Infection/Inflammation

Transcription

1 Infection/Inflammation Compliance with American Urological Association Guidelines for Post-Percutaneous Nephrolithotomy Antibiotics Does Not Appear to Increase Rates of Infection Sameer Deshmukh, Kevan Sternberg, Natalia Hernandez and Brian H. Eisner*, From the Department of Urology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, and Division of Urology, University of Vermont College of Medicine (KS), Burlington, Vermont Abbreviations and Acronyms AUA ¼ American Urological Association PCN ¼ percutaneous nephrostomy PCNL ¼ percutaneous nephrolithotomy POD ¼ postoperative day UTI ¼ urinary tract infection Accepted for publication April 21, * Correspondence: Department of Urology, GRB 1102, 55 Fruit St., Boston, Massachusetts (telephone: ; FAX: ; beisner@partners.org). Financial interest and/or other relationship with Boston Scientific, Bard, Cook, Olympus and Ravine Group. See Editorial on page 869. Editor s Note: This article is the third of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 1174 and Purpose: We compared infection rates after percutaneous nephrolithotomy in a group of patients without a history of infection or struvite calculi who received 24 hours or less of antibiotics postoperatively (ie compliance with AUA guidelines) vs a group that received 5 to 7 days of antibiotics postoperatively. Materials and Methods: We retrospectively reviewed the records of consecutive percutaneous nephrolithotomy procedures in patients without a history of urinary tract infection. Group 1 received 24 hours or less of antibiotics postoperatively and group 2 received a mean of 6 days of antibiotics postoperatively. Results: A total of 52 patients in group 1 (24 hours or less of antibiotics) and 30 in group 2 (mean 6 days of antibiotics) met study inclusion criteria. In 5 group 1 patients (9.6%) fever developed within 72 hours of percutaneous nephrolithotomy but none demonstrated bacteriuria or bacteremia on cultures. No patient in group 1 was treated for urinary tract infection on postoperative days 3 to 14. In 4 group 2 patients (13.3%) fever developed within 72 hours of percutaneous nephrolithotomy. A single patient showed bacteriuria (less than 10,000 cfu mixed grampositive bacteria) on culture while no patient demonstrated bacteremia. No patient in group 2 was treated for urinary tract infection on postoperative days 3 to 14. There was no difference in stone-free rates or the need for additional procedures between the 2 groups. Conclusions: In this pilot series compliance with AUA guidelines for antibiotic prophylaxis did not result in higher rates of infection than in a comparable group of 30 patients who received approximately 6 days of antibiotics postoperatively. Key Words: kidney; nephrostomy, percutaneous; anti-bacterial agents; standards; compliance THE overuse of antibiotics is an important public health issue as it contributes to antibiotic resistance. 1 The AUA Best Practice Statement (updated in 2012) cites level IIb and level III evidence in its recommendation that antibiotic prophylaxis is indicated in all patients undergoing percutaneous nephrolithotomy for a duration of 24 hours or less. 1 A single randomized study evaluated antibiotic therapy after PCNL, comparing single dose to short course antibiotics, and found no difference in the incidence of septic complications after PCNL. 2 We hypothesized that compliance with antibiotic guidelines does not result in an increase in infectious complications after PCNL. We report a retrospective study 992 j /15/ /0 THE JOURNAL OF UROLOGY Ó 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH,INC. Vol. 194, , October 2015 Printed in U.S.A.

2 COMPLIANCE WITH AUA GUIDELINES FOR POST-NEPHROLITHOTOMY ANTIBIOTICS 993 comparing 2 antibiotic regimens for PCNL, including 24 hours or less of antibiotics vs several days (mean 5.9) of postoperative antibiotics. METHODS We retrospectively reviewed the records of consecutive PCNL procedures performed at 2 institutions. Group 1 comprised patients who received 24 hours or less of antibiotics for PCNL. All of these procedures were performed at a single institution (Massachusetts General Hospital). The first dose of antibiotics was routinely given 30 to 60 minutes prior to the start of the procedure (ie flexible cystoscopy and placement of a ureteral catheter). The final dose of antibiotics in patients in this group was given less than 24 hours after the first dose. Group 2 comprised patients who received antibiotics for several days after PCNL (mean 5.9 as described). All of these procedures were performed at a single institution (University of Vermont). During the inpatient stay if a patient demonstrated a temperature of greater than 101.5F, urine and blood cultures were obtained before the initiation of empirical intravenous antibiotics. All patients in this study were seen within 14 days of discharge home and assessed for any infectious complications in the perioperative period by office urinalysis and culture if indicated. In addition, any patient who reported receiving antibiotics during or after discharge was considered to have had cystitis. Rates of infectious complications were compared using the Student t-test. Exclusion criteria were chronic indwelling catheters such as a suprapubic tube, history of infection within 60 days before PCNL based on documented urine cultures, history of recurrent cystitis or a prior struvite stone. In our practices all of these patients receive 3 to 7 days of antibiotics before PCNL. Patients underwent percutaneous access on the same day as PCNL. Access was performed by the treating urologist. If postoperative drainage consisted of a nephrostomy tube, it was removed on POD 1 or 2. If drainage was achieved with a ureteral stent, it was removed from POD 7 to 14. RESULTS Group 1d24 Hours or Less of Antibiotics (Compliance with AUA Guidelines) A total of 52 patients met study inclusion criteria. In 5 patients (9.6%) fever developed during the inpatient stay. No patients (0%) were admitted to the intensive care unit for presumed sepsis. However, none of these patients (0%) had positive urine or blood cultures, suggesting that the source of fever was likely atelectasis or another nongenitourinary source. No patients (0%) were treated for infection from PODs 3 to 14. Mean age was 51.9 years and 29% of the patients were female. The most common intraoperative antibiotics used were cefazolin in 37% of patients, ampicillin plus gentamicin in 19% and ceftriaxone in 17%. Mean SD operative time was minutes and 34 accesses (65.0%) were below the 12th rib. Stone size was greater than 2.0 cm in 79% of patients and 23% of the stones were staghorn calculi. Stone composition was pure or mixed calcium in 25 patients (46%), mixed calcium and uric acid in 10 (19%), pure uric acid in 3 (6%), struvite in 4 (8%), cystine in 2 (4%) and unknown in 8 (15%). A ureteral stent without nephrostomy served as drainage for 35% of procedures and the remainder had nephrostomy tube drainage for less than 48 hours. Tables 1 to 3 show these data. The stone-free rate was 64% and 10 patients (19%) underwent a secondary procedure for removal of residual stones. There were no significant differences between the 2 groups in Table 1. Patient characteristics Group 1 Group 2 p Value Mean age * % Male/female 71/29 47/ % Antibiotic: Intraop Cefazolin (46), ampicillin þ gentamicin (21), ceftriaxone (17), fluoroquinolone only (3.8), vancomycin/ gentamicin (3.8), vancomycin only (3.8), ampicillin/ciprofloxacin (1.9), cefazolin/ gentamicin (1.9) Ampicillin þ gentamicin (70), quinolone alone (20), vancomycin/gentamicin (6.6) Not applicable Postop Not applicable Quinolone (80), trimethoprim-sulfamethoxazole (10), cephalexin (6.6) Not applicable Mean antibiotic duration 24 Hrs 5.9 Days <0.001 Median SD operative time (mins) % Renal access location: <0.001 Below 12th rib Supracostal 35 3 % Drainage PCN only (29), PCN þ stent (36), stent only (35) PCN þ stent (100) <0.001 No. stone greater than 2.0 cm (%) 41 (79) 24 (80) 0.99* No. large staghorn calculus (%) 12 (23) 4 (13.3) 0.4* * Not significant. Greater than 50% of collecting system.

3 994 COMPLIANCE WITH AUA GUIDELINES FOR POST-NEPHROLITHOTOMY ANTIBIOTICS Table 2. Postoperative stone composition, fever and culture data Group 1 Group 2 No. pts No. stone composition (%): Pure/mixed calcium 25 (46) 22 (73) Mixed calcium/uric acid 10 (19) 2 (6.6) Pure uric acid 3 (6) 3 (10) Struvite 4 (8) 2 (6.6) Cystine 2 (4) 0 Unknown 8 (15) 1 (3.3) No. fever/total No. (%): POD 0e3 (during hospital admission) 5/52 (9.6) 4/30 (13.3) POD 3e14 (after discharge home) 0/52 1/30 (3.3) No. pos culture/total No. during fever evaluation: Urine 0/5 1/4 (25)* Blood 0/5 0/4 No. postop intensive care unit admission/total No. 0/52 0/30 * Sole positive culture showed less than 10,000 cfu/ml mixed gram-positive organisms. stone-free status or secondary procedures (chisquare test p ¼ 0.8 and 0.9, respectively). Group 2dSeveral Days of Postoperative Antibiotics A total of 30 patients met study inclusion criteria. In 4 patients (13.3%) fever developed during the inpatient stay. No patients (0%) were admitted to the intensive care unit for presumed sepsis. One of those 4 patients had a positive urine culture and no patients had positive blood cultures. One patient (3.3%) was treated for infection from PODs 3 to 14. The culture grew fewer than 10,000 cfu mixed organisms. Mean age was 55.7 years and 53% of the patients were female. The most common intraoperative antibiotics used were ampicillin plus gentamicin in 70% of patients, quinolone alone in 20% and vancomycin/gentamicin in 6.6%. Patients were treated postoperatively with antibiotics for an average of 5.9 days (163 hours, range 4 to 7 days) following PCNL. The most common postoperative antibiotic regimens were quinolone in 80% of patients, trimethoprim-sulfamethoxazole in 10% and cephalexin in 6.6%. Mean operative time was minutes and 29 accesses (97%) were below the 12th rib. Stone size was greater than 2.0 cm in 80% of patients and 13.3% of the stones were staghorn calculi. Stone composition was pure or mixed calcium in 21 patients (76.7%), mixed calcium and uric acid in 2 (6.7%), pure uric acid in 4 (13.3%), struvite in 2 (6.7%), cystine in 0 (0%) and unknown in 1 (3.3%). Ureteral stent and nephrostomy served as postoperative drainage in all patients (tables 1 to 3). The stone-free rate was 67% and 8 patients (26%) underwent a secondary procedure to remove residual stones. DISCUSSION The current AUA recommendation for antibiotic prophylaxis in all patients during PCNL regardless of the presence or absence of specific risk factors is largely based on the study by Charton et al. 3 They found that in 35% of 107 patients undergoing PCNL without antibiotic prophylaxis a UTI developed postoperatively, although a urine culture showed none preoperatively. This trial stands in comparison to a prospective, nonrandomized trial in 49 patients undergoing PCNL who received oral ciprofloxacin, intravenous ciprofloxacin or no antibiotic prophylaxis. 4 UTI developed postoperatively in 17%, 0% and 40% of patients, respectively. Similarly established evidence to guide the duration of antibiotic therapy after PCNL is sparse. While the ideal duration of therapy with prophylactic antibiotics for PCNL is unknown, the current AUA recommendation for 24 hours or less of antibiotics for PCNL is based primarily on 1 major Turkish series. 5 In that prospective comparative study in 2002 Dogan et al compared antibiotic prophylaxis with a single preoperative intravenous dose of ofloxacin (200 mg) to a second group of patients who received treatment dose therapy (400 mg per day) until eventual removal of the nephrostomy tube. They found no statistical difference between these groups with regard to bacteriuria, bacteremia, positive stone culture or incidence of postoperative fever. Table 3. Clinical data on patients with postoperative fever in groups 1 and 2 Pt No. Preop Antibiotic Postop Quinolone (No. days) Access above 12th Rib Operative Time (mins) Ureteral Stent Nephrostomy Tube Stone Composition (%) Group 1: e 1 Cefazolin No 180 No Yes Struvite (40), calcium oxalate (10), calcium phosphate (50) 2 Ampicillin/gentamicin No 80 Yes Yes Calcium oxalate (100) 3 Cefazolin Yes 109 Yes Yes Calcium oxalate (80), uric acid (20) 4 Ampicillin/gentamicin No 101 Yes No Calcium oxalate (100) 5 Cefazolin No 102 No Yes Unknown Group 2: No Yes Yes 1 Ampicillin/gentamicin Calcium phosphate (90), calcium oxalate dihydrate (10) 2 Ampicillin/gentamicin Calcium oxalate monohydrate (60), uric acid (40) 3 Quinolone Calcium oxalate monohydrate (80), calcium phosphate (20) 4 Vancomycin/gentamicin Magnesium ammonium phosphate hexahydrate (80), ammonium urate (20)

4 COMPLIANCE WITH AUA GUIDELINES FOR POST-NEPHROLITHOTOMY ANTIBIOTICS 995 More recently Seyrek et al performed a randomized, controlled trial comparing ampicillinsulbactam with cefuroxime for prophylaxis of infection following PCNL. 6 The duration of antibiotic therapy was also randomized to single dose therapy, antibiotic therapy for 12 hours after the procedure and continuous antibiotic therapy until the time of nephrostomy tube removal. No significant difference in incidence was found between any of the groups. The conclusion of this study was that single dose antibiotic therapy with either of the 2 studied antibiotics was sufficient for prophylaxis of infection during PCNL. In our pilot series of 52 patients who underwent PCNL, in whom we were compliant with AUA guidelines for antibiotic prophylaxis (ie 24 hours or less of antibiotics for PCNL), the overall rate of postoperative documented urinary tract infection was 0%. Five of 52 patients (9.6%) in group 1 were treated with additional antibiotics for postoperative fever, which developed within the first 72 hours after surgery. However, none demonstrated positive urine or blood cultures and none were admitted to the intensive care unit for observation. Given these data, we believe that the postoperative fevers were likely due to atelectasis or other nongenitourinary causes. In addition, we did not capture any infectious complications that developed in more delayed fashion (ie from PODs 3 to 14). In the comparison group 30 patients (group 2) were treated with antibiotics in a fashion that was not compliant with current AUA guidelines. These patients received antibiotics at the start of the case in the operating room and were also treated with antibiotics postoperatively for an additional 5.9 days (163 hours). In this group fever developed in 4 of 30 patients (13.3%) within the first 72 hours after surgery. None of these 4 patients had sepsis, required intensive care unit admission or had subsequent positive blood culture. In 3 patients urine cultures showed no growth while the fourth culture yielded fewer than 10,000 mixed gram-positive bacteria. Patient 4 (1 of 30 patients or 3.3%) had fever in delayed fashion (PODs 3 to14). However, notably this patient had also experienced fever during the first 72 hours postoperatively and was being treated with quinolone for a 10-day period (table 3). Previous studies have reported the incidence of sepsis causing mortality after PCNL. In 1993 O Keeffe et al identified 6 patients who died of sepsis among 700 who underwent PCNL. 7 A more recent series, the CROES (Clinical Research Office of the Endourological Society) study, examined an international cohort of more than 5,000 patients who underwent PCNL and identified only 2 who died of sepsis. 8 However, less serious complications develop far more frequently after PCNL. 9 In the CROES study postoperative fever due to noninfectious and infectious causes was the most common postoperative event, reported in 598 patients (10.5%). 8 Urinary tract infection consisting of fever and positive urine culture requiring prolonged antibiotic course was less common with fewer than 30 occurrences (0.6%) and sepsis developed in fewer than 10 patients (0.2%). 10 In other studies the rates of post-pcnl sepsis have been reported to range from 0.3% to 4.7%. 11,12 There are several limitations to our study. We acknowledge that there is selection bias in our patient group. We chose to study only patients without a history of infection and excluded patients with struvite calculi. However, we think that this is appropriate for certain reasons. We believe strongly that patients with a history of infection or struvite stones should be treated for 5 to 7 days with preoperative antibiotics based on prior studies. 13,14 We also currently treat these patients with antibiotics for several days after surgery. Further, while infection/struvite stones are often in staghorn configuration, requiring PCNL, in practice they do not make up the majority of stones for which PCNL is performed in many countries. In fact, large series have demonstrated that PCNL is performed far more commonly for calcium stones and noninfected stones than for struvite calculi. 15 We believe that the select patient group studied in this series (ie healthy patients with no history of UTI or struvite stones) are the most appropriate group of patients undergoing PCNL in terms of limiting the use of antibiotics. In patients with a history of infection or struvite stones we use a different treatment algorithm, which often includes several days of antibiotics before and after PCNL. There is also some heterogeneity regarding the antibiotics used during the study. As the study was nonrandomized, this was not a controlled variable. Notably, however, the antibiotic regimens were compliant with AUA Best Practice Guidelines and to date no single regimen suggested by the guidelines has been shown to have superior antimicrobial efficacy. An additional point is that while fever greater than 101.5F was used as a cutoff point for the initiation of antibiotics in study group 1, other systemic inflammatory response syndrome criteria were not evaluated. Finally, we acknowledge that bacteria that cause infection may be harbored in the interstices of stones of all compositions. While struvite stones are associated with urease-splitting bacteria, other types of stones (calcium containing stones and uric acid stones) may also be associated with infection. Thus, when considering an antibiotic regimen, it is important to consider the patient history for this reason and not only the stone composition.

5 996 COMPLIANCE WITH AUA GUIDELINES FOR POST-NEPHROLITHOTOMY ANTIBIOTICS CONCLUSIONS In this pilot series of PCNL in patients without a history of infection who underwent PCNL compliance with AUA guidelines for antibiotic prophylaxis did not result in higher rates of infection than in a comparable group of 30 patients who received approximately 6 days of antibiotics postoperatively. These results are encouraging and suggest that in properly selected patients responsible use of antibiotics may not increase infectious complications. Further prospective and randomized studies would be helpful to confirm our findings. REFERENCES 1. Wolf JS Jr, Bennett CJ, Dmochowski RR et al: Best practice policy statement on urologic surgery antimicrobial prophylaxis. J Urol 2008; 179: Tuzel E, Aktepe OC and Akdogan B: Prospective comparative study of two protocols of antibiotic prophylaxis in percutaneous nephrolithotomy. J Endourol 2013; 27: Charton M, Vallancien G, Veillon B et al: Urinary tract infection in percutaneous surgery for renal calculi. J Urol 1986; 135: Darenkov AF, Derevianko II, Martov AG et al: The prevention of infectious-inflammatory complications in the postoperative period in percutaneous surgical interventions in patients with urolithiasis. Urol Nefrol (Mosk) 1994; Dogan HS, Sahin A, Cetinkaya Y et al: Antibiotic prophylaxis in percutaneous nephrolithotomy: prospective study in 81 patients. J Endourol 2002; 16: Seyrek M, Binbay M, Yuruk E et al: Perioperative prophylaxis for percutaneous nephrolithotomy: randomized study concerning the drug and dosage. J Endourol 2012; 26: O Keeffe NK, Mortimer AJ, Sambrook PA et al: Severe sepsis following percutaneous or endoscopic procedures for urinary tract stones. Br J Urol 1993; 72: de la Rosette J, Assimos D, Desai M et al: The Clinical Research Office of the Endourological Society Percutaneous Nephrolithotomy Global Study: indications, complications, and outcomes in 5803 patients. J Endourol 2011; 25: Kreydin EI and Eisner BH: Risk factors for sepsis after percutaneous renal stone surgery. Nat Rev Urol 2013; 10: Labate G, Modi P, Timoney A et al: The percutaneous nephrolithotomy global study: classification of complications. J Endourol 2011; 25: Armitage JN, Irving SO, Burgess NA et al: Percutaneous nephrolithotomy in the United Kingdom: results of a prospective data registry. Eur Urol 2012; 61: Vorrakitpokatorn P, Permtongchuchai K, Raksamani EO et al: Perioperative complications and risk factors of percutaneous nephrolithotomy. J Med Assoc Thai 2006; 89: Kumar S, Bag S, Ganesamoni R et al: Risk factors for urosepsis following percutaneous nephrolithotomy: role of 1 week of nitrofurantoin in reducing the risk of urosepsis. Urol Res 2012; 40: Mariappan P, Smith G, Moussa SA et al: One week of ciprofloxacin before percutaneous nephrolithotomy significantly reduces upper tract infection and urosepsis: a prospective controlled study. BJU Int 2006; 98: Soucy F, Ko R, Duvdevani M et al: Percutaneous nephrolithotomy for staghorn calculi: a single center s experience over 15 years. J Endourol 2009; 23: 1669.