Eur J Vasc Endovasc Surg (2011) 41, 68e75
In Vivo Detection of Staphylococcus aureus in Biofilm on Vascular Prostheses Using Non-invasive Biophotonic Imaging ¨fer b, K. Ohlsen b, G.C. Tiurbe a, C. Bu ¨hler a, U. Lorenz a,*, T. Scha C.-T. Germer a, R. Kellersmann a a
Section of Vascular and Endovascular Surgery, Department of General, Visceral, Vascular & Paediatric Surgery, University Clinic of Wuerzburg, Oberdu¨rrbacher Str. 6, 97080 Wuerzburg, Germany b Institute for Molecular Infection Biology, University of Wuerzburg, Germany Submitted 4 June 2010; accepted 6 September 2010 Available online 12 October 2010
KEYWORDS S. aureus; Biofilm; Bioluminescence; Vascular graft infection
Abstract Objectives: Biophotonic imaging was compared to standard enumeration method both for counting Staphylococcus aureus in biofilm and bacterial susceptibility tests of different graft materials. Design: Prospective, randomized, controlled animal study. Material and methods: Five types of vascular grafts were placed subcutaneously in 35 mice and challenged with bioluminescent S. aureus. The mice were divided into equal groups as follows: group A (polyester), group B (polytetrafluoroethylene), group C and D (two types of silver acetate-coated polyester) and group E (bovine pericardium). Controls were given only the bacteria. The bioluminescence signal of S. aureus, able to predict number of viable bacteria in biofilm without any manipulation, was measured at different time points. Five days postinfection, regular cultures of adherent bacteria on grafts were obtained. Comparative analyses between bioluminescence activity and culture enumeration were performed. Results: The number of viable bacteria on silver-coated prostheses was the slightest, indicating superior bacterial resistance. The density of bacteria on polytetrafluoroethylene and polyester was comparable, with a non-significant advantage for polytetrafluoroethylene. Moreover, bioluminescence detected the number of viable S. aureus in biofilm more exactly compared to enumeration of bacteria. Conclusion: Bioluminescence imaging can be considered a useful tool to characterize susceptibility of any graft material to bacterial biofilm prior to implantation. ª 2010 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. E-mail address:
[email protected] (U. Lorenz). 1078-5884/$36 ª 2010 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ejvs.2010.09.007
S. aureus Biofilm Graft Infection
Introduction The increasing use of alloplastic bypass materials in modern vascular surgery is associated with a variable but definite risk of bacterial infection, which occurs in 1e4% of patients with aortic grafts and in up to12% of patients with peripheral lower extremity bypass surgeries.1 Prosthetic vascular graft infection (PVGI) causes substantial morbidity and mortality, increased length-of-stay in hospital, and a high amputation rate.1,2 Its treatment also imposes a considerable financial burden on any health-care system. The prevention of PVGI is a great challenge in clinical practice. Although the aetiology of PVGI in individual cases is often unknown, intra-operative contamination with bacteria is thought to be the major cause of early graft infection, followed by wound infection and haematogenous bacterial spread.3 Haematogenous and lymphogenous seeding may occur early or late after prosthetic graft implantation. Intracavitary graft infections (involving aorto-iliac, aortofemoral or iliofemoral grafts) tend to present late. The interpretation of clinical PVGI data is sometimes hampered by the small number and heterogeneity of patients, differences in the underlying microorganism, and the variable clinical course of the infection. Staphylococcus aureus is the leading cause of PVGI and the capacity to adhere and to form multilayered communities, termed ‘biofilm’ on the surfaces of implanted prostheses appears to be one of the major causes of persistent infection following surgery.4 To better understand and control infection on vascular prostheses, preclinical model systems should be used to determine the properties of different materials in resisting bacterial biofilm.5 These systems should closely simulate the in situ conditions for each graft, while at the same time providing reproducible, accurate results. In the past, the most common method to determine the extent of PVGI required the sequential sacrificing of the study animals to count the number of bacteria at a certain time point. Such an invasive approach makes it difficult to follow the kinetics of growth of bacteria within the biofilm. Furthermore, enumeration based on colony-forming unit (cfu) method has not been advised to quantify bacteria in biofilm.6 Recently, a new method of bioluminescence imaging was developed for evaluating bacterial biofilm infection on medical devices.7 This technique can be used for noninvasive continuous monitoring of bacterial density and growth on biomaterials.8 In the present study, we evaluated the ability of this method to monitor bacterial biofilm on vascular grafts in an experimental in vivo model in mice. We compared the susceptibilities of silver-coated, polytetrafluoroethylene, polyester and bovine pericardium grafts to S. aureus biofilm. In addition, the goal of this article was to determine the advantages of the bioluminescence method in comparison to the bacterial enumeration method.
Materials and Methods Animals All animal studies were approved by the Ethics Committee of the University of Wuerzburg. Age- and gender-matched
69 NMRI (Naval Medical Research Institute) mice (Charles River, Sulzfeld, Germany) were studied.
Prostheses material The following five types of vascular prostheses material were used: polyethylene terephthalate, a resin of the polyester family (InterGard, Intervascular, La Ciotat Cedex, France), polytetrafluoroethylene (ePTFE; Impra, Bard, Tempe, AZ, USA), silver acetate-coated knitted polyester (I) (InterGard Silver, InterVascular, La Ciotat Cedex, France), silver acetate-knitted polyester (II) (Silver Graft, Braun, Melsungen, Germany) and bovine pericardium (VascuGuard, Synovis, Minnesota, USA).
Surgical model All the implanted grafts were 1 cm2 in size. Mice of each group were intra-peritoneally anesthetised with xylazin (8 mg kg 1 body weight) and ketamine (100 mg kg 1 body weight), the hair on the back was shaved and the skin was cleansed with Octenisept solution (Schu ¨lke & Mayr., Norderstedt, Germany). One subcutaneous pocket was made on each side of the median line in the lower back area through a 1-cm-long incision. One square centimetre of each sterile prosthesis material was implanted aseptically into the pockets and the incision closed with sutures. Sterile saline solution (0.1 ml) containing the S. aureus Xen29 strain at a concentration of 1 108 cfu ml 1 was inoculated onto the graft surface using a syringe to create a subcutaneous fluid-filled pocket. Control mice (n Z 7) received only the bacteria without any graft material. The animals were returned to individual cages and thoroughly examined daily.
Bacterial strain Bioluminescent S. aureus strain Xen29 was used for challenging. S. aureus Xen29 is a genetically engineered isolate of the parental strain ATCC 12600 expressing a modified Photorhabdus luminescence lux operon.7 The strain was obtained commercially from Xenogen Corporation (now part of Caliper Life Sciences, Hopkinton, MA, USA). The ability of this strain to produce biofilm and in vivo virulence has been shown.7,9 A single colony of the strain was used to inoculate a 25-ml 2 YT broth culture overnight at 37 C. The culture was washed in phosphate-buffered saline (PBS) and serially diluted to obtain a concentration of 1 108 cfu ml 1 cfus as confirmed by quantitative culture analysis. The bacteria were suspended in 0.1 ml of physiologic NaCl solution for application.
Experimental protocol Vascular graft materials were placed subcutaneously in 35 mice with polyester graft material in group A (n Z 7), polytetrafluoroethylene graft material in group B (n Z 7), two types of silver acetate-coated polyester graft materials in groups C and D (n Z 7 each) and bovine pericardium graft material in group E (n Z 7).
70
U. Lorenz et al.
In vivo bioluminescence imaging
Statistical methods
After infection with bioluminescent S. aureus Xen29, the mice were anaesthetised with inhaled isoflurane and imaged according to a defined time schedule (15 min, 2 h and 1e5 days postinfection) using the Berthold NightOWL low-light imaging system (Berthold, Bad Wildbad, Germany). Total photon emissions from predefined regions of interest (ROIs) with metabolically active S. aureus in each animal were acquired for 1 min. The captured images were then quantified using the indiGo software package (Berthold, Bad Wildbad, Germany). Bioluminescent signals from the ROI were expressed in a pseudo-colour image, with red representing the most intense luminescence and blue representing the least. This image was overlaid on a grey-scale reference image to generate a two-dimensional picture of the distribution of bioluminescent bacteria on the graft surface; the data are presented as the cumulative photon counts (photons per second (ps 1)) collected within each individual ROI. The image data were analysed and pictures were created using Adobe Photoshop CS2.
Bioluminescence intensity is reported as mean S.D. Bacterial counts are visualised as scatter dot plot with mean line graphics. Statistical comparisons between the two groups were made using the ManneWhitney-U test. The GraphPad Prism program (GraphPad Software, Inc., La Jolla, CA, USA) was used to analyse and graphically present the data.
Assessment of graft contamination The grafts were aseptically harvested from mice 5 days after infection, gently washed in sterile saline solution and placed in tubes containing 10 ml of PBS solution. They were then sonicated for 5 min to remove the adherent bacteria from the grafts and homogenised in sterile PBS. Quantification of viable bacteria was done by culturing serial 10fold dilutions (0.1 ml) of the bacterial suspension onto agar plates. All plates were incubated at 37 C for 48 h and evaluated for the presence of the S. aureus Xen29 strain. The organisms were quantified by counting the number of cfu per plate.
Results Clinical course None of the animals died or had clinical evidence of adverse effects, such as local signs of perigraft inflammation, anorexia, vomiting, diarrhoea or changes in behaviour.
Biophotonic imaging of infected prostheses material Real-time in vivo bioluminescence of subcutaneous S. aureus Xen29 correlated to the total number of bacteria (r2 Z 0.75), suggesting that bioimaging provides an accurate evaluation of bacterial burden (Fig. 1). The bioluminescent images of a representative animal selected from each group and associated mean bioluminescent signals (total flux in ps 1) after contamination of implanted grafts are shown in Fig. 2 (panels A and B). At 15 min after bacterial infection, each graft group had a similar mean bioluminescent signal at the site of contamination. The bioluminescent signal intensity declined in the two groups with silver-coated graft material (C and D) over the course of 2 h. In the other groups (A, B and E), the mean bioluminescent signal remained steady from 15 min to 2 h, and
Figure 1 Correlation between non-invasive bioluminescence imaging and bacterial burden. Bioluminescent S. aureus Xen29 was subcutaneously applied to the median line of the lower back area of whole animals. The intensity of the transcutaneous photon emission is represented as a pseudocolour image. In vivo bioluminescence signal in photons per second (p/s) and plate count for enumerating bacteria in colony-forming units (cfu) were plotted against each other. Regression analysis yielded r2 Z 0.75, indicating a positive correlation between the two methods.
S. aureus Biofilm Graft Infection then declined by day 1. A progressive reduction in the bioluminescent signal was observed in all graft groups after 24 h, dropping as low as approximately 5 105e1 106 ps 1 in all but group E (bovine pericardium), which fell to 2.41 106 ps 1. In control animals, the bioluminescent signal showed a continued increase until 24 h post-contamination, after which it declined to undetectable levels by the end of the observation period. The bioluminescent signals peaked again 2 days after graft contamination in all groups except group E, where the peak value was reached on day 3. Afterwards, the bioluminescent signals declined gradually to the lowest level in the respective groups until the end of the observation
71 period on day 5, but were undetectable in groups C and D (silver-coated prostheses). Overall, the curve shapes in the groups were closely similar on different kinetic levels. The highest overall signal intensity was seen for animals in group E with bovine pericardium graft material, followed in descending order by groups A, C, B and D. At the end of the observation period on day 5 of graft infection, the mean total flux (ps 1) reached 4.75 105 in group A (polyester), 1.86 105 in group B (ePTFE) and 1.77 106 in group E (bovine pericardium). One of seven animals (14%) had no detectable signal in group B and 2 of 7 (29%) in group A. All animals in group E had traceable bioluminescence signals. In groups C (silver I) and D (silver
Figure 2 Time series of photon detection from metabolically active S. aureus within biofilms on different vascular prostheses in mice during a 5-day follow-up. (A) The dynamic of bioluminescence levels was monitored 15 min, 2 h and on day 1e5 after contamination of vascular graft materials with 1 107 cfu S aureus Xen29. Pseudocolor representation of bioluminescence intensity of a representative mouse of each group with indicated graft material (legend) is presented. The time of monitoring corresponds to the time points indicated in the graph of panel B. (B) Corresponding mean luminescence (SD) (total flux in photons per second (p/s)) of mice were plotted as a function of time. ()) Significant differences between the indicated groups on day 5 after contamination are displayed.
72 II), the mean bioluminescent signal was zero, with only a marginally detectable signal in two animals in group C.
Bacterial counting on contaminated prostheses materials The comparative analyses between bioluminescence and the traditional colony-counting method after 5 days of graft challenge with S. aureus Xen29 are summarised in Fig. 3 and Table 1. No evidence of graft contamination was detected on any explanted graft in group D (silver II). In group C (silver I), only 2 of 7 (29%) explanted grafts showed a low count of 10 and 30 cfu cm 2. In accordance, the bioluminescence signal was also positive but not quantifiable in these two animals. The quantitative graft cultures of the other groups exhibited variable bacterial growth. In group A (polyester), 5 of 7 (71%) grafts showed bacterial growth, three of which had 1 103, 2 103 and 4 103 bacteriacm 2. The other two grafts had low counts of 10 and 30 cfu cm 2. By contrast, all seven animals in group A had a detectable bioluminescence signal. In group B (PTFE), 5 of 7 (71%) grafts showed bacterial growth. Only two had moderate counts of 3 103 and 1.5 104 cfu cm 2. The other three grafts had low counts of 20 and 10 cfu cm 2. In this group, only 1 of 7 animals had no detectable bioluminescence signal. In group E, every graft exhibited moderate bacterial growth with a mean (range) count of 2 103 (6 102e9 103) cfu cm 2. In this group, all animals showed consistently detectable bioluminescence signals.
Discussion In 1993, the Ad Hoc Committee of the Joint Councils of the Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter, proposed
Figure 3 Recovery of S. aureus strain Xen29 following subcutaneous challenge of vascular graft materials with 1 107 bacteria. The number of cfu was determined by differential plating analysis. The number of cfu from the investigated graft materials (polyester, polytetrafluoroethylene, silver I, silver II, bovine pericardium) of animals were visualised as scatter dot blot with the horizontal line denotes the mean value. Significant differences between the groups are indicated.
U. Lorenz et al. a comprehensive performance and procedure standard for arterial prostheses.5 Besides other characteristics, the ideal prosthetic material should be resistant to infection as are human arteries. Although several attempts have been made to improve resistance to bacterial contamination, no currently available prosthetic graft completely fulfils this criterion. The creation of a prosthetic vascular graft that is able to eliminate the natural ability of bacteria to adhere and to develop biofilm requires considerable further scientific investigation. In this context, reliable preclinical in vivo model systems to test new concepts of altering surface properties by physical, chemical and biological modifications serve as preliminary steps for the improvement of the antiinfectious characteristics of vascular grafts. To date, the design of animal studies usually requires removal of infected grafts at each sampling point to indirectly estimate the pathogen burden by measuring the bacterial count.10e12 This method needs large numbers of animals and it does not permit direct monitoring of the infection over time.13,14 Moreover, the enumeration of bacteria is a great challenge, especially in graft-associated biofilm infection.15 In particular, the manipulation of graft materials by vortex or sonication to successively count adherent bacteria presents disadvantages as they may not detach all bacteria or may alter bacterial viability leading to inaccurate counts. A potential solution to these drawbacks is biophotonic imaging, as first described by Contag et al.16 and later by Burns et al.17 for monitoring Gram-negative bacterial infections. Gram-positive bacteria including S. aureus strains can also be made bioluminescent through the stable integration of an optimised P. luminescence luxABCDE cassette onto their chromosome as first described by Francis et al.18 In the present study, we examined whether the in vivo model of biophotonic imaging could be adapted to allow continuous monitoring of bacterial biofilm colonisation for a certain period after graft contamination. The study design was chosen for several reasons. First, direct inoculation of bacteria mimics contamination of vascular grafts at the time of implantation, the foremost reason for early graft infection.19 Although transient bacteraemia as a source of infection is well described in experimental investigations,20 the clinical evidence remains speculative. Second, staphylococci are the most important pathogens in PVGI.21 Third, although the prosthetic material was not implanted into the continuity of an arterial vessel, the subcutaneous graft placement may reproduce an environment similar to the clinical situation. Further, fourthly, the direct correlation between S. aureus density on medical devices and the measured bioluminescent signal intensity has been shown.18,22 In our model, the kinetics of graft infection as a measure of bioluminescence exhibited some specific features. In the control group without prosthetic material implantation, for example, no bacterial proliferation was found prior to 24 h post-contamination. This suggests that the implantation of any prosthetic material supports bacterial survival and growth. In agreement, a foreign body reaction has been identified as a predisposing factor to infection.23 The time course of bioluminescence intensity in all prosthesis groups showed a wavelike rhythmic pattern with an overall
S. aureus Biofilm Graft Infection
73
Table 1 Bioluminescence and traditional bacterial determination method to count S. aureus Xen29 in biofilm after 5 days of graft infection in a standardized mouse model. Graft material
Plate count method Detected
Polyester PTFE Silver I Silver II Bovine pericardium
5/7 5/7 2/7 0/7 7/7
Bioluminescence signal
Mean cfu (Range)/cm2 3 101 (0e4.0 104) 1 101 (0e1.5 104) 0 (0e3.0 101) 0 2.0 103 (6 102e9103)
decreasing signal amplitude up to day 5 after infection. This phenomenon mirrors probably the cyclic presence of metabolically active bacteria in biofilm under the control of the animal’s host defence mechanisms, which causes finally a decrease in the bacterial density.24,25 In contrast to our results, some previous studies have reported higher colonisation rates over a certain time in similar mouse or rat models with subcutaneously implanted vascular graft materials.26,27 The apparent conflicting data might be explained by using higher inoculum sizes and different strains in these models. The number of bacteria needed to produce a consistent growth and the biofilm formation capacity varies from one strain to another. Another issue is the choice of the experimental animal. Reproducible infections in mice and rats are difficult to establish because of their high natural resistance to staphylococci. For example, it has been previously reported that various rat strains behaved remarkably differently in prosthetic contamination studies. Highly variable numbers of bacteria adherent to the same foreign material were detected.28 Conclusively, it is difficult to compare directly the number of biofilm-associated bacteria across different studies. Nevertheless, the variable extent of bioluminescence signals among graft materials suggests truly different abilities to resist bacterial growth on the surface. Specifically, in the group with silver-coated prostheses, the bioluminescence signal decreased rapidly after contamination, whereas, with uncoated synthetic prosthetic materials, a decline was observed only after several hours. Moreover, in the groups with silver-coated prostheses, almost no more signal activity could be detected on the day of graft explantation. Hence, silvercoating appears to confer superior in vivo resistance to staphylococcal growth compared to the other uncoated materials. Comparative analysis of the commonly used prosthetic grafts polyester and ePTFE suggests the latter has better resistance to staphylococcal adherence. Finally, the bovine pericardium was significantly less resistant to staphylococci compared to all other prostheses in this model. Of note, biophotonic imaging was able to detect more often metabolically active bacteria within biofilm than culture enumeration at the end of the study period. In polyester, the study strain was identified in all cases by biophotonic imaging, while the same bacterium was found by enumeration only in five of seven cases. Similarly, one more infected ePTFE graft was identified by biophotonic imaging. The possibility to find more often low numbers of
Detected
Mean photons (Range)/sec
7/7 6/7 2/7 0/7 7/7
4.7 105 1.86 105 0 0 1.77 106
(1.2 105e9.10 105) (0e9.68 105) (0e5.36 105) (7.3 105e1.8 106)
viable bacteria on graft material by biophotonic imaging may have great importance for interpreting bacterial resistance of prostheses. The differential adherence of bacteria to a prosthetic vascular graft determines its relative infectivity. In other words, the higher the load of vital bacteria on the graft surface at any time, the greater is the risk for late prosthetic vascular graft infection.29,30 The presented findings on the individual prosthetic materials accord with clinical or preclinical observations. The superior resistance to infection of silver-coated prostheses has been reported in clinical studies.31,32 Furthermore, bacterial strains have been shown to have greater affinity to polyester than to PTFE,33,34 although the literature is somewhat inconsistent.35 Some authors have advised extracellular matrix biomaterials in the management of graft infection.29,30 One such graft material with excellent healing properties during implantation is bovine pericardium, whose use as a replacement for other infected prosthetic grafts has been discussed.36 This is in contrast with our presented finding that bovine pericardium has low resistance to infection. Previously published observations are in line with our data.37 There are some limitations to our study. The used experimental model is not directly comparable with the implantation of grafts in human vessels and, therefore, conservative interpretations of the presented results seem advisable. Furthermore, we studied exclusively the early, extracavitary graft contamination and only a single strain of S. aureus was included. Long-term studies are required for confirmation of the data. In the present study, we show that biophotonic imaging can quantify not only differences in resistance to infection of various graft materials, but also that bacterial growth can be monitored longitudinally, in real-time and noninvasively in live animals. Bioluminescence allows more sensitive detection of bacteria within biofilm on graft surfaces. The experimental model presented here may be ideally suited as an adjunct to other studies for pre-implant evaluation of graft resistance to bacterial biofilm.
Conflict of Interest None.
Acknowledgements The authors are indebted to Mr. J. Heywood for critical reading of the manuscript and helpful remarks. This study
74 was supported by a grant from the Federal Ministry of Education and Research (BMBF) Network of Competence ‘Pathogenomics’ Alliance ‘Gram-Positive Cocci’ and by the German Research Foundation (TRR34). We kindly thank Berthold Technologies for providing the low-light imaging system and technical support.
References 1 Perera GB, Fujitani RM, Kubaska SM. Aortic graft infection: update on management and treatment options. Vasc Endovascular Surg 2006;40:1e10. 2 Armstrong PA, Back MR, Bandyk DF, Johnson BL, Shames ML. Selective application of sartorius muscle flaps and aggressive staged surgical debridement can influence long-term outcomes of complex prosthetic graft infections. J Vasc Surg 2007;46:71e8. 3 Zdanowski Z, Danielsson G, Jonung T, Norgren L, Ribbe E, Thorne J, et al. Intraoperative contamination of synthetic vascular grafts. Effect of glove change before graft implantation. A prospective randomised study. Eur J Vasc Endovasc Surg 2000;19:283e7. 4 Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a common cause of persistent infections. Science 1999;284: 1318e22. 5 Abbott WM, Callow A, Moore W, Rutherford R, Veith F, Weinberg S. Evaluation and performance standards for arterial prostheses. J Vasc Surg 1993;17:746e56. 6 Berlutti F, Rosso F, Bosso P, Giansanti F, Ajello M, De Rosa A, et al. Quantitative evaluation of bacteria adherent to polyelectrolyte HEMA-based hydrogels. J Biomed Mater Res A 2003; 67:18e25. 7 Kadurugamuwa JL, Sin L, Albert E, Yu J, Francis K, DeBoer M, et al. Direct continuous method for monitoring biofilm infection in a mouse model. Infect Immun 2003;71:882e90. 8 Engelsman AF, van der Mei HC, Francis KP, Busscher HJ, Ploeg RJ, van Dam GM. Real time noninvasive monitoring of contaminating bacteria in a soft tissue implant infection model. J Biomed Mater Res B Appl Biomater 2009;88:123e9. 9 Lorenz U, Huttinger C, Schafer T, Ziebuhr W, Thiede A, Hacker J, et al. The alternative sigma factor sigma B of Staphylococcus aureus modulates virulence in experimental central venous catheter-related infections. Microbes Infect 2008;10:217e23. 10 Hernandez-Richter T, Wittmann F, Mayr S, Schmitt-Sody M, Blasenbreu S, Wichmann M, et al. Experimental examination concerning the efficacy of silver-coated Dacron prostheses in vascular graft infections following subcutaneous implantation in a standardized infection model. Zentralbl Chir 2003;128: 244e9. 11 Sato S, Nitta Y, Saiki Y, Kawamoto S, Iguchi A, Kaku M, et al. Enhanced perigraft angiogenesis prevents prosthetic graft infection. Ann Thorac Surg 2008;86:1278e84. 12 Sacar M, Sacar S, Kaleli I, Onem G, Turgut H, Goksin I, et al. Linezolid alone and in combination with rifampicin prevents experimental vascular graft infection due to methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis. J Surg Res 2007;139:170e5. 13 Bergamini TM, Corpus Jr RA, McCurry TM, Peyton JC, Brittian KR, Cheadle WG. Immunosuppression augments growth of graft-adherent Staphylococcus epidermidis. Arch Surg 1995; 130:1345e50. 14 Rupp ME, Ulphani JS, Fey PD, Mack D. Characterization of Staphylococcus epidermidis polysaccharide intercellular adhesin/hemagglutinin in the pathogenesis of intravascular catheter-associated infection in a rat model. Infect Immun 1999;67: 2656e9.
U. Lorenz et al. 15 Pantanella F, Valenti P, Frioni A, Natalizi T, Coltella L, Berlutti F. BioTimer Assay, a new method for counting Staphylococcus spp. in biofilm without sample manipulation applied to evaluate antibiotic susceptibility of biofilm. J Microbiol Methods 2008;75:478e84. 16 Contag CH, Contag PR, Mullins JI, Spilman SD, Stevenson DK, Benaron DA. Photonic detection of bacterial pathogens in living hosts. Mol Microbiol 1995;18:593e603. 17 Burns SM, Joh D, Francis KP, Shortliffe LD, Gruber CA, Contag PR, et al. Revealing the spatiotemporal patterns of bacterial infectious diseases using bioluminescent pathogens and whole body imaging. Contrib Microbiol 2001;9:71e88. 18 Francis KP, Yu J, Bellinger-Kawahara C, Joh D, Hawkinson MJ, Xiao G, et al. Visualizing pneumococcal infections in the lungs of live mice using bioluminescent Streptococcus pneumoniae transformed with a novel Gram-positive lux transposon. Infect Immun 2001;69:3350e8. 19 O’Brien T, Collin J. Prosthetic vascular graft infection. Br J Surg 1992;79:1262e7. 20 Leport C, Goeau-Brissonniere O, Lebrault C, Vilde JL, Pechere JC. Mechanisms of ceftriaxone prophylaxis against late bacteremic vascular graft infection caused by Staphylococcus aureus in a dog model. Ann Vasc Surg 1991;5:500e5. 21 Chiesa R, Astore D, Frigerio S, Garriboli L, Piccolo G, Castellano R, et al. Vascular prosthetic graft infection: epidemiology, bacteriology, pathogenesis and treatment. Acta Chir Belg 2002;102:238e47. 22 Wiles S, Clare S, Harker J, Huett A, Young D, Dougan G, et al. Organ specificity, colonization and clearance dynamics in vivo following oral challenges with the murine pathogen Citrobacter rodentium. Cell Microbiol 2004;6:963e72. 23 Olofsson P, Rabahie GN, Matsumoto K, Ehrenfeld WK, Ferrell LD, Goldstone J, et al. Histopathological characteristics of explanted human prosthetic arterial grafts: implications for the prevention and management of graft infection. Eur J Vasc Endovasc Surg 1995;9:143e51. 24 Kuklin NA, Pancari GD, Tobery TW, Cope L, Jackson J, Gill C, et al. Real-time monitoring of bacterial infection in vivo: development of bioluminescent staphylococcal foreign-body and deep-thigh-wound mouse infection models. Antimicrob Agents Chemother 2003;47:2740e8. 25 Kadurugamuwa JL, Modi K, Yu J, Francis KP, Purchio T, Contag PR. Noninvasive biophotonic imaging for monitoring of catheter-associated urinary tract infections and therapy in mice. Infect Immun 2005;73:3878e87. 26 Hernandez-Richter T, Schardey HM, Wittmann F, Mayr S, Schmitt-Sody M, Blasenbreu S, et al. Rifampin and Triclosan but not silver is effective in preventing bacterial infection of vascular dacron graft material. Eur J Vasc Endovasc Surg 2003; 26:550e7. 27 Bergamini TM, Peyton JC, Cheadle WG. Prophylactic antibiotics prevent bacterial biofilm graft infection. J Surg Res 1992;52: 101e5. 28 Van Wijngaerden E, Peetermans WE, Vandersmissen J, Van Lierde S, Bobbaers H, Van Eldere J. Foreign body infection: a new rat model for prophylaxis and treatment. J Antimicrob Chemother 1999;44:669e74. 29 Ehrenfeld WK, Wilbur BG, Olcott CN, Stoney RJ. Autogenous tissue reconstruction in the management of infected prosthetic grafts. Surgery 1979;85:82e92. 30 Ruotolo C, Plissonnier D, Bahnini A, Koskas F, Kieffer E. In situ arterial allografts: a new treatment for aortic prosthetic infection. Eur J Vasc Endovasc Surg 1997;14(Suppl. A):102e7. 31 Ricco JB. InterGard silver bifurcated graft: features and results of a multicenter clinical study. J Vasc Surg 2006;44: 339e46. 32 Strathmann M, Wingender J. Use of an oxonol dye in combination with confocal laser scanning microscopy to monitor damage
S. aureus Biofilm Graft Infection to Staphylococcus aureus cells during colonisation of silvercoated vascular grafts. Int J Antimicrob Agents 2004;24: 234e40. 33 Schmitt DD, Bandyk DF, Pequet AJ, Towne JB. Bacterial adherence to vascular prostheses. A determinant of graft infectivity. J Vasc Surg 1986;3:732e40. 34 Camiade C, Goldschmidt P, Koskas F, Ricco JB, Jarraya M, Gerota J, et al. Optimization of the resistance of arterial allografts to infection: comparative study with synthetic prostheses. Ann Vasc Surg 2001;15:186e96.
75 35 Yasim A, Gul M, Ciralik H, Ergun Y. Gelatin-sealed dacron graft is not more susceptible to MRSA infection than PTFE graft. Eur J Vasc Endovasc Surg 2006;32:425e30. 36 Odero A, Argenteri A, Cugnasca M, Pirrelli S. The crimped bovine pericardium bioprosthesis in graft infection: preliminary experience. Eur J Vasc Endovasc Surg 1997;14(Suppl. A):99e101. 37 Derksen WJ, Verhoeven BA, van de Mortel RH, Moll FL, de Vries JP. Risk factors for surgical-site infection following common femoral artery endarterectomy. Vasc Endovascular Surg 2009;43:69e75.
