Impact of urinary catheter on resistance patterns and clinical outcomes on complicated urinary tract infection

Introduction and hypothesis Complicated urinary tract infection (cUTI) is highly prevalent and costly for health systems. The impact of the indwelling urinary catheter on etiologic agents and clinical outcomes has been poorly studied in Latin America. Methods Cross-sectional study including patients with cUTI, with positive urine culture, treated at Hospital Universitario San Ignacio, Bogotá (Colombia) between 2017 and 2020. Clinical and microbiologic characteristics, treatments and outcomes are explored, comparing those with and without indwelling urinary catheter. Results Seven hundred thirty-five patients with non-catheter-associated cUTI (NC-cUTI) and 165 with catheter-associated cUTI (CAUTI) were included. CAUTI group had a higher proportion of recurrent UTI (18% vs 33.3%, p < 0.001), ICU requirement (2.7% vs 8.5%, p < 0.001), longer hospital stay (6 vs 10 days, p < 0.001) and > 30 days unplanned readmission rate (5.8% vs 10.3%, p < 0.001). In the same group, we found a higher frequency of Pseudomonas spp (2.6% vs 9.4%, p < 0.001), Enterococcus spp. (2.4% vs 3.3%, p = 0.016), Serratia marcescens (0.6% vs 3.3%, p < 0.001) and Citrobacter freundii (0.5% vs 5.7%, p < 0.001). It implied a higher number of patients treated with fourth-generation cephalosporins (1.4% vs 4.8%, p = 0.004), ertapenem (32.9% vs 41.8%, p = 0.027) and carbapenems associated with a second antibiotic (1.9% vs 8.5%, p < 0.001). Conclusions Patients with CAUTI have a higher frequency of resistant germs, require greater use of resources and have worse clinical outcomes than patients who do not require such devices. Measures should be strengthened to minimize its use, in both the hospital and outpatient setting.


Introduction
Complicated urinary tract infection (cUTI) corresponds to a heterogeneous entity characterized by urinary tract infection (UTI) manifestations and risk factors associated with urinary tract structural anomalies, presence of a catheter or devices in the urinary tract, and comorbidities such as diabetes, neoplasms, immune disorders or isolation of multiresistant germs [1]. The UTI represents 1.8% of US hospitalizations with costs per hospitalization near to 10,000 dollars [2,3]. Additionally, cUTI is associated with a high rate of therapeutic failure (26.6%) 4 , hospital readmission around 9% 2 and 30-day mortality of 8.7% [4]. cUTI is associated with elevated costs and high health services requirement; therefore, cUTI is a relevant entity for health systems.
Brief review Complicated urinary tract infection (cUTI) is prevalent. A single-center descriptive study evidenced that catheter-associated cUTI is associated with resistant germs and worse clinical outcomes. In Colombia, there are descriptions of microbiologic isolations in community-acquired UTI [5,6] and UTI associated with health care in users of vesical catheter [7,8]. However, in both Colombia and Latin America, there is a lack of information on the clinical manifestations, complications and resistance patterns in patients with catheter-associated UTI (CAUTI) and its differences with non-catheter-associated cUTI (NC-cUTI).
The aim of this study is to describe the clinical and microbiologic characteristics, treatments and outcomes in patients with cUTI and to compare CAUTI vs NC-cUTI, in a reference university hospital in Colombia.

Study design and participants
A cross-sectional study was carried out including patients with diagnosis of cUTI treated at the Hospital Universitario San Ignacio (HUSI) in Bogotá, Colombia, between January 2017 and May 2020. The inclusion criteria were: patients > 18 years old, urinary tract infection discharge diagnosis (ICD-10 code N10, N12, N13.6, N15.1, N15.9, N30.0, N30.8, N30.9 or N39.0), reported cUTI in clinical history, hospitalization ≥ 48 h, presence of clinical symptoms (example: dysuria, urgency, frequent urination, flank pain, positive closed fist percussion test, suprapubic pain or fever) and positive urine culture with ≥ 10 5 colony-forming units (CFU)/ml and no more than two microorganisms isolated. The urine collection method depended on the presence or absence of urinary catheter or external devices (see below). A clean-catch sample was obtained on patients without urinary catheter. For users of urinary catheters or external devices who required replacement of the catheter, a new urinary sample was obtained through the catheter. Patients with neurogenic bladder were included if there was no clinical suspicion of another infection site and accomplished other cUTI diagnosis criteria. cUTI diagnosis criteria were defined according to diagnostic criteria recommended by European Association of Urology [9]. Pregnant women, patients referred to another hospital and those who completed hospital care in a home care service were excluded. The institutional research ethics committee approved the study (FM-CIE-0174-22).
Patient information was obtained from institutional electronic medical records. Sociodemographic data are systematically collected during patient care. Clinical presentation, attention year, comorbidities, antibiotics administered, intensive care unit (ICU) stay, length of hospital stay, 30 days hospital readmission after discharge and 30-day mortality were collected using standardized formats.
CAUTI was defined as cUTI in nephrostomy, suprapubic cystostomy or indwelling catheter users (external devices). NC-cUTI was defined as cUTI not associated with external devices. Immunosuppression was defined as: HIV infection, transplant, active neoplasia or prednisolone use ≥ 10 mg/day. Culture isolates and phenotypic resistance patterns were obtained according to microbiology laboratory reports. Resistance phenotypes were defined as follows [10][11][12] Cephalosporin resistance (except cephamycin) with amoxicillin-clavulanate (AC) and carbapenem sensitivity, confirmed by laboratory. • Carbapenemase production pattern: Microbiologic isolation resistant (or diminished sensitivity) to carbapenems and positive confirmatory test (Hodge test, EDTA or boronic acid test) [13]. Although non-enzymatic resistance may be present, positive detection is defined according to carbapenems resistance (or diminished sensitivity) and negative confirmatory test. • Others: Gram-negative germs with alternative resistance patterns to those mentioned above, gram-positive germs and candida.

Discussion
In this study, we describe clinical and microbiologic characteristics, treatments and outcomes of patients with cUTI in a reference hospital in Bogota, Colombia. Our results show that patients with CAUTI presented: (1) a higher recurrent UTI rate, antibiotic use in the last 3 months and 30-day unplanned hospital readmission; (2) a higher percentage of AMPc and carbapenem resistance; (3) a higher requirement of fourth-generation cephalosporins, ertapenem and carbapenems associated with another antibiotic. Demographic characteristics of our cUTI patients are similar to those reported worldwide, although some variations exist. Descriptive studies of cUTI show median ages between 65.1 and 73 years [3,14,15] and a similar male proportion [15,16]. Other studies report sepsis or septic shock prevalence between 16% and 27% [15,16], similar to our results (20.8%). Antibiotic use in the last 3 months was similar to European reports [4]. However, we found a lower ICU stay requirement (3.8%) compared to other reports in the USA (18.6 %) [16].
Regarding comorbidities, Charlson index is different between different populations. We found that 22.6% of our patients presented a Charlson index ≥ 6 [median 4 (IQR 2-6)], similar to a study developed in the USA that reported a Charlson index ≥ 5 in 18.22% [16]. Another study developed in the USA showed an average Charlson index of 1.08 [standard deviation (SD) 1.83] [2] while in Europe they reported an average Charlson index of 2.4 (SD 2.39) [15]. This difference could be explained because our institution is a reference hospital, treating more complex and comorbid patients.
CAUTI patients had some clinical characteristics different from those reported in the international literature. In the USA, 66.39% of CAUTI patients were male [3], which is higher than the 34.5% reported in our results. In Colombia, a study carried out in two hospitals in Antioquia [8] showed 51.1% male patients. It is possible that the lower prevalence of males reported in our study is associated with HUSI condition as a cancer center reference hospital with higher requirement of catheters or external devices due to a malignant urinary tract obstruction.
Main cUTI isolation profiles are E. coli (64.2%), Klebsiella spp. (11.2%), Proteus spp. (7%) and Enterococcus spp. (3.1%). This profile is similar to international literature [1,16,17] and to that reported by the Bacterial Resistance Control Group in Bogotá (GREBO, in Spanish) in 2017 [18]. However, there are differences in the isolates found in patients with CAUTI compared to NC-cUTI. In the first group, we found a lower prevalence of E. coli and a higher number of AMPc constitutive germs (Pseudomonas spp., Enterobacter spp., Serratia marcescens and Citrobacter freundii). There was a similar result compared to the 25.8% of constitutive AMPc germs found in Europe [16] or the 22% of AMPc constitutive germ isolates found in a systematic review of patients managed in ICUs [19].
Sensitivity profile reported in patients with cUTI, regardless of the isolated germ, shows a high prevalence of ESBL germs (17.9%), without statistically significant differences between patients with CAUTI vs NC-cUTI (18.9% vs 14.5%, p = 0.111) (see Fig. 1). A study developed in 2010 with data from nine hospitals in Colombia reported a lower prevalence of ESBL laboratory confirmation, between 3.4 and 6.3% for E. coli and 3.4 to 17.2% for K. pneumoniae [20]. GREBO 2017 reports ceftriaxone resistance of E. coli and Klebsiella spp. with ESBL confirmation in 18.7% and 44.9%, respectively [18]. Our results present an intermediate resistance profile between these two studies of 26% in E. coli and 9% in Klebsiella spp. Our findings suggest there is a local increase in E. coli and Klebsiella spp. ESBL prevalence. Therefore, epidemiologic surveillance of these germs in cUTI should continue.
We found a carbapenems resistance (3.8%) higher than that reported in China (imipenem resistance of 0.5% for E. coli and 1.3% for Klebsiella spp.) [17]. A Colombian study in 2013 reported an E. coli resistance to ertapenem of 0% and Klebsiella of 6.9% [19]. GREBO 2017 [18] reported ertapenem resistance of E. coli of 1.5%, Klebsiella spp. of 23.3% and Pseudomonas spp. of 6.2%. Our results are similar to those found by the GREBO group (E. coli 0%, Klebsiella spp. 21%). Once again, these findings highlight the importance of epidemiologic surveillance.
Specifically, in CAUTI patients we found a high prevalence of carbapenem resistance (8.2%). With a higher prevalence in Klebsiella spp. (32%), although lower for Pseudomonas spp. (10%) compared with studies evaluating US ICUs (Klebsiella spp. 13% and Pseudomonas spp. 36%) [19]. Resistance profiles in CAUTI patients (AMPc and carbapenem resistance) explain the greater use of fourthgeneration cephalosporins (4.8%), ertapenem (41.8%) and carbapenems associated with other antibiotics (8.5%). Therefore, CAUTI patients should receive a closer followup and periodic reassessment of catheter indication. This is the first study in Latin America that reports the clinical characteristics, isolates and resistance profiles in cUTI according to the presence of a catheter or external devices. It increases the knowledge about the local microbiologic profile of our patients and provides important information for epidemiologic follow-up. Additionally, it is based on clinical diagnostic criteria for cUTI, thus facilitating clinical practice implementation of the results.
There are some limitations. A selection bias of patients with cUTI cannot be ruled out since the operational definition of this disease remains heterogeneous and could have been different at the moment of recording discharge diagnosis by the treating physician. However, we reviewed the recorded information and classified patients according to internationally accepted criteria. Also, this study was developed in a single center. Therefore, the external validity in other populations (especially in hospitals with less complexity) should be evaluated in the future. Lastly, we did not categorize the cUTI population according to whether it was community-based or healthcare-related, or by time since catheter or external device placement. This information was not reliably recorded in the medical records; therefore, this study is considered exploratory and will promote new studies to confirm our findings.
In conclusion, this study reports the clinical characteristics, isolates and resistance profiles in patients with cUTI. Patients with CAUTI were associated with a higher percentage of recurrent UTI, antibiotic use in the last 3 months, 30-day unplanned hospital readmission, higher prevalence of AMPc resistance profile, greater carbapenem administration, and greater fourth-generation cephalosporins and carbapenems associated with another antibiotic administration. Finally, empiric administration of ertapenem or fourthgeneration cephalosporins in patients with CAUTI should be assessed in future studies. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.