ among specialties (e.g., urology versus microbiology), multidisciplinary collaboration in both the design and execution of the study is essential. This approach will ensure that the study results are incorporated into the guidelines of various professional associations and can find their way to clinical practice.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.