- Email: [email protected]
Age as an independent risk factor for surgical site infections in a large gastrointestinal surgery cohort in Japan
Journal of Hospital Infection 75 (2010) 183–187
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Age as an independent risk factor for surgical site infections in a large gastrointestinal surgery cohort in Japan M. Utsumi a, *, J. Shimizu b, f, A. Miyamoto c, f, K. Umeshita a, f, T. Kobayashi d, f, M. Monden e, f, K. Makimoto a a
Department of Health Science, Graduate School of Medicine, Osaka University, Osaka, Japan Department of Surgery, Toyonaka Municipal Hospital, Osaka, Japan c Department of Surgery, National Hospital Organization, Osaka National Hospital, Osaka, Japan d Department of Surgery, Ikeda Municipal Hospital, Osaka, Japan e Department of Surgery, Graduate School of Medicine, Osaka University, Osaka, Japan f Multicenter Clinical Study Group of Osaka, Risk Management Group, Osaka, Japan b
a r t i c l e i n f o
s u m m a r y
Article history: Received 3 August 2009 Accepted 13 January 2010 Available online 30 April 2010
A large scale survey was conducted to examine risk factors for surgical site infections (SSIs) among Japanese patients undergoing gastrointestinal surgery. The purposes of the study were: (i) to investigate age as a risk factor for SSIs in gastrointestinal surgery; and (ii) to examine the differences in risk factors for SSIs between laparoscopic cholecystectomy and open cholecystectomy. Surveillance data were prospectively collected from 20 participating hospitals in Japan between July 2003 and November 2007. SSIs were identified by use of the Centers for Disease Control and Prevention criteria. SSIs were identified in 1471 of 12 015 available cases, with an overall incidence of 12.2%. In the final logistic regression model, age was a risk factor in open cholecystectomy, gastrectomy and appendicectomy. Length of operation was a risk factor for SSIs for six surgical procedures, and wound class and drain use were also risk factors in most procedures. When comparing laparoscopic surgery against open procedure, use of silk sutures was a risk factor for SSIs in laparoscopic cholecystectomy. Drain use, wound class, operation duration, male gender and age were additional risk factors for SSIs in open cholecystectomy. In summary, patient age is a significant predictor for SSIs in some gastrointestinal procedures, although risk factors for SSIs in laparoscopic procedures appear quite different from those in open procedures. Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Gastrointestinal surgical procedures Laparoscopy Nosocomial infections Surgical wound infection
Introduction Surgical site infections (SSIs) account for approximately 15% of all nosocomial infections and occur in 10–30% of all patients undergoing gastrointestinal surgery.1,2 SSIs are an important cause of increased length of hospital stay, and directly affect the morbidity and risk of mortality in surgical patients.3,4 Epidemiological studies have identified risk factors as well as protective factors for SSIs, such as demographic factors, preoperative preparation and laparoscopic procedures.3 Earlier studies have reported an increased risk of SSIs in older patients.3 However, studies conducted in recent years have shown conflicting results. A large cohort study (144 485 patients) showed * Corresponding author. Address: 1–7 Yamada-oka, Suita, Osaka, 565-0871, Japan. Tel.: þ81 6 6879 2541; fax: þ81 6 6879 2541. E-mail address: [email protected] (M. Utsumi).
that the risk of SSIs increased linearly between 17 and 65 years of age, and that the risk decreased linearly in older groups in a multivariate analysis.5 Other studies have reported that patients aged 65 years had a higher risk of SSIs than younger patients, and some reported no association between age and the risk of SSIs.6,7 Studies on gastrointestinal surgery also showed conflicting results on the role of age in the development of SSIs. Increased age was a significant predictor of SSIs in oesophageal surgery, cholecystectomy and appendicectomy, while the lack of an association between age and the risk of SSIs has been reported for various types of gastrointestinal surgery.8–11 Discrepancies in the reported role of age in SSIs among studies can be ascribed to the heterogeneity of age groupings, types of procedure examined, and mixtures of procedures, since SSI rates are known to vary among procedures.3 One of the protective factors of SSIs is the use of laparoscopy, and Centers for Disease Control and Prevention (CDC) proposed a National Nosocomial Infection Surveillance system (NNIS) risk
0195-6701/$ – see front matter Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2010.01.021
184
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
index encompassing procedure-specific laparoscopic surgery.3 For laparoscopic surgery of the gallbladder, colon and rectum, 1 was subtracted from the NNIS risk index scores.12 In recent years, laparoscopic surgery, especially cholecystectomy, has become much more common than when this risk index was proposed. Nevertheless, the differences in the risk factors of SSIs between open and laparoscopic cholecystectomy have not been reported, mainly because the majority of the studies treated the laparoscopic procedure as a covariate rather than as a separate procedure for analysis.3,13 The purposes of the study were: (i) to investigate age as a risk factor for SSIs in gastrointestinal surgery; and (ii) to examine the differences in the risk factors for SSIs between laparoscopic cholecystectomy and open cholecystectomy in a large cohort of patients undergoing gastrointestinal surgery in Japan. Methods Two university hospitals and 18 university-affiliated hospitals in the Kansai area of Japan participated in prospective SSI surveillance between March 2003 and November 2007. The procedures studied were cholecystectomy, gastrectomy, appendicectomy, colectomy and rectal resections. Not all the participating hospitals registered for surveillance of all the procedures. Before beginning data collection, study sessions were held in order to standardise the data collection method. Data were prospectively collected by the infection control professionals in each hospital. The definition of SSIs proposed by the CDC was used, and nurses or surgeons (non-primary) in the project team determined the presence of an SSI.3 Patients were routinely followed-up at the hospital where they had surgery, and surgeons were requested to report post-discharge SSIs to the surveillance team. One author (J.S.) was in charge of data collection and held semi-annual feedback sessions at the participating hospitals. Variables collected were demographic data, the American Society of Anesthesiologists (ASA) score, dates of admission and discharge, date of operation, type of procedure, emergency status, use of laparoscopy, prophylactic antibiotics, multiple dose antibiotics, use of drains, stoma formation, silk suture use, wound class, duration of operation, and the status of SSIs. Patient ages were grouped into decades. ASA scores were aggregated into 2 and 3, and wounds were categorised as clean contaminated (class 2) and contaminated or dirty-infected (class 3 or 4). Two cut-off points were defined for operation duration. One definition used the CDC guidelines; the other was defined as the 75th centile of the distribution of the operation duration for each procedure because operation durations in Japan tend to be longer than those reported in the USA, especially for gastrectomy. 14 If a laparoscopic procedure was attempted and then progressed to open surgery, the procedure was classified as an open surgery.
Statistical analysis SPSS II for Windows was used for statistical analysis. The Student’s t-test was used for continuous variables and the c2-test was used for categorical variables to test statistical significance. Variables that reached statistical significance (P < 0.05) were entered into the logistic regression to estimate adjusted risks. In order to examine the validity of data collected from 20 institutions, SSI rates were stratified by the NNIS risk index and type of procedure. The NNIS index is a composite score consisting of ASA score, wound class, duration of surgery and application of laparoscopic procedure and has been used for risk adjustment worldwide.12 The surveillance study was part of a quality improvement project for the participating hospitals. This research was approved by the ethics committee of Osaka University. Results During the study period, 12 126 patients were registered. After excluding cases with missing data on SSI status, age and NNIS risk index, 12 015 cases were analysed. The patients’ mean age was 60.3 years (SD: 16.5), and the distribution of risk factors differed among procedures (Table I). Appendicectomy had a lower risk profile, except for wound class, than other procedures. Patients with laparoscopic cholecystectomy had better ASA scores and fewer class 3 wounds than those with open cholecystectomy. Our cut-off points for operation duration in open cholecystectomy, gastrectomy and rectal resections were longer than the cut-off points defined by the CDC. Laparoscopic procedures were used in fewer than 10% of each surgery with the exception of cholecystectomy (Table I). More than 80% of the patients received antibiotic prophylaxis, and 50% of the patients received an additional dose of antibiotics for operation duration >180 min. Silk sutures were used in approximately 50% of procedures, except for laparoscopic cholecystectomy and appendicectomy (Table I). The median length of hospital stay was 19 days (interquartile range: 11–28 days). The total number of SSIs was 1471 with an overall SSI incidence of 12.2%. The SSI incidence ranged from a low of 1.4% for laparoscopic cholecystectomy to a high of 24.2% for rectal resections (Table I). SSI incidence was stratified by the NNIS risk index and procedure (Figure 1). The risk of SSIs increased with increasing NNIS index scores in each procedure, and SSI rates for laparoscopic surgery were lower than those for open surgery. In univariate analysis, significant age differences between those with SSIs and those without were found for open cholecystectomy, gastrectomy and appendicectomy. The procedure-specific SSI incidence was then tabulated and plotted by age decade (Figure 2). For
Table I Demographics and clinical characteristics
SSI incidence (%) Age (mean, SD) Women (%) ASA classification 3 (%) Wound class 3 (%) Laparoscope use (%) Single dose antibiotics (%) Silk suture (%) Drain use (%)
Laparoscopic cholecystectomy
Open cholecystectomy
Gastrectomy
Appendicectomy
Colon resection
Rectal resection
N ¼ 2142 (17.8%)
N ¼ 641 (5.3%)
N ¼ 3493 (29.1%)
N ¼ 1298 (10.8%)
N ¼ 2939 (24.5%)
N ¼ 1502 (12.5%)
1.4 57.7, 13.6 54.2 3.4 8.3 100.0 90.0 2.4 59.2
8.0 64.6, 12.8 40.4 13.4 23.6 0.0 92.3 53.7 75.2
10.2 65.6, 11.4 32.7 9.0 4.4 3.3 97.5 53.8 89.5
12.4 36.7, 21.5 45.1 4.3 46.1 8.7 83.1 37.9 29.7
17.3 67.1, 11.9 44.7 13.2 10.4 7.5 94.0 52.1 70.6
24.2 64.6, 11.1 37.0 8.1 8.7 2.7 95.4 55.0 90.2
SSI, surgical site infection; ASA, American Society of Anesthesiologists.
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
appendicectomy and open cholecystectomy, SSI incidence rates linearly increased with each decade. For gastrectomy, SSI rates tended to increase with age, although the SSI rate for the 80–89year-old group declined (n ¼ 298). In contrast, for laparoscopic cholecystectomy, SSI incidence rates varied little by age. For colectomy and rectal resections, younger age groups had higher SSI rates than older age groups (Figure 2). For appendicectomy, the proportion of class 3 wounds increased sharply with age, from 33% for 20–29-year-olds to 75% for 70–79-year-olds. To a lesser extent, a similar trend was seen for open cholecystectomy; the proportion of class 3 wounds increased with age, from 15% for 40–49-year-olds to 35% for patients 80 years of age and older. For laparoscopic cholecystectomy, gastrectomy, colectomy and rectal resections, the proportion of class 3 wounds varied little among age groups. Multivariate analysis was conducted for each procedure and each operation time, and the results were similar. Table II displays the significant factors associated with SSIs using the adjusted operation times. Operation duration was a risk factor for SSIs in all six procedures, and wound class and drain use were also risk factors for SSIs in most procedures. Age was a risk factor in open cholecystectomy, gastrectomy, and appendicectomy (Table II). Use of silk sutures was a risk factor in laparoscopic cholecystectomy and appendicectomy. Antibiotic prophylaxis use was not a protective factor for any procedure, and administration of multiple doses of antibiotics was a risk factor for SSI in gastrectomy (Table II). Further analysis of gastrectomy with operation duration >180 min showed that patients receiving multiple doses of antibiotics had a significantly
185
greater duration of surgery than those who did not receive multiple doses (P < 0.001). The risk factors for SSIs in laparoscopic cholecystectomy were quite different from those in open procedures, with or without adjustment for relevant variables, and the only common risk factor was longer duration of surgery (Table II). In the other procedures, use of laparoscopy was a significant protective factor in gastrectomy and colon resections. Discussion This study found variations in SSI incidence by age for different gastrointestinal surgery procedures, suggesting the importance of examining procedure-specific SSI incidence. Although patient age cannot be altered, identifying it as a risk factor is important for communicating the risk to the patient and for exploring risk reduction strategies. Our study was consistent with two large studies in which increased age was associated with increased risk for SSIs in appendicectomy and cholecystectomy.9,13 Studies in which age was not a risk factor for SSIs were mostly colorectal surgery; our study supports these findings.11,15 A common characteristic of studies that failed to show age as a risk factor for SSIs is the lack of procedure-specific analysis.7,10,16 Increased age is accompanied by reduced immunity and increased comorbidity, which may account for an increased rate of SSIs in the elderly.17,18 If this is so, SSI incidence would increase in the elderly with any procedure. However, SSI rates only increased with age in three out of six procedures examined in our study. Of
60
Incidence of SSI (%)
50 40 30 20 10 0
M 0 1 2
*
*
3
0-Y0-N 1 2 3
CHOL
GAST
0-Y 0-N 1 2
*
*
3
M 0 1 2 3
APPY
* M 0
COLN
Colectomy,
Category = NNIS risk index – 1 with laparoscope use.
cholecystectomy,
M means NNIS risk index = 0 and laparoscope use.
* 1
2
3
REC
rectum resection Gastrectomy,
0-Y: NNIS risk index = 0 and laparoscope use.
appendectomy
0-N: NNIS risk index = 0 and no laparoscope use. No other adjustment.
Figure 1. Surgical site infection rates by selected operative procedure and modified risk index category incorporating laparoscope use.12 *Sample N < 50; y0% (0/67). CHOL, cholecystectomy; GAST, gastrectomy; APPY, appendicectomy; COLN, colectomy; REC, rectum resection.
186
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
35.0 30.0
SSI (%)
25.0
LapaCHOL OpenCHOL
20.0
GAST 15.0
APPY COLN
10.0
REC 5.0 0.0
<20
20-29
30-39
40-49 50-59 Age (years)
60-69
70-79
80-89
Figure 2. Incidence of surgical site infection (SSI) by age group and type of surgical procedure. Samples of N 50 are recorded. Lapa, laparoscopy; CHOL, cholecystectomy; GAST, gastrectomy; APPY, appendicectomy; COLN, colectomy; REC, rectum resection.
these, appendicectomy and open cholecystectomy share inflammation as a common condition, and the proportion of class 3 wounds increased with age in both procedures. Even after adjustment, age remained an independent risk factor for these two procedures. Inflammation in older patients may be an indication of reduced local defence against infection and/or increased co-infections in other sites, raising the risk of SSIs. However, in laparoscopic cholecystectomy, patient age was not a risk factor for SSIs, and these hypotheses do not explain the difference in risk factors between open and laparoscopic Table II Logistic regression analysis of surgical site infection risk factors Surgical procedure Laparoscopic cholecystectomy
Variable
Adjusted OR
95% CI
P-value
Silk suture Operation duration
4.59 4.17
1.50–14.03 0.008 1.97–8.80 <0.001
Open cholecystectomy Drain use Wound class 3 Operation duration Male Age
5.24 2.61 2.14 2.10 1.37
1.22–22.45 1.38–4.93 1.13–4.05 1.03–4.25 1.05–1.79
Gastrectomy
Wound class 3 Operation duration Multi-dose antibiotics Age Laparoscope use
4.44 2.33 1.77 1.15 0.33
2.99–6.81 1.82–2.98 1.36–2.29 1.04–1.27 0.13–0.84
<0.001 <0.001 <0.001 0.008 0.002
Appendicectomy
Wound class 3 Drain use Operation duration Silk suture Age
3.62 3.21 1.93 1.65 1.17
2.15–6.10 2.02–5.13 1.30–2.87 1.13–2.41 1.07–1.28
<0.001 <0.001 0.001 0.009 0.001
Colon resection
Wound class 3 Drain use Stoma Operation duration Laparoscope use
3.49 2.13 2.04 1.33 0.54
2.61–4.66 1.60–2.82 1.46–2.85 1.05–1.67 0.32–0.91
<0.001 <0.001 <0.001 0.016 0.021
Rectal resection
Wound class 3 Drain use Stoma Operation duration Male
3.51 3.12 2.26 1.34 1.42
2.34–5.26 1.67–5.84 1.70–2.99 1.00–1.80 1.08–1.87
<0.001 <0.001 <0.001 0.048 0.013
OR, odds ratio; CI, confidence interval.
0.026 0.003 0.019 0.040 0.021
cholecystectomy. Although laparoscopy is now used in nearly 80% of cholecystectomy, expanding the application of laparoscopic procedures may reduce the overall SSI incidence further in cholecystectomy. Since the risk factors for SSIs following laparoscopic cholecystectomy are quite different from those in open procedures, risk estimates for laparoscopic procedures should be separated from open procedures in order to obtain accurate risk assessments, if the sample size permits. For gastrectomy, the association between advanced age and the risk of SSIs cannot be explained by the presence of infection, as the proportion of class 3 wounds did not vary by age. Stomach cancer incidence in Japan is much higher than in western countries where SSI incidence rates for gastrectomy with stomach cancer have not been reported. Further research is necessary to explore differences between younger and older patients for gastrectomy. In contrast to these procedures, younger age groups had higher SSI rates than older groups in colectomy and rectal resection. Younger patients might have undergone more complicated surgery with greater risk of SSIs. Differences in indication for surgery between younger and older patients need to be compared in future research. Drain use increased the risk of SSIs 2–5-fold in open gastrointestinal surgery with the exception of gastrectomy, and our study concurs with previous studies of open gastrointestinal surgery.19 A high prevalence of drain use is not unique to our study as others reported 70–80% of drain use in gastrointestinal surgery in Japan and in Taiwan.16,20 All the articles reviewing prophylactic drainage showed the lack of benefit or increased likelihood of negative outcomes in colorectal surgery, open cholecystectomy and appendicectomy.19,21,22 In a meta-analysis of prophylactic drainage in gastrointestinal surgery, an indication of prophylactic drainage use was limited to oesophageal resection and total gastrectomy with an evidence level of ‘D’ – namely expert opinion.22 Concerted efforts are necessary to reduce drain use in order to lower SSI rates as well as other postoperative complications. The use of silk sutures was a predictor for SSIs in laparoscopic cholecystectomy and appendicectomy. Silk sutures are nonabsorbable, multifilament suture materials, both of which are risk factors for SSIs.23 In Japan, silk is still commonly used for intraabdominal suturing and ligation because it is less expensive and
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
easier to handle than absorptive synthetic suture material.16 A recent retrospective cohort study of hepatectomy patients reported an adjusted SSI odds ratio of 3.4 for silk suture use when compared with absorptive materials.23 Silk suture use has been decreasing in Japan, which would contribute to the decline in SSI rates. Although the lack of benefit from antimicrobial prophylaxis was unexpected, a few other studies have also reported a lack of benefit.11,13 Effectiveness of prophylactic antibiotics varies among gastrointestinal surgery. A meta-analysis of laparoscopic cholecystectomy showed that prophylactic antibiotics were ineffective in preventing postsurgical complications, while antibiotic prophylaxis was shown to reduce SSIs in colectomy and appendicectomy.24–26 In colon surgery, all randomised clinical trials for antibiotic prophylactic evaluation were conducted in the 1980s, and placebo groups had SSI rates of approximately 40%, nearly twice as high as the rate in our corresponding group.26 Advances in pre- and intraoperative management in the past 30 years may have contributed to the risk reductions. The changes include abandoning preoperative shaving and prevention of hypothermia, etc. Overall reductions in SSIs in colectomy may make it difficult to demonstrate an effect of antibiotic prophylaxis. An additional antibiotic dose was a risk factor in gastrectomy. As the patients with multiple doses of antibiotics had longer operation duration than those without, it may reflect an unadjusted risk. Nevertheless, timing, dosage, and type of antibiotic use in prophylaxis need to be examined to monitor proper use of antibiotics. One of the limitations of the study was the lack of clinical indicators that were considered to be risk factors for SSIs, such as history of smoking and blood transfusion.27,28 However, these indicators are unlikely to affect elderly patients going through only appendicectomy, open cholecystectomy, or gastrectomy. Strong support for our research comes from its size. Twenty hospitals participated in the study using standard protocols, providing a large body of data, which enabled us to demonstrate the importance of procedure-specific SSIs in examining risk factors. SSI rates by NNIS index are in accordance with western studies, demonstrating the validity of the data collected by multiple institutions. In addition, postoperative hospital stays in Japan are much longer than in the UK because fee-for-service was still in place at the time of the study, which made it easier to establish the presence of SSIs. However, a prospective payment system is being introduced in acute care hospitals in Japan, and post-discharge surveillance will become more important in the near future. The proportion of elderly among surgical patients has been increasing due to population ageing. Patient age is a significant predictor of SSIs in some gastrointestinal procedures. Accurate risk estimates would contribute to the prevention, early detection and initiation of treatment in high risk groups. The current study also suggests the need for separate risk assessments for laparoscopic procedures and open procedures. Acknowledgements The authors wish to acknowledge the nurses and physicians who cooperated in this project. Special thanks go to NTT West Osaka Hospital, Osaka Senin Hoken Hospital, Osaka Medical Center for Cancer and Cardiovascular Diseases, Kanou Hospital, Kansai Rosai Hospital, Kinki Daigaku Igakubu Nara Hospital, Kinki Central Hospital Aid Association of Public School Teachers, Kobe Ekisaikai Hospital, National Hospital Organization, Osaka National Hospital, Shitennouji Hospital, Ikeda Municipal Hospital, Itami City Hospital, Kawanishi City Hospital, Sakai Municipal Hospital, Toyonaka Municipal Hospital, Suita Municipal Hospital, Higashiosaka City General Hospital, Fujimoto Hospital, Minoh City Hospital, and Yao Municipal Hospital.
187
Conflict of interest statement None declared. Funding sources None. References 1. Smyth ET, McIlvenny G, Enstone JE, et al. Four country healthcare associated infection prevalence survey 2006: overview of the results. J Hosp Infect 2008;69:230–248. 2. Smith RL, Bohl JK, McElearney ST, et al. Wound infection after elective colorectal resection. Ann Surg 2004;239:599–605. 3. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guidelines for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999;20:250–278. 4. Kirkland KB, Briggs JP, Trivette SL, Wilkinson WE, Sexton DJ. The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol 1999;20:725–730. 5. Kaye KS, Schmit K, Pieper C, et al. The effect of increasing age on the risk of surgical site infection. J Infect Dis 2005;191:1056–1062. 6. Razavi SM, Ibrahimpoor M, Sabouri Kashani A, Jafarian A. Abdominal surgical site infections: incidence and risk factors at an Iranian teaching hospital. BMC Surg 2005;5:2. 7. Petrosillo N, Drapeau CM, Nicastri E, Martini L, Ippolito G, Moro ML. Surgical site infections in Italian hospitals: a prospective multicenter study. BMC Infect Dis 2008;8:34. 8. Breigeiron R, de Souza HP, Sidou JP. Risk factors for surgical site infection after surgery for esophageal perforation. Dis Esophagus 2008;21:266–271. 9. Rotermann M. Infection after cholecystectomy, hysterectomy or appendectomy. Health Rep 2004;15:11–23. 10. de Oliveira AC, Ciosak SI, Ferraz EM, Grinbaum RS. Surgical site infection in patients submitted to digestive surgery: risk prediction and the NNIS risk index. Am J Infect Control 2006;34:201–207. 11. Walz JM, Paterson CA, Seligowski JM, Heard SO. Surgical site infection following bowel surgery: a retrospective analysis of 1446 patients. Arch Surg 2006;141:1014–1018. 12. National Nosocomial Infections Surveillance (NNIS) System. NNIS Report, Data Summary from January 1990–May 1999. Am J Infect Control 1999;27:520–532. 13. Romy S, Eisenring MC, Bettschart V, Petignat C, Francioli P, Troillet N. Laparoscope use and surgical site infections in digestive surgery. Ann Surg 2008;247:627–632. 14. Imai E, Ueda M, Kanao K, et al. Surgical site infection risk factors identified by multivariate analysis for patient undergoing laparoscopic, open colon, and gastric surgery. Am J Infect Control 2008;36:727–731. 15. Tang R, Chen HH, Wang YL, et al. Risk factors for surgical site infection after elective resection of the colon and rectum: a single-center prospective study of 2,809 consecutive patients. Ann Surg 2001;234:181–189. 16. Watanabe A, Kohnoe S, Shimabukuro R, et al. Risk factors associated with surgical site infection in upper and lower gastrointestinal surgery. Surg Today 2008;38:404–412. 17. Schwab R, Walters CA, Weksler ME. Host defense mechanisms and aging. Semin Oncol 1989;16:20–27. 18. Gardner ID. The effect of aging on susceptibility to infection. Rev Infect Dis 1980;2:801–810. 19. Gurusamy KS, Samraj K. Routine abdominal drainage for uncomplicated open cholecystectomy. Cochrane Database Syst Rev 2007;2. CD006003. 20. Yeh CY, Changchien CR, Wang JY, et al. Pelvic drainage and other risk factors for leakage after elective anterior resection in rectal cancer patients: a prospective study of 978 patients. Ann Surg 2005;241:9–13. 21. Jesus EC, Karliczek A, Matos D, Castro AA, Atallah AN. Prophylactic anastomotic drainage for colorectal surgery. Cochrane Database Syst Rev 2004;4. :CD002100. 22. Petrowsky H, Demartines N, Rousson V, Clavien PA. Evidence-based value of prophylactic drainage in gastrointestinal surgery: a systematic review and meta-analysis. Ann Surg 2004;240:1074–1084. 23. Togo S, Kubota T, Takahashi T, et al. Usefulness of absorbable sutures in preventing surgical site infection in hepatectomy. J Gastrointest Surg 2008;12: 1041–1046. 24. Choudhary A, Bechtold ML, Puli SR, Othman MO, Roy PK. Role of prophylactic antibiotics in laparoscopic cholecystectomy: a meta-analysis. J Gastrointest Surg 2008;12:1847–1853. 25. Andersen BR, Kallehave FL, Andersen HK. Antibiotics versus placebo for prevention of postoperative infection after appendicectomy. Cochrane Database Syst Rev 2005;20. CD001439. 26. Bratzler DW, Houck PM. Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical Infection Prevention Project. Clin Infect Dis 2004;38:1706–1715. 27. Sorensen LT, Horby J, Friis E, Pilsgaard B, Jorgensen T. Smoking as a risk factor for wound healing and infection in breast cancer surgery. Eur J Surg Oncol 2002;28:815–820. 28. Leal-Noval SR, Rincon-Ferrari MD, Garcia-Curiel A, et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest 2001;119:1461–1468.
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Age as an independent risk factor for surgical site infections in a large gastrointestinal surgery cohort in Japan M. Utsumi a, *, J. Shimizu b, f, A. Miyamoto c, f, K. Umeshita a, f, T. Kobayashi d, f, M. Monden e, f, K. Makimoto a a
Department of Health Science, Graduate School of Medicine, Osaka University, Osaka, Japan Department of Surgery, Toyonaka Municipal Hospital, Osaka, Japan c Department of Surgery, National Hospital Organization, Osaka National Hospital, Osaka, Japan d Department of Surgery, Ikeda Municipal Hospital, Osaka, Japan e Department of Surgery, Graduate School of Medicine, Osaka University, Osaka, Japan f Multicenter Clinical Study Group of Osaka, Risk Management Group, Osaka, Japan b
a r t i c l e i n f o
s u m m a r y
Article history: Received 3 August 2009 Accepted 13 January 2010 Available online 30 April 2010
A large scale survey was conducted to examine risk factors for surgical site infections (SSIs) among Japanese patients undergoing gastrointestinal surgery. The purposes of the study were: (i) to investigate age as a risk factor for SSIs in gastrointestinal surgery; and (ii) to examine the differences in risk factors for SSIs between laparoscopic cholecystectomy and open cholecystectomy. Surveillance data were prospectively collected from 20 participating hospitals in Japan between July 2003 and November 2007. SSIs were identified by use of the Centers for Disease Control and Prevention criteria. SSIs were identified in 1471 of 12 015 available cases, with an overall incidence of 12.2%. In the final logistic regression model, age was a risk factor in open cholecystectomy, gastrectomy and appendicectomy. Length of operation was a risk factor for SSIs for six surgical procedures, and wound class and drain use were also risk factors in most procedures. When comparing laparoscopic surgery against open procedure, use of silk sutures was a risk factor for SSIs in laparoscopic cholecystectomy. Drain use, wound class, operation duration, male gender and age were additional risk factors for SSIs in open cholecystectomy. In summary, patient age is a significant predictor for SSIs in some gastrointestinal procedures, although risk factors for SSIs in laparoscopic procedures appear quite different from those in open procedures. Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Gastrointestinal surgical procedures Laparoscopy Nosocomial infections Surgical wound infection
Introduction Surgical site infections (SSIs) account for approximately 15% of all nosocomial infections and occur in 10–30% of all patients undergoing gastrointestinal surgery.1,2 SSIs are an important cause of increased length of hospital stay, and directly affect the morbidity and risk of mortality in surgical patients.3,4 Epidemiological studies have identified risk factors as well as protective factors for SSIs, such as demographic factors, preoperative preparation and laparoscopic procedures.3 Earlier studies have reported an increased risk of SSIs in older patients.3 However, studies conducted in recent years have shown conflicting results. A large cohort study (144 485 patients) showed * Corresponding author. Address: 1–7 Yamada-oka, Suita, Osaka, 565-0871, Japan. Tel.: þ81 6 6879 2541; fax: þ81 6 6879 2541. E-mail address: [email protected] (M. Utsumi).
that the risk of SSIs increased linearly between 17 and 65 years of age, and that the risk decreased linearly in older groups in a multivariate analysis.5 Other studies have reported that patients aged 65 years had a higher risk of SSIs than younger patients, and some reported no association between age and the risk of SSIs.6,7 Studies on gastrointestinal surgery also showed conflicting results on the role of age in the development of SSIs. Increased age was a significant predictor of SSIs in oesophageal surgery, cholecystectomy and appendicectomy, while the lack of an association between age and the risk of SSIs has been reported for various types of gastrointestinal surgery.8–11 Discrepancies in the reported role of age in SSIs among studies can be ascribed to the heterogeneity of age groupings, types of procedure examined, and mixtures of procedures, since SSI rates are known to vary among procedures.3 One of the protective factors of SSIs is the use of laparoscopy, and Centers for Disease Control and Prevention (CDC) proposed a National Nosocomial Infection Surveillance system (NNIS) risk
0195-6701/$ – see front matter Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2010.01.021
184
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
index encompassing procedure-specific laparoscopic surgery.3 For laparoscopic surgery of the gallbladder, colon and rectum, 1 was subtracted from the NNIS risk index scores.12 In recent years, laparoscopic surgery, especially cholecystectomy, has become much more common than when this risk index was proposed. Nevertheless, the differences in the risk factors of SSIs between open and laparoscopic cholecystectomy have not been reported, mainly because the majority of the studies treated the laparoscopic procedure as a covariate rather than as a separate procedure for analysis.3,13 The purposes of the study were: (i) to investigate age as a risk factor for SSIs in gastrointestinal surgery; and (ii) to examine the differences in the risk factors for SSIs between laparoscopic cholecystectomy and open cholecystectomy in a large cohort of patients undergoing gastrointestinal surgery in Japan. Methods Two university hospitals and 18 university-affiliated hospitals in the Kansai area of Japan participated in prospective SSI surveillance between March 2003 and November 2007. The procedures studied were cholecystectomy, gastrectomy, appendicectomy, colectomy and rectal resections. Not all the participating hospitals registered for surveillance of all the procedures. Before beginning data collection, study sessions were held in order to standardise the data collection method. Data were prospectively collected by the infection control professionals in each hospital. The definition of SSIs proposed by the CDC was used, and nurses or surgeons (non-primary) in the project team determined the presence of an SSI.3 Patients were routinely followed-up at the hospital where they had surgery, and surgeons were requested to report post-discharge SSIs to the surveillance team. One author (J.S.) was in charge of data collection and held semi-annual feedback sessions at the participating hospitals. Variables collected were demographic data, the American Society of Anesthesiologists (ASA) score, dates of admission and discharge, date of operation, type of procedure, emergency status, use of laparoscopy, prophylactic antibiotics, multiple dose antibiotics, use of drains, stoma formation, silk suture use, wound class, duration of operation, and the status of SSIs. Patient ages were grouped into decades. ASA scores were aggregated into 2 and 3, and wounds were categorised as clean contaminated (class 2) and contaminated or dirty-infected (class 3 or 4). Two cut-off points were defined for operation duration. One definition used the CDC guidelines; the other was defined as the 75th centile of the distribution of the operation duration for each procedure because operation durations in Japan tend to be longer than those reported in the USA, especially for gastrectomy. 14 If a laparoscopic procedure was attempted and then progressed to open surgery, the procedure was classified as an open surgery.
Statistical analysis SPSS II for Windows was used for statistical analysis. The Student’s t-test was used for continuous variables and the c2-test was used for categorical variables to test statistical significance. Variables that reached statistical significance (P < 0.05) were entered into the logistic regression to estimate adjusted risks. In order to examine the validity of data collected from 20 institutions, SSI rates were stratified by the NNIS risk index and type of procedure. The NNIS index is a composite score consisting of ASA score, wound class, duration of surgery and application of laparoscopic procedure and has been used for risk adjustment worldwide.12 The surveillance study was part of a quality improvement project for the participating hospitals. This research was approved by the ethics committee of Osaka University. Results During the study period, 12 126 patients were registered. After excluding cases with missing data on SSI status, age and NNIS risk index, 12 015 cases were analysed. The patients’ mean age was 60.3 years (SD: 16.5), and the distribution of risk factors differed among procedures (Table I). Appendicectomy had a lower risk profile, except for wound class, than other procedures. Patients with laparoscopic cholecystectomy had better ASA scores and fewer class 3 wounds than those with open cholecystectomy. Our cut-off points for operation duration in open cholecystectomy, gastrectomy and rectal resections were longer than the cut-off points defined by the CDC. Laparoscopic procedures were used in fewer than 10% of each surgery with the exception of cholecystectomy (Table I). More than 80% of the patients received antibiotic prophylaxis, and 50% of the patients received an additional dose of antibiotics for operation duration >180 min. Silk sutures were used in approximately 50% of procedures, except for laparoscopic cholecystectomy and appendicectomy (Table I). The median length of hospital stay was 19 days (interquartile range: 11–28 days). The total number of SSIs was 1471 with an overall SSI incidence of 12.2%. The SSI incidence ranged from a low of 1.4% for laparoscopic cholecystectomy to a high of 24.2% for rectal resections (Table I). SSI incidence was stratified by the NNIS risk index and procedure (Figure 1). The risk of SSIs increased with increasing NNIS index scores in each procedure, and SSI rates for laparoscopic surgery were lower than those for open surgery. In univariate analysis, significant age differences between those with SSIs and those without were found for open cholecystectomy, gastrectomy and appendicectomy. The procedure-specific SSI incidence was then tabulated and plotted by age decade (Figure 2). For
Table I Demographics and clinical characteristics
SSI incidence (%) Age (mean, SD) Women (%) ASA classification 3 (%) Wound class 3 (%) Laparoscope use (%) Single dose antibiotics (%) Silk suture (%) Drain use (%)
Laparoscopic cholecystectomy
Open cholecystectomy
Gastrectomy
Appendicectomy
Colon resection
Rectal resection
N ¼ 2142 (17.8%)
N ¼ 641 (5.3%)
N ¼ 3493 (29.1%)
N ¼ 1298 (10.8%)
N ¼ 2939 (24.5%)
N ¼ 1502 (12.5%)
1.4 57.7, 13.6 54.2 3.4 8.3 100.0 90.0 2.4 59.2
8.0 64.6, 12.8 40.4 13.4 23.6 0.0 92.3 53.7 75.2
10.2 65.6, 11.4 32.7 9.0 4.4 3.3 97.5 53.8 89.5
12.4 36.7, 21.5 45.1 4.3 46.1 8.7 83.1 37.9 29.7
17.3 67.1, 11.9 44.7 13.2 10.4 7.5 94.0 52.1 70.6
24.2 64.6, 11.1 37.0 8.1 8.7 2.7 95.4 55.0 90.2
SSI, surgical site infection; ASA, American Society of Anesthesiologists.
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
appendicectomy and open cholecystectomy, SSI incidence rates linearly increased with each decade. For gastrectomy, SSI rates tended to increase with age, although the SSI rate for the 80–89year-old group declined (n ¼ 298). In contrast, for laparoscopic cholecystectomy, SSI incidence rates varied little by age. For colectomy and rectal resections, younger age groups had higher SSI rates than older age groups (Figure 2). For appendicectomy, the proportion of class 3 wounds increased sharply with age, from 33% for 20–29-year-olds to 75% for 70–79-year-olds. To a lesser extent, a similar trend was seen for open cholecystectomy; the proportion of class 3 wounds increased with age, from 15% for 40–49-year-olds to 35% for patients 80 years of age and older. For laparoscopic cholecystectomy, gastrectomy, colectomy and rectal resections, the proportion of class 3 wounds varied little among age groups. Multivariate analysis was conducted for each procedure and each operation time, and the results were similar. Table II displays the significant factors associated with SSIs using the adjusted operation times. Operation duration was a risk factor for SSIs in all six procedures, and wound class and drain use were also risk factors for SSIs in most procedures. Age was a risk factor in open cholecystectomy, gastrectomy, and appendicectomy (Table II). Use of silk sutures was a risk factor in laparoscopic cholecystectomy and appendicectomy. Antibiotic prophylaxis use was not a protective factor for any procedure, and administration of multiple doses of antibiotics was a risk factor for SSI in gastrectomy (Table II). Further analysis of gastrectomy with operation duration >180 min showed that patients receiving multiple doses of antibiotics had a significantly
185
greater duration of surgery than those who did not receive multiple doses (P < 0.001). The risk factors for SSIs in laparoscopic cholecystectomy were quite different from those in open procedures, with or without adjustment for relevant variables, and the only common risk factor was longer duration of surgery (Table II). In the other procedures, use of laparoscopy was a significant protective factor in gastrectomy and colon resections. Discussion This study found variations in SSI incidence by age for different gastrointestinal surgery procedures, suggesting the importance of examining procedure-specific SSI incidence. Although patient age cannot be altered, identifying it as a risk factor is important for communicating the risk to the patient and for exploring risk reduction strategies. Our study was consistent with two large studies in which increased age was associated with increased risk for SSIs in appendicectomy and cholecystectomy.9,13 Studies in which age was not a risk factor for SSIs were mostly colorectal surgery; our study supports these findings.11,15 A common characteristic of studies that failed to show age as a risk factor for SSIs is the lack of procedure-specific analysis.7,10,16 Increased age is accompanied by reduced immunity and increased comorbidity, which may account for an increased rate of SSIs in the elderly.17,18 If this is so, SSI incidence would increase in the elderly with any procedure. However, SSI rates only increased with age in three out of six procedures examined in our study. Of
60
Incidence of SSI (%)
50 40 30 20 10 0
M 0 1 2
*
*
3
0-Y0-N 1 2 3
CHOL
GAST
0-Y 0-N 1 2
*
*
3
M 0 1 2 3
APPY
* M 0
COLN
Colectomy,
Category = NNIS risk index – 1 with laparoscope use.
cholecystectomy,
M means NNIS risk index = 0 and laparoscope use.
* 1
2
3
REC
rectum resection Gastrectomy,
0-Y: NNIS risk index = 0 and laparoscope use.
appendectomy
0-N: NNIS risk index = 0 and no laparoscope use. No other adjustment.
Figure 1. Surgical site infection rates by selected operative procedure and modified risk index category incorporating laparoscope use.12 *Sample N < 50; y0% (0/67). CHOL, cholecystectomy; GAST, gastrectomy; APPY, appendicectomy; COLN, colectomy; REC, rectum resection.
186
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
35.0 30.0
SSI (%)
25.0
LapaCHOL OpenCHOL
20.0
GAST 15.0
APPY COLN
10.0
REC 5.0 0.0
<20
20-29
30-39
40-49 50-59 Age (years)
60-69
70-79
80-89
Figure 2. Incidence of surgical site infection (SSI) by age group and type of surgical procedure. Samples of N 50 are recorded. Lapa, laparoscopy; CHOL, cholecystectomy; GAST, gastrectomy; APPY, appendicectomy; COLN, colectomy; REC, rectum resection.
these, appendicectomy and open cholecystectomy share inflammation as a common condition, and the proportion of class 3 wounds increased with age in both procedures. Even after adjustment, age remained an independent risk factor for these two procedures. Inflammation in older patients may be an indication of reduced local defence against infection and/or increased co-infections in other sites, raising the risk of SSIs. However, in laparoscopic cholecystectomy, patient age was not a risk factor for SSIs, and these hypotheses do not explain the difference in risk factors between open and laparoscopic Table II Logistic regression analysis of surgical site infection risk factors Surgical procedure Laparoscopic cholecystectomy
Variable
Adjusted OR
95% CI
P-value
Silk suture Operation duration
4.59 4.17
1.50–14.03 0.008 1.97–8.80 <0.001
Open cholecystectomy Drain use Wound class 3 Operation duration Male Age
5.24 2.61 2.14 2.10 1.37
1.22–22.45 1.38–4.93 1.13–4.05 1.03–4.25 1.05–1.79
Gastrectomy
Wound class 3 Operation duration Multi-dose antibiotics Age Laparoscope use
4.44 2.33 1.77 1.15 0.33
2.99–6.81 1.82–2.98 1.36–2.29 1.04–1.27 0.13–0.84
<0.001 <0.001 <0.001 0.008 0.002
Appendicectomy
Wound class 3 Drain use Operation duration Silk suture Age
3.62 3.21 1.93 1.65 1.17
2.15–6.10 2.02–5.13 1.30–2.87 1.13–2.41 1.07–1.28
<0.001 <0.001 0.001 0.009 0.001
Colon resection
Wound class 3 Drain use Stoma Operation duration Laparoscope use
3.49 2.13 2.04 1.33 0.54
2.61–4.66 1.60–2.82 1.46–2.85 1.05–1.67 0.32–0.91
<0.001 <0.001 <0.001 0.016 0.021
Rectal resection
Wound class 3 Drain use Stoma Operation duration Male
3.51 3.12 2.26 1.34 1.42
2.34–5.26 1.67–5.84 1.70–2.99 1.00–1.80 1.08–1.87
<0.001 <0.001 <0.001 0.048 0.013
OR, odds ratio; CI, confidence interval.
0.026 0.003 0.019 0.040 0.021
cholecystectomy. Although laparoscopy is now used in nearly 80% of cholecystectomy, expanding the application of laparoscopic procedures may reduce the overall SSI incidence further in cholecystectomy. Since the risk factors for SSIs following laparoscopic cholecystectomy are quite different from those in open procedures, risk estimates for laparoscopic procedures should be separated from open procedures in order to obtain accurate risk assessments, if the sample size permits. For gastrectomy, the association between advanced age and the risk of SSIs cannot be explained by the presence of infection, as the proportion of class 3 wounds did not vary by age. Stomach cancer incidence in Japan is much higher than in western countries where SSI incidence rates for gastrectomy with stomach cancer have not been reported. Further research is necessary to explore differences between younger and older patients for gastrectomy. In contrast to these procedures, younger age groups had higher SSI rates than older groups in colectomy and rectal resection. Younger patients might have undergone more complicated surgery with greater risk of SSIs. Differences in indication for surgery between younger and older patients need to be compared in future research. Drain use increased the risk of SSIs 2–5-fold in open gastrointestinal surgery with the exception of gastrectomy, and our study concurs with previous studies of open gastrointestinal surgery.19 A high prevalence of drain use is not unique to our study as others reported 70–80% of drain use in gastrointestinal surgery in Japan and in Taiwan.16,20 All the articles reviewing prophylactic drainage showed the lack of benefit or increased likelihood of negative outcomes in colorectal surgery, open cholecystectomy and appendicectomy.19,21,22 In a meta-analysis of prophylactic drainage in gastrointestinal surgery, an indication of prophylactic drainage use was limited to oesophageal resection and total gastrectomy with an evidence level of ‘D’ – namely expert opinion.22 Concerted efforts are necessary to reduce drain use in order to lower SSI rates as well as other postoperative complications. The use of silk sutures was a predictor for SSIs in laparoscopic cholecystectomy and appendicectomy. Silk sutures are nonabsorbable, multifilament suture materials, both of which are risk factors for SSIs.23 In Japan, silk is still commonly used for intraabdominal suturing and ligation because it is less expensive and
M. Utsumi et al. / Journal of Hospital Infection 75 (2010) 183–187
easier to handle than absorptive synthetic suture material.16 A recent retrospective cohort study of hepatectomy patients reported an adjusted SSI odds ratio of 3.4 for silk suture use when compared with absorptive materials.23 Silk suture use has been decreasing in Japan, which would contribute to the decline in SSI rates. Although the lack of benefit from antimicrobial prophylaxis was unexpected, a few other studies have also reported a lack of benefit.11,13 Effectiveness of prophylactic antibiotics varies among gastrointestinal surgery. A meta-analysis of laparoscopic cholecystectomy showed that prophylactic antibiotics were ineffective in preventing postsurgical complications, while antibiotic prophylaxis was shown to reduce SSIs in colectomy and appendicectomy.24–26 In colon surgery, all randomised clinical trials for antibiotic prophylactic evaluation were conducted in the 1980s, and placebo groups had SSI rates of approximately 40%, nearly twice as high as the rate in our corresponding group.26 Advances in pre- and intraoperative management in the past 30 years may have contributed to the risk reductions. The changes include abandoning preoperative shaving and prevention of hypothermia, etc. Overall reductions in SSIs in colectomy may make it difficult to demonstrate an effect of antibiotic prophylaxis. An additional antibiotic dose was a risk factor in gastrectomy. As the patients with multiple doses of antibiotics had longer operation duration than those without, it may reflect an unadjusted risk. Nevertheless, timing, dosage, and type of antibiotic use in prophylaxis need to be examined to monitor proper use of antibiotics. One of the limitations of the study was the lack of clinical indicators that were considered to be risk factors for SSIs, such as history of smoking and blood transfusion.27,28 However, these indicators are unlikely to affect elderly patients going through only appendicectomy, open cholecystectomy, or gastrectomy. Strong support for our research comes from its size. Twenty hospitals participated in the study using standard protocols, providing a large body of data, which enabled us to demonstrate the importance of procedure-specific SSIs in examining risk factors. SSI rates by NNIS index are in accordance with western studies, demonstrating the validity of the data collected by multiple institutions. In addition, postoperative hospital stays in Japan are much longer than in the UK because fee-for-service was still in place at the time of the study, which made it easier to establish the presence of SSIs. However, a prospective payment system is being introduced in acute care hospitals in Japan, and post-discharge surveillance will become more important in the near future. The proportion of elderly among surgical patients has been increasing due to population ageing. Patient age is a significant predictor of SSIs in some gastrointestinal procedures. Accurate risk estimates would contribute to the prevention, early detection and initiation of treatment in high risk groups. The current study also suggests the need for separate risk assessments for laparoscopic procedures and open procedures. Acknowledgements The authors wish to acknowledge the nurses and physicians who cooperated in this project. Special thanks go to NTT West Osaka Hospital, Osaka Senin Hoken Hospital, Osaka Medical Center for Cancer and Cardiovascular Diseases, Kanou Hospital, Kansai Rosai Hospital, Kinki Daigaku Igakubu Nara Hospital, Kinki Central Hospital Aid Association of Public School Teachers, Kobe Ekisaikai Hospital, National Hospital Organization, Osaka National Hospital, Shitennouji Hospital, Ikeda Municipal Hospital, Itami City Hospital, Kawanishi City Hospital, Sakai Municipal Hospital, Toyonaka Municipal Hospital, Suita Municipal Hospital, Higashiosaka City General Hospital, Fujimoto Hospital, Minoh City Hospital, and Yao Municipal Hospital.
187
Conflict of interest statement None declared. Funding sources None. References 1. Smyth ET, McIlvenny G, Enstone JE, et al. Four country healthcare associated infection prevalence survey 2006: overview of the results. J Hosp Infect 2008;69:230–248. 2. Smith RL, Bohl JK, McElearney ST, et al. Wound infection after elective colorectal resection. Ann Surg 2004;239:599–605. 3. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guidelines for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999;20:250–278. 4. Kirkland KB, Briggs JP, Trivette SL, Wilkinson WE, Sexton DJ. The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol 1999;20:725–730. 5. Kaye KS, Schmit K, Pieper C, et al. The effect of increasing age on the risk of surgical site infection. J Infect Dis 2005;191:1056–1062. 6. Razavi SM, Ibrahimpoor M, Sabouri Kashani A, Jafarian A. Abdominal surgical site infections: incidence and risk factors at an Iranian teaching hospital. BMC Surg 2005;5:2. 7. Petrosillo N, Drapeau CM, Nicastri E, Martini L, Ippolito G, Moro ML. Surgical site infections in Italian hospitals: a prospective multicenter study. BMC Infect Dis 2008;8:34. 8. Breigeiron R, de Souza HP, Sidou JP. Risk factors for surgical site infection after surgery for esophageal perforation. Dis Esophagus 2008;21:266–271. 9. Rotermann M. Infection after cholecystectomy, hysterectomy or appendectomy. Health Rep 2004;15:11–23. 10. de Oliveira AC, Ciosak SI, Ferraz EM, Grinbaum RS. Surgical site infection in patients submitted to digestive surgery: risk prediction and the NNIS risk index. Am J Infect Control 2006;34:201–207. 11. Walz JM, Paterson CA, Seligowski JM, Heard SO. Surgical site infection following bowel surgery: a retrospective analysis of 1446 patients. Arch Surg 2006;141:1014–1018. 12. National Nosocomial Infections Surveillance (NNIS) System. NNIS Report, Data Summary from January 1990–May 1999. Am J Infect Control 1999;27:520–532. 13. Romy S, Eisenring MC, Bettschart V, Petignat C, Francioli P, Troillet N. Laparoscope use and surgical site infections in digestive surgery. Ann Surg 2008;247:627–632. 14. Imai E, Ueda M, Kanao K, et al. Surgical site infection risk factors identified by multivariate analysis for patient undergoing laparoscopic, open colon, and gastric surgery. Am J Infect Control 2008;36:727–731. 15. Tang R, Chen HH, Wang YL, et al. Risk factors for surgical site infection after elective resection of the colon and rectum: a single-center prospective study of 2,809 consecutive patients. Ann Surg 2001;234:181–189. 16. Watanabe A, Kohnoe S, Shimabukuro R, et al. Risk factors associated with surgical site infection in upper and lower gastrointestinal surgery. Surg Today 2008;38:404–412. 17. Schwab R, Walters CA, Weksler ME. Host defense mechanisms and aging. Semin Oncol 1989;16:20–27. 18. Gardner ID. The effect of aging on susceptibility to infection. Rev Infect Dis 1980;2:801–810. 19. Gurusamy KS, Samraj K. Routine abdominal drainage for uncomplicated open cholecystectomy. Cochrane Database Syst Rev 2007;2. CD006003. 20. Yeh CY, Changchien CR, Wang JY, et al. Pelvic drainage and other risk factors for leakage after elective anterior resection in rectal cancer patients: a prospective study of 978 patients. Ann Surg 2005;241:9–13. 21. Jesus EC, Karliczek A, Matos D, Castro AA, Atallah AN. Prophylactic anastomotic drainage for colorectal surgery. Cochrane Database Syst Rev 2004;4. :CD002100. 22. Petrowsky H, Demartines N, Rousson V, Clavien PA. Evidence-based value of prophylactic drainage in gastrointestinal surgery: a systematic review and meta-analysis. Ann Surg 2004;240:1074–1084. 23. Togo S, Kubota T, Takahashi T, et al. Usefulness of absorbable sutures in preventing surgical site infection in hepatectomy. J Gastrointest Surg 2008;12: 1041–1046. 24. Choudhary A, Bechtold ML, Puli SR, Othman MO, Roy PK. Role of prophylactic antibiotics in laparoscopic cholecystectomy: a meta-analysis. J Gastrointest Surg 2008;12:1847–1853. 25. Andersen BR, Kallehave FL, Andersen HK. Antibiotics versus placebo for prevention of postoperative infection after appendicectomy. Cochrane Database Syst Rev 2005;20. CD001439. 26. Bratzler DW, Houck PM. Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical Infection Prevention Project. Clin Infect Dis 2004;38:1706–1715. 27. Sorensen LT, Horby J, Friis E, Pilsgaard B, Jorgensen T. Smoking as a risk factor for wound healing and infection in breast cancer surgery. Eur J Surg Oncol 2002;28:815–820. 28. Leal-Noval SR, Rincon-Ferrari MD, Garcia-Curiel A, et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest 2001;119:1461–1468.