ARTICLE IN PRESS

Biomaterials 25 (2004) 5547–5556

Silver-coated megaendoprostheses in a rabbit model—an analysis

of the infection rate and toxicological side effects
Georg Goshegera,*, Jendrik Hardesa, Helmut Ahrensa, Arne Streitburgera, Horst Buergerb,
Michael Errenc, Andreas Gunseld, Fritz H. Kemperd, Winfried Winkelmanna,
Christof von Eiffe
a
Department of Orthopedics, University of Muenster, Albert-Schweitzer-Str. 33, 48149 Muenster, Germany
b
Institute of Pathology, University of Muenster, DomagkstraX e 17, 48149 Muenster, Germany
c
Institute of Clinical Chemistry and Laboratory Medicine, University of Muenster, Albert-Schweitzer-Str. 33, 48149 Muenster, Germany
d
Environmental Specimen Bank for Human Tissues, University of Muenster, DomagkstraX e 11, 48149 Muenster, Germany
e
Institute of Medical Microbiology, University of Muenster, DomagkstraX e 10, 48149 Muenster, Germany
Received 11 August 2003; accepted 27 December 2003

Abstract

Deep infection of megaprostheses remains a serious complication in orthopedic tumor surgery. Despite the use of systemic and
local antibiotic prophylaxis the reported infection rate is between 5% and 35%. Silver-coated medical devices proved their
effectiveness in reducing infections. The objective of this study was to examine in vivo the antimicrobial efficacy and possible side-
effects of a silver-coated megaprosthesis. In a first study, 30 rabbits (15 titanium versus 15 silver-coated Mutarss—endoprostheses)
were infected with Staphylococcus aureus. In a second study, toxicological side effects were analyzed in 10 rabbits with a silver-
coated megaprosthesis.
The silver group showed significantly (po0.05) lower infection rates (7% versus 47%) in comparison with the titanium group.
Measurements of the C-reactive-protein, neutrophilic leukocytes, rectal temperature and body weight showed significant (po0.05)
lower signs of inflammation in the silver group. The analysis of the silver concentration in blood (median 1.883 ppb) and in organs
(0.798–86.002 ppb) showed elevated silver concentrations without pathologic changes in laboratory parameters and without
histological changes of organs. In conclusion, the new silver-coated Mutarss—megaprosthesis resulted in reduced infection rates
without toxicological side effects, suggesting that this prosthesis might be a promising device in tumor surgery exhibiting
antimicrobial activity.
r 2004 Elsevier Ltd. All rights reserved.

Keywords: Antimicrobial; Antibacterial; Bone; Metal ion release; Metal ion toxicity; Metal surface treatment

1. Introduction gical response was observed despite repeated revision

Since the early 1980s endoprosthetic replacement of
large bone defects after tumor resection has been
performed with satisfactory long-term results. The
overall endoprosthetic survival rate has been quoted to
be 79–87% at 5 years, 71–80% at 10 years and 56% at
15 years [1,2]. Deep infection has been reported to be the
most common complication. In the last decade, infec-
tion rates between 5% and 35% have been described
[2–7]. In some patients a poor clinical and bacteriolo-
*Corresponding author. Tel.: +49-251-83-47901/47963; fax: +49-
251-83-47989.
E-mail address: goshegg@uni-muenster.de (G. Gosheger).
surgery and adequate intravenous antibiotic therapy.
Capanna et al. [3] reported a reinfection rate of 43%
after revision surgery in cases of endoprosthetic infec-
tion. Secondary amputation or hip disarticulation is in
some cases the only solution in order to control the
infection [1,2,5,7–10].
Recently, the silver coating of foreign materials such
as heart valves, cardiac catheters and urinary catheters
has proved to reduce the infection rate of medical
devices [11–16]. The silver coating of orthopedic
endoprostheses has not been reported so far. In the
animal model described here, the infection rate of
silver-coated versus noncoated prostheses after chal-
lenge with Staphylococcus aureus was determined

0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2004.01.008
 ARTICLE IN PRESS
5548 G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556

and the silver concentrations in blood, urine and organs

with possible toxic side effects were documented.
Although experimental, these findings may be of clinical
relevance and provide the basis for further studies
in humans.

2. Methods

The study was approved by the institutional review

board of the Animal Care and Use Committee (G 10/
2000) and was conducted with reference to the OECD
Principles of Good Laboratory Practice.
Healthy adult female New Zealand white rabbits,
weighing 3300–5200 g were used. The rearing of the
animals was nearly identical concerning vaccination,
accommodation and feeding. The rabbits were obtained
2 weeks prior to surgery to allow acclimatization to the
housing in the central animal laboratory. The animals
were caged individually and fed orally with K 810
(Hoeveler Corp.) and water ad libitum. Air temperature
was 20 C and humidity 60%.
Preoperative inclusion criteria were a weight of at
least 3500 g, a physiological differential blood cell count
and no inflammatory processes or wounds. Preopera-
tively the mean weight in both groups did not
significantly differ.
The infection rate was established in 15 rabbits with a
titanium–vanadium prosthesis without silver coating (T-
group) (Fig. 1a) and in 15 rabbits with a silver-coated
endoprosthetic replacement of the diaphyseal femur (A-
group) (Fig. 1b) after injecting a bacterial suspension
(0.8 ml) of 5  104 cfu S. aureus. Preoperatively, a
randomization was performed. On each day one silver-
coated and one noncoated prosthesis were implanted.
The operation sequence was randomized. During the
follow-up (third week) one rabbit in the A-group had to
be sacrificed because of a rotational failure of the
endoprosthesis.
The silver coating was realized by a galvanic deposi-
tion of elementary silver (percentage purity of 99.7%)
on the prosthetic surface. The layer thickness ranges
from 10 to 15 mm. Silver ions are not able to dissolve, if
the silver is directly applied to a titanium surface.
Therefore, a sustained release of the antimicrobial silver
ions cannot be accomplished. A more noble metal like
gold is necessary to serve as a cathode material so
as to drive the release of silver ions as the anode.
For this reason a 0.2 mm thick gold layer was applied
by the vapor deposition technique on the titanium
surface before.
Preoperatively, the operating field (left leg and abdo-
men) was shaved. The operations were performed under
strict aseptic conditions. Anesthesia and analgesia were
done with intramuscularly injected xylacin (30 mg/kg)
and ketamin (50 mg/kg). No artificial ventilation was
Fig. 1. (a) Diaphyseal titanium implant; (b) Silver-coated implant.
necessary because the narcotics did not cause a
respiratory depression.
The skin incision (length 8 cm) was performed
laterally between the greater trochanter and the distal
lateral epicondyle of femur. First, the distal osteotomy
was performed and after removing the remaining dorsal
muscles the proximal osteotomy was done. After
removing the shaft of femur, the proximal and distal
part of the prosthesis were cemented. The cement did
not contain any antibiotics. The proximal and distal
part of the prosthesis were connected with two screws.
The operation time was on average 67 min (standard
deviation of 15.5 min in the silver-coated group and
11.6 min in the noncoated group).
The S. aureus strain (Strain number A 22616-5) which
was used in these experiments was isolated from bone
specimen of a 34 years-old female patient with chronic
osteomyelitis. Preoperatively a defined suspension was
prepared with a final inoculum of 5  104 colony
forming units. On each screw 0.4 ml of the suspension
were injected with a syringe. The fascial closure was
done continuously, so that the bacterial suspension did
not spread in the surrounding tissue.
Postoperatively, the rabbits were observed for 90 days.
Then the animals were sacrificed by a hard blow behind
the ears followed by exsanguination. Clinically, beside
the general condition, weight (weekly) and temperature
 ARTICLE IN PRESS
G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556
(biweekly) were documented. Furthermore, inflamma-
tory laboratory parameters were documented on the 7th,
14th, 21st, 28th, 56th and 84th postoperative day. These
included a differential blood count and CRP levels. In
the literature no standard values for the differential
blood count and the C-reactive protein (CRP) have been
reported for New Zealand rabbits. Therefore, preopera-
tively standard and threshold values of the differential
blood cell count and the CRP were determined in 40
healthy rabbits. Standard values were assumed, if they
ranged between 2.5% and 97.5% percentile. The values
were as follows: leukocytes: 6340–14280/ml, neutrophilic
granulocytes: 19.0–37.6%, monocytes: 3.0–11.2%, lym-
phocytes: 50.9–72.5%, mean corpuscular volume (MCV)
53.6–72.6 fl, CRP: 0–1 mg/dl.
In the course of the experiments endoprosthetic
surface was punctured directly through the scar by a
needle. Samples were taken alongside the whole surface
available by one puncture site on the 7th, 14th, 28th and
56th postoperative day and evaluated for S. aureus in all
cases. Specimens were cultured on Columbia blood agar
and also incubated in a brain–heart infusion (35 C, for
X48 h). Liquid culture was incubated up to 14 days, if
there was no growth. Identification of S. aureus was
based on conventional criteria, including the coagulase
tube test and the API Staph system (ATB 32 staph,
bioMerieux, Marcy-L’Etoile, France).
Postmortem examinations were conducted within
2–4 h of death. The examination focused on the tissue
around the endoprosthesis. For this reason the implant
and bone with surrounding tissues were divided into
three pieces by a non sterile diamond saw. Macro-
scopically, the tissue was examined for necrotic areas
and especially pus. After decalcification with 15%
trichloroacetic acid, 4% formalin and 2% zinc chloride
microscopical examination was done by means of 3–
4 mm thick sections of the proximal, diaphyseal and
distal femur after removing the implant. Specimens were
particularly searched for signs of acute or chronic
inflammatory infiltrates using a hematoxylin/eosin and
gram stain. Leukocytes in the pus were documented in a
counting chamber. Infection was established, if the
isolation of S. aureus was possible or if there were
histological signs of infection or leukocyte counts of
more than 50,000/mm3.
Toxicological examinations concerning the silver
coating were done in 10 rabbits with a silver-coated
prosthesis and in 10 animals with noncoated prosthesis.
Postoperatively, one rabbit with a silver-coated pros-
thesis was sacrificed because of a rotational failure of the
prosthesis. Preoperatively, the silver concentrations in
blood and urine were measured. Silver concentrations
were determined as parts per billion (ppb) by the
inductively coupled plasma mass spectrometry (ICPMS)
(see Table 1). Additionally, the following parameters
were determined: protein, glucose, urea, creatinine,
5549
Table 1
Silver concentration of blood and organs measured by an inductively
coupled plasma mass spectrometry (ICPMS)
Ti–Al–V n=10 Silver n=9
Brain 0.347 4,561
Heart 0.222 0,798
Liver 0.256 86,002
Lung 0.368 9,022
Spleen 0.254 27,960
Kidney 0.152 1,385
Blood 0.049 1,883
Silver concentration (median ppb) po0.01.
glutamic–oxalo-acetic transaminase (GOT), glutamic–
pyruvic transaminase (GPT), g-glutamyltransferase
(GGT), alkaline phosphatase (AP), lactate dehydrogen-
ase (LDH), creatine kinase (CK) and lipase. Since there
were no standard values available for New Zealand
rabbits, these values were determined preoperatively in
40 healthy rabbits. Standard values were assumed, if
they ranged between 2.5% and 97.5% percentiles. The
standard values were as follows: protein 4.9–6.9 g/dl;
glucose 104–147 mg/dl; urea 11–27 mg/dl; creatinine
0.7–1.3 mg/dl; GOT 5–14 U/l; GPT 15–43 U/l; GGT
3–10 U/l; AP 1–33 U/l; LDH 48–259 U/l; cholinesterase
120–2919 U/l; creatine kinase 105–1117 U/l and lipase
333–933 U/l. Postoperatively, the former mentioned
values were analyzed at the second, fourth, eighth and
12th postoperative week. The last analyses were
performed postmortem and blood samples were taken
from the common carotid artery.
Postmortem examination included the inspection of
the skin after shaving. The inspection focused on the
presence of an ash-colored skin, which can occur in
argyria (systemic silver intoxication). Macroscopic and
microscopic examination of the femur were performed
to evaluate the presence of necrotic tissue, inflammatory
infiltrates and foreign body granulomas. Special interest
was laid on the stability of the silver coating and the
presence of silver sedimentations in the surrounding soft
tissues. Histological samples were obtained from the
brain, heart, thymus, liver, kidney, spleen, ovary and
testes. The organs were examined histologically after
formalin fixation.
The statistical evaluation was performed with the
programs Excel and Statistical Package for Social
Sciences (10.0). All values for each single parameter,
from one group, were compared with the corresponding
values from the other group over the entire trial period.
The parameters of the differential blood count, CRP,
temperature, weight and silver concentration of the A-
and T-group were compared using the Mann–Whitney
U test. The tests had a significance level of at least 5%
and a power of 80%. The microscopic and macroscopic
results were analyzed with the Chi-square test.
 ARTICLE IN PRESS
5550 G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556
3. Results
3.1. Weight
Preoperatively, the mean weight in the A-group was
4510 and 4665 g in the T-group. While in the A-group
the weight increased continuously, the weight showed
fluctuations in the T-group with a notable reduction
during the postoperative period. Overall, 3 months
postoperatively animals of the A-group revealed a mean
weight gain of 558 g and the animals of the T-group a
mean weight gain of 151 g (Fig. 2). These results were
statistically significant (p=0.001).
3.2. Temperature
The mean temperature of all animals at every
measurement (biweekly) was in between the standard
values (range 38.5–39.5 C). However, in the A-group
the temperatures were significantly lower (po0.001).
3.3. Laboratory studies
The white blood cell count preoperatively and in the
first postoperative week showed no or only slight
differences between the two groups. After the second
week an opposing development between the two groups
was observed. Concern to the leukocytes, in the A-group
the mean value decreased from 10956/ml in the first
postoperative week to 9911/ml in the second week. In the
course of the trial period there was a slight increase up
to 10585/ml at the last measurement. In contrast the
mean leukocytes count in the T-group increased until
the 56th day up to 15173/ml. At the last measurement the
leukocytes were 13.791/ml in the T-group compared to
10.585/ml in the A-group. These differences were
statistically significant (p=0.043).
In the first postoperative week the number of
neutrophilic granulocytes was relatively increased in
Fig. 2. Weight gain/loss in comparison to the starting weight.
the T-group with a mean value of 44.5% (standard value
19.0–37.6%). The values in the A-group were also
initially moderately elevated in comparison with the
preoperative values. However, during the follow-up
period the values normalized. The overall difference
between both groups was statistically significant
(po0.014) (Fig. 3). The number of monocytes showed
a greater elevation in the T-group, particularly at the
56th and at the last measurement, than in the A-group
(p=0.021). However the threshold value of 11.0% was
never exceeded.
The number of lymphocytes decreased in the T-group
from 61.0% preoperatively to 39.5% at day 56 (normal
values, range 50.9–72.5%). At the last measurement an
increase to 46.0% was observed. In the A-group the
mean values decreased from 60.1% preoperatively to
49.3% in the first postoperative week. In the following
weeks a continuous increase to 56.3% was observed
(Fig. 4). The overall difference between the groups was
statistically significant (p=0.001).
It must be noticed that two animals of the A-group
had a subcutaneus abscess of the abdominal wall at the
90th postoperative day (rabbit 13 and 14 in Table 3).
Fig. 3. Neutrophilic distribution in mean percentage. Threshold value
19.0–37.6%.
Fig. 4. Lymphocytes distribution in mean percentage. Threshold value
50.9–72.5%.
 ARTICLE IN PRESS
G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556 5551

Leucocytes Leucocytes Granulocytes Granulocytes Lymphocytes Lymphocytes Monocytes Monocytes

Therefore, the values for the differential blood count for

3 months
these two animals were excluded for this point in time.

14,2
10,2
16,7

12,3

18,1

12,9
7,2

7,9

8,5

7,1
8,1
7,7
8,4
9,1
8,4
However, postmortem section revealed no evidence of
deep prosthetic infection in both cases. Including these

Overview of the macroscopic/microscopic and microbiologic results, the weight and the laboratory values preoperatively and on the 90th postoperative day in the T-group (n=15)
values in the statistical analysis there is still an overall

Preop.
significance for the parameters neutrophilic granulo-

10,3

11,1
9,4
7,9
4,6
8,7
6,3
3,3
6,2
7,6
6,7
7,1

7,5
6,3
8,4
cytes (p=0.032), lymphocytes (p=0.009) and monocytes
(p=0.034). Concerning the leukocytes (p=0.065) there

3 months
is still a trend to notice but not a significant result.

66

8,8
71
19,9
13,5
20,6
60,2
18,1
65,1
19,9

47,6
66,8
62,1

55,1
62,8
The CRP values in the T-group were always higher
when compared to the A-group (p=0.015). Never-
theless, the CRP in the T-group was only increased in
the first two postoperative weeks and at the 56th

Preop.

61
63,2
61,8
58,8
66,3
52,8
56,2
58,2

63,5
69,2
64,6
50,9
51,1
72,6
64,2
postoperative day above the threshold value of 1 mg/
dl. In the A-group the CRP values were within the
normal range.

3 months

59

60
64,5
73,2

31,5
67,3
26,1

22,9
37,3
24,5
28,1
79,2
19,6
33,8
27,4
3.4. Macroscopic and microscopic examination

Macroscopically, pus was detected in one of 14

rabbits (7.1%) in the A-group. In the T-group pus was

Preop.

25,1
27,6
31,6
27,8
36,8
36,4
36,2

25,7
23,1
22,8
35,9
38,5
18,9
25,8
30
observed in seven from 15 rabbits (46.6%) (po0.025).
Microscopically, in four animals of the A-group (28.6%)

3 months
and in 10 animals of the T-group (66.7%) inflammatory

8930

9130

9060

9500
11230
19350
13620

15090
19120
18570

10850

11710
32910
10540
12810
infiltrates were observed (po0.05) (Tables 2 and 3).

3.5. Microbiologic examination

Postmortem Preop 3 months Preop.

9460
9050

7680

7260

8310

6340
12090
11520
12790
17000

12580

10860

16910
15350

10620
Microbiologic examination identified S. aureus in
three rabbits in the A-group. In one rabbit S. aureus was
isolated on the 56th and 90th day without macroscopi-
CRP

1,2
0,4
1,8
0,4
0,4
0,2
3,3
0,0
1,4
0,2
0,3
1,7
0,0
0,2
0,0
cally visible pus on postmortem section (rabbit 1 in
Table 3). In a second rabbit S. aureus grew on the 14th
CRP

0,4
0,4
0,3
0,1
0,1
1,0
1,4
0,1
0,0
0,2
0,0
1,0
0,0
0,1
1,0
postoperative day, but was not subsequently isolated
(rabbit 14 in Table 3). In a third animal S. aureus was
cultivated from the pus on the 90th day (rabbit 9 in
Weight Weight

4700
5110
5020
4850
3550
3150
3700
5600
4900
5850
4800
4200
4600
5300
5900
Table 3). In this case inflammation was also proven
histologically.
Inflammatory Bacteria at section Preop.

In the T-group S. aureus was isolated in five rabbits

5250
5560
4730
5120
4350
3300
4430
5520
4630
5450
4150
4550
3670
4700
5540
with visible pus on the 90th postoperative day. In two
rabbits with macroscopically visible pus S. aureus was
not cultivated (rabbits 21 and 26 in Table 2). More than
Microbiology

50.000 Leucocytes/mm3 was measured in the aspiration

Gram neg.
S. aureus

S. aureus

S. aureus

S. aureus

S. aureus

fluid of animals with macroscopic evidence of pus.










3.6. Toxicological results

Macroscopic Microscopic

infiltrates

Nine rabbits with a silver-coated endoprosthesis were

Neg.

Neg.

Neg.

Neg.

Neg.
Pos.
Pos.
Pos.

Pos.
Pos.
Pos.

Pos.

Pos.
Pos.

Pos.

examined concerning toxicological side effects. The

mean protein concentration, glucose concentration,
(Titanium) Pus/no pus

urea, GOT, GPT, GGT, AP, LDH, choline esterase

no pus

no pus

no pus

no pus

no pus
no pus

no pus

no pus

and CK were within normal limits at all times.

pus

pus

pus

pus

pus

pus

pus

Postmortem examination showed no macroscopically

pathological findings in the femur. All prostheses
Table 2

Animal

became well incorporated (Fig. 5). Microscopically, no

flaking of the silver coating could be noted. No attrition,
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
 ARTICLE IN PRESS
5552 G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556

Monocytes

3 months

13,7

10,2

13,4
9

6
6,2
5,5
7,5

6,5

9,7
9,7
5,6
9,4
6,4
Monocytes
Overview of the macroscopic/microscopic and microbiologic results, the weight and the laboratory values preoperatively and on the 90th postoperative day in the A-group (n=14)

Preop.

13,9

10,1
5
3
6,4
6,3
6,8
8,6
9,6
5,5

7,7

0,7
5,2

7,6
Lymphocytes

3 months

69
45,6
63,3
61,2

53,2
58,9
52,3
70,2
16,8
55,7
58,1
71,5
33,5
17,2
Lymphocytes

Fig. 5. Macroscopic findings demonstrating a well incorporated

Preop.

implant and stable silver coating (three axial sections of the implant).
65

50

62
64,2
61,6
63,7

58,6
67,6

62,9
63,1
51,6
60,1
57,5
53,6
In tight connection to the implant a rim of ossified tissue could be seen
macroscopically. The presence of grey material at the periphery of
Granulocytes

section is a technical artefact due to the sectioning of the prosthesis and

3 months

consequent impregnation of the metal powder.

39

32
26,9

21,3
35,9
32,6
35,9
22,2
68,1
32,8
30,2
21,9
52,8
72,6
Granulocytes

Preop.

34
27,1
29,7
28,1
23,5
30,7
25,5
34,5
27,2
23,6

33,1
35,9
33,3
26,6
Leucocytes

3 months

9530
6810
9350

9540

8660

6510
14580

12700

13490

13240
12100

10410
16670
16010
Leucocytes

Preop.

8540
7650

7660

8780
7280
16340
10770

13930
14280
11190

10430

13260
13090
10190
3 months
CRP

0
0

0
0
0
0
1,6
0,2
1,5
1,8

0,1

3,4
0,7
Preop.
CRP

0
0
0

0
1

0
0
0,2
0,3
0,1
0,1

0,1

0,2

0,4

Fig. 6. Microscopic sections of the periprosthetic environment with no

Postmortem

evidence of inflammation or foreign body granuloma.

Weight

(10  amplification).
4400
5600
5200
4830
4350
5300
5400
4780
4750
5180
4900
4550
4830
4700
Weight

Preop.

5180
5200
4650
4450
3900
4500
4900
4050
5180
3470
4400
3910
4900
4620

chronic inflammatory infiltrates or foreign body gran-

ulomas were detected (Fig. 6). Histological examination
Bacteria at section

of the organs revealed no abnormal pathologic findings.

Microbiology

The silver concentrations were determined in nine

S. aureus

S. aureus

rabbits with a silver-coated prosthesis. The mean silver

concentrations in the blood were significantly elevated














compared to the noncoated group (p=0.0003). The

Inflammatory

concentration was 1.88 ppb (range, 1.06–5.88) in the

Microscopic

infiltrates

silver-coated group and 0.029 ppb (range, 0.022–0.054)

in the T-group (Table 1).
Neg.
Neg.
Neg.

Neg.

Neg.

Neg.
Neg.
Neg.
Neg.
Neg.
Pos.

Pos.

Pos.

Pos.

The mean silver concentrations in the organs were

Macroscopic

always significantly elevated in the silver-coated group

Pus/no pus

compared to the noncoated group (Table 1). The highest

no pus
no pus
no pus
no pus
no pus
no pus
no pus
no pus

no pus
no pus
no pus
no pus
no pus

silver concentrations were found in the liver with a mean

pus

concentration of 90.4 ppb (range, 16.4–357.2). The

Table 3

Animal

(Silver)

spleen showed also an elevated silver concentration with

on average 27.96 ppb (range, 2.9–268.9).
10
11
12
13
14
1
2
3
4
5
6
7
8
9
 ARTICLE IN PRESS
G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556
4. Discussion
Foreign materials—e.g. intravascular catheters, pros-
thetic heart valves, cardiac pacemakers and orthopedic
devices—are frequently used in almost all fields of
modern medicine and are associated with a definitive
risk of bacterial infection. According to the underlying
patient characteristics, the microorganisms which are
implicated, and the type of the device, morbidity and
mortality of device-associated infections may vary,
however, foreign body associated infections significantly
contribute to the increasing problem of nosocomial
infections [17,18]. Staphylococci account for the major-
ity of infections both of temporarily inserted and of
permanently implanted orthopedic devices [18]. In
recent years, a battery of virulence factors involved in
the pathogenesis of foreign body-associated staphylo-
coccal infection has been defined and characterized [19].
Most important in the pathogenesis is the colonization
of the device surface by formation of a biofilm [19,20].
The clinical experience with foreign body-associated
staphylococcal infections clearly shows that host defense
mechanisms as well as antibacterial chemotherapy,
despite the use of antimicrobial agents with proven
in vitro activity, are often unable to cure these
infections, and in particular, to eliminate the staphylo-
cocci from the infected device [17,21]. Therefore, the
removal of the infected medical device is often
inevitable. Nevertheless, orthopedic devices cannot be
removed without significant morbidity [22]. Therefore,
prevention of these infections would have an important
impact on patient morbidity and the cost effectiveness of
hospital care [17,21].
Among metals with antimicrobial activity, silver has
raised the interest of many investigators because of its
good antimicrobial action and low toxicity [12,20,23,24].
Bacteria are quite susceptible to silver with bactericidal
activity having been reported at silver concentrations as
low as 35 ppb [25]. Previous studies have shown that
silver ions are released from the surface coating and
reach antimicrobially effective silver concentrations in
the surrounding tissue [26]. The antimicrobial effect is
caused by the silver ions, which bind to membranes,
enzymes and to nucleic acid. Thereby various reversible
and irreversible cellular interferences occur [12,24].
Silver ions inhibit the respiratory chain, disturbing the
aerobe metabolism of microorganisms [27].
In comparison to an antibiotic surface coating, silver-
coated megaprostheses might have major advantages.
Firstly, no resistance has been described so far.
Secondly, the effect of silver ions is continuous and
long lasting due to the oligodynamic effect of elemen-
tary silver [13,28]. Thirdly, the ability of many bacteria
to produce a biofilm is reduced and the likelihood of
bacterial colonization is decreased [11,20]. Olson et al.
[20] found that the use of silver-coated endotracheal
5553
tubes in dogs significantly lowered colonization rates
and reduced lung inflammation after bacterial challenge
with Pseudomonas aeruginosa.
These striking features of silver are reflected by
clinical studies with silver-based coatings on medical
devices. Sioshansi et al. [29] used ion implantation to
deposite silver-based coatings on a silicone rubber,
which thereafter demonstrated antimicrobial activity.
Also silver–copper surface films, sputter-coated onto
catheter materials, showed antibacterial activity against
Pseudomonas aeruginosa biofilm formation [30].
Silver has also been used extensively for the develop-
ment of infection-resistant urinary catheters [14,15,
31,32]. A novel central venous catheter containing
silver-sulfadiazine and chlorhexidine was evaluated in
a randomized, comparative trial in 204 patients in an
intensive care unit and proved to have a fourfold
reduced risk to produce bacteremia [33]. Other trials
have also demonstrated the clinical usefulness of
this catheter, which is already commercially available
[34–37]. In another randomized prospective study in
hemato-oncological patients a novel silver sulfate–
polyurethane catheter was associated with a significantly
lower rate of bacteremia than in the control group
(10.2% versus 22.5%) [38].
Collinge et al. [39] described a reduced infection rate
of silver-coated external fixation pins. Thirty six silver-
coated and 12 conventional stainless steel pins were
placed in the iliac crest of six sheep and inoculated with
Staphylococcus aureus. After 2 12 weeks the pin sites were
examined for inflammation and the pin tips were
quantitatively cultured and examined with scanning
electron microscopy (SEM). Eighty four per cent of the
uncoated pins were infected in comparison to 62% of
the silver-coated pins.
Furthermore, reduced infection rates of silver-coated
heart valves are reported in Refs. [12,16,26]. However,
the antimicrobial efficacy of silver-coated heart valves
still remains controversial, which can be seen in the
publication of Darouiche et al. [40] describing no
evidence of antimicrobial efficacy. Additionally, large
clinical trials revealed no significant differences between
silver-coated and uncoated urinary and central venous
catheters [41,42]. Schierholz et al. [28] raise concern that
free silver ions precipitate in albumin containing
environments leading to concentrations to low to
achieve bactericidal effects. In our opinion the local
concentration of free silver is additionally reduced in all
mentioned studies with direct contact to body fluids by
dilution.
In the present animal experiment, the superficial silver
coating in diaphyseal femoral replacements resulted in a
significantly reduced infection rate after injection of S.
aureus. Only one of 14 rabbits with a silver-coated
prosthesis macroscopically developed an infection. In
the noncoated group, seven rabbits (46.6%) had visible
 ARTICLE IN PRESS
5554 G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556
suppuration. The general condition concerning weight
gain and laboratory values (leucocytes and CRP) were
significantly better in the A-group. These findings
suggest, that in the A-group, the animals were able to
resist the injected bacteria. S. aureus was not able to
cause infection with systemic illness.
In view of silver’s potential for toxicity, regardless of
how limited it may appear to be, extensive research is
still essential [24]. The most common condition in silver
exposed people is argyria, a gray–blue discoloration of
the tissues (e.g. skin). It was seen more commonly in the
19th century associated with occupational exposure in
silversmiths, miner and photographs [24,43]. Argyria
can also appear from the use of silver-containing
medications [24,43–45]. In the literature the total silver
concentration, which is associated with a systemic
argyrosis, amounted to 4–6 g silver [12,16,46,47]. We
calculated the amount of silver, which is necessary for
the coating of a total femoral replacement in humans
(prosthesis length 350 mm, diameter 25 mm, thickness of
the coating 0.004 mm), to be 1.154 g. Therefore, systemic
argyrosis is unlikely.
In most cases the silver deposition does not cause
apparent ill-effects, despite markedly elevated silver
levels [43,45,48]. Nevertheless, there are case reports,
describing fatty degeneration of the liver, kidneys and
heart [43]. Toxic damage in the former mentioned
organs were described in Ref. [46]. Inhibition of the
proliferation of keratinocytes and fibroblasts after
treatment with silver-sulfadiazin has been described
[49,50].
There are some reports describing a chronic inflam-
matory reaction with thinner pannus and microabscess
formation around the sewing cuff in patients treated by
a silver-coated heart valve. The poor tissue ingrowth can
lead to loosening of the sutures. The authors assume a
toxic reaction deriving from the silver released from the
impregnated sewing cuff, which may lead to an
inhibition of normal fibroblast response [51–53]. Since
silver concentrations are not measured in the environ-
ment tissue, coherence of this factors cannot be proved
until now. Tweden et al. [16] reported no evidence of
silver toxicity to cultured fibroblasts until concentra-
tions reached levels of 1200 ppb. Therefore, the forma-
tion of a thinner pannus has not to be an expression of
cytotoxicity, but can be related to a possibly chronic
inflammatory reaction by elementary silver in the very
close environment of a silver-coated medical device. A
poor tissue ingrowth does not have any comparable
dramatic effect in orthopedic implants like it has in heart
valve replacement surgery. Nevertheless, inhibition of
osteoblasts by elementary silver cannot be excluded.
Therefore, in the current study the silver coating was not
expanded to the prosthetic stem. In a further animal
study the effect of silver-coated stems in a total hip
replacement concerning the osseous ingrowth and
cellular response will be assessed by histology, electron
microscopy and measuring bone proteins like, e.g.
osteocalcin and collagen I.
In contrast, other authors reported about no cyto-
toxic effects of silver and its good biocompatibilty
[11,54–57]. These conflicting results probably can be
explained by varying silver concentrations acting on
different cell types. It is well known that silver toxicity is
a dose-dependent process. Elementary silver is an inert
metal and is ionized only slowly in the organism [12,47].
Due to the oligodynamic effect, the silver release is
comparatively low and has been evaluated in previous
studies [13]. Therefore, elementary silver is regarded as
low cytotoxic only [54,58]. In the current study the silver
concentrations around the prosthesis and in the organs
did not cause histological changes or functional deficits
of the organs.
In this study silver concentrations in the blood were
on average 1.88 ppb. Silver levels below 10 ppb were
considered as normal [12]. Toxic side effects were
described for blood concentrations of 300 ppb in the
form of argyrosis, leukopenia, liver and kidney damage
[12,16,25]. In studies using silver-coated heart valves, the
blood concentration did not exceed 22 ppb [12]. In this
study no toxicological side effects were observed. In fact,
the silver concentration was significantly elevated in all
analyzed organs, especially in the liver and the spleen
with mean values of 90.4 and 27.9 ppb, respectively.
Although the functional parameters and histological
examination of the organs revealed no pathologic
findings.
Nevertheless, there are case reports describing adverse
effects below a concentration of 300 ppb. Sudmann et al.
[59] report about a 76-year old woman, in whom a
silver-impregnated bone cement was used in a total hip
replacement. Five years after insertion of the prosthesis
she developed serious neurological deficits with senso-
motoric deficits of the ipsilateral extremity despite silver
concentration in the joint cavity and in the serum did
not exceed 103 and 6.3 ppb, respectively. In general
neurologic side effect are extremely rare. In a study of 30
silver-intoxicated patients no paralysis of muscles are
reported [48].
In summary, silver coating of materials is easy to
perform with a galvanic coating of the endoprosthetic
device. The cost of a silver-coated endoprosthesis will be
about 5–7% higher than a non coated implant, whereas
a reduction in the incidence of infections would result in
less revision surgery including the reimplantation of a
new endoprosthesis. Therefore, prevention of these
infections would have an important impact on patient
morbidity and the cost effectiveness of hospital care
[17,21]. The data from the current animal experiment
also confirms the results of former studies, in which the
infection rate of silver-coated medical devices has been
reduced. In the future, it will be necessary to implant
 ARTICLE IN PRESS
G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556
silver-coated megaprostheses in humans in order to
assess possible toxic side effects and to determine the
infection rate of tumor patients with various diseases
and immunodeficiency caused by the tumor disease itself
and by chemo- and radiation therapy.
Acknowledgements
The authors gratefully thank Robert Grimer, M.D.,
Royal Orthopaedic Hospital, Birmingham for revising
the paper as a native speaking orthopedic surgeon who
is specialized in orthopedic oncology. The study was
supported by grants of implantcast corp., Buxtehude
Germany and by equipment of Merck company,
Darmstadt Germany and by Ethicon, Hamburg Ger-
many.
References
[1] Ham SJ, Schraffordt Koops H, Veth RP, van Horn JR, Molenaar
WM, Hoekstra HJ. Limb salvage surgery for primary bone
sarcoma of the lower extremities: long-term consequences of
endoprosthetic reconstructions. Ann Surg Oncol 1998;5(5):
423–36.
[2] Mittermayer F, Krepler P, Dominkus M, Schwameis E, Sluga M,
Heinzl H, et al. Long-term followup of uncemented tumor
endoprostheses for the lower extremity. Clin Orthop 2001;388:
167–77.
[3] Capanna R, Morris HG, Campanacci D, Del Ben M, Campanacci
M. Modular uncemented prosthetic reconstruction after resection
of tumours of the distal femur. J Bone Joint Surg Br 1994;76(2):
178–86.
[4] Malawer MM, Chou LB. Prosthetic survival and clinical results
with use of large-segment replacements in the treatment of high-
grade bone sarcomas. J Bone Joint Surg Am 1995;77(8):1154–65.
[5] Ritschl P, Capanna R, Helwig U, Campanacci M, Kotz R.
KMFTR (Kotz modular femur tibia reconstruction system)
modular tumor endoprosthesis system for the lower extremity.
Z Orthop Ihre Grenzgeb 1992;130(4):290–3.
[6] Safran MR, Kody MH, Namba RS, Larson KR, Kabo JM,
Dorey FJ, et al. 151 endoprosthetic reconstructions for patients
with primary tumors involving bone. Contemp Orthop
1994;29(1):15–25.
[7] Wirganowicz PZ, Eckardt JJ, Dorey FJ, Eilber FR, Kabo JM.
Etiology and results of tumor endoprosthesis revision surgery in
64 patients. Clin Orthop 1999;358:64–74.
[8] Hillmann A, Gosheger G, Hoffmann C, Ozaki T, Winkelmann
W. Rotationplasty-surgical treatment modality after failed limb
salvage procedure. Arch Orthop Trauma Surg 2000;120(10):
555–8.
[9] Kotz R, Ritschl P, Trachtenbrodt J. A modular femur-tibia
reconstruction system. Orthopedics 1986;9(12):1639–52.
[10] Shih LY, Sim FH, Pritchard DJ, Rock MG, Chao EY. Segmental
total knee arthroplasty after distal femoral resection for tumor.
Clin Orthop 1993;292:269–81.
[11] Boswald M, Mende K, Bernschneider W, Bonakdar S, Ruder H,
Kissler H, et al. Biocompatibility testing of a new silver-
impregnated catheter in vivo. Infection 1999;27(Suppl. 1):S38–42.
[12] Brutel de la Riviere A, Dossche KM, Birnbaum DE, Hacker R.
First clinical experience with a mechanical valve with silver
coating. J Heart Valve Dis 2000;9(1):123–9; discussion 129–30.
5555
[13] Jansen B, Rinck M, Wolbring P, Strohmeier A, Jahns T. In vitro
evaluation of the antimicrobial efficacy and biocompatibility of a
silver-coated central venous catheter. J Biomater Appl 1994;9(1):
55–70.
[14] Karchmer TB, Giannetta ET, Muto CA, Strain BA, Farr BM. A
randomized crossover study of silver-coated urinary catheters in
hospitalized patients. Arch Intern Med 2000;160(21):3294–8.
[15] Plowman R, Graves N, Esquivel J, Roberts JA. An economic
model to assess the cost and benefits of the routine use of silver
alloy coated urinary catheters to reduce the risk of urinary tract
infections in catheterized patients. J Hosp Infect 2001;48(1):33–42.
[16] Tweden KS, Cameron JD, Razzouk AJ, Holmberg WR, Kelly SJ.
Biocompatibility of silver-modified polyester for antimicrobial
protection of prosthetic valves. J Heart Valve Dis 1997;6(5):
553–61.
[17] von Eiff C, Proctor RA, Peters G. Coagulase-negative staphylo-
cocci. Pathogens have major role in nosocomial infections.
Postgrad Med 2001;110(4):63–4, 69–70, 73–6.
[18] Brause BD. Infected orthopedic prostheses. In: Bisno AL,
Waldrogel FA, editors. Infections associated with indwelling
medical devices. Washington, DC: American Society for Micro-
biology; 1989.
[19] von Eiff C, Peters G, Heilmann C. Pathogenesis of infections due
to coagulase-negative staphylococci. Lancet Infect Dis 2002;
2(11):677–85.
[20] Olson ME, Harmon BG, Kollef MH. Silver-coated endotracheal
tubes associated with reduced bacterial burden in the lungs of
mechanically ventilated dogs. Chest 2002;121(3):863–70.
[21] Herrmann M, Peters G. Catheter-associated infections caused by
coagulase-negative staphylococci: clinical and biological aspects.
New York: Marcel Dekker; 1997.
[22] Grimer RJ, Belthur M, Chandrasekar C, Carter SR, Tillman RM.
Two-stage revision for infected endoprostheses used in tumor
surgery. Clin Orthop 2002;395:193–203.
[23] Klasen HJ. Historical review of the use of silver in the treatment
of burns. I. Early uses. Burns 2000;26(2):117–30.
[24] Hollinger MA. Toxicological aspects of topical silver pharma-
ceuticals. Crit Rev Toxicol 1996;26(3):255–60.
[25] Chambers C, Proctor C, Kabler P. Bactericidal effect of low
concentrations of silver. J Am Water Works Association 1962:
208–216.
[26] Illingworth B, Bianco RW, Weisberg S. In vivo efficacy of silver-
coated fabric against Staphylococcus epidermidis. J Heart Valve
Dis 2000;9(1):135–41.
[27] Bragg PD, Rainnie DJ. The effect of silver ions on the respiratory
chain of Escherichia coli. Can J Microbiol 1974;20(6):883–9.
[28] Schierholz JM, Lucas LJ, Rump A, Pulverer G. Efficacy of silver-
coated medical devices. J Hosp Infect 1998;40(4):257–62.
[29] Sioshansi P. New processes for surface treatment of catheters.
Artif Organs 1994;18(4):266–71.
[30] McLean RJ, Hussain AA, Sayer M, Vincent PJ, Hughes DJ,
Smith TJ. Antibacterial activity of multilayer silver-copper
surface films on catheter material. Can J Microbiol 1993;39(9):
895–9.
[31] Newton T, Still JM, Law E. A comparison of the effect of early
insertion of standard latex and silver-impregnated latex foley
catheters on urinary tract infections in burn patients. Infect
Control Hosp Epidemiol 2002;23(4):217–8.
[32] Saint S, Elmore JG, Sullivan SD, Emerson SS, Koepsell TD. The
efficacy of silver alloy-coated urinary catheters in preventing
urinary tract infection: a meta-analysis. Am J Med 1998;105(3):
236–41.
[33] Sheng WH, Ko WJ, Wang JT, Chang SC, Hsueh PR, Luh KT.
Evaluation of antiseptic-impregnated central venous catheters for
prevention of catheter-related infection in intensive care unit
patients. Diagn Microbiol Infect Dis 2000;38(1):1–5.
 ARTICLE IN PRESS
5556 G. Gosheger et al. / Biomaterials 25 (2004) 5547–5556
.
[34] Bach A, Bohrer .
H, Bottiger BW, Motsch J, Martin E. Reduction
of bacterial colonization of triple-lumen catheters with antiseptic
bonding in septic patients. Anaesthesiology 1994;81:261.
[35] Maki DG, Wheeler SJ, Stolz SM. Study of a novel antiseptic-
coated central venous catheter. Crit Care Med 1999;19(Suppl.):99.
[36] Pai MP, Pendland SL, Danziger LH. Antimicrobial-coated/
bonded and -impregnated intravascular catheters. Ann Pharmac-
other 2001;35(10):1255–63.
[37] Ramsay J, Nolte F, Schwarzmann S. Incidence of catheter
colonization and catheter related infections with an antiseptic
impregnated triple lumen catheter. Crit Care Med 1994;22:
A115.
[38] Goldschmidt H, Hahn U, Salwender HJ, Haas R, Jansen B,
Wolbring P, et al. Prevention of catheter-related infections by
silver coated central venous catheters in oncological patients.
Zentralbl Bakteriol 1995;283(2):215–23.
[39] Collinge CA, Goll G, Seligson D, Easley KJ. Pin tract infections:
silver vs uncoated pins. Orthopedics 1994;17(5):445–8.
[40] Darouiche RO, Meade R, Mansouri M, Raad II. In vivo efficacy
of antimicrobial-coated fabric from prosthetic heart valve sewing
rings. J Heart Valve Dis 1998;7(6):639–46.
[41] Dahlberg PJ, Agger WA, Singer JR, Yutuc WR, Newcomer KL,
Schaper A, et al. Subclavian hemodialysis catheter infections: a
prospective, randomized trial of an attachable silver-impregnated
cuff for prevention of catheter-related infections. Infect Control
Hosp Epidemiol 1995;16(9):506–11.
[42] Riley DK, Classen DC, Stevens LE, Burke JP. A large
randomized clinical trial of a silver-impregnated urinary catheter:
lack of efficacy and staphylococcal superinfection. Am J Med
1995;98(4):349–56.
[43] White JM, Powell AM, Brady K, Russell-Jones R. Severe
generalized argyria secondary to ingestion of colloidal silver
protein. Clin Exp Dermatol 2003;28(3):254–6.
[44] Fisher NM, Marsh E, Lazova R. Scar-localized argyria secondary
to silver sulfadiazine cream. J Am Acad Dermatol 2003;49(4):
730–2.
[45] Steininger H, Langer E, Stommer P. Generalized argyrosis. Dtsch
Med Wochenschr 1990;115(17):657–62.
[46] Wan AT, Conyers RA, Coombs CJ, Masterton JP. Determination
of silver in blood, urine, and tissues of volunteers and burn
patients. Clin Chem 1991;37(10 Part 1):1683–7.
[47] Perrelli G, Piolatto G. Tentative reference values for gold, silver
and platinum: literature data analysis. Sci Total Environ
1992;120(1–2):93–6.
[48] Rosenman KD, Moss A, Kon S. Argyria: clinical implications of
exposure to silver nitrate and silver oxide. J Occup Med
1979;21(6):430–5.
[49] Kuroyanagi Y, Kim E, Shioya N. Evaluation of a synthetic
wound dressing capable of releasing silver sulfadiazine. J Burn
Care Rehabil 1991;12(2):106–15.
[50] McCauley RL, Linares HA, Pelligrini V, Herndon DN, Robson
MC, Heggers JP. In vitro toxicity of topical antimicrobial agents
to human fibroblasts. J Surg Res 1989;46(3):267–74.
[51] Butany J, Scully HE, VanArsdell G, Leask R. Prosthetic heart
valves with silver-coated sewing cuff fabric: early morphological
features in two patients. Can J Cardiol 2002;18(7):733–8.
[52] Tozzi P, Al-Darweesh A, Vogt P, Stumpe F. Silver-coated
prosthetic heart valve: a double-bladed weapon. Eur J Cardi-
othorac Surg 2001;19(5):729–31.
[53] Schaff HV, Carrel TP, Jamieson WR, Jones KW, Rufilanchas JJ,
Cooley DA, et al. Paravalvular leak and other events in silzone-
coated mechanical heart valves: a report from avert. Ann Thorac
Surg 2002;73(3):785–92.
[54] Greil J, Spies T, Boswald M, Bechert T, Lugauer S, Regenfus A,
et al. Analysis of the acute cytotoxicity of the Erlanger silver
catheter. Infection 1999;27(Suppl. 1):S34–7.
[55] Bosetti M, Masse A, Tobin E, Cannas M. Silver coated materials
for external fixation devices: in vitro biocompatibility and
genotoxicity. Biomaterials 2002;23(3):887–92.
[56] Hachem RY, Wright KC, Zermeno A, Bodey GP, Raad II.
Evaluation of the silver iontophoretic catheter in an animal
model. Biomaterials 2003;24(20):3619–22.
[57] Oloffs A, Grosse-Siestrup C, Bisson S, Rinck M, Rudolph R,
Gross U. Biocompatibility of silver-coated polyurethane catheters
and silver-coated Dacron material. Biomaterials 1994;15(10):
753–8.
[58] Liedberg H, Lundeberg T. Assessment of silver-coated urinary
catheter toxicity by cell culture. Urol Res 1989;17(6):359–60.
[59] Sudmann E, Vik H, Rait M, Todnem K, Andersen KJ, Julsham
K, et al. Systemic and local silver accumulation after total hip
replacement using silver-impregnated bone cement. Med Prog
Technol 1994;20(3–4):179–84.