Professional Documents
Culture Documents
DOI 10.1007/s10156-011-0298-y
ORIGINAL ARTICLE
Hiroshige Mikamo
Received: 16 March 2011 / Accepted: 16 August 2011 / Published online: 9 September 2011
Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases 2011
Abstract Surgical-site infection (SSI) is a major contrib- Keywords Surgical-site infection (SSI) Colorectal
utor to patient mortality rates and health care costs. Due to the surgery Antimicrobial prophylaxis
high risk of bacterial contamination, colorectal surgery is
associated with a particularly high risk of postoperative
infection. The surveillance reported here was conducted at Introduction
Aichi Medical University Hospital on 304 patients who
underwent elective colorectal resectiontotal or partial Surgical-site infection (SSI) is dened as infections
from June 2006 to May 2009. To determine risk factors for occurring within 30 daysor within 1 year in the case of
SSI, multivariate analysis was used. Forty-six (15.1%) implantation of a foreign bodyafter surgery and
patients were diagnosed with SSI. Patients who received affecting either the incision (supercially or deeply),
cefotiam for prophylaxis showed the highest incidence of SSI organs, or body spaces at the site of operation [1]. In
(26.6%), and patients who were administered omoxef 2002, the US Centers for Medicare and Medicaid Services
showed the lowest incidence (8.1%). Patients who developed implemented the National Surgical Infection Prevention
SSI were more likely to intraoperative blood loss Project. This project builds on experience that the US
(308.1 29.8 vs. 153.9 12.2; p \ 0.05), longer postop- Centers for Disease Control and Prevention (CDC) have
erative antimicrobial administration (5.3 2.2 vs. 4.5 1.5; gained from implementing the National Nosocomial
p \ 0.05), and longer operative time (3.3 1.6 vs. Infections Surveillance System (CDC-NNIS). According
2.7 1.2; p \ 0.05). Intraoperative bleeding, antimicrobial to data from the system, SSI accounts for 1416% of
choices to cover both anaerobic and aerobic bacteria, and reported nosocomial infections among all hospitalized
length of antimicrobial administration were independently patients and 38% among surgical patients [1, 2]. In Europe,
predictive of SSI development according to multivariate available data show that the incidence of SSI can be as high
logistic regression analysis. These results suggest that the as 20% depending on surgical procedure, surveillance
degree of operative invasion and anaerobic bacteria contrib- criteria, and study design.
ute to SSI following colorectal surgery. As for patient resistance, intrinsic patient characteristics
strictly correlating with an increased risk of SSI include
advanced age, an American Society of Anesthesiologists
(ASA) score of III, obesity, pre-existing illness, and host-
M. Hagihara (&) M. Suwa Y. Muramatsu Y. Kato defense deciency [38]. Moreover, risk factors for SSI
Y. Yamagishi H. Mikamo
related to the surgical procedure include quality of surgical
Department of Infection Control and Prevention,
Aichi Medical University School of Medicine, care, diabetes mellitus, surgery type and duration, emer-
Nagakute-cho, Aichi-gun, Aichi 480-1195, Japan gency procedure, blood transfusion, intraoperative hypo-
e-mail: hagimao@hotmail.com thermia, and systemic hypoxemia [912]. Additionally, SSI
is a major contributor to patient mortality rates and health
Y. Ito
Department of Pharmacy, Aichi Medical University School care costs. Mortality rates were two to three times higher in
of Medicine, Nagakute-cho, Aichi-gun, Aichi 480-1195, Japan patients in whom SSI developed compared with uninfected
123
84 J Infect Chemother (2012) 18:8389
patients, and hospital readmission rates were signicantly conrm patients case record; cases with no explicitly
increased [5, 1315]. documented incision time or antimicrobial administration
In large-bowel surgery, the predominant isolates are time; infection, including peritonitis and perforation, doc-
Bacteroides fragilis and Escherichia coli [16]. Due to the umented intraoperatively or within 48 h after the end of
high risk of bacterial contamination, colorectal surgery is surgery; obstruction of the small bowel; stomal or bypass
associated with a particularly high risk of postoperative surgery; preoperative infectious diseases; and oral antibi-
infection. SSI rates of up to 40% and of about 25% have otic or steroid administration before surgery. Antimicro-
been found in patients not receiving or receiving periop- bials were considered prophylactic if given before surgery,
erative antibiotic prophylaxis, respectively [3, 17]. Indeed, intraoperatively, or within 24 h after the end of surgery.
the efcacy of perioperative systemic antimicrobial agents
is mainly related to suppression of bacterial growth in the Statistical analysis
tissues of the operative eld once contamination occurrs.
As previous studies demonstrate, inappropriate timing of Statistical analysis was performed using JMP software
antimicrobial administration and inappropriate selection of (5.1.1 SAS Institute). The univariate relation between each
the antimicrobial agent extend patient admission [1822]. independent variable and SSI was evaluated using a t test
It is universally accepted that antibiotic prophylaxis, i.e., for continuous variables. Pearsons v2 test for categorical
perioperative administration of suitable antimicrobial variables or Fishers exact test, used when variables were
agents, must be considered one of the most important \5, was used to analyze categorized variables. In multi-
measures for preventing SSI [2329]. However, the use of variate analysis, logistic regression analysis with a Wald
antimicrobial prophylaxis is often suboptimal. Therefore, statistic step-forward selection was used. One-way analysis
we performed a retrospective SSI surveillance focusing on of variance (ANOVA) was used to compare SSI risk fac-
patients undergoing colorectal surgery at the Aichi Medical tors in each group (cefmetazole, cephamycin and a second-
University Hospital. The aim of the study was to determine generation cephalosporin; cefotiam, a second-generation
the actual incidence of SSI and to identify risk factors for cephalosporin; omoxef, oxacephem and identied third-
developing SSI in patients undergoing colorectal surgery. generation piperacillin). Signicance was dened as
p \ 0.05.
123
J Infect Chemother (2012) 18:8389 85
Table 1 Patient demographics on some antimicrobial based on clinical symptoms and data
Characteristics
(fever and inammation markers). Consequently, mean
term of postoperative antimicrobial administration for
The number of operations 304 (209/95) prophylaxis was 4.6 days.
(colon/rectal)
Age (year) 66 12 Univariate analysis
(19 to 91)
Sex (male/female) 181/123
Table 2 summarizes comparisons of patient characteristics.
Diabetic mellitus (yes/no) 51/253
In the univariate analysis, patients were divided into those
Operation time (h) 2.8 1.3
(0.41 to 8.52)
with or without SSI. When we evaluated perioperative and
Blood loss (ml) 191.8 252.1
operative characteristics, patients who developed SSI were
(3.0 to 920.0) more likely to have intraoperative blood loss (308.1 29.8
Intraoperative body temperature (C) 36.4 2.8 vs. 153.9 12.2; p \ 0.05), longer postoperative antimi-
(35.8 to 38.9) crobial administration (5.3 2.2 vs. 4.5 1.5; p \ 0.05),
Postoperation body temperature (C) 37.1 0.6 and longer operative time (3.3 1.6 vs. 2.7 1.2;
(35.1 to 40.0) p \ 0.05). The prolonged length of time from incision to
Blood sugar (mg/dl) (day 0) 125.9 29.3 antimicrobial dosing in the piperacillin group was signi-
(119.5 to 192.0)
cantly associated with the development of SSI (25.4 5.2
Blood sugar (mg/dl) (day 1) 138.9 35.8 vs. 12.2 2.1; p \ 0.05). However, none of the other
(74.8 to 258.0)
factors recorded were statistically associated with the
Blood sugar (mg/dl) (day 2) 158.9 36.6
(151.1 to 231.0) development of incision SSI.
ASA score
1 73 (24.0%)
Comparison of SSI risk factors in each antimicrobial
2 194 (63.8%)
group
3 26 (8.6%)
Table 3 summarized results of the comparison of risk
4 1 (0.3%)
factors in each antimicrobial group. There were signicant
Antimicrobial prophylaxis 302/2
(yes/no) (case) differences between piperacillin and omoxef groups in
Cefmetazole 36 (11.9%) diabetes mellitus (p \ 0.05), blood sugar level (day 0)
Cefotiam 64 (21.2%) (p \ 0.05), blood sugar level (day 2) (p \ 0.05), term of
Flomoxef 136 (45.0%) postoperative administration of antimicrobial agents
Piperacillin 51 (16.9%) (p \ 0.05); there were signicant differences between
Other 15 (5.0%)
piperacillin and cefmetazole groups in terms of postoper-
SSI incidence (yes/no)
ative administration of antimicrobial agents (p \ 0.05).
Cefmetazole 5/46 (9.8%)
Multivariable analysis
Cefotiam 17/47 (26.6%)
Flomoxef 11/125 (8.1%)
Table 4 summarizes results from the multivariate analysis.
Piperacillin 5/31 (13.9%)
Following the univariate analysis, variables, operative
First antimicrobial dosing (min)a 14.9 17.1
(-109.0 to 86.0) blood loss, selection of cefotiam for prophylaxis, and
Term of postoperative administration of 4.6 1.7 length of administration of postoperative antimicrobials
antimicrobial agents (day) (1.0 to 14.0) were selected with stepwise logistic regression analysis.
Intraoperative redosing (case) 42 (13.8%)
Laparoscopic surgery (yes/no) 41/263
Discussion
ASA American Society of Anesthesiologists, SSI surgical-site
infection
a
The time from antimicrobial dosing to incision
Antimicrobial prophylaxis to prevent SSI is one of the most
widely accepted practices in surgery. Our results showed
that 99.4% of all patients who received elective colorectal
and 5% other antimicrobials. Moreover, patients who surgery were administered antimicrobial prophylaxis. SSI
received cefotiam showed the highest incidence of SSI incidence in for colorectal resections during the study
(26.6%), followed by piperacillin (13.7%), cefmetazole period was 15.1%, which was in range with a general
(9.6%), and omoxef (8.1%). Most patients were continued review of the literature [26, 30, 31].
123
86 J Infect Chemother (2012) 18:8389
Table 2 Comparison between surgical-site infection (SSI) group and non-SSI group
SSI Non-SSI
As previous studies demonstrated inappropriate timing attain the best results. Indeed, Nelson et al. [32], in a
of antimicrobial administration, inappropriate selection of systemic review on antibiotic prophylaxis in colorectal
antimicrobial agent extended patient admission [1822]. surgery, studied 182 randomized controlled trials and
Optimal prophylaxis requires application in appropriate found that the addition of an anaerobic to an aerobic cov-
types of operations, selection of safe and effective anti- erage resulted in a statistically signicant reduction in the
microbials, initial administration and redosing to maintain incidence of SSI. In our study, 45% of patients received
effective serum and tissue levels throughout the operation, omoxef, 21% cefotiam, 17% piperacillin, and 12% cef-
and discontinuation when the patient is no longer receiving metazole. Patients administered cefotiam showed the
a benet [1]. Stone et al. [26] showed that the lowest rates highest incidence of SSI among all patients during the
of SSI in abdominal operations were associated with pro- study period. Cefotiam is not sold in Western countries,
phylaxis started within 1 h before incision [29]. However, and many drugs in this study, except cefotiam, are thought
our results showed that 90.8% patients received antimi- to be effective for anaerobic bacteria. Our results also
crobial prophylaxis within 1 h before incision. In patients suggest that antimicrobial prophylaxis needs to cover both
who received prophylactic piperacillin and developed SSI, aerobic and anaerobic bacteria to prevent SSI. Indeed, the
the time from antimicrobial dosing to incision were sig- colon and distal small intestines contain an enormous res-
nicantly longer than for other patients. Piperacillin has a ervoir of facultative and anaerobic bacteria, which is sep-
shorter half-life than other antimicrobials. Therefore, our arated from the rest of the body by the mucous membrane.
result suggest the necessity of considering an antimicrobial Hence, colorectal surgery might have a higher risk for SSI
blood concentration based on antimicrobial half-time and development. Furthermore, although intraoperative redos-
operation time, even though all antimicrobials we used ing is generally recommended only when the operation
were within 1 h prior to operation. lasts longer than 23 h, Song and Glenny [33] showed a
As choosing an antimicrobial agent for prophylaxis in single-dose or short-term regimen is as effective as long-
colorectal surgery, there is general agreement that antimi- term postoperative administration and that no signicant
crobials that cover both anaerobic and aerobic bacteria difference in the incidence of SSI can be found between a
123
J Infect Chemother (2012) 18:8389 87
Table 3 Comparison of surgical-site infection (SSI) risk factors in each antimicrobial group
Cefmetazole Cefotiam Flomoxef Piperacillin
single-dose regimen and a multidose regimen. In our study, criticized as being unsuitable for risk evaluation in colo-
13.8% of all patients received intraoperative redosing, rectal surgery because most patients undergoing such the
although mean operation time was 2.8 h; 75.0% patients procedure have an ASA score of 1 or 2 and a wound
received postoperative prophylaxis, although prophylactic classication of cleancontaminated.
antibiotics are recommended to be given for \24 h after There are several important limitations to this study.
operation. Patients with longer operative time and longer First, this was retrospective analysis and not a randomized
postoperative antimicrobial prophylaxis had a higher inci- controlled study to examine the signicance of specic risk
dence of SSI. These results reect the degree of invasion factors for SSI. Therefore, there was some variability
for patients undergoing colorectal surgery. among practices [the use of antimicrobial-coated sutures,
Previous studies suggest that SSI rates might not be the mechanical bowel preparation (glycolelectrolyte solution
same between colon and rectum surgery [34, 35]. Although or 2-Hydroxy-1,2,3-propanetricarboxylic acid magnesium
our results showed no signicant difference in SSI inci- salt)]. Second, we did not take into account stratication by
dence between colon and rectum surgery groups, there was surgeon because this study was conducted in a hospital.
a trend to SSI development in the rectum surgery group. In Although the term of investigation was 3 years, a few
addition, the reason ASA score did not selected as a risk surgeons in a group conducted colorectal surgeries. Addi-
factor of SSI in our study is the NNIS Risk Index has been tionally, SSI is categorized into incisional and organ/space
123
88 J Infect Chemother (2012) 18:8389
123
J Infect Chemother (2012) 18:8389 89
surgical procedures. The Antibiotic Prophylaxis Study Group. administered preoperatively at different times. Clin Pharm.
Am J Surg. 1996;171:54852. 1984;3:647.
19. Matuschka PR, Cheadle WG, Burke JD, Garrison RN. A new 29. Classen DC, Evans RS, Pestotnik SL, Horn SD, Menlove RL,
standard of care: administration of preoperative antibiotics in the Burke JP. The timing of prophylactic administration of antibiotics
operating room. Am Surg. 1997;63:5003. and the risk of surgical-wound infection. N Engl J Med.
20. Gorecki P, Schein M, Rucinski JC, Wise L. Antibiotic adminis- 1992;326:2816.
tration in patients undergoing common surgical procedures in a 30. Clarke JS, Condon RE, Bartlett JG, Gorbach SL, Nichols RL,
community teaching hospital: the chaos continues. World J Surg. Ochi S. Preoperative oral antibiotics reduce septic complications
1999;23:42932. of colon operations: results of prospective, randomized, double-
21. Trick WE, Scheckler WE, Tokars JI, Jones KC, Reppen ML, blind clinical study. Ann Surg. 1977;186:2519.
Smith EM, et al. Modiable risk factors associated with deep 31. Goldring J, McNaught W, Scott A, Gillespie G. Prophylactic oral
sternal site infection after coronary artery bypass grafting. antimicrobial agents in elective colonic surgery. A controlled
J Thorac Cardiovasc Surg. 2000;119(1):10814. trial. Lancet. 1975;2:9971000.
22. Burke JP. Maximizing appropriate antibiotic prophylaxis for 32. Nelson RL, Glenny AM, Song F. Antimicrobial prophylaxis for
surgical patients: an update from LDS Hospital, Salt Lake City. colorectal surgery. Cochrane Database Syst Rev. 2009;21:
Clin Infect Dis. 2001;33(Suppl 2):S7883. CD001181.
23. Miles AA, Miles EM, Burke J. The value and duration of defence 33. Song F, Glenny AM. Antimicrobial prophylaxis in colorectal
reactions of the skin to the primary lodgement of bacteria. Br J surgery: a systematic review of randomised controlled trials.
Exp Pathol. 1957;38(1):7996. Health Technol Assess. 1998;2:1110.
24. Burke JF. The effective period of preventive antibiotic action in 34. Nichols RL, Choe EU, Weldon CB. Mechanical and antibacterial
experimental incisions and dermal lesions. Surgery. 1961; bowel preparation in colon and rectal surgery. Chemotherapy.
50:1618. 2005;51:S11521.
25. Polk HC Jr, Lopez-Mayor JF. Postoperative wound infection: a 35. Coppa GF, Eng K. Factors involved in antibiotic selection in
prospective study of determinant factors and prevention. Surgery. elective colon and rectal surgery. Surgery. 1988;104:8538.
1969;66:97103. 36. Eagye KJ, Nicolau DP. Deep and organ/space infections in
26. Stone HH, Hooper CA, Kolb LD, Geheber CE, Dawkins EJ. patients undergoing elective colorectal surgery: incidence and
Antibiotic prophylaxis in gastric, biliary and colonic surgery. Ann impact on hospital length of stay and costs. Am J Surg.
Surg. 1976;184:44352. 2009;198:35967.
27. Polk HC Jr, Trachtenberg L, Finn MP. Antibiotic activity in 37. de Lalla F. Antimicrobial prophylaxis in colorectal surgery: focus
surgical incisions. The basis of prophylaxis in selected opera- on ertapenem. Ther Clin Risk Manag. 2009;5:82939.
tions. JAMA. 1980;244:13534.
28. DiPiro JT, Vallner JJ, Bowden TA Jr, Clark B, Sisley JF. Intra-
operative serum concentrations of cefazolin and cefoxitin
123