SEARCH

SEARCH BY CITATION

Keywords:

  • antimicrobial prophylaxis;
  • ileal conduit urinary diversion;
  • radical cystectomy;
  • surgical-site infection

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Objective:  To determine the optimum schedule for perioperative antimicrobial prophylaxis (AMP) for bladder cancer patients submitted to radical cystectomy with ileal conduit urinary diversion.

Methods:  We studied 77 consecutive bladder cancer patients who underwent radical cystectomy with ileal conduit. The 1-day group (n = 33) received pre-, intra- and postoperative administrations of 2 g of piperacillin on the operation day alone; the 3-day group (n = 44) received antibiotics for 3 days or more (same schedule as the 1-day group on the operation day and every 12 h thereafter). The study was designed and postoperative complications including surgical-site infection (SSI) were defined according to the modified Centers for Disease Control and Prevention criteria.

Results:  No significant differences were found between the 1-day group and 3-day group in terms of total SSI (18.1% vs 20.5%), superficial incisional SSI (12.1% vs 13.6%), deep incisional SSI (12.1% vs 13.6%), space SSI (12.1% vs 11.4%), postoperative ileus (18.2% vs 11.4%), febrile urinary tract infections (15.2% vs 15.9%) or pneumonia (3.0% vs 4.3%), respectively. In both groups, disease stage and patients' underlying conditions such as diabetes did not have an influence on the incidence of postoperative complications.

Conclusion:  One-day AMP had equivalent efficacy to that of the standard prophylaxis protocol for preventing septic complications following radical cystectomy with ileal conduit. This finding supports the hypothesis that delivery of antibiotics on the operation day is critical in this setting


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Radical cystectomy with urinary diversion is the mainstay for the treatment of muscle-invasive bladder cancer. Recent advances in surgical techniques and postoperative management have led to a significant decrease of postoperative complications.1,2 Also, prophylactic administration of antibiotics is the current standard precaution in this surgery, and is recommended in the latest clinical guidelines from various societies.3–6 Nevertheless, septic complications have been reported in between 15% and 60% of patients after elective radical cystectomy with urinary diversion using the bowel, even when appropriate antibiotics are used for several days.1,78 These unfavorable conditions bring about a significant increase in morbidity and mortality,1,8 and radical cystectomy has been one of the most troublesome procedures among general operations.9 Despite extensive studies and universal agreement on the need for application of antimicrobial prophylaxis (AMP), the optimum drug type, dose and administration period for this surgery remain unknown. Some investigators insist that pre- and intraoperative antibiotic administration is crucial for reducing the population of microorganisms in the surgical-site before wound closure, whereas advocates of postoperative administration emphasize the significance of adequate and continuous tissue concentrations of antibiotics.10

Seley was the first to report the usefulness of antibiotics for preventing surgical wound infection in 1939,11 and since his description, numerous studies have been carried out regarding this practical but challenging problem. Yet, the minimal AMP yielding adequate prophylaxis for radical cystectomy with urinary diversion remains controversial, and there has been no study that examined the role of prophylactic antimicrobial agents on the operation day in this surgery, based on consensus criteria from the infectious disease and urological societies. The present study was undertaken to assess the efficacy of prophylactic antibiotic administration on the operative day alone (1-day AMP) in elective radical cystectomy with ileal conduit urinary diversion, in comparison with ordinary AMP protocols, to gain information about the optimal AMP in contaminated urological surgery.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Between April 2002 and July 2007, 83 consecutive patients having invasive bladder cancer, who underwent elective radical cystectomy with urinary diversion using an ileal conduit, were enrolled in the study. Informed consent was obtained from all patients. Exclusion criteria were: pregnancy, penicillin allergy and contra-indications for surgery. Patients with far-advanced disease, such as those having systemic metastasis or requiring pelvic exenteration, were also excluded. The procedure for this research project was approved by a suitably constituted Ethics Committee of our institution, and it conforms to the provisions of the Declaration of Helsinki.

Clinical and pathological stages were determined according to the sixth International Union Against Cancer (UICC) classification of 2002.12 Clinical staging routinely included chest radiograph or computerized tomography (CT), abdominal and pelvic CT, isotope bone scanning, and intravenous urography. Also, all patients underwent prior transurethral resection of the bladder tumor (TURBT), and patients with muscle-invasive or high-grade recurrent cancer underwent radical cystectomy.

For standardized cystectomy for invasive bladder cancer, we amended the cystectomy protocol described previously,13 and carried out urinary diversion in accordance with the World Health Organization (WHO) consensus.14 In all operations, the procedure from skin incision to cystectomy was carried out by two surgeons, and that from urinary diversion to wound closure was done by another two. Also, all patients underwent pelvic lymphadenectomy prior to cystectomy according to a defined lymph node dissection protocol. Briefly, all connective, fatty, and lymphatic tissue was dissected between the obturator fossa and the crossing of the ureter with the common iliac artery along the external iliac vessels. In addition, the internal iliac artery and the branches thereof were freed of all connective and lymphatic tissue with the medial branches ligated (internal iliac region). All patients were treated by experienced urologists.

We referred to the guidelines for prevention of surgical-site infection (SSI) in the modified Centers for Disease Control and Prevention criteria (CDC guidelines).5 Preoperative hair removal, patient skin preparation in the operating room, and preoperative hand/forearm antisepsis of surgeons were carried out according to the recommendations. We did not control the bacterial population of the ileum before surgery by means of empiric antibiotic treatment. On the day before the operation, patients received mechanical bowel cleansing with perioral Niflec (Ajinomoto, Tokyo, Japan), which consisted of 2.93 g of sodium chloride, 1.485 g of potassium chloride 3.37 g of sodium bicarbonate and 11.37 g of anhydrous sodium sulfate, diluted in 2 L of water. During open surgery, the surgical wound was surrounded by a Steri-Drape wound protector (3M Health Care, St. Paul, MN, USA), which was attached inside the wound surface. For contaminated procedures such as transection of the ileal conduit and anastomosis thereof, we used another set of basic surgical instruments, including forceps, scissors, and scalpels. Just before wound closure, the surgical field was washed with adequate saline. The closed wound was washed with saline, and the closed site was sealed using OpSite skin-sealers (Smith & Nephew, Tokyo, Japan). The seal was left until suture removal and sutures were removed on the seventh postoperative day unless SSI was suspected. We used closed drain systems, and the drain tube was removed when the irrigation volume per day decreased to 50–120 mL without signs of infections. Ureteral stents were generally removed on the fourteenth postoperative day in ileal conduit diversion.

Perioperative antimicrobial agents were given based on the quality standard for AMP recommended by the Infectious Diseases Society of America and CDC guidelines.5,6 We used piperacillin in the present study, which is one of the most universally available antibiotics.5,6 The patients were studied in the following two treatment groups according to the AMP period: between April 2005 and July 2007, the group of 1-day patients (n = 38) received AMP on the operation day alone, and the data were prospectively accumulated; the group of 3-day patients (n = 46) treated with AMP for 3 days or longer between April 2002 and March 2005 were reviewed in their clinical record. The 1-day group patients received intravenous AMP with piperacillin 2 g just prior to skin incision, following cystectomy just before manipulation of the bowel, every 3 h during the subsequent procedure and postoperatively at 2 h. When cystectomy took 3 h or longer, patients received an additional dose every 3 h before diversion. The 3-day group patients were treated with intravenous AMP of piperacillin 2 g for 3 days or more, and the administration schedule was the same on the operation day as in the 1-day group and thereafter, with an additional equivalent dose every 12 h. Intraoperative administration intervals in the extended operations were the same as those described above for the 1-day group patients. In both treatment groups, no patient had severe renal dysfunction, and all patients received the dose as planned.

Patients were hospitalized at least until the ureteral stents were removed, and were followed in outpatient settings to evaluate all possible postoperative complications. The study was designed and SSI within 30 days after surgery was defined according to the modified CDC criteria.5 When patients showed symptoms of infection, we examined blood cell count, blood biochemistry, urinalysis, chest and abdominal radiograph and bacteriological tests. Further modalities such as CT were applied to patients with possible intrapelvic or abdominal abscess.

Statistical comparisons were made using the Welch-corrected t-test, χ2 tests and regression analysis with Prism, Version 4.02 (GraphPad software, Inc., San Diego, CA, USA) for Windows-based computers.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Of the 1-day group patients, two were withdrawn from the study because far-advanced disease was noticed intraoperatively, and two were withdrawn due to a protocol violation. Of the 3-day group patients, two patients were excluded because of an intraoperative diagnosis of far-advanced disease. Seventy-seven patients (58 men and 19 women; mean age 65.7 years [range 41–78]) comprised the final study groups, with 33 in the 1-day group and 44 in the 3-day group. The mean operative time was 348 min (range 265–440), although it included waiting time of 20–60 min for the intraoperative surgeons exchange as described in Methods. The pathological disease stage included stage I (n = 5), stage II (n = 26), stage III (n = 29) and stage IV (n = 17). In the 1-day group, unilateral hydronephrosis was observed in six patients (18.2%), and six patients and one patient of the 3-day group (in total, 15.9%) presented with unilateral and bilateral hydronephrosis, respectively. The two treatment groups were similar regarding age, sex, medical history of diabetes mellitus, hypertension and chronic obstructive pulmonary disease (COPD), preoperative serum albumin and creatinine levels, hematocrit, preoperative hospital stay and preoperative clinical stage (Table 1). Body mass index (BMI) was also uniformly distributed between the two groups. No difference in operative time was observed between the two groups. Adverse events possibly induced by antibiotics appeared in two patients in the 3-day group. One patient reported mild skin eruption, and another had pseudomembranous enterocolitis/clostridium difficile-associated diarrhea. Between the 1-day and 3-day groups, no difference was found in terms of intraoperative blood loss or intra- and postoperative blood transfusion (800 mL and 800 mL, in nine and seven patients, respectively) or in the quality of ileum cleansing noticed intraoperatively by the surgeons.

Table 1.  Comparison of demographics, analytical measurements, diagnoses and procedures between the 1-day group (receiving antibiotic prophylaxis on the operation day alone) and 3-day group (receiving it for 3 days or more)
 1-day group3-day groupP-value
  1. BMI, body mass index; COPD, chronic obstructive pulmonary disease.

Gender (male/female)23/1035/90.321
Age (years, mean)65–75 (66.9)41–78 (64.8)0.768
BMI (mean)16.7–26.3 (20.9)17.5–26.7 (21.7)0.114
Diabetes (yes/no)5/286/380.851
Hypertension (yes/no)9/249/350.484
COPD (yes/no)3/303/410.683
Serum albumin (g/L, mean)2.1–4.6 (3.11)2.1–4.1 (3.01)0.796
Serum creatinine (mg/dL, mean)0.6–2.0 (1.11)0.6–2.1 (1.17)0.468
Haematocrit (%, mean)21–46 (33.4)21–46 (32.9)0.816
Neoadjuvant chemotherapy (yes/no)4/292/420.220
Preoperative hospital stay (days, median)2–18 (3.0)2–21 (3.0)0.558
Pathological stages (n)  0.124
 I; pT1 and pN041 
 I I; pT2 and pN01115 
 III; pT3–4a and pN01415 
 IV; pT4b or pN1–3413 

Between the 1-day and 3-day groups, there were no differences in the occurrence rate of total SSI (18.1% vs 20.5%) (superficial incisional SSI [12.1% vs 13.6%], deep incisional SSI [12.1% vs 13.6%] and space SSI [12.1% vs 11.4%]), postoperative ileus (18.2% vs 11.4%), febrile urinary tract infections (15.2% vs 15.9%) or pneumonia (3.0% vs 4.3%), respectively (Table 2). All patients with space SSI had purulent drainage from a drain, but had no evident abscess on imaging. There were no patients who experienced organ SSI in both groups. Two patients of the 3-day group required a reoperation for postoperative ileus. Also, the length of postoperative hospital stay was not different between the two treatment groups (hospital stay days included 14 days of the ureteral stents indwelling period as described in Methods and days required for stoma-care instruction in all patients). There was no in-hospital mortality.

Table 2.  Comparison of postoperative complications between the two treatment groups
 1-day group3-day groupP-value
  1. All patients with space surgical-site infection (SSI) had purulent drainage from a drain, but had no evident abscess on imaging. Hospital stay days included those of the ureteral stents indwelling period (14 days) plus days required for stoma-care instruction in all patients.

SSI total (yes/no)6/279/350.803
 Superficial incisional SSI (yes/no)4/296/380.845
 Deep incisional SSI (yes/no)1/321/430.836
 Space SSI (yes/no)3/304/40>0.999
Postoperative ileus (yes/no)6/275/390.398
Urinary tract infection (yes/no)5/287/370.928
Pneumonia (yes/no)1/322/420.734
Postoperative hospital stay (days, mean)16–74 (35.0)16–110 (30.1)0.198

Microbes detected in the infected wound or abscess of most patients with SSI were typical of normal bowel flora, such as streptococci, Escherichia coli, aerobic and Bacteroides spp. Methicillin-sensitive Staphylococcus aureus (MSSa) was found in one patient of the 3-day group and two of the 1-day group, and methicillin-resistant Staphylococcus aureus (MRSa) was isolated from one 3-day group patient and two 1-day group patients.

In both groups, patient age, presence of underlying diseases such as diabetes mellitus or hypertension and preoperative laboratory data did not significantly influence the increase of the incidence of postoperative septic complications (data not shown).

We compared the drainage-tube-indwelling period and the incidence of SSI, including wound infection and intra-abdominal abscess, between localized disease (stage I and II) and disease with extravesical extension (stage III and IV). In the 1-day group, the tube indwelling period was longer in patients with advanced disease stage (15.30 ± 11.47 versus 6.78 ± 0.83, P = 0.041), but the incidence of SSI was not significantly different (Table 3). In the 3-day group, the disease stage did not affect the SSI incidence or tube removal. We could not find a contribution of any other stage-related factors, such as operative time or blood loss, to the incidence of SSI.

Table 3.  Incidence of surgical-site infection (SSI) and drainage tube indwelling period (drainage period); comparison between localized disease (Stage I and II) and disease with extravesical extension (Stage III and IV)
 Pathological stageSSI (yes/no)Drainage period (days)
  • *

    = 0.510;

  • **

    = 0.828;

  • ***

    = 0.041;

  • ****

    = 0.350.

Group 1-dayStage I and II2/13*6.78 ± 0.83***
Stage III and IV4/14*15.30 ± 11.47***
Group 3-dayStage I and II3/15**9.94 ± 4.78****
Stage III and IV5/21**11.82 ± 7.18****

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Recent Japanese guidelines for prevention of perioperative infections in the urological field defined a surgical wound classification based on the CDC criteria, and stressed that it is necessary to choose antimicrobials not for prophylaxis, but for treatment against unclean or infected operations.15 SSI in radical surgery with urinary diversion for invasive bladder cancer may depend on two biological and two technical factors. The former involve the patient's underlying condition and extent of the primary disease, and the latter are associated with the surgical technique and urinary diversion.1,13, 16 We discuss the present results concerning these factors.

In the present study, the patient's age and underlying disease such as diabetes mellitus or COPD were not associated with an increased incidence of postoperative septic complications. In our series, the fraction of patients with these underlying disorders was small, and most patients having diabetes were well-controlled for their diseasepreoperatively. This might explain the low incidence of SSI in complicated cases.

The present series included patients with locally advanced disease or lymph node metastases in each treatment group, and their fraction was thought to be higher compared with that of general studies. In addition, the number of neoadjuvant-setting cases might be small. Our institution frequently treats such patients with progressive disease, in whom prior chemotherapy is hard to carry out due to severe local symptoms or poor renal functions. It is equivocal whether the postoperative complication rate is higher in cases of progressive disease.7,13 This was also the case in our patients in both groups, although the differences were not significant. We hypothesized that SSI in advanced cases is potentially implicated in cancer-related morbidity and may also be associated with sustained intrapelvic lymphorrhea and late removal of the drainage tube. In the present 1-day group, however, the longer tube-indwelling period in patients with advanced disease did not lead to a higher incidence of SSI (Table 3). To determine the optimal antiseptic protocols for locally advanced disease, further studies are currently underway.

Radical cystectomy, still the standard in therapeutic option for muscle-invasive or highly recurrent bladder cancer, necessarily requires urinary diversion. As also seen in the present study, ileal conduit or substitution is applied most frequently after removal of the diseased bladder. A mixed population of yeast and gram-positive cocci was shown to develop in the conduit before withdrawal of antibiotic protection,17 and urinary diversion using the bowel contributes to contamination of bowel microbes, increasing the chance for SSI. The present study showed that most SSI accompanied by diversion can be prevented by AMP on the operative day. Although conduit urine is bacteriuric in most cases,16 its contribution to SSI is thought to be small. Our results may be useful also in cases requiring urinary diversion for congenital malformations, neurological disorders and incurable inflammatory disease of the urinary tract.

The present occurrence rate of SSI was lower in both treatment groups compared with those in many previous reports.6,8, 13 Although we cannot clearly explain the reason, minimal antibiotic use, which is the standard in our institution, might be associated with a lower population of antimicrobial-resistant microbes. We have been applying 1-day AMP or just preoperative administration of piperacillin, second-generation cephalosporins or fluoroquinolones as the routine prophylactic protocol in patients receiving radical nephrectomy, total nephroureterectomy, radical prostatectomy and other elective major urological surgery, and the occurrence rate of SSI was less than 1–4% in these cases (Hara N et al. 2007, unpublished data, in Niigata Cancer Center Hospital). One-day AMP has recently been applied in clean or clean-contaminated urological surgery.18,19 The next step was the challenge of contaminated urological surgery represented by radical cystectomy with urinary diversion, and it was tested in the current study. We do not consider that it matters whether 1-day AMP is the necessary minimal protocol. What is of interest is to find out the critical point when and where infection is established during or following urinary tract surgery using the bowel, and the results of the current study suggested that pre- and intraoperative delivery of antibiotics has a significant effect thereon. Yet, we also experienced many cases of SSI in the present study despite recent advances in surgical devices and techniques, and thus further studies are warranted to overcome this ubiquitous problem. Studies using wide-spectrum agents are currently underway.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

One-day AMP had an equivalent efficacy with AMP for 3 days or more for preventing postoperative septic complications in patients undergoing radical cystectomy with urinary diversion. The present study suggests that antibiotic administration on the operative day is critical for preventing such sepsis, and additional postoperative use may be selectively applied.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References