Documente Academic
Documente Profesional
Documente Cultură
J Orthop Trauma Volume 30, Number 10, October 2016 www.jorthotrauma.com | 531
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Tennent et al J Orthop Trauma Volume 30, Number 10, October 2016
allow for the formation of biolms, which are intrinsically lateral right thigh and the femoral shaft exposed. A 24-mm
resistant to antibiotics and complicate clinical manage- poly acetyl plate was xed to the femur with six 0.9-mm
ment.2225 The purpose of this study was to evaluate the threaded K-wires, and a 6-mm segment of bone was removed
effectiveness of locally applied vancomycin powder to with a saw under saline irrigation. Each animal was then
decrease the bacterial burden in a contaminated traumatic inoculated with 1.2 106 6 1.9 105 colony-forming units
orthopaedic fracture model at different times that resemble (CFUs) of UAMS-1, a clinical S. aureus osteomyelitis isolate,
different clinical scenarios. We hypothesize that the local through a bovine collagen matrix (Sigma-Aldrich), and the
application of vancomycin powder would decrease bacterial wounds closed.
load in this contaminated fracture model. After the initial procedure, all animals were recovered
and allowed food and water ad libitum. At their designated
time-to-treatment point, the animals were again anesthetized,
MATERIALS AND METHODS the wound reopened, debrided, and irrigated with 60 mL of
sterile normal saline. The assigned treatment was then placed
Study Design into the wound, either 50 mg vancomycin powder or four
This study was conducted in compliance with the 6-mm vancomycin-impregnated beads as described above
Animal Welfare Act, the implementing Animal Welfare (Fig. 1), the wound closed in layers, and the animal recov-
Regulations, and in accordance with the principles of the ered. Beginning at the time of debridement, each animal
Guide for the Care and Use of Laboratory Animals. All received systemic antibiotics for 3 days consisting of twice-
procedures were performed in a laboratory accredited by the daily subcutaneous injections of cefazolin (5 mg/kg). Four-
Association and Accreditation of Laboratory Animal Care teen days after debridement, the animals were anesthetized
following a protocol approved by our Institutional Animal and killed with sodium pentobarbital. The plates and wires,
Care and Use Committee. bone, beads, and the surrounding muscle were taken and
Fifty-four Lewis rats were rst randomly assigned into analyzed for microbial colonization, inammatory cytokines
treatment groups: debridement and irrigation (D and I) [interleukin 6 (IL-6) and tumor necrosis factor a (TNFa)],
without local antibiotics (Standard Treatment, n = 18); and local antibiotic concentration.
D and I with 4 vancomycin-loaded PMMA beads (n = 18);
or D and I with 50 mg local vancomycin powder (n = 18),
Bacterial Enumeration
which is a sufcient application to coat the entirety of the
wound bed before closure. The rats were then assigned to At euthanasia, the femurs, implants (plates and wires),
a time-to-treatment group: 6 hours (early treatment) (n = and beads were removed. The femurs were ash frozen in
10) or 24 hours (late treatment) (n = 8). The 24-hour treatment liquid nitrogen, crushed to a ne powder, and resuspended
time was chosen as previous in vitro work with Staphylococ- in normal saline. The implants and beads were suspended in
cus aureus has shown reliable biolm formation at 24 normal saline and sonicated for 15 minutes to remove
hours.2628 At the specied time-to-treatment point, all ani- attached bacteria. Serial dilutions of the supernatants were
mals underwent D and I with placement of their previously plated onto 5% sheeps blood agar plates (Remel, Lenexa,
assigned local antibiotic treatment. All animals were killed 2 KS) and incubated overnight at 378C. CFUs were counted
weeks after treatment placement and samples analyzed. and normalized to the weight of explanted material.25
532 | www.jorthotrauma.com Copyright 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 30, Number 10, October 2016 Locally Applied Vancomycin
collagenase (Sigma-Aldrich). Vancomycin was extracted and B). Locally applied vancomycin powder reduced bacteria
from plasma or muscle homogenate and concentrations were both within the bone (Fig. 2A) and on the implants (Fig. 2B)
measured by HPLC using UV detection at 210 nm. compared with the standard D and I treatment alone when
local vancomycin treatment was applied 6 hours after con-
Statistics tamination (P , 0.001). Similar results were seen for the
An infection was dened as detectable bacteria greater vancomycin bead group when treatment was performed at 6
than 103 CFU/g tissue as this level is approximately 1-log hours (P , 0.05). However, when local vancomycin applica-
CFU greater than the level of detection and represents the tions were delayed until 24 hours after the initial bacterial
bacterial burden where bone and implant are expected to inoculation, the vancomycin beads and powder both failed
show detectable levels of bacteria.25 Fisher exact test using to decrease infection (P = 1). Bacterial counts on the implants
absolute bacterial counts was performed comparing the treat- in those animals treated at 6 hours were signicantly lower
ment groups with the standard treatment of those animals only in the powder group (P , 0.001). Furthermore, there
infected ($103 CFU/g) and not infected (,103 CFU/g). Stu- were no signicant differences in bacteria between the pow-
dent t test was used to compare vancomycin concentrations in der and bead groups when treatment was given at either 6 or
the muscle. A 2-way analysis of variance with Bonferroni 24 hours on bone (P = 0.016 and P = 0.564, respectively) or
t test was used to compare local inammatory markers. Sta- implants at 24 hours (P = 0.0679).
tistical signicance was set at P , 0.05. All values are ex- Vancomycin was detectable in the blood of all vanco-
pressed as mean 6 SEM. mycin powder animals at 6 and 24 hours after application,
with an average of 23.5 6 4.0 and 8.6 6 1.8 mg/mL, respec-
tively. These levels declined over time, with less than 30% of
RESULTS the animals having detectable quantities of vancomycin in the
All bone and implant samples taken from control serum by day 14 (Fig. 3A). In contrast to the vancomycin
animals developed infection at 6 and 24 hours (Figs. 2A powder group, vancomycin beads resulted in negligible
Copyright 2016 Wolters Kluwer Health, Inc. All rights reserved. www.jorthotrauma.com | 533
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Tennent et al J Orthop Trauma Volume 30, Number 10, October 2016
vancomycin being detected in serum at 6 or 24 hours; how- (Fig. 3B), indicating persistent antibiotic presence in the sur-
ever, low levels of vancomycin became detectable in serum at rounding musculature at the time of sacrice.
days 7 and 14 in those animals treated with vancomycin beads Consistent with the previous results, measurable levels
(Fig. 3A). Vancomycin was also detectable in the adjacent of local proinammatory markers, IL-6 and TNFa, corre-
musculature for both beads, 2.87 (SEM: 1.14) and powder sponded with reductions in microbial burden in all groups
groups, 5.67 (SEM: 2.53), at 14 days after treatment at each time point (Figs. 4A and B), with an exception of
534 | www.jorthotrauma.com Copyright 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 30, Number 10, October 2016 Locally Applied Vancomycin
FIGURE 4. Levels of local inflammatory proteins IL-6 and TNFa. A, When compared with the standard treatment, there was
a decrease in IL-6 for both vancomycin beads and vancomycin powder at 6 hours. There was also a decrease in IL-6 among the
vancomycin beads at 24 hours. B, There was a significant decrease in TNFa in both vancomycin beads and vancomycin powder at
6 hours but not at 24 hours. *P , 0.05; #P , 0.001. Editors Note: A color image accompanies the online version of this article.
IL-6 of the vancomycin bead group at 24 hours. There was no interesting to note that there were no differences in bacteria
difference in levels of IL-6 or TNFa between vancomycin load between vancomycin powder and beads groups at 6
beads and vancomycin powder groups at 6 hours (P = hours or 24 hours, at 24 on the implants or in inammatory
0.999 and P = 0.061, respectively) or 24 hours (P = 0.116 markers at 6 hours or 24 hours. This may imply that
and P = 0.133). Lastly, whereas those animals receiving treat- vancomycin powder is a viable, more cost-effective alterna-
ment at 6 hours showed little clinical evidence of infection, tive to beads that can allow immediate closure of a traumatic
including purulence and necrotic tissue, 5 of 8 in the vanco- wound when additional procedures are not required.
mycin bead group and 5 of 8 in the vancomycin powder Previous studies have demonstrated that early antibiotic
group showed purulence and tissue necrosis when the treat- administration decreases infection rates when administered
ment was delayed 24 hours. sooner after injury; however, the application of local anti-
biotics has historically relied on secondary delivery mecha-
nisms, such as PMMA, calcium sulfates, or other absorbable
DISCUSSION materials.24,25,3038 Much like the spine literature where van-
Using a well-established contaminated orthopaedic comycin powder has been used successfully when placed in
wound model, this study found that the local application of a clean surgical wound before closure, vancomycin powder
vancomycin powder was an effective treatment to decrease drastically reduced evidence of infection only when used
infection rates when treatment was performed at 6 hours. early.1421,39 This time-dependent decrease in effectiveness
Although the antibiotic beads reduced the infection rate when seen with vancomycin powder is consistent with previous
administered at 6 hours postcontamination, it is interesting preclinical studies evaluating the early treatment of contami-
that there was a dichotomous treatment response; half of the nated wounds and the need for early treatment, presumably
wounds had large amount of bacteria within the wound and before mature biolm formation could occur, which has
half were not infected. In contrast, this was not seen within been shown to be within the rst 510 hours after contam-
the antibiotic powder group at the early time point where all ination for some organisms.24,25,28,33,34 Furthermore, the
animals displayed an approximately 5-log CFU reduction in presence of vancomycin powder in the surrounding mus-
bacteria when compared with D and I alone which were culature at 2 weeks seen in this study is similar to the
similar to the corresponding decreases in inammatory previous literature showing continued vancomycin pres-
markers in those groups that had decreased levels of intra- ence in spine wound exudates several days after applica-
wound bacteria. Vancomycin was also shown to be present at tion.15 This time dependency stresses the limitations of
signicant levels up to 2 weeks after application in the current treatment interventions and the importance of having
surrounding musculature in those animals assigned to the other antiinfective agents at the surgeons disposal with
vancomycin powder group. Furthermore, both the local and increased effectiveness against biolms when treatment is
serum vancomycin levels were similar between the antibiotic delayed or if there is clinical concern for persistent infection,
bead and powder groups 14 days after application. Lastly, it is such as rifampin which has shown in vitro effectiveness at
Copyright 2016 Wolters Kluwer Health, Inc. All rights reserved. www.jorthotrauma.com | 535
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Tennent et al J Orthop Trauma Volume 30, Number 10, October 2016
536 | www.jorthotrauma.com Copyright 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 30, Number 10, October 2016 Locally Applied Vancomycin
26. Cardile AP, Sanchez CJ, Samberg ME, et al. Human plasma enhances the 36. Penn-Barwell JG, Murray CK, Wenke JC. Local antibiotic delivery by
expression of staphylococcal microbial surface components recognizing a bioabsorbable gel is superior to PMMA bead depot in reducing infec-
adhesive matrix molecules promoting biolm formation and increases tion in an open fracture model. J Orthop Trauma. 2014;28:370375.
antimicrobial tolerance in vitro. BMC Res Notes. 2014;17(7):457. 37. Stinner DJ, Hsu JR, Wenke JC. Negative pressure wound therapy re-
27. Sanchez CJ, Mende K, Beckius ML, et al. Biolm formation by clinical duces the effectiveness of traditional local antibiotic depot in a large
isolates and the implications of chronic infections. BMC Infect Dis. 2013; complex musculoskeletal wound animal model. J Orthop Trauma.
13:47. 2012;26:512518.
28. Sanchez CJ Jr, Shiels SM, Tennent DJ, et al. Rifamycin derivatives are 38. Hafeman AE, Zienkiewicz KJ, Carney E, et al. Local delivery of tobra-
effective against staphylococcal biolms in vitro and elutable from mycin from injectable biodegradable polyurethane scaffolds. J Biomater
PMMA. Clin Orthop Relat Res. 2015;473:28742884. Sci Polym Ed. 2010;21:95112.
29. Chen X, Kidder LS, Lew WD. Osteogenic protein-1 induced bone for- 39. Theologis AA, Demirkiran G, Callahan M, et al. Local intrawound
mation in an infected segmental defect in the rat femur. J Orthop Res. vancomycin powder decreases the risk of surgical site infections in
2002;20:142150. complex adult deformity reconstruction: a cost analysis. Spine. 2014;
30. Harley BJ, Beaupre LA, Jones CA, et al. The effect of time to denitive 39:18751880.
treatment on the rate of nonunion and infection in open fractures. 40. Li B, Brown KV, Wenke JC, et al. Sustained release of vancomycin from
J Orthop Trauma. 2002;16:484490. polyurethane scaffolds inhibits infection of bone wounds in a rat femoral
31. Lack WD, Karunakar MA, Angerame MR, et al. Type III open tibia segmental defect model. J Control Release. 2010;145:221230.
fractures: immediate antibiotic prophylaxis minimizes infection. J Or- 41. Kallala R, Graham SM, Nikkhah D, et al. In vitro and in vivo effects of
thop Trauma. 2015;29:16. antibiotics on bone cell metabolism and fracture healing. Expert Opin
32. Patzakis MJ, Wilkins J. Factors inuencing infection rate in open fracture Drug Saf. 2012;11:1532.
wounds. Clin Orthop Relat Res. 1989;243:3640. 42. Rathbone CR, Cross JD, Brown KV, et al. Effect of various concentra-
33. Branstetter JG, Jackson SR, Haggard WO, et al. Locally-administered anti- tions of antibiotics on osteogenic cell viability and activity. J Orthop Res.
biotics in wounds in a limb. J Bone Joint Surg Br. 2009;91:11061109. 2011;29:10701074.
34. Brown KV, Walker JA, Cortez DS, et al. Earlier debridement and antibiotic 43. Eder C, Schenk S, Trinopoulos J, et al. Does intrawound application of
administration decrease infection. J Surg Orthop Adv. 2010;19:1822. vancomycin inuence bone regeneration in spinal fusion?. Spine J. 2014;
35. Penn-Barwell JG, Murray CK, Wenke JC. Early antibiotics and debride- 14:S2S3.
ment independently reduce infection in an open fracture model. J Bone 44. Guelcher SA, Brown KV, Li B, et al. Dual-purpose bone grafts improve
Joint Surg Br. 2012;94:107112. healing and reduce infection. J Orthop Trauma 2011;25:477482.
Copyright 2016 Wolters Kluwer Health, Inc. All rights reserved. www.jorthotrauma.com | 537
Copyright 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.