Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Nov 1.
Published in final edited form as: J Surg Res. 2016 Jul 5;206(1):41–47. doi: 10.1016/j.jss.2016.06.099

Antibiotic-Releasing Microspheres Prevent Mesh Infection in Vivo

Kevin T Grafmiller a,*, Sean T Zuckerman d,*, Clayton Petro a, Lijia Liu a, Horst A von Recum b, Michael J Rosen c, Julius N Korley d
PMCID: PMC5141609  NIHMSID: NIHMS808796  PMID: 27916373

Abstract

Background

Infection remains a dreaded complication after implantation of surgical prosthetics, particularly after hernia repair with synthetic mesh. We previously demonstrated the ability of a newly developed polymer to provide controlled release of an antibiotic in a linear fashion over 45 days. We subsequently showed that coating mesh with the drug-releasing polymer prevented a Staphylococcus aureus (SA) infection in vivo. In order to broaden the applicability of this technology, the polymer was synthesized as isolated “microspheres” and loaded with vancomycin (VM) before conducting a non-inferiority analysis.

Materials and Methods

Seventy-three mice underwent creation of a dorsal subcutaneous pocket that was inoculated with 104 CFU of green fluorescent protein (GFP)-labeled SA (105 CFU/ml). Multifilament polyester mesh (7*7mm) was placed into the pocket and the skin was closed. Mesh was either placed alone (n=16), coated with VM-loaded polymer (n=20), placed next to VM-loaded microspheres (n=20) or unloaded microspheres (n=10), or flushed with VM solution (n=7). Quantitative tissue/mesh cultures were performed at 2 and 4-weeks. Mice with open wounds and explanted mesh were excluded.

Results

Twenty-two of twenty-three (96%) tissue-mesh samples from mesh alone or empty miscrospheres were positive for GFP-labeled SA at two and four-weeks. Six of seven (86%) samples from the VM flush group were positive for GFP SA at 4 weeks. Thirty-eight of thirty-eight (100%) VM-loaded pCD-coated mesh or VM-loaded microspheres were negative for GFP SA at two and four weeks.

Conclusion

Slow affinity based drug-releasing polymers in the form of microspheres are able to adequately clear a bacterial burden of SA and prevent mesh infection.

Keywords: Microsphere, Hernia, Mesh, Infection, Cyclodextrin, Antibiotic releasing

1 - INTRODUCTION

The use of synthetic mesh in ventral hernia repair allows the surgeon to achieve a more durable repair that is associated with a lower recurrence rate compared to traditional auto-tissue repair[1]. However, mesh-related complications such as acute or delayed prosthetic infection has become a concerning clinical problem. Current prevention and non-surgical treatments for mesh infection are limited, and infection of the mesh often necessitates surgical removal to avoid serious complications, such as enteric fistula[2]. A multisite cohort study of patients undergoing incisional hernia repair with permanent mesh prosthesis at 16 Veterans Affairs hospitals from 1998 to 2002 showed that 5.1% of the 1,071 mesh repairs were complicated by subsequent mesh explantation at a median of 7.3 months (interquartile range 1.4 –22.2)[3]. Infection was the most common reason for mesh explantation(69%)[3].

Mesh infection can arise weeks after implantation, but the main prophylactic options available to prevent mesh infection, including pulse lavage, perioperative systemic antibiotics, and antibiotic impregnated mesh and sutures, are all short acting. Additionally, pulse lavage with or without antibiotics lacks evidence to support its efficacy[4,5] and perioperative systemic antibiotics can produce serious side effects, especially with prolonged administration[6], limiting their use in long-term prophylaxis.

We previously demonstrated the ability of a newly developed polymer to provide controlled release of antibiotic in a linear fashion over 45 days[7,8]. The von Recum lab at Case Western Reserve University found that crosslinked cyclodextrin polymers (pCD) loaded with antibiotic release drug in a longer, more linear fashion than chemically similar dextran gels[7,8]. Cyclodextrin (CD) is a ring of 6, 7, or 8 glucose molecules that forms a pocket where drugs can bind and is commonly used in pharmaceutics to increase solubility of poorly soluble drugs. The molecular interaction between the CD pocket and drug is the rate-limiting step in release that determines the amount of free drug able to diffuse out of the polymer. Once a drug is out of the CD pocket it is able to diffuse freely throughout the material and may either diffuse out of the polymer or stick into another CD pocket. This type of release mechanism has been termed affinity-based release[9]. The pCD is capable of extended antimicrobial activity for more than 7 months even following commercial sterilization[10]. Given the extended time window following surgery during which mesh infections can arise, we believe the extended antibiotic delivery demonstrated by this pCD platform represents a novel approach to prevent mesh infections. We previously demonstrated that pCD-coated polyester mesh loaded with vancomycin (VM) prevented subcutaneous Staphylococcus aureus (SA) infection in vivo[11]. Despite the great potential of this polymer, its clinical applicability is limited by the fact that each different polymer coated product requires individual FDA approval[12]. Therefore, a drug delivery mechanism independent of the prosthesis would be preferable and more broadly applicable to preventing infection of any device. We reformulated the polymer as microspheres that could be administered independently from the prosthetic device, loaded the microspheres with VM and, conducted a non-inferiority analysis. We designed this study to determine if microsphere polymer complexes are technically practical and efficacious for use against prosthetic infection and to compare antibiotic-loaded microspheres against antibiotic-loaded, polymer-coated mesh with respect to their ability to prevent prosthetic infection.

2 – MATERIALS AND METHODS

2.1 - Polymer Materials

b-Cyclodextrin prepolymer (CD) was purchased from CycloLab, Budapest, Hungary. Ethylene glycol diglycidyl ether (EGDGE, Mn = 174) was purchased from Polysciences, Inc., Warrington, PA. All other reagents were purchased from Fisher Scientific (Pittsburgh, PA) at reagent grade.

2.2 - Mesh and Polymer Coating

A 25% w/w CD solution in dimethylformamide was prepared under stirring. Hexamethylene diisocyanate (HDI) was added to the above solution to become glucose:HDI molar ratio of 1: 0.32. Known volumes of pCD-HDI solution as prepared above were uniformly coated onto standard pieces (0.7 cm x 0.7 cm) of a three-dimensional market approved polyester mesh (Parietex TET) (Covidien, Mansfield, MA) in a glass dish, and crosslinked at room temperature 3 days. The polymer coated samples, hereafter called pCD-coated mesh, were carefully removed from the glass and were immersed in solvent, solvent/water 50/50, followed by Milli-Q water for one day each. The samples were then dried at room temperature in a dust free environment. pCD-coated meshes used for drug loading and animal studies were prepared using this methodology.

2.3 - Microspheres

The pCD microspheres were synthesized via single water-in-oil emulsification. Briefly, epichlorohydrin crosslinked βCD prepolymer (CycloLab R&D; Hungary) was dried overnight at ~70 °C prior to solubilization in 0.2 M KOH to make a 25% w/v solution by vortexing. After the polymer was fully dissolved, EGDGE (Mn ~174) was added dropwise and vortexed on high for 2 min. The CD/EGDGE mixture was added to 50 ml light paraffin oil (Fisher Scientific) and homogenized for 5 min at 13,000 RPM (Brinkman Kinematica Polytron PT3000 homogenizer equipped with a PT-DA 3012/2 TS blade). The polymer emulsion was then added to 100 ml light paraffin oil stirred at 400 RPM by an overhead stirrer (IKA-Werke Eurostar Euro-St PCV P4S1) to allow crosslinking.

After 48 h the stirrer was stopped and the oil was filtered off using a coarse frit glass filter. The crosslinkned pCD microspheres were washed successively with excess hexanes, excess acetone, and finished with excess Milli-Q grade water. Microspheres were allowed to dry at room temperature for 2 d.

2.4 - Drug Loading Determination

2.4.1 - Coated mesh

VM was chosen because it is a clinically relevant antibiotic and ease of loading into our polymer. Analysis of drug loading was performed as detailed previously [11]. Briefly, dry pCD-coated mesh samples were loaded in aqueous solution of VM (5 wt%) for 3 d at room temperature, dried, extensively extracted into water, and VM analyzed spectrophotometrically at 282 nm.

2.4.2 – Microspheres

Microspheres were immersed in VM (5 wt % in water) room temperature for 3 d. The drug-loaded microspheres were then briefly washed 3 times in pure water to remove surface-adsorbed drug and dried at room temperature for 2 d followed by vacuum drying. VM present in the samples was determined spectrophotometrically at 282 nm after extensive (>30d) extraction in water.

2.5 - Animals

All animal experimental protocols were approved by the Institutional Committee for Use and Care of Laboratory Animals (IACUC), Case Western Reserve University, Cleveland, OH. Female mice aged 6–8 wk (strain: C57BL/6J; The Jackson Laboratory, Bar Harbor, ME) were used. We used females instead of males to maintain consistency with our previous study[11]. Mice were maintained on a light/dark, 12/12 h cycle at 22°C, and received food and water ad libitum.

2.6 - Bacterial Inoculum Preparation

The same clinical strain (Seattle 1945) of SA transformed with a green fluorescent protein (GFP)-labeled plasmid to produce SA 1945GFPuvr was used in this study as previously[11]. Briefly, the strain was recovered from–80°C, inoculated in brain heart infusion, diluted 1:50 and cultured 2.5 h to 108 CFU/mL via optical density. The bacteria was then diluted to 105 CFU/mL in sterile saline, and the wound was inoculated with 100ul 105 CFU/ml GFP-labeled SA for a final inoculum of 104 CFU/mesh

2.7 - Mesh

Commercially available polyester mesh (Parietex TET) (Covidien, Mansfield, MA) was tailored into standard pieces (7mm* 7mm) prior to implantation into the mice’s dorsal wound. For the mesh plus microsphere groups, either VM-loaded microspheres or empty (i.e. unloaded) microspheres were placed on top of and adjacent to the mesh. For the polymer-coated mesh group, a piece of VM-loaded, pCD-coated mesh was implanted into the subcutaneous pocket on the animal. Control groups received uncoated mesh.

2.8 - Surgical model

After hair clipping and sterile preparation, a dorsal subcutaneous pocket was created in each mouse. A 7mm by 7mm piece of sterile multifilament polyester mesh was implanted into the wound for each group: mesh alone (n=16), VM-loaded pCD-coated mesh (n=20), VM-loaded microspheres (n=20), empty microspheres (n=10), or flushed with VM solution (n=7). All wounds were closed with sterile clippers prior to inoculation with 104 CFU of GFP-labeled SA (100 ul of 105 CFU/ml) into the wounds. VM flush was introduced into the surgical wound in a dose of 0.7 mg (100 mL of a 7 mg/L solution) following wound closure. All study groups were followed for 2 and 4 weeks before euthanasia for analysis. The two exceptions were the empty microspheres group that was only sacrificed at two weeks and the VM flush group that was only sacrificed at 4 weeks.

2.9 - Study Groups

Our study groups included: (1) control – mesh alone (n=16), (2) VM-loaded pCD-coated mesh (n=20), (3) VM-loaded microspheres (n=20); (4) empty microspheres (n=10), and (5) VM flush (n=7). Quantitative tissue/mesh cultures were done at 2 and 4-weeks postoperatively. A portion of the animals from the mesh alone group (n=10), the VM-loaded pCD-coated mesh group (n=10), and the VM-loaded microsphere group (n=10) were followed for 2 weeks prior to euthanasia. The rest were followed for 4 weeks prior to euthanasia. At necropsy, mesh with surrounding tissue was excised under sterile conditions and weighed. Each specimen was then placed in 1000 uL of sterile 0.9% normal saline and homogenized. Mice with open wounds and explanted mesh were excluded.

2.10 - Microbiologic Examination

Quantitative microbiologic culture was performed as detailed previously [11]. Briefly, serial 10-fold dilutions of the mesh and tissue homogenate were plated in duplicate, cultured for 72 h and counted. For plates showing no growth, the minimum number of detectable colonies (300 CFU/g) was used in statistical analyses. Cultured bacteria was checked for green fluorescence, and standard microbiological methods used to identify those not displaying green fluorescence.

2.11 - Statistical Analysis

Data were analyzed using STATA (ver. 10; College Station, TX, USA) as detailed previously [11]. Median and interquartile ranges were obtained for CFU/g counts. The minimum number of colonies detectable (300 CFU/g) was used for samples not showing bacterial growth. A P value < 0.05 was considered significant. Post-hoc analysis was performed when Kruskal-Wallis rank test was found to be significant.

3 - RESULTS

3.1 - Drug Loading Efficiency

Percent drug loading was calculated as: % drug loading = Wdrug / (Wsample + Wdrug) x 100 where Wdrug and Wsample are the weight of drug and sample, respectively. pCD-coated mesh loaded 11.8 ± 0.6% VM, which is consistent with previously published work using this HDI-crosslinked pCD [7]. Microspheres loaded 9.1 ± 0.7 % VM.

3.2 - Study groups

All animals survived throughout the duration of the study. One animal in the mesh alone control group (n=1/10), two animals in the VM-loaded microspheres (n=2/10), and two animals in the empty microspheres group (n=2/10) were excluded due to open wounds and explanted mesh. All excluded animals were in groups scheduled for sacrifice at 2 weeks.

3.3 - Microbiology Results

GFP SA was isolated from tissue-mesh homogenates of inoculated animals from the mesh alone, empty microsphere, and VM flush groups. Additionally, low numbers of GFP negative, coagulase negative Staphylococci were isolated from a few samples. These GFP negative Staphylococci were likely derived from skin contamination at the time of surgical necropsy and thus were excluded from analysis.

Median bacterial counts from homogenates at different survival time points were as follows. At 2 weeks: the empty microspheres group had the highest median GFP SA growth (n=8; 1.4*108CFU/g), followed by the mesh alone group (n=9; 8.6*106CFU/g). Both the VM-loaded pCD-coated mesh group (n=10; sterile cultures) and the VM-loaded microspheres group (n=8; sterile cultures) achieved complete clearance of GFP SA. The difference in median GFP SA bacterial counts at 2 weeks was statistically significant (P<0.05; Table 1). On post-hoc analysis of 2-week data, the mesh alone group and the empty microspheres group both had higher GFP SA bacterial counts than the VM-loaded pCD-coated mesh group and the VM-loaded microspheres group (P<0.05). No statistical difference was found at 2 weeks on post-hoc analysis between the mesh alone group and the empty microspheres group (P>0.05). At four weeks: the mesh alone group had the highest median GFP SA growth (n=6; 4.9*102CFU/g), followed by the VM flush group (n=7; 4.2*102CFU/g). Both the VM-loaded pCD-coated mesh group (n=10; sterile cultures) and the VM-loaded microspheres group (n=8; sterile cultures) had no GFP SA growth at 4 weeks. The difference in median GFP SA bacterial counts at 4 weeks was statistically significant (P<0.05; Table 2). On post-hoc analysis of 4-week data the mesh alone control group and the VM flush group had higher GFP SA bacterial counts than both the VM-loaded pCD-coated mesh group and the VM-loaded microspheres group (P<0.05). No statistical difference was found at 4 weeks on post-hoc analysis between the mesh alone group and the VM flush group (P>0.05).

Table 1. Two week median GFP positive SA bacterial growth from tissue and polyester mesh homogenates following various treatment regimens.

Mesh alone and empty microspheres groups had significantly higher GFP SA counts than VM-loaded pCD-coated mesh and VM-loaded microspheres groups (P<0.05). No statistical difference was found between mesh alone and empty microspheres (P>0.05) or between VM-loaded pCD-coated mesh and VM-loaded microspheres (P>0.05).

Study groups GFP+Animals Median(CFU/g) IQR(CFU/g) P value
Control – mesh alone (n=9) 9 8.6*106 1.2*108 P < 0.05
VM-loaded pCD-coated mesh (n=10) 0 0 0
VM-loaded microspheres (n=8) 0 0 0
Empty microspheres (n=8) 8 1.4*108 4.2*108
Open in a new tab

Abbreviations: VM = vancomycin; SA = Staphylococcus aureus; GFP = green fluorescent protein; pCD = crosslinked cyclodextrin polymer; CFU = colony forming units; IQR = interquartile range.

Table 2. Four week median GFP positive SA bacterial growth from tissue and polyester mesh homogenates following various treatment regimens.

Mesh alone and VM flush groups had significantly higher GFP SA counts than VM-loaded pCD-coated mesh and VM-loaded microspheres groups (P<0.05). No statistical difference was found between Mesh alone and empty microspheres groups (P>0.05) or between VM-loaded pCD-coated mesh and VM-loaded microspheres groups (P>0.05).

Study groups GFP+Animals Median(CFU/g) IQR(CFU/g) P value
Control - Mesh alone(n=6) 5 4.9*102 1.1*103 P < 0.05
VM-loaded pCD-coated mesh (n=10) 0 0 0
VM-loaded microspheres (n=8) 0 0 0
VM flush (n=7) 6 4.2*102 2.0*103
Open in a new tab

Abbreviations: VM = vancomycin; SA = Staphylococcus aureus; GFP = green fluorescent protein; pCD = crosslinked cyclodextrin polymer; CFU = colony forming units; IQR = interquartile range.

Comparisons were also made between 2 week and 4 week results for the VM-loaded pCD-coated mesh group and the VM-loaded microspheres group. There was no difference in bacterial growth recovered at 2 and 4 weeks for these two groups (P>0.05).

4 - DISCUSSION

Mesh infection remains a significant challenge facing hernia surgeons and lacks a gold-standard method for prevention and treatment. SA has been found to be one of the pathogens most commonly involved in mesh infection [13]. We previously created an affinity-based pCD drug delivery platform that demonstrated the capacity to release drug in a linear fashion, which allowed therapeutically active levels of drug to be maintained for a longer duration than with traditional drug impregnated polymers that release drugs via diffusion in a logarithmic fashion[7,8]. Our previous study demonstrated that polyester mesh coated with this affinity based, drug delivering pCD was able to effectively prevent a SA mesh infection with efficacy demonstrated at 2 and 4 weeks[11]. We have now demonstrated that this pCD polymer can be synthesized as microspheres and easily placed adjacent to implanted mesh. Compared to the pCD-coated mesh, the microspheres have an equivalent ability to prevent SA mesh infection at 2 and 4 weeks.

Despite the demonstrated efficacy of the pCD coating, using the polymer as a coating for prosthetics limits its clinical applicability because each different prosthetic coated with polymer requires individual FDA approval, which significantly affects physician choice and preference [12]. Our solution to this problem was to develop a drug delivery mechanism that would allow the antibiotic delivering polymer to be administered adjacent to the mesh or prosthetic, but not as part of the mesh or prosthetic, without interfering with the surgical repair. We synthesized the antibiotic delivering polymer as isolated microspheres that could be administered separately from the prosthetic and conducted a non-inferiority analysis. Altering the polymer from a coating to microspheres did not impair the ability of the polymer to prevent mesh infection in vivo in a mouse model at 2 and 4 weeks following mesh implantation.

Compared to other polymers, our pCD offers a more linear drug release profile, has demonstrated long-term antimicrobial activity beyond 7 months and in vivo efficacy out to 28 days, and has the potential to provide full wound area coverage. Logarithmic release of antibiotic produces a spike in drug concentration that quickly drops below therapeutic levels, whereas linear release can maintain active drug concentrations longer, which is ideal when treating hernia infections that can arise weeks after mesh implantation. Most currently available polymers do not possess ideal linear release patterns. Guillaume et al. 2012 developed an ofloxacin and rifiampicin impregnated biodegradable polymer that could be airbrushed onto mesh; however, despite triple coating the mesh to improve antibiotic release, 25% of the rifampicin and 40% of the ofloxacin were released in the first 5 hours [14]. The triple coated mesh effectively prevented infection throughout the entire experiment, but the study was concluded after only 72 hours [14]. Laurent et al. 2011 studied a ciprofloxacin impregnated polymer mesh coating over 24 hours and found a logarithmic decrease in the zone of inhibition for S. aureus, S. epidermidis, and E. coli, with the zone of inhibition completely disappearing by 24 hours for S. epidermidis [15]. Letouzey et al. 2011 examined an amoxicillin and ofloxacin polymer over a longer period of 10 days, but found a rapid, logarithmic release of both antibiotics over the first 2 days followed by low level release dropping to zero by day 8 ½ [16]. Antibacterial efficacy was only demonstrated out to 72 hours [16].

The polymer with performance most similar to ours is a VM impregnated coating studied by Fernandez-Gutierrez et al. 2013 that demonstrated an antimicrobial effect for at least 14 days in an agar diffusion test and prevented infection in a rabbit model out to 30 days [17]. The polymer was shown to produce near-linear release of antibiotic over 40 days, similar to our pCD with demonstrated near-linear release over 45 days [7,8,17]. This polymer and our pCD both possess ideal release profiles and have demonstrated long-term efficacy in animal models. Unlike our pCD, the polymer studied by Fernandez-Gutierrez et al. 2013 has not demonstrated antimicrobial activity out to 7 months, and it has not been successfully synthesized as microspheres, limiting the polymer’s versatility [10,17].

Antibacterial products with demonstrated efficacy that have received FDA approval include the Xenmatrix AB graft and the AIGISRx ST polymer coated mesh [18,19]. Both of these products release minocycline and rifampin and have only been studied out to 7 days post-implantation [18,19]. The Xenmatrix AB graft initially releases antibiotic rapidly with approximately 50% of rifampin and 40% of minocycline eluted in the first 24 hours, followed by a more linear release up to day 7 [18]. It has been shown to prevent graft infection in a rabbit model for 7 days [18]. The AIGISRx mesh releases 80% of antibiotics in the first 24 hours[20], and when used as an envelope to surround an implantable battery pack, it prevents infection for 7 days [17]. Other FDA approved, drug releasing products such as Vicryl suture and DualMesh Plus have failed to consistently demonstrate their antimicrobial efficacy [21,22,23]. Compared to our pCD, current FDA approved antibiotic releasing products have less desirable drug release profiles and have not demonstrated long-term antimicrobial activity or long-term efficacy in an animal model.

Finally, the ability of our polymer to be synthesized as microspheres that distribute throughout the wound site means that the antibacterial effect of our polymer is not limited to the immediate proximity of the prosthetic. Our microspheres can disperse throughout the wound to provide full area antibacterial coverage, unlike polymers used as prosthetic coatings.

In addition to ideal drug release properties, our polymer possesses additional chemical properties that make this new technology adaptable, safe, and convenient. First, the polymer can be chemically adapted to accommodate the loading and delivery of a diverse range of antimicrobials[8]. Second, the polymer is composed of cyclodextrin and ethylene glycol, both of which safely degrade into biocompatible components that are physiologically eradicated[8]. Third, the polymer releases antibiotic directly at the desired site of action decreasing the risk of systemic side effects. Fourth, the microspheres are convenient to use compared to other common surgical adjuncts, such as hemostatic agents. The microspheres can be prepared ahead of time, unlike products such as Flowseal that must be prepared less than 8 hours before use, and there is no need to wash away excess product to avoid adverse events, which is required when using Flowseal or Aristra [24,25].

5 - CONCLUSION

pCD microspheres were easily placed adjacent to prosthetic mesh in a small animal subcutaneous model. These pCD microspheres loaded with VM were able to prevent an SA prosthetic mesh infection as effectively as the previously validated pCD-coated mesh loaded with antibiotic. pCD microspheres capable of affinity based drug delivery are a promising technology that allows local drug delivery adjacent to prosthetic surgical devices. Large animal studies and reintroduction of bacteria at staged time intervals will help elucidate the full potential of this delivery platform.

Figure 1. Scanning electron microscope image of microspheres.

Figure 1

Open in a new tab

Image taken with accelerating voltage of 5kV. Microsphere size equals 256 ± 74 μm.

Acknowledgments

Research reported in this publication was supported by National Institute of General Medical Sciences of the National Institutes of Health under award number R43GM100525.

Footnotes

Author contributions: Sean Zuckerman, Horst A. von Recum, and Julius N. Korley developed the polymer, synthesized it as microspheres, and experimentally determined its characteristics. They helped with analysis and interpretation of data and they edited the manuscript. Kevin T. Grafmiller, Clayton Petro, Lijia Liu, and Michael J. Rosen helped conceive, design, and conduct the study and analyzed and interpreted the data. They drafted and edited the manuscript. All authors gave approval for submission of this manuscript.

DISCLOSURES

Research reported in this publication was supported by National Institute of General Medical Sciences of the National Institutes of Health under award number R43GM100525. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Michael J. Rosen is a Bard speaker, Ariste Medical board member, and has received research support from W.L. Gore & Associates, Inc. and from Miromatrix Medical, Inc.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Burger JW, Luijendijk RW, Hop WC, et al. Long-term follow-up of a randomized controlled trial of suture versus mesh repair of incisional hernia. Ann Surg. 2004;240:578–83. doi: 10.1097/01.sla.0000141193.08524.e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Sanchez VM, Abi-Haidar YE, Itani KM. Mesh infection in ventral incisional hernia repair: incidence, contributing factors, and treatment. Surg Infect (Larchmt) 2011;12:205–10. doi: 10.1089/sur.2011.033. [DOI] [PubMed] [Google Scholar]
  • 3.Hawn MT, Gray SH, Snyder CW, et al. Predictors of mesh explantation after incisional hernia repair. Am J Surg. 2011;202:28–33. doi: 10.1016/j.amjsurg.2010.10.011. [DOI] [PubMed] [Google Scholar]
  • 4.Aguilar B, Chapital AB, Madura JA, et al. Conservative management of mesh-site infection in hernia repair. J Laparoendosc Adv Surg Tech A. 2010;20:249–52. doi: 10.1089/lap.2009.0274. [DOI] [PubMed] [Google Scholar]
  • 5.Falagas ME, Vergidis PI. Irrigation with antibiotic-containing solutions for the prevention and treatment of infections. Clin Microbiol Infect. 2005;11:862. doi: 10.1111/j.1469-0691.2005.01201.x. [DOI] [PubMed] [Google Scholar]
  • 6.Rios A, Rodriguez JM, Munitiz V, et al. Antibiotic prophylaxis in incisional hernia repair using a prosthesis. Hernia. 2001;5:148–52. doi: 10.1007/s100290100026. [DOI] [PubMed] [Google Scholar]
  • 7.Thatiparti TR, Shoffstall AJ, Von Recum HA. Cyclodextrin based device coatings for affinity-based release of antibiotics. Biomaterials. 2010;31:2335–47. doi: 10.1016/j.biomaterials.2009.11.087. [DOI] [PubMed] [Google Scholar]
  • 8.Thatiparti TR, Von Recum HA. Cyclodextrin complexation for affinity-based antibiotic delivery. Macromol Biosci. 2010;10:82–90. doi: 10.1002/mabi.200900204. [DOI] [PubMed] [Google Scholar]
  • 9.Wang NX, von Recum HA. Affinity-based drug delivery. Macromol Biosci. 2011;11:321–32. doi: 10.1002/mabi.201000206. [DOI] [PubMed] [Google Scholar]
  • 10.Halpern JM, Gormley CA, Keech M, von Recum HA. Thermomechanical properties, antibiotic release, and bioactivity of a sterilized cyclodextrin drug delivery system. J Master Chem B Mater Biol Med. 2014;2:2764–2772. doi: 10.1039/C4TB00083H. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Harth KC, Rosen MJ, Thatiparti TR, et al. Antibiotic-releasing mesh coating to reduce prosthetic sepsis: an in vivo study. J Surg Res. 2010;163:337–43. doi: 10.1016/j.jss.2010.03.065. [DOI] [PubMed] [Google Scholar]
  • 12.FDA (21 CFR 807.81(a)(3))
  • 13.Mangram AJ, Horan TC, Pearson ML, et al. Guideline for Prevention of Surgical Site Infection, 1999. Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee. Am J Infect Control. 1999;27:97–132. [PubMed] [Google Scholar]
  • 14.Guillaume O, Garric X, Lavigne JP, et al. Multilayer, degradable coating as a carrier for the sustained release of antibiotics: Preparation and antimicrobial efficacy in vitro. J Control Release. 2012;162:492–501. doi: 10.1016/j.jconrel.2012.08.003. [DOI] [PubMed] [Google Scholar]
  • 15.Laurent T, Kacem I, Blanchemain N, et al. Cyclodextrin and maltodextrin finishing of a polypropylene abdominal wall implant for the prolonged delivery of ciprofloxacin. Acta Biomater. 2011;7:3141–3149. doi: 10.1016/j.actbio.2011.04.020. [DOI] [PubMed] [Google Scholar]
  • 16.Letouzey V, Lavigne JP, Garric X, et al. Is degradable antibiotic coating for synthetic meshes provide protection against experimental animal infection after fascia repair? J Biomed Mater Res B Appl Biomater. 2012;100:471–479. doi: 10.1002/jbm.b.31973. [DOI] [PubMed] [Google Scholar]
  • 17.Fernandez-Gutierrez M, Olivares E, Pascual G, et al. Low-density polypropylene meshes coated with resorbable and biocompatible hydrophilic polymers as controlled release agents of antibiotics. Acta Biomater. 2013;9:6006–6018. doi: 10.1016/j.actbio.2012.12.012. [DOI] [PubMed] [Google Scholar]
  • 18.Xenmatrix™ AB surgical graft antibacterial-coated regenerative collagen matrix brochure. Davol Inc., subsidiary of C. R. Bard, Inc., Warwick, RI. Copyright 2014. Hansen LK, Brown M,
  • 19.Johnson D, Palme li DF, Love C, Darouiche R. In vivo model of human pathogen infection and demonstration of efficacy by an antimicrobial pouch for pacing devices. Pacing Clin Electrophysiol. 2009 Jul;32(7):898–907. doi: 10.1111/j.1540-8159.2009.02406.x. [DOI] [PubMed] [Google Scholar]
  • 20.E-mail correspondance with Medtronic representative Randy Mansfield. 1/5/2016. Informaiton from unpublished in vivo data from finished good specifications.
  • 21.Gomezalonso A, Garciacriado F, Parrenomanchado F, et al. Study of the efficacy of Coated VICRYL Plus Antibacterial suture (coated Polyglactin 910 suture with Triclosan) in two animal models of general surgery. J Infect. 2007;54:82–8. doi: 10.1016/j.jinf.2006.01.008. [DOI] [PubMed] [Google Scholar]
  • 22.Chilukuri D, Shah J. Local delivery of vancomycin for the prophylaxis of prosthetic device-related infections. Pharm Res. 2005;22:563–72. doi: 10.1007/s11095-005-2497-7. [DOI] [PubMed] [Google Scholar]
  • 23.Carbonell AM, Matthews BD, Dreau D, Foster M, Austin CE, Kercher KW, Sing RF, Heniford BT. The susceptibility of prosthetic biomaterials to infection. Surg Endosc. 2005 Mar;19(3):430–5. doi: 10.1007/s00464-004-8810-4. [DOI] [PubMed] [Google Scholar]
  • 24.Arista™ AH absorbable hemostatic particles instructions for use. Davol Inc., subsidiary of C. R. Bard, Inc., Warwick, RI. Manufacturer: Medafor, Inc. Minneapolis, MN. Revision date: 5/22/2014.
  • 25.FLOSEAL hemostatic matrix instructions for use. Baxter Healthcare Corporation; Hayward, CA: Revision date: 6/5/2014. [Google Scholar]

RESOURCES