www.advhealthmat.de www.MaterialsViews.

com

FULL PAPER

Mitigation of Staphylococcus aureus-Mediated Surgical Site Infections with IR Photoactivated TiO2 coatings on Ti Implants
Asem Aboelzahab, Abdul-Majeed Azad,* Shawn Dolan, and Vijay Goel
such as spinal tumors.[3] One study pertaining to the assessment of the cost and Surgical site infections caused by methicillin-resistant and methicillin-susceptible length of patient hospitalization found Staphylococcus aureus (MRSA, MSSA) lead to patient hospitalization for an that SSI onset after surgery resulted in extended period coupled with concomitant hospitalization resources and cost. an average increase in cost up to $26,000 The detrimental effect resulting from the onset of these infections poses great per patient, in addition to longer hospihealth risks, leading to death in some instances. Titanium dioxide (TiO2) is talization for 8 days.[4] These outstanding endowed with the unique capability of photoactivity which has been extensively outcomes of SSIs make it increasingly important to focus attention on ways to exploited in antibacterial activities. It has been shown to be very effective in its mitigate the very onset of infection, folbactericidal efcacy against infection-causing bacterial strains, namely, E. coli lowing the surgery. It is therefore, relevant and S. aureus. In this study, the use of IR-photoactivated TiO2 nanocoatings on and particularly important to explore new titanium implants to mitigate the onset of surgical site infections is described. and radical strategies in lieu of traditional TiO2 coatings were created on implantable materials by way of an aqueous treatments via antibiotics that have been shown to fail in many cases to eradicate plasma electrodeposition technique and were used to mitigate the harmful bactethe infection buildup fully.[59] rial growth upon brief activation by an infrared (IR) laser source. The necrosis of In this connection, systematic research S. aureus cells was found to exceed 90% within 30 min. following a 30s exposure was launched where the unique photoacof the titania-coated model implants (Ti mesh and plate). Promising potential of tive attributes of nanostructured titanium antibacterial coatings in mitigating surgical site infections has been shown. dioxide (titania, TiO2) were examined. The bactericidal efcacy of infrared (IR) photoactivated TiO2-coated Ti plates and wires against E. coli has been reported previously, which demon1. Introduction strated their effectiveness in causing quantitative cell necrosis Surgical site infections (SSI) are the leading cause of increased in record time.[10] In another application, the antibacterial effect patient hospitalization, and in some cases patient mortality. of IR photoactivated electrospun nanobrillar titania has been According to some studies, approximately 500,000 cases of demonstrated with bacterial necrosis exceeding 90%.[11,12] Other patient infection occur per year in the United States alone.[1,2] researchers have also reported the efcacy of titania powder These are exacerbated with the presence of other complications and thin lms in their cell necrosis potency against S. aureus and other infection-causing bacterial species upon activation by ultraviolet radiation (UV).[1315] A. Aboelzahab, V. Goel The photocatalytic properties of titanium dioxide were disDepartment of Bioengineering covered by Akira Fujishima,[16] who showed the photo-induced The University of Toledo Toledo, OH 43606-3390, USA cleavage of water molecules, known as the Honda-Fujishima A.-M. Azad effect,[17] on TiO2 electrodes. Since then, photocatalysts based Chemical Engineering on titania have been extensively studied for the past 40 years, The University of Toledo mainly for the destructions of organic compounds in water and Toledo, OH 43606-3390, USA in the environment. Titania is a semiconducting oxide with a E-mail: abdul-majeed.azad@utoledo.edu bandgap of 3.2 eV. Upon illumination by light, the photoenergy S. Dolan Henkel Corporation generates an electron-hole pair which comes to the titania surMadison Heights, MI, 48071, USA face, as shown below:
V. Goel Department of Orthopedic Surgery Health Science Campus The University of Toledo Toledo, OH 43606-3390, USA
(Ti O2 ) + h + TiO2 e cb v b ( Ti O2 ) h

DOI: 10.1002/adhm.201100032

The electron in the conduction band can reduce O2 molecule and produce superoxide ions (O2), and the hole in the valence band can react with H2O and produce hydroxyl radical (OH). 1

Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com

www.advhealthmat.de www.MaterialsViews.com

Both these reactive oxygen species (ROS) are short-lived but extremely potent. In addition, other ROS such as hydrogen peroxide and singlet oxygen radicals have also been detected. The reaction of ROS with organic compounds is extremely agile leading to their complete oxidation to carbon dioxide. The process of cell destruction in the case of living microorganisms also involves this oxidation mechanism, where the photo-induced generation of these radical oxygen species (ROS) creates a hostile environment that mitigates bacterial proliferation and survival, thereby inducing bacterial cell necrosis.[18,19] Killing of microbial cells in contact with a Pt-TiO2 catalyst in water, upon illumination with near-UV light for 1 to 2 h was rst reported by Matsunaga et al.[20] It should be pointed out that in the absence of O2 or other suitable electron acceptor, no photocatalytic reaction is likely to occur, due to the equally rapid and deleterious process of electron-hole recombination.[21] This paper describes the development of titania coating on Ti plates and mesh samples by using a patented aqueous plasma electro-deposition process.[22] Photoactivation using a hand-held IR laser was used to illustrate the advantage of shorter exposure time while obtaining comparable and in some case, increased cell necrosis as compared to far longer illumination by UV radiation.[1015] Although plates and other net-shaped congurations are more robust mechanically in cases where geometrically standard implants are preferred, wires and mesh give more exibility in those instances where reaching the intricate areas of injury poses difculty to the rm plates and rods. This exible behavior is also of relevance in the use of shape memory alloys in conjunction with the bactericidal coatings being developed in this research program.

to 800 C for 4 h in static air at a ramp rate of 10/min in both heating and cooling cycles. Systematic structural and microstructural analysis of the samples was carried out using X-ray diffraction (XRD-PANalytical XPert Pro MPD) and scanning electron microscopy (Hitachi S-4800 UHR SEM). The quantitative elemental analysis was accomplished by using the energy dispersive spectroscopy (EDS) attached to the SEM unit. Prior to SEM imaging, the sample surfaces were sputter-coated with a gold-palladium target for approximately 40s. While the coated plates were analyzed with both XRD and SEM/EDS, the analysis on mesh sample was done by SEM/EDS alone because of the difculty of doing XRD on mesh congurations. 2.2. Bacterial Growth For bacterial growth, two sub-culture asks with 150 mL of TSB superbroth (32g of tryptone, 20g of yeast extract and 5g of sodium chloride per 1L DI water, Fisher Scientic, Waltham, MA) each were inoculated with 1mL of S. aureus (NRS72, Sanger476; MSSA), which was taken from a previously puried batch culture growth. A standard calibration curve was created with the aid of streaking of cell suspensions on TSB/agar to enumerate cell concentration. The S. aureus cells were grown to stationary phase, diluted to a concentration of 9.52 108cells mL1, and then divided into 1mL aliquots in 3mL capacity eppendorf tubes for individual experiments. Stock solutions were also stored in 1mL aliquots with the addition of 30% glycerol for future growth of more bacterial samples. All samples were stored at 80 C until use. 2.3. In-Situ Image Analysis

FULL PAPER

2. Experimental Section
Titanium plates (of commercial purity 99.99%, Alfa-Aesar, MA) and titanium mesh (99% purity, Cleveland Wire Cloth & Manufacturing Company, OH) were used as implant surrogates. The use of diverse materials also permitted a direct comparison between the nanostructured morphologies formed on each, as well as their corresponding bactericidal potency.

2.1. TiO2 Coating on cp Ti Plates and Mesh The two titanium implant surrogates were coated by Henkel Corporation (Madison Heights, MI, USA) using a patented aqueous plasma electrodeposition (PED) process. The method employs a pulsed DC voltage of 240V for 10 ms on and 30 ms off with a current density of 1500 A/m2, which creates a very adherent TiO2 coating from the precursor solution on the substrate. In this case, the Ti substrates (plate or mesh) were used as the anode, onto which titanium dioxide deposits are created from solution; the coatings are cured in situ by the plasma glow on the surface.[22] By increasing the time of PED process, sequentially thicker titania coatings were obtained on the Ti substrates. The TiO2-coated samples were subsequently heated

For analysis and imaging of cell necrosis of the bacterial suspension during experimentation, an Invitrogen Baclight Live/ Dead Assay kit (Invitrogen, Carlsbad, CA) was used in addition to confocal microscopy to obtain real-time imaging of the bacterial cells while in suspension with the titania-coated Ti substrates, upon being activated by the IR laser. This assay uses SYTO9 and propidium iodide stains, which were stored in 1 l aliquots for individual experimental use. 100 L of S. aureus suspension were diluted in ultra-pure water (with a conductivity of 5.5 108 S cm1) to a 1:3 v/v ratio to a volume of 300 L and cell concentration of 3.17 108 cells/ ml, with SYTO9 and PI stains added in 1L quantities, respectively. The suspension was set in the dark for 20 min. to complete staining. 10 L of the suspension was then mixed with 30 uL of ultra-pure water (to cause 4-fold dilution) and added to a 35 mm 14 mm glass bottom microwell dish (No. 15 cover glass). This gave a nal concentration of 7.93 107 cells/ml for imaging. A 1 cm 1 cm TiO2-coated Ti specimen was rst activated by a handheld IR laser ( = 808 nm; power = 1 W) from www. freaklasers.com for 30s, then introduced to the bacterial suspension. This laser introduces much greater power than the IR ashlight used in previous studies.[10,12] The time-lapse imaging was recorded for 40 min. using a confocal multiphoton

wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

www.advhealthmat.de www.MaterialsViews.com

FULL PAPER

Figure 1. SEM images of TiO2-coatings of increasing thickness (a = 1.3 m, b = 2.4 m, c = 3.7 m d = 5.0 m) on Ti plates; the samples were red at 800 C for 4 h after coating by PED process.

photoactive surface. On the other hand, the microscale platelets and rods offer a larger surface area, which also could be benecial in the bacterial necrosis per unit area. The XRD analysis showed the presence of TiO2 in rutile phase in all the four samples, as can be seen from Figure 2. The identical XRD signatures conrm the reproducibility of the PED coating technique developed by Henkel and the uniformity in lm growth. The morphological features of the titania lms deposited on Ti mesh are distinctly different from those on Ti plates, as seen in Figure 3. Figure 4 shows the systematic variation in the thickness of TiO2 lm coated by PED process on the mesh samples. The lm thickness increases monotonically and almost linearly, which is also corroborated by the SEM images of the successive lms, taken at same magnication (inset). A similar trend was observed in the case of plates as well, but is not shown here to avoid redundancy.

microscope system (Leica TCS SP5 MP, Leica Microsystems, Bannockburn, IL). Upon photoexcitation, the S. aureus cells stained with SYTO9 uoresce green and those stained with propidium iodide uoresce red. The excitation/emission of SYTO9 and propidium iodide occur at 480/500 nm and 490/645 nm, respectively. The survival/necrosis rate of the microorganism was determined by using the imaging software called ImageJ. By post-confocal imaging, this software allows one to count the bacteria in order to determine the amount of live cells compared to total number of cells.

3.2. Evaluation of Bactericidal Effect of Photoactivated Titania-Coated Plates and Mesh We have previously reported the fabrication of titania coatings by anodization, hydrothermal processing and the PED technique.[10] Their bactericidal efcacy against E. coli was also successfully demonstrated. Activation by IR laser ( = 808 nm) for 30 s proved effective in the necrosis of E. coli cells up to 40%. The photocatalytic response of free-standing titania in other formats (powder, nanobers, etc.) in the presence of UV radiations has also been reported.[10,1215]
<110>

3. Results and Discussion

3.1. Surface Analysis of Henkel Coated Samples The Henkel coatings of titania on the plates and mesh were created with different thicknesses. They were imaged by SEM and analyzed for their gross chemical composition by EDS; the TiO2-coated plates were also subjected to x-ray diffraction for phase identication. Figure 1 shows the systematic evolution of the surface morphological features, with increasing thickness of titania coating on the plate samples. As can be seen, with increasing lm thickness, the structure goes gradually from all high aspect (L/D) ratio 1-D bers, to a mixture of 1-D bers and 2-D platelets, to 2-D platelets and 3-D rods. Moreover, the surface densication also increases; a thinner titania lm is endowed with more nanofeatures and more void space which may allow for higher and direct exposure of the bacterial species with the

14000

A<112>

<220>

A<101>

Intensity

8000

6000

4000

2000

22

32

42

52

62

<002> <310>

<210>

Sample 4

Sample 3 Sample 2

Sample 1

<301> <112>

10000

<111>

<211>

12000

<101>

72

2-
Figure 2. XRD signatures of the coating on Ti plates after ring at 800 C for 4 h. The diffraction peaks predominantly belong to rutile titania with minor peaks conforming to the anatase (A) modication.

Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com

www.advhealthmat.de www.MaterialsViews.com

FULL PAPER
Figure 3. Morphology of a) pure Ti-mesh and b) Henkel TiO2-coated Ti mesh red at 800 C for 4 h; c) a higher magnication image of coating (b).

3.5

2.5

1.5

1 1 2 3 4 5

Number of TiO2 coatings on Ti mesh substrate

Figure 4. Variation in the thickness of TiO2 coating on Ti mesh. Inset: SEM images of TiO2Ti cross-sectional interface for sample with varying titania lm thickness (red arrow is a measure of the thickness of titania lm on the Ti substrate).

The UV radiations ( = 365 nm) employed hitherto are of lower intensity though of larger photon energy (de Broglies equation of wave length-energy equivalence: E = hc/; where h is Plancks constant, c the speed of light and the radiation wave length). Consequently, longer exposure time (reaching 60 min. to 2h in some cases) is needed to obtain signicant results. On the other hand, the IR laser source used by the present authors in this as well as previous work, is of larger wave lengths, and hence of smaller photon energy. However, these laser beams are of much higher source intensity (5 W/cm2). This effectively creates a larger population of incident photons per unit area and it is this parameter that functions as a deciding factor in the observed photoactivation-mediated effect in the cell necrosis. A larger photon ux (number per unit incident area) in the case of IR source as opposed to UV source is likely the reason for observed agile and more quantitative bactericidal activities in relatively shorter span of time on the titania surfaces. A shorter exposure would limit the time the patient is exposed to radiations, thereby making the effectiveness of the technique more attractive and applicable Figure 5. Confocal images of the bacterial colonies at different times in S. aureus suspension for clinical applications. Furthermore, in the exposed to a handheld IR laser ( = 808 nm) for 30 s (green/SYTO9 = live cells; red/PI = dead cells). 4

light of the fact that longer exposure to UV radiation (up to 60 min. or more) does not induce quantitative necrosis, repeated exposures would be needed. Under these conditions, one might reach dose levels that are harmful to human tissues. This explanation corroborates the brief but effective use of an IR source and strengthens the following results obtained on the titania-coated plate and mesh samples; to the best of our knowledge no such data exists with a parallel study using UV activation by others. Figure 5 shows the images of an S. aureus suspension not subjected to any TiO2 or IR activation over an interval of 20 min.; with no cell necrosis occurring, this was used to serve as a control. The titania-coated plate and mesh samples were evaluated for their antibacterial efcacy against S. aureus (MSSA). Confocal imaging using SYTO9 and PI stains was employed to monitor the bacterial cells as a function of time-lapse subsequent to irradiation of the sample and its immersion in the microbial bath. A Ti mesh coupon with a 1.5 m thick TiO2 coating (sample #2) was subjected to IR activation for 30 s followed by immersion in a bacterial suspension of stained S. aureus cells. The specimen was placed so as to expose the bacteria to the photoactivated region of the mesh. Confocal imaging was recorded in real-time as the bactericidal effects of the titania coating took effect. Figure 6 illustrates the quick effect of the Henkel-coated mesh in causing S. aureus cell necrosis. Within approximately 5 min. of exposure, only 23.71% of the S. aureus cells survived; the concentration of the surviving cells

Thickness, m

wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

www.advhealthmat.de www.MaterialsViews.com

FULL PAPER

Figure 6. Confocal images of the bacterial colonies at different times in S. aureus suspension containing Henkel TiO2-coated mesh samples (top to bottom) 2, 4, and 5 (red at 800 C/4 h) and exposed to a handheld IR laser ( = 808 nm) for 30 (green/SYTO9 = live cells; red/PI = dead cells).

already taken place at 0 min. in some of the images. Figure 6 also shows the bactericidal efcacy of the mesh samples with titania coating thicknesses of 2.6 m (#4) and 3.2 m (#5), respectively. Although these mesh samples did cause evident cell necrosis, they were not found to be as effective as the sample with a smaller coating thickness. The survival count after 5 min. of exposure was 50.7% for sample 4, which dropped down to 21.5% after a time-lapse of 30 min. In comparison to sample (#2) with a thinner titania coating, the cell survival concentration was slightly more than twice on sample #4. Increasing the coating thickness of the Ti mesh to 3.2 m (sample 5) did increase the bactericidal efcacy as compared to Sample 4 by 10% after only 5 min., but the survival after 30 min. for this sample was approximately 22.3%. This is slightly less effective than sample 4, but portrays relatively equivalent results. The bactericidal efcacy of Ti plates also was tested against S. aureus. It was found that coated mesh samples were more effective because of their more immediate access to the bacterial suspension as is apparent by the structural difference between plates and mesh. Therefore, cell necrosis was apparently detected in one Henkel-coated plate sample of thickness 5 m, the results of which are seen in Figure 7. It should be noted here that placement of the plates among the bacterial suspension incorporated a further degree of difculty because of the dense and bulky nature of the material. It was necessary for the plate to be fully exposed to the S. aureus cells while not crushing the cells or distorting bacterial movement, which therefore leads to distortion of the uorescent imaging by

was 9% after 30 min. The green pixels indicate live cells while the red pixels denote dead ones. A note of clarication with regard to the time stamp of 0 min. on the confocal images is warranted. The images marked as 0 min. were realistically taken a few moments after actual exposure to the titania-coated mesh due to the time elapsed between activation and placement in the suspension, before being placed inside the slide holder of the apparatus. An additional amount of time also elapsed in positioning the center of the microscopic lens on the activated area prior to imaging. This would explain the observed cell necrosis
Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

Figure 7. Confocal images of the bacterial colonies at different times in S. aureus suspension containing Henkel TiO2-coated plate sample (red at 800 C/4 h) and exposed to a handheld IR laser ( = 808 nm) for 30 s (green/SYTO9 = live cells; red/PI = dead cells).
wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advhealthmat.de www.MaterialsViews.com

FULL PAPER

Table 1. Summary of the efcacy of the TiO2-coated mesh and plate samples activated by IR-laser source ( = 808 nm; power = 5 W cm2) for 30s against S. aureus. The data represent an average of 10 measurements with each of the mesh and plate samples and has an uncertainty of 2%
Material Titania lm thickness Post-exposure cell survival [%] after a time-lapse of 5 min. Control (no exposure to TiO2/IR) TiO2-coated Ti mesh (2) TiO2-coated Ti mesh (4) TiO2-coated Ti mesh (5) TiO2-coated Ti plate (4) 1.5 m 2.6 m 3.2 m 5.0 m 100 23.7 50.7 42.6 55.7 30 min. 100 9.0 21.5 22.3 12.9

4. Conclusion
Ti mesh and plate coupons were coated with uniform TiO2 lm using a patented aqueous plasma electrodeposition (PED) process. The bactericidal efcacy of the coated samples was tested against S. aureus (MSSA) with cell necrosis of greater than 90% achieved after a 30 min. time-lapse following 30 s of IR activation. In comparison to the bactericidal efcacy of titania with UV illumination, activation by IR laser adopted in this study illustrated the superiority of the method used to create the photoactive lms. The advantage of IR activation in terms of approximately 90% cell death after only 30 s of exposure is outstanding. Cell necrosis rates less than 10% have been reported by other groups, and comparable results were obtained only after very long UV activation (cf. 60 min.).

confocal microscopy. For this reason, bactericidal efcacy of the Henkel-coated plates was not consistent throughout. Still, the results for Henkel-coated plate showed its successful antibacterial effect against S. aureus, indicating that only 55.7% of cells survived after 5 min.; the survival diminished to about 12.9% after 30 min. The data is summarized in Table 1. From the data sown in Table 1, it can be concluded that titania-coated Ti mesh samples had an increased efcacy towards cell necrosis as compared to Ti plates. This increased effect can be attributed to the ability of mesh to be more accessible to the bacterial suspension because of its porous structure. In addition, thinner coatings were more effective in causing necrosis of the bacterial cells. Ti plates and mesh with increasing coating thicknesses caused slightly lower cell death after activation, though still comparable with the highest survival count being around 20%. At rst sight, these results might appear counterintuitive, but the increased efcacy observed in this study can be explained as a combination of a number of factors related mainly to the structural artifacts of the coating. For example, the particle density of TiO2 is much greater in thinner coatings, thus allowing for more titania particles packing within a unit space. Second, there is a gradual but denite change in shape and particle distribution between thin and thick coatings. This results in a corresponding change in the aspect ratio (length to diameter; l/d) as seen clearly from the SEM images in Figure 1. The effectiveness of the thinner lm (with larger aspect ratio) is therefore due to the availability of larger surface area on the thinner coating within the contour of limited activation area of the laser beam. This behavior of titania coatings could be used for the implant activation in both pre and post-surgery situations. In post-surgery cases, this may simplify the reactivation step if the infection persists after surgery. Furthermore, IR radiations are capable of penetrating skin and other soft tissue, which might necessitate little or no surgical incision. Currently, the persistence of surgical site infection often requires re-opening and removal of implants for ex-vivo sterilization and even replacement in some cases. With a titania pre-coated implant in place, that could be activated and re-activated as required, the necessity of additional surgical procedures would potentially diminish.

Acknowledgements
Image processing and analysis was carried out in the in JavaNational Institute of Health (http.://rsbweb.nih.gov/ij/download.html). The authors wish to express their gratitude to Ms. Tamara Phares of the Bioengineering Department, Dr. Andrea Kalinoski of the Advanced Microscopy and Imaging Center (Department of Surgery, Health Science Campus), Mr. Robert Kinner of the Chemical Engineering Department and Dr. Kristen Williams of the Microbiology Department (Health Science Campus) for their assistance in various stages of this work. Financial support from DePuy Spine (Johnson & Johnson, Raynham, MA) is also gratefully acknowledged. Received: February 13, 2012 Published online:

[1] R. Nichols, Emerging Infect. Dis. 2001, 7, 220224. [2] J. Engemann, Y. Carmeli, S. Cosgrove, V. Fowler, M. Bronstein, S. Trivette, J. Briggs, D. Sexton, K. Kaye, Clin. Infect. Dis. 2003, 36, 592598. [3] M. Olsen, J. Mayeld, C. Lauryssen, L. Polish, M. Jones, J. Vest, V. Fraser, Journal of Neurosurgery: Spine. 2003, 98, 149155. [4] S. Cosgrove, Y. Qi, K. Kaye, S. Harbarth, A. Karchmer, Y. Carmeli, Infect. Control Hospital Epidemiol. 2005, 26, 166174. [5] K. Sieradzki, R. Roberts, S. Haber, A. Tomasz, New Engl. J. Med. 2010, 340, 517523. [6] B. Howden, P. Ward, P. Charles, T. Korman, A. Fuller, P. du Cros, E. Grabsch, S. Roberts, J. Robson, K. Read, N. Bak, J. Hurley, P. Johnson, A. Morris, B. Mayall, M. Grayson, Clin. Infect. Dis. 2004, 38, 521528. [7] G. Siberry, T. Tekle, K. Carroll, J. Dick, Clin. Infect. Dis. 2003, 37, 12571260. [8] T. Jones, M. Yeaman, G. Sakoulas, S. J. Yang, R. Proctor, H. G. Sahl, J. Schrenzel, Y. Xiong, A. Bayer, Antimicrob. Agents Chemother. 2008, 52, 269278. [9] S. Cosgrove, Y. Qi, K. Kaye, S. Harbarth, A. Karchmer, Y. Carmeli, Infect. Control Hospital Epidemiol. 2005, 26, 166174. [10] A.-M. Azad, R. Hershey, A. Aboelzahab, V. Goel, Adv. Orthop. 2011, 113. [11] A.-M. Azad, S. McKelvey, Z. Al-Firdaus, Adv. Mater. Manuf. Testing Information Anal. Center. 2008, 3, 27. [12] A.-M. Azad, R. Hershey, S. Ali, V. Goel, J. Mater. Res. 2010, 25, 17611770.

wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

www.advhealthmat.de www.MaterialsViews.com

FULL PAPER

[13] H. Koseki, K. Shiraishi, T. Tsurumoto, T. Asahara, K. Baba, H. Taoda, N. Terasaki, H. Shindo, Surf. Interface Anal. 2009, 41, 771774. [14] J.C. Yu, W. Ho, J. Lin, H. Yip, P. K. Wong, Environ. Sci. Technol. 2003, 37, 22962301. [15] Y. Oka, W.-C. Kim, T. Yoshida, T. Hirashima, H. Mouri, H. Urade, Y. Itoh, T. Kubo, J. Biomed. Mater. Res. Part B. 2008, 86B, 530540. [16] A. Fujishima, K. Honda, S. Kikuchi, Kogyo Kagaku Zasshi 1969, 72, 108113. [17] A. Fujishima, K. Honda, Nature 1972, 238, 3738.

[18] E. Cabiscol, J. Tamarit, J. Ros, Int. Microb. 2000, 3, 38. [19] H. U. Simon, A. Haj-Yehia, F. Levi-Schaffer, Apoptosis 2000, 5, 415418. [20] T. Matsunaga, R. Tomada, T. Nakajima, H. Wake, FEMS Microbiol. Lett. 1985, 29, 211214. [21] C. Wei, W.-Y. Lin, Z. Zaina, N. E. Williams, K. Zhu, A. P. Kruzic, R. L. Smith, K. Rajeshwar, Environ. Sci. Technol. 1994, 28, 934938. [22] A.-M. Azad, S. Dolan, S. A. Akbar, Int. J. Appl. Ceram.Technol. 2008, 5, 480489.

Adv. Healthcare Mater. 2012, DOI: 10.1002/adhm.201100032

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com