Surgical site infection (SSI) is one of the common causes of nosocomial infections. Since the Centers for Disease Control and Prevention (CDC) proposed guidelines for antimicrobial prophylaxis (AMP) in 1999,1 a number of studies and reviews on SSI prevention and the use of AMP agents have been reported for several surgical categories. The general remarks described in the CDC guidelines awakened urologists to the importance of perioperative management for prevention of SSI, but available information was limited. In 2001, the European Association of Urology (EAU) published their guidelines for AMP in urologic surgery as a part of the management of urinary and male genital tract infections.2 These were the first recommendations corresponding to the various surgical procedures in urology. However, to date, there has not been much evidence to support the guidelines demonstrated for open as well as laparoscopic procedures in urologic surgery,3–10 whereas strategies for the prevention of febrile infection after transurethral prostatectomy (TURP)11,12 and transrectal prostate biopsy13,14 have been exhaustively investigated.
Recently, the Japanese Urological Association (JUA) published their guidelines for the prevention of perioperative infections in urology.15 The general remarks are almost similar to the previously published guidelines, but the durations for AMP in the JUA guidelines are relatively longer than those recommended in the CDC and EAU guidelines. One reason for this is that the JUA guidelines were designed for the prevention of not only SSI but also remote infection (RI), which occurs more frequently than SSI.4,8 Therefore, the administration of antimicrobial drugs is intended to achieve therapeutic effects as well as preventive ones. Another reason is that urinary tract infection (UTI) was the most common RI following urologic procedures,4,8 indicating that manipulation of the urinary tract could be the main cause of postoperative UTI. UTI could be a preventative target for AMP following urologic surgery. In this regard, the EAU guidelines recommend 1-day administration of AMP in procedures using bowel segments, they refrained from making a recommendation on whether AMP administration may be required for one or more days in procedures with continent pouches or bladder replacement.2
Thus, although the concepts of these guidelines are largely in accord with each other, their differences leave several questions that need to be answered. The main reason for this is that most parts of these guidelines were constructed by reviewing previous studies on general surgery since few well-controlled randomized studies have been reported on urologic surgery. Therefore, several controversies have arisen by applying the evidence on general surgery to urologic surgery. Assessing previous published reports, this review aims to develop more optimal guidelines in urology by clarifying agreements and differences in urologic interventions.
General recommendations for perioperative managements
In the CDC guidelines, risk factors for perioperative infections are divided into two groups; patient and environmental factors.1 Patient risk factors are advanced age, malnutrition, diabetes, smoking, obesity, infections in nonsurgical sites, immunocompromized status, and long preoperative hospital stay. Environmental risk factors are inappropriate skin preparation, preoperative hair removal, prolonged operation time, inappropriate AMP, poorly controlled operating room ventilation system, inadequate sterilization of surgical instruments, foreign body use in operation, inappropriate drain use, and immature surgical techniques. However, some of these risk factors are somewhat controversial because there is insufficient evidence to confirm them.
The CDC guidelines recommend that preoperative hair removal should be avoided, but if necessary hair should be removed immediately before an operation, preferably with electric clippers.1 This followed a report that demonstrated that preoperative shaving of the surgical site the night before an operation is associated with a significantly higher SSI risk compared with the use of depilatory agents or no hair removal (5.6% vs 0.6%).16 However, a recent meta-analysis concluded that no significant difference was found in SSI when patients were shaved or clipped 1 day before surgery or on the day of surgery; however, both clipping and depilatory creams resulted in fewer SSI than shaving with a razor.17 If it is necessary to remove hair, then it appears preferable to use clippers rather than a razor.
Preoperative eradication of Methicillin-resistant Staphylococcus aureus
Staphylococcus aureus including Methicillin-resistant Staphylococcus aureus (MRSA) is one of the most common species causing SSI, with most of these infections thought to originate from the patient's own flora.18 Some papers have demonstrated that perioperative intranasal mupirocin, a topical antibiotic ointment against Staphylococcus species, effectively prevents SSI in limited procedures,19,20 but this is not the case in all surgical procedures.19,21 A meta-analysis concluded that perioperative intranasal mupirocin appears to decrease the incidence of SSI in nongeneral surgeries such as cardiac, orthopedic, and neurosurgery; however, no reduction in the SSI rate was seen in randomized general surgery trials, suggesting that it would not be effective to reduce the SSI rate in urologic surgery.22
Increasing age as a risk factor for SSI
Although the risks for many infectious diseases increase with aging, the association between age and the risk of SSI remains unclear, with papers reporting conflicting results. A recent study that analyzed 144 485 surgical procedures reported that an increase in risk was noted until the age of 65 years, whereas the risk for SSI decreased after this age.23 In this regard, interestingly, a review article commented that older patients who are at increased risk for SSI undergo surgical procedures less frequently than do their healthy peers, ‘hardy survivors’, possibly because clinicians have judged the risk of an adverse clinical outcome to be high.24 However, mortality rate, duration of hospitalization, and costs associated with hospital care are greater for elderly than for younger patients with SSI,25 demonstrating how important the prevention of SSI is in the elderly.
Nonpharmacologic prevention of SSI
Hypothermia can cause various adverse outcomes, including morbid myocardial events, increased blood loss and the requirement for transfusion, postsurgical wound infections, and prolonged hospitalization.26 Several papers have shown that hypothermia may delay healing and predispose patients to SSI, suggesting that perioperative normothermia should be maintained unless therapeutic hypothermia is specifically indicated.27,28
Other papers have reported that supplemental perioperative oxygen (inspired fraction of 80% instead of 30%) significantly reduces postoperative nausea and vomiting, diminishes the decrease in phagocytosis and bacterial killing usually associated with anesthesia and surgery, and reduces the rate of postoperative wound infection among patients who undergo colon resection.26 Although only a few studies have addressed this method, available data suggests that supplemental perioperative oxygen may improve surgical outcome with little or no associated risk.26
Timing and duration of AMP administration
Two decades ago, antimicrobial agents were generally initiated after operations finished; however, now it is well known that antimicrobial agents should be administered 30 min before the beginning of the operation so that a bactericidal concentration of drugs is established in serum and tissue by the time an incision is made.1,2 Additional doses should be given every 3–4 h during the procedure to maintain therapeutic levels of antimicrobial agents throughout the operation and for a few hours after the incision is closed.1 Over 40 years ago, this issue was addressed and clearly detailed by Burke.29 An AMP was effective only when given during a period from the time bacteria entered the body until 3 h later in an experimental model using cutaneous inoculation with Staphylococcus aureus in guinea pigs.29 Classen et al. prospectively monitored the timing of the administration of AMP administration and analyzed occurrences of SSI in 2847 patients undergoing clean and clean-contaminated surgical procedures; showing that the SSI rate was 0.6% in patients who received AMP preoperatively, whereas it was 1.4% and 3.6% in those who received AMP perioperatively (during the 3 h after an incision) or postoperatively (more than 3 but less than 24 h after an incision), respectively.30
Duration of AMP administration is another issue that needs to be carefully addressed. The CDC guidelines recommend that AMP should not be continued for more than 24 h in clean and clean-contaminated procedures.1 A review article that analyzed 29 well-designed prospective studies of various major surgical procedures showed that there was no clear advantage in either a single-dose regimen (the definition allows for the administration of a second dose of an antimicrobial during surgery if the procedure is long) or multiple-dose regimens (the definition allows for antimicrobial agents to be given at the end of the procedure, in the recovery room or at a later time over a period of more than 24 h) in preventing SSI.31 The study recommended that a single-dose AMP regimen be implemented.31
On the other hand, a recent study analyzing 2374 clean surgeries reported that AMP did not reduce the infectious complication rate in a low-risk group, whereas it significantly reduced the rate of SSI in high-risk patients, suggesting that AMP may be necessary for high-risk but not for low-risk patients undergoing clean surgery.32
Recommendations for urologic surgery
Wound classification of urologic surgery
In the CDC guidelines, surgical wounds are classified as clean (class I), clean-contaminated (class II), contaminated (class III) and dirty/infected (class IV). Under these classifications, operative wounds, including those of the urinary tract, represent clean-contaminated procedures.1 However, this classification may not be generally available for surgical procedures in urology, which comprises various different techniques such as open surgery, laparoscopic surgery, endoscopic-instrumental surgery, and extracorporeal shock wave lithotripsy (ESWL); therefore, the respective SSI rates will not be similar.4,8 On the other hand, in the EAU guidelines, surgical procedures are classified as:
- 1Open operations including: (i) urinary tract with bowel segments; (ii) urinary tract without bowel segments; (iii) implant/prosthesis for penis and sphincter, and reconstructive genital operation; and (iv) other interventions outside of the urinary tract.
- 3Diagnostic intervention including: (i) transrectal biopsy of the prostate; and (ii) endoscopic-instrumental examinations such as urethrocystoscopy and laparoscopic procedures.2
Reviewing previous studies on general surgery, Grabe estimated the rates of infection in open urologic surgery and classified them in three categories; clean surgery (nephrectomy), clean-contaminated surgery (open urinary tract, bowel segment), and surgery for the implant of a prosthesis.33
In our recent multicenter study of 1156 open and laparoscopic operations, SSI rates were analyzed by eight surgical categories.8 As Grabe noted,33 our study demonstrated that, regardless of open or laparoscopic procedures, surgery on the upper abdomen showed low SSI rates and can be classified into clean surgery, since the SSI rates were 3.0% or less for laparoscopic nephrectomy/adrenalectomy (1.5%), open nephrectomy/adrenalectomy (0%), and other clean abdominal surgeries (3.0%).8 On the other hand, surgery including that of the lower urinary tract can be classified into clean-contaminated surgery, since the SSI rates were relatively high after genital surgery (3.1%), open/laparoscopic nephroureterectomy (8.2%), open radical prostatectomy (6.0%), and open lower urinary tract surgery (10.7%).8 Of note, the SSI rate was the highest in surgery using bowel segments (23.4%), associated with a high percentage of deep or organ/space, indicating that this procedure may be classified into contaminated surgery rather than clean-contaminated surgery8 (Table 1).
|Clean||Nephrectomy, adrenalectomy, retroperitoneal tumor dissection, lymph node dissection, etc.|
|Clean-contaminated||Nephroureterectomy, partial nephrectomy, pyeloplasty, total cystectomy without bowel segments, partial cystectomy, operations for vesicoureteral reflux, radical prostatectomy, genital surgery, etc.|
|Contaminated (using bowel segments)||Neobladder, continent pouches, bladder augmentation, ureteral replacement, etc.|
|Dirty||Open trauma of urinary tract, operation for infected kidney, etc.|
Transurethral prostatectomy has been well studied by a number of authors. Berry and Barratt analyzed 32 randomized controlled trials and concluded that AMP use in patients at low risk undergoing TURP significantly decreased bacteriuria (26% to 9.1%) and the incidence of clinical septicemia (4.4% to 0.7%), highlighting that TURP is not clean surgery.11 Another review by Qiang et al. reached similar conclusions, that AMP decreases the incidence of post-TURP complications such as bacteriuria, high fever, bacteremia and additional antibiotic treatment in patients with sterile urine (<100 000 bacteria/mL).12 Although treatment protocols of any duration with quinolones, cephalosporins, cotrimoxazole and aminoglycosides were effective, subgroup analysis of cephalosporin-based trials indicated that short treatment protocols appeared more effective than single dose protocols.10 As well, the incidence of bacteremia is significantly high in patients with chronic catheterization undergoing TURP.34 Also, in the presence of bacteriuria, the incidence of genitourinary infectious complications is 10 times as high as that without bacteriuria.35 Therefore, patients undergoing TURP should have sterile urine at operation. In a review, Grabe suggested that impaired renal function, high age, duration of the operation (>70 min), severe bleeding, and the skill of the surgeon are potential risk factors for postoperative infections.36 Taken together, although the best regimen and choice of antibiotics has not been determined, a short AMP protocol is required for patients, especially high-risk ones, undergoing TURP. The JUA guidelines recommend administration of single- or multiple-dose treatment by broad-spectrum penicillins, and first- or second-generation cephalosporins within 72 h for TURP,15 while the EAU guidelines recommend no AMP for all endoscopic-instrumental procedures including TURP in low-risk patients (Table 2).
|Procedures||EAU guidelines2||Grabe33||JUA guidelines15|
|TURBT||No||No||All patients (single dose)|
|TURP||No||All patients (<72 h)||All patients (<72 h)|
|Clean||No||No||All patients (single dose)|
|Clean-contaminated||No||Suggested (single dose)||All patients (48–72 h)|
|Contaminated (using bowel segments)||All patients (<24 h)||All patients (<24 h)||All patients (<96 h)|
|Laparoscopic procedures||No||As for open surgery||As for open surgery|
There have been just a few studies addressing other transurethral interventions such as transurethral resection of bladder tumors (TURBT). In the JUA guidelines, single-dose or 1-day prophylaxis are recommended for patients with sterile urine undergoing TURBT; however, AMP prophylaxis within 72 h should be considered for patients with lower urinary obstruction, diabetes, immunocompromized status, or residual tumors.15
Generally, penicillins with beta-lactamase inhibitor (BLI), or first- or second-generation cephalosporins are recommended as AMP agents, since these are effective against many gram-positive or gram-negative microorganisms contaminating from the skin, urinary tract, and intestines. In the EAU guidelines, third-generation cephalosporins are listed as alternative antibiotics for use in high-risk patients undergoing operations using bowel segments. However, Matsukawa et al. reported that the incidence of SSI did not increase after the regulation of AMP by the prohibition of third-generation cephalosporins.3 In our recent study, although the difference was not statistically significant, the SSI rate using third-generation cephalosporins following urinary diversion with bowel segments was much higher (42.9%) than those using other AMP agents (11.1–25.0%),8 indicating that third-generation cephalosporins should not be used as AMP agents in these procedures.
In the JUA guidelines, single-dose or 1-day AMP protocols are recommended for clean operations, while 2 or 3-day AMP is recommended for clean-contaminated operations, and 4-day AMP for bowel-utilizing operations.2 To our knowledge, in urologic open surgery there are no reports of a comprehensive randomized trial to determine the optimal duration of AMP administration, though some reports have demonstrated that shortening the duration of AMP did not increase the incidence of SSI in open or laparoscopic clean, clean-contaminated, and contaminated procedures.3,4,8,9 Our opinion is that, 2-day or longer AMP is not required in clean and clean-contaminated operations, as previously reported.5,9, 31 The EAU guidelines set out that AMP is not required in open operations without bowel segment use, but that AMP should be considered in patients with increased risk of infection such as diabetes or immunosuppressed status.2 Grabe noted that clean operations will usually not require antimicrobial prophylaxis except for those including the implantation of a prosthetic device, while clean-contaminated will benefit from preventive antimicrobials33 (Table 2). These opinions may be proven correct, but further study is still required before recommendations based on clinical evidence are established as accepted practices. A recent paper reported that no increase of SSI rate was observed in the group who did not receive AMP when compared with those who received oral or parenteral administration of AMP in minimum incision endoscopic surgery of adrenal and renal tumors, indicating that AMP could be discarded without increasing the incidence of SSI in operation with minimum incision.10
In the EAU guidelines, there is little concerning laparoscopic procedures. To date, there has not been sufficient data concerning AMP for laparoscopic surgery in urology accumulated; therefore, a consensus has not been reached on its application in spite of the common use of AMP. Our recent study showed no significant difference between the SSI rates following 227 laparoscopic and 177 open clean surgeries (1.3% vs 1.1%).8 Fahlenkamp et al. reported that the incidence of SSI was as low as 0.8% in clean laparoscopic procedures for the kidney, adrenal gland, ureter, varicocele, and lymph nodes, etc.37 Therefore, at least for now, it appears preferable that AMP is used as in open surgery (Table 2).
As laparoscopic surgery is thought less invasive, the incidence of SSI in laparoscopic procedures could be lower than in open procedures, especially clean-contaminated and contaminated ones. A review of general surgery reported that SSI rates were significantly lower with laparoscopic procedures than open ones following cholecytectomy (1.1% vs 4.0%), colorectal incision (5.0% vs 9.5%), and appendectomy (2% vs 8%).38 Rassweiler et al. reported that SSIs occurred in five of 219 (2.3%) patients undergoing open radical prostatectomy while in just one of 438 (0.3%) undergoing laparoscopic radical prostatectomy.39 Castillo et al. found a SSI in one of 59 (1.7%) laparoscopic radical cystectomy patients.40
Novitsky et al. in their review noted that the less surgical trauma induced by laparoscopic surgery may result in reduced inflammatory response and minimal immunosuppression, since postoperative levels of the inflammatory cytokines such as interleukin-6 (IL-6) have been demonstrated to be consistently lower after laparoscopic procedures.41 Mutoh et al. evaluated the incidence of systemic inflammatory response syndrome (SIRS) to compare the surgical invasiveness between laparoscopic adrenalectomy and open surgery; reporting that the mean duration of SIRS was significantly shorter in the laparoscopic (1.2 days) than in the open (1.9 days) group, indicating that laparoscopic adrenalectomy is less invasive than open surgery.42