Journal of Hospital Infection
The effect
19, 175-l
80
of surgical theatre head-gear bacterial counts
H. Humphreys*,
Department
(1991)
A. J. Russell, R. J. Marshall, D. S. Reeves
on air
V. E. Ricketts
of Medical Microbiology and Infection Control, Hospital, Westbury-onTrym, Bristol BSlO 5NB Accepted
for publication
27 August
and
Southmead
1991
Summary:
The wearing of disposable head-gear in operating theatres is currently recommended for scrubbed and non-scrubbed staff. However, there is little evidence of its effectiveness as an infection control measure in casual or non-scrubbed theatre staff. The effect of head-gear on bacterial air counts was studied, using six volunteers, in a sealed room, with and without ventilation. Using a Casella slit sampler and a SAS Air Sampler, air counts ranged from 0.08 to greater than 2.9 colony forming units (cfu) m-r. The wearing of head-gear was not associated with a reduction in air counts but counts were lower with ventilation. We recommend that non-scrubbed staff no longer wear head-gear as effective ventilation probably counteracts any possible increased bacterial shedding. Scrubbed staff should continue to wear disposable head-gear because of their proximity to the operative field. Keywords: microbiology.
Operating
theatre;
wound
infection;
nosocomial
infection;
air
Introduction The wearing of disposable head-gear or hats by all staff is a routine procedure in most operating theatres because it is considered effective in reducing bacterial air counts and, by implication, postoperative infection. Human hair has long been recognized as a potential reservoir of Staphylococcus aureus in hospital patients and staff, especially where skin or scalp diseases such as eczema and psoriasis are present.“2 Dineen & Drusin3 have described two outbreaks of postoperative wound infection traced to staff carrying S. aweus in human hair. Twelve infections were associated with a doctor working on one ward and an additional five with a nurse complaining of sores on her scalp. Consequently, most official recommendations regarding infection control measures for theatre staff suggest that hair be kept covered at all times.4’5
* Correspondence Medical Centre, 0195~6701/91/110175+06
to: Dr Nottingham
Humphreys, NG7
Department
of
Microbiology,
University
Hospital,
Queen’s
2UH.
$03.00/O
c
175
1991 The Hospital
Infectron
Socaety
176
H. Humphreys
et al.
However, there is little convincing evidence to confirm the effectiveness of such head-gear. 6,7 The results from one study carried out in a modern hospital operating room suggest that head cover, of whatever type, but excluding helmet exhaust apparatus employed in orthopaedic theatres, does not affect environmental air counts.8 It is difficult to prove that the wearing of head-gear by staff not directly involved in the operation affects postoperative infection because of its relative infrequency and multi-factorial aetiology. Therefore, the effect on bacterial air counts is probably the best indirect measurement of the efficacy of this practice in reducing infection rates. This study was initiated to examine the effect of one type of conventional head-gear on bacterial air counts in a sealed room with and without ventilation. Materials
and methods
Experimental design The study was carried out in a room, total volume 57.82 m3, previously used to isolate infectious patients. The room is ventilated with 18-20 air changes per hour and all internal and external windows were sealed for the duration of the study (Figure 1). The ventilation is by means of a Vokes Unipac system: input air is heated, humidified and conducted through coarse and fine filters. Air is expelled through biofilters to the outside and remains completely separate at all times from input air. Doors and all other interfaces were also sealed with masking tape as was the entrance/exit door at the start of each study or control period. The walls, windows, floors, all other surfaces and the ceiling were cleaned with Hospec BP (5 ml in 5 1 of water) on the day prior to the start of the study. In addition, the floor and all horizontal surfaces were cleaned at the end of each period with Hospec BP. Non-removable items, such as electric lights, and wash-hand basin, were covered and sealed. The six volunteers who participated in the study were healthy adults without any skin or hair conditions. Each subject was dressed in conventional operating theatre dress, i.e. theatre clothes (Cromaco, 50% polyester and 50% cotton), with the trouser legs taped to minimize perineal shedding, disposable gloves (LRC products) and masks (Vanguard Surgical, Universal Hospital Supplies Ltd). The head-gear worn was Barrier Surgeons Hoods (Surgikos) which cover the head, side of the face and chin. These hats are made of 73% rayon and 27% binder (ethylene, vinyl, acetate, copolymer) and contain small perforations which ensure they are comfortable to wear while at the same time preventing hair from extruding to the outside. Bacterial air counts were sampled during four test periods on different days: with ventilation and head-gear, with ventilation but no head-gear, without ventilation but with head-gear and finally without either ventilation or head-gear. Between each period the room ventilation remained on. Sampling was also carried out during two control periods, i.e. with and
Head-gear
and bacterial 5.36
Door c---------~
air counts
m
entrance/exit I
I
------m---q
Corridor
Figure 1. Plan of ventilated external sealed window; 0 position of air sampler.
room with position ‘--: I internal sealed
of air samplers. A window; @ position
Internal of SAS
sealed door; air sampler;
without ventilation when one volunteer sampled air with the minimum of activity. Throughout each test period three volunteers walked at a normal pace for 15 min while three remained stationary but touched the opposite shoulder with the hands coming down to hang by the side after each exercise, to the beat of a metronome (40 per min). The roles were reversed for a second period of 15 min and conversation was kept to a minimum throughout. Air sampling and microbiology During the last five mins of each 30 min period, 165 m3 (33 m3 min-‘) of air was sampled using a Casella slit sampler with a blood agar plate 90 mm in diameter. Half way through each session, 450 m3 of air was sampled using a SAS Air Sampler (SC099) with a contact plate 55 mm in diameter. During two study and both control periods there was additional sampling (240 m3) with the SAS Air Sampler during the final ten min. Contact plates contained trypticase soya agar(TSA) with 0.2% Tween 80 and both these and the blood agar plates were incubated in air for 48 hours. Total colony counts and individual counts for S. aweus were recorded and expressed as colony forming units per m3 (cfu mP3). Corrections to the
H. Humphreys
178
et al.
observed colony counts from the SAS Air Sampler results were made as recommended by the suppliers using correction charts. These are based upon standard statistical formulae for estimating the probability of an event occurring concurrently, i.e. for co-incidence of two or more cfus passing through the same hole which occurs except at very low air counts. Hair counts from ail six volunteers were carried out after the fourth test period by passing a sterile D6 scalp massage brush, three inches in diameter,’ through the hair six times, subsequently dipping the brush in quarter strength Ringer’s solution with 0.2% Tween 80 and inoculating in duplicate with a spiral plater on TSA. Plates were incubated for 48 hours. Total colony- and S. aureus counts were recorded. Results were expressed as cfu per litre (cfu 1-l). Results
Bacterial colony counts ranged from 0.14 to 1.55 cfumm3 with the slit sampler and from 0.08 to greater than 2.9 cfume3 using the SAS Air Sampler (Table I). Staphylococcus aureus was not isolated during any of the study or control periods. The wearing of head-gear did not appreciably reduce air counts but counts in general were higher in the absence of ventilation. Air counts using the SAS Air Sampler for 450 m3 and 240 m3 were 0.12 cfu me3 and 0.08 cfu me3 with ventilation and 1.03 cfu mm3 and O-53 cfu mW3 without ventilation respectively, during the control period. Individual hair colony counts for the six volunteers ranged from zero to 15 with a mean of 4 per volunteer. Staphylococcus aureus was not recovered from the hair of any of the volunteers. Table
I.
Bacterial
air counts (cfu m-“) during study and control periods Slit
sampler 165 m3*
SAS Air 450 m3*
Sampler 240 m3*
Head-gear With ventilation Without ventilation
0.53 1.55
>2.9 2.01
1.59
0.66 0.35
0.43 0.13
0.16
0.25 0.14
0.12 1.03
0.08 0.53
No head-gear With ventilation Without ventilation
Control With ventilation Without ventilation * Total
volume
of air sampled.
Discussion
Postoperative infection rates are governed by many factors including good surgical technique and sterile procedures throughout the operation. Many
Head-gear
and bacterial
air counts
179
measures which have never been scientifically validated to minimize postoperative sepsis have been adopted over the years. Some were introduced before the advent of well-ventilated operating theatres when no distinction was made between measures adopted by those directly involved in the surgical procedure and other theatre staff. As a result, such measures have been incorporated into theatre ritual and some indeed may contribute to a false sense of complacency amongst theatre staff. At present there are no universally agreed standards for operating theatre bacterial air counts. Air counts during a lo-year review of commissioning theatres in one theatre suite ranged from 26 to 145 bacteria-carrying particles per cubic metre of air. lo Thirty-five bacterial particles per m3 or less and the absence of Clostridium perfringens or S. aureus is an acceptable standard but it is widely acknowledged that air counts vary with the number of people, activity, sex, type of clothing and method of air sampling.” In this study, two methods of air sampling were used and considerable volumes of air were sampled. The slit sampler has traditionally been employed to measure air counts but the portable SAS Air Sampler used here is capable of sampling larger volumes and has an efficiency close to 100% of that of the slit sampler. l2 We recorded considerable variation in bacterial air counts both between different sampling periods and between the two methods and do not believe that additional sampling would have affected our conclusions. The conditions simulated, to a large extent, those pertaining to a non-orthopaedic theatre with c. 18-20 air changes per hour. Head-gear did not lead to a reduction in air counts and reduced counts, when they occurred, were related to the use of ventilation. Notwithstanding the limitations of this study, i.e. variation in air counts between methods, the nature of the experimental room and simulated theatre activity, we feel that the wearing of head-gear by casual theatre staff does not significantly contribute to infection control practice. This applies especially to those wandering into and out of the anaesthetic room. However, at this stage we would not extend this recommendation to orthopaedic theatres where prosthetic joint surgery takes place, because of the catastrophic consequences of joint sepsis caused by skin organisms. In conventional theatres, effective ventilation probably counteracts any possible increased bacterial shedding from the absence of head-gear in staff not directly involved in the operative procedure but those assisting should continue to adopt this practice because of their close proximity to the operative field. There is no substitute for good surgical technique and high standards of personal hygiene which include regular showers or baths, and keeping long hair tidy. These recommendations are practical, inexpensive and preferable to misleading theatre staff about the effectiveness of head-gear in reducing infection. Perceval is correct in his assertion that the virulence and not just the number of microorganisms in the unit is important.13 However, this study demonstrates that the wearing of head-gear by casual theatre staff is unlikely
180
H. Humphreys
et al.
to affect either. Moreover, in this cost-conscious era, it seems reasonable to advise discontinuation of a practice which does not add significantly to infection control and instead, concentrate resources on those procedures of proven benefit. Consequently, we recommend discontinuation of this practice amongst casual or non-scrubbed theatre staff. References I. Summers MM, Lynch PF, Black T. Hair as a reservoir of staphylococci. J CEin Puthol 1965; 18: 13-15. Noble WC. Staphylococcus aureus on the hair. J Clin Path01 1966; 19: 57&572. i: Dineen P, Drusin L. Epidemics of postoperative wound infections associated with carriers. Lancet 1973; 2: 1157-1159. 4. A report to the Medical Research Council by the sub-committee on aseptic methods in operating theatres of their Committee on Hospital Infection. Lancer 1968; 1: 705-709. 5. CDC Guidelines for the prevention and control of nosocomial infections. Guidelines for prevention of surgical wound infections, 1985. Am J Infect Control 1986; 14: 71-82. 6. Hambraeus A, Laurel1 G. Protection of the patient in the operating suite. J Hasp Infection 1980; 1: 15-30. 7. Lehr PS, Palmer PN. Operating room practices: myth or science. Aorn J 1989; 49: 645-649. 8. Ritter MA, Eitzen HE, Hart JB, French MLV. The surgeon’s garb. Clin Orthop 1980; 153: 204-209. 9. Black WA, Bannerman CM, Black DA. Carriage of potentially pathogenic bacteria in the hair. Br J Surg 1974; 61: 735-738. 10. Holton J, Ridgway GL, Reynoldson AJ. A microbiologist’s view of commissioning theatres. J Hasp Infection 1990; 16: 29-34. 11. Air sampling in operating theatres (Editorial). J Hasp Infection 1984; 5: l-2. 12. Lath V. Performance of the surface air system samplers. J Hasp Infection 1985; 6: 102-107. 13. Perceval A. Why isolate theatre suites? Lancet 1991; 1: 337.
The effect
19, 175-l
80
of surgical theatre head-gear bacterial counts
H. Humphreys*,
Department
(1991)
A. J. Russell, R. J. Marshall, D. S. Reeves
on air
V. E. Ricketts
of Medical Microbiology and Infection Control, Hospital, Westbury-onTrym, Bristol BSlO 5NB Accepted
for publication
27 August
and
Southmead
1991
Summary:
The wearing of disposable head-gear in operating theatres is currently recommended for scrubbed and non-scrubbed staff. However, there is little evidence of its effectiveness as an infection control measure in casual or non-scrubbed theatre staff. The effect of head-gear on bacterial air counts was studied, using six volunteers, in a sealed room, with and without ventilation. Using a Casella slit sampler and a SAS Air Sampler, air counts ranged from 0.08 to greater than 2.9 colony forming units (cfu) m-r. The wearing of head-gear was not associated with a reduction in air counts but counts were lower with ventilation. We recommend that non-scrubbed staff no longer wear head-gear as effective ventilation probably counteracts any possible increased bacterial shedding. Scrubbed staff should continue to wear disposable head-gear because of their proximity to the operative field. Keywords: microbiology.
Operating
theatre;
wound
infection;
nosocomial
infection;
air
Introduction The wearing of disposable head-gear or hats by all staff is a routine procedure in most operating theatres because it is considered effective in reducing bacterial air counts and, by implication, postoperative infection. Human hair has long been recognized as a potential reservoir of Staphylococcus aureus in hospital patients and staff, especially where skin or scalp diseases such as eczema and psoriasis are present.“2 Dineen & Drusin3 have described two outbreaks of postoperative wound infection traced to staff carrying S. aweus in human hair. Twelve infections were associated with a doctor working on one ward and an additional five with a nurse complaining of sores on her scalp. Consequently, most official recommendations regarding infection control measures for theatre staff suggest that hair be kept covered at all times.4’5
* Correspondence Medical Centre, 0195~6701/91/110175+06
to: Dr Nottingham
Humphreys, NG7
Department
of
Microbiology,
University
Hospital,
Queen’s
2UH.
$03.00/O
c
175
1991 The Hospital
Infectron
Socaety
176
H. Humphreys
et al.
However, there is little convincing evidence to confirm the effectiveness of such head-gear. 6,7 The results from one study carried out in a modern hospital operating room suggest that head cover, of whatever type, but excluding helmet exhaust apparatus employed in orthopaedic theatres, does not affect environmental air counts.8 It is difficult to prove that the wearing of head-gear by staff not directly involved in the operation affects postoperative infection because of its relative infrequency and multi-factorial aetiology. Therefore, the effect on bacterial air counts is probably the best indirect measurement of the efficacy of this practice in reducing infection rates. This study was initiated to examine the effect of one type of conventional head-gear on bacterial air counts in a sealed room with and without ventilation. Materials
and methods
Experimental design The study was carried out in a room, total volume 57.82 m3, previously used to isolate infectious patients. The room is ventilated with 18-20 air changes per hour and all internal and external windows were sealed for the duration of the study (Figure 1). The ventilation is by means of a Vokes Unipac system: input air is heated, humidified and conducted through coarse and fine filters. Air is expelled through biofilters to the outside and remains completely separate at all times from input air. Doors and all other interfaces were also sealed with masking tape as was the entrance/exit door at the start of each study or control period. The walls, windows, floors, all other surfaces and the ceiling were cleaned with Hospec BP (5 ml in 5 1 of water) on the day prior to the start of the study. In addition, the floor and all horizontal surfaces were cleaned at the end of each period with Hospec BP. Non-removable items, such as electric lights, and wash-hand basin, were covered and sealed. The six volunteers who participated in the study were healthy adults without any skin or hair conditions. Each subject was dressed in conventional operating theatre dress, i.e. theatre clothes (Cromaco, 50% polyester and 50% cotton), with the trouser legs taped to minimize perineal shedding, disposable gloves (LRC products) and masks (Vanguard Surgical, Universal Hospital Supplies Ltd). The head-gear worn was Barrier Surgeons Hoods (Surgikos) which cover the head, side of the face and chin. These hats are made of 73% rayon and 27% binder (ethylene, vinyl, acetate, copolymer) and contain small perforations which ensure they are comfortable to wear while at the same time preventing hair from extruding to the outside. Bacterial air counts were sampled during four test periods on different days: with ventilation and head-gear, with ventilation but no head-gear, without ventilation but with head-gear and finally without either ventilation or head-gear. Between each period the room ventilation remained on. Sampling was also carried out during two control periods, i.e. with and
Head-gear
and bacterial 5.36
Door c---------~
air counts
m
entrance/exit I
I
------m---q
Corridor
Figure 1. Plan of ventilated external sealed window; 0 position of air sampler.
room with position ‘--: I internal sealed
of air samplers. A window; @ position
Internal of SAS
sealed door; air sampler;
without ventilation when one volunteer sampled air with the minimum of activity. Throughout each test period three volunteers walked at a normal pace for 15 min while three remained stationary but touched the opposite shoulder with the hands coming down to hang by the side after each exercise, to the beat of a metronome (40 per min). The roles were reversed for a second period of 15 min and conversation was kept to a minimum throughout. Air sampling and microbiology During the last five mins of each 30 min period, 165 m3 (33 m3 min-‘) of air was sampled using a Casella slit sampler with a blood agar plate 90 mm in diameter. Half way through each session, 450 m3 of air was sampled using a SAS Air Sampler (SC099) with a contact plate 55 mm in diameter. During two study and both control periods there was additional sampling (240 m3) with the SAS Air Sampler during the final ten min. Contact plates contained trypticase soya agar(TSA) with 0.2% Tween 80 and both these and the blood agar plates were incubated in air for 48 hours. Total colony counts and individual counts for S. aweus were recorded and expressed as colony forming units per m3 (cfu mP3). Corrections to the
H. Humphreys
178
et al.
observed colony counts from the SAS Air Sampler results were made as recommended by the suppliers using correction charts. These are based upon standard statistical formulae for estimating the probability of an event occurring concurrently, i.e. for co-incidence of two or more cfus passing through the same hole which occurs except at very low air counts. Hair counts from ail six volunteers were carried out after the fourth test period by passing a sterile D6 scalp massage brush, three inches in diameter,’ through the hair six times, subsequently dipping the brush in quarter strength Ringer’s solution with 0.2% Tween 80 and inoculating in duplicate with a spiral plater on TSA. Plates were incubated for 48 hours. Total colony- and S. aureus counts were recorded. Results were expressed as cfu per litre (cfu 1-l). Results
Bacterial colony counts ranged from 0.14 to 1.55 cfumm3 with the slit sampler and from 0.08 to greater than 2.9 cfume3 using the SAS Air Sampler (Table I). Staphylococcus aureus was not isolated during any of the study or control periods. The wearing of head-gear did not appreciably reduce air counts but counts in general were higher in the absence of ventilation. Air counts using the SAS Air Sampler for 450 m3 and 240 m3 were 0.12 cfu me3 and 0.08 cfu me3 with ventilation and 1.03 cfu mm3 and O-53 cfu mW3 without ventilation respectively, during the control period. Individual hair colony counts for the six volunteers ranged from zero to 15 with a mean of 4 per volunteer. Staphylococcus aureus was not recovered from the hair of any of the volunteers. Table
I.
Bacterial
air counts (cfu m-“) during study and control periods Slit
sampler 165 m3*
SAS Air 450 m3*
Sampler 240 m3*
Head-gear With ventilation Without ventilation
0.53 1.55
>2.9 2.01
1.59
0.66 0.35
0.43 0.13
0.16
0.25 0.14
0.12 1.03
0.08 0.53
No head-gear With ventilation Without ventilation
Control With ventilation Without ventilation * Total
volume
of air sampled.
Discussion
Postoperative infection rates are governed by many factors including good surgical technique and sterile procedures throughout the operation. Many
Head-gear
and bacterial
air counts
179
measures which have never been scientifically validated to minimize postoperative sepsis have been adopted over the years. Some were introduced before the advent of well-ventilated operating theatres when no distinction was made between measures adopted by those directly involved in the surgical procedure and other theatre staff. As a result, such measures have been incorporated into theatre ritual and some indeed may contribute to a false sense of complacency amongst theatre staff. At present there are no universally agreed standards for operating theatre bacterial air counts. Air counts during a lo-year review of commissioning theatres in one theatre suite ranged from 26 to 145 bacteria-carrying particles per cubic metre of air. lo Thirty-five bacterial particles per m3 or less and the absence of Clostridium perfringens or S. aureus is an acceptable standard but it is widely acknowledged that air counts vary with the number of people, activity, sex, type of clothing and method of air sampling.” In this study, two methods of air sampling were used and considerable volumes of air were sampled. The slit sampler has traditionally been employed to measure air counts but the portable SAS Air Sampler used here is capable of sampling larger volumes and has an efficiency close to 100% of that of the slit sampler. l2 We recorded considerable variation in bacterial air counts both between different sampling periods and between the two methods and do not believe that additional sampling would have affected our conclusions. The conditions simulated, to a large extent, those pertaining to a non-orthopaedic theatre with c. 18-20 air changes per hour. Head-gear did not lead to a reduction in air counts and reduced counts, when they occurred, were related to the use of ventilation. Notwithstanding the limitations of this study, i.e. variation in air counts between methods, the nature of the experimental room and simulated theatre activity, we feel that the wearing of head-gear by casual theatre staff does not significantly contribute to infection control practice. This applies especially to those wandering into and out of the anaesthetic room. However, at this stage we would not extend this recommendation to orthopaedic theatres where prosthetic joint surgery takes place, because of the catastrophic consequences of joint sepsis caused by skin organisms. In conventional theatres, effective ventilation probably counteracts any possible increased bacterial shedding from the absence of head-gear in staff not directly involved in the operative procedure but those assisting should continue to adopt this practice because of their close proximity to the operative field. There is no substitute for good surgical technique and high standards of personal hygiene which include regular showers or baths, and keeping long hair tidy. These recommendations are practical, inexpensive and preferable to misleading theatre staff about the effectiveness of head-gear in reducing infection. Perceval is correct in his assertion that the virulence and not just the number of microorganisms in the unit is important.13 However, this study demonstrates that the wearing of head-gear by casual theatre staff is unlikely
180
H. Humphreys
et al.
to affect either. Moreover, in this cost-conscious era, it seems reasonable to advise discontinuation of a practice which does not add significantly to infection control and instead, concentrate resources on those procedures of proven benefit. Consequently, we recommend discontinuation of this practice amongst casual or non-scrubbed theatre staff. References I. Summers MM, Lynch PF, Black T. Hair as a reservoir of staphylococci. J CEin Puthol 1965; 18: 13-15. Noble WC. Staphylococcus aureus on the hair. J Clin Path01 1966; 19: 57&572. i: Dineen P, Drusin L. Epidemics of postoperative wound infections associated with carriers. Lancet 1973; 2: 1157-1159. 4. A report to the Medical Research Council by the sub-committee on aseptic methods in operating theatres of their Committee on Hospital Infection. Lancer 1968; 1: 705-709. 5. CDC Guidelines for the prevention and control of nosocomial infections. Guidelines for prevention of surgical wound infections, 1985. Am J Infect Control 1986; 14: 71-82. 6. Hambraeus A, Laurel1 G. Protection of the patient in the operating suite. J Hasp Infection 1980; 1: 15-30. 7. Lehr PS, Palmer PN. Operating room practices: myth or science. Aorn J 1989; 49: 645-649. 8. Ritter MA, Eitzen HE, Hart JB, French MLV. The surgeon’s garb. Clin Orthop 1980; 153: 204-209. 9. Black WA, Bannerman CM, Black DA. Carriage of potentially pathogenic bacteria in the hair. Br J Surg 1974; 61: 735-738. 10. Holton J, Ridgway GL, Reynoldson AJ. A microbiologist’s view of commissioning theatres. J Hasp Infection 1990; 16: 29-34. 11. Air sampling in operating theatres (Editorial). J Hasp Infection 1984; 5: l-2. 12. Lath V. Performance of the surface air system samplers. J Hasp Infection 1985; 6: 102-107. 13. Perceval A. Why isolate theatre suites? Lancet 1991; 1: 337.
