Journal
of Hospital
Infection
Microbial
(1992)
21, 103-110
contamination preoperative
of brushes shaving
used for
S. Oie and A. Kamiya Department
of Pharmacy, Accepted
Yamaguchi
University
for publication
Hospital,
7 February
Ube, Japan
1992
Summary:
Microbial contamination of brushes used for preoperative shaving was investigated. Of the 24 brushed examined, 18 were contaminated with 10”lo9 colony forming units (cfu) per brush. Non-fermentative Gramnegative bacilli such as Pseudomonas aeruginosa and Xanthomonas maltophilia, and yeast-like fungi such as Candida parapsilosis were the primary contaminants. The mean bacterial count on the skin after the use of contaminated brushes (having a mean bacterial count 2.2 x lo8 cfu) in 14 subjects was 4.6 X lo* cfu 25 cm-*, which was about 100 times @J < 0.001) the control level. Contaminated brushes could not be disinfected with 80% ethyl alcohol, 0.1% sodium hypochlorite or 0.5% chlorhexidine. These findings suggest that the use of brushes should be avoided for preoperative shaving with a razor, and that sterile gauze and shaving foam should be used instead of a brush and soap. Keywords:
Shaving
brush;
microbial
contamination;
Pseudomonas
aeruginosa.
Introduction Extensive preoperative shaving with a razor produces many microscopic wounds even when it is done by a skilled member of staff’ and may cause postoperative wound infection. 2,3 For this reason, shaving with a razor has mostly been replaced by depilatories and electric shavers in the USA.46 In Japan, however, extensive shaving using a razor is still performed routinely on the day before operation, and brushes are often used to apply lather prior to shaving. The person who performs the shaving, the razor blade, the soap and the brush used to apply the lather, as well as the patient himself, are possible sources of infection of wounds. In particular, brushes are used repeatedly although they are difficult to clean, are often heavily contaminated, and might infect wounds. In fact, outbreaks of postoperative infection due to bacterial contamination of brushes for shaving were reported recently in the Correspondence Yamaguchi 0195-6701/92/060103
to: University
Akira Kamiya, Hospital, 1144
Ph.D., Kogushi,
Professor Ube 755,
+08 $03.00/O
and Director, Japan.
Department
0 1992 The Hospml
103
of
Pharmacy,
Infection
Society
104
S. Oie and A. Kamiya
UK and Spain. 7,8We studied microbial contamination of brushes used in our hospital for shaving, and the microbial transfer from contaminated brushes to the skin. We also evaluated methods for disinfecting contaminated brushes.
Materials
and
methods
Microbial contamination of brushes Twenty-four brushes used for shaving in six departments of our hospital were examined. Each brush was placed in a 500 ml bottle containing 200 ml sterile physiological saline and shaken for 30 min in a recipro-shaker (Taiyo SR-1; frequency 250 min-‘; amplitude 50mm). The number of bacteria contaminating each brush was estimated by performing viable counts on the saline. The predominant organisms were identified. Each saline sample was diluted 10, 102, lo3 and lo’-fold in sterile saline; 0.2 ml of the undiluted or diluted samples were transferred to trypticase soy agar and Sabouraud dextrose agar plates and spread with a glass ‘hockey stick’. Plates were incubated at 30°C for 24-72 h (trypticase soy agar) and at 25°C for 2-7 days (Sabouraud dextrose agar). The number of colonies were counted on each plate and the organisms were identified by Gram-staining, morphological examination, the OF test, the cytochrome-oxidase test and the ‘API’ system. Ten unused brushes were similarly examined. Five of them were made in Japan of raccoon dog hair and the remaining five were of pig hair and from Taiwan. The Taiwanese brushes were marked ‘sterilized’. Nurses at the six departments of our hospital were interviewed about the management of shaving brushes. Transfer of organisms from contaminated brushes to the skin Brushes that were found to be contaminated with 106-lo9 cfu per brush were immersed in the physiological saline in which the microbes had been released in the above test, and were allowed to dry at room temperature for 3 days. These ‘contaminated’ brushes were then moistened with tap water, lathered with sterilized solid soap, and applied to the skin of the left or right forearm region of 14 nurses who showed no skin lesions. Lather was applied to the skin of the other forearm with sterilized brushes that were autoclaved as a control. The sites where the lather was applied were immediately wiped with sterile gauze, and the number of microbes at these sites was estimated by microbial assay of the gauze. The same method as the above microbial assay of the brushes was used. The predominant microbes isolated were identified. The numbers of organisms in the brushes were also ascertained. Fisher’s t-test was used to measure any significant differences between the number of organisms on forearms obtained after application of lather with ‘contaminated’ or sterilized brushes.
Microbial
contamination
of brushes
105
Disinfection of contaminated brushes Contaminated brushes were prepared as above, i.e. by immersing the brushes used in the above test in contaminated physiological saline and allowing them to dry. When brushes were used repeatedly, they were contaminated by the same method before each test. Brushes were immersed in 80% ethyl alcohol (Dainippon Pharmaceutical Co.) for 10 min, 0.1% (1010 ppm available chlorine) sodium hypochlorite (‘Milton’, P&G) for 30 min, or 0.5% chlorhexidine (‘Hibitane’, ICI Pharmaceuticals) for 30 min. Microbial counts were performed on the brushes after this treatment.
Results
Microbial contamination of brushes Brushes for shaving were treated similarly in the six departments of our hospital. They were washed with tap water after use, immersed in chlorhexidine solution, washed again with water, and left in the room without covering. The concentration of chlorhexidine solution and the time of immersion were unknown. Table I shows the microbial counts and predominant microbial species detected in the 24 brushes examined. The level of microbial contamination was 103-lo9 cfu per brush, with 18 (75%) brushes giving counts of 106-lo9 cfu. The predominant microbial groups detected were non-fermentative Gram-negative bacilli such as Pseudomonas aeruginosa, Xanthomonas maltophilia, and Pseudomonas cepacia and yeast-like fungi such as Candida parapsilosis. Figure 1 shows the frequencies of these organisms. P. aeruginosa was detected most frequently (10 times out of 24, 41.6%). In all of the five unused raccoon dog hair brushes the level of microbial contamination was about lo6 cfu per brush, and Bacillus spp. were the predominant organisms. In the five unused pig hair brushes, the level of contamination was 2 x lo2 cfu per brush or less. Transfer of microbes from contaminated brushes to the human skin Figure 2 shows the number of organisms transferred from contaminated brushes to the human skin in 14 subjects. The mean log count ( f standard deviation) of organisms in a skin area of 25 cm* was 3.71 f 0.62 with sterilized brushes that were autoclaved, but was 5.66f 0.54 with contaminated brushes. The microbial count on the skin was increased about 100 times (PC 0.001) by treatment with a contaminated brush. The microbial species detected on the skin treated with contaminated brushes were the same as those isolated from the contaminated brushes. The mean log count (f SD) of organisms isolated from the brushes used to contaminate the skin was 8.34 ( f 0.67).
S. Oie and A. Kamiya
106 Table
I. Culture
Sample
results
from
shaving
Microbial
No.
counts
1
2.0 x IO9
2
4.2 x IO*
3
4.0 x IO7
4
3.1 x lo7
5
2.6 x 10’
6
2.2 x IO’
7 8 9
I.5 x IO’ I.2 x IO’ I.1 x 10’
10
I.1 x IO’
11
9.6 x IO6
12
9.0 x IO6
13
8.4 x IO’
14
4.6 x IO6
15 16 17 18
4.1 2.5 I.4 I.0
19
4.0 x 105
20
I.6 x IO5
21
5.0 x IO4
22 23 24
4.5 x IO4 8.0 x IO3 4.0 x IO3
* Glucose
non-fermentative
x x x x
IO6 106 IO6 IO6
Gram-negative
brushes
used for
preoperative
Contaminants Achromobacter xylosoxidans Pseudomonas cepacia Pseudomonas luteola Achromobacter xylosoxidans Pseudomonas spp. Rhodotorula rubra Achromobacter xylosoxidans Pseudomonas aeruginosa Rhodotorula rubra Pseudomonas aeruginosa Pseudomonas cepacia Pseudomonas fluorescens Candida parapsilosis Pseudomonas aeruginosa Xanthomonas maltophilia Xanthomonas maltophilia Pseudomonas cepacia Pseudomonas aeruginosa Xanthomonas maltophilia Pseudomonas cepacia Pseudomonas aeruginosa Acinetobacter lwoffii Pseudomonas spp. Pseudomonas fluorescens Pseudomonas aeruginosa Pseudomonas spp. Xanthomonas maltophilia Pseudomonas aeruginosa Pseudomonas stutzeri Xanthomonas maltophilia Pseudomonas spp. Agrobacterium radiobacter GNGB* Aeromonas sobria Candida parapsilosis Pseudomonas aeruginosa Pseudomonas cepacia Candida parapsilosis Rhodotorula spp. Candida parapsilosis Pseudomonas aeruginosa Candida parapsilosis Xanthomonas maltophilia Candida parapsilosis Pseudomonas aeruginosa Aeromonas sobria GNGB GNGB bacilli.
hair
removal
(N
= 24)
Microbial Pseudomonas Condida
contamination
107
of brushes
aeruginoso porapsilosis
Xanthomonos
ma/faphi/io
Pseudomonas
cepacio
Achromobacter
xylosoxl’dans IO
20
30
40
%
Figure
1. Frequency
(%)
of isolation
Sterile
Figure 2. Microbiai counts from application of soap by contaminated * Sterile vs. contaminated brushes
of organisms
brushes (control)
the skin of the brushes. (mean cfu).
from
24 shaving
brushes.
Contaminated brushes
forearms
of
14 subjects
(25
cm’)
after
S. Oie and A. Kamiya
108 Table Disinfection
80%
Ethyl
0.1% 0.5%
II.
Disinfection
method
Time
lO/lO
Hypochlorite
30 min
616
Chlorhexidine
30 min
S/IO
cfus
of contaminated
brushes
brushes*
No. of brushes with counts of at least IO5
10 min
* Estimated
alcohol
of contaminated
were
in the
order
Range
(mean) organisms
of surviving (cfu)
7.3 x 105-5.0 x 10’ (1.8 x 107) 7.2 x 106-4.0x 10’ (2.7 x IO’) 2.5 x 1044.5x 107 (I.8 X IO’) of 10’109.
Disinfection of contaminated brushes Table II shows the effects of disinfectants on contaminated b ushes (107-lo9 cfu per brush). The level of contamination remained atad cfu per brush or above in most of the brushes after immersion in 80% ethyl alcohol for 10 min, 0.1% sodium hypochlorite for 30 min, or 05% chlorhexidine for 30 min. Therefore, these agents had little disinfectant effect on contaminated brushes.
Discussion
The handling of brushes used for preoperative shaving at our hospital was considered to be similar to that of most Japanese institutions. Eighteen of the 24 brushes examined showed high levels of microbial contamination (106-lo9 cfu per brush) and P. aeruginosa, a common hospital pathogen,‘*” was detected most frequently. In addition, microbes on contaminated brushes were found to be transferred to the skin at a level of lo4 cfu cmP2 by applying lather with the brushes. Therefore, if the skin is shaved after applying lather with contaminated brushes, organisms from the brushes may enter the small wounds produced by the razor blade. Discontinuation of the use of contaminated brushes is considered to be important for prevention of hospital-acquired infection. Bacillus spp. were occasionally detected at relatively high levels even in unused brushes, probably because these brushes were made of animal hair. With frequent repeated use, microbes contaminating the brushes tended to change from Bacillus spp. to. Gram-negative bacilli. Contaminated brushes could not be effectively disinfected although they were immersed in strong disinfectants such as 80% ethyl alcohol and 0.1% sodium hypochlorite” for relatively long periods. In suspension tests, these agents rapidly eradicate bacteria such as P. aeruginosa and P. cepacia.‘2j’3 This discrepancy in the disinfectant potency between the in-vitro condition and the actual use for disinfection of brushes indicates inactivation or failure
Microbial
contamination
of brushes
109
of penetration of disinfectants under clinical conditions.“,” Bacteria and yeast-like fungi attached to the hair of brushes are considered to protect themselves from external factors by secreting protectants such as not described in the extracellular polymeric materials. 16-18 Although methods, we found that non-fermentative Gram-negative bacilli such as P. aeruginosa,19~20which are believed to die readily in dry conditions, survived for over 1 month in brushes left in the room to dry. Organisms may have survived because the brushes did not dry properly. These observations show that shaving brushes can be heavily contaminated, and that their use in hospitals should be abandoned. For shaving with a razor, brushes and soap lather should be replaced with sterile gauze and shaving foam. 21Also, in countries such as Japan where shaving with a razor is still a routine procedure, depilation methods with depilatory creams, electric hair clippers, or electric shavers should be substituted for the razor.22 References 1. Hamilton 2. 3. 4. 5. 6. 7. 8.
9. 10.
11.
12. 13. 14. 15. 16. 17.
HW,
Hamilton
KR,
Lone
FJ. Preoperative hair removal. CanJ Surg 1977; 20: 269-275. Seropian R, Reynolds BM. Wound infections after preoperative depilatory versus razor preparation. AmJ Surg 1971; 121: 251-254. Alexander JW, Fischer JE, Boyajian M, Palmquist J, Morris MJ. The influence of hair-removal methods on wound infections. Arch Surg 1983; 118: 347-352. Craig CP. Preparation of the skin for surgery. Infect Control 1986; 7: 257-258. Terry BA. Cost-effective application of the Centers for Disease Control guideline for prevention of surgical wound infections. AmY Infect Control 1985; 13: 232-235. Garner J’S Guideline for prevention of surgical wound infections, 1985. Am J Infect Control 1986; 14: 71-82. Lewis AM, Stephenson JR, Garner J, Afshar F, Tabaqchali S. A hospital outbreak of Serratia marcescens in neurosurgical patients. Epidemiol Infect 1989; 102: 69-74. Wilhelmi I, Bernaldo de Quir6s JCL, Romero-vivas J, Duarte J, Rojo E, Bouza E. Epidemic outbreak of Serratia marcescens infection in a cardiac surgery unit. r Clin Microbial 1987; 25: 1298-1300. Emmerson AM. The role of the skin in nosocomial infection: a review. J Chemother 1989; l(Supp1. 1): 12-18. Olson MM, MacCallum J, McQuarrie DG. Preoperative hair removal with clippers does not increase infection rate in clean surgical wounds. Surg - Gynecol Obstet 1986: 162: 181-182. Rutala WA, Weber DJ. Environmental issues and nosocomial infections. In: Farber BF, Ed. Infection Control in Intensive Care. New York: Churchill Livingstone 1987; 152-153, 159-160. Koshiro A, Oie S. Bactericidal activity of ethanol against glucose nonfermentative Gram-negative bacilli. Microbios 1984: 40: 3340. Coates Dy A comparison of sodium ‘hypochlorite and sodium dichloroisocyanurate products. J Hosp Infect 1985; 6: 31-40. Best M, Kennedy ME, Coates F. Efficacy of a variety of disinfectants against Listeria spp. Appl Environ Microbial 1990: 56: 377-380. L&yd-&ans N, Springthorpe ?S, Sattar SA, Chemical disinfection of human rotavirus-contaminated inanimate surfaces. 3’Hyg (Camb) 1986; 97: 163-173. Miller MJ, Ahearn DG, Adherence of Pseudomonas aeruginosa to hydrophilic contact lenses and other substrata. J Clin Microbial 1987; 25: 1392-1397. Marrie TJ, Costerton JW. Prolonged survival of Serratia marcescens in chlorhexidine. Appl Environ Microbial 1981; 42: 1093-l 102.
110 18.
19. 20. 21. 22.
S. Oie and A. Kamiya Sticker D, Dolman J, Rolfe S, Chawla J. Activity of antiseptics against Escherichia co& growing as biofilms on silicone surfaces. Eur J Clin Micro&o1 Infect Dis 1989; 8: 974-978. Musa EK, Desai N, Casewell MW. The survival of Acinetobacter calcoaceticus inoculated on fingertips and on formica. J Hasp Infect 1990; 15: 219-227 McDade JJ, Hall LB. Survival of Gram-negative bacteria in the environment I. Effect of relative humidity on surface-exposed organisms. Am J Hyg 1964; 80: 192-204. Lowbury EJL, Ayliffe GAJ, Geddes AM, Williams, JD, Eds. Control of Hospital Infection. London: Chapman & Hall 1982; 80. Daschner F. Cost-effectiveness in hospital infection control-lessons for the 1990s. r Hasp Infect 1989; 13: 325-336.
of Hospital
Infection
Microbial
(1992)
21, 103-110
contamination preoperative
of brushes shaving
used for
S. Oie and A. Kamiya Department
of Pharmacy, Accepted
Yamaguchi
University
for publication
Hospital,
7 February
Ube, Japan
1992
Summary:
Microbial contamination of brushes used for preoperative shaving was investigated. Of the 24 brushed examined, 18 were contaminated with 10”lo9 colony forming units (cfu) per brush. Non-fermentative Gramnegative bacilli such as Pseudomonas aeruginosa and Xanthomonas maltophilia, and yeast-like fungi such as Candida parapsilosis were the primary contaminants. The mean bacterial count on the skin after the use of contaminated brushes (having a mean bacterial count 2.2 x lo8 cfu) in 14 subjects was 4.6 X lo* cfu 25 cm-*, which was about 100 times @J < 0.001) the control level. Contaminated brushes could not be disinfected with 80% ethyl alcohol, 0.1% sodium hypochlorite or 0.5% chlorhexidine. These findings suggest that the use of brushes should be avoided for preoperative shaving with a razor, and that sterile gauze and shaving foam should be used instead of a brush and soap. Keywords:
Shaving
brush;
microbial
contamination;
Pseudomonas
aeruginosa.
Introduction Extensive preoperative shaving with a razor produces many microscopic wounds even when it is done by a skilled member of staff’ and may cause postoperative wound infection. 2,3 For this reason, shaving with a razor has mostly been replaced by depilatories and electric shavers in the USA.46 In Japan, however, extensive shaving using a razor is still performed routinely on the day before operation, and brushes are often used to apply lather prior to shaving. The person who performs the shaving, the razor blade, the soap and the brush used to apply the lather, as well as the patient himself, are possible sources of infection of wounds. In particular, brushes are used repeatedly although they are difficult to clean, are often heavily contaminated, and might infect wounds. In fact, outbreaks of postoperative infection due to bacterial contamination of brushes for shaving were reported recently in the Correspondence Yamaguchi 0195-6701/92/060103
to: University
Akira Kamiya, Hospital, 1144
Ph.D., Kogushi,
Professor Ube 755,
+08 $03.00/O
and Director, Japan.
Department
0 1992 The Hospml
103
of
Pharmacy,
Infection
Society
104
S. Oie and A. Kamiya
UK and Spain. 7,8We studied microbial contamination of brushes used in our hospital for shaving, and the microbial transfer from contaminated brushes to the skin. We also evaluated methods for disinfecting contaminated brushes.
Materials
and
methods
Microbial contamination of brushes Twenty-four brushes used for shaving in six departments of our hospital were examined. Each brush was placed in a 500 ml bottle containing 200 ml sterile physiological saline and shaken for 30 min in a recipro-shaker (Taiyo SR-1; frequency 250 min-‘; amplitude 50mm). The number of bacteria contaminating each brush was estimated by performing viable counts on the saline. The predominant organisms were identified. Each saline sample was diluted 10, 102, lo3 and lo’-fold in sterile saline; 0.2 ml of the undiluted or diluted samples were transferred to trypticase soy agar and Sabouraud dextrose agar plates and spread with a glass ‘hockey stick’. Plates were incubated at 30°C for 24-72 h (trypticase soy agar) and at 25°C for 2-7 days (Sabouraud dextrose agar). The number of colonies were counted on each plate and the organisms were identified by Gram-staining, morphological examination, the OF test, the cytochrome-oxidase test and the ‘API’ system. Ten unused brushes were similarly examined. Five of them were made in Japan of raccoon dog hair and the remaining five were of pig hair and from Taiwan. The Taiwanese brushes were marked ‘sterilized’. Nurses at the six departments of our hospital were interviewed about the management of shaving brushes. Transfer of organisms from contaminated brushes to the skin Brushes that were found to be contaminated with 106-lo9 cfu per brush were immersed in the physiological saline in which the microbes had been released in the above test, and were allowed to dry at room temperature for 3 days. These ‘contaminated’ brushes were then moistened with tap water, lathered with sterilized solid soap, and applied to the skin of the left or right forearm region of 14 nurses who showed no skin lesions. Lather was applied to the skin of the other forearm with sterilized brushes that were autoclaved as a control. The sites where the lather was applied were immediately wiped with sterile gauze, and the number of microbes at these sites was estimated by microbial assay of the gauze. The same method as the above microbial assay of the brushes was used. The predominant microbes isolated were identified. The numbers of organisms in the brushes were also ascertained. Fisher’s t-test was used to measure any significant differences between the number of organisms on forearms obtained after application of lather with ‘contaminated’ or sterilized brushes.
Microbial
contamination
of brushes
105
Disinfection of contaminated brushes Contaminated brushes were prepared as above, i.e. by immersing the brushes used in the above test in contaminated physiological saline and allowing them to dry. When brushes were used repeatedly, they were contaminated by the same method before each test. Brushes were immersed in 80% ethyl alcohol (Dainippon Pharmaceutical Co.) for 10 min, 0.1% (1010 ppm available chlorine) sodium hypochlorite (‘Milton’, P&G) for 30 min, or 0.5% chlorhexidine (‘Hibitane’, ICI Pharmaceuticals) for 30 min. Microbial counts were performed on the brushes after this treatment.
Results
Microbial contamination of brushes Brushes for shaving were treated similarly in the six departments of our hospital. They were washed with tap water after use, immersed in chlorhexidine solution, washed again with water, and left in the room without covering. The concentration of chlorhexidine solution and the time of immersion were unknown. Table I shows the microbial counts and predominant microbial species detected in the 24 brushes examined. The level of microbial contamination was 103-lo9 cfu per brush, with 18 (75%) brushes giving counts of 106-lo9 cfu. The predominant microbial groups detected were non-fermentative Gram-negative bacilli such as Pseudomonas aeruginosa, Xanthomonas maltophilia, and Pseudomonas cepacia and yeast-like fungi such as Candida parapsilosis. Figure 1 shows the frequencies of these organisms. P. aeruginosa was detected most frequently (10 times out of 24, 41.6%). In all of the five unused raccoon dog hair brushes the level of microbial contamination was about lo6 cfu per brush, and Bacillus spp. were the predominant organisms. In the five unused pig hair brushes, the level of contamination was 2 x lo2 cfu per brush or less. Transfer of microbes from contaminated brushes to the human skin Figure 2 shows the number of organisms transferred from contaminated brushes to the human skin in 14 subjects. The mean log count ( f standard deviation) of organisms in a skin area of 25 cm* was 3.71 f 0.62 with sterilized brushes that were autoclaved, but was 5.66f 0.54 with contaminated brushes. The microbial count on the skin was increased about 100 times (PC 0.001) by treatment with a contaminated brush. The microbial species detected on the skin treated with contaminated brushes were the same as those isolated from the contaminated brushes. The mean log count (f SD) of organisms isolated from the brushes used to contaminate the skin was 8.34 ( f 0.67).
S. Oie and A. Kamiya
106 Table
I. Culture
Sample
results
from
shaving
Microbial
No.
counts
1
2.0 x IO9
2
4.2 x IO*
3
4.0 x IO7
4
3.1 x lo7
5
2.6 x 10’
6
2.2 x IO’
7 8 9
I.5 x IO’ I.2 x IO’ I.1 x 10’
10
I.1 x IO’
11
9.6 x IO6
12
9.0 x IO6
13
8.4 x IO’
14
4.6 x IO6
15 16 17 18
4.1 2.5 I.4 I.0
19
4.0 x 105
20
I.6 x IO5
21
5.0 x IO4
22 23 24
4.5 x IO4 8.0 x IO3 4.0 x IO3
* Glucose
non-fermentative
x x x x
IO6 106 IO6 IO6
Gram-negative
brushes
used for
preoperative
Contaminants Achromobacter xylosoxidans Pseudomonas cepacia Pseudomonas luteola Achromobacter xylosoxidans Pseudomonas spp. Rhodotorula rubra Achromobacter xylosoxidans Pseudomonas aeruginosa Rhodotorula rubra Pseudomonas aeruginosa Pseudomonas cepacia Pseudomonas fluorescens Candida parapsilosis Pseudomonas aeruginosa Xanthomonas maltophilia Xanthomonas maltophilia Pseudomonas cepacia Pseudomonas aeruginosa Xanthomonas maltophilia Pseudomonas cepacia Pseudomonas aeruginosa Acinetobacter lwoffii Pseudomonas spp. Pseudomonas fluorescens Pseudomonas aeruginosa Pseudomonas spp. Xanthomonas maltophilia Pseudomonas aeruginosa Pseudomonas stutzeri Xanthomonas maltophilia Pseudomonas spp. Agrobacterium radiobacter GNGB* Aeromonas sobria Candida parapsilosis Pseudomonas aeruginosa Pseudomonas cepacia Candida parapsilosis Rhodotorula spp. Candida parapsilosis Pseudomonas aeruginosa Candida parapsilosis Xanthomonas maltophilia Candida parapsilosis Pseudomonas aeruginosa Aeromonas sobria GNGB GNGB bacilli.
hair
removal
(N
= 24)
Microbial Pseudomonas Condida
contamination
107
of brushes
aeruginoso porapsilosis
Xanthomonos
ma/faphi/io
Pseudomonas
cepacio
Achromobacter
xylosoxl’dans IO
20
30
40
%
Figure
1. Frequency
(%)
of isolation
Sterile
Figure 2. Microbiai counts from application of soap by contaminated * Sterile vs. contaminated brushes
of organisms
brushes (control)
the skin of the brushes. (mean cfu).
from
24 shaving
brushes.
Contaminated brushes
forearms
of
14 subjects
(25
cm’)
after
S. Oie and A. Kamiya
108 Table Disinfection
80%
Ethyl
0.1% 0.5%
II.
Disinfection
method
Time
lO/lO
Hypochlorite
30 min
616
Chlorhexidine
30 min
S/IO
cfus
of contaminated
brushes
brushes*
No. of brushes with counts of at least IO5
10 min
* Estimated
alcohol
of contaminated
were
in the
order
Range
(mean) organisms
of surviving (cfu)
7.3 x 105-5.0 x 10’ (1.8 x 107) 7.2 x 106-4.0x 10’ (2.7 x IO’) 2.5 x 1044.5x 107 (I.8 X IO’) of 10’109.
Disinfection of contaminated brushes Table II shows the effects of disinfectants on contaminated b ushes (107-lo9 cfu per brush). The level of contamination remained atad cfu per brush or above in most of the brushes after immersion in 80% ethyl alcohol for 10 min, 0.1% sodium hypochlorite for 30 min, or 05% chlorhexidine for 30 min. Therefore, these agents had little disinfectant effect on contaminated brushes.
Discussion
The handling of brushes used for preoperative shaving at our hospital was considered to be similar to that of most Japanese institutions. Eighteen of the 24 brushes examined showed high levels of microbial contamination (106-lo9 cfu per brush) and P. aeruginosa, a common hospital pathogen,‘*” was detected most frequently. In addition, microbes on contaminated brushes were found to be transferred to the skin at a level of lo4 cfu cmP2 by applying lather with the brushes. Therefore, if the skin is shaved after applying lather with contaminated brushes, organisms from the brushes may enter the small wounds produced by the razor blade. Discontinuation of the use of contaminated brushes is considered to be important for prevention of hospital-acquired infection. Bacillus spp. were occasionally detected at relatively high levels even in unused brushes, probably because these brushes were made of animal hair. With frequent repeated use, microbes contaminating the brushes tended to change from Bacillus spp. to. Gram-negative bacilli. Contaminated brushes could not be effectively disinfected although they were immersed in strong disinfectants such as 80% ethyl alcohol and 0.1% sodium hypochlorite” for relatively long periods. In suspension tests, these agents rapidly eradicate bacteria such as P. aeruginosa and P. cepacia.‘2j’3 This discrepancy in the disinfectant potency between the in-vitro condition and the actual use for disinfection of brushes indicates inactivation or failure
Microbial
contamination
of brushes
109
of penetration of disinfectants under clinical conditions.“,” Bacteria and yeast-like fungi attached to the hair of brushes are considered to protect themselves from external factors by secreting protectants such as not described in the extracellular polymeric materials. 16-18 Although methods, we found that non-fermentative Gram-negative bacilli such as P. aeruginosa,19~20which are believed to die readily in dry conditions, survived for over 1 month in brushes left in the room to dry. Organisms may have survived because the brushes did not dry properly. These observations show that shaving brushes can be heavily contaminated, and that their use in hospitals should be abandoned. For shaving with a razor, brushes and soap lather should be replaced with sterile gauze and shaving foam. 21Also, in countries such as Japan where shaving with a razor is still a routine procedure, depilation methods with depilatory creams, electric hair clippers, or electric shavers should be substituted for the razor.22 References 1. Hamilton 2. 3. 4. 5. 6. 7. 8.
9. 10.
11.
12. 13. 14. 15. 16. 17.
HW,
Hamilton
KR,
Lone
FJ. Preoperative hair removal. CanJ Surg 1977; 20: 269-275. Seropian R, Reynolds BM. Wound infections after preoperative depilatory versus razor preparation. AmJ Surg 1971; 121: 251-254. Alexander JW, Fischer JE, Boyajian M, Palmquist J, Morris MJ. The influence of hair-removal methods on wound infections. Arch Surg 1983; 118: 347-352. Craig CP. Preparation of the skin for surgery. Infect Control 1986; 7: 257-258. Terry BA. Cost-effective application of the Centers for Disease Control guideline for prevention of surgical wound infections. AmY Infect Control 1985; 13: 232-235. Garner J’S Guideline for prevention of surgical wound infections, 1985. Am J Infect Control 1986; 14: 71-82. Lewis AM, Stephenson JR, Garner J, Afshar F, Tabaqchali S. A hospital outbreak of Serratia marcescens in neurosurgical patients. Epidemiol Infect 1989; 102: 69-74. Wilhelmi I, Bernaldo de Quir6s JCL, Romero-vivas J, Duarte J, Rojo E, Bouza E. Epidemic outbreak of Serratia marcescens infection in a cardiac surgery unit. r Clin Microbial 1987; 25: 1298-1300. Emmerson AM. The role of the skin in nosocomial infection: a review. J Chemother 1989; l(Supp1. 1): 12-18. Olson MM, MacCallum J, McQuarrie DG. Preoperative hair removal with clippers does not increase infection rate in clean surgical wounds. Surg - Gynecol Obstet 1986: 162: 181-182. Rutala WA, Weber DJ. Environmental issues and nosocomial infections. In: Farber BF, Ed. Infection Control in Intensive Care. New York: Churchill Livingstone 1987; 152-153, 159-160. Koshiro A, Oie S. Bactericidal activity of ethanol against glucose nonfermentative Gram-negative bacilli. Microbios 1984: 40: 3340. Coates Dy A comparison of sodium ‘hypochlorite and sodium dichloroisocyanurate products. J Hosp Infect 1985; 6: 31-40. Best M, Kennedy ME, Coates F. Efficacy of a variety of disinfectants against Listeria spp. Appl Environ Microbial 1990: 56: 377-380. L&yd-&ans N, Springthorpe ?S, Sattar SA, Chemical disinfection of human rotavirus-contaminated inanimate surfaces. 3’Hyg (Camb) 1986; 97: 163-173. Miller MJ, Ahearn DG, Adherence of Pseudomonas aeruginosa to hydrophilic contact lenses and other substrata. J Clin Microbial 1987; 25: 1392-1397. Marrie TJ, Costerton JW. Prolonged survival of Serratia marcescens in chlorhexidine. Appl Environ Microbial 1981; 42: 1093-l 102.
110 18.
19. 20. 21. 22.
S. Oie and A. Kamiya Sticker D, Dolman J, Rolfe S, Chawla J. Activity of antiseptics against Escherichia co& growing as biofilms on silicone surfaces. Eur J Clin Micro&o1 Infect Dis 1989; 8: 974-978. Musa EK, Desai N, Casewell MW. The survival of Acinetobacter calcoaceticus inoculated on fingertips and on formica. J Hasp Infect 1990; 15: 219-227 McDade JJ, Hall LB. Survival of Gram-negative bacteria in the environment I. Effect of relative humidity on surface-exposed organisms. Am J Hyg 1964; 80: 192-204. Lowbury EJL, Ayliffe GAJ, Geddes AM, Williams, JD, Eds. Control of Hospital Infection. London: Chapman & Hall 1982; 80. Daschner F. Cost-effectiveness in hospital infection control-lessons for the 1990s. r Hasp Infect 1989; 13: 325-336.
