Antibiotic Prophylaxis After Uncomplicated Ureteroscopic Stone Treatment: Is There a Difference?

Introduction

T he management of kidney and ureteral stones has improved dramatically over the years, with endoscopic techniques such as ureteroscopy (URS) constituting a large portion of the therapeutic procedures. ¹ Symptomatic relief and reasonable stone-free rates with URS are possible, reducing morbidity and hospital stay, but symptomatic urinary tract infection (UTI) and urosepsis are potential postoperative complications. ^2,3

Infections can increase postoperative length of stay, hospital readmission rates, and the use of antimicrobial agents—all of which are associated with drastically increased cost and use of medical resources. ⁴ Conversely, the overuse of antimicrobials can lead to Clostridium difficile (C difficile) colitis, bacterial colonization, yeast infections, and future infection with resistant organisms. The American Urological Association (AUA) best practice policy statement on urologic surgery antibiotic prophylaxis recommends prophylaxis for all patients who are undergoing URS, but the duration of antibiotic prophylaxis in patients who are undergoing uncomplicated URS remains controversial. ⁵

There is compelling evidence that the use of perioperative prophylactic antibiotics significantly lowers incidence of postoperative bacteriuria when compared with no treatment, 2% vs 13%, respectively. ² Furthermore, another study suggests that patients who are undergoing URS without antibiotic prophylaxis may be at risk for febrile UTIs in up to 25% of cases, with bacteriuria rates as high as 30%. ⁶

There are no specific data on antibiotic prophylaxis with cystoscopy and stent removal after uncomplicated URS; rather, the data are extrapolated from other cystoscopy with manipulation procedures, such as transurethral resection of the prostate or bladder tumor. With an internalized device, such as a ureteral stent, however, there are possible further risk factors for development of a postoperative UTI, such as biofilm development on the stent and potential of intermittent urinary tract obstruction while passing stone fragments after stent removal.

We attempt to evaluate the risk of development of a symptomatic UTI or other complications based on the antibiotic prophylaxis given to a patient who is undergoing uncomplicated URS for the treatment of urolithiasis.

Patients and Methods

We retrospectively reviewed the medical charts of all patients who underwent ureteroscopic treatment of stones with subsequent cystoscopy and stent removal from 2004/2005 (group 1)—before the AUA best practice policy statement in 2008—and July 2009–June 2010 (group 2) (demographic information in Table 1). We excluded all patients with preoperative positive urine cultures, preoperative antibiotics, urinary diversion, who underwent concomitant percutaneous nephrolithotomy, had preoperative stent placement, or had strings attached to the ureteral stents for stent removal to keep the study population uniform. All patient data were analyzed retrospectively.

All patients in both groups received a single dose of intravenous first-generation cephalosporin or fluoroquinolone at the time of initial intervention; adjustments to our typical protocol were rarely made, based only on allergies and renal function, if needed. All patients in our study had ureteral stents placed intraoperatively after laser lithotripsy. Group 1 received an oral fluoroquinolone continuously for 1 week postoperatively, and the stent was removed within 5 to 7 days. Group 2 received an oral first-generation cephalosporin (cephalexin) peri-stent removal only, and the stent was also removed within 5 to 7 days. We define peri-stent removal antibiotics as a three-day course starting the day before stent removal.

Based on the recent AUA best practice policy statement, an oral fluoroquinolone or trimethoprim/sulfamethoxazole (TMP-SMX) is the antimicrobial of choice for the peristent removal time period, and a first-generation cephalosporin is an alternative antimicrobial. ⁵ We created our antibiotic regimen change before the release of the best practice policy and therefore it was used for the purposes of this study. This change was made based on the overuse of fluoroquinolones at our institution and the high resistance pattern seen at our institute to TMP-SMX at that time.

Our primary end point was symptomatic UTI. All patients had a urinalysis and/or urine culture at the time of stent removal. If any symptoms of UTI developed (urgency, frequency, dysuria), a urinalysis and a urine culture were repeated after stent removal. In addition, a urinalysis and urine culture were repeated for all patients at the time of their first postoperative visit approximately 4 to 6 weeks after stent removal when they underwent postoperative imaging. Other factors evaluated were baseline demographics, renal function, degree of preoperative hydronephrosis, and stone composition. Chi-square tests were used for categorical variables, and the two-tailed Student t test was used for continuous variables. All statistical analysis was performed using STATA^© version 11.

Results

There were 86 ureteroscopic interventions for stone disease in the group 1 time period and 87 in the group 2 time period. After the exclusion criteria, group 1 had 49 patients, group 2 had 48. Both groups were similar in terms of basic demographics. There was no statistical difference in the incidence of symptomatic UTI between the two groups; each group had one incident of UTI (approximately 2% risk) (P=0.988) (Table 2). There were no other adverse events, such as pyelonephritis, sepsis, surgical reintervention, C difficile, or any complication that necessitated hospital readmission. The UTI in group 1 was secondary to Escherichia coli and in group 2, Staphylococcus species; both patients were treated for a total of 7 days with an oral culture-specific antibiotic with resolution.

The UTI in group 1 occurred in a male who had no comorbidities. The one symptomatic infection in group 2 occurred in a female who had a history of breast cancer and long-standing diabetes. Both patients had stones composed of calcium oxalate. There were two patients in group 1 in whom asymptomatic funguria/bacteriuria developed (Candida/E coli) and none in group 2—the difference not being statistically significant. There were no other significant findings with regard to all other parameters that were evaluated (Tables 1 and 2).

Discussion

The use of prophylactic antibiotics in patients who are undergoing urologic procedures has clearly been shown to reduce the incidence of postoperative infectious complications. ^{7 –10} Although there have been a paucity of randomized control trials demonstrating the efficacy of perioperative antibiotics, the evidence is compelling. ^1,2,11 Available information about the exact prophylactic antibiotic regimen for urologic procedures and, specifically, uncomplicated endourologic treatment of stone disease is sparse. There is a wide assortment of recommendations from various investigators about the recommended regimen of antibiotics, the course of therapy, and route of treatment. ^3,7,12,13 Furthermore, some authors have even debated the necessity of antibiotics in patients without risk factors, sterile urine, and absence of intraoperative complications. ¹ The guidelines of the European Association of Urology and the AUA best practice policy statement for the prevention of UTIs recommend perioperative antibiotic prophylaxis for the prevention of infection related to surgical procedures at the time of cystoscopy with manipulation, which includes stent removal. ^5,14

The goal of our study was to demonstrate that the duration of antibiotic prophylaxis after uncomplicated URS does not affect the rate of symptomatic UTIs or other complications related to the surgery, additionally addressing the issue of prophylactic antibiotics at the time of cystoscopic stent extraction.

The UTI rate in our cohorts was 2% in both study groups, similar to what Knopf and associates ² obtained with antibiotic prophylaxis at the time of surgery. Their prospective study compared single dose levofloxacin vs no prophylaxis in uncomplicated URS for stone treatment. Their primary end point was bacteriuria, 1.8% vs 12.5%—the lower incidence in the group with antibiotic prophylaxis. They, however, did not mention whether they used any specific treatment around the time of stent removal. Furthermore, Takahashi and colleagues ¹⁵ compared the incidence of postoperative febrile complications in 135 patients who underwent URS and laser lithotripsy. Their retrospective study showed no statistically significant difference between the groups receiving a single dose of antibiotic vs a multiday course, 4.5% and 11.6%, respectively; however, there was a trend for increasing febrile complications when a longer course of antibiotics was used. The aforementioned study may suggest that a longer course of antibiotics may lead to higher complication rates, but one is to also note that a higher percentage of patients (7.2% to 3%) had a positive preoperative urine culture in the group who received a longer course of antibiotics.

In our study we excluded patients with positive preoperative urine cultures and other possible mitigating factors to attempt to eliminate potential confounders and specifically analyze an uncomplicated group.

There was no discernable similarity between the two patients who had symptomatic UTIs except that they both had calcium oxalate stones. An interesting observation by Hugosson and coworkers ¹⁶ showed that approximately 48% of struvite stones had the presence of bacteria, but up to 32% of calcium oxalate stones harbored bacteria. While we did not specifically culture all stones unless we anticipated infection, one can infer that asymptomatic bacteria or a bacterial biofilm may be present in a traditionally “noninfected” stone like calcium oxalate. The two cases of UTIs both had calcium oxalate stones, so there may have been bacteria harbored on or within these stones that possibly could have been a factor in the postoperative infection.

General risk factors for infection related to the patient include elderly age, nutritional deficiency, immune-compromised (diabetes mellitus, cancer), smoking, obesity, coexisting infection at remote site, or long-term urinary obstruction. ^1,5,12,14,17 Other factors that we supposed would necessitate a longer antimicrobial course are patients with urinary diversion, concomitant other urologic procedures such as percutaneous nephrolithotomy, preoperative positive urine cultures, or indwelling stent. Therefore, we excluded them from our study. In the absence of risk factors, while performing an uncomplicated URS, a single dose of antimicrobial prophylaxis at the time of the surgery and antibiotics peri-stent removal were sufficient to prevent postoperative complications in our patients.

It has been established that excessive and/or improper use of antimicrobials increase cost, especially financial, personal, and public costs. ^4,18,19 The antimicrobials themselves differ in cost. Christiano and associates ²⁰ determined that using a single prophylactic dose of oral ciprofloxacin was equally as effective as intravenous cefazolin in patients undergoing endourologic surgery, but was markedly cheaper. They estimated that it would save approximately $7300 per 100 patient treated at their institution. In addition, personal costs of antimicrobial use can vary from mild allergy to life-threatening anaphylaxis and suppression of normal flora to the development of C difficile colitis. Furthermore, the public health risk is also profound. Antibiotic usage has likely lead to the development of various resistant bacterial strains that have considerably increased morbidity, mortality, and overall cost in healthcare. ²¹ In general, the costs (financial and health) of antibiotic prophylaxis may be kept at bay by using the cheapest and safest, yet most effective agent for the shortest duration that is compatible with excellent clinical judgment and practice.

Our study has limitations. First, our study was not prospective in design; however, patients were treated identically, with the exception of the antibiotic regimen, during the two time periods, minimizing selection bias. Second, because of the low number of postoperative UTIs or complications, our study may be underpowered; however, this is the first study of its kind to specifically evaluate the short antibiotic regimen at the time of cystoscopic stent removal after uncomplicated URS. To precisely identify the true difference in UTI rates, a prospective, multi-institutional study would be necessary and is currently under way.

Moreover, some may question why we did not use TMP-SMX as part of our antibiotic prophylaxis regimen. We purposely did not use TMP-SMX because of recommendations by our institution's infectious disease department. The antibiogram released by the department regarding our hospital had shown that common nosocomial and community pathogens known to cause UTIs (eg, E coli, Klebsiella) were exceedingly resistant to the drug during the time period of our study. It is important to stress the use of antibiograms that expose local resistance patterns unique to each institution in choosing the appropriate antibiotic prophylaxis.

Finally, it would also be interesting in the future to have an additional arm to the study with no antibiotics at the time of stent removal to see if that leads to an increased occurrence of symptomatic UTI and bacteriuria. Additional studies with much larger numbers may be able to identify further risk factors related to UTI development, specifically related to use of peri-stent removal antibiotics.

Our practice has changed over the last several years based on the AUA best practice policy statement, and this study supports our clinical decision to reduce the duration of antimicrobial prophylaxis for patients who are undergoing uncomplicated URS for stone disease.

Conclusions

The use of peri-stent removal antibiotic prophylaxis is sufficient to prevent symptomatic UTIs in patients who have undergone therapeutic ureteroscopic treatment for urolithiasis. The judicious use of antibiotics in uncomplicated cases may help in lowering the incidence of resistant organisms and other complications related to the widespread use of antibiotics. Future prospective studies are warranted to corroborate our findings and identify cases in which antibiotic use can possibly be limited further.