To the Editor—Urinary tract infections and prostatitis account for most postoperative infectious complications following prostate procedures, by direct inoculation of bacteria from skin or rectal mucosa into the prostate or urinary tract. Escherichia coli is the most common pathogen in 75%–90% of cases. Reference Williamson, Barrett, Rogers, Freeman, Hadway and Paterson1 Rates of sepsis from a urinary source are also rising, which may be secondary to increasing prevalence of antibiotic resistant uropathogens. Reference Halpern, Sedrakyan, Dinerman, Hsu, Mao and Hu2 Although the mechanisms leading to antibiotic resistance are complex, inappropriate utilization of antibiotics, particularly fluoroquinolones (FQs), has been shown to correlate with postprocedural infection risk due to ease of bacterial acquisition of resistance gene mutations. Reference Toner, Bolton and Lawrentschuk3 The Infectious Diseases Society of America (IDSA) now recommends avoiding empiric FQ use in genitourinary infections unless regional antibiogram data shows <20% FQ resistance to E. coli.
Previous studies have demonstrated an increasing prevalence of FQ-resistant E. coli in the rectal vault of men undergoing transrectal ultrasound-guided prostate biopsy (TRUS-PBx). Reference Cheung, Patel, Landis, Carter and Han4 Despite these trends, the most current American Urological Association (AUA) best-practice statement, last amended in 2014, still recommends FQs as primary antimicrobial agents for preoperative prophylaxis for TRUS-PBx and transurethral procedures. Reference Wolf, Bennett and Dmochowski5
Fluoroquinolone resistance to E. coli has reached 50% in some regions of the United States. Reference Bouchillon, Badal, Hoban and Hawser6 As of 2018, our institution had been using ciprofloxacin preoperatively due to a lack of data supporting alternative agents. Our institution-wide antibiogram reports ∼30% FQ resistance to E. coli, but we were unsure of the generalizability to men undergoing prostate procedures. We aimed to evaluate FQ resistance among E. coli isolates in this population at Cooper University Hospital. We assessed E. coli susceptibility as a marker for the presence or absence of antibiotic-resistant GU flora.
Methods
We utilized TheraDoc (Premier Healthcare, Charlotte, NC) to identify men ≥18 years of age who underwent a primary prostate procedure between January 2014 and December 2017 and had positive E. coli isolates from either urine or blood within the preceding 12 months. For men requiring multiple surgical procedures, only data prior to the first prostate procedure were included for analysis. Medical records were reviewed via EPIC (EPIC Systems, Verona, WI). The primary end point was the prevalence of FQ-resistant E. coli in men undergoing prostate procedures. This evaluation was approved by the institutional review board as a performance improvement project.
Results
In total, 57 men met criteria for chart evaluation. The most common procedure was radical prostatectomy (44%), followed by prostate photovaporization (23%), transurethral resection of the prostate (21%), and TRUS-PBx (12%). All patients received preoperative antibiotics and most received a single agent. Cefazolin were administered to 49% and FQs were administered to 26% of patients. Of 57 E. coli isolates, 31(53%) of 57 were FQ resistant whereas only 8(14%) of 57 were ceftriaxone resistant (Fig. 1). The 32% overall hospital rate of FQ-resistant E. coli was significantly lower than the 54% rate in our study population (P = .0010). Overall, 41 patients (72%) received FQs within 1 year of their procedure. Fluoroquinolone resistance was significantly associated with prior FQ use (P = .0091). Of 8 patients with a β-lactam allergy, 6 received an FQ (75%) compared to 14(29%) of 49 patients without allergy (P = .011).
Fig. 1. Antibiotic susceptibility pattern of E. coli isolates to common preoperative antibiotics.
Discussion
The rate of FQ-resistant E. coli was very high in our population of men undergoing invasive urologic procedures. Our E. coli FQ resistance rate of 54% was considerably higher than the 32% rate predicted from our hospital antibiogram. These findings indicate that regional or hospital-wide antibiograms may not accurately reflect resistance in a subset of male urologic patients, and they underscore the need to tailor the selection of empirical prophylactic antibiotics to accommodate patient-specific factors. Although patients with available prior positive E. coli cultures may not be entirely representative of all men undergoing prostate procedures, both are groups with high risk of prior antibiotic and FQ exposures.
We also demonstrated that prior FQ use was associated with the presence of FQ-resistant isolates. In the era of prostate-specific antigen (PSA)–based prostate cancer screening, appropriate management of elevated PSA remains an intensely debated topic in the urological literature. Benign etiologies, such as prostatitis or benign prostatic hyperplasia, can result in PSA elevations. To avoid a potentially unnecessary prostate procedure, clinicians often prescribe antibiotics to treat a possible subclinical infection to determine whether PSA levels normalize. It is estimated that up to a third of nonurologist physicians still adhere to this practice despite the paucity of scientific evidence supporting it and the potential for unfavorable sequelae including antimicrobial resistance and a higher likelihood of post–TRUS-PBx sepsis. Reference Atalay, Canat, Alkan, Cakir and Altunrende7,Reference Eggener, Large and Gerber8 We believe that our findings of a direct association between prior FQ use and subsequent resistance is significant and adds to the growing body of data supporting utilization of alternative antibiotics for empirical prophylaxis and treatment.
In conclusion, FQ resistance to E. coli is unacceptably high (53%) in this population of adult men undergoing prostate procedures. Whole-hospital antibiograms may not be reliable to predict E. coli resistance in patients undergoing urologic procedures. If preprocedure culture data are available, the choice of prophylaxis should be targeted toward the bacterial isolate and its corresponding susceptibilities. In the absence of preprocedural culture data, an alternative agent such as ceftriaxone should be considered for standard preoperative prophylaxis for transurethral or transrectal procedures. First-generation cephalosporins remain the first choice for radical prostatectomy.
Acknowledgments
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
