Protocol-based perioperative antimicrobial prophylaxis in urologic surgeries: Feasibility and lessons learned.

INTRODUCTION: Rational use of antibiotics and strict adherence to practice guidelines is essential to prevent antibiotic resistance. The best surgical prophylaxis protocol requires tailoring of the available guidelines in accordance to the local bacterial flora. We designed a protocol for surgical prophylaxis to check the rampant abuse of antibiotics in the department of urology and evaluated its feasibility.
MATERIALS AND METHODS: Patients admitted for elective major surgeries under a single unit of our department over a period of 5 months were included in the study. A protocol for antibiotic prophylaxis was designed based on the European Association of Urology guidelines and the local hospital antibiogram. Single-dose intravenous cefuroxime was administered to the patients undergoing clean and clean-contaminated surgeries. Extended protocols were formulated for contaminated surgeries. Postoperative course and complications were recorded. Effectiveness was defined as adherence to the protocol (without an addition or a change in antibiotic regimen) along with an uneventful postoperative course. Prospectively maintained data were analyzed using descriptive statistics.
RESULTS: Data of 277 patients were analyzed. The mean age was 48.37 ± 17.39 years and 27.1% had comorbidities. Majority of the surgeries were clean contaminated (81%), and 60.3% of the total were endoscopic. The protocol was effective in 89.5% of the patients (248/277). The failure rate was higher for the contaminated procedures (41.7%) (odds ratio - 6.43; confidence interval = 1.51-27.2, P < 0.001). Post-operative sepsis with or without shock was the commonest cause (16/29, 55.2%) of protocol failure. Fourteen out of the 16 patients who developed sepsis had undergone endourological surgeries.
CONCLUSIONS: Protocol-based perioperative antibiotic prophylaxis in urological surgeries is feasible. Similar protocols should be developed and validated at other major centers to limit the unnecessary use of antibiotics and prevent the emergence of antibiotic resistance.

INTRODUCTION

As per the global prevalence study on infections in urology 2003–2010, the incidence of multi-resistant hospital-acquired urinary tract infection (UTI) is 9.4% and the prevalence of carbapenemase-producing Enterobacteriaceae is rising.[1] The data from European and other international governmental organizations have also reported on the emerging threat of multidrug-resistant strains, particularly for the primary uropathogen, Escherichia coli.[234] A unique problem to our urological setup (tertiary care center in a developing country) is indwelling catheters and percutaneous nephrostomies which remain in-situ for long durations and get colonized by resistant bacteria and become a source of postoperative sepsis and cross infection.

Inappropriate or prolonged administration of antibiotics post surgery is the commonest reason for emergence of multidrug-resistant strains.[4] The operating surgeon, in the apprehension of anticipatory post-operative infectious complications, usually prescribes such an inappropriate regimen. Around 30%–50% of the antibiotics used in the hospitals are prescribed for post-surgical prophylaxis and almost 30%–90% of these prophylaxis regimes are inappropriate.[56] Failure to adhere to surgical prophylaxis guidelines is another reason for the emergence of antibiotic resistance.

A study previously reported from India showed that approximately 55% of the surgeons prescribed a single antibiotic for clean surgeries.[7] A combination of two or three antimicrobial agents were preferred for clean-contaminated (42.3%) and dirty (46.9%) procedures, respectively.[7] The most commonly prescribed prophylactic antibiotics (80%) are the third-generation cephalosporins (particularly ceftriaxone and cefotaxime).[57] We, prior to this study, were also prescribing a combination of a third-generation cephalosporin with aminoglycoside for a variable duration as the routine post surgical prophylaxis protocol.

There is an urgent need to reappraise the antibiotic prophylaxis protocols. To respond to the threat of antibiotic resistance, a number of health-care systems have instituted antibiotic stewardship programs. These programs promote the appropriate use of antibiotics to improve patient outcomes, check microbial resistance, and limit the spread of infections caused by multidrug-resistant organisms.[489] These programs also enforce and monitor the adherence to relevant evidence-based guidelines on antibiotic use, thereby limiting the unnecessary use of antibiotics for prolonged periods, which will reduce the health care cost significantly.

In India, due to lack of adequate information and guidelines on post-surgical antimicrobial prophylaxis, there is an unmet need to generate baseline data on the patterns of prophylactic antibiotic use. At our institute, the department of urology in association with the antibiotic stewardship committee, designed a protocol for perioperative surgical prophylaxis and assessed the feasibility of the same.

MATERIALS AND METHODS

Patients admitted for elective major urological surgeries in a single unit of the department of urology over 5 months' duration were included in this study. The study was undertaken as a part of antibiotic stewardship program running at our center. The European Association of Urology (EAU) guidelines for perioperative antibiotic prophylaxis strongly recommend obtaining a urine culture prior to all urological surgeries. In general, many antibiotics are suitable for perioperative antibacterial prophylaxis including co-trimoxazole, second-generation cephalosporins, aminopenicillins plus a beta-lactamase inhibitor, aminoglycosides, and fluoroquinolones. Knowledge of local pathogen profile, susceptibility, and virulence is mandatory to establish the local antibiotic prophylaxis guidelines.[10] For patients undergoing contaminated surgeries (positive urine culture), control of bacteriuria prior to surgery and consideration of a prolonged regimen (no specific duration) is recommended. A protocol for antibiotic prophylaxis was designed based on the EAU guidelines[10] and adapting it to local hospital antibiogram [Table 1]. Local antibiogram was developed from 758 urinary samples collected from the urology ward of our hospital over the past 6-months duration, of which 601 (79.3%) samples were sterile. E Coli was isolated in 64 samples, of which around 50% were sensitive to cefoperazone + sulbactam. abl Since the majority of the elective urological surgeries were to be performed after documenting a sterile urine culture, we considered second-generation cephalosporin (cefuroxime) as our antibiotic of choice.

Table 1

Antibiogram of Escherichia coli from urine of inpatients of urology ward over 6 months

Antibiotic	Sensitive	Resistant	Intermediate sensitive	Not done
Amikacin	47	16	1	0
Ciprofloxacin	3	58	3	0
Gentamicin	33	29	2	0
Co-trimoxazole	37	27	0	0
Nitrofurantoin	63	1	0	0
Nalidixic acid	0	64	0	0
Imipenem	30	12	1	21
Cefoperazone + sulbactam	10	17	6	31
Colistin	42	1	0	21

All surgical procedures were classified as clean, clean-contaminated and contaminated. They were also classified according to the approach, namely endoscopic, laparoscopic, robot assisted, and open. Clean, clean-contaminated, and contaminated surgeries were defined as per the EAU guidelines as follows: clean surgeries included uninfected surgical site, no entry into the urogenital tract (UT), no evidence of inflammation, and no break in sterile technique. Clean contaminated surgeries included the entry into the UT or the gastrointestinal tract (GIT) with no or little (controlled) spillage and no break in sterile technique. Contaminated surgeries included the entry into the UT and/or the GIT with spillage of the contents, presence of inflammatory tissue, presence of bacteriuria (UT), a major break in sterile technique or open, fresh accidental wounds. Single-dose 1.5-g intravenous cefuroxime was administered for clean and clean-contaminated surgeries. Extended protocols were formulated for contaminated surgeries as shown in Figure 1. Patients with risk factors such as morbid obesity, diabetes mellitus, steroid intake, malnutrition, and chronic kidney disease received antibiotics for 48h. Patients with positive pre-operative culture were treated with antibiotics as per the culture report and the indwelling tubes and catheters were changed in an attempt to make the urine sterile and were subsequently operated under antibiotic cover. This protocol was followed for all elective major urological surgeries, the postoperative course and complications were noted and the outcomes were assessed. Effectiveness of prophylaxis protocol was defined as adherence to the protocol without additional use or change of antibiotics and an uneventful postoperative course. Prospectively maintained data were analyzed using descriptive statistics. All quantitative variables were estimated using measures of central location (mean and median) and measures of dispersion (standard deviation and interquartile range). Normality of data was checked by Kolmogorov–Smirnov tests of normality. Odds ratio (OR) was calculated for clean, clean-contaminated, and contaminated procedures. Chi-square test was used to compare the effectiveness between two groups. All statistical tests were seen at two-tailed level of significance (P ≤ 0.01 and P ≤ 0.05). Statistical analysis was carried out using IBM SPSS (Statistical Package for Social Sciences) statistical version 22.0 for windows. (IBM Corp, Version 22.0. Armonk, NY: IBM Corp.)

Figure 1

Modified perioperative surgical prophylaxis protocol in urologic surgeries

RESULTS

A total of 277 patients were analyzed. The mean age of the study population was 48.37 ± 17.39 years and 27.1% had comorbidities. Majority of the surgeries were clean contaminated (80.5%), followed by clean (10.9%) and contaminated (8.6%). Regarding the approach, endoscopic surgeries comprised 60.3% of the total number followed by open, robot assisted, and laparoscopic [Table 2]. Our antibiotic prophylaxis protocol was effective in 89.5% of the patients (248/277). The cases that failed on the study protocol (n = 29) are summarized in Table 3. The failure rate was highest in the contaminated procedures group (41.7%) (OR – 6.43; confidence interval = 1.51–27.2, P < 0.001) [Table 4]. Post-operative sepsis with or without shock was the commonest reason (16/29, 55.2%) for failure. Fourteen of the 16 patients who developed sepsis underwent endoscopic surgeries (percutaneous nephrolithotomy (PNL) =8, transurethral resection of the prostate (TURP) =2, transurethral resection of bladder tumor = 2, bladder neck incision (BNI) =1, and optical internal urethrotomy = 1). Among these, six patients had a prior positive urine culture which was made sterile preoperatively and four patients had persistent positive urine culture at the time of surgery.

Table 2

Demographic data and type of surgical approach

Variable	Result
Total number of surgeries (n)	277
Mean age (years)±SD	48.37±17.39
Male/female	205/72
Presence of comorbidities (%)	75 (27.1)
Type of surgery (%)
Laparoscopic	30 (10.8)
Robot assisted	35 (12.6)
Open	45 (16.2)
Endoscopic	167 (60.3)

SD=Standard deviation

Table 3

Details of cases having failure of modified protocol

Postoperative complication	Clavien-Dindo grade	n (n=29)	Approach
Sepsis±shock	II	16	PNL=8; TURP=2; TURBT=2; OIU=1 BNI=1; LUL=1; RN=1
Persistent fever	II	4	PNL=1; LN=1; RN-1; IILND=1
Superficial surgical site infection	II	4	RC + IC=1; RARP=1; Hypospadias repair=1; HIO-1
Others	II	3	RARC + ONB=1; RC+IC=1; RANSS=1
Deep surgical site infection	III	1	RARP=1
UTI with pyelonephritis	II	1	RC+IC=1

UTI=Urinary tract infection, PNL=Percutaneous nephrolithotomy, TURP=Transurethral resection of prostate, TURBT=Transurethral resection of bladder tumor, RN=Radical nephrectomy, LUL=Laparoscopic ureterolithotomy, OIU=Optical internal urethrotomy, BNI=Bladder neck incision, LN=Laparoscopic nephrectomy, IILND=Ilioinguinal lymph node dissection, RC+IC=Radical cystectomy + ileal conduit, RARP=Robot-assisted radical prostatectomy, RARC + ONB=Robot-assisted radical cystectomy + Orthotopic neobladder, HIO=High inguinal orchidectomy, RANSS=Robot-assisted nephron-sparing surgery

Table 4

Outcome of our modified antibiotic prophylaxis protocol according to type of surgery

Type of surgery	Success per protocol	Failure per protocol
Clean (n=30)	27 (90)	3 (10)
Clean contaminated (n=223)	207 (92.8)	16 (7.2)
Contaminated (n=24)	14 (58.3)	10 (41.7)*

*P<0.001 Odds ratio 6.429 (CI 1.51-27.2). CI=Confidence interval

On further analysis of the entire cohort, 42 patients had a previous positive urine culture, of which, 30 could be made sterile by administering oral/intravenous sensitive antibiotics and changing indwelling tubes. However, 12 patients remained culture positive [Table 5]. Of these 12, four cases failed on the prophylaxis protocol, all being endoscopic surgeries (PNL = 2; TURP = 1, and BNI = 1). Of the rest eight procedures, where the antimicrobial prophylaxis was successful despite the positive culture, three were endoscopic and five were laparoscopic or open cases. Nine of the 29 failures had undergone PNL. In three of these nine patients, the postoperative urine culture revealed Enterococcus spp. sensitive only to vancomycin. All these three patients responded to addition of vancomycin to the ongoing antibiotics for the treatment of sepsis.

Table 5

Antibiotic-sensitive pattern of patients operated with positive urine culture (n=12)

Patient	Organism	Sensitive to	Protocol failure
Case 1	Pseudomonas	Amikacin, ciprofloxacin, cefoperazone-sulbactam, imipenem, colistin	Not failed
Case 2	E. coli	Nitrofurantoin, amikacin, piperacillin-tazobactam, Imipenem	Not failed
Case 3	Klebsiella	Co-trimoxazole, norfloxacin, piperacillin-tazobactam, imipenem	Failed
Case 4	Pseudomonas	Piperacillin-tazobactam, colistin	Failed
Case 5	Acinetobacter	Amikacin, norfloxacin, ofloxacin, piperacillin-tazobactam, imipenem	Not failed
Case 6	E. coli	Nitrofurantoin, amikacin	Not failed
Case 7	Enterococcus	Nitrofurantoin, gentamicin, amoxicillin, vancomycin	Failed
Case 8	Pseudomonas	Imipenem, colistin	Not failed
Case 9	E. coli	Nitrofurantoin, amikacin, piperacillin-tazobactam, imipenem	Not failed
Case 10	E. coli	Amikacin, cefoperazone-sulbactam, piperacillin-tazobactam, colistin	Not failed
Case 11	E. coli	Nitrofurantoin, amikacin	Failed
Case 12	Pseudomonas	Colistin	Not failed

E. coli=Escherichia coli

DISCUSSION

In this study, we presented the outcomes of following a specific perioperative antibiotic prophylaxis protocol in patients undergoing elective urological surgeries at a tertiary care center. The antibiotic protocol was based on the institutes antibiogram and EAU guidelines.[10] For clean and clean-contaminated surgeries, single-dose cefuroxime (1.5 g intravenous) was administered. The EAU recommends control of bacteriuria before surgery and considers “prolonged regimen” of antibiotics for contaminated surgeries.[10] However, the duration of this “prolonged regimen” is not defined. In our protocol, for contaminated surgeries (those operated on positive urine culture or when there is an anticipatory intraoperative contamination of the surgical field) and for patients with risk factors, we started an intravenous antibiotic (based on the preoperative urine culture report) 1 day prior to the planned procedure and continued the same for 48 h as the part of prolonged regimen. For patients with positive urine culture, a genuine attempt was made to make the urine culture sterile or to decrease the bacterial load by changing the indwelling catheters/stents and administering appropriate sensitive antibiotics.

For procedures such as vesicovaginal fistula repair and radical cystectomy and urinary diversion, to cover for anaerobes, intravenous ornidazole (500 mg) was added. Majority of our patients undergoing ilioinguinal lymph node dissection for carcinoma penis were malnourished and had local infection superimposed on the metastases to the inguinal nodes. Thus, oral amoxicillin and clavulanic acid was added for 5 days in the postoperative period for this procedure.

A systematic review by Bootsma et al. highlighted the lack of well-controlled trials evaluating the need of antibiotic prophylaxis for urological procedures other than for TURP and prostate biopsy.[3] We included all routine urological procedures in our study and our protocol was effective in 89.5% of the patients. A uniform prophylaxis with an earlier generation antibiotic for a finite duration is advocated in antibiotic stewardship programs. A multicentric study by Togo et al. found that a single dose antimicrobial prophylaxis was effective in preventing peri-operative infections post urological surgeries.[11] On multivariate analysis of clean-contaminated and contaminated surgeries, specific risk factor for post operative infectious complications could not be found.[11] Similarly, adherence to protocol and use of a single dose of cefuroxime was found to be feasible as prophylaxis for majority of the urological procedures in our study. This changed our practice from the use of a combination of third-generation cephalosporin (cefotaxime) and aminoglycoside (amikacin) to a single-dose of cefuroxime (second generation) in majority of our patients.

On further analyzing the protocol failure cases, we found a higher failure rate in contaminated procedures. These procedures had six times higher chance of failure as compared to clean and clean-contaminated surgeries. This failure rate (41.7%) was also higher than the reported incidence of surgical site infection (SSI) as per the historical controls and guidelines (10%–15%).[1011] As compared to other surgical specialty procedures, the urological procedures are unique in being predominantly endourological and thus are not conferred by the conventional definitions of SSI. As these procedures are performed under irrigation fluid at high pressures, an intra-luminal contamination (positive urine culture), can gain access to systemic circulation via the pyelovenous, pyelolymphatic, and pyelotubular reflux and can result in urosepsis.[12] In our study, urosepsis was the commonest cause of protocol failure (16/29). Out of these, 14 procedures were endourological. Another major contributing factor at a tertiary care center like ours with prolonged waiting list for surgery is catheters/Double J stents which are left indwelling for prolonged periods and get colonized with resistant organisms and become a potential source of bacteremia.

In our study, the patients undergoing PNL (9/29) comprised the major bulk of patients developing postoperative sepsis. The postoperative cultures were an important guide to therapy in these patients as the cultured organism changed to Enterococcus in post-operatively collected samples. Studies have showed that cultures from stone, pelvic urine, and postoperative urine can grow different organisms. The common organisms isolated following percutaneous stone extraction are Enterobacteriaceae, enterococci, and staphylococci.[13141516] This new bacterial strain, causing post operative sepsis, is harbored inside the stone and is released on fragmentation and is not picked up pre-operatively. Thus intra or postoperative cultures are a better guide for postoperative sepsis treatment, as in our study.

Benefits of antibiotic stewardship are far reaching. A large study from Europe by Cai et al. compared the data of 3529 urologic procedures (between January 2011 and December 2013) after the protocol introduction with the data of 2619 procedures (between January 2008 and December 2010) before the protocol implementation and found that the proportion of patients with symptomatic postoperative infection did not differ (180/3529 [5.1%] vs. 117/2619 [4.5%]; P = 0.27).[17] Further, they analyzed a total of 342 isolates from all the patients with symptomatic postoperative infections and found that the rate of resistance of E. coli to piperacillin/tazobactam (9.1% vs. 5.4%; P = 0.03), gentamicin (18.3% vs. 11.2%; P = 0.02), and ciprofloxacin (32.3% vs. 19.1%; P = 0.03) decreased significantly after protocol introduction. Also, the antibiotic drug costs (€76 980 vs. €36 700) and costs related to postoperative infections (€45 870 vs. €29 560) decreased following the introduction of the protocol (P < 0.001). Thus, they concluded that adherence to the EAU guidelines on antibiotic prophylaxis reduced antibiotic usage without increasing postoperative infection rate and lowered the prevalence of resistant uropathogens. A similar comparative study is underway at our center as well.

With multidrug resistance being an emerging problem, protocol-based antibiotic prophylaxis is the need of hour. Our study shows the feasibility of developing and successfully implementing such protocols. These protocols are likely to decrease the antibiotic resistance and the costs of treatment. A multicentric study or a study on larger population is required to validate the protocol-based prophylaxis in tertiary care setups in developing world.

CONCLUSIONS

Protocol-based perioperative antibiotic prophylaxis in urological surgeries is feasible with good efficacy. Higher antibiotics for a longer duration can be prescribed to patients undergoing contaminated surgeries or those with positive cultures. Similar protocols should be developed and validated at other major centers so as to limit the unnecessary use of antibiotics and prevent the emergence of antibiotic resistance.