• Embed Doc
  • Copy Link
  • Readcast
  • Collections
  • CommentGo Back
Download
 
JOURNAL OF SURGICAL RESEARCH
44, 1-7 ( 1988)
Control of Prosthetic Bacterial infection: Evaluation of an EasilyIncorporated, Tightly Bound, Silver Antibiotic PTFE Graft’
ALANI.BENVENISTY,M.D.,GARYTANNENBAUM, M.D., THOMASN.AHLBORN, M.D.,CHARLESL.FOX,M.D.,SHANTAMODAK, PH.D., LESTERSAMPATH,B.S.,KEITHREEMTSMA, M.D., ANDROMAN
NOWYGROD,
M.D.
Department of Surgery, Columbia University, College of Physicians and Surgeons,630 West 168 Street, New York, New York 10032Submitted for publication November 6, 1986Despite the use of prophylactic antibiotics in vascular surgery, prosthetic infection rate remains2-Y&. Antibiotics bound to vascular prostheses can control experimentally induced infection butprolonged antibacterial activity has not been achieved. This study evaluates the in vivo efficacy andantibiotic retention of an easily prepared silver-antibiotic prosthesis. Prostheses were prepared bycombining silver with oxacillin or amikacin using an organic solvent. After evaporation of the solvent,the graft was left impregnated with the antibiotic complex. In vivo retention studies were performed byimplanting PTFE“‘Ag-oxacillin prostheses n four canine abdominal aortas. When prostheses wereexplanted at 1 week, mean antibiotic retention was found to be 20% of original activity, higher than themean inhibitory concentration for Staphylococcus aureus. In three groups of five dogs, 20 X 7-mmprostheses of PTFE alone, PTFE silver-oxacillin, or PTFE silver-amikacin were implanted in theabdominal aorta and the grafts were inoculated with 10’ S. aureus of a known bacteriophage type, in aclosed retroperitoneal pocket. The animals were sacrificed at 1 week and the prostheseswere excisedfor quantitative bacterial culture. PFTE silver-oxacillin, and PTFE silver-amikacin prostheseshad 1.7X 10’
and
2.0 X 10’ colonies, respectively, significantly less (P i 0.05) than controls (1.3
X
lo6colonies). These data suggest hat antibiotic prosthesescan be easily prepared without binding agents.They retain the bound antibiotic for a prolonged period and are effective in reducing graft infection in astringent direct contamination model.
0 1988 Academic FT~SS,
X.
INTRODUCTION
Vascular prosthetic graft infection is a seri-ous problem that occurs in 2-5% [l-3] ofpatients with subsequent limb loss in40-75% and mortality in 25-50% [2, 41. In-fection is more commonly due to contami-nation at the time of implantation than fromhematogenous seeding. [5, 61. Bonding ofantimicrobial agents to prosthetic materialshas been suggested as a way of decreasinggraft infectibility and as a method of produc-ing grafts that can be placed in infected fields[7]. Techniques for the application of antibi-otics to grafts have been developed usingsurfactant agents [8] and collagen binding[9]. We recently reported a simple method
’ Presented at the Annual Meeting of the Associationfor Academic Surgery, Washington, DC., November5-8, 1986.
for the direct application of silver-containingantibiotics to Dacron grafts [ 161.These mod-ified grafts are easy o prepare with a numberof antibiotics. The silver moiety providesadded antimicrobial effect in addition toprolonging the graft’s retention of the antibi-otic complex. We noticed in earlier studieswith silver antibiotics bound to knitted Da-cron prostheses that immediately after im-plantation there was an early loss of antibi-otic due to brisk bleeding through the inter-stices of the porous prosthesis [ 111. Sincethere is little interstitial bleeding with poly-tetrafluorethylene (PTFE) grafts, we felt thistype of prosthesis would better retain the an-tibiotic. In this study, we have evaluated theefficacy of several commonly used antibi-otics on prosthetic infection induced by lo-cally introduced
Staphyloccocus aweus
afterexamining the kinetics of the time course ofelution of the antibiotic on the prosthesis.
0022-4804/88 $1 SO
Copyright Q 1988 by Academic Press, nc.All rights of reproduction in any form reserved.
 
2
JOURNAL OF SURGICAL RESEARCH: VOL. 44, NO. 1, JANUARY 1988MATERIALS AND METHODS
Preparation of GraftsPTFE (Gore-tex, W. L. Gore & Associates,Inc.) 7-mm internal-diameter graft materialwas used throughout this study. Silver-anti-biotic grafts were prepared by first soaking asegment of PTFE prosthesis in a solution ofoxacillin (100 mmole) or amikacin (100mmole) in acetic acid:chloroform (1:25, v/v)mixture for 5 min at 37°C. The graft wasair-dried and then suspended n a 25mmolesolution of silver nitrate in ethanol for 5 minat room temperature. Following drying, thegraft was returned to the first solution foranother 5 min, redried, and then washed indistilled water. A final drying period was car-ried out in a vacuum desiccator prior to gassterilization. Grafts were stored in a dark re-frigerator until used. Control grafts were un-treated as the organic agents have previouslybeen shown to have no antimicrobial ef-fect [ 141.Antibiotic Retention StudiesQuantitation of drug in grafts: Grafts wereprepared as described above using ’ “AgNo3(New England Nuclear, Inc.). Silver incorpo-ration into the graft was determined by mea-suring gamma scintillation on a Nuclear-Chicago Gamma counter and extrapolatingfrom the specific activity of the radioisotope.The presence of antibiotic on graft surfaceswas quantitated using a disk-inhibitionassay. Pieces of graft 2 mm in diameter wereplaced on blood agar plates coated with lo4organisms of S. aureus and incubated at37°C for 24 hr. The diameter of the zone ofinhibition surrounding the graft was mea-sured and compared to standards.Animal studies. Conditioned adult femalemongrel dogs weighing 20-30 kg were thesubjects of all experiments. Animals werehoused and cared for in accordance withNIH Publication No. 85-23, Guide for Careand Use
of
Laboratory Animals. Surgicalprocedures were performed under pentobar-bital anesthesia and endotracheal tube venti-latory support on a Harvard respirator. Ster-ile conditions were preserved in all proce-dures. Aortic interposition grafts were placedvia a midline abdominal incision after mobi-lization of the infrarenal aorta. Metal surgi-cal clips were used to ligate one or two pairsof lumbar arteries as well as the inferior mes-enteric artery. Grafts were sutured end-to-end with 5-O polypropylene (Prolene, Ethi-con, Inc.) monofilament suture after exci-sion of a short segment of aorta. Noanticoagulants were administered. In long-term studies the retroperitoneum was thenclosed over the graft with Prolene suture.The midline fascia and skin were reapproxi-mated with absorbable sutures.Short-term elution kinetics were deter-mined by implanting 2-cm lengths of PTFEimpregnated with ’ ’ Ag-oxacillin (’ loAgOXA) or’ “Ag-amikacin (’ ’ AgAMI)in the canine aorta. Grafts were left in placefor 10 min and then excised. Radioactivesilver retention and antibiotic activity werethen determined as described above. Fivegrafts of each type were tested.Long-term elution kinetics were deter-mined after 1 week of graft implantation forfour ’ “AgOXA grafts. No additional antibi-otics were used.In vivo efficacy of the antibiotic grafts wasstudied in groups of five animals for each ofthe experimental grafts. Graft implantationwas performed in the standard fashion. Priorto closure of the retroperitoneum, 0.1 cc of108/cc S. aureus (coagulase positive andpenicillin resistant) of a known phage type(6/42e/47/54/75/77/83a/84/81) wasinjectedover the graft. Care was taken not to contam-inate any other site in the operative field.One week following implantation/infectionanimals were reexplored under sterile condi-tions and the grafts were excised. Cultures ofthe graft bed were taken for quantitative cul-ture. The excised graft was placed in 5 cc ofnutrient broth and agitated vigorously inorder to free organisms into suspension.Quantitative cultures were performed byplating 0.2 cc of the resulting suspension on ablood agar plate and counting the number ofcolonies after 24 hr of growth.
 
BENVENISTY ET AL.: CONTROL OF PROSTHETIC INFECTION
3
TABLE I
ONE-WEEK ELUTION KINETICS OF PTFE AgOXA PROSTHESISAND ANTIMICROBIAL Acrrvm
Expt.1234x&SD’ “Ag in graft (pmole)PreimplantPostimplant4.5 0.954.0 0.503.8 0.903.6 0.753.98 + 0.33 0.78 f 0.18Percentageoriginal21.012.523.621.019.5 f 4.2Zone inhibition (mm)1515131414.3 I!C .83
Microscopic studies were performed ontreated and nontreated PTFE grafts. Sectionsof grafts prior to or following 20 min of im-plantation were fixed in Karnovsky’s solu-tion, dehydrated in ethanols, and criticalpoint dried. Scanning electron microscopy(SEM) was performed on specimens sputtercoated with 60:40 gold:palladium mixturewith a Cambridge Stereoscan 250 Mk2. En-ergy dispersive X-ray microanalysis was per-formed after carbon coating on a Kevex8000 analyzer (see Ref. [ 121 or detailed dis-cussion of these techniques).
RESULTS
Antibiotic Retention Studies
Analysis of the ” ‘AgOXA and ” ‘AgAMIPTFE grafts shows that they retain 3.98 and4.2 pmole of antibiotic complex, respec-tively, per 2 cm of graft length.
Short-Term Elution Kinetics
74% of AgOXA and 47% AgAMI were re-tained after 10 min of perfusion in the canineaorta as determined by ’ “Ag gamma scintil-lation. These values correspond to 3 pmoleof AgOXA and 2 pmole of AgAMI.Zone of inhibition results for the antibioticgrafts were 40 and 33 mm, respectively.
Long-Term Elution Kinetics
After 1 week of implantation, the’ “AgOXA graft retained 19.5% of radioac-tivity as shown in Table 1. This is equivalentto 0.78 pmole of AgOXA and produced a14.3-mm zone of inhibition when placed ona
S. aureus
plate. This is one-third the diam-eter of the zone formed around a preimplantgraft. AgAMI was not tested at 1 week be-cause of the superior characteristics of theAgOXA graft.
In vivo Eficacy Studies
After 1 week of implantation, untreatedPTFE contained 1.3
X
lo6 colonies of S.
aureus
while AgOXA had 1.7
X
lo2 andAgAMI had 2.0 X lo2 colonies per graft (P< 0.05 for each graft). Cultures of the graftbeds contained 36.8 and 2.2 colonies forAgOXA and AgAMI, respectively. Controlbeds contained 2.75
X
lo5 colonies (P = 0.06for both experimental groups vs control).These results are summarized in Table 2.Neither the control beds nor the antibioticgroups appeared grossly purulent but all ap-peared mildly inflamed.
TABLE 2
CONTROL OF INFECTION BY PTFE-ANTIBIOTIC
GRAFTS: n Vivo
EFFICACY
Quantitative bacterial countGraftGraft bedControl 1.30 x lo62.75 x lo5ASOXA 1.72 X 102*3.68 x lo’**
AgAMI
2.00 x loz*2.20 x loo*** P < 0.05 vs control (Student’s t test).** P = 0.06 vs control.
of 00

Leave a Comment

You must be to leave a comment.
Submit
Characters: ...
You must be to leave a comment.
Submit
Characters: ...