Journal of Hospital Infection (1987) 10, 287-291

‘Tegaderm’ dressings prevent recolonization of

chlorhexidine-treated skin

Birgitta HolmstrGm* and Carin SvenssonT

Departments oj *Clinical Bacteriology, Central Hospital, Kristianstad and

tMedica1 Microbiology, University of Lund, Malm6 General Hospital, Sweden

Accepted for publication 15 February 198 7

Summary: ‘Tegaderm’, a semi-permeable, self-adhesive dressing made of

polyurethane film, applied to intact upper arm skin in Intensive Care Unit
(ICU) patients reduced the aerobic skin flora compared to adjacent exposed
skin sites. We also applied ‘Tegaderm’ to skin disinfected with chlorhexidine;
the aerobic skin flora remained unchanged at a low level for 5 days.

Introduction

Transparent, self-adhesive, semi-permeable dressings made of

polyurethane film (‘Tegaderm’, ‘OpSite’) have been introduced in the
health services during the last few years. The dressings were originally
mainly intended for burns but their range of uses has been extended and
they are increasingly used in connection with central venous catheters.
The effect of these dressings on the skin flora has not been fully
investigated. The protection they may afford against bacterial
contamination and their effect on skin prepared with chlorhexidine are not
known.
The dressings have an antibacterial effect in in-vitro studies (Holland et
al., 1984) but the effect on the skin flora has not been studied. In a study
comparing ‘OpSite’ and traditional dressings, applied to intravenous
catheter puncture sites (Kelsey & Gosling, 1984), the routines concerning
the cannulae and the treatment of the skin before application of the
dressings differed from ours. We therefore decided to study the effect of
‘Tegaderm’ on intact skin, and following skin preparation with 4%
chlorhexidine gluconate detergent.

Patients and methods

To achieve as realistic conditions as possible we tested ‘Tegaderm’ on
patients treated in an Intensive Care Unit (ICU). The diagnoses and
Correspondence to: Carin Svensson, Dept of Medical Microbiology, University of Lund, M&n6
General Hospital, 214 01 Malmii, Sweden

01954701/87/060287+05 $03.00/O 0 1987 The Hospital Infectmn Society

287
288 B. HolmstrSm and C. Svensson

treatments varied but the subjects may be considered representative of

patients treated in ICUs at Swedish regional hospitals. Seventy-five
patients were originally included in the study but 20 were subsequently
excluded for various reasons, e.g., the patient’s entire body was washed with
chlorhexidine or the dressing was removed in error. Of the 5.5 patients
remaining for evaluation 41 were men. The age range was 22-94 years
(mean 66 years).
The outside of the patients upper arms was chosen as the test site. In each
patient, one arm was disinfected with a chlorhexidine 4% gluconate
impregnated sponge for 1 min and dried with a clean towel. The other arm
(control arm) was left untreated. Samples were then taken with four contact
Rodac plates using blood-letheen agar from four sites assayed in vertical
order on both upper arms. After these initial cultures, a ‘Tegaderm’
dressing size 10 x 12 cm was applied to each arm so as to cover the middle
two cultured sites.
After 5 days, the dressings were removed and samples taken with contact
plates from the same sites as the initial cultures. The two outermost sites,
i.e., the upper and lower most, on each arm had thus been left exposed,
whereas the two middle sites had been covered.
The contact plates were incubated at 37°C for 48 h. The number of
colonies was counted and staphylococci, Gram-negative bacilli and
suspected pathogenic streptococci were typed. The degree of bacterial
colonization was described as the number of colony-forming units (cfu) on
two plates, i.e. approximately 50 cm2 area. The degree of colonization on
two skin sites was compared by means of Student’s t-tests on the mean
values.

Results

The results are summarized in Table I and Figure 1.

Chlorhexidine-disinfected arm
The mean bacterial count after chlorhexidine disinfectant was 9 cfu per
50 cm’. After 5 days, the flora on the sites not covered by the dressing had

Table I. Culture of aerobic skin flora from control and test sites

Day 0 Day 5

Skin left exposed Skin under

‘Tegaderm’

Control Chlorhexidine Control Chlorhexidine Control Chlorhexidine

cfu per 50 cm2

Median 178 2 241 237 27 8

Mean 431 9 522 378 202 35
Range 8-2000 o-148 12-2000 1 S-2000 O-2000 O-666
 Antibacterial effect under ‘Tegaderm’ 289

>I20 -

110 -

100 - .

90 - .

80 -
..
NE 70 -
0
::
\3 60 - .
.
x .
.
.
50 - * 0

.
.
40 - .
00 .
.

30 -

20 -
A40 . 00 .
. 000 .
IO -“D n n ooo” .
0 000 . .
**.
l .*0.=*I I I I I
0 5 IO I5 20 25 30 35 40 45 50 55
No. of patients

Figure 1. Quantitation of the aerobic skin flora. Each mark represents cfus per 50 cm2.
0, Control arm day 0; 0, chlorhexidine-washed arm day 0; A, control arm day 5; A,
chlorhexidine-washed arm day 5; E, control arm day 5 under ‘Tegaderm’; 0,
chlorhexidine-washed arm day 5 under ‘Tegaderm’.

recovered. Under the dressing, the bacterial count increased but the
difference between the counts at day 0 and 5 is not significant (P> 0.05). On
day 0, 49 of 55 patients (89%) were below 25 cfu per 50 cm2. On day 5
under ‘Tegaderm’ dressing, 45 patients (82%) were still at that low level.
The difference in the number of colonies on day 5 between skin covered by
the dressing and skin left exposed is significant.

Control arm
The control arm showed no difference between the initial cultures and those
after 5 days from exposed skin. However, the cultures from the skin sites
covered by the dressing for 5 days showed some reduction of the bacterial
count. Twenty-six of 55 patients (47%) were below 2.5 cfu per 50 cm*. The
290 B. HolmstrGm and C. Svensson

difference between the mean values of 431 and 202 cfu per 50 cm* is weakly
significant (PcO.05). This reduction in bacteria was not as great as that
obtained when the dressing was applied to skin disinfected with a
chlorhexidine impregnated sponge, where the difference between the mean
values of 202 and 35 cfu per 50 cm2 is statistically significant (PC 0.01).
Staphylococcus aweus, coliforms, Pseudomonas aeruginosa and other
bacteria, not belonging to the normal skin flora, were occasionally isolated
from the exposed skin sites but not from skin covered by the dressing.

Discussion
Some reduction of the aerobic skin flora could be demonstrated on intact,
untreated upper arm skin covered with a ‘Tegaderm’ dressing for 5 days. If
the aerobic skin flora was first reduced by disinfection with a chlorhexidine
impregnated sponge, the low bacterial count obtained remained largely
unchanged under the dressing for 5 days. Although ‘Tegaderm’ had an
antibacterial effect on untreated skin, the reduction did not lead to a low
level.
In order to achieve as low a level of skin bacteria as possible, the skin must
be disinfected before ‘Tegaderm’ is applied. Our results show a significant
difference between the chlorhexidine-disinfected arm (mean value 35 cfu
per 50 cm*) and the untreated arm (mean value 202 cfu per 50 cm2) after
protection with ‘Tegaderm’ for 5 days. We were unable to find S. aweus,
coliform bacteria or P. aeruginosa colonization when the skin was covered by
the dressing. Our study thus shows that the dressing has a protective effect
and is not permeable to bacteria from the surroundings.
We chose upper arms as the test site in order not to interfere with planned
or existing vascular catheterization, while obtaining conditions comparable
with those during central venous catheterization, e.g., distance from a
tracheostomy. A trial period of 5 days was chosen because the dressing
usually starts to loosen at the edges after this time.
Our results do not answer the question whether ‘Tegaderm’ dressings
offer advantages over conventional dressings in connection with, for
example, central venous catheterization, though they may be relevant
(Kelsey & Gosling, 1984). A detailed study of skin colonization, catheter
colonization and rate of infection at central venous catheter puncture sites
covered with ‘Tegaderm’ or a similar dressing after disinfection with
chlorhexidine is warranted, as the number of bacteria under the dressing
can thus be maintained at a very low level. A multicentre trial will probably
be necessary to obtain sufficient data within a reasonable period of time.

References
Holland, K. T., Davis, W., Ingham, E. & Gowland, G. (1984). A comparison of the in-vitro
antibacterial and complement activating effect of ‘OpSite’ and ‘Tegaderm’ dressings.
rournal of Hospital Infection 5, 323-328.
 Antibacterial effect under ‘Tegaderm’ 291

Kelsey, M. C. & Gosling, M. (1984). A comparison of the morbidity associated with

occlusive and non-occlusive dressings applied to peripheral intravenous devices. rournal
of Hospital Infection 5, 313-321.