RCRK

Sslert

DPc

R2F(oCCRH

©P

isk Factors for Surgical Site Infections afterolorectal Resection in Diabetic Patients

ishabh Sehgal, MB, BCh, Arthur Berg, PhD, Rafael Figueroa, MD, Lisa S Poritz, MD,evin J McKenna, MD, David B Stewart, MD, Walter A Koltun, MD, FACS

BACKGROUND: Surgical site infections (SSIs) are a well known complication of gastrointestinal surgery andassociated with an increased morbidity, mortality and overall cost. Diabetes mellitus (DM) is arisk factor for SSI. However, there is no clear consensus as to which other risk factors play asignificant role. The goal of this study was to identify risk factors associated with SSI in patientswith DM undergoing colorectal resection.

STUDY DESIGN: A retrospective review was conducted of DM patients who underwent colorectal resection fromJune 2000 to June 2009 at Milton S Hershey Medical Center, Division of Colorectal Surgery.Individual measures were analyzed using chi-square, t-test, and Mann-Whitney U tests, andstatistical significance was confirmed using a multiple logistical regression model.

RESULTS: There were 183 DM patients included in the study, 28 (15%) of whom developed SSI. Glucoselevels were significantly higher in the SSI group for each time interval, 0 to 6 hours (211 mg/dL,p � 0.03), 0 to 48 hours (176 mg/dL, p � 0.001), and 48 to 96 hours (167 mg/dL, p � 0.012)postoperatively. Other measures significantly associated with SSI included the use of drains(p � 0.05) and the use of prophylactic antibiotics for more than 24 hours (p � 0.02). Bodymass index and stoma creation approached statistical significance (p � 0.08, 0.07, respectively).The type of hypoglycemic regimen, immunosuppression, and emergency surgery were notassociated with an increased rate of SSI.

CONCLUSIONS: Higher than normal glucose control at all postoperative time intervals was associated with SSI.The majority of glucose levels were below the American Diabetes Association recommendedlevel of 200 mg/dL, but patients still developed SSI. Type of perioperative glucose control didnot affect the incidence of SSI. These data suggest that DM patients undergoing colectomyshould have glucose tightly controlled, avoid placement of drains, and receive antibiotics for lessthan 24 hours. ( J Am Coll Surg 2011;212:29–34. © 2011 by the American College of

Surgeons)

eAefio

ctrtshadMf

urgical site infections (SSI) pose an ongoing problem inurgical practice. They lead to patient dissatisfaction, de-ayed wound healing, increased risk of incisional hernias,xtended length of hospital stay (LOS), and an increasedisk of other nosocomial complications. Furthermore,here is an increased expense associated with SSI.The CDC

isclosure Information: Nothing to disclose.resented at the American College of Surgeons 95th Clinical Congress, Chi-ago, IL, October 2009.

eceived July 15, 2010; Revised September 7, 2010; Accepted September 10,010.rom the Division of Colon and Rectal Surgery, Department of SurgerySehgal, Berg, Figueroa, Poritz, McKenna, Stewart, Koltun), and Departmentf Public Health Sciences (Berg), Penn State Milton S Hershey Medicalenter, Hershey, PA.orrespondence address: Rishabh Sehgal, MB, BCh, Division of Colon andectal Surgery, Penn State Milton S Hershey Medical Center, PO Box 850,

tershey, PA 17033. email: rsehgal@hmc.psu.edu or wkoltun@hmc.psu.edu

292011 by the American College of Surgeons

ublished by Elsevier Inc.

stimates an additional cost of $2,734 to $26,019 to themerican health care system per SSI, with an annual figurestimated between $130 million and $845 million.1 There-ore, reducing the incidence of SSIs is important in improv-ng both the individual patient’s health and reducing theverall cost of health care.

Patients undergoing colorectal resection are at an in-reased risk for developing SSI due to the bacterial load ofhe colon. The incidence of SSIs after colorectal resectionanges from 3% to 30%.2 A wide variety of risk factorshought to play a role in the pathogenesis of SSIs have beentudied. Obesity, steroid use, anemia, bowel preparation,ypothermia, nutritional status, and blood transfusionre some of the factors considered to be associated withevelopment of an SSI. The Center for Medicare andedicaid Services (CMS) has implemented protocols

or prophylactic antibiotic choice, timing of administra-

ion, and duration of use aimed at reducing SSI, but as

ISSN 1072-7515/11/$36.00doi:10.1016/j.jamcollsurg.2010.09.011

os

isgpSobhfsaittu

MSA(sCCastecTC

DDdapw((m

ds

iotb4pspictaw

alppauubb

STwm2ttV

TS

I

1

2

3

4

30 Sehgal et al Colorectal Resection and Surgical Site Infection J Am Coll Surg

f yet it is unclear if these interventions have had aignificant impact.

Diabetes mellitus (DM) has been shown to play a signif-cant predisposing role for SSIs in many surgical disciplinesuch as orthopaedic, cardiac, and general surgery.3-6 Poorlycemic control, especially in the immediate postoperativeeriod, has been associated with an increased incidence ofSI.6 The American Diabetes Association (ADA) has rec-mmended that perioperative blood glucose levels be keptelow 200 mg/dL to minimize SSI.7 However, in thisigher risk group of DM patients, it is unclear what otheractors play a role in SSI occurrence. For example, whethereverity of DM defined by the need for insulin versus oralgents, or whether prolonged use of prophylactic antibiot-cs might alter SSI incidence is unknown. The purpose ofhis study was to investigate other potential risk factors inhe development of SSIs in this high risk diabetic groupndergoing colorectal resection.

ETHODSubjectsll patients with DM who underwent colorectal resection

segmental colectomy, total proctocolectomy, anterior re-ection, or abdominal perineal resection) by the Division ofolon and Rectal Surgery at the Milton S Hershey Medicalenter, Penn State College of Medicine, between June 2000

nd June 2009, were identified. Patients who underwent aimple colostomy closure without segmental resection and pa-ients who did not have a primary closure of their wound werexcluded. Data were collected from patient hospital records,linic notes, operative records, and electronic medical records.he study was approved by the Milton S Hershey Medicalenter Institutional Review Board.

ata collectionemographic data collected included age at surgery, sex,

iagnosis, procedure performed, length of stay (LOS), re-dmission for SSI or any other reason, the presence of areoperative stoma (ileostomy or colostomy), surgicalound classification, American Society of Anesthesiology

ASA) score (I to IV), preoperative hypoglycemic regimendiet controlled, oral hypoglycemics, or insulin), and body

Abbreviations and Acronyms

BMI � body mass indexDM � diabetes mellitusLOS � length of staySSI � surgical site infection

ass index (BMI). Patients were subdivided into those who g

id or did not develop an SSI according to the definitionet by the CDC (Table 1).

Postoperative glucose levels were divided into 3 timentervals. The first postoperative glucose (within 6 hoursf operation) was a single value. The second and thirdime values were a mean of all the glucose readings takenetween 0 and 48 hours (including the first glucose) and8 and 96 hours, respectively. The 0- to 6-hour timeeriod was in part chosen because intraoperative bloodugars were infrequently measured, so this 0- to 6-houreriod was effectively the first glucose after operation, sot was the best available measure of intraoperative glu-ose control. The 0- to 48-hour measurement is thoughto be an indicator of early postoperative glucose controlnd by definition, included the first glucose value asell.Operative factors analyzed included creation of a stoma,

pproach (open, laparoscopic, including hand assisted, oraparoscopic converted to open), mode of skin closure (sta-les vs subcutaneous closure), use of intra-abdominal orelvic drains, preoperative bowel preparation, prophylacticntibiotic use, preoperative steroid or immunosuppressionse, evidence of intra-abdominal infection at operation,se of antibiotics longer than 24 hours, requirement forlood products, estimated blood loss, core intraoperativeody temperature, and operative time.

tatistical analyseshe R statistical software system (version 2.9.2, http://ww.r-project.org/) was used to perform univariate andultivariate statistical analyses. The Fisher’s exact test,

-sample t-test, and Mann-Whitney U test were used toest the association of individual predictors with SSI. Forhe multivariate analysis, logistic regression was performed.ariable selection for consideration in the multivariate lo-

able 1. Criteria for Defining a Surgical Site Infectionuperficial incisional surgical site infection

nfection occurs within 30 days after the operation and infectioninvolves only skin or subcutaneous tissue of the incision and atleast one of the following:

. Purulent drainage, with or without laboratory confirmation,from the superficial incision.

. Organisms isolated from an aseptically obtained culture of fluidor tissue from the superficial incision.

. At least one of the following signs or symptoms of infection:pain or tenderness, localized swelling, redness, or heat andsuperficial incision is deliberately opened by surgeon, unlessincision is culture-negative.

. Diagnosis of superficial incisional surgical site infection by thesurgeon or attending physician.

istic regression was based on statistical significance or

mwa

RAwtCht1

Samodot

aLeb

wpn(h

hwg

Tw

V

P

L

B

E

C

L

B

EEA

W

S

A

*S

Ta

V

AS

D

P

P

*fS

31Vol. 212, No. 1, January 2011 Sehgal et al Colorectal Resection and Surgical Site Infection

arginal statistical significance in the univariate analysishile also considering the medical relevance of the vari-

bles as potential predictors.

ESULTStotal of 191 diabetic patients were identified who under-ent colorectal resection from June 2000 to June 2009 at

he Milton S Hershey Medical Center by the Division ofolon and Rectal Surgery. Eight patients, none of whomad a wound infection, had no glucose readings recorded inhe chart and were therefore excluded. Out of the evaluable83 patients, 28 (15%) patients developed an SSI.

Patient characteristics not significantly associated withSI on univariate analysis are listed in Table 2 and includedge, sex, diagnosis (colorectal cancer, diverticulitis, inflam-atory bowel disease, or miscellaneous) and the presence

f a preoperative stoma. Interestingly, the type of homeiabetic regimen the patient used, ie, diet controlled versusral hypoglycemics versus insulin, did not have any rela-ionship to the development of an SSI.

Operative measures that were not significantly associ-ted with an SSI included the use of perioperative steroids,OS, the use of blood products, estimated blood loss, av-rage core body temperature, and the use of preoperative

able 2. Patient Characteristics Not Significantly Associ-ted with Surgical Site Infection on Univariate Analysis

ariableSSI

(n � 28, 15%)No SSI

(n � 155, 85%)p

Value

ge, y*, (mean � SE) 63.6 � 12 65.6 � 12.8 0.15ex, n (%) 0.67

Female 12 (43) 64 (41)Male 16 (57) 91 (59)iagnosis, n (%) 0.29Colorectal cancer 10 (36) 70 (45)Diverticulitis 6 (21) 33 (21)Inflammatory bowel

disease 9 (32) 25 (16)Other 3 (11) 27 (18)

reoperative stoma, n(%) 0.14

Yes 3 (10) 6 (4)No 25 (90) 149 (96)

reoperativehypoglycemicregime, n (%) 0.38

Diet only 4 (14) 24 (15)Oral hypoglycemics 14 (50) 94 (60)Insulin 10 (36) 37 (25)

Statistical analysis performed using Mann–Whitney test. Statistical analysisor remaining variables performed using Fisher’s exact test.SI, surgical site infection.

owel preparation. Furthermore, whether the procedure A

as done as an elective or emergent colectomy, the patient’shysical status (ASA grade), and wound classification didot correlate with SSI. The approach to the colectomyopen or laparoscopic) and the mode of skin closure alsoad no influence on development of SSI (Table 3).Glucose values at all 3 time points were significantly

igher in the patients who developed an SSI than thoseho did not (Table 4, Fig. 1). The largest difference inlucose values between patients with or without SSI was

able 3. Operative Measures Not Statistically Associatedith Surgical Site Infection on Univariate Analysis

ariable SSI (n � 28)No SSI

(n � 155)p

Value

erioperative steroids, % 0.77Yes 14 13No 86 87

ength of stay, d*,(mean � SD) 8.03 � 3.75 7.58 � 4.93 0.55

lood transfusion, % 0.28Yes 27 19No 73 81

stimated blood loss, mL*,(mean � SD) 388 � 445 290 � 353 0.11

ore body temperature, oC*,(mean � SD) 36.15 � 0.76 35.96 � 1.31 0.35

ength of operation, min*,(mean � SD) 161 � 81 164 � 83 0.77

owel preparation, % 1Yes 67 66No 33 34

lective, % 81 82 0.43mergency, % 19 18SA grade, % 1II 32 32III 61 58IV 7 10ound classification,% 0.72II 93 91IV 7 (III–IV) 9

ide of colectomy, % 0.75Left 79 74Right 21 26

pproach, % 0.25Open 89 87Laparoscopic to open 7 2Laparoscopic 4 11

Skin closure 0.33Clips 78 85Sutures 10 8

Statistical analysis performed using Mann–Whitney test.tatistical analysis for remaining variables performed using Fisher’s exact test.

SA, American Society of Anesthesiologists; SSI, surgical site infection.

sibv

sBsdwcfwt

DTfOvsatm

o2difttbsteSt

Fch4(sg

TI

F

BP

S

D

*fB

TIM

GDABS

L

TTi

4

GMS

32 Sehgal et al Colorectal Resection and Surgical Site Infection J Am Coll Surg

een with the immediate postoperative glucose. Also, thismmediate postoperative glucose was the highest glucose inoth the SSI and non-SSI patient groups, with the highestalue (211 mg/dL) seen in the SSI group.

Measures other than glucose that were significantly as-ociated with SSI on univariate analysis included higherMI, use of prophylactic antibiotics greater than 24 hours,

toma creation and the use of pelvic or intra-abdominalrains (Table 5). When these significant univariate factorsere put into a multiple logistical regression model, glu-

ose levels, the use of drains, and prophylactic antibioticsor more than 24 hours remained significantly associatedith SSI; BMI and stoma creation missed absolute statis-

ical significance (Table 6).

ISCUSSIONhe aim of this study was to identify potential risk factors

or SSI in diabetic patients undergoing colorectal resection.ur results concur with those from other studies in the

ascular, cardiothoracic, and orthopaedic literature thathow high perioperative glucose levels to be associated withn increased risk of developing SSI.6 The American Diabe-es Association recommends glucose levels be less than 200g/dL perioperatively in order to avoid SSI. However, in

able 5. Significant Factors Associated with Surgical Sitenfection on Univariate Analysis

actorsSSI

(n � 28)No SSI

(n � 155) p Value

MI, kg/m2*, (mean � SD) 35 � 8.6 32 � 16.5 0.03rophylactic antibiotics used

�24 h, % 0.006Yes 70 45No 23 55

toma creation, % 0.003Yes 67 41No 33 59rains, % 0.04Yes 60 42No 40 58

Statistical analysis performed using Mann–Whitney test. Statistical analysisor remaining variables performed using Fisher’s exact test.

able 4. Glucose Measurements (Normal 70 to 120 mg/dL)imenterval, h SSI (n � 28) No SSI (n � 155) p Value

0 – 6 211.11 � 13.62 g/dL 182.39 � 4.34 g/dL 0.030 – 48 176.77 � 7.41 g/dL 160.44 � 2.64 g/dL 0.0018 – 96 167.85 � 5.03 g/dL 155.50 � 2.43 g/dL 0.012

lucose measurements are mean � SE. Statistical analysis performed usingann–Whitney test.

SI, surgical site infection.

MI, body mass index; SSI, surgical site infection. p

ur cohort, most of the glucose levels measured were below00 mg/dL in the patients who still developed SSIs. Byividing our patients’ glucose measurements into 3 time

ntervals, we sought to identify if there was a specific timerame during which hyperglycemia may be more critical inhe development of an SSI. In both groups of our patients,hose with and without SSI, the immediate postoperativelood glucose was the highest of the 3 time points mea-ured, suggesting this may be a key time point for interven-ion. Other studies in cardiac patients have also shown thatlevated intraoperative blood glucose is associated withSI.8 Our data suggest, however, that the American Diabe-es Association recommendation of 200 mg/dL may still be

igure 1. Postoperative glucose values. The following blood glu-ose values were recorded: the first postoperative glucose (0 to 6ours), all glucose between 0 and 48 hours and all glucose between8 and 96 hours. Patients with and without surgical site infectionSSI) were compared. At all time points the mean glucose wasignificantly higher in the patients with an SSI. Orange bar, SSI;reen bar, no SSI.

able 6. Significant Factors Associated with Surgical Sitenfection on Multivariate Analysiseasurement p Value

lucose 0.02rains 0.05ntibiotic �24 h 0.02ody mass index, kg/m2 0.08toma creation 0.07

ogistic regression using enter method and incorporating all variables with

�0.1 on univariate analysis.

til

ldttaistlpwlsca

tmsshpUsdututa

sfhmceita

suphst

dfsasgwpoogf

gsewocfsttaltas

cclsargBafwnwc

tssrio

33Vol. 212, No. 1, January 2011 Sehgal et al Colorectal Resection and Surgical Site Infection

oo high, and lower blood sugars should be sought to min-mize SSI, because the majority of the measured glucoseevels were below 200 mg/dL in those who developed SSIs.

The majority of the studies in the literature that haveooked at perioperative glucose management have beenone primarily in the critical care and cardiac surgical set-ing. In 2003, Furnary and colleagues8 published a studyhat looked at 3,554 diabetic patients undergoing coronaryrtery bypass grafting. All patients were treated with eitherntraoperative subcutaneous insulin or with continuous in-ulin infusion for hyperglycemic control. Observed mor-ality with continuous insulin infusion was significantlyower than with subcutaneous insulin. Furthermore, betterostoperative glucose control was also shown to decreaseound infections.8,9 Data from the critical care and cardiac

iterature may be able to be extrapolated to other types ofurgery, but large studies looking at perioperative glycemicontrol in patients with diabetes outside of the critical carend cardiac surgical realm are still limited.

One might postulate that the more severe the diabeteshe higher the risk of SSI. The type of hypoglycemic regi-en can be used as a surrogate for disease severity. One can

uggest that diet controlled patients might have the leastevere diabetes, those on oral hypoglycemic medicationsave intermediate DM severity, and those with insulin de-endent diabetes have the most severe category of disease.sing these severity categories, our data did not show a

ignificant difference between these groups in the risk ofeveloping an SSI. Ideally, hemoglobin A1c levels could besed to assess longer term glycemic control. However dueo the retrospective nature of this study these values werenavailable. It appears, however, that absolute periopera-ive glucose control is more important than severity of DMs reflected by hypoglycemic regimen.

It can be argued that elevated glucose may be an earlyign of postoperative infection as opposed to a causativeactor. Certainly it is possible that during our 48- to 96-our time point, patients who would ultimately get an SSIay have already started to develop such an infection that

ould have been reflected in increased blood sugars. How-ver, the highest blood glucose levels were found in themmediate postoperative period within 6 hours of opera-ion, which would be very unlikely to have been caused bywound infection at such an early time point.Apart from glucose control, other measures that were

ignificantly associated with development of SSIs were these of pelvic or intra-abdominal drains and the use of pro-hylactic broad-spectrum antibiotics for longer than 24ours. Prolonged use of both of these measures has beenhown to be associated with SSI in other patient popula-

ions.10,11 Both prolonged antibiotic use and placement of i

rains may have occurred because there were intraoperativeindings that made the operating surgeon concerned aboutubsequent infection, suggesting that it was the intraoper-tive finding, as opposed to the drain or antibiotics them-elves, that increased the SSI risk. However, in the SSIroup of patients with drains, only 1 of 18 of those drainsas placed in an abscess cavity. Therefore, the drains in ouratients were primarily placed for reasons other than obvi-us findings of infection at the time of operation. Theveruse of drains may increase the risk of SSI due to retro-rade contamination along the drain, creating a pathwayor subcutaneous soft tissue contamination.

All of our patients received appropriate anerobic andram-negative antibiotic coverage on induction of anesthe-ia. Using a single dose of preoperative antibiotic is consid-red acceptable in operations lasting up to 3 hours andithout gross soilage during the operation.10 However, usef antibiotics beyond 24 hours can be indicated in moreomplicated operative procedures in which infection isound. Increasing the length of operation has also beenhown to be a risk factor for SSI.10,12 In theory, prolonginghe length of the operation could potentially increase bac-erial exposure to the wound and reduce efficacy of thentibiotic within the tissues, suggesting the need for pro-onged antibiotics.10 However, in our study, the duration ofhe procedure was not a significant variable relating to SSI,lthough this could be due to the somewhat limited sampleize.

Obesity not only makes the procedure technically morehallenging and potentially longer, but there is also an in-reased amount of fat or dead space in the wound that mayead to an increased risk of developing a wound infection. Atudy conducted by Itani and associates13 looking at thessociation of BMI and the type of antibiotic prophylaxis,evealed that in patients undergoing elective colorectal sur-ery there was a higher incidence of SSI in patients withMI � 30 kg/m2 irrespective of the type of prophylacticntibiotic used. Obesity is already a well known risk factoror development of SSI.14,15 In our study, increased BMIas significant on univariate analysis and approached sig-ificance on multivariate analysis. This suggests thateight loss in the obese patient before purely elective pro-

edures may improve overall SSI risk in the DM patient.Creation of a stoma was also significantly associated with

he development of SSI in this study. These results echoeveral results from other similar studies.16 Creation of atoma may increase the risk of developing an SSI for 2easons. First, the stoma may have been created because ofntraoperative findings that could lead to an increased riskf infection, and second, the potential for enteric contam-

nation of a fresh abdominal wound from the exposed

bac

tvtwqtgtTl

taopSod

A

SAADC

R

1

1

1

1

1

1

1

34 Sehgal et al Colorectal Resection and Surgical Site Infection J Am Coll Surg

owel lumen. Stoma creation should be used judiciously,nd bowel opened or matured only after the incision islosed and protected from possible contamination.

This study has some limitations. Though suggestive,ighter glucose control was not definitively shown to pre-ent wound infection rates. Larger multicenter prospectiverials will be necessary to prove this, but would be worth-hile in order to obtain data that can then be used inuality control programs. However, based on our findings,he American Diabetes Association recommendations oflucose levels being kept less than 200 mg/dL may still beoo high in DM patients undergoing colorectal resection.ighter control of perioperative blood glucose levels may

ead to lower SSI rates.To summarize, in diabetic patients undergoing colorec-

al surgery, to minimize wound infections, prophylacticntibiotics should be stopped within 24 hours, and the usef drains should be kept to a minimum. Weight loss beforeurely elective colorectal surgery may also improve SSI risk.tomas should be used only when necessary and tight peri-perative blood glucose controls may improve the risk ofeveloping SSI.

uthor Contributions

tudy conception and design: Sehgal, Figueroa, Poritz, Koltuncquisition of data: Sehgal, Figueroanalysis and interpretation of data: Bergrafting of manuscript: Sehgal, Poritz, Koltunritical revision: Stewart, McKenna, Poritz, Koltun

EFERENCES

1. Center of Disease Control and Prevention. Surgical site infec-tions. Available at: http://www.cdc.gov/ncidod/dhqp/FAQ_SSI.html. Accessed October 10, 2010.

2. Smith RL, Bohl JK, McElearney ST, et al. Wound infection afterelective colorectal resection. Ann Surg 2004;239:599–605; dis-cussion 605–607.

3. Chen S, Anderson MV, Cheng WK, Wongworawat MD. Dia-

betes associated with increased surgical site infections in spinalarthrodesis. Clin Orthop Relat Res 2009;467:1670–1673.

4. Ogihara H, Takeuchi K, Majima Y. Risk factors of postoperativeinfection in head and neck surgery. Auris Nasus Larynx 2009;36:457–460.

5. Trussell J, Gerkin R, Coates B, et al. Impact of a patient carepathway protocol on surgical site infection rates in cardiotho-racic surgery patients. Am J Surg 2008;196:883–889.

6. McConnell YJ, Johnson PM, Porter GA. Surgical site infectionsfollowing colorectal surgery in patients with diabetes: associa-tion with postoperative hyperglycemia. J Gastrointest Surg2009;13:508–515.

7. American Diabetes Association. Standards of medical care indiabetes–2008. Diabetes Care 2008;31:S12–54.

8. Furnary AP, Gao G, Grunkemeier GL, et al. Continuous insulininfusion reduces mortality in patients with diabetes undergoingcoronary artery bypass grafting. J Thorac Cardiovasc Surg 2003;125:1007–1021.

9. Golden SH, Peart-Vigilance C, Kao WH, Brancati FL. Periop-erative glycemic control and the risk of infectious complicationsin a cohort of adults with diabetes. Diabetes Care1999;22:1408–1414.

0. Tang R, Chen HH, Wang YL, et al. Risk factors for surgical siteinfection after elective resection of the colon and rectum: asingle-center prospective study of 2,809 consecutive patients.Ann Surg 2001;234:181–189.

1. Drapeau CM, D’Aniello C, Brafa A; Italian Group on SurgicalInfections in Plastic Surgery, et al. Surgical site infections inplastic surgery: an Italian multicenter study. J Surg Res 2007;143:393–397.

2. Cruse PJ, Foord R. The epidemiology of wound infection. A10-year prospective study of 62,939 wounds. Surg Clin NorthAm 1980;60:27–40.

3. Itani KM, Jensen EH, Finn TS, et al. Effect of body mass indexand ertapenem versus cefotetan prophylaxis on surgical site in-fection in elective colorectal surgery. Surg Infect (Larchmt)2008;9:131–137.

4. Lynch RJ, Ranney DN, Shijie C, et al. Obesity, surgical siteinfection, and outcome following renal transplantation. AnnSurg 2009;250:1014–1020.

5. Mehrabi A, Fonouni H, Wente M, et al. Wound complicationsfollowing kidney and liver transplantation. Clin Transplant2006;20:97–110.

6. KonishiT,WatanabeT, Kishimoto J, Nagawa H. Elective colon andrectal surgery differ in risk factors for wound infection: results of

prospective surveillance. Ann Surg 2006;244:758–763.