Are computer keyboards a cross-infection risk

in a dental clinic?
Sarika Patel, Kathryn Porter, Rachel L Sammons*

The School of Dentistry, College of Medical and Dental Sciences, University of Birmingham, St Chad’s Queensway,
Birmingham, B4 6NN, UK. Email: [Link]@[Link]
*Corresponding author

Accepted for publication: 26 August 2010

Key words: Computer keyboards, cross-infection, dentistry, bacterial survival, disinfection

Abstract
omputers in dental surgeries located within the record procedures. The equipment is often located in the patient treat-
A
C patient treatment room could potentially pose a
risk of cross-infection of patients and operators.
ment room as a ‘chair-side’ unit (Schleyer et al, 2006), where con-
tamination may occur both via contact with the operators’ hands and
The aims of this investigation were to determine the potentially from splashes and air-borne oral micro-organisms resulting
degree of contamination of computer keyboards located from dental procedures (Motta et al, 2005; Motta et al, 2007; Decraene
in clinics and study rooms in a teaching dental hospi- et al, 2008). In this short report we present the results of a pilot study
tal, to determine the survival time of micro-organisms on to investigate the degree of contamination of computer keyboards in
keyboards and to compare the efficiency of two common clinics and study areas in a dental teaching hospital, to determine how
disinfectants in disinfecting keyboards. All keyboards long common bacterial species could survive on the keys and whether
were found to be contaminated with a variety of micro- the weekly use of isopropanol wipes to disinfect keyboards in the
organisms including Staphylococcus aureus, coagulase- clinics is effective.
negative staphylococci, Gram-negative rods and cocci. A
range of Gram-positive and -negative bacteria and Can- Methods and materials
dida albicans were able to survive up to 10 days in dried Bacterial colonisation on computer keyboards: initial screening
suspensions on computer keys. Seventy per cent isopro- Eight computers in four different areas of a university teaching dental
panol wipes proved effective in reducing the numbers of hospital were used in the investigation: two separate study areas used
viable test organisms on keys by at least 96% and reduc- by undergraduate students (Dell, Berkshire, computers) and two
ing contamination of keyboards in routine use. The data patient clinics on separate floors, used by dental surgeons, dental
suggest that computers in dental study areas and clin- students and nursing staff (Viglen, Hertfordshire, computers). Two
ics act as reservoirs of potentially pathogenic organisms computers in each area were swabbed. Preliminary experiments estab-
including S. aureus and should be regularly disinfected lished that the method of collection that yielded most organisms was
to reduce the risk of cross-infection. the use of a dry swab moistened by immersion for 1 second in sterile
distilled water (detection limit approximately 10 organisms/key). The
Introduction keyboards were swabbed by running the tip of the swab from left to
A number of investigations have indicated that computer keyboards in right over the entire length covering the tops of all the keys and then
clinical settings such as intensive care units, wards and operating the- turning the swab and returning over the same surface, so that the
atres are contaminated with a variety of micro-organisms (Isaacs et al, swab swept the keyboard twice. One single operator carried out the
Peer reviewed article

1998; Neely et al, 1999; Neely et al, 2005; Bures et al, 2000; Coia and procedure to ensure consistency of regime and pressure. The key-
Masterton, 2001; Devine et al, 2001; Man et al, 2002; Schultz et al, 2003; boards were swabbed three times and swabs used to directly inoculate
Hartmann et al, 2004; Wilson et al, 2006; Wilson et al, 2008; Fukada blood agar plates (one plate per swab), which were incubated aerobi-
et al, 2008). The study by Wilson et al (2006) revealed that over one cally at 37°C for 48 hours. Because the study was an undergraduate
third of bedside computers in a London hospital intensive care unit were final year project with a limited time scale, plates were only incubated
contaminated with meticillin resistant Staphylococcus aureus (MRSA). aerobically. Figure 1 shows a typical blood agar plate with a variety of
Computers are increasingly used in dental clinics to access morphologically distinct colonies of bacteria and fungi collected from
patients’ medical and dental records, for treatment planning and to one computer. Morphologically similar colonies were assumed for

© The Author(s) 2010

Reprints and permissions:
[Link]
206 Journal of Infection Prevention Downloaded from [Link] at FLORIDA INTERNATIONAL UNIV on June 5, 2015
NOVEMBER
R 2010 VOL. 11 NO. 6 10.1177/1757177410384892
 practical purposes to be the same organism and representative colo- were determined by swabbing an approximate 5 mm2 area of each key
nies were Gram stained. Further standard tests were carried out to (area contacted by the tip of the swab). The keys were stored in Petri
identify potential pathogens such as Staphylococcus aureus, coliforms dishes in a closed (non-operating) Aura B3 biological safety cabinet
and pseudomonad-like organisms. for 12 days at room temperature, sampling by swabbing every two to
three days. Surviving organisms were cultured on blood agar or tryp-
Survival of organisms on computer keys tone soya agar for B. subtilis for 48 h. S. sanguinis was incubated
Escherichia coli NCTC 10418, Bacillus subtilis 168, Candida albicans anaerobically in a CO2-enriched atmosphere.
ATCC 10231, Pseudomonas aeruginosa, ATCC 27853, Staphylococ-
cus aureus NCTC 6571, Staphylococcus epidermidis 11047 and Strep- Disinfection
tococcus sanguinis RW2 were selected for testing as representatives The micro-organisms listed above were used to contaminate isolated
of organisms previously reported to have been isolated from keyboards keys with a dried suspension as described above. Each key was then
in dental clinics (Motta et al, 2005, Motta et al, 2007). Isolated (Viglen wiped four times with the same side of an isopropanol wipe or sterile
keyboard) keys were sterilised using ethylene oxide and contaminated absorbent paper (Tork, SCA, Goteborg, Sweden) cut to the same size
with approximately 104 colony forming units (cfu)/ml of each organ- soaked in Virkon (Antec International, Suffolk, UK). This is a surface
ism in 0.1 ml TSB (three keys per organism). The suspension was disinfectant containing dipotassium peroxodisulphate as the active
dried on the keys for approximately 1 hour at room temperature. Three ingredient. Non-wiped keys and keys that had been wiped with absor-
keys were inoculated per organism. Initial numbers of viable bacteria bent paper soaked in sterile distilled water were used as controls. After
wiping, each key was swabbed and organisms cultured as described
above.

Effectiveness of isopropanol disinfection in routine use

In this dental hospital the surfaces of keyboards in the clinics are
wiped over with isopropanol wipes approximately once a week. In this
experiment two computer keyboards in clinical and study areas were
disinfected twice a day (morning and mid-afternoon, before the start
of clinics) using isopropanol wipes. No other form of cleaning was
performed. After five days the numbers of organisms were determined
by swabbing as described above.

Contamination during routine use

To monitor the build up of contaminants on a keyboard, one used by
clinical students for accessing patient records and (as controls) iso-
lated individual keys removed from a spare Viglen keyboard placed on
the work surface adjacent to it, were first disinfected by wiping with
70% isopropanol wipes (Azo wipes; Vernon-Carus Ltd. Lancashire,
UK). The keyboard was then used as normal and swabbed, with oper-
ator compliance, once approximately every 30 minutes for a period of
150 minutes. During this period three to four different people used the
Figure 1. Typical appearance of a blood agar plate resulting from swabbing a computer computer for approximately two minutes at each visit. Removal of
keyboard, illustrating the numbers and variety of organisms isolated. Arrows indicate gloves and hand-washing before use is mandatory.
possible fungal colonies

Table 1. Micro-organisms isolated from computer keyboards. Numbers are colony forming units (cfu) or
percentages where indicated (shaded rows)

Type of organism Clinical areas Study areas

Gram-positive cocci (GPC) 1120 2700

% coagulase positive GPC (S. aureus) 53 42
% Catalase negative (Enterococci/Streptococci) 6 17
Gram-positive rods 650 650
Gram-negative cocci 630 780
Gram-negative rods (GNR) 440 670
% lactose non-fermenting, citrate positive, oxidase positive GNR (pseudomonads) 47 21
Peer reviewed article

% lactose fermenting, citrate negative, oxidase negative, urease negative GNR 0 48

(coliforms)
Gram-variable rods 10 32
Fungi 7 14
Total cfu (From four computers/area) 2580 4832
Average no. of cfu removed/keyboard. 646 1208

Downloaded from [Link] at FLORIDA INTERNATIONAL UNIV on June 5, 2015

VOL. 11 NO. 6 NOVEMBER 2010 Journal of Infection Prevention 207
 Results and the lower starting numbers (Day 0) of P. aeruginosa and S. san-
Initial screening guinis suggests that some of the organisms did not survive the drying
Initial screening of computers in study and clinical areas indicated that process and after a few days their numbers fell below the detection
they were all contaminated, each swab yielding several hundred bac- limit.
terial cfu and occasional fungi (Figure 1; Table 1). Higher numbers of
organisms were isolated from the keyboards in the study areas, pos- Disinfection
sibly because these keyboards were never cleaned whereas the ones in Distilled water was effective in reducing numbers of S. epidermidis but
the clinic were supposedly cleaned weekly with 70% isopropanol not S. aureus or P. aeruginosa. Virkon was more effective but isopro-
wipes (compliance was not checked). However as the computers in panol wipes removed all C. albicans, P. aeruginosa and S. sanguinis,
the different areas were of different makes direct comparison was 99.9% of S. epidermidis and 96% of all the other organisms tested
impossible. Coagulase negative staphylococci/micrococci were the (Figure 3). The same pattern of responses was reproduced on two
most frequently isolated organisms but potential pathogens including separate keyboards.
S. aureus and pseudomonad-like organisms were isolated from com-
puters in clinics and study areas. Coliforms were isolated from com- Effectiveness of isopropanol disinfection
puters in the study areas. The use of isopropanol wipes to disinfect the computers in the morn-
ing and afternoon for five working days significantly reduced the
Survival of organisms number of organisms recovered at the end of the week (paired t-test
A proportion of all the test organisms tested survived at least three for means, p < 0.001; n = 8). Organisms recovered from clinical and
days and viable S. aureus, C. albicans, B. subtilis and [Link] were study area computers were reduced by factors of 10 and approximately
recovered after 10 days (Figure 2). Initially the keys were contami- 100, respectively (Figure 4).
nated with approximately equal numbers of organisms (104 cfu/key)
Contamination during routine use
The results of the experiment to determine how quickly bacteria can
build up on computer keyboards during routine use are shown in
Figure 5. There was a significant increase in the number of cfu recov-
ered from the keyboard over time compared with the number recov-
ered from isolated keys on the bench beside it (Figure 5; paired t-test
for means; p < 0.001; n = 6).

Discussion
This investigation revealed that keyboards in study areas and dental
clinics were contaminated with bacteria and fungi. Consistent with
other studies (Bures et al, 2000; Devine et al, 2001; Schultz et al,
2003; Rutala et al, 2006; Wilson et al, 2006), staphylococci were the
most common isolates together with other organisms that could orig-
inate from the resident flora of the fingertips, which includes the
genera Staphylococcus, Corynebacterium, Proprionibacterium, Strep-
tococcus and Pseudomonas (Cogen et al, 2008). S. aureus was
Figure 2. Survival of organisms on keys from a computer keyboard. A suspension of each isolated from study area and clinic keyboards and organisms presumed
test organism initially containing approximately 105 cfu/ml in tryptone soya broth was
dried onto the keys (n = 3 keys/organism). The ‘time 0’ result indicates the number of
to be coliforms were also isolated from those in the study areas.
viable bacteria after drying. The graph shows the average numbers and standard Differences in numbers on the computers in the different areas may
deviation of bacteria recovered by swabbing an area of 5 mm2 on the keys after the relate to the difference in type of computer, number of operators
number of days indicated. Sampling was continued for 12 days but no viable organisms (which was not controlled), the degree of contamination on individu-
were recovered after 10 days
als’ hands, or cleaning frequency.
Peer reviewed article

Figure 3. Comparison of the effectiveness of disinfectants on computer keys. The graph shows numbers of viable bacteria (average +//− standard deviation; n = 3 keys) recovered following
disinfection by wiping with filter paper soaked in distilled water, a surface disinfectant (Virkon) and disinfectant wipes containing 70% isopropanol. Control keys were not wiped

208 Journal of Infection Prevention NOVEMBER

R 2010 VOL. 11 NO. 6
Downloaded from [Link] at FLORIDA INTERNATIONAL UNIV on June 5, 2015
 Start of the week End of the week

1.0E+03

Total number of bacterial isolated (cfu/ml)

1.0E+02

1.0E+01

1.0E+00
Clinical area Study area
Location of computers
Figure 4. Comparison of bacterial growth before and after the use of isopropanol wipes for one week. The graph shows the average number and standard deviation of organisms recovered
from the keyboard at the start of the week and after the use of isopropanol wipes for a week (average number +/− standard deviation; n = 3). Each pair of columns represents a keyboard;
the first 4 are within two clinical areas and the last 4 within two study areas

50
Total number of bacterial isolated (cfu/ml)

45 Isolated key

40 Keyboard

35

30

25

20

15

10

0
0 30 60 90 120 150
Time (minutes)
Figure 5. Numbers of bacteria recovered from a keyboard during routine use in a patient clinic and from an isolated untouched key in the same environment. Swabs were inoculated into
tryptone soya broth and 0.1 ml of the resulting suspension was plated on 3 separate plates to determine the average cfu/ml +//− standard deviation

Transfer by direct skin contact is probably the major route to keyboard Streptococci may not have grown well under these conditions and may
contamination but in a dental clinic this could also occur via splashes have been inhibited by the distilled water in the initial swabbing. As
and air-borne micro-organisms released from the oral cavity during treat- many oral bacteria have complex culture requirements, in future studies
ment (Bennett et al, 2000; Harrel and Molinari, 2004; Rautemaa et al, a combination of aerobic and anaerobic culture techniques, and partial
2006). Many dental procedures produce air-borne droplets (reviewed by 16S ribosomal RNA gene sequencing (as used by Decraene et al, 2008),
Harrel and Molinari, 2004), which may also contain blood (Harrel and or PCR could provide a more comprehensive indication of the range of
Molinari, 2004; Cristina et al, 2008). Any surfaces on which these settle organisms present.
could become reservoirs of antibiotic-resistant micro-organisms and This study indicated that contamination quickly builds up on key-
viruses (Decraene et al, 2008). During procedures such as mechanical boards and the rate and variety of organisms will be dependent on
Peer reviewed article

scaling (Bennet et al, 2000) and the use of high-speed rotating instru- numbers of users (Anderson and Palombo, 2009). It also demon-
ments, aerosols may travel long distances: Rautemaa et al (2006) identi- strated that a range of organisms were able to survive on keys for sev-
fied bacteria including the viridans-group of streptococci, which indicate eral days and the results are consistent with those of Neely and Maley
contamination with oral flora, on agar plates placed up to 2 m from the (2000), which demonstrated long-term survival of staphylococci on
dental chair following such procedures. The culture conditions in this hospital fabrics; Smith et al (1996), which showed survival of
pilot study limited it to the isolation of organisms that could grow in air Pseudomonas aeruginosa and staphylococci in dried coagulum on sur-
on blood agar and hence the numbers of organisms recovered (and faces for several months; and Rangel-Frausto et al (1994), which
types) are undoubtedly underestimates of the actual number present. reported C. albicans survival on inanimate surfaces for up to 10 days.

VOL. 11 NO. 6 NOVEMBER 2010

Downloaded from [Link] at FLORIDA INTERNATIONAL UNIV on June 5, 2015
Journal of Infection Prevention 209
 Isopropanol wipes, as already used in the clinics for disinfection of cleaning and disinfection. Hands should be washed before and after
equipment surfaces, are convenient and easy to use and proved to be keyboard use.
effective in removing at least 96% of the test organisms from keys. The Factors affecting contamination and survival on keyboards include
results of the short clinical assessment suggest that their daily use indentations, surface roughness or other features that facilitate soil
reduced the number of organisms present, although this requires further retention and the composition of the key surface including any addi-
investigation. Rutala et al (2006) recommended daily disinfection with tives that may influence bacterial adhesion. Soil containing micro-or-
a quaternary ammonium disinfectant. The difference in the relative effi- ganisms may accumulate around and beneath the keys. The use of
ciency of isopropanol and Virkon could have been due to the difference washable keyboards, disposable fitted keyboard covers (Neely et al,
in carrier (absorbent paper versus wipe) and this should be addressed in 2005) or even plastic food wrap to protect the keyboard and mouse
any future comparative studies. Hassoun et al (2004) found that isopro- have been suggested USAF Dental evaluation and consultation service
panol swabs were effective in reducing contamination of personal dig- Fact sheet No. 26, 2007). However disposable covers deteriorate and
ital assistants (‘palm-top’ computers) used by healthcare workers. are not easy to clean, whilst the reflective surface of food wrap makes
Several studies have provided evidence to suggest a link between the keys difficult to see. Keyboards that are easy to clean are recom-
computer use in hospitals and cross-contamination of patients (for mended: the results of a study by Wilson et al (2008) demonstrated
example, Neely et al, 1999; Bures et al, 2000; Devine et al, 2001; that smooth, silicone-coated keyboards proved more difficult to con-
Lu et al, 2009). It is difficult to know whether the level of contamina- taminate and were easier to clean than conventional ones and those
tion presents a serious risk of cross-infection in a dental hospital fitted with alarms that indicated when cleaning was required improved
where the patients are mostly outpatients and are not usually immu- cleaning compliance in an intensive care unit.
nocompromised. However, students and staff should be more aware
of the possibility of transfer of micro-organisms between computers in Conflict of interests
study and clinical areas and adhere to rules concerning hand hygiene, None declared.

References Isaacs D, Daley A, Dalton D, Hardiman R, Nallusamy R. (1998)

Anderson G, Palombo EA. (2009) Microbial contamination of computer Swabbing computers in search of nosocomial bacteria. Journal of
keyboards in a university setting. American Journal of Infection Paediatric Infectious Diseases 17(6): 533.
Control 37(6): 507–09. Lu PL, Siu LK, Chen TC, Ma L, Chiang WG, Chen YH, Lin SF, Chen TP.
Bennett AM, Fulford MR, Walker JT, Bradshaw DJ, Martin MV, Marsh (2009) Methicillin-resistant Staphylococcus aureus and Acinetobacter
PD. (2000) Microbial aerosols in general dental practice. British Dental baumannii on computer interface surfaces of hospital wards and asso-
Journal 189(12): 664–67. ciation with clinical isolates. BMC Infectious Diseases 9: 164–71.
Bures S, Fishbain JT, Uyehara CFT, Parker JM, Berg BW. (2000) Man GS, Olapoju M, Chadwick MV, Vuddamalay P, Hall AV, Edwards
Computer keyboards and faucet handles as reservoirs of nosocomial A, Kerr JR. (2002) Bacterial contamination of ward-based computer
pathogens in the intensive care unit. American Journal of Infection terminals. Journal of Hospital Infections 52: 314–18.
Control 28(6): 465–71. Motta RHL, Ramacciato JC, Groppo FC, Pacheco A, De Mattos TR.
Cogen AL, Nizet V, Gallo RL. (2008) Skin microbiota: a source of dis- (2005) Environmental contamination before, during, and after dental
ease or defence? British Journal of Dermatology 158: 442–55. treatment. American Journal of Dentistry 18(5): 340–4.
Coia JE, Masterton RG. (2001) Computer keyboards as a risk for noso- Motta RHL, Groppo FC, Bergamaschi CDC, Ramacciato JC, Pharm SB,
comial infection. American Journal of Infection Control 29(5): 345. de Mattos-Filho TR. (2007) Isolation and antimicrobial resistance of
Cristina ML, Spagnolo AM, Sartini M, Dallera M, Ottria G, Lombardi R, Staphylococcus aureus isolates in a dental clinic environment.
Perdelli, F. (2008) Evaluation of the risk of infection through exposure Infection Control and Hospital Epidemiology 28(2): 185–90.
to aerosols and spatters in dentistry. American Journal of Infection Neely AN, Maley MP, Warden GD. (1999) Computer keyboards as res-
Control 36(4): 304–07. ervoirs for Acinetobacter baumannii in a burn hospital. Clinical
Decraene V, Ready D, Pratten J, Wilson M. (2008) Air-borne microbial Infectious Diseases 29: 1358–60.
contamination of surfaces in a UK dental clinic. Journal of General Neely AN, Maley MP. (2000) Survival of enterococci and staphylococci
and Applied Microbiology 54(4): 195–203. on hospital fabrics and plastic. Journal of Clinical Microbiology 38:
Devine J, Cooke RPD, Wright EP. (2001) Is methicillin-resistant 724–6.
Staphylococcus aureus (MRSA) contamination of ward-based compu- Neely AN, Weber JM, Daviau P, MacGregor A, Miranda C, Nell M,
ter terminals a surrogate marker for nosocomial MRSA transmission and Bush, P Lighter, D. (2005) Computer equipment used in patient care
handwashing compliance? Journal of Hospital Infection 48: 72–5. within a multihospital system: recommendations for cleaning and
Fukada T, Iwakiri H, Ozaki M. (2008) Anaesthetists’ role in computer disinfection. American Journal of Infection Control 33(4): 233–7.
keyboard contamination in an operating room. Journal of Hospital Rangel-Frausto MS, Houston AK, Bale MJ, Fu C, Wenzel RP. (1994) An
Infection 70: 148–53. experimental model for study of Candida survival and transmission in
Harrel SK, Molinari J. (2004) Aerosols and splatter in dentistry – a brief human volunteers. European Journal of Clinical Microbiology and
review of the literature and infection control implications. Journal of Infectious Diseases 13: 590–5.
the American Dental Association 135(4): 429–37. Rautemaa R, Nordberg A, Wuolijoki-Saaristoe K, Meurman JH. (2006)
Peer reviewed article

Hartmann B, Benson M, Junger A, Quinzio L, Röhrig R, Fengler B, Bacterial aerosols in dental practice – a potential hospital infection
Färber UW, Wille B, Hempelmann G. (2004) Computer keyboard and problem? Journal of Hospital Infection 64(1): 76–81.
mouse as a reservoir of pathogens in an intensive care unit. Journal of Rutala WA, White MS, Gergen MF, Weber DJ. (2006) Bacterial con-
Clinical Monitoring and Computingg 18(1): 7–12. tamination of keyboards: efficacy and functional impact of disinfect-
Hassoun A, Vellozzi EM, Smith MA. (2004) Colonization of personal ants. Infection Control and Hospital Epidemiology 27(4): 372–7.
digital assistants carried by health care professionals. Infection Control Schleyer KL, Thyvalikakath TP, Spallek H, Torres-Urquidy MH,
and Hospital Epidemiology 25:1000–01. Hernandez P, Yuhaniak J. (2006) Clinical computing in general

210 Journal of Infection Prevention NOVEMBER

R 2010 VOL. 11 NO. 6
Downloaded from [Link] at FLORIDA INTERNATIONAL UNIV on June 5, 2015
 dentistry. Journal of the American Medical Informatics Association Available at [Link]
13: 344–52. documents/afms/ctb_109779.pdf (accessed 21 June 2010).
Schultz M, Gill J, Zubairi S, Huber R, Gordin F. (2003) Bacterial con- Wilson APR, Hayman S, Folan P, Ostro PT, Birkett A, Batson S, Singer
Peer reviewed article

tamination of computer keyboards in a teaching hospital. Infection M, Bellingan G. (2006) Computer keyboards and the spread of MRSA.
Control and Hospital Epidemiology 24(4): 302–3. Journal of Hospital Infection 62(3): 390–2.
Smith SM, Eng RHK, Padberg FT. (1996) Survival of nosocomial Wilson APR, Ostro P, Magnussen M, Cooper B for the Keyboard Study
pathogenic bacteria at ambient temperature. Journal of Medicine 27: Group (2008) Laboratory and in-use assessment of methicillin-resist-
293–302. ant Staphylococcus aureus contamination of ergonomic computer
USAF Dental evaluation and consultation service Fact Sheet No. 26 keyboards for ward use. American Journal of Infection Control 36:
(May 2007) Computer Equipment in the Dental Operatory. e19–e25.

VOL. 11 NO. 6 NOVEMBER 2010

Downloaded from [Link] at FLORIDA INTERNATIONAL UNIV on June 5, 2015
Journal of Infection Prevention 211