IiAJJ

O

M

R

C

IcIcc

ptp

M

Wu

RnS

vTN

3

©A

Reconstructive Urology

ncisional Surgical Site Infectionn Kidney Transplantationntonio Ramos, Angel Asensio, Elena Muñez, Julian Torre-Cisneros, Miguel Montejo,ose Maria Aguado, Federico Cofán, Jordi Carratalá, Oscar Len, andose Miguel Cisneros

BJECTIVES Incisional surgical site infections are common bacterial infections in kidney transplantation. Thepurpose of this study was to determine the incidence, timing, etiology, and risk factors forincisional surgical site infections.

ETHODS We performed a prospective study that included a population of 1400 consecutive patients(58.4% males) who underwent kidney transplantation in Spanish hospitals pertaining to theRESITRA research network.

ESULTS A total of 55 patients developed 63 episodes of incisional surgical site infections. Median timefrom transplant to incisional surgical site infections was 20 days (range, 2 to 76 days). Allinfected patients recovered from incisional surgical site infections. The most frequently isolatedpathogens were Escherichia coli (31.7%), Pseudomonas aeruginosa (13.3%), Enterococcus faecalis(11.6%), Enterobacter spp. (10%), and coagulase-negative staphylococci (8.3%). Diabetic pa-tients had an increased risk of incisional surgical site infections (7.5%, P � 0.013). We usedseveral different regimens of antimicrobial prophylaxis. None were found to be associated withan increased risk of incisional surgical site infections. The use of sirolimus was associated with anincreased risk of incisional surgical site infections (7.4%, P � 0.018).

ONCLUSIONS Diabetic patients, and those who received sirolimus-based immunosuppressive regimens,showed an increased risk of developing incisional surgical site infections after kidney

transplantation. UROLOGY 72: 119 –123, 2008. © 2008 Elsevier Inc.

tRftiHlptnoDt3i(cdwtmwwwat

ncisional surgical site infection (ISSI) remains animportant problem after kidney transplantation(KT).1,2 The immunosuppression required after KT

ould have an influence on the incidence or severity ofSSI.2 Antibiotic prophylaxis has been proven to de-rease the risk of ISSI in KT, but the optimal antibiotichoice has still not been defined.3–7

Few studies have analyzed ISSI etiology in KT.8,9 Theurpose of this study was to determine the incidence,iming, etiology, risk factors (specifically the antibioticrophylaxis), and prognosis for ISSIs after KT.

ATERIAL AND METHODS

e performed a prospective study that included all consec-tive patients who underwent deceased-donor whole liver

ESITRA (Network for the Study of Infections in Transplantation) is a researchetwork consisting of several Spanish public hospitals, with funding granted by thepanish National Health Service.

From the Infectious Diseases Unit, Hospital Universitario Puerta de Hierro, Uni-ersidad Autónoma de Madrid; and RESITRA (Network for the Study of Infections inransplantation)-GESITRA (Infection in Transplant Study Group), REIPI (Spanishetwork for the Research in Infectious Diseases), SpainReprint requests: Dr. Antonio Ramos, San Martín de Porres 4, 28035 Madrid, TF

c4 913445000, 34 646 829 554, Spain. E-mail: antrammar@yahoo.esSubmitted: July 22, 2007, accepted (with revisions): November 8, 2007

2008 Elsevier Inc.ll Rights Reserved

ransplantation in 11 Spanish hospitals pertaining to theESITRA research network (Network for the Study of In-

ections in Transplantation) between August 2003 and Sep-ember 2005. RESITRA, which was founded in April 2003,s a research network financed by the Spanish Ministry ofealth. It includes 16 transplant centers and 3 reference

aboratories and prospectively includes all solid organ trans-lant and hematopoietic stem cell transplant recipients fromhese hospitals. Each hospital has a research team (doctor,urse, and technician) who manage the research tools. Annline database and a serum, plasma, microorganism, andNA bank are used as research tools. Pretransplant, peri-

ransplant, and follow-up (days 0, 7, 14, 30, 60, 90, 180, 270,60, and 720 after transplantation) data are prospectivelyncluded in the online database, as well as all infectionsdiagnostic workup, clinical presentation, therapy, and out-ome) and rejection episodes, always according to the pre-efined definitions, based on universally accepted criteria,hich are published on the RESITRA website (www.resi-

ra.retics.net/web/index.asp).10 Antibiotic prophylaxis is ad-inistered 30 minutes before surgery and for 48 hours after-ard. Data collection is performed through PDF e-forms,hich are sent to an SQL-server database located in theebsite. Data exploitation is carried out through managerialnd statistical databases generated from the SQL-server da-abase, after a validating process managed by each hospital

oordinator.

0090-4295/08/$34.00 119doi:10.1016/j.urology.2007.11.030

DWdtmdiwfcoi

(lgasa

oMCaamaaDs

RApomTpthc

optw1pdstata

1m(fn

yatyiI(twepa

ro2NPlti

cmoaclpvaTpc

ay

1

efinitions of Infectionse adopted Centers for Disease Control and Prevention Criteria

efinitions for Surgical Site Infection, which were slightly modifiedo include incisional wound infections that developed up to 3onths after surgery.3,11,12 Incisional surgical site infection was

efined as an infection occurring within 30 days of surgery thatnvolved skin or subcutaneous or deep soft tissues at the surgicalound site, with at least one of the following: purulent discharge

rom the incision, organisms isolated from an aseptically obtainedulture of fluid or tissue, or one or more features of infection (painr tenderness, swelling, redness, heat, or fever) with a diagnosis ofnfection by the surgeon or physician.

The following variables were assessed as risk factors for ISSIs:1) pretransplant variables, including age, gender, diabetes mel-itus, and underlying disease that elicited KT, elective, or emer-ency surgery and previous transplantation; and (2) operativend posttransplant variables including duration of transplanturgery, red blood cell transfusion, immunosuppression regimen,nd antibiotic prophylaxis regimen.

We compared continuous variables using the Student t-testr, when a normal distribution could not be assumed, theann-Whitney test. Categorical data were compared using thehi-square, Chi-square for trends, or Fisher exact test. We usedlogistic regression model to evaluate variables found to be

ssociated with ISSI by univariate analysis. We selected a finalodel by a backward strategy from a maximal model including

ll variables associated with a p value less than 0.2.13 Allnalyses were performed with Stata SE version 8.0 software.ifferences at two-tailed p values less than 0.05 were considered

tatistically significant.

ESULTStotal of 1400 patients underwent KT during the study

eriod; 219 patients received a second transplant (15.6%)wing to rejection. A toal of 819 patients (58.4%) wereen, with an age range of 26 to 71 years (mean, 51 years).he predominant underlying diseases were glomerulone-hritis, 311 patients (22.2%); renal polycystosis, 201 pa-ients (14.3%); diabetic nephropathy, 131 patients (9.4%);ypertensive nephropathy, 105 patients (7.5%); andhronic tubulointerstitial nephropathy, 78 patients (5.6%).

Fifty-five patients developed 63 episodes of ISSI. The riskf developing at least 1 ISSI was 3.9% (55 of 1400) peratient, and the incidence of ISSI was 4.3 episodes per 100ransplanted patients. Median time from transplant to ISSIas 20 days (range, 2 to 76 days). Of the infected patients,2 (22%) developed renal failure. Every case received em-irical and/or directed antibiotic treatment plus woundrainage. One patient required reoperation to resect exten-ive necrotic tissue from the wound site. All infected pa-ients recovered from ISSI. We detected 229 episodes ofcute rejection in 200 patients (16.3 episodes per 100 pa-ients). Sixteen patients (1.1%) died during the first yearfter transplantation.

We recovered 63 organisms from 50 episodes of ISSI,1 of which episodes were polymicrobial infections. Theost frequently isolated pathogens were Escherichia coli

31.7%), Pseudomonas aeruginosa (13.3%), Enterococcusaecalis (11.6%), Enterobacter spp. (10%), and coagulase-

egative Staphylococci (8.3%) (Table 1). C

20

We examined risk factors for ISSI by univariate anal-sis, including 6 pretransplant variables and 4 operativend posttransplant variables. Mean age in infected pa-ient and noninfected counterparts was 53.1 and 50.3ears, respectively (P � 0.172). Diabetic patients had anncreased risk of ISSI (7.5%, P � 0.013) and had secondSSI episodes more frequently (33%) than other patients9%, P � 0.058). There were no relevant differences inhe ISSI causative pathogens in these patients comparedith those of nondiabetic patients. We found no differ-nces in ISSI by gender, urgent or elective indication,revious renal transplantation, duration of surgery, ormount of blood or plasma transfused (Table 2).

We used several different antimicrobial prophylaxisegimens. The most frequent regimens included the usef cefazolin (n � 658), amoxicillin/clavulanate (n �49), and third-generation cephalosporin (n � 221).one were associated with an increased risk of ISSI.atients with ISSI who had received antibiotic prophy-

axis with amoxicillin/clavulanate presented with infec-ion resulting from to Pseudomonas aeruginosa, with pos-tive cultures in 4 of 11 cases (36%, P � 0.059).

The most frequent immunosuppressive regimens wereomposed of triple therapy with steroids, mycophenolateofetil and calcineurin inhibitors, cyclosporine (n � 643),

r tacrolimus (n � 153); steroids, mycophenolate mofetil,nd sirolimus (n � 97); double therapy with steroids andalcineurin inhibitors (n � 72); and steroids, mycopheno-ate mofetil, and basiliximab (n � 64). A total of 1150atients received mycophenolate mofetil, 45 of whom de-eloped ISSI (3.9%, P � 0.448). The use of sirolimus wasssociated with an increased risk of ISSI (7.4%, p � 0.018).here were no relevant differences in the ISSI causativeathogens in these patients compared with those who re-eived other immunosuppressive regimens.

Using a logistic regression model based on the vari-bles identified by univariate analysis, multivariate anal-sis revealed that diabetic patients (odds ratio, 2.5; 95%

Table 1. Organisms recovered from incisional surgical siteinfections after kidney transplantation

Microorganism Isolate(n � 63)

E. coli 20 (31.7)P. aeruginosa 8 (13.3)Enterobacter spp 6 (10)A. baumanii 3 (4.8)Klebsiella spp 2 (3.3)P. mirabilis 1 (1.6)M. morganii 1 (1.6)S. maltophilia 1 (1.6)E. faecalis 7 (11.6)SCN 5 (8.3)S. aureus 3 (4.8)S. viridans 1 (1.6)S. agalactiae 1 (1.6)Bacteroides spp 1 (1.6)Candida spp 3 (4.8)

I, 1.3 to 5.0; P � 0.009) and those who received an

UROLOGY 72 (1), 2008

i2r

CUiwen(nba2oi

piiwd

sosn

wuasGht

iabpgc

wwd

oIiirrt

U

mmunosuppression regimen with sirolimus (odds ratio,.4; 95% CI, 1.2 to 4.9; P � 0.016) had an independentisk of developing ISSI.

OMMENTrine spillage after bladder opening during surgery and

mmunosuppressive drugs could increase the risk ofound infections in patients receiving a KT.2 How-ver, the ISSI rate in KT is usually low and its prog-osis is generally good as observed in this series3.9%).1–3,12,14 –17 Rapid advances in surgical tech-iques and medical treatment during recent years haveeen associated with a progressive decline in ISSI ratesnd severity.17–22 Most ISSIs occurred during the firstto 3 weeks, as noted in previous reports, but could

ccasionally appear later, particularly in cases of deepncisional infections or fungal ISSIs.11,15,23

The most striking finding in this study was the highercentage of ISSI caused by gram-negative bacilli, includ-ng Pseudomonas spp.17 In contrast, a gram-positive predom-nance was observed in series describing the first experiencesith KT (in the 1970s and 1980s) and among patients whoid not receive antibiotic prophylaxis.7,9,18

ISSI infections in KT caused by Staphylococcus spp.uggest that endogenous skin flora are the infecting in-culum.3 However, contamination cannot be excluded inome of our ISSI cases that are attributed to coagulase-

Table 2. Incisional surgical site infection (ISII) by potentia

ISII

VariableAll patientsMaleFemaleDiabetes mellitusMain diagnosis

GlomerulonephritisPolycystosisDiabetic nephropathyHypertensive nephropathyTubulointerstitial nephropathyLupic nephropathy

Type of surgeryEmergencyElective

Kidney retransplantAntibiotic prophylaxis

CefazolinAmoxicillin/clavulanateThird-generation cephalosporinGlycopeptide AztreonamQuinolonesGycopeptide/third-generation cephalosporin/

Amoxicillin/aztreonamCarbapenems

Ampicillin

Sirolimus

CI, confidence interval.

egative staphylococci. t

ROLOGY 72 (1), 2008

The relevant role of those gram-negative bacilli thatere observed could be related to contamination fromrine during the bladder anastomosis, or to improvedntibiotic prophylaxis activity against staphylococcalpecies.2,11,17,24,25 Colonization with nonfermenterram-negative during the period of hospitalization oremodialysis sessions could have occurred before theransplant surgery.26

Fungal infection resulting from Candida spp. occurredn 3 patients (5.4%). Fungi are rarely implicated in ISSIssociated with conventional kidney surgery15,21 and maye related to prolonged hospitalization and immunosup-ression in KT patients.17,21 Although ISSI in KT has aood overall prognosis, cases resulting from Candidaould be more serious.17–21

Increased age has been related to different problemsith the surgical wound.11 However, although patientsith ISSI were older than noninfected KT recipients, theifference did not reach statistical significance.Diabetic patients have an increased risk of ISSI, as

bserved in previous studies, and frequently present withSSI relapses during the post-transplant period.1,11,16,24 Its not known whether inadequate glycemic control dur-ng the perioperative period could be related to a higherate of surgical site infection in these patients.27 It is alsoemarkable that other known ISSI risk factors, such ashe duration of the operation or kidney retransplanta-

factors.

n/No. (Risk in %) Odds Ratio (95% CI) P-Value

/1400 (3.92) (3.0–5.1)8/818 (3.4)7/578 (4.7)

0.86 (0.7–1.3) 0.237

2/160 (7.5) 2.0 (1.2–3.3) 0.013

7/311 (5.5) 1.2 (0.8–1.9) 0.3416/201 (3.0)9/131 (6.9)6/105 (5.7)

2/78 (2.6)1/19 (5.3)

0/15 (5.6) 0.662/1385 (3.97)9/219 (4.1) 1.05 (0.6–1.9) 0.880

0/658 (4.6) 1.1 (0.7–1.5) 0.8362/249 (2.1)6/221 (2.7)3/64 (4.7)1/39 (2.6)0/36 (0)2/26 (7.7)1/22 (4.5)0/20 (0)0/19 (0)

1/144 (7.6) 2.28 (1.7–4.7) 0.015

l risk

Risk

55221

1

55

31

1

ion, were not detected in this study.24

121

taKcnnibsawotr

dHowss

rpapr

sTipwpwswtnef

nmflnictto

CDid

Asam

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

1

Although cefazolin has been considered a suitable op-ion for antibiotic prophylaxis in KT,4 no prophylacticntibiotic regimen has been definitively established forT.5–7 An antibiotic prophylaxis regimen that did notover many pathogens isolated in ISSI cases (e.g., gram-egative nonfermenter bacteria or resistant cocci) wasot associated with an increased risk of ISSI. This could

ndicate that the antibiotic prophylaxis regimen may note the most important factor in ISSI risk. Some authorsupport the use of local antibiotics in the wound irrigantlone, whereas others perform KT with no antibioticshatsoever.3,7,23 It seems that factors such as the qualityf the surgical technique, or the development of post-ransplant urine leakage, could have a more importantole in the development of ISSI in KT.19

The results of this study do not corroborate with aefinitive recommendation for antibiotic prophylaxis.owever, cefazolin could still be considered a suitable

ption for antibiotic prophylaxis in KT.4 Institutionsith a high rate of ISSI owing to resistant pathogens

hould adapt their antibiotic prophylaxis regimen to theirpecific bacteria sensitivity patterns.24

Considering the pathogens isolated in this study, weecommend that the empirical treatment regimen in KTatients with ISSI include an agent with broad coveragegainst gram-negative bacilli. The specific regimen ap-lied in each institution should take into account theesistance profile of the aforementioned bacteria.28

Patients who received sirolimus-based immunosuppres-ive regimens presented with a higher incidence of ISSI.his feature has previously been observed and related to

mpaired wound healing owing to the antiproliferativeroperties of sirolimus on fibroblasts.13,29 However, thereere no differences in the causative ISSI pathogens inatients who received sirolimus compared with patientsho did not. Some transplant centres prefer not to use

irolimus immediately after transplantation but choose toait until the wound is healed, particularly in cases with

he concomitant use of steroids.29 Changes in the immu-osuppression regimen, resulting from acute rejectionpisodes, have also been identified as an independent riskactor associated with SSI.24

One limitation of the present study is that we didot analyze obesity, cigarette smoking, and duration ofechanical ventilation.30 In addition, the possible in-

uence of ISSI on the duration of hospitalization wasot studied. Nevertheless, these results provide an

nsight into the risk of ISSI in relation to patientharacteristics, immunosuppression regimens, and an-ibiotic prophylaxis which could motivate new studieshat lead to a better understanding of the pathogenesisf ISSI in KT.

ONCLUSIONSiabetic patients and those who received sirolimus-based

mmunosuppressive regimens have an increase risk of

eveloping incisional surgical site infections in KT.

22

cknowledgments. The authors thank Isabel Millán for as-isting with statistical analysis, and Martin Hadley-Adams forssisting with the English language and preparation of theanuscript.

eferences1. Alangaden GJ, Thyagarajan R, Gruber SA, et al: Infectious diseases

alter kidney transplantation: current epidemiology and associatedrisk factors. Clin Transplant 20: 401-409, 2006.

2. Humar A, and Matas AJ: Surgical complications after kidneytransplantation. Semin Dial 18: 505-510, 2005.

3. Stephan RN, Munschauer CE, and Kumar MS: Surgical woundinfection in renal transplantation: outcome data in 102 consecutivepatients without perioperative systemic antibiotic coverage. ArchSurg 132: 1315-1358, 1997.

4. American Society of Health-System Pharmacists. ASHP Thera-peutic Guidelines on Antimicrobial Prophylaxis in Surgery. Am JHealth Syst Pharm 56: 1839-1888, 1999.

5. Capocasale E, Mazzoni MP, Tondo S, et al: Antimicrobial prophy-laxis with ceftriaxone in renal transplantation: prospective study of170 patients. Chemotherapy 40: 435-440, 1994.

6. Tillegard A: Renal transplant wound infection: the value of prophy-lactic antibiotic treatment. Scand J Urol Nephrol 18: 215-221, 1984.

7. Midtvedt K, Hartmann A, Midtvedt T, et al: Routine perioperativeantibiotic prophylaxis in renal transplantation. Nephrol DialTransplant 13: 1637-1641, 1998.

8. Sussman M, and Russell RB: Infection after cadaveric renal trans-plantation. Proc R Soc Med 65: 471-473, 1972.

9. Tillegard A: Renal transplant wound infection: the value of prophy-lactic antibiotic treatment. Scand J Urol Nephrol 18: 215-221, 1984.

0. RESITRA: Definiciones RESITRA. Available: www.resitra.net/web/index.asp.2004. Accessed September 15, 2006. RESITRA net-work, 2004.

1. Humar A, Ramcharan T, Denny R, et al: Are wound complicationsafter a kidney transplant more common with modern immunosup-pression? Transplantation 72: 1920-1923, 2001.

2. Horan TC, Gaynes RP, Martone WJ, et al: CDC definitions ofnosocomial surgical site infections, 1992: a modification of CDCdefinitions of surgical wound infections. Infect Control Hosp Epi-demiol 13: 606-608, 1992.

3. Hosmer DV, and Lemeshow S: Model-building strategies andmethods for logistic regression. In: Hosmer DV (Ed): Applied Lo-gistic Regression. New York, Wiley and Sons, 1989, pp 82-134.

4. Ferraresso M, and Berardinelli L: Nosocomial infection in kidneytransplant recipients: a retrospective analysis of a single-centerexperience. Transplant Proc 37: 2495-2496, 2005.

5. Pourmand G, Pourmand M, Salem S, et al: Posttransplant infec-tious complications: a prospective study on 142 kidney allograftrecipients. Urology 3: 23-31, 2006.

6. Osca JM, Alonso M, Pallardo L, et al: Surgical wound infection inrenal transplant recipients. Actas Urol Esp 17: 176-180, 1993.

7. Kyriakides GK, Simmons RL, and Najarian JS: Wound infectionsin renal transplant wounds: pathogenetic and prognostic factors.Ann Surg 182: 770-775, 1975.

8. Muakkassa WF, Goldman MH, Mendez-Picon G, et al: Woundinfections in renal transplant patients. J Urol 130: 17-19, 1983.

9. Judson RT: Wound infection following renal transplantation. AustN Z J Surg 54: 223-224, 1984.

0. Koyle MA, Ward HJ, Twomey PA, et al: Declining incidence ofwound infection in cadaveric renal transplant recipients. Urology31: 103-106, 1988.

1. Wai PH, Ewing CA, Johnson LB, et al: Candida fasciitis followingrenal transplantation. Transplantation 72: 477-479, 2001.

2. Schweizer RT, Kountz SL, and Belzer FO: Wound complications inrecipients of renal transplants. Ann Surg 177: 58-62, 1973.

3. Mallat SG, Aoun M, Moussalli A, et al: Cutaneous aspergillosis in

a renal transplant recipient. J Med Liban 52: 111-114, 2004.

UROLOGY 72 (1), 2008

2

2

2

2

2

2

3

U

4. Dantas SR, Kuboyama RH, Mazzali M, et al: Nosocomial infectionsin renal transplant patients: risk factors and treatment implicationsassociated with urinary tract and surgical site infections. J HospInfect 63: 117-123, 2006.

5. Takeyama K, Matsukawa M, Kunishima Y, et al: Incidence of and riskfactors for surgical site infection in patients with radical cystectomywith urinary diversion. J Infect Chemother 11: 177-181, 2005.

6. Abdulrahman IS, Al-Mueilo SH, Bokhary HA, et al: A prospectivestudy of hemodialysis access-related bacterial infections. J InfectChemother 8: 242-246, 2002.

7. Guvener M, Pasaoglu I, Demircin M, et al: Perioperative hypergly-

cemia is a strong correlate of postoperative infection in type II

ROLOGY 72 (1), 2008

diabetic patients after coronary artery bypass grafting. Endocr J 49:531-537, 2002.

8. Surucuoglu S, Gazi H, Kurutepe S, et al: Bacteriology of surgicalwound infections in a tertiary care hospital in Turkey. East Afr MedJ 82: 331-336, 2005.

9. Valente JF, Hricik D, Weigel K, et al: Comparison of sirolimus vs.mycophenolate mofetil on surgical complications and wound heal-ing in adult kidney transplantation. Am J Transplant 3: 1128-1134, 2003.

0. Singh D, Lawen J, and Alkhudair W: Does pretransplant obesityaffect the outcome in kidney transplant recipients? Transplant Proc

37: 717-720, 2005.

123