Heparin-mediated antibiotic delivery from an electrochemically-aligned collagen sheet

Abstract

BACKGROUND:

Implantable medical devices and hardware are prolific in medicine, but hardware associated infections remain a major issue.

OBJECTIVE:

To develop and evaluate a novel, biologic antimicrobial coating for medical implants.

METHODS:

Electrochemically compacted collagen sheets with and without crosslinked heparin were synthesized per a protocol developed by our group. Sheets were incubated in antibiotic solution (gentamicin or moxifloxacin) overnight, and in vitro activity was assessed with five-day diffusion assays against Pseudomonas aeruginosa. Antibiotic release over time from gentamicin-infused sheets was determined using in vitro elution and high performance liquid chromatography (HPLC).

RESULTS:

Collagen-heparin-antibiotic sheets demonstrated larger growth inhibition zones against P. aeruginosa compared to collagen-antibiotic alone sheets. This activity persisted for five days and was not impacted by rinsing sheets prior to evaluation. Rinsed collagen-antibiotic sheets did not produce any inhibition zones. Elution of gentamicin from collagen-heparin-gentamicin sheets was gradual and remained above the minimal inhibitory concentration for gentamicin-sensitive organisms for 29 days. Conversely, collagen-gentamicin sheets eluted their antibiotic load within 24 hours. Overall, heparin-associated sheets demonstrated larger inhibition zones against P. aeruginosa and prolonged elution profile via HPLC.

CONCLUSION:

We developed a novel, local antibiotic delivery system that could be used to coat medical implants/hardware in the future and reduce post-operative infections.

Keywords

Local antibiotic delivery electrochemically compacted collagen sheet collagen sheet with heparin implantable medical devices and hardware

1. Introduction

Implantable devices and hardware are used across many medical disciplines [1,2]. Infection represents a common complication associated with these materials due to local immune response suppression and biofilm formation [2–4]. Implants provide a surface where bacteria and fungi adhere and produce biofilms [5–7]. These biofilms create a barrier to antibiotic penetrance, elude the host’s immune system, and demonstrate antibiotic resistance [3,8].

Patients are typically treated with antibiotics at the time of implant insertion [1,9,10]. If they develop an infection post-operatively, treatment starts with oral or intravenous antibiotics [11], which may be extended and/or require hospitalization. If antibiotics fail to clear the infection, patients require removal of their device/hardware [3,12]. Risk factors for infection include diabetes, immunosuppression, renal failure [11], and poor nutrition. Considering the burden of these infections, increased use of implantable devices, and rising bacterial resistance, newer strategies are needed to prevent implant-related infections. Recently, local infection prophylaxis and treatment has garnered interest [4,5], and many systems have been developed to prevent infection of orthopedic and cardiac implants [13–16].

We sought to develop a novel, biologic antimicrobial coating that may be particularly suited for use with head and neck hardware as well as other medical devices. In this manuscript, we detail the design and synthesis of electrochemically compacted type 1 collagen sheets with covalently-bonded heparin and electrostatically-associated antibiotics. Heparin has previously been used for delivery of bioinductive growth factors via charge-based affinity [17,18]. To our knowledge, heparin has not been used for delivery of small antibiotic molecules.

The sheets are 100 μm thick, flexible, robust, and can be cut to size to coat hardware during surgery. We hypothesize these sheets will prevent infection in a two-fold manner: as a physical barrier to biofilm formation and through modulated, affinity-based local release of antibiotics. Our in vitro studies assess the antimicrobial activity of these sheets via diffusion assays against Pseudomonas aeruginosa (major cause of hardware infections). Additionally, we evaluate the in vitro temporal antibiotic release profile using high performance liquid chromatography (HPLC).

2. Materials and methods

2.1. Preparation of collagen and collagen-heparin sheets

Electrochemically compacted collagen sheets were synthesized as previously described [19]. Briefly, acid-soluble monomeric type I collagen solution from bovine hide (Collagen Solutions Inc., CA, USA) was dialyzed free of salts overnight. The solution was loaded between two planar carbon electrodes and 30 volts was applied for 2 minutes. Collagen monomers are mobilized electrokinetically within the electric field and compacted at their isoelectric point to form a densely packed sheet. Sheets were incubated in 10× phosphate buffered saline (PBS) for 30 minutes at 37 °C.

1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) and N-hydroxysuccinimide (NHS) (Life Technologies) were used to crosslink heparin to collagen sheets. Heparin sodium salt (molecular weight 6,000–30,000 Daltons, porcine intestinal mucosa, Celsus Inc.) was dissolved in EDC-NHS solution (EDC:NHS ratio 1.67:1, pH 7–8 in deionized (DI) water). Heparin concentration was 50 mg/mL and heparin:collagen molar ratio was 80:1. This heparin:collagen molar ratio was determined to maximize heparin crosslinking without noticeably altering the structural properties of the collagen sheet. Collagen sheets were incubated in heparin crosslinking solution for two hours. Sheets were rinsed three times for 5 seconds and a fourth for 15 minutes in DI water. A second crosslinking procedure was performed in EDC-NHS 80% ethanol solution without heparin for 30 minutes. Sheets were rinsed in DI water and stored in PBS at 4 °C.

Collagen and collagen-heparin sheets were dissolved in 1 N HCl at a temperature of 55 °C overnight. Dimethylmethylene blue (DMMB) assay for sulfated glycosaminoglycans confirmed the presence of heparin in the heparin associated sheets based on the purple color change of the sheet.

2.2. Bacterial diffusion assay

Collagen and collagen-heparin sheets were divided into 8 mm discs using punch biopsy. Discs were dried for 2 hours at 37 °C and incubated overnight in 150 μL of 2 mg/mL gentamicin sulfate solution (Sigma-Aldrich) at 4 °C to allow for electrostatic association of gentamicin.

Collagen-gentamicin and collagen-heparin-gentamicin discs were divided into two groups based on additional wash steps (unwashed and washed groups). Discs in the washed group were rinsed in 1 mL of DI water 5 times for five seconds. Next, P. aeruginosa (clinical isolate [20]) at 10⁻⁵ concentration was spread onto agar plates (100 μL/plate). Plates were dried for 60 minutes. Collagen discs were placed onto the agar plates, and the plates were incubated at 37 °C. Inhibition zones were photographed and measured (diameter) using digital calipers at 18 hours and 5 days post-plating.

To determine the applicability of this material over a range of antibiotics, moxifloxacin was also assessed. Collagen and collagen-heparin sheets were prepared as described above. Prior to the bacterial diffusion assay, sheets were incubated in moxifloxacin hydrochloride 0.5% (Vigamox solution) overnight instead of gentamicin. The same four groups were created and tested against P. aeruginosa using a bacterial diffusion assay.

2.3. In vitro elution and HPLC analysis

Collagen and collagen-heparin sheets were prepared as described above. Sheets were dried for 2 hours at 37 °C and weighed. Five experimental sheets were designed: (1) Collagen only, (2) Collagen-Gentamicin (50 mg/mL gentamicin solution), (3) Collagen-Heparin, (4) Collagen-Heparin-Gentamicin (50 mg/mL) and (5) Collagen-Heparin-Gentamicin (25 mg/mL). Two different concentrations of gentamicin were utilized in this initial analysis to determine the antibiotic release rate from the collagen-heparin sheets.

Groups 2, 4 and 5 were incubated in 5 mL of their respective gentamicin solution at 4 °C overnight. Sheets were then removed from solution, blotted with Kim Wipes, and weighed. Each sheet was placed into its own microcentrifuge tube with 1 mL of 1× PBS and stored at 37 °C. At different time points (2, 6 and 24 hours and every 48 hours until day 29) the elution volume was removed and replaced with fresh PBS. Elution samples were frozen at −20 °C until HPLC analysis.

To increase detection sensitivity for HPLC, gentamicin was derivatized with 2,4-dinitrofluorobenzene (DNFB, Sigma Aldrich) as described previously [21,22]. Briefly, 100 μL of sample (in PBS) or gentamicin standards (Sigma, diluted with PBS, 15.6--500 μg/mL) was mixed with 100 μL of 2% NaHCO₃ followed by 400 μL of acetonitrile. After vortexing, samples were centrifuged at 10,000 rpm for 5 minutes at room temperature. Four hundred μL of supernatant and 40 μL of 10% DNFB (V/V in acetonitrile) were transferred to a glass tube with screw cap. Glass tubes were placed in a 77 °C water bath for 45 minutes. Reaction was stopped by adding 25 μl of 1N HCl. HPLC separation was performed as described by Erdem et al. [21]. Derivatized gentamicin was applied to Perkin Elmer Flexar HPLC system, with C18 column (Perkin Elmer Peptide ES-C18, 4.6 × 100 mm). The mobile phase consists of 30% NaHCO₃ (0.1%) and 70% acetonitrile, with flow rate of 1 mL/minute for 10 minutes. Gentamicin component C_1a had a retention time of 5.2 minutes, and components C₁ and C₂ merged into one peak with a retention time of 5.6 minutes. The areas of the two peaks were analyzed with Chromera software (version 4.1.0.6386). The concentrations of the two peaks of samples were calculated according to the respective peaks of standards, and the mean value of the two peaks was recorded for the final gentamicin concentration. HPLC analysis was run in triplicate for each group and time point.

2.4. Statistical methods

Five or six replicates were run per group for the bacterial diffusion assays. Mean and standard deviation for inhibition zones were calculated for each group at 18-hour and 5-day time-points. To determine whether zone sizes were statistically different among the four groups within and across the two timepoints, a three-way analysis of variance (ANOVA) model was carried out using wash status, heparinization state and time point as the three variables. The three-way ANOVA was followed by a post-hoc Tukey’s test comparing the mean inhibition zone for each group to every other group across both time points. A P-value < 0.05 was considered statistically significant (Origin software). For HPLC, standard deviation for gentamicin concentration was calculated for each group at every time point.

Fig. 1.

Electrochemically compacted collagen sheet. (A) Photograph of an electrochemically compacted collagen sheet. Each sheet is pliable and measures approximately 2 × 7 cm and is 100 μm thick. (B) Example of collagen sheet cut to size and wrapped around a titanium plate that is used for mid-face fixation of facial fractures or osseous free flap reconstruction in head and neck surgery.

3. Results

3.1. Synthesis of collagen-heparin-antibiotic sheets

Collagen sheets were produced as described in the methods (Fig. 1A). Each sheet was 2 × 4 cm and 100 μm thick. Sheets are flexible, robust and can be cut to size to coat medical implants/hardware (Fig. 1B). DMMB assay confirmed binding of heparin to sheets based on the purple color change of collagen-heparin sheets (Fig. 2).

Fig. 2.

Dimethylmethylene blue (DMMB) assay for sulfated glycosaminoglycans. (A) DMMB staining in heparin–EDC/NHS crosslinking solution (top) and EDC-NHS crosslinking solution (bottom). (B,C) DMMB staining of heparin conjugated EDC/NHS crosslinked collagen sheet (top) and EDC-NHS crosslinked collagen sheet (bottom). Based on the purple color change, the presence of heparin in the heparinized collagen sheet is confirmed.

3.2. Bacterial diffusion assay with gentamicin

Growth inhibition zones against P. aeruginosa were photographed for all groups after 18 hours (Fig. 3). At this time, agar plates with unwashed collagen-gentamicin sheets produced inhibition zones measuring 1.42 ± 0.14 cm (mean ± SD) while the washed collagen-gentamicin sheets did not show any inhibition (Fig. 4). The unwashed collagen-heparin-gentamicin sheets had a 2.42 ± 0.10 cm inhibition zone, which was not statistically different from the washed collagen-heparin-gentamicin sheets (2.40 ± 0.05 cm, P = 1.00). Both unwashed and washed collagen-heparin-gentamicin groups had statistically significant larger inhibition zones compared to the unwashed collagen-gentamicin group (P < 0.01 both comparisons).

Fig. 3.

Agar plates at 18 hours from bacterial diffusion assay with gentamicin. Example of bacterial diffusion assay with P. aeruginosa. These two agar plates have collagen-gentamicin (A) or collagen-heparin-gentamicin (B) discs that were either unwashed (left half) or washed (right half) prior to placement on the plate. The growth inhibition zone appears most robust on the plate with the collagen-heparin-gentamicin discs whether they were washed or not washed prior to placement on the agar plate.

Fig. 4.

Growth inhibition zones from bacterial diffusion assay with gentamicin. A 10⁻⁵ concentration of P. aeruginosa was plated onto agar plates and dried for 60 minutes. Four different formulations of collagen discs were applied to the plates. Plates were incubated at 37 °C, and the diameter across the inhibition zone was measured at 18 hours and five days. The bar graphs display the mean for each group along with the error bars representing standard deviation. There is an “x” for the washed collagen-gentamicin group as it did not show any measurable zone of inhibition. Zones were most robust for the collagen-heparin-gentamicin discs, and this persisted up to five days (*P < 0.05, **P < 0.01).

At day five, the unwashed collagen-gentamicin group showed a stable inhibition zone at 1.39 ± 0.12 cm (Fig. 4). This was not statistically different compared to the inhibition zone for this same group at 18 hours (P = 1.00). The unwashed and washed collagen-heparin-gentamicin sheets maintained stable inhibition zones at five days with measurements of 2.08 ± 0.44 cm and 2.26 ± 0.33 cm, respectively. Variability in bacterial lawn growth and doubling time are reflected in the size of the inhibition zone and associated standard deviation. Even with the predictable variability in P. aeruginosa growth, the observations were statistically similar compared to their 18 hour time-points for unwashed (P = 0.12) and washed (P = 0.98) groups, respectively. Finally, washed collagen-gentamicin sheets did not show any bacterial growth inhibition after five days.

3.3. Bacterial diffusion assay with moxifloxacin

An identical study was carried out assessing growth inhibition zones against P. aeruginosa using moxifloxacin associated collagen sheets (Fig. 5). Once again, at 18 hours and five days the washed collagen-moxifloxacin sheets did not generate any inhibition zone. The unwashed collagen-moxifloxacin discs had stable inhibition zones of 1.85 ± 0.09 cm and 1.72 ± 0.09 cm at 18 hours and five days, respectively (P = 0.99). The collagen-heparin-moxifloxacin sheets trended towards a larger inhibition zone at 18 hours: unwashed group was 2.28 ± 0.29 cm and washed group 2.10 ± 0.23 cm. Compared to the unwashed collagen-moxifloxacin group at the same time point, the unwashed group showed statistically significant growth inhibition (P = 0.02), while the washed group did not reach statistical significance (P = 0.51).

Fig. 5.

Growth inhibition zones from bacterial diffusion assay with moxifloxacin. P. aeruginosa at a concentration of 10⁻⁵ was plated onto agar plates and collagen discs of four different formulations were applied to the plates. Plates were incubated at 37 °C, and zones of inhibition were measured at 18 hours and five days. The bar graphs display the mean for each group along with the error bars depicting standard deviation. The washed collagen-moxifloxacin group did not show any zone of inhibition, so it is marked in the graph with a “x”. Growth inhibition zones were most robust for the collagen-heparin-moxifloxacin discs, and this continued up to five days (*P < 0.05, **P < 0.01).

At five days, the inhibition zones for the collagen-heparin-moxifloxacin groups remained robust: 2.53 ± 0.24 cm for unwashed and 2.18 ± 0.22 cm for washed. These were statistically unchanged compared to their 18 hour time-points for unwashed (P = 0.52) and washed (P = 0.99) groups. Finally, as compared to the unwashed collagen-moxifloxacin group at five days, both of the collagen-heparin-moxifloxacin cohorts had statistically significant larger inhibition zones (P < 0.01 both comparisons).

3.4. High performance liquid chromatography

Control groups that were not incubated in gentamicin solution (collagen alone and collagen-heparin sheet) produced elution samples that did not show detectable gentamicin at any timepoint (data not shown). The group without heparin (collagen-gentamicin with 50 mg/mL gentamicin solution) only eluted detectable gentamicin within the first 24 hours (Fig. 6). On the other hand, the two collagen-heparin-gentamicin groups eluted both the highest initial quantities of gentamicin and continued to elute gentamicin above the reported minimum inhibitory concentration (MIC) for gentamicin-sensitive organisms (4 μg/mL) [23] throughout the duration of the experiment. Importantly, comparing collagen-heparin-gentamicin (50 mg/mL) to collagen-gentamicin (50 mg/mL) at the 2 hour time-point revealed that heparinization resulted in a 1.8-fold greater uptake of antibiotic at the time of initial loading (1425 versus 802 μg/mL, respectively).

Fig. 6.

High performance liquid chromatography quantification of gentamicin elution. Elution was performed on collagen-gentamicin and collagen-heparin-gentamicin sheets as depicted in this line graph. The collagen-gentamicin sheet only eluted detectable gentamicin for the first 24 hours, while the groups with heparin eluted detectable gentamicin more slowly over the entire time course of this experiment.

4. Discussion

Implantable medical devices and hardware are ubiquitous in medicine, but they remain plagued by infectious complications [2,3]. We sought to develop and provide initial in vitro validation of a novel local antibiotic delivery system. First, we synthesized electrochemically-compacted collagen sheets conjugated with heparin and electrostatically loaded with antibiotic. Next, we assessed their antimicrobial efficacy and elution profile.

Antimicrobial efficacy was demonstrated against P. aeruginosa in bacterial diffusion assays using gentamicin and moxifloxacin-loaded collagen-heparin sheets. Pseudomonal growth inhibition lasted the entire five-day experiment, and heparinized collagen sheets had greater anti-microbial efficacy compared to sheets without heparin. Furthermore, collagen-heparin-antibiotic sheets retained their function even if they were washed before placement on the plates, while the collagen-antibiotic sheets had no antimicrobial activity after rinsing. Thus, heparinization enhances the association between the antibiotic and collagen sheet, resulting in improved therapeutic function.

Interestingly, in the bacterial diffusion assays there was an increase in the standard deviation for gentamicin, but not for moxifloxacin, between 18 hours and 5 days (Figs 4 and 5). We can only speculate about the factors that may have led to this outcome. It is possible that there may have been plate to plate variations in bacterial colony density that could have become more pronounced for the gentamicin group as the time progressed. Alternatively, the heparinization of the collagen sheets, and by extension the antibiotic uptake, may not have been as homogeneous for this particular group, which could have resulted in greater variation in growth inhibition zone sizes.

The sheet’s drug elution profile was characterized using HPLC analysis of an in vitro elution assay. Collagen-gentamicin sheets eluted detectable gentamicin for only 24 hours while the collagen-heparin-gentamicin groups eluted antibiotic above the reported MIC for gentamicin-sensitive organisms for the full 29-day experiment. This highlights that heparinization of the collagen sheet leads to greater association with the antibiotic and a slower release of gentamicin, improving sustainable antimicrobial efficacy of gentamicin at the site of release. The absence of the heparin linker allows for an initial blast of gentamicin (Fig. 6), whereas the heparin coated product sustains release and allows for extended potency. Clinically, this slow antibiotic release will be important to ensure surgical wounds remain sterile post-operatively to prevent biofilm formation.

It is necessary to highlight the complementary results that the bacterial diffusion assays and HPLC analysis hold in this work. In the HPLC analysis, the collagen-gentamicin sheets, in the absence of heparin, eluted all of their antibiotic within 24 hours, but during the diffusion assay these sheets were still able to inhibit bacterial growth up to five days. This apparent discrepancy is a result of methodological differences. For the HPLC analysis the PBS used to incubate the samples was fully exchanged at each time point. For the diffusion assays there was no such exchange, so none of the original antibiotic was ever washed off the plate. This could lead to persistent bacterial growth inhibition even if the sheet was no longer eluting antibiotic. Indeed, these analyses demonstrate two facets: the bacterial diffusion assays showed the efficacy of the formulation in vitro while the HPLC analysis characterized the antibiotic release profile. In this manner, the HPLC results could not be used to predict the outcome of the bacterial diffusion assay reliably, but rather both of these analyses are needed to understand the bigger picture on the potential clinical utility.

Heparin has been utilized in medical biomaterials previously. This includes synthetic vascular grafts that take advantage of the anti-coagulant properties of heparin to decrease thrombogenicity [24]. Recently, groups have employed the functional groups of heparin to conjugate a variety of molecules, including cancer therapeutics to improve biocompatibility and therapeutic efficacy [25]. Additionally, heparin has been used to design affinity-based drug delivery systems. These systems utilize non-covalent interactions to hold medications, growth factors or other molecules within a biological matrix, protecting their activity and slowing their diffusion [17,18].

We used heparin to create an electrostatic association with each antibiotic to improve biocompatibility and generate a slow release of the medication. This represents an affinity-based drug delivery system, and it is promising that we were able to associate multiple antibiotics with this sheet. Certainly, other antibiotics, antibiotic combinations or anti-fungals [26] could be used with this sheet for different indications in the future. Based on our literature review, heparin has not been used for delivery of small antibiotic molecules previously.

Numerous local antibiotic delivery systems have been developed, with advances especially in orthopedics and cardiology. Medtronic developed a bio-absorbable, anti-bacterial envelope containing minocycline and rifampin for cardiac devices [13]. Orthopedic groups have studied vancomycin-modified titanium plates [14], gentamicin-impregnated hydroxyapatite cement [27], minocycline and rifampin coated titanium pins [15], and many other formulations to either treat or prevent infections across multiple animal models. Indeed many local antibiotic delivery systems exist, and their designs are specific for certain medical specialties.

We did not identify any local antibiotic delivery systems for use with head and neck hardware. Many procedures in head and neck surgery violate the oral cavity, exposing the surgical sites and hardware to oral flora leading to increased rates of post-operative infections. A meta-analysis of hardware complications for patients with facial fractures requiring open reduction and internal fixation (ORIF) reported a 10.3% complication rate. Infection represented almost half of the complications and was the primary reason for hardware removal [12]. For head and neck cancer patients undergoing osseous free flap reconstruction, early infection carries a significant risk of late flap failure [28]. Moreover, exposure and infection represent the most common reasons for hardware removal [29]. Due to the frequency of P. aeruginosa related infections in head and neck surgery, we carried out these initial studies with P. aeruginosa and anti-pseudomonal antibiotics [10,30,31].

We propose wrapping head and neck hardware or other medical devices in collagen-heparin-antibiotic sheets prior to implantation. The goal is to sterilize the surgical bed as the antibiotic is slowly released over time and thereby prevent early bacterial contamination of the hardware. Synthesizing these sheets is quick and wrapping the hardware in the operating room would add minimal time to surgery. This is an early proposed model of delivery and may be modified as the collagen sheets are further optimized. Electrocompaction of type I collagen is biocompatible and biodegradable in vivo [32]. Moreover, this biologic coating may enhance wound healing [33,34]. All in all, we believe our local antibiotic delivery system could be beneficial in head and neck surgery as well as for other devices/hardware in the future.

There are several limitations to this study. As this was an initial in vitro assessment, we did not evaluate all pathogenic organisms nor all antibiotics that are clinically relevant. It will be important to examine the ability for this collagen-heparin sheet to associate with a range of antibiotics and assess efficacy against a spectrum of bacteria. Second, we did not determine the in vitro elution of the collagen-heparin-moxifloxacin sheets via HPLC. It would be valuable to understand if this slow antibiotic elution is maintained with antibiotics other than gentamicin. Finally, we did not delve into an in vivo assessment of this sheet.

Although the in vitro work is promising, there are many avenues that will be explored in the future before this system can be employed clinically. We developed an in vivo infection model in mice using P. aeruginosa and head and neck titanium plates and will assess the ability for this sheet to reduce bacterial burden, biofilm formation and rate of clinical infection in the mice. Next, we need to verify this sheet does not impair osseointegration or wound healing in an animal fracture model. Lastly, we plan to evaluate the degradation profile of the collagen sheet to determine how long the antibiotic remains active in vivo. Once we have robust in vitro and in vivo results, we can move towards investigating the utility of this sheet to prevent hardware associated infections in clinical trials.

Footnotes

Acknowledgements

We would like to acknowledge the Clinical and Translational Science Collaborative Bioanalyte Core for their assistance with both bacteriology and high performance liquid chromatography analyses carried out in this study.

Conflict of interest

No benefit of any kind was received either directly or indirectly by the authors.

References

Atreya

A.R.

Cook

J.R.

and Lindenauer

P.K.

, Complications arising from cardiac implantable electrophysiological devices: Review of epidemiology, pathogenesis and prevention for the clinician, Postgr Med 128 (2016), 223–230.

VanEpps

J.S.

and Younger

J.G.

, Implantable device-related infection, Shock 46 (2016), 597–608.

Vinh

D.C.

and Embil

J.M.

, Device-related infections: A review, J Long Term Eff Med Implant 15 (2005), 467–488.

Schierholz

J.M.

and Beuth

, Implant infections: A haven for opportunistic bacteria, J Hosp Infect 49 (2001), 87–93.

Arciola

C.R.

Campoccia

Speziale

Montanaro

and Costerton

J.W.

, Biofilm formation in Staphylococcus implant infections. A review of molecular mechanisms and implications for biofilm-resistant materials, Biomaterials 33 (2012), 5967–5982.

Dibartola

A.C.

Swearingen

M.C.

Granger

J.F.

Stoodley

and Dusane

D.H.

, Biofilms in orthopedic infections: A review of laboratory methods, APMIS 125 (2017), 418–428.

Chandra

Kuhn

D.M.

Mukherjee

P.K.

Hoyer

L.L.

McCormick

and Ghannoum

M.A.

, Biofilm formation by the fungal pathogen Candida albicans: Development, architecture, and drug resistance, J Bacteriol 183 (2001), 5385–5394.

Davidson

Poon

Saunders

Shapiro

I.M.

Hickok

N.J.

and Adams

C.S.

, Tetracycline tethered to titanium inhibits colonization by Gram-negative bacteria, J Biomed Mater Res B Appl Biomater 103(7) (2015), 1381–1389.

Domingo

Dale

Gao

Groves

Stanley

Maxwell

R.A.

, A single-center retrospective review of postoperative infectious complications in the surgical management of mandibular fractures: Postoperative antibiotics add no benefit, J Trauma Acute Care Surg 81 (2016), 1109–1114.

10.

Durand

M.L.

Yarlagadda

B.B.

Rich

D.L.

Lin

D.T.

Emerick

K.S.

Rocco

J.W.

, The time course and microbiology of surgical site infections after head and neck free flap surgery, Laryngoscope 125 (2015), 1084–1089.

11.

Nielsen

J.C.

Gerdes

J.C.

and Varma

, Infected cardiac-implantable electronic devices: Prevention, diagnosis, and treatment, Eur Hear J 36 (2015), 2484–2490.

12.

Hernandez Rosa

Villanueva

N.L.

Sanati-Mehrizy

Factor

S.H.

and Taub

P.J.

, Review of maxillofacial hardware complications and indications for salvage, Craniomaxillofac Trauma Reconstr 9 (2016), 134–140.

13.

Kolek

M.J.

Patel

N.J.

Clair

W.K.

Whalen

S.P.

Rottman

J.N.

Kanagasundram

, Efficacy of a bio-absorbable antibacterial envelope to prevent cardiac implantable electronic device infections in high-risk subjects, J Cardiovasc Electrophysiol 26 (2015), 1111–1116.

14.

Stewart

Barr

Engiles

Hickok

N.J.

Shapiro

I.M.

Richardson

D.W.

, Vancomycin-modified implant surface inhibits biofilm formation and supports bone-healing in an infected osteotomy model in sheep: A proof-of-concept study, J Bone Jt Surg Am 94 (2012), 1406–1415.

15.

Darouiche

R.O.

Mansouri

M.D.

Zakarevicz

Alsharif

and Landon

G.C.

, In vivo efficacy of antimicrobial-coated devices, J Bone Jt Surg Am 89 (2007), 792–797.

16.

Alt

Bitschnau

Osterling

Sewing

Meyer

Kraus

, The effects of combined gentamicin-hydroxyapatite coating for cementless joint prostheses on the reduction of infection rates in a rabbit infection prophylaxis model, Biomaterials 27 (2006), 4627–4634.

17.

Sakiyama-Elbert

S.E.

, Incorporation of heparin into biomaterials, Acta Biomater 10(4) (2014), 1581–1587.

18.

Vulic

and Shoichet

M.S.

, Affinity-based drug delivery systems for tissue repair and regeneration, Biomacromolecules 15 (2014), 3867–3880.

19.

Younesi

Islam

Kishore

Panit

and Akkus

, Fabrication of compositionally and topographically complex robust tissue forms by 3D-electrochemical compaction of collagen, Biofabrication 7 (2015), 35001.

20.

Bonfield

T.L.

Hodges

C.A.

Cotton

C.U.

and Drumm

M.L.

, Absence of the cystic fibrosis transmembrane regulator (Cftr) from myeloid-derived cells slows resolution of inflammation and infection, J Leukoc Biol 92 (2012), 1111–1122.

21.

Erdem

Gundogan

N.U.

Usubutun

Kilinc

Erdem

S.R.

Kara

, The protective effect of taurine against gentamicin-induced acute tubular necrosis in rats, Nephrol Dial Transpl 15 (2000), 1175–1182.

22.

Ryan

J.A.

, Colorimetric determination of gentamicin, kanamycin, tobramycin, and amikacin aminoglycosides with 2,4-dinitrofluorobenzene, J Pharm Sci 73 (1984), 1301–1302.

23.

Stravinskas

Horstmann

Ferguson

Hettwer

Nilsson

Tarasevicius

, Pharmacokinetics of gentamicin eluted from a regenerating bone graft substitute: In vitro and clinical release studies, Bone Jt Res 5(9) (2016), 427–435.

24.

Hoshi

R.A.

Van Lith

Jen

M.C.

Allen

J.B.

Lapidos

K.A.

and Ameer

, The blood and vascular cell compatibility of heparin-modified ePTFE vascular grafts, Biomaterials 34 (2013), 30–41.

25.

Liang

and Kiick

K.L.

, Heparin-functionalized polymeric biomaterials in tissue engineering and drug delivery applications, Acta Biomater 10 (2014), 1588–1600.

26.

Zumbuehl

Ferreira

Kuhn

Astashkina

Long

Yeo

, Antifungal hydrogels, Proc Natl Acad Sci USA 104 (2007), 12994–12998.

27.

Solberg

Gutow

and Baumgaertner

, Efficacy of gentamycin-impregnated resorbable hydroxyapatite cement in treating osteomyelitis in a rat model, J Orthop Trauma 13(2) (1999), 102–106.

28.

Zender

C.A.

Mehta

Pittman

A.L.

Feustel

P.J.

and Jaber

J.J.

, Etiologic causes of late osteocutaneous free flap failures in oral cavity cancer reconstruction, Laryngoscope 122 (2012), 1474–1479.

29.

Day

K.E.

Desmond

Magnuson

J.S.

Carroll

W.R.

and Rosenthal

E.L.

, Hardware removal after osseous free flap reconstruction, Otolaryngol - Head Neck Surg (United States) 150(1) (2014), 40–46.

30.

Yarlagadda

B.B.

Deschler

D.G.

Rich

D.L.

Lin

D.T.

Emerick

K.S.

Rocco

J.W.

, Head and neck free flap surgical site infections in the era of the Surgical Care Improvement Project, Head Neck 38 Suppl 1 (2016), E392–E398.

31.

Zhou

Zeng

Qing

and Tang

, Retrospective study on multidrug-resistant bacterium infections after rigid internal fixation of mandibular fracture, J Oral Maxillofac Surg 74 (2016), 770–777.

32.

Kishore

Uquillas

J.A.

Dubikovsky

Alshehabat

M.A.

Snyder

P.W.

Breur

G.J.

, In vivo response to electrochemically aligned collagen bioscaffolds, J Biomed Mater Res B Appl Biomater 100 (2012), 400–408.

33.

Varga

Sixta

Bem

Matia

Jirkovska

and Adamec

, Application of gentamicin-collagen sponge shortened wound healing time after minor amputations in diabetic patients - A prospective, randomised trial, Arch Med Sci 10 (2014), 283–287.

34.

Yetim

Ozkan

O.V.

Dervisoglu

Erzurumlu

and Canbolant

, Effect of local gentamicin application on healing and wound infection in patients with modified radical mastectomy: A prospective randomized study, J Int Med Res 38 (2010), 1442–1447.