Journal of Antimicrobial Chemotherapy (2004) 54, 1019–1024

DOI: 10.1093/jac/dkh478
Advance Access publication 10 November 2004
JAC
Silver nanoparticles and polymeric medical devices: a new
approach to prevention of infection?
Franck Furno1,2, Kelly S. Morley2, Ben Wong2, Barry L. Sharp3, Polly L. Arnold2, Steven M.
Howdle2, Roger Bayston1*, Paul D. Brown4, Peter D. Winship3 and Helen J. Reid3
1
Biomaterials-Related Infection Group, School of Medical and Surgical Sciences, University of Nottingham,
Nottingham NG7 2UH; 2School of Chemistry, University of Nottingham, Nottingham NG7 2RD; 3Department

Downloaded from http://jac.oxfordjournals.org/ at Southern Methodist University on August 31, 2014

of Chemistry, University of Loughborough, Loughborough LE11 3TU; 4School of Mechanical, Materials and
Manufacturing Engineering, University of Nottingham, Nottingham NG7 2RD, UK

Received 19 May 2004; returned 13 July 2004; revised 24 August 2004; accepted 27 September 2004

Objectives: Implantable devices are major risk factors for hospital-acquired infection. Biomaterials
coated with silver oxide or silver alloy have all been used in attempts to reduce infection, in most
cases with controversial or disappointing clinical results. We have developed a completely new
approach using supercritical carbon dioxide to impregnate silicone with nanoparticulate silver metal.
This study aimed to evaluate the impregnated polymer for antimicrobial activity.
Methods: After impregnation the nature of the impregnation was determined by transmission electron
microscopy. Two series of polymer discs were then tested, one washed in deionized water and the
other unwashed. In each series, half of the discs were coated with a plasma protein conditioning film.
The serial plate transfer test was used as a screen for persisting activity. Bacterial adherence to the
polymers and the rate of kill, and effect on planktonic bacteria were measured by chemiluminescence
and viable counts. Release rates of silver ions from the polymers in the presence and absence of
plasma was measured using inductively coupled plasma mass spectrometry (ICP-MS).
Results: Tests for antimicrobial activity under various conditions showed mixed results, explained by
the modes and rates of release of silver ions. While washing removed much of the initial activity there
was continued release of silver ions. Unexpectedly, this was not blocked by conditioning film.
Conclusions: The methodology allows for the first time silver impregnation (as opposed to coating) of
medical polymers and promises to lead to an antimicrobial biomaterial whose activity is not restricted
by increasing antibiotic resistance.

Keywords: antimicrobial biomaterials, implantable devices, medical polymers

Introduction serious2 with a mortality rate of up to 28%.3 An essential event

in initiation of biomaterials-related infection (BRI) is microbial
Infection is a well-recognized complication of implantable adhesion to the device, or more often to the patient-derived
devices, of which a wide variety is now in use. The devices can glycoprotein coating (conditioning film) which begins to be
be categorized according to the nature and risk of infection.1 deposited immediately on implantation in most devices.4,5 Once
While Category I devices are totally implanted and are at risk adhesion has occurred, proliferation leads to the development of
mainly at insertion, Category II devices are incompletely a biofilm, which is insusceptible to most therapeutic agents at
implanted and have periods of risk extending throughout their achievable concentrations.6 – 8 Device removal is often required
use. Central venous catheters (CVC), wound drains and catheters in order to ensure eradication of infection and to avoid relapse.
for continuous ambulatory peritoneal dialysis (CAPD) are Several approaches to prevention of BRI involving surface
examples. Approximately 3 million CVC are inserted each year coatings have been proposed. Many, such as surface modifi-
in the USA, resulting in 850 000 infections, almost 20% of them cation by gas plasma, appear to reduce microbial adhesion
..........................................................................................................................................................................................................................................................................................................................................................................................................................

*Corresponding author. E-mail: roger.bayston@nottingham.ac.uk

..........................................................................................................................................................................................................................................................................................................................................................................................................................

1019
JAC vol.54 no.6 q The British Society for Antimicrobial Chemotherapy 2004; all rights reserved.
 F. Furno et al.

in vitro but are ineffective in vivo, probably because of oblitera-

tion by conditioning film.9 Silver has long enjoyed a reputation
for antimicrobial properties, and has therefore been seen as a
candidate for coating of medical devices. While various coat-
ings, using silver salts or ion beam implantation of metallic sil-
ver, have been devised they have shown disappointing clinical
results10 – 13 or even further complications.14 Possible reasons for
this are obliteration or inactivation of the silver coating by blood
plasma, and the lack of durable activity inherent in coatings.
True impregnation of polymers15 with antimicrobials has led to
only one clinical application so far, but this has demonstrated
the superiority of impregnation over coating.16 Hitherto, silver
could not be impregnated into biomaterials. Attempts to admix
silver metal or salts into biomaterials have failed to improve
antimicrobial activity and have risked impaired mechanical prop-
erties. In an attempt to address these shortcomings while taking

Downloaded from http://jac.oxfordjournals.org/ at Southern Methodist University on August 31, 2014

advantage of the broad antimicrobial spectrum of silver, we have
devised a process whereby nanoparticles of silver metal can be
impregnated into medical polymers. This paper reports studies
on their release and antimicrobial activity in vitro.

Figure 1. Transmission electron microscopy image of nanoparticulate silver

Methods
Biomaterial
Medical grade unfilled silicone sheet 0.45 mm thick (Dow Corning
Ltd, Meriden, UK) was cut into discs of 15 mm diameter.
Impregnation process
The impregnation was carried out using a novel method.17 Briefly,
organic complexes of silver were synthesized in anaerobic con-
ditions. A solution of the organometallic precursors dissolved in
supercritical carbon dioxide (scCO2) was then used to swell and
permeate the polymer at 4000 psi, 408C, for 24 h. For this stage, the
silicone discs were placed in a 10 mL stainless steel high-pressure
autoclave (Thar Technologies, Pittsburgh, PA, USA) itself placed
inside a metal heating block and attached to a transducer and a
thermocouple to monitor the pressure and temperature. The scCO2
was then vented and the polymers were exposed to H2 gas
(1500 psi, 408C, 24 h) in order to decompose the organometallic pre-
cursors, leading to a homogeneous distribution of nanoparticles of
silver (10 – 100 nm) in the polymer (Figure 1).18,19 The residues of
the organic precursors were then removed completely by flushing
with scCO2. This was checked carefully by thermogravimetric ana-
lysis of the final polymer composite. The formation of silver nano-
particles in the silicone matrix was confirmed by detailed analyses
using transmission electron microscopy (Figure 1). The overall load-
ing, particle size and particle distribution of the silver nanoparticles
was controlled by careful moderation of the supercritical fluid con-
ditions.18,19
(dark, electron-dense particles) dispersed in silicone. The associated diffrac-
tion pattern (inset) has been indexed against standards to confirm that the
particles are metallic silver. Note the extremely small size of the silver
particles.
iment to maintain viability without proliferation in these test
conditions) to 1  108 cfu/mL, corresponding to A490 0.7 – 0.8.
Application of conditioning film
One series of discs was washed after impregnation. This consisted
of immersion in sterile deionized water for 1 h. A second series
remained unwashed. Both sets of discs were immersed in 50%
human plasma (National Blood Authority, Sheffield, UK) for 30 min
at 378C after which excess plasma was rinsed away by gentle
immersion three times in sterile water. In an additional series, discs
were immersed in plasma for 1 h.
Serial plate transfer test (SPTT)
Four sets of discs were tested: washed and unwashed after impreg-
nation, and with or without 30 min conditioning film; 20 mL ISA
plates (Oxoid, Basingstoke, UK) were inoculated with S. epidermi-
dis F22 using a rotary inoculator and the discs placed centrally on
each plate. After overnight incubation, any inhibition zones were
recorded (diameter minus disc diameter, mm) and the discs were
transferred to fresh inoculated plates, taking care to present the
same surface to the agar. The process was continued until no more
inhibition was seen.21

Test bacterium
A clinical isolate of Staphylococcus epidermidis (F22) whose adher-
ence and biofilm characteristics were known was kept at 208C in
cryoprotectant until use. To resuscitate, 20 mL of tryptone soya
broth (TSB; Oxoid, Basingstoke, UK) was inoculated and incubated
overnight at 378C. One drop of this was then inoculated into 20 mL
of fresh TSB and incubated with shaking at 378C for 4 h to ensure
expression of adhesins in early – mid log phase.20 The cells were
then washed in PBS and resuspended in 2% TSB (found by exper-
Quantitative adherence
As with the SPTT, four sets of discs were tested. They were
immersed in a suspension of S. epidermidis F22 at A490 for 1 h and
after rinsing to remove non-adhered bacteria they were sonicated
at 50 Hz/10 min (Ultrawave, Cardiff, UK). Triplicates of 125 mL of
the sonicate were added to the wells of chemiluminescence trays
(Zeptogen Ltd, Middlesex, UK) and read in a luminometer
(Berthold Technologies GmbH, Bad-Wildbad, Germany) using
a Lumitech Vialite kit (BioWhittaker Ltd, Berks, UK). Plate viable
counts were also carried out.

1020
 Silver nanoparticle impregnation and infection
Figure 2. Serial plate transfer test (SPTT) clearly showing a 15 mm diameter zone of growth inhibition (arrow, circled) surrounding the silicone disc with sil-
ver nanoparticles (unwashed) on the third day of the experiment (b), compared with the control (a).
Planktonic killing
Groups of discs as above were immersed in a suspension of S. epi-
dermidis F22 at A490 0.7– 0.8 for 1, 2, 3, 4 and 5 h at 378C and the
suspensions assayed by chemiluminescence and plate counting. An
additional two groups of discs, one washed and the other unwashed,
were plasma-coated for 1 h before testing.
Rate of kill
Triplicates of washed and unwashed discs with or without condition-
ing film were immersed as above in a suspension of S. epidermidis
F22 at A490 0.7 – 0.8 for 1 h. They were then rinsed and immersed in
2% TSB for 2, 3, 4 and 5 h at 378C and after rinsing to remove non-
adhered bacteria they were sonicated. The sonicates were assayed
by chemiluminescence and plate counting.
Release of silver ions
The four groups of discs as above were thoroughly washed in sterile
deionized water, then immersed in 5 mL of either sterile deionized
water or 50% human plasma for 1 day at room temperature. The
next day they were removed, blotted free of excess fluid, and trans-
ferred to a fresh 5 mL of the same elutant. The process was contin-
ued for 5 days. The suspending fluids from days 3, 4 and 5 were
then analysed for silver by inductively coupled plasma mass spec-
trometry (ICP-MS: VG PQ ExCell, Winsford, UK). While the water
samples remained clear, the plasma samples contained amorphous
material that in some cases assumed a web-like appearance. Unused
suspending fluids were also analysed as controls. A 100 mL aliquot
of each of the deionized water samples was diluted 1000-fold before
ICP-MS analysis. Then 1 mL of each plasma sample was centri-
fuged (2500 rpm, 5 min), and 100 mL of the resulting supernatant
was removed and diluted 1000-fold before analysis by ICP-MS.
This ensured the absence of any particulate matter in the plasma
samples that were analysed. Standard solutions of inorganic silver
salts were also included.
Results
SPTT
The results may be presented as two sets of data, one referring
to discs washed after processing, and the other to those left
1021
Downloaded from http://jac.oxfordjournals.org/ at Southern Methodist University on August 31, 2014
unwashed. In the case of the unwashed discs, zones of inhibition
were seen for 10 days, decreasing from a mean of 8.5 mm on
day 1 to 3 mm on day 5, showing clear evidence of persisting,
diffusible activity (Figure 2). No residual bacteria were seen
under the discs after removal, but further studies were not car-
ried out to detect them. This activity remained unchanged in the
presence of a conditioning film, that is in those discs which had
been immersed in plasma before being applied to the plate.
Tests for residual organic complexes that could have accounted
for this were negative. However, when the second set of discs
were washed after impregnation, all the diffusible inhibitory
activity was extinguished and no zones were seen, even on day 1.
No inhibition zones were seen with plain unprocessed silicone
discs or with those which had undergone supercritical fluid pro-
cessing but without silver.
Adherence
Again, there was a difference between washed and unwashed
discs. In the unwashed series, no viable bacteria were found
after 1 h of exposure, confirmed by chemiluminescence (data not
shown). The control unprocessed discs, with or without a con-
ditioning film, showed 1  107 cfu/mL, whereas the washed
impregnated discs, again irrespective of a conditioning film,
showed a 4 log reduction to 1103 cfu/mL.
Killing of adhered bacteria
A series of the above discs, after exposure to bacterial suspen-
sion for 1 h, were rinsed and suspended in 2% TSB. Triplicates
were removed and sonicated at 2, 3, 4 and 5 h. The results in
Table 1 show again that bacteria attached to unwashed discs
were all killed within 1 h, irrespective of conditioning film. The
washed discs showed an initial reduction compared with controls
of 4 logs but the numbers of viable attached bacteria gradually
increased by 2–3 logs over 5 h. A slight difference of 1 log was
seen when a conditioning film was present.
Killing of planktonic bacteria
The results are shown in Figure 3. A clear difference was seen
between washed and unwashed discs. While the former showed
 F. Furno et al.

Table 1. Rate of killing of adhered bacteria Table 2. Comparison of the mean Ag isotope concentrations, each
assayed in triplicate by inductively coupled plasma mass
Unwashed Washed spectrometry, in water and plasma extracts
Control
107 109
Time (h) after without with without with without Ag isotope Ag isotope
bacterial adherence CF CF CF CF CF Sample day concentration (ppm) concentration (ppm)

1 0 0 103 103 107 Water

2 0 0 103 104 108 3 0.270 ± 0.004 0.267 ± 0.010
3 0 0 105 105 108 4 0.147 ± 0.006 0.144 ± 0.006
4 0 0 105 106 108 5 0.131 ± 0.003 0.129 ± 0.004
5 0 0 105 106 108 Plasma
3 4.132 ± 0.010 4.135 ± 0.033
Silver-impregnated silicone discs were either washed in water or left 4 0.859 ± 0.016 0.854 ± 0.021
unwashed, then exposed to a suspension of S. epidermidis for 1 h. One set 5 0.568 ± 0.003 0.562 ± 0.006
was plasma-coated (conditioning film, CF) before bacterial adherence. They
were then rinsed and immersed in 2% TSB. Triplicate discs were sonicated

Downloaded from http://jac.oxfordjournals.org/ at Southern Methodist University on August 31, 2014

after 1, 2, 3, 4 and 5 h and viable counts carried out on the sonicates. The silver-impregnated discs were extracted for 5 days and sampled daily.
Controls consisted of discs that went through the whole process but without Results are means ± S.D .
silver complexes.

the first 3 days. A significant amount of silver ions was still

being released into the plasma on days 4 and 5 ( 0.859/0.854
and 0.568/0.562 ppm, respectively). The precipitated material
seen in the plasma series was also analysed for silver content,
and concentrations were found to exceed the range covered
by the standard curve. However, an approximate value of
70–90 ppm was obtained. Further analysis of this material could
not be completed but initial results from a Biuret test suggested
that it was proteinaceous.

Figure 3. Chemiluminescence assays showing the antimicrobial effect of Discussion

silicone impregnated with silver nanoparticles. Results are expressed as per-
centage of viable planktonic bacteria at time zero. Column 1, unwashed
discs, no conditioning film (CF); column 2, unwashed discs, 30 min CF;
column 3, unwashed discs, 1 h CF; column 4, washed discs, no CF. Control
(unimpregnated) discs had no effect on numbers of viable planktonic bac-
teria. The unwashed silver nanoparticle-loaded discs showed a dramatic and
sustained activity over 5 h. The rate of reduction of viability was reduced
with a 30 min conditioning film, and even more with a 1 h film. The washed
discs showed no significant antimicrobial activity against planktonic bacteria
over the test period. The controls remained consistently negative (i.e. no
activity) over the duration of the experiment (data not shown). All tests were
repeated nine times.
no effect on planktonic S. epidermidis, in the unwashed discs
there was a progressive decline in viability as shown by chemi-
luminescence over the 5 h test period. This was confirmed by
viable counts (data not shown). The presence of plasma coating
slowed the rate of killing, and this was more evident when
plasma coating was applied for 1 h.
Release of silver ions
The impregnated materials used in this part of the work had first
been washed as described above. From the data generated by
this analysis, calibration curves were plotted for both 107Ag and
109
Ag isotopes. Using these curves and the equations associated
with them, the concentrations of Ag isotopes in the deionized
water and plasma samples were calculated and are shown in
Table 2. Very little Ag ion was released into the water samples
(<0.5 ppm). A greater quantity of Ag ions was released into
the plasma samples with the bulk (4 ppm) being removed during
The method used here to introduce silver into silicone elastomer
has been shown to result in distribution of nanoparticles of met-
allic silver throughout the polymer, and is the only method
known to us of impregnation of polymers with silver. However,
the results show that much of the antimicrobial activity was
removed by washing the impregnated discs. This clearly shows
that a significant level of surface (or near surface) deposited sil-
ver nanoparticles is contributing to an initial ‘burst’ effect—one
which is abolished if the samples are carefully washed before
use. A variety of tests was used in order to explore several poss-
ible modes of action. The serial plate transfer test was originally
formulated to test biomaterials containing antimicrobials that
could readily diffuse through agar to produce a zone of
nhibition. This is not usually the case with silver, and we were
surprised to find zones of inhibition against S. epidermidis that
suggested a diffusible antimicrobial effect. However, these were
abolished by washing the discs. Killing of bacteria adhered to
the silicone was complete within 1 h but again this was dimin-
ished by washing. The fact that it was not completely abolished
by washing suggests that while there appears to be a high
surface concentration of silver initially, there is also a much
slower release of silver from nanoparticles buried deeper into
the polymer matrix. This would probably not be detected by the
SPTT method, but would show on the adherence studies. The
data presented in Table 1 clearly demonstrate that there is an
effect of the silver from the matrix in the washed samples,
demonstrating release of silver from below the polymer surface.
The killing of planktonic S. epidermidis was again abolished by
washing, but the cultures with unwashed discs showed a

1022
 Silver nanoparticle impregnation and infection
progressive decline in viable bacteria over 5 h, again indicating
high surface activity. Clearly, if there were residues of silver
precursor compound, or organic ligands, these might also have
an antimicrobial effect. However, thermogravimetric analyses of
a series of the silicone materials after careful and exhaustive
extraction using supercritical carbon dioxide demonstrated that
all such residues had been removed in the processing. That the
antimicrobial effect was due to silver is supported by the ICP-
MS results, which were accumulated over a much longer period
of 5 days. The ICP-MS analyses were carried out on washed
discs, yet they clearly showed release of silver over the test
period. Interestingly, this occurred most strikingly in the pre-
sence of plasma, reaching a peak at 3 days. This finding should
be seen in conjunction with the results of tests on discs bearing
a plasma conditioning film: while this appeared to slow the
release of silver ions, as shown by comparison of the effect of
30 min and 1 h conditioning films on planktonic killing results
(Figure 3), the silver ions were clearly able to penetrate the con-
ditioning film. This finding is interesting as silver is known to
have a high avidity for protein, and the presence of a protein
conditioning film, or any extraneous plasma protein, has pre-
viously been assumed to inactivate any silver ions released.22
The finding has obvious clinical implications. The surfaces of
implanted devices rapidly become coated with glycoproteins
from tissue and plasma, with the possible exception of the inner
surface of central venous catheters, and this has been cited as
one of the main reasons for clinical failure of silver-coated
devices.22 However, in the case of impregnation, we have now
demonstrated that there is both a depot effect and a diffusion
pressure available to ‘push’ the silver ions through the condition-
ing film. In order to do this, there must be enough silver ions
available over a sufficient period to exceed those lost to protein
binding. Indeed, this is one of the objectives of impregnation.
The precipitate found during the ICP-MS analysis was likely to
be plasma proteins complexed with released silver, and this was
supported by the finding of extremely high silver concentrations
in the precipitated material.
However, another clinically important advantage of impreg-
nation is the protection of both inner and outer surfaces of cath-
eters against bacterial colonization.23 This has been shown to be
crucial in clinical trials of efficacy.24 These two advantages of
impregnation, that is the continued release of silver ions in anti-
microbial concentrations even in the presence of a conditioning
film, and the ability to protect both inner and outer surfaces of
catheters, could be expected from the use of this novel method
to impregnate polymeric biomaterials with silver. A further
advantage is the known wide antimicrobial spectrum of activity
of silver.
The data presented here clearly demonstrate that supercritical
fluid impregnation of silver precursor followed by controlled
decomposition and extraction leads to a distribution of silver
throughout the silicone matrix. The impregnation process also
yields a significant surface coating of silver which is easily
removed by simple washing. However, underlying this is a sus-
tained release which shows great promise, and provides signifi-
cant scope for optimization of these release kinetics.
Acknowledgements
This study was supported by grants from the Engineering and
Physical Sciences Research Council, The Medical Research
1023
Council, The European Community and the Wade Charitable
Trust. S.M.H is a Royal Society Wolfson Research Merit Award
holder.
References
1. Bayston, R. (1999). Medical problems due to biofilms: clinical
impact, aetiology, molecular pathogenesis, treatment and prevention.
In Dental Plaque Revisited (Newman, H. N. & Wilson, M., Eds), pp.
111– 23. BioLine, Cardiff, UK.
2. Elliott, T. S. J., Faroqui, M. H., Tebbs, S. E. et al. (1995). An
audit programme for central venous catheter-associated infections.
Journal of Hospital Infection 30, 181–91.
3. Arnow, P. M., Quimosing, E. M. & Beach, M. (1993).
Consequences of intravascular catheter sepsis. Clinical Infectious
Diseases 16, 778–84.
4. Baumgartner, J. N. & Cooper, S. L. (1996). Bacterial
Downloaded from http://jac.oxfordjournals.org/ at Southern Methodist University on August 31, 2014
adhesion on polyurethane surfaces conditioned with thrombus
components. Journal of American Society of Artificial Internal
Organs 42, M476 –9.
5. Green, R. J., Davies, M. C., Roberts, C. J. et al. (1999).
Competitive protein adsorption as observed by surface plasmon
resonance. Biomaterials 20, 385–91.
6. Allison, D. G. & Gilbert, P. (1995). Modification by surface
association of antimicrobial susceptibility of bacterial populations.
Journal of Industrial Microbiology 15, 311 –7.
7. Proctor, R. A. & von Humboldt, A. (1998). Bacterial energetics
and antimicrobial resistance. Drug Resistance Updates 1, 227– 35.
8. Zimmerli, W., Widmer, A. F., Blatter, M. et al. (1998). Role of
rifampin for treatment of orthopedic implant-related staphylococcal
infections—a randomised controlled trial. Journal of the American
Medical Association 279, 1537– 41.
9. Everaert, E. P. J. M., Van de Belt-Gritter, B., Van der Mei, H. C.
et al. (1998). In vitro and in vivo microbial adhesion and growth on
argon plasma-treated silicone rubber voice prostheses. Journal of
Materials Science—Materials in Medicine 9, 147–57.
10. Morris, N. & Stickler, D. (1998). Encrustation of indwelling
urethral catheters by Proteus mirabilis biofilms growing in human urine.
Journal of Hospital Infection 39, 227–34.
11. Lai, K. K. & Fontecchio, S. A. (2002). Use of silver-hydrogel
urinary catheters on the incidence of catheter-associated urinary tract
infections in hospitalized patients. American Journal of Infection
Control 30, 221–5.
12. Crabtree, J. H., Burchette, R. J., Siddiqi, R. A. et al. (2003).
The efficacy of silver-ion implanted catheters in reducing peritoneal
dialysis-related infections. Peritoneal Dialysis International 23,
368– 74.
13. Walder, B., Pittet, D. & Tramer, M. R. (2002). Prevention of
bloodstream infections with central venous catheters treated with anti-
infective agents depends on catheter type and insertion time: evidence
from a meta-analysis. Infection Control and Hospital Epidemiology 23,
748– 56.
14. Riley, D. K., Classen, D. C., Stevens, L. E. et al. (1995). A large
randomised clinical trial of a silver-impregnated urinary catheter: lack
of efficacy and staphylococcal superinfection. American Journal of
Medicine 98, 349–56.
15. Bayston, R., Ashraf, W. & Bhundia, C. (2004). Mode of action of
an antimicrobial biomaterial for use in hydrocephalus shunts. Journal of
Antimicrobial Chemotherapy 53, 778– 82.
16. Govender, S. T., Nathoo, N. & Van Dellen, J. R. (2003).
Evaluation of an antibiotic-impregnated shunt system for the treatment
of hydrocephalus. Journal of Neurosurgery 99, 831– 9.
17. Webb, P. B., Parsons, A. J., Marr, P. C. et al. (2000). Green
chemistry: dissolving biomolecules and modifying biomedical implants
with supercritical carbon dioxide. Pure and Applied Chemistry 72,
1347–55.
 F. Furno et al.
18. Morley, K. S., Marr, P. C., Webb, P. B. et al. (2002). Clean
preparation of nanoparticulate metals in porous supports: a supercriti-
cal route. Journal of Materials Chemistry 12, 1898– 905.
19. Morley, K. S., Licence, P., Marr, P. C. et al. (2004). Supercritical
fluids: a route to palladium-aerogel nanocomposites. Journal of
Materials Chemistry 14, 1212–7.
20. Patti, J. M., Allen, B. L., McGavin, M. J. et al. (1994).
MSCRAMM—mediated adherence of microorganisms to host tissues.
Annual Review of Microbiology 48, 585–617.
21. Bayston, R., Grove, N., Siegel, J. et al. (1989). Prevention of
hydrocephalus shunt catheter colonisation in vitro by impregnation with
1024
antimicrobials. Journal of Neurology, Neurosurgery, and Psychiatry 52,
605–9.
22. Schierholz, J. M., Lucas, L. J., Rump, A. et al. (1998). Efficacy
of silver-coated medical devices. Journal of Hospital Infection 40,
257–62.
23. Wilcox, M., Kite, P. & Dobbins, B. (1998). Antimicrobial
intravascular catheters—which surface to coat? Journal of Hospital
Infection 40, 322– 3.
24. Darouiche, R. O., Raad, I. I., Heard, S. O. et al. (1999). A
comparison of two antimicrobial-impregnated central venous catheters.
New England Journal of Medicine 340, 1 –8.
Downloaded from http://jac.oxfordjournals.org/ at Southern Methodist University on August 31, 2014