C Surgery, Vol. 4, No. 5, pp. 570-572, 1996 Copyright Cl 1996 The International Society for Cardiovascular Surgery Published by Elsevier Science Ltd. Printed in Great Britain PII: SO%7-21O9(96)OOO1-4 0967-2109/96 $15.(30 + 0,00 VASCULAR REVIEW Conservative surgeryfor aorticgraftinfection J. J. Earnshaw GloucestershireRoyalHospital,GreatWesternRoad,Gloucester GLI 3NN, UK Surgeryfor aortic graft infectionis a majorchallengeoften characterizedby the need for ingenuityandimprovisation. Traditional treatmentis bytotalgraft excisionand extra-anatomic bypass. /n situ replacementofthe infectedgraft usingeitherautogenoustissue or antibiotic- impregnatedDacronis effectivein selectedcases but it is not clearwhen such conservative treatment maybe employed.Graftexcisionand thoroughdebridementof infectedtissue are important,whichever techniqueisused.Itwould‘seemunwiseto perform in situ reconstruction unlessthe remainingoperativefieldis free from contamination. Whenin situ replacementis selected,a rifampicin-soaked Dacrongraftisthe easyoption,but largestudieswillbeneededto determinewhetherthis is a suitablelong-termalternativeto conventional treatment. Copyright01996 The International Socie@ for Cardiovascular Surgery. Keywords: vascularreconstruction, graft infection ‘For every complex problem there is a solution that is simple, obvious and probably wrong’. Graft infection is the worst disaster which can follow aortic surgery and is fatal in up to 500/. of cases*. Traditional management involves total graft excision and replacement by an axillobifemoral graft through clean tissues. This is a ‘tour de force’ for both surgeon and patient. As in other branches of surgery, present trends for conservatism have led to suggestionsfor a less aggressive (extensive) approach, and published results with a variety of methods have been as good, if not better than historic reports. The problem is that most surgeons encounter aortic graft infection on only a handful of occasions throughout their career. Individual knowledge is based on retrospective reports in the literature, often experienced over a decade or more, from a few large centres. How can surgeons who practise in smaller units, distil the available knowledge for their occasional patient? Incidenceand aetiology Graft infection complicates 1–2’Yo of aortic reconstruc- tions, though the true rate can only be known if patients are followed prospectivelyfor many years. The Based on a lecture given to the Surgical Infection Study Group in Cambridge, 1994 Correspondenceto: Mr J. J. Earnshaw aetiology is usuallyimplantation of cutaneous or enteric organisms at the time of surgeryor direct spread from a groin wound infection; late graft infection secondary to bacteraemia is probably an extreme rarity. The delay in presentation following initial surgery depends on the virulence of the infecting bacteria. Organisms such as Staphylococcus aureus or Escherischia coli usually cause problems within a few weeks or months of graft in5ertion3,4.Less virulent organisms such as Stapbyb coccus epidermidis may colonize a graft for months or even years before complications occur. These organisms may inhabit the area adjacent to the graft in a layer of ‘serum’, preventingincorporation of the graft into host tissues: the bacterial biofilm5. This fascinating occurrence is the subject of investiga- tion into the immunological balance between host defences and invading bacteria. Does, for example, an alteration in host immune status permit bacterial overgrowth and the presentation of late graft infection? It is known that Dacron grafts explanted from symp- tomless patients for reasons other than infection are often contaminated with S. epidermidis6. Diagnosis The diagnosisof overt graft infection is seldom in doubt when the presentation includes complications such as false aneurysmor aortoenteric fistula. Low-grade illness associated with bacterial biofilm may be more difficult to confirm. Perigraft seroma is diagnostic of bacterial biofilm infection. Computed tomography and magnetic 570 CARDIOVASCULAR SURGERYOCTOBER1996 VOL4 NO5