BERKELEY, CA (UroToday.com) - Urinary tract infection is a common admitting diagnosis of hospitalized patients, estimated to account for > 350 000 hospital admissions and $3.4 billion in healthcare costs each year in the United States. Rates of antimicrobial resistant among pathogens causing urinary tract infections are increasing worldwide. E. coli remains the most common uropathogen, accounting for 75-90% of UTIs. Along with other Enterobacteriaceae, E. coli can produce hydrolytic enzymes, specifically extended-spectrum β-lactamases (ESBL), which result in high rates of bacterial resistance to frequently used oral agents (e.g., trimethoprim-sulfamethoxazole, fluoroquinolones, and β-lactams). Bloodstream infection caused by ESBL-producing bacteria has been associated with delays in receipt of appropriate antibiotic therapy, prolonged hospitalization, and increased cost of care; however, the impact on outcomes specifically in UTI remains unknown.
While globally the prevalence of ESBL-producing bacteria is increasing, rates vary across different geographical regions. The Study for Monitoring Antimicrobial Resistance Trends (SMART) program analyzed trends in antibiotic resistance of E. coli among hospitalized patients with urinary tract infections from 88 international hospitals in 2009-2010 (Hoban DJ, et al. 2011. Diagn Microbial Infect Dis. 70:507-11). The overall rate of ESBL-production was 17.9%. ESBL rates were highest from the Asia/Pacific (27.7%) and Latin America (23.3%) regions, while the lowest rates were reported from North America (7.4%).
We used medical records and the institutional accounting system to determine clinical and economic differences in outcomes and costs between cases of ESBL UTI and non-ESBL UTI in adults admitted to hospital (MacVane et al. 2014. J Hosp Med. 9(4):232-8). Eighty-four patients were admitted to Hartford Hospital between September 1, 2011 and August 31, 2012 with ESBL UTI. We identified 55 patients with ESBL UTI (cases) who were matched 1:1 based on demographics and infecting pathogen with controls (non–ESBL UTI) that were included in the clinical and economic analysis.
Regarding the risk factors for urinary tract infection due to ESBL-producing bacteria, we found similarities to other studies that have reported diabetes mellitus, history of recurrent UTIs, recent antibiotics, recent hospitalization, and previous isolation of an ESBL-producing organism compared with non-ESBL UTIs. These features can be used as a tool to identify patients potentially carrying ESBL-producing bacteria on admission to hospital.
With respect to clinical outcomes, patients with ESBL UTI experienced higher failure rates of initial antibiotic regimen and time to appropriate antibiotic compared with non-ESBL UTI. Moreover, the median cost of care was greater (additional $3658) and the median length of stay prolonged for ESBL UTI (6 vs 4 days) despite similar hospital reimbursement (additional $469).
In our study population, the percentage of ESBLs was on the lower end of the range observed in the SMART study, consistent with the North American rates. Among our patients, we observed prolonged hospital stays and increased cost of care were endured by ESBL patients compared with non-ESBL UTI. Although clinical response was largely unchanged between cohorts in our analysis, ESBL UTI was associated with numerical higher rates of infection-related mortality (7.2% vs 1.8%) and 30-day UTI readmission (7.2% vs 3.6%). The lack of difference for these clinical findings may be related to the less severe nature of urinary tract infection compared with more invasive infectious processes and the relatively low incidence of ESBLs at our institution. As a result, the impact of these resistant infections may have even greater implications on outcomes and costs of care in areas with higher rates of ESBL, such as Asia and Latin America.
Unlike other studies on the impact of drug-resistant infections, we assessed the significance of ESBL production on hospital reimbursement. The cost of care for patients with ESBL UTI was ~50% greater than their non-ESBL controls. However, there was no appreciable difference in median hospital reimbursement among ESBL UTI, representing a potential loss of income if optimal treatment is not initiated on admission for this population. We believe the drivers of the prolonged length of stay and increased costs of care in the ESBL cohort are due to higher rates of inappropriate initial antibiotic therapy and corresponding initial clinical failure. Therefore, we propose these data provide rationale for the empiric use of ESBL active antibiotics (i.e., carbapenems) in certain high-risk individuals (previous history or multiple risk factors for ESBL). The application of this strategy may be easiest to justify and most valued in areas with high rates of ESBL infection.
Although the use of ESBL-active agents like the carbapenems are associated with a higher drug acquisition cost than conventional therapies such as ceftriaxone or levofloxacin, the added cost of care associated with inappropriate therapy trumps drug cost which represented less than 1% of the overall cost of care in our analysis. Importantly, while we advocate for the use of agents with activity against ESBL-producing organisms for empiric therapy in the setting of the previously identified risk factors, consideration should be given to the susceptibility of the infecting organism identified, and therapy should be carefully monitored to ensure de-escalation is performed when narrower spectrum antibiotics are suitable. Particularly during discharge planning, efforts should be made to utilize the most practical therapy, such as the selection of an oral antimicrobial (i.e., fosfomycin, nitrofurantoin, β-lactams or fluoroquinolone) when possible, based on susceptibility results.
In summary, urinary tract infection caused by ESBL-producing bacteria is an important public health concern. Using risk factors to facilitate early detection and timely appropriate antibiotics may improve patient outcomes and minimize the additional cost of these infections.
Shawn H. MacVane, PharmD, BCPSa and David P. Nicolau, PharmD, FCCP, FIDSAa, b,* as part of Beyond the Abstract on UroToday.com. This initiative offers a method of publishing for the professional urology community. Authors are given an opportunity to expand on the circumstances, limitations etc... of their research by referencing the published abstract.
aCenter for Anti-Infective Research and Development, Hartford Hosp., Hartford, CT USA
bDivision of Infectious Diseases, Hartford Hosp., Hartford, CT USA
David P. Nicolau, PharmD, FCCP, FIDSA
Center for Anti-Infective Research and Development
80 Seymour Street
Hartford, CT 06102
Tel: 860-545-3941; Fax: 860-545-3992