~~~

Cl-

~~

Yaper

~

~GuideJinestor Clinical Care

.--, c,. ,,- .'

~a!:elP..Pag~"MO; John "1. Á, 8Qhnen,..A:10;), RaYrt!°nd ,F/e!ch~r,~O;A!bert T,McManus; PhO;, : JosephS. So/Qmk{n¡ MO;. Oietmar H,Wittmaf:ln,MO,PhO

.Prophylactic administration of antibiotics can decreasepostoperative morbidity, shorten hospitalization, and re-duce the oyera" costs attributablé to infections. Principiesof prophylaxis include providing effective levels of antibi-otics in the decisive interval, and, in most instances, limit-ing the course to intraoperative coverage only. Use in TheNational Research Council clean contaminated operationsis appropriate and, in many instances, has been provenbeneficial. Antibiotic prophylaxis is also indicated forcleanoperations, such as those involved with insertion of pros-thetic devices, that are associated with low infection riskand high morbidity. Extension of antibiotic prophylaxis toother categories of clean wounds should be limited to pa-tients with two or more risk factors established by criteriain the study of the efficacy of nosocomial infection control(SENIC) because the baseline infection rafe in these patientsis high enough to justify their use. Cefazolin (or cefoxitinwhen anaerobic coverage is necessary) remains the main-stay of prophylactic therapy. Selection of an alternate agentshould be based on specific contraindications, local infec-tion control surveillance data, and the results of clinical tri-als. Newer criteria for determining the risk of "site infec-tion" (wound and intracavitary) are in evolution and marlead to modification of these recommendations over thenext several years.

(Arch Surg. 1993;128:79-88)

20% cost to the total hospital bill. Nationwide, the cost of thisexcessive hospitalization is likely to be more than $1.5 billionper year? Ehrenkranz et al3 have proposed that antibioticprophylaxis for patients undergoing cesarean section couldresult in a net national annual saving of $9 million for thiscategory of operation alone.

On the other hand, inappropriate and indiscriminate useof prophylactic antibiotics mar increase costs through un-necessary drug use, requisite laboratory monitoring, andselection of resistant organisms. These undesirable effectsnecessitate more expensive infection control measures andantibiotics. As with other adjunctive measures in surgery,the use of antibioticprophylaxis is not a substitute fOl goodinfection control practices, appropriate patient prepara-tion, good judgment, an adequate operating environment,and good technique. Their proper use is not a substitute forexcellent patient careo

This article was developed by the Antimicrobial AgentsCommittee and was approved by the Executive Commit-tee of the Surgical Infection Society as a set of guidelinesfor selection and use of prophylactic antibiotics for surgi-cal wounds. It does not deal with endoscopic operations,image-directed percutaneous procedures, or endocarditisprophylaxis.

P rophylactic administration of antibiotics represents theirmost common use in surgery. Designed to reduce the

incidence of postoperative infection, antibiotic prophylaxismar reduce overall costs by avoiding the expenses attribut-able to infectionsand by shortening hospital stay. Haley etaP have indicated that a surgical wound we-mon prolongsho~pitalization for approximately 1 week and adds a 10% to

DEFINITION OF ANTIMICROBIAL PROPHYLAXISFOR SURGICAL WOUNDS

Rigorously defined, prophylactic antibiotics are thosegiven to patients before contamination or infection has oc-curred. A broader, more practical definition includes clin-ical situations in which infection or contarnination isalready present, primary treatment of the infection is sur:'gical, and anticipatory antibiotic administration mainlyserves to minimize postoperative wound infection (eg, pa-tients with acuteappendicitis or acute cholecystitis). Initia!antibiotic therapy is anticipatory, presumptive, or empírico Adiagnosisof infection or possible infectionhas been rnadebut is not definitive, and there are no culture data. If sim-ple acute appendicitis (cholecystitis) is found at operation,additional doses of antibiotics may be avoided or givenina yery short course to minimize subsequent wound infec-tion (prophylactic). For findings ofcomplicated appendid-tis;-contmuationof antibíotic therapy is -'aÍ111ed at theunderlying infection and is therapeutic, not prophylactic.

Antimicrobia/ Prophy/axi~age et al 79

Accepted for~iion July 25, 1992.' -~

From the DepartmentSof Surgeiy, The University ofTexas HealthSci-enceCenter, San Antonio{Dr Page); University ofToronto, Ontario (DrBohnen); Universityof South Afabama, Mobile (DrFletcher); Microbi-ology Branch, US Army Instituteof SurgicaJ Research, Fort Sam Hous-

ton~ Tex(DrMct.:1al)~s);U!)iy.ersity qfC!ncinn~ti Co~lege?f Med(cJne..Ohlo (Dr SoJomkm); and Medlcal College ofWlsconsm, Mllwaukee (DrWittmann):,'..-'"';: C'",,!,.',; ,'"' '\"c;"'c :'

Reprint !~uests tQ !?epartment,of Su!gery,.. The Uniyersity of. TexasHealth Sclence Center,. 7703 Floyd Curl Dr} SanAntonio, TX 78284-7842 (Dr Page).

Arch Surg~V~1128..'an~arY) 993'.':,

~~~~-i;:~

2 ()l¡

.." c\+","'"r

;:""..'."".""'4""\"""""""" ..,.,y,,\,,","" E"""."""" ,!"'"'",," ""XCand;kií1étic: "io e'rties"ofthedril -;'~and:on: th'éresults:of

~ng~Qh" atient\

fact9:rs;:' EgriÓl~~~~i,~~";menti,Tabre: L: surnmariiessome fac1ófSthilf iijCiéased the

sificanon:scheme: hasser:voo asthebasis for recornmend;.:;

, -Table1.-Factors Associated With Increased Risk

; of Postoperative Infection,

IntraoperativeFactors

PerioperativeF:ldors

Patient ~J

esofage, clong p.reo~ratlve Intraoperatlve 'c.h6spitalization contamination

:"lutrition No preo~rative Lengthy operation

'.",;-,shower(, 'Ea.rly~having Excess.ive"', ofslte electrocautery'c" ""

problems, Hairremoval Foreign material"i"" ,Priorantibiotic Wound drainagetherapy ,

,Dx,ejj}i~ High hematocrit, ,

, wound flu "ldcc, .."'!-" c.otei~fection

Epinephfinewoun'", c"..' , ,.."'.o", ' injection'"c,"' ; ",", " , 'Icosterold 'c Intraoperatlve '

~rapy:,!';,;" hypotension, '"--'ation Massivetransf.'-, " , Al h 11 ' .

, co o , ,,lmation I "__L_-

""

iei,rradiation ,",c C .'

~~

~err

~

)ciatedliabetes

~~~

bblOtiCS ~nd"lt$reducbon wrln;tne) use oí. ~tiblOtiCS m, '

wotmd, c'6" ., c, c"c".,'" "7 '"'mañ~ C'ementandtRé:~se'of"ro' ñ lactic antibiQti&~lf iS'.,

testedi'a!: credictiveimooel "orstratifiéatiori: of NRC cleancc'{?: ,'""'pcJi""c',,'-""""cc,, ,,"" c".,c"'"" c"

woiliids mi2 communi"Ros ifalS: From'¡:'C6'Uléition oí.cC ,'Cc '~"C"'c'c"'Co/"é""' Ec",c\ v, ,""R, P... "

risk oí. in-fecgon,Slmilarly,~aley~t al,6 m the 1985 study oí. the ef-ficacyof"rios&o~al..inféCtiori control (SENIC), identifiedfourfudependent #dadditiveriSk factors forpostopera-tivewound inféCtion.. These factors are operations on fue

hours,and the

pr~~~~, 9fat)E}~~t, th!ee m~lcal~la~~ses",..'Jable;:?c~n:'pares :~e,~ection rat~coí.t~e, various

SI;:NI~~d~C classifications froma popUlation of morethan 59 QqQ patien,ts~ Inspectionfudicate,s a wider separa-tionof ris~qf infection by the SENI~ classificati~~ than bythe NRC classification. Two or more SENIC risk factors

c ~ .,

abrogatefhevalyeoftpeNRC classific~tion. Patients withNRCcleanwounds andtwo or more nsk factors have an

, c' c'

inf~onrate fhatrangesfrom 8% tQ 15% and, thus, qual-ify for antibiotic prophyIaxis. These patients accounted forappro~at~ly 10%of al1 cleanwp'un~s ~nd about 6% ~fthe popUlationatlarge.Thus, conslderation of the c.ümbl-natio~ of NRCclassificatil;>n and the SENI~ risk categoriessee~app~op.riate~s ~ w.de for administering prophy-lacti~"a~tiblotics,.f~r dl!ectíng wound management, andforconducting~-hospital epidemiolo~c research.

~ ccomb~tioni~ a particularly important consider-ationbecause~f r~ent~nth~i~mfor extendfug prophy-lac~tic~tibiotic~, toca. yane'!;y ~f.NRCclean operationsthatconstitute 59% to 6Q% oí. al1,p'pera#ons performed.R~ently~ Pla~~t~ ~~ proposedben~fit from the prophylac-tic,~e.,<?:f ..cefo~cig!W patients undergofug a varie'!;yofbreast operations.,and fu those undergofug inWnal herni-orrhaphy. Althoughfhe numerical tr~pd.,andthe authors'futerpretation favor this position, they have been criticized

Iyp,

th

Recent opel"Ch'ronic,inflan nexacrnoropn~ne

skin preparation

ior

511

.National Research Council Wound~fication C~*Table 2.

Criteria

~

ted l00/of

(20'

Elective (not urgen(or emergency), primarily closed; no acute inflammation or transectlon LgastrointestinaJ, óropharyngeal, genitourinary, biliary, or tracheobronchialtracts; no

'technique break (eg; elective inguinal herniorrhaphy), 'Urgent or emergency case that is otherwise "cJeanff;elective, controlled opening of

gástrointes,inal,.oropharyngeaJ, biliary, or tracheobronchial tracts; minimal spillage and/oníinor techniqué break; reoperationvia"cJéan" iriCisjonwithin 7 days; blunttiauma,

..'. .intact skm;negatlve exploratlonjeg, vagotomy~nd pyloroplasty)c'

Acute;' ribnpurulent iil'flammation (~ote absence'6fpurulence); m~¡6r;echnique break ormajorspillfron'í hollow organ;penetrating traúína'<4 h old; chronic open wounds to be

grafte~orcCovered {eg, acute, nonperforated, nongangrenousappendicitis)"' ; "" Cc' "',PurÜlenté or'ab$cess;preoperativeperfo~ationof gastrointestinal¡oropharyngeal, biliary, 01

tracheobronchialtracts; penetrating trauma >4 h old (eg, ~rlorated áppendicitis with

abscess)\~;,;::'.;,;'i~i}§i --~c."'."'"'"'

)1ftY(40'

~O6)973;10-Page: etal

Antimjcrobial

ProphylaxJ

* Modifled from Ann Siiig. ': é ".,;, ¡,'!tWoJn({f~fectionráteafter Cruse et~I(Arch Surg

Table(3:~Distribution'of Patients and Infectionscby SENI~sk FáCtfu~c.c':~'andNátiona.l ResearchcCouncil (N RC) WoúridClassification*c,cc"

Infectión by NRC Wound ClassificationSENIC, % ofPatients With

InfectionSENIC

Risk FactQrsSENIC¡ % ofAl! Patients

SENIC,. %of Group, ,With Infectión

CreanContaminated Cóntaminated

'.c.N/A(c::;"

Clean &rtyN/A

..,"6.7

10.9,'"18.8

10'1;,,¡~;,29.',"is"::c

v1

4b!32

16

5

3.9.11',;(\8.4 8.9',:,~""'"

~n{S

Yo of

~nts Wit::tion

4:In

nfection control; and N/A, mutually exclusive criteria. The...ounds accounted for 55% (t) of all patients and 39% (*) c§).

Patients with O 5ENIC risk factors accounted for 46% () SENIC risk factors was 1 % ti). SENIC risk factors include

---fied from Haley et al.6 SENIC indicates study of the ~fficacy ofnosocomiare examplestatements indicating how toread this table: The NRC Cleanpatients. The infection!atefor patien~ with NRC clean wounds was 2.9'Its and 10% ('1 )of all infected patients. The infection rate for patients witloperations, operations lónger than 2 hours, three or more associated draglth Contaminated or Dirty wounds cannot have O SENIC risk factors. Patie

1I1

for combining groups with nonhomogeneous risk, forproviding a liberal definition of infection, and for citing arate of infection for breast operations that is higher thanreported

by others.8 Mostof the issues have to do with theinability to stratify risk within the broad group of NRCclean

operations. Until more definitive studies are avail-able, the use of prophylactic antibiotics in patients withNRC clean wounds should be limited to those with two ormore risk factors or those with potentially serious morbid-ity or mortality from an infection at the operative site.

Clean Wounds for Which the Use of ProphylacticAntibiotics Is Validated

In some operations, the use of prophylactic antibiotics isllStified in spite of a theoreticalIy smalI risk of infection.Jost of these operations necessitate insertion of prosthet-c

materials. Infection in these instances results in either;reat morbidity or mortality (eg, insertion of cardiacralves, prosthetic joints, and prosthetic intravascular¡rafts).

Similarly, antibiotic prophylaxis is usualIy used in:lean operations involving the central nervous system andor those involving cardiopulmonary bypass, whether orlot a prosthetic device is inserted. Neither the risk factorsiescribed

by Ehrenkranz nor those established by the;ENIC classification have been systematicalIy evaluated in

hese patients with high morbidity and low risk of infec-ion. It is likely, however, that many manifest two or morejENIC riskfactors. In general, these groups are character-zed by advanced age (a usual marke:of associatedmed-calproblems) and prolonged operatlons.

SpectrumThe microbiologic characteristics of both the operative

site and the hospital environment should influence thechoice of antibiotics. The usual flora encountered duringsurgery and the organisms responsible for postoperativeinfection have been carefully documented in a number ofgeneral and site-specific studies. While both infiuence pro-phylactic antibiotic selection, coverage is directed prima-rily at those organisrriS that cause postoperative irlfection.These flora are usually some, but notall, of the endogenousorganisms encouritered. AntiInicrobiaJ prophylaxis in op-erations that do not violate a hollow viscus or mucosa needto cover only gram-positive skin flora, primarily Staphylo-coccus epidermidis and Staphylococcus aureus. Operations in-volving the gastrointestinal tract, the genitourinary tract,and the hepatobiliary system, as well as some pu1ri1onaryoperations, must cover both skin flora andadditiónal site-specific organismS, Operations that feature póstóperativeinfectionswith both aerobic and anaerobic florashould usean inclusive antibiotic regimen. Choosing the most limit-ed effectivespectrum is difficult, but important. Precisionin this effort avoids unnecessari1y broad coverage iiridminimizes risk for the patient and the environment (over-growth with opporlunistic bacteria and fungi and devel-opment of resistant strains).

Risk of Antibiotic ToxicitrThe potential toxicity of antibiotics represents an im-

portant risk of prophyIaxis. The least toxic, effectiveanti-biotic regilnen should be chosen. Table 4 represents someimportant toxicities of commonly usedprophylactic anti-biotics. In general, the cephalosporin group manifests ac-ceptable toxicities, the majority of which are reversiblé af-ter withdrawal oÍ,. the drug. The mostsignificant andfrightening toxicity, an anaphyIactic reaction to a ~:-lactamantibiotic\, (penicilliri,.! cephalosporini' carbapenéiri;: or

Ant;m;crobíalProphylaxí~agé etal 81

Clioice of Antibiotics ,The choice of anantibiotic for prophylaxis is muItifacto-lL It depends on the operation to be performed, the or-!nisms that must be incIuded, the kinetics and toxicitiesthe drug, and its performance in proFerir designed

nical trialS;

:h Surg-Vot 128, January1993

development of multiply resistant strains of S epidennidisin response to prophylaxis with rifampin, rifampin andnafcillin, or cefazolin sodium. Furthermore, they docu-mented coloniiation of adjacent patients and hospital staff.Sanders and Sanders14 cite potential evolution of "stáblyderepressed" ¡3-Iactamase-produc:ingmutants as a reasonto limit the use of third-generation cephalosporins forprophylaxis. Clostridium difficile coloniiation mar compli-cate prophylaxis, even with a single-dose strategy}5

~

monobactam), can often be avoided by obtaimnga carefulhistory. lf the patient reportsa history ofanim:mediatehypersensitivity reaction,analtemateantibi~tic should beselected. Aztreonam, a monobactam antibiotic, does notcross-react ~th penicillin antibodies. lt may providealtemative gram-negative prophylaxiS for patients allergicto ~enicillin.~)f gram-P?sitive ?rga~s~ are the prop~y-lachc target,.. vancomycm orc~da~ycm may be ~onsld-ered.The ~'r~ man s~drome" l~! howeyer, verycommo?after the fu::s,t dose oí. vancomycm,eyen ~hen thedrug 15admi~ster~ ~ppropriately:1O~issmdrome may includeflushing, pruntus, chest pam, musclespas~,orhypoten,.sion.~hese sFptqms shouldnotbe misipterpreted as~nallerglc reaction that preclud~s ~her use ofv~~omycm.

~minoglycosides representtljeother e~d 9fthe t~xicityspectrum. Nephrotoxicity andot9toxicity,both f!~uent"[iththerapeuticc?urses, areextremelyrare as complica-hons of prophylax15 for48 hour,sQrless. Nevertheless, theototoxicity iS irreversible, aminoglycosides are extremelyvaluable therapeutic agents, andseveralerrors commonlymade in the use ofprophylactic antibiotics (see below) maysignificantly increase toxicity. For these reasons, we do not.recommend parenteral aminoglycosides tor routine pro-phylaxis. lf no other agent is acceptable, aminoglycosidesmust be used in strict compliance with the pharmacoki-netic and duration considerations enumerated below.

~,:; c', "{!~;.':; :'

c;~,;, Risk ofChanging Ecology-Changesin flora and the deyeJopment of'resiStantstrainS are possibles~uelae of prophylacticantibiotic ÍlSe.Many rando#ed prospectivestudies have documentedinfectioÍí WithorganismsresiSt'ant. to the antibiotic used,

an adJun~t.Jo?penheartoperations. Archer and collea~es1:13 descnbed:,.,"

'

82¡;Arch Surg-'-Vol\128} Januar'Y:,1993::;'¡","'cc;;',;c: ccc," ""C"cc C Cc"",c "'c5cc.c""'C'j~','c""";;::"'~C""",;$C';.c"C;

PharmacokineticsBurke'sdescriptionof the decisive interval (the first 3

hours afterwounding and/ or contamination) in experi-mental arumalS has been clearly confirmed in clinicalpraCtice}6.1! To be maximally effective, the antibioticshould be present inadequate concentrations at the oper-ative site as early as possible in the decisive interval andfor as long as the wound is apeno lf a systeinic antibiotic(s)is used, intravenous administration of an appropriate dosein the operating room just before the induction of anesthe-sia is practical. This timing provides the interval requiredfor distribution from the central compartment (blooq-stream) to the tissue compartment (wound). "On call" dos-ing is no longeracceptable beca use itmay result in prematureadministration 01 theantibiotic regimen and jnsufficient tissueconcentrations of drug during the decisive intervaV8 lnitialdosing dependson thevolume of distribution, peak levels,plasmacqearance;protein bindmg, and bioávailability ofthe ~ntibiotic r~~,~se~ected.Rep.eat~ dosing intraop-erativelymaybe necessarydependmg on the agent usedand the duration oí the operation.For agents that are rap-idly cleared; repeat dosing ataRinterval two timeS theplaSma half-li!eisap~~?p~a te. lliug: ~d~~ati~~cac-'cording to these tenets assures appropnate tissue levels ofthe antibiotic, arequiSite foreffective prophylaxis.!~,;.f

et al

;,"'c..A'Cc,"-,i~

,"'C"""C"-,,,;Cé;CJ;,;,,-,';Jcc',:éééc,,' """,../"""g"",c".""",",;""""" g"," 'CC'c-/.",

sary, All represent failuré to'adhéiefótnetenetsií'idicated,;"" ":;""""""""coc/¡..above, P~rhapsthe mostegregl?,~S erfQ~~~~~,~ap~r?pp.,¡

ate use of valuable, therapeubc,:agents...,~lth¡ exp~degspectrum and excessively long-ferro adminisf:ratioi1'~ Indis;'cririlinate antibioticuse iscostly, expose$the:" pati'é'i)ts'fQ~adverseeffects,.and'promotes resistante tothe ditigs: Re;¡.' ,. C ...'""centdatasuggest thatmvolvement of chmcal pha~~clst~¡in the overalI PJógraq\ófantibioticéprophyIaxiS, th~'i.isebfi

co~ ~uteriZed re~~~~*~,satellite phaririac~es; ~ndsttong~'pOSltiOns by ,the prof~s~!onaI st~~far~ effective~ co~bat-¡éérng these errors.i2-24: " "

Theshortesteffectiveéóurseofró h Iactic'arttibi6tics~C 'c" ""C jP,;c!?'"~",c"':"c';'¡;ic

¡nIy 1Vc~I~the pa~e~~~; mc~,h~;~pe~a~~fi ;r~?T',~~t~J?ng\alf;pfe a&~~~

!~~r~dI;Ugs fqrshqrter, p~!?C~~~S" !~s~ay amount to

1trac?pe~ati.ve,p:p~d lsone of t~emo~t slgrufic~nt r~ce~thang~~é~an~blohcprophylaxIS and lSgraFaticallydjf~~rentfromthe 2~to48-ho~co.verage prey¡ously recom;'1erid~c!, ~\ I~ ha~p~oved eff~tive Jor op~rations~nvolvingle ga~tro~testinal tract, for orthopedlc operahons, and)r

~esarean section'iind gynecologic procedures. It re-la~ debatable for cardiac procedures!l Extending theJurse of antibiotic proppylaxis to"cover" lines, robes, and1theters is unwarranted.

Alternate StrategiesWhile the foregoing considers primarily the strategy oí

parenteral prophylaxis to reduce wourid infection, othetoptions are available. These iriclude topical prophylaxisand technical approaches. ; -,

Topical prophylaxis has also proved beneficialusing an-tibiotic irrigation or mstillation at the time of wound clo-sure. Conceptually, these techniquesprovide antibiotics invery high concentrations at the site of maximum risk earlyenough in the decisive interval to beofberiefit. Eleven of13 prospective clinical trials reported up to 1977 docu-mented benefit using either "nonabsorbable" antibiotics(mainIy aminoglycoside/bacitracin cómbinations) or anantibiotic likely to be absorbed from the wound(mainlyampicillin).23 This technique has proved most efficient inthe absence of established infection}6 Luminal decontami-nation with antibiotics is designed to lower the risk ofwound infection by reducing the number of organismspresented to the surgical wound. It has been most fre-quently applied to colorectal operations (see below).Effective regimens include anaerobic and aerobic coverageand at least one antibiotic with some systemic overlap. In-tuitively, topical, luminal, and systemic therapy shouldenhance one another, but this advantage is difficult to sci-entificalIy validate in the clinical setting.27,28

Surgical technique, wound management, and overalIpatientcare are of great importance in minimizing the in-cidence of wound infection. Rarely is one aspect of man-agement of singular importance. It is the sum of the partsthat yields favorable results. Delayed primary wound clo-sure around the fifthpostoperative dar is a specific tech-nique for substantialIy reducing wound infection.29,30 Theuse of delayed primary wound closure, in lieu of primarywound closure, is appropriately applied to situations thatfeature excessive intraoperative contamination or estab-lished infection. Success is reported in 80% to 90% of de-layed primary wound closure efforts and failure does notthreaten invasive infection or prolong the course of antibi-otics. In some hospitals, selective application of this tech-roque has reduced the incidence of wound infections incontaminated and di~ wounds to an attack rate lowerthan that observed for clean contaminated wounds.

Other Considerations

Ideally, the prophylactic antibiotic(s) selected should!Ve been proved effective in randomized, prospective,Id

clinical trials. Ad hoc choices, however, are manda-ry in instances of conflicting allergies or other contrain-cationsto

the use ofthe usual drug. In addition, the reg-len chosen shoUld be compatible with the findings frome

hospital' s infection control wound surveillance reportois particUlarly important for hospitals with a high inci-

nce of infection with methicillin-resistant S aureus[RSA) and/or S epidermidis (MRSE). These isolates are'quent enough in some hospitals to represent the"en-mic flora" Staphylococcus and to dicta te the basis for theoice

of antistaphylococcal prophylaxis. While all strainsMRSA and MRSE are resistant to alll3-lactams, they aremore virulent than methicillin-sensitive organisms.

eir majar risk is 3 to 4 days of ineffective 13-lactam ther-y while waiting for microbiologic identification. The~sence of MRSA and/ or MRSE as endemic hospitalra mandates recognition of the problem and the choiceanappropriate (usually more expensive) antibiotic. Onationallevel, this trend is likely to recapitulate the de-'opment of penicillin-resistant staphylococci witnessed:.he late 1950s.me additional consideration in the choice of a prophy-tic antibiotic is avoidance of a drug valuable for defin-e therapy. If a num~er of drugs appear equally accept-e for prophylaxis, one shoUld pick the agent least likelybe used for defirutive therapy. This strategy shouldl~mize selecting organisms resistant to valuable thera-tIC agents.

CasIostis appropriately the last item considered in the choicerophylactic

antibiotics. Among otherwise equal antibi-; in the selection criteria mentioned above, the least ex-;ive

should be chosen. The least expensive is not alwaysjrugwith the lowest procurement costo Total expenses1de the costs of laboratory monitoring, drug adminiscra-:both supplies and personnel),adverse effects! and fail-.f prophyIaxis (ie, wound infection).

SITE-SPECIFICCONSIOERA TIONS ANORECOMMENOA TIONS

Prophylactic antibiotics for specific sites are discussedbelow and summafÍzed in Table 5. These recommenda-tions are similar to others recently published arid arelargely cephalosporin-based.31-33 Cefazolin, the standard ofcomparison, fulfills most of the criteria indicated above, isnot the drug of first choice for definitive treatmentóf anyknown infection, and is relatively inexpensive. Its half-life,

Antím¡crob;ajProphyl.iix;~ageetal 83

Common Errors in Antibiotic Prophylaxis

nmon errors in aritibiotic prophylaxis include choos-le wrong agent, administering the initial clase toa

omitting critical incrao~rative closes in long opera-ur8:-c-VoI128,

January 19~3é

'1

i il ~, ,:;:,C\'i;i¡ii~:'¡~;!"NoncardiacVasrular O~rations'c".'"";,'j1$;¡,¿,,

,J~;;iS' 'featrmcreased when vascwat 'rostl1éses'are,re" uired

ti¡::'}~~tidhar () ~riitibriS arealso at ili8!~as~drisk;1!K~1!&

a "'IK~6b::CoKtfolle¡$fud" Iiffi1t~io"'atientsPWith "üiri;m~í~i8f{~:To'itar tl1'(!:iii"

"'"'c,'~,c P.,.,P, .."",c "é , "c;, '..éC"'CC", ";'~é;'wounds:.'Therewas no dell'\onstiated advantage ofcom-,'.,,' 'C.!,...' -c'", c",'bmed therapy. Prophylacticantiblotics should not beused

forcarotid'endarterectomy orbrachiaIartery repair. n1ereéC \C"'C'cc-c CCC'

ar~~~~~~ie~t data to,p,r?vide a~~~9~e~~at~o~..forextia-anatOffilc cerebrovascular reconsfuichons:,' " .C

:!:':~;;';ié',",'" ",,'C " 'c

infectiori:i[tateg¡...from~:' ro' h laXisY;lri:clude~ofthó nathic'

laxiS isnot iridÍcated fórdentoalveólar"rócooures"exce' t.,"'. ,..,"'J.."","".."""" ,,,,"",: ¡"f"";""""+""'~""C'C"!P:"'J' ,C'" ""'A!"..." Pfor,immüi1osu- "'résséd:"atienfs..~A.'review;óU 92. NRC

siZe:'w~~;,.'to6diffeiei1¿~s..~~:~;i ':;'11:;",;,:,;, .,,:;R, y: 'c,'

!hé~v~~a~I~ p:o~-,

phyIactic ~nqbIOtiC~~~c stand~rq p~actlc~, A,ya~ety oi

slde.~éd. r~s~ti9nf~r)ung ~ancer..T~es~dyby~~~s.et aIfXJprqvid~sthe,best ~ataf?r the e!fic~g ~fantiblotiC pr°I;'~y~a:xIS.COmpa~atiye trlaIs e~aI~atmgtq:urdoses of peruclIIin vscefuroXIme or comparmg one vs SIX clases of cefazoIin haveshown no difference in out~ome. 61,62

Gastroduodelial OperationsThe bacterioIogic condition of the~tomach is a function

of gastric acidity. Conditionsthat decreasé gastric acidityremarkabIy increase gastric flora (primariIy with nasopha-ryngeaI aerobes) and increase' the incidence of post-operative wound infection.63 The overaII incidence c;>fgastroduodenaioperations has decreased since the intro-duction of appropriate concepts of antibiotic prophyIaxis.Most proceduresare now either bariatric, for cancer, or forcomplicated uIcer disease. VirfualIy alI in the Iatter twogroups have been treated with acid-reducing agents.Reports substantiate the vaIueorcephaIosporins and theapplicability of singIe-dose prophyIaxis when measuredby clínicaI outcome and by antibiotic concentrations inbIood, gastric mucosa, and'subcutaneous tissue.64,65

,::' Orlhopedic Operations" ';, ,

The.,fuaj9rity°t pos!operative~~ons afe~ausedbys aureus or S epldennldlS. The ~Jor problem IS that of

prosthetic joint infections. Antibiotics are jusWied in thesepriinary ,clean oren operations because themorbldity as-soci.ate~ ~ith infecti~n ~so ~gh, aIthoughtherisk of in-fection IS low.44 The high infection rafe forsome operations(eg, revisionsof totalknees, ankIefusions,ortotal hIpre-~sio~s) justifies their use.45 CefazolinreducesoveraQinfection to less than 3% (vs control of 5% to 15%) and isthe agent of choice because of low toxicity, low cost, andadequate bone, serum, and soft-tissue leveIs.46 Antibioticcourses extended beyond 48 hours are of no benefit, andthe trend is toward intraoperative coverage only.44,46 Theuse of antibiotic prophylaxis is aIso indiéated for hip-fracrure operations.47 1t appears unnecessary inopenreduction of low-energy closed fractures.48

pata suggest the benefit of prophylactic antibiotics forpatientswith lower'extremity amputation, but thestudiesaré not large enough for det.initive recommendations. Ki-

netic studies andtissueleveIs depict adequate concentra:tionsof penicillins and cephaIosporiris in bone, serum, andsoft tissue, correlating somewhat with transcutaneous ox-ygen pressure measurements.49,so

Neurosurgical OperationsProphylactic antibiotics directed at S aureus and/or

S epider1Jlidis are commonly used in clean neurosurgicalproced ures in spite ofthe absence of definitive randomizedprospective trials. The review by Haines51 of the bestavailable data justifies this approach.Effectiveagents thathave been used include combinations of cefazolin andgentamicin, gentamicin and vancomycin, and vancomy-cin, penicillin,c!Qxac~li~, or piperacillin as single agents.Overall wound infechonrates were 4% to 8% without and

a smgleq?se..r;f,:~~p~t~regardmg antiblotic prophylaxlS f~rc~re~rospma!fl~lq sh1,1:r1t Pfocedur~ are equivocal: An~-

mwhich prosth~tlc m~tega~ are ~erted. "

Biliary TractOperationsMost postoperative infections occur in patients with

posinve hile Cultlires, and most arecaused bygram-negative organisms. The majorityofwound iruections inpatients with negative hile Cultures' are caused by S aureus;Postoperative infections with Enterococcus species are notrafe in patients who receive cephalosporin prophylaxis.Risk stratification for the likelihoodof positive hile culturesby established criteria is sometimes used to select patientsfor prophylactic antibiotics.66,67 Series reporting the inci-dence of positive hile cultures fróm low-riskpatientsgivevalues ranging fi:o~ 10%' to 50%. ~,69Accordingly, prophy-'lacticantibioticsare used with9utstratification bymany,surgeo~.' ;",,'j,,; ,éc ,;

Chol~stectoriiy is among themost common opérations'p~rfo~ed and hasbee~thesu~jecf9fa numberofantibi-otIctrialS. PlacebO-controlled tnalShave proved the value:of cefazolin, and ¿omparative trials have demonstratedthél

tIon.~¿l In addlti~n,toplc~l therapyhas !'een .tested agal~t,systeri1ican tib iotics a ndcombin~. ~o p~~al.and:syste~~,;

, " .c" cies.~ There are no datato'support thepreferentialuse of~;,expanded s pectrtiril cep ha los porins!A ppió Fria te r'&om 'j¡

Prophylactica~ti?iotics:areai.med p~arily. at oralanaerobes and aerob~cco~9~esp~Ially streptococCl.56 They

"C"'¡""f,.:-",!""-/';'é:t~7"""""""~"""T.c,,,"C c"c."

luü'"ñ~Pró 11' lactic antibodies,a commohI");¡sedas'jor

associaf'éd"Witna hí h incldence'ot"'ósfb' 'eranve'mréétion

¡~&~9~~¡:"f9;neWfantioi6tics' and straté" "es.¿Thc~iSK~'.bfósfO erative

,dI~~~~~U~i~t!fl'9%r, fór;"süi1lé aéndicitjS,~,~ Perioeiativea:nti~iotics,"'~-""c,,~'c,," R""pp~""C". , ~., -"-""'C_- E...,"cv;""v"~,,, ,""reduce the'rate'of woÜifd irifé'ctiorito 1% tó 5%; "'c":'.'

Witf\;':i'ÉiiObic' andjanae:robic""'covera 'e' 'rovid'é effectÍve

ccc '. '"C c' .'c g "é' c"

roiúdaz6le,'although deficierit rnitsspectrum, has been;"1'C c ,; .-'" " ..; " ,su~~ess~y use~as. a smgleage~t.~3Toplcal. agentshavealso proved effective prophylaXls lar, sImple acuteappendicitis.26 "," c,

Colorectal Operations

Mechanical bowel preparation decreases fecalbulk butdoes not alter the concentrationof the orgariisms in thestool.77 Extensive study of prophylactic antibiotics in col-orectaloperations has ~elded proponentsof oral (luminal)and parenteral re~ens. In general, North American sur-geons focus on luminal prophylactic antibiotics and em-phasizea reduction in the total number of orgaiúsms po-tentially available for intraoperative contamination.Meanwhile,European surgeons mar employonly sys-temic coverage aimed at ameliorating theconsequences ofany orgaiúsms spilled. Both approaches recognize a high-er risk of operations involving extraperitonealrectum thanof those limited to fue colon arid agree on several princi-pIes: (1) mechaiúcal preparationis essential, (2) antimicro-bialprophylaXisis effective,and (3) gram-positiye aerobesas well as gram-negative aerobes andanaerobes shoUld becovered;.' Clarke et al78 documented reduction oÍ postoperativeinfection from A3% Withmechaiúcal bowey preparationalone to9r.oWith the addition.of tmee ..pr~perative claseso.fneomycln and erythromycm base m a .large VeteransAdministration Cooperative study. This is the only studydemonstrating a reduction in intra-abdominal infection inthe treatment group. Another investigation showed theresUlts of antibiotic preparation of thecolon with neomy-cixi and erytmomycin to be similar toroetroiúdazole anderYthro~yc.in in ~e~ ofoverall infec~on.~ Ina~~ti-~~~t~rT~1 mvolvmg morethan 500 pahents, theaddltionRfp~r~nteral cephalothin tothe luminal antib~oticsshowedc~g;~~f~rence.~5.~%"~s!8~):80A repeat ~ft~ la~~er studydeslgn;substitutingcefoXltin for cephalothm,: mdepen-dently by_tW;o separategroups,~eldedcoriflictingresUlts.

while in theother, its additioriappeared tobeoÍvalue.8J~2:"*~StU:dies that have foCUsedonc' mechamcal piepa:ration.', c '" ..."" ."" c ..a»4 parenteral antiblotics have show» efficacyof~yanety

i(){~ingl~ agents orcombinations providilig broad aerobicimpor.,

"""'""""-" 'C,c,"M""""'C',,'"""'~""~'-'~"~'="""""""'" "",."-Qfiiztteonam-cliÍ\dam 'ciri"or:céfofaxime"á:s a sin le ii' i1t"~!cc,"co c,..,. ,YC"'*'"C" CC,," g C ~

'oéldg!a' similar infectión ráté"ofaboút6%.84.85 " :::7:'

ic iCi:iC, GynecologicOperations'

mybenefit from prophylactic antibiotics. Benefit ismagni-fied in patients wit~~~k!ai~tors efilq1:J~ocioeConomic ~ta-~s; extrem~~ of age~ ope~l~! i~!" .dla~etes;.or preceding~t~entatio~..Postoperati.ve slte infecti~ns mar becaused by a vanety of aeroblc and anaeroblc orgamsms.C i iBacteroides species are usually )he dominant anaerobe.Singledoses of a váriety of cephalosporins are effective.

C c.c .Expa~<jed-spt;ct~ cephalosporms represent no benefit?ve~cefazolinin spiteof.the bacteriology. Use ofcefazolinm lieu of later-generahon drugs mar also reduce theselection of ~-lactamase-producingresistant organisms.90

i

Urologic OperationsThe principIes of antibiotic prophylaxis for patients un-

dergoing nephrectomy and cystectomy ,vith urinary con-duit construction are sufficiently similar to those forabdominal clean contaminated and colorectal operations,respectively,thatthey donot warrant separate discussion.Luminal preparation for cystectomy and ileal conduitshould be enhanced with specific coverage of any organ-ism cultured from the urinary tract. There are, however, nodefinitivepublished data.

Abdominal TraumaPostoperative infections after abdominal trauma vary

with the age of the patient, the organ and number oforgansinjured, and the extent of transfusion required.91 As withappendicitis, useofantibiotics in pénetrating or blunt ab-dominal trauma is initiallypresumptive and necessarilydirected at a variety of aerobes and anaerobes.92 Prophy-lactic therapy is administered only in the Cace of a lapa-rotomy without injury to a hollow viscus.

A number of antibiotic trials comparing a variety ofexpanded-spectrum single agents with aminoglycoside-based combinations (usually including gentamicin andclindamycin) show similar efficacy .?1.93-96 Single agents areas effective as aminoglycoside-based combinations. Ce-foxitin isusually thestandardwith which othér singleagents are'comparedarid, therefore, represents the antibi-otic withwhicli the largest experience has been accrued.Short-duration prophylaxis (12hours) appears as effectiveas prolongedcoverage. 97~

ThemanUSCrlpf was: prepared bythe Antimicrobial Agenfs Com-~tlee~f11eSurgi~1

~ection Society.. c! References ":!

!1..Ha!ey RW, Schaberg 9R, Crossley K.B, Von Allmen so! ~cc;:;o~a.n jE.EXtra charg~and prolongatlon of stayattrlbutable tonosocomlalmfectlons:

al

r~;~',:".,,;C:'~'i;';é~é:"t.'

~

cefamandole, cefazolin,cefurox.ime proplíylaxis íií"ópénhé¡¡rtsurgery ..An"

¿'~6:Mit!dz!nsl<ilaxls for

37. Gelfand MS; S.immons BP¡ Sc~oettle P; et al. CefamandoJeversusce.,fonicid proplíy!axis incárdióvástularsurgery: ápróspectivéstudy~Ann Thd':racSiJrg..

38. Szilagi DE; Smitlí RF, Efliott JP¡ yrandeci,é MP~lnfection inarterialrei,constructión witlí synthetié grafts:AnnSurg: 1972;186:321-333.."'.'"":';'39. Goldstorie Ji. Moore: WS,1nfection in"vaséularprostlíesés:"cllnicalmanifestations and surgicalrnanagernent tn1¡ Surg,'1974;128:22S-233.;

"40.landreneau MO,. RajúS:lnfectionsaftereleétlve bypass surgery forlower 1imb isclíemia: tlíe influenceof preoperativétrariscutaneous arteriog:

Japlíy,Surg~iY; 1981;90:956-961..:* r41. Ourlí~m JR, Malone JM, Bernlíard VM. Tlíe Impact ofmultlple oper-

atióris on tlíe importance ofarterialwaWcultures.j Vasc Surg; 1987;5:160-169;:. c" ""':';'f,C.-;" J::":.."::;':c;..

42. KaIser AB, Cfayson KR, Mullíerm Jl Jr, et a). AntlblotlC proplíylaxls mvascular surgery. Ann Surg.1978;188:283.:189. :c' , '! ...

43; Pitt HA, Pós~ier RG,MacGowari WAl" FranklW. Proplíylacti¿ anti-biotics in vascularsurgery: topical, systemic, or botlí?Ann Surg. 1980;192:356-364. ,

44. Nelson Cl. Prevention of sepsis. C/ín Orthop. 1987;222:66-72.45. Taylor GJ, Bannister GC, Calder S. Per(operativewound infection in

elective ortlíopaedic surgery.) Hosp Infett..1990;16:141~147.46. Van Meirlíaeglíe J,Verdonk R..VérsclíraegenG!MynyP, Paeme G,

Claessens H. Flucloxacillin compared witlí ce(azolin il}slíort-term p,ophy-laxis for clean ortlíopedic surgery. Arch Orthop Traúma Surg. 1989;108:308-313. ,

o.. o.. o..47. Tengve B, Kjellander J. Antibiotic proplíylaxis in operations ontroclíanteric femoral fractures.] Bone Joínt Surg Am. 1978;60:97-99.

48. Huglíes SP, Miles RS, littlejolín M, Brown E,ls antibiotic próplíylaxisnecessary for interna/. fixation of low-energy fractures? Injury. 1991 ;22:111-113. o.. o.. o..

49. Jrgensen lN, Andreasen JJ, Nielsen PT, Kondradsen lA, Hartzen SH.Oicloxacillin concentrations In amputation. Acta Orthop Scand. 1989;60:617-620. o..

50. Mars M, Elson KI, Salisbury RT, Robbs JV. Do pre:operative antibiot-ics reaclí tlíe operative field in amputation surgery for periplíeral va~culardisease? a pilot study. S Afr J Surg. 1990;28:58-61. o..

51. Haines SJ. Efficacy of antibiotic proplíylaxis ino..clean neurosurgiéaloperations. Neurosurgery.1989;24:401-405.. o..

52. Geraglíty J, Feely M. Antibiotic prophylaxis in neurosurgery. J Neuro-surg. 1984;60:724-726. o..

53. BullockR,van Oelfen JR; Ketelbéy W, ~einacIíSG. A double-blindplacebo-controlled trial of perloperative proplíylactic antibiotics forelective

o.. neurosurgery. J Neurosurg. 1988;69:687-691, '54. van Ek B, Oijkmans BA, van OulkenH, Mouton RP!Herrnal}s J, van

Furtlí R. Effect of cloxacillin proplíylaxis on tlíe bacterial flora of craniotomywounds. Scand J Infect Oís. 1990;22:345-352. " o..

55. Cartmill TO, al Zalíawi MF, Sisson PR..etal. Five days versusone dayof penicillin as próplíylaxis in elective neurosurgical operations. J Hosp In-fect.1989;14:63-68.

56. Jolínson JT, Yu Vl. Role of aerobic gram-negative rods, anaerobes, andfungi in wound infection after líead and necksurgery: implications foranti-biotic proplíylaxis. Head Neck. 1989;11 :27-29.

57. Friberg O, lundbergC. Antibiotic proplíylaxis in majar líead and necksurgery wlíen clean-contaminated wounds are establislíed. Scand J Infect OísSuppl.1990;70:87-90.

58. Peterson lJ. Antibiotic proplíylaxis against wound infections in oraland maxillofacial surgery. J Oral Maxí/lofacSurg. 1990;48:617-620.

59. Carrau Rl, ByzakisJ, Wagner Rl, JolínsonlT. RoJeof proplíylactican-tibiotics in uncontaminated neck dissections. A!ch Otolaryngol.HeadNeckSurg.1991;117:194-195. , o..;;c:. o..'c;;;c~z':\

60. Ilves R, Cooper JOi TOdd TRJ, Pearson fG. Prospectlve, randomlzed

double-bli~d study ~sing proplíylacticc.ephalotlíi~!or major.. elective, gen-eral tlíoraclc operatlons. J Thorac Card,ovasc Surg. 1981 ;81 :813-817.

61. Krasnik M, Tlíi.is J, Frimoot-Miller N,'Antibiotic proplíylaxis innon-cardiac tlíoracic surgery: a double-blind study of penicillin vs ce-o.. o..furoxime. Scand J /horac Cardíovasc Surg. 199.1;~5:73-76. o..

62.. Olak J, Jeyasinglíam K, Forrester-Wood C, Hutter j, al-Zeeralí M;o..Brown E. Randomized triar of one-dose versussix-dose cefazolin proplíylaxisinelectivegeneral tlíoracic surgery. Ann Thorac Surg. 1991 ;51 :956-958.

63. Gatelíouse O, Oimock F, Burdon OW, Alexander-Williams J, Keiglí-cley MR. Prediction o( wound sepsis (ollowing gastric operation. Br J Surg.1978;65:551-554.. o.. o.. Cc o..

64.PoriesWJ..Va.n .Rij AM: Bur(inglíam BT!Fu!glíum RS, Meellíeim O.P.roplíy(actic cefazolm In gastrtC bypass surgery.Surgery. 19.~1;90:42~-;432..

65; lewis RT, G~all RG!MarienB",ParkM! lIo">:d-~~lt~:VV~ WJegand,fM'. Efficacy and distribution ofsing/e.,dosepreop~!a~v~c.antlb,ot,c proplí~-;/, o..é o..'o..o..c,cCCé o..o..".ceo.. io "",';i' o.. \cc " o..

é2;,WenzeJ RP;Preop~ratlyea!!t!b!ot'!CP.ropllylaxls. NEngHM~,.;.19?2;

:!,!~i;;;,<;:,!t...3.,: EllrenkranZNJ;8'ac~elde.rWC¡ Pfaff Sj; PoppeO,YergOEpKaslow

RA:!nfect!o~S coi!lp'ifat.i ñ? low~rjs~'c~s.areañ~ections i.ri:CQ~1!~ry.¡~hospj"taJs:efficacy ofantlm!c.f!?p'a! propllylaxls'.'~m J O~tl;t~yn~9Ih990;.162:,

" 337-34;;1,; ,";4, Natlonal Academy of Sclences, Natlonal Researcli CounclJ¡ OIVIslon of

fections: tlle)nfluence of ultraviolet irradiationoñthe'operatlng'room and of

",:i::'5:Ellrenkr~nz'Nj, SúrgicaJ wound infectionoccurrente i;'clean 'opera-

tions: risk stratificátiori for interlidspital compariSón~; Am} Med, 1981;70:

!';"é~.,:,;, ,;'" "";','"',"':6. ;H~leYR~,Cu.J\ier.OH,. M?rg~n~M,~llite~, E~or!.JG, Hoo!ón T.M,

1985;121:206-215fc!'";;"",f.'" !,".",' .';"'(\C'",::,.",! c7; Platt R, Zaleznik OF, Hopkiñs CC, eral, Perioperativeantibiotic'

, ,propllylaxis forllerniorrllapllyand breast surgery. N Engl J Med.1990;322:153-160.' "'!",,! C ': "'""...,

8. Antibiotic própllylaxis for Ilérniorraplly and breastsurgery: N Engr¡Med; .1990;322:1884-1886; lettersto tlle editor:,: ;'" ,

9; Saxon A, 8ea1lGN, Rohr AS, AdeJman OC. Imrhediatehypersensitivi~reactions to beta'-lactanl antibiotics. Ann Intern Med.1987;107:204-215.

10. Wallace MR,.Mascola JR, Oldfield EC: Redman syndr<?me: incidence,etiology, and propllyJaxis. 1.!nfect Oís, 1991;164:.1.180-1185.

.11:Conte JE Jr, Collen SN, Roe B8, EJaslloff RM. Antib.iotic propllylaxisandcardiacsurgery: a prospec_tivedouble-blind comparison of single-dose'versusmultipleC:dose regimens. Ann Intern Med. .1972;76:943-949.

12. Arcller Gl, Tenenbaum MJc Antibiotic-resistant Staphylococcus epí-dermídís in patients undergoing cardiac surgery. Antímícrob AgentsChemother.1980;17:269-272, ! '

13. Arcller Gl, Armstrong 8C. Alteration of stapllylococcal flora in cardi-ac surgery patients receivingantibioticpropllylaxis. 1. Infect Ois. 1983;147:642-649.

14. Sanders CC, SandersWE. Type I beta-lactamases of gram negativebacteria: interactions witll beta-lactain antibiotics: 1. Infect Oís. .1986;154:792-800c

.15; Priv.itera G, Scarpe/lini P, Ortisi G, NicastroG, Nicolin R, de lalla F.Prospective study of C/ostridíum dífficíle intestinal colonization and diseasefollowing single.dose antibioticP.ropllylaxis in surgery. Antimícrob AgentsChemother.1991;35:208-210.

.16. 8urke JF. Tlle effectiveperiod of preventive antibiotic action in exper-imental incisions and dermallesions. Surgery.1961;50:161-168.

17. Po/k HC Jr, lopez-Mayor JF, Postoperative wound infection: aprospectivestudy of determinant factors and prevention. Surgery. 1969;66:97-.103.

18. Galanduik S, Polk HC Jr, Jagelman OG, Fazio VW. Re-empllasis ofpriorities in surgical antibiotic propllylaxis. Surg Gynecol Obstet. 1989;.169:2.19-222.

19. 8ergamini TM; Po/kHC Jr. Pllarmacodynamicsof antibiotic penetra-tion of tissue and surgical propllylaxis. Surg Gynecol Obstet. 1989;168:283-289.

20. OiPiro lT, Clleung RPF, 80wden TA, Mansberg JA. Single dosesystem.ic antibiotic propllylaxis of surgical wound infections. Am 1. Surg,.1986;152:552-559.

21, Ariano RE, Zllanel GG. Antimicrobial propllylaxis in coronary bypasssurgéry: a critical appraisal. OICP. 199.1 ;25:478-484.

22. Peterson CO, Sllu/tz NJ, Goldberg DE. Pllarmacist monitoring of tlletiming of preoperative antibiotic administration. Am 1. Hosp Pharm. 1990;47:384-386.

23. Oobrzanski S, lawley DI, McOermott 1, SelbyM, Ausobsky JR. Tlleimpact of guidelines on perioperative antibiotic administrarían. 1. Clín PharmTher.1991;16:19-24.

24. Scller KS, 8ernstein JM, Arenstein Gl, Sorensen C. Reducing tlle casIof surgical prophylaxis. Am Surco 1990;56:32-35.

25. Halasz NA. Wound infettion and topicaJ antibiotics: tlle surgeon's di-lemma. Arch Surco 1977;.112:1240-1244.

26. Andersen 8, Bendtsen A, Holbraad l, Scllantz A. Wound infectionsafte¡ appendicectomy. Acta Chír Scand. 1972;138:53.1-536.

27. Bergaminl TM, lamont PM, Clleadle WG, PoJk HC Jr. Combined top-ícal and systemic antibioticpropllylaxis in experimental wound infedion. Am1. Surg, 1984;147:753-756.

28. Seco Jl, Ojeda E, Reguilon C, et al. Combined topical and systemicantibioti¡; propllylaxis in acure appendicitis. Am ISurg. .1990;159:226-230.

29. lowry KF, Curtis GM, Oe.layed suture in tlle management of wounds:analysis of 72.1 traumaticwounds illustrating tlle influence oftime intervalin wound repair. Am J Surco 1950;80:280-287. c

30. lipton S, Estrin J. Kamatlle M~ Haq l8erkowitz S. Surface hydrogen:on concentration as a determinant intiming deJayed cJosúre ofwounds. Surg

,

':;yneroIObstet,I?7~;1.39:1~9-192. .." ",'3.1- Kaiser AB. Antlmlcroblal propllylaxlS msurgery. N Engl1. Med. .1986;

1.15:1129-1.138.

32. Gorbacll Sl. The raJe of cepllalosporins in surgical propllylaxis; I An,"

-\rdJ Su +VoW8 anu jj,,'.;,j,c;'c\:,'

.Add!~ionof ~áre~~~ral:cefo~iti".to regimen ?fo.~910f~alo~rilt'~n~:a!andomlz~ prospectlve s209-211,.';:.'; :::¡83¿:M6rris DL;WilsonSR¡Pain tetal) Ac

metronidazoleandcefotaxime/metronidáZole in I"'.

at1t!mi!;f~bial. prop!lx'axismust!ncfudegram-poChemother.1990;25:673-678;.c' "

;(.84.: Rooo.lico.G; Pul~S,BlandinoG..etal.Col

inder-Willial;iatedwith 1:

kéighléy MRB¡)érativefjnd!n~

al parison of aztrective colorectalsul{ecover;JAnti~

',-iu!tivariateSép~is:8ij~

'6;e

LJ.

lo~'~at!Q~;J¡lg!¡-I:!s~factQrs.

and ~.st.operatIV~ S.epSIS .I!) p~~I-:n~ undelP!l.lalYt!~go~rat!,ops~hll~ recelvlng a P!op~y.!agl<; .~!1~!blo~lc;,W~!

)!I~l)d$l!rg!l;:aUnfectl<?nStudy;qroup; Br J,5urg, 1989;7~:374;-37?:;'O~,Str~~h~.~.CJl;~lackJ,~°:'Y.!S. SJA, Watef"rort/l] A. Prophyl~ct!C::cOS)haz911.fí:~a!nst,wo!!!1~SfPSIS af!er<;~olecystectomy, BMJ,1971;1:12

;senA, Kahn'.". .Iectlve operatll12-114:. .Icidence

of inlrith,antibiotics

~,. ." ..C' ., ., ".-

¡8s: Raa,navi!¡ 1;>1.!, tfesselfeldt p.. Pedersen T,Zacha!i.ian~enOH.Nq¡!.,ffe<:íof topical ampjcillin prophylaxis ini

)f th~coIQ!1::9i~~fufJJ:;Su'8 Gynecolpbstet.19.89;168; 86: Mugford M, Klngston J, Chalmers 1; Reduclng me Iion'aftercaesa:rean.s~ction:

implications of prophylaxis '

1ósp(tal.'r,~u~f~~[51J;"1~8?;,299.:) 003-1006. .''. 87¡ Ford LC;Hammll HA; Lebherz TB;Cost-effectlvl

:f

úsf<?t C~~!eál1s~jorlsonC, Duff Po Antib:spectrumagent neceitts DH, Hillier Sl, Es15 prédictors of post-(prophylaxis.Obstet

veátíd randc99Ó;107:32772:SarrMC¡bt1csin

high

exténded.

89;\""deliverv c

990;76:343':341nital tra<;t isolatEIg women recer

>phy)axis lor CE)bstet Gynecol.:h DA. Upper ~i infections amc

Kj. féy H,t.cal páti.e

anti

ud)

!mseAro

Hc lylaxGrol

~ , -..91. Nichols Rl, Smith jW,Klein DB, et al. Risk of infectionng

abdOminal trauma. !\I Englj Med. 1984;311:1065-1070.92. Thadepa)li H;GorbachSl, Norsenj, BroidoPW, Nors

I.bdominal trauma, anaerobes, andantibiotics. Surg Gynecol

BauerT, VI!rforated ap¡:1989;209:3C

Di

111.

Iticenteren), t-Obste

mlzea sruoy 01 moxalaaam versus:ting abdominal trauma. Surg Gyne.94.

Nelson RM, Benitez PR, Newell MA, Wilson RF. ~.penetrating abdominal trauma. Arch Surg. 1986;1295.

O'Donnell V, Mandal AK, Lnu MA, Thadepalli 1-lamycin and gentam

ingle litdo-er

Nichols!rative ni dence

alual

~wis RT

9:99-103,

colon?(JCanjSL MR, Dellin¡In

of prevenl. Arch Surg

P. Wertz ~

a trial ó;urg.19

Jerts PL

A, Chafar:>minal injDphyl;

hoetz

99. DemetriadesD, Lakhoo M, Pezi~Intibiotic prophylaxis in penetrating at:efoxitin.

Injury. 1991 ;22:20-24.

leal:-PageAntim!crobiaJ

Prophylaxi~8, Jar