- Email: [email protected]
National survey of MRSA: Ireland, 1995
Journal
of Hospital
Infection
National
(1997)
35,
survey
175-184
of MRSA:
Ireland,
1995
Z. Johnson, P. Fitzpatrick, C. Hayes, G. Sayers, H. Pelly, B. McDonnell, L. Thornton and J. Buttimer*
Health
Information
Received
Unit, Eastern Health Board; and *Department Hawkins House, Dublin, Ireland
27 March
1996; revised manuscript
of Health,
accepted 30 October 1996
Summary:
The objective of this survey was to obtain an indication of the size of the methicillin-resistant Staphylococcus aureus (MRSA) problem in Ireland prior to introducing national MRSA control guidelines. A survey of all microbiology laboratories in Ireland was carried out over two weeks in Spring 1995. For patients from whom MRSA was isolated during the study period standard demographic and clinical data were requested and period prevalence/1000 discharges was calculated. All 45 microbiology laboratories surveyed responded. MRSA was isolated from 448 patients during the twoweek period. The period prevalence of MRSA was 16.5/1000 discharges. Males aged 365 had the highest rate (SO/l000 discharges). Half of all isolates were from patients in surgical or medical wards, but 4% were from community-based sources such as GPs, nursing homes and hospices. Thirtytwo percent of MRSA patients were infected rather than colonized. MRSA is clearly a significant problem in Ireland. While it is largely a hospital problem at present, the increasing trend towards day procedures and shorter hospital stay means that infection will increase in the community.
Keywords:
Methicillin
resistance;
Staphylococcus
aureus.
Introduction Methicillin-resistant Staphylococcus aureus (MRSA) is an increasingly important cause of morbidity, and is spreading from large hospitals to smaller community-based facilities and nursing homes.’ A survey in a university hospital in the USA revealed 41% of MRSA cases to be communityacquired.* Although symptomless carriage of MRSA predominates, since the mid-1970s there has been an increasing number of reports of severe invasive infection.3’4 As a consequence, expenditure on infection control measures in hospitals and nursing homes has grown.5-7 Before the completion of this survey there were no national guidelines in Ireland on the prevention and management of MRSA; a Department of Health committee was formed to address this issue. The objectives of this study were to obtain an indication of the size of the MRSA problem in the Republic of Ireland prior to the introduction of national guidelines on MRSA control; this would serve as a baseline for Ol95-6701/97/030175+10
$12.00/O
0 1997 The Hospital
175
Infection
Society
Z. Johnson et al.
176
future studies of the effectiveness of the guidelines; screening practices in the Republic of Ireland.
and to describe
MRSA
Methods
Survey of microbiological
laboratories
A survey of all microbiology laboratories in the country was carried out over a two-week period from 20 February to 5 March 1995 on patients from whom MRSA was isolated during the study period (inpatients and outpatients, both newly-diagnosed and previously-positive patients). Each laboratory was requested also to provide information on the screening policy for MRSA in each hospital for which the laboratory performs bacteriological testing and the presence in each hospital of patients known to be infected or colonized with MRSA, but from whom a specimen had not been received by the laboratory during the study period.
Survey
of hospital
administration
departments
A survey of hospital administrators/secretary-managers was also performed. Data were requested regarding the following: the number of beds occupied on 27 February 1995 (the midpoint of the two-week study period); the number of discharges during the period 20 February to 5 March 1995; the breakdown of discharges by age and sex; the breakdown of discharges by county of residence of patient. The data were analysed using SAS (a registered trademark of SAS Institute Inc., Cary, NC, USA) on the Eastern Health Board computer system. The chi square test was used for comparison of proportions. Where the result was not significant at the 5% level this is indicated by (NS). Results
Survey
of microbiological
laboratories
All 45 microbiology laboratories surveyed responded. The total number of patients from whom MRSA was isolated over the two-week period was 448. The ratio of males:females was l-7:1 (279:169). Three hundred and forty-seven patients (77.5%) were 45 years or older; 154 (34.4%) were 75 years or older. Nineteen patients (4.2%) were under five years. Table I documents frequency of patients with MRSA confirmed by a laboratory during the period of the survey by county of residence, and the rate per 100 000 of the population. While almost one third of isolates were from patients resident in Dublin county and city, the rate of 13*6/100 000 is tenth in ranking of counties. MRSA was isolated from patients in almost every type of hospital ward/unit (Table II). The greatest number of specimens were from patients in surgical wards (30*3%), while medical and geriatric patients together accounted for another 29.2%. No data were given
Survey
of MRSA:
Ireland
1995
177
I. County of residence of patients from whom methicillin-resistant Staphylococcus aureus (MRSA) was isolated during the period of the study ranked in descending order per 100 000 population
Table
County
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
Number with
Mayo Galway Cavan Roscommon Louth Donegal Monaghan Tipperary Kildare Dublin Offaly Limerick Meath Sligo Carlow Wicklow Clare Waterford Kilkenny Leitrim Wexford Longford Cork Kerry Laois Westmeath
Total * This data the period treatment indicating + Data on fixed abode.
of patients MRSA 25 38 ::, 3;: 8 :; 139 2; 14 7
5 11
Population
*Rate
per 100 000 population
110713 180 364 52 796 51897 90 724 128 117 51 293 132772 122 656 1025 304 58 494 161956 105 370 54 756 40 942 97 265 90 918 91624 73 635 25 301 102 069 30 296 410 369 121 894 52 314 61880
21.7 21.1 20.8 19-3 18.7 15.6 15.6 14.3 13.9 13.6 13.7 13.6 13.3 12.8 12.2 11.3 11.0 9.8
3525
12.7
719
3:; 6.9 6.6 g:; i:;
refers to the number of patients confirmed MRSA positive by laboratory testing during of the survey per 100 000 residents in each county. It does not imply that they were receiving in a hospital in that county at the time of MRSA isolation and should not be regarded as the level of MRSA in any particular hospital. county of residence missing on one patient, one resident in Northern Ireland, one of no
as to the specific ward type of some patients who were in a hospital outside that in which the laboratory is situated (i.e., external). Of the 42 MRSApositive individuals who were not hospital inpatients at the time of the study, 31 were attending hospital as outpatients, day ward or accident and emergency department patients, and 11 were attending their own GP in the community. Information was not sought on recent hospital admissions for these cases. MRSA was isolated from 633 sites. The commonest sites of isolation are indicated in Table III, stratified by ward and by infection or colonization status of the patient. Infection or colonization status was unknown for 57 (12.7%) patients. There were seven isolates from blood and 14 from body
178
et al.
Z. Johnson Table
Location
I I.
of methicillin-resistant
Staphylococcus
aureus
positive patients
Location
N
%
Surgical unit Medical ward Geriatric ward Intensive care unit Outpatients department Orthopaedics External Paediatric ward Special care baby unit Psychiatric ward Burns unit Specimen sent in by GP Isolation unit Home Accident and emergency Respiratory ward Other*
135 100 30
30.3 22.5 6.7 6.1 ::; 3-4 ;:; 2.0 1.6 ::I 1.1 ::; 6.1
Total
100-o
* Other: dialysis unit, renal plastic surgery unit, cardiac t Data
on location
Table
III.
(a)
Isolates
missing
unit, hospice, day ward, nursing unit, gynaecology ward, ear/nose/throat ward, high-dependency unit, on
three
patients.
Common sites of isolation of methicillin-resistant Staphylococcus by infection or colonization status from
infected
aureus,
strati’ed
patients
Site
Ward Surgical
Wound Nose Skin Sputum Throat
(b)
ward, oncology unit, tuberculosis ward.
Isolates
from
Medical
Orthopaedic
ICU
Other
Total
25 z
77 27 15 25 5
24 z
11 1
8
:
i
4 0
7 2
(: 1
: 0
0 7 0
Geriatric
ICU
colonized
52
patients
Site
Wound Nose Skin Sputum Throat
Geriatric
Ward Surgical
Medical
21 33 18
13 23 20 14 11
Ei
This table refers to isolates, not the total of 633 isolates (70.9%).
Orthopaedic
patients. The ICU, intensive
:
i
i
4 2 4
1 five commonest care unit.
4 8 ; 4 sites
of isolation
Other ii 12 :: accounted
Total 62 108 59 36 35 for
449
of
Survey
of MRSA:
Ireland
1995
179
100 90
/
- 80 70 60 2
50 40 30 20 10 0
5-14
15-24
25-44
45-64
65-74
(.........e..),
Infected;
75+
Age group Figure
1. Age groups
of infected/colonized
patients.
( -),
colonized.
cavities. Figure 1 shows the age groups of the 391 patients (241 male, 150 female) where infection/colonization status was known; 62.2% of infections occurred in the 65 years and over age group. Of the females 34% were infected compared with 32% of the males. Sixty percent of infections occurred in male patients. Over one quarter (35; 27.6%) of infected patients and 84 (31.9%) colonized patients were in surgical wards; 26 (20.5%) and 62 (23*6%), respectively, were in medical wards. Other frequent locations of infected patients were orthopaedic wards (10; 7.9%), geriatric wards (9; 7.1%) and intensive care units (8; 6.3%). Over the two-week period the total number of MRSA screening specimens from both patients and staff received by laboratories for bacteriological analysis was 5830. The number received by individual laboratories ranged from O-689 (mean 130). Eighteen different media or media combinations are used for MRSA detection by the 45 laboratories. Over three quarters (80.5%) state that methicillin is used to detect MRSA; 14.8% oxacillin; 3.7% methicillin discs with oxacillin medium; and 1.2% methicillin and gentamicin. Screening practice protocols for 82 hospitals served by the 45 microbiology laboratories were obtained. Forty-four (53.7%) hospitals currently have an infection control nurse, either full-time or part-time. Fifty-five (67.1%) hospitals screen certain patient groups routinely for MRSA; the remaining 27 (32.9%) screen only if an outbreak occurs. Of the hospitals that screen routinely, the following sites are screened: nose 55 (loo%), axilla 42 (76*4%), throat 32 (58.2%), perineum 32 (58.2%), groin 23 (41*8%), wound 21 (38.2%), abnormal skin 18 (32*7%), catheter specimen of urine 18 (32*7%), sputum 13 (23*6%), hairline 10 (18.2%), intravenous access site
Z. Johnson
180 Table Patient
IV.
et al.
Patient groups screened in the 55 hospitals which routinely screen for MRSA
group
Transfers from other hospitals Patients previously infected/colonized with MRSA Transfers from nursing homes On admission to ICU Patients being transferred to another hospital Weekly screen of patients in ICU Admissions for cardiac surgery Pre-total hip/knee replacement surgery Admissions from the community Admissions to liver unit Monthly screen of long-stay patients Patients returning to outpatients department Patients who have been in another hospital during the preceding Preoperative screen of patients who will need to go to ICU postoperatively Pre-orthopaedic surgery Admissions to baby unit Bedridden/chronically-ill patients on antibiotics on admission Admissions to isolation unit Weekly screen of patients in liver unit ICU,
intensive
N
%
48 46 33 10
87.3 83.6 60.0 18-2 12.7
z
;::
; 2
3.6
: 1
1.8 1.8 1.8 1.8
year :
I”:;
1 1
::;
: 1
1.8 1.8 1.8
care unit.
8 (14-S%), hands 7 (12.7%) and umbilicus 7 (12.7%). The umbilicus is screened in neonates only. Where wound, abnormal skin, sputum, catheter specimen of urine and intravenous access site are cited these are dependent on the patient status at the time of screening. The patient groups routinely screened are listed in Table IV, screening practice focuses on patients transferred from other hospitals or nursing homes and those previously infected/colonized with MRSA. Seventy-nine (89.0%) hospitals screen staff; details of screening practices were available for 49 (62.0%) hospitals: all screen during outbreaks of MRSA; eight screen all doctors and nurses on employment also; one hospital screens in addition all doctors commencing working in surgery and nurses commencing in ICU and one hospital screens all anaesthetic staff every six months. Fifty-one (62.2%) hospitals screen the environment for MRSA. Data were provided on 71 hospitals by the microbiology laboratories that serve them regarding the existence of a known reservoir of MRSA infected/ colonized patients who were in the hospital during the period of the study, but from whom a screening specimen was not received by the laboratory ( i.e., diagnosis of MRSA made before commencement of the study); 16 hospitals had such a reservoir of infected patients and data regarding the number of patients were provided on 15. The total number of such patients was 111, with the number in individual hospitals ranging from one to 24 (median five). An estimate of the period prevalence rate of MRSA infection/colonization
Survey
of MRSA:
Ireland
1995
181
for the study duration can be obtained from the sum (559) of the number of patients diagnosed as MRSA positive during the period of study (448) plus the number of MRSA positive patients in the reservoirs of infection in the hospitals during the period (111); because data are missing on reservoirs of infection in 12 hospitals, and the data only include those patients screened for MRSA, this period prevalence rate of 15.9/100 000 population [using the population of Ireland (3.525 million) as denominator] must be considered as the minimum rate.
Survey of hospital
administration
departments
Fifty-seven of the 59 hospitals surveyed responded to the questionnaire sent to administrators/secretary-managers (96.6%). Due to the inability of one laboratory to distinguish isolates from a group of hospitals, the administrative data on this group were treated as if from a single hospital, giving a total of 55 hospitals. While all 55 hospitals provided the numbers of total discharges and occupied beds, data regarding the breakdown by age and sex were given by only 44 hospitals, and the breakdown by county of residence of patient by 42 hospitals. The period prevalence rate of MRSA was 16.5/1000 discharges and 36*2/ 1000 beds occupied. The rate of MRSA in males was 25*5/1000 male discharges and the rate in females was 13*0/1000 female discharges. The rate of MRSA was greater in males than in females in all age groups, with the difference most marked in the over 65 age group. Period prevalence rates were also calculated for the number of cases of MRSA in patients resident in each health board area per 1000 patients discharged from all hospitals who were resident in the same health board area. The rates were as follows: Western Health Board 25.3, North-Eastern Health Board 24.2, Eastern Health Board 20.3, Mid-Western Health Board 19.5, Southern Health Board 15.3, North-Western Health Board 15.1, South-Eastern Health Board 10.2 and Midland Health Board 8.7. These rates were calculated using county of residence of patients as numerator and denominator. In many cases, patients were being treated in hospitals located in health board areas other than their area of residence. Therefore the rates cannot be used to infer a problem in any particular hospital or group of hospitals. Discussion
MRSA is clearly a problem in Irish hospitals as a substantial number of isolates were recorded during the relatively short survey window. The majority of MRSA positive patients are elderly men, and at all ages males predominate; male sex has been previously documented as a predictor of MRSA colonization.’ The largest proportion of isolates came from patients in surgical and medical wards, while the documentation of isolates from intensive care units, special care baby units and oncology wards indicates
182
2. Johnson
et al.
a potentially very serious problem. A high prevalence of MRSA in medical, surgical and intensive care units has been well documented internationally.9~‘0 It has been shown that in hospitals where infection control measures are strictly enforced, the incidence of infections due to resistant microorganisms can be reduced.” Inappropriate use of antibiotics must be curtailed also to address the problem.” Considerable microbiology laboratory support is essential.” The costs of controlling a 91-week outbreak of MRSA in the United Kingdom were estimated in 1994 as 4403 600 (i.e., j(14435/week);’ the cost-effectiveness of infection control can only truly be appreciated in the light of sequential surveillance data.i2 While MRSA is largely a hospital-acquired pathogen, the increasing trend for day-case procedures, shorter hospital stays and home-based parenteral therapy means that hospital-acquired infection will likely increase as a problem in the community also.13 A recent study of MRSA at a university hospital in the USA revealed 41% of MRSA cases to be community-acquired, with 8% of these having no identified risk factors (hospital admissions, antibiotic use, nursing home residence or intravenous drug use).’ Canadian hospital studies have revealed that 62% of MRSA-positive patients had MRSA present at admission to hospital.14 Rates of infection in individual hospitals may fluctuate from week to week, thus, for a given hospital, the rate calculated within the time scale of a two-week survey period may either over or underestimate the true rate. However, on a nationwide basis, the overall study rates should adequately reflect the general pattern of MRSA. Detection of colonization by MRSA depends in part on screening practices which clearly vary, hence inter-institutional comparisons were inappropriate. While this study highlights the volume of activity imposed on microbiological laboratories by MRSA it must be remembered that our figures will underestimate this because the figure of 448 represents only one isolate from each patient, and these patients may have had multiple sites screened and repeat testing performed. The current variation in screening practices in different hospitals means that the activity of individual laboratories varies greatly. We did not set out in this study to quantify costs in terms of labour and materials but we feel this may merit further study. Bed occupancy, hospital discharges, total and county populations were used as denominators in this study. We used the number of discharges as one denominator because it was readily available. Clearly the number of admissions would be equally useful. The use of hospital catchment populations as denominator would be an appropriate alternative in another setting; use of hospital catchment populations was not possible in this study because such populations are not readily defined in Ireland, with relatively free choice of hospital for patients, especially within urban settings. Ongoing surveillance of MRSA infection is obviously important. Routine electronic data transmission from microbiological laboratories to a central
Survey
of MRSA:
Ireland
1995
183
surveillance centre would facilitate collection of numerator data. Bloodstream isolates of MRSA could also be used as a sentinel of infection trends; use of such a limited sentinel would simplify data collection. The need for standardized guidelines on MRSA control is apparent. The Department of Health has issued guidelines”-” on control of MRSA following recommendations of the Departmental Committee on MRSA;18 this study should serve as a useful baseline against which to measure their effect, We wish to acknowledge the support of the following people for this study: Mr Frank Aherne, Principal Officer, Department of Health, for funding this study; Mr Brian O’Boyce, Health Information Unit, Eastern Health Board, for data entry services; Dr Eric Mulvihill, consultant microbiologist, St James’ Hospital; Dr Mary Cafferkey, consultant microbiologist, The Children’s Hospital, Temple Street; Professor Conor Keane, consultant microbiologist, St James’ Hospital; Dr Rosemary Hone, consultant microbiologist, Mater Misericordiae Hospital; the members of the National MRSA Committee; the consultant microbiologists who responded to the survey; the staff of microbiology laboratories who assisted with data collection; the hospital administrative staff who provided discharge data.
References 1. Emori TG, Gaynes RP. An overview of nosocomial infections, including the role of the microbiology laboratory. Clin MicrobioE Rev 1993; 6: 428442. 2. Layton MC, Hierholzer WJ, Patterson JE. The evolving epidemiology of methicillinresistant Staphylococcus aureus. Am J Med 1991; 91 (Suppl. 3B): 228S-2328. 3. Brumfit W, Hamilton-Miller J. Methicillin resistant Staphylococcus aureus. N Engl J Med 1989; 320: 1188-1196. 4. Keane CT. Methicillin-resistant Staphylococcus aureus (MRSA). IMJ 1993; 86: 151. 5. Cox RA, Mallaghan C, Conquest C, King J. Epidemic methicillin-resistant Staphylococcus aureus: controlling the spread outside hospital. J Hasp Infect 1995; 29: 107-119. 6. Hone R. The epidemiology of methicillin-resistant Staphylococcus aureus. In: Cafferkey MT, Ed. Methicillin-resistant Staphylococcus aureus. Clinical management and laboratory aspects. New York: Marcel Dekker, 1992: 91-105. 7. Cox RA, Conquest C, Mallaghan C, Marples RR. A major outbreak of methicillinresistant Staphylococcus aureus caused by a new phage-type (EMRSA-16). J Hosp Infect 1995; 29: 87-106. 8. Murphy S, Denman S, Bennett RG, Greenough WB, Lindsay J, Zelesnick LB. Methicillin-resistant Staphylococcus aureus colonization in a long-term care facility. J Am Geriatr Sot 1992; 40: 213-217. 9. Cheong I, Tan SC, Wong YH, Zainudin BM, Rahman MZ. Methicillin-resistant m hospital. Med J Malaysia 1994; 49: Staphylococcus aureus (MRSA) . a Malaysian 24-28. 10. Tomasz A. Multiple-antibiotic-resistant pathogenic bacteria-a report on the Rockefeller University workshop. N Engl J Med 1994; 330: 1247-1251. 11. Murray BE. Can antibiotic resistance be controlled? N EnglJ Med 1994; 330: 1229-1230. 12. Chaudhuri AK. Infection control in hospitals: has its quality-enhancing and costeffective role been appreciated? J Hasp Infect 1993; 25: l-6. 13. Rosenberg J. Methicillin-resistant Staphylococcus aureus (MRSA) in the community: who’s watching? Lancet 1994; 346: 132-133. 14. Embil J, Ramotar K, Romance L. Methicillin-resistant Staphylococcus aureus in tertiary care institutions on the Canadian prairies 1990-1992. Infect Control Hasp Epidemiol 1994; 15: 646-657.
184 15.
16. 17.
18.
Z. Johnson
et al.
Department of Health. Guidelines for the Control of Methicillin Resistant Staphylococcus aureus (MRSA ) in Acute Hosbital Wards. Includine - Sbecialist Units. Dublin: De* partment of Health, 1995. a Department of Health. Guidelines for the Control of Methicillin Resistant Staphylococcus aureus (MRSA) in the Irish Healthcare Setting. Dublin: Department of Health, 1995. Department of Health. Guidelines for the Control of Methicillin Resistant Staphylococcus aureus (MRSA) in DistrictfCommunity Hospitals, Special Hospitals and Nursing/Residential Homes. Dublin: Department of Health, 1995. Committee established by the Department of Health. Control of Methicillin Resistant Staphylococcus aureus in the Irish Healthcare Setting. Dublin: Department of Health, 1994.
of Hospital
Infection
National
(1997)
35,
survey
175-184
of MRSA:
Ireland,
1995
Z. Johnson, P. Fitzpatrick, C. Hayes, G. Sayers, H. Pelly, B. McDonnell, L. Thornton and J. Buttimer*
Health
Information
Received
Unit, Eastern Health Board; and *Department Hawkins House, Dublin, Ireland
27 March
1996; revised manuscript
of Health,
accepted 30 October 1996
Summary:
The objective of this survey was to obtain an indication of the size of the methicillin-resistant Staphylococcus aureus (MRSA) problem in Ireland prior to introducing national MRSA control guidelines. A survey of all microbiology laboratories in Ireland was carried out over two weeks in Spring 1995. For patients from whom MRSA was isolated during the study period standard demographic and clinical data were requested and period prevalence/1000 discharges was calculated. All 45 microbiology laboratories surveyed responded. MRSA was isolated from 448 patients during the twoweek period. The period prevalence of MRSA was 16.5/1000 discharges. Males aged 365 had the highest rate (SO/l000 discharges). Half of all isolates were from patients in surgical or medical wards, but 4% were from community-based sources such as GPs, nursing homes and hospices. Thirtytwo percent of MRSA patients were infected rather than colonized. MRSA is clearly a significant problem in Ireland. While it is largely a hospital problem at present, the increasing trend towards day procedures and shorter hospital stay means that infection will increase in the community.
Keywords:
Methicillin
resistance;
Staphylococcus
aureus.
Introduction Methicillin-resistant Staphylococcus aureus (MRSA) is an increasingly important cause of morbidity, and is spreading from large hospitals to smaller community-based facilities and nursing homes.’ A survey in a university hospital in the USA revealed 41% of MRSA cases to be communityacquired.* Although symptomless carriage of MRSA predominates, since the mid-1970s there has been an increasing number of reports of severe invasive infection.3’4 As a consequence, expenditure on infection control measures in hospitals and nursing homes has grown.5-7 Before the completion of this survey there were no national guidelines in Ireland on the prevention and management of MRSA; a Department of Health committee was formed to address this issue. The objectives of this study were to obtain an indication of the size of the MRSA problem in the Republic of Ireland prior to the introduction of national guidelines on MRSA control; this would serve as a baseline for Ol95-6701/97/030175+10
$12.00/O
0 1997 The Hospital
175
Infection
Society
Z. Johnson et al.
176
future studies of the effectiveness of the guidelines; screening practices in the Republic of Ireland.
and to describe
MRSA
Methods
Survey of microbiological
laboratories
A survey of all microbiology laboratories in the country was carried out over a two-week period from 20 February to 5 March 1995 on patients from whom MRSA was isolated during the study period (inpatients and outpatients, both newly-diagnosed and previously-positive patients). Each laboratory was requested also to provide information on the screening policy for MRSA in each hospital for which the laboratory performs bacteriological testing and the presence in each hospital of patients known to be infected or colonized with MRSA, but from whom a specimen had not been received by the laboratory during the study period.
Survey
of hospital
administration
departments
A survey of hospital administrators/secretary-managers was also performed. Data were requested regarding the following: the number of beds occupied on 27 February 1995 (the midpoint of the two-week study period); the number of discharges during the period 20 February to 5 March 1995; the breakdown of discharges by age and sex; the breakdown of discharges by county of residence of patient. The data were analysed using SAS (a registered trademark of SAS Institute Inc., Cary, NC, USA) on the Eastern Health Board computer system. The chi square test was used for comparison of proportions. Where the result was not significant at the 5% level this is indicated by (NS). Results
Survey
of microbiological
laboratories
All 45 microbiology laboratories surveyed responded. The total number of patients from whom MRSA was isolated over the two-week period was 448. The ratio of males:females was l-7:1 (279:169). Three hundred and forty-seven patients (77.5%) were 45 years or older; 154 (34.4%) were 75 years or older. Nineteen patients (4.2%) were under five years. Table I documents frequency of patients with MRSA confirmed by a laboratory during the period of the survey by county of residence, and the rate per 100 000 of the population. While almost one third of isolates were from patients resident in Dublin county and city, the rate of 13*6/100 000 is tenth in ranking of counties. MRSA was isolated from patients in almost every type of hospital ward/unit (Table II). The greatest number of specimens were from patients in surgical wards (30*3%), while medical and geriatric patients together accounted for another 29.2%. No data were given
Survey
of MRSA:
Ireland
1995
177
I. County of residence of patients from whom methicillin-resistant Staphylococcus aureus (MRSA) was isolated during the period of the study ranked in descending order per 100 000 population
Table
County
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
Number with
Mayo Galway Cavan Roscommon Louth Donegal Monaghan Tipperary Kildare Dublin Offaly Limerick Meath Sligo Carlow Wicklow Clare Waterford Kilkenny Leitrim Wexford Longford Cork Kerry Laois Westmeath
Total * This data the period treatment indicating + Data on fixed abode.
of patients MRSA 25 38 ::, 3;: 8 :; 139 2; 14 7
5 11
Population
*Rate
per 100 000 population
110713 180 364 52 796 51897 90 724 128 117 51 293 132772 122 656 1025 304 58 494 161956 105 370 54 756 40 942 97 265 90 918 91624 73 635 25 301 102 069 30 296 410 369 121 894 52 314 61880
21.7 21.1 20.8 19-3 18.7 15.6 15.6 14.3 13.9 13.6 13.7 13.6 13.3 12.8 12.2 11.3 11.0 9.8
3525
12.7
719
3:; 6.9 6.6 g:; i:;
refers to the number of patients confirmed MRSA positive by laboratory testing during of the survey per 100 000 residents in each county. It does not imply that they were receiving in a hospital in that county at the time of MRSA isolation and should not be regarded as the level of MRSA in any particular hospital. county of residence missing on one patient, one resident in Northern Ireland, one of no
as to the specific ward type of some patients who were in a hospital outside that in which the laboratory is situated (i.e., external). Of the 42 MRSApositive individuals who were not hospital inpatients at the time of the study, 31 were attending hospital as outpatients, day ward or accident and emergency department patients, and 11 were attending their own GP in the community. Information was not sought on recent hospital admissions for these cases. MRSA was isolated from 633 sites. The commonest sites of isolation are indicated in Table III, stratified by ward and by infection or colonization status of the patient. Infection or colonization status was unknown for 57 (12.7%) patients. There were seven isolates from blood and 14 from body
178
et al.
Z. Johnson Table
Location
I I.
of methicillin-resistant
Staphylococcus
aureus
positive patients
Location
N
%
Surgical unit Medical ward Geriatric ward Intensive care unit Outpatients department Orthopaedics External Paediatric ward Special care baby unit Psychiatric ward Burns unit Specimen sent in by GP Isolation unit Home Accident and emergency Respiratory ward Other*
135 100 30
30.3 22.5 6.7 6.1 ::; 3-4 ;:; 2.0 1.6 ::I 1.1 ::; 6.1
Total
100-o
* Other: dialysis unit, renal plastic surgery unit, cardiac t Data
on location
Table
III.
(a)
Isolates
missing
unit, hospice, day ward, nursing unit, gynaecology ward, ear/nose/throat ward, high-dependency unit, on
three
patients.
Common sites of isolation of methicillin-resistant Staphylococcus by infection or colonization status from
infected
aureus,
strati’ed
patients
Site
Ward Surgical
Wound Nose Skin Sputum Throat
(b)
ward, oncology unit, tuberculosis ward.
Isolates
from
Medical
Orthopaedic
ICU
Other
Total
25 z
77 27 15 25 5
24 z
11 1
8
:
i
4 0
7 2
(: 1
: 0
0 7 0
Geriatric
ICU
colonized
52
patients
Site
Wound Nose Skin Sputum Throat
Geriatric
Ward Surgical
Medical
21 33 18
13 23 20 14 11
Ei
This table refers to isolates, not the total of 633 isolates (70.9%).
Orthopaedic
patients. The ICU, intensive
:
i
i
4 2 4
1 five commonest care unit.
4 8 ; 4 sites
of isolation
Other ii 12 :: accounted
Total 62 108 59 36 35 for
449
of
Survey
of MRSA:
Ireland
1995
179
100 90
/
- 80 70 60 2
50 40 30 20 10 0
5-14
15-24
25-44
45-64
65-74
(.........e..),
Infected;
75+
Age group Figure
1. Age groups
of infected/colonized
patients.
( -),
colonized.
cavities. Figure 1 shows the age groups of the 391 patients (241 male, 150 female) where infection/colonization status was known; 62.2% of infections occurred in the 65 years and over age group. Of the females 34% were infected compared with 32% of the males. Sixty percent of infections occurred in male patients. Over one quarter (35; 27.6%) of infected patients and 84 (31.9%) colonized patients were in surgical wards; 26 (20.5%) and 62 (23*6%), respectively, were in medical wards. Other frequent locations of infected patients were orthopaedic wards (10; 7.9%), geriatric wards (9; 7.1%) and intensive care units (8; 6.3%). Over the two-week period the total number of MRSA screening specimens from both patients and staff received by laboratories for bacteriological analysis was 5830. The number received by individual laboratories ranged from O-689 (mean 130). Eighteen different media or media combinations are used for MRSA detection by the 45 laboratories. Over three quarters (80.5%) state that methicillin is used to detect MRSA; 14.8% oxacillin; 3.7% methicillin discs with oxacillin medium; and 1.2% methicillin and gentamicin. Screening practice protocols for 82 hospitals served by the 45 microbiology laboratories were obtained. Forty-four (53.7%) hospitals currently have an infection control nurse, either full-time or part-time. Fifty-five (67.1%) hospitals screen certain patient groups routinely for MRSA; the remaining 27 (32.9%) screen only if an outbreak occurs. Of the hospitals that screen routinely, the following sites are screened: nose 55 (loo%), axilla 42 (76*4%), throat 32 (58.2%), perineum 32 (58.2%), groin 23 (41*8%), wound 21 (38.2%), abnormal skin 18 (32*7%), catheter specimen of urine 18 (32*7%), sputum 13 (23*6%), hairline 10 (18.2%), intravenous access site
Z. Johnson
180 Table Patient
IV.
et al.
Patient groups screened in the 55 hospitals which routinely screen for MRSA
group
Transfers from other hospitals Patients previously infected/colonized with MRSA Transfers from nursing homes On admission to ICU Patients being transferred to another hospital Weekly screen of patients in ICU Admissions for cardiac surgery Pre-total hip/knee replacement surgery Admissions from the community Admissions to liver unit Monthly screen of long-stay patients Patients returning to outpatients department Patients who have been in another hospital during the preceding Preoperative screen of patients who will need to go to ICU postoperatively Pre-orthopaedic surgery Admissions to baby unit Bedridden/chronically-ill patients on antibiotics on admission Admissions to isolation unit Weekly screen of patients in liver unit ICU,
intensive
N
%
48 46 33 10
87.3 83.6 60.0 18-2 12.7
z
;::
; 2
3.6
: 1
1.8 1.8 1.8 1.8
year :
I”:;
1 1
::;
: 1
1.8 1.8 1.8
care unit.
8 (14-S%), hands 7 (12.7%) and umbilicus 7 (12.7%). The umbilicus is screened in neonates only. Where wound, abnormal skin, sputum, catheter specimen of urine and intravenous access site are cited these are dependent on the patient status at the time of screening. The patient groups routinely screened are listed in Table IV, screening practice focuses on patients transferred from other hospitals or nursing homes and those previously infected/colonized with MRSA. Seventy-nine (89.0%) hospitals screen staff; details of screening practices were available for 49 (62.0%) hospitals: all screen during outbreaks of MRSA; eight screen all doctors and nurses on employment also; one hospital screens in addition all doctors commencing working in surgery and nurses commencing in ICU and one hospital screens all anaesthetic staff every six months. Fifty-one (62.2%) hospitals screen the environment for MRSA. Data were provided on 71 hospitals by the microbiology laboratories that serve them regarding the existence of a known reservoir of MRSA infected/ colonized patients who were in the hospital during the period of the study, but from whom a screening specimen was not received by the laboratory ( i.e., diagnosis of MRSA made before commencement of the study); 16 hospitals had such a reservoir of infected patients and data regarding the number of patients were provided on 15. The total number of such patients was 111, with the number in individual hospitals ranging from one to 24 (median five). An estimate of the period prevalence rate of MRSA infection/colonization
Survey
of MRSA:
Ireland
1995
181
for the study duration can be obtained from the sum (559) of the number of patients diagnosed as MRSA positive during the period of study (448) plus the number of MRSA positive patients in the reservoirs of infection in the hospitals during the period (111); because data are missing on reservoirs of infection in 12 hospitals, and the data only include those patients screened for MRSA, this period prevalence rate of 15.9/100 000 population [using the population of Ireland (3.525 million) as denominator] must be considered as the minimum rate.
Survey of hospital
administration
departments
Fifty-seven of the 59 hospitals surveyed responded to the questionnaire sent to administrators/secretary-managers (96.6%). Due to the inability of one laboratory to distinguish isolates from a group of hospitals, the administrative data on this group were treated as if from a single hospital, giving a total of 55 hospitals. While all 55 hospitals provided the numbers of total discharges and occupied beds, data regarding the breakdown by age and sex were given by only 44 hospitals, and the breakdown by county of residence of patient by 42 hospitals. The period prevalence rate of MRSA was 16.5/1000 discharges and 36*2/ 1000 beds occupied. The rate of MRSA in males was 25*5/1000 male discharges and the rate in females was 13*0/1000 female discharges. The rate of MRSA was greater in males than in females in all age groups, with the difference most marked in the over 65 age group. Period prevalence rates were also calculated for the number of cases of MRSA in patients resident in each health board area per 1000 patients discharged from all hospitals who were resident in the same health board area. The rates were as follows: Western Health Board 25.3, North-Eastern Health Board 24.2, Eastern Health Board 20.3, Mid-Western Health Board 19.5, Southern Health Board 15.3, North-Western Health Board 15.1, South-Eastern Health Board 10.2 and Midland Health Board 8.7. These rates were calculated using county of residence of patients as numerator and denominator. In many cases, patients were being treated in hospitals located in health board areas other than their area of residence. Therefore the rates cannot be used to infer a problem in any particular hospital or group of hospitals. Discussion
MRSA is clearly a problem in Irish hospitals as a substantial number of isolates were recorded during the relatively short survey window. The majority of MRSA positive patients are elderly men, and at all ages males predominate; male sex has been previously documented as a predictor of MRSA colonization.’ The largest proportion of isolates came from patients in surgical and medical wards, while the documentation of isolates from intensive care units, special care baby units and oncology wards indicates
182
2. Johnson
et al.
a potentially very serious problem. A high prevalence of MRSA in medical, surgical and intensive care units has been well documented internationally.9~‘0 It has been shown that in hospitals where infection control measures are strictly enforced, the incidence of infections due to resistant microorganisms can be reduced.” Inappropriate use of antibiotics must be curtailed also to address the problem.” Considerable microbiology laboratory support is essential.” The costs of controlling a 91-week outbreak of MRSA in the United Kingdom were estimated in 1994 as 4403 600 (i.e., j(14435/week);’ the cost-effectiveness of infection control can only truly be appreciated in the light of sequential surveillance data.i2 While MRSA is largely a hospital-acquired pathogen, the increasing trend for day-case procedures, shorter hospital stays and home-based parenteral therapy means that hospital-acquired infection will likely increase as a problem in the community also.13 A recent study of MRSA at a university hospital in the USA revealed 41% of MRSA cases to be community-acquired, with 8% of these having no identified risk factors (hospital admissions, antibiotic use, nursing home residence or intravenous drug use).’ Canadian hospital studies have revealed that 62% of MRSA-positive patients had MRSA present at admission to hospital.14 Rates of infection in individual hospitals may fluctuate from week to week, thus, for a given hospital, the rate calculated within the time scale of a two-week survey period may either over or underestimate the true rate. However, on a nationwide basis, the overall study rates should adequately reflect the general pattern of MRSA. Detection of colonization by MRSA depends in part on screening practices which clearly vary, hence inter-institutional comparisons were inappropriate. While this study highlights the volume of activity imposed on microbiological laboratories by MRSA it must be remembered that our figures will underestimate this because the figure of 448 represents only one isolate from each patient, and these patients may have had multiple sites screened and repeat testing performed. The current variation in screening practices in different hospitals means that the activity of individual laboratories varies greatly. We did not set out in this study to quantify costs in terms of labour and materials but we feel this may merit further study. Bed occupancy, hospital discharges, total and county populations were used as denominators in this study. We used the number of discharges as one denominator because it was readily available. Clearly the number of admissions would be equally useful. The use of hospital catchment populations as denominator would be an appropriate alternative in another setting; use of hospital catchment populations was not possible in this study because such populations are not readily defined in Ireland, with relatively free choice of hospital for patients, especially within urban settings. Ongoing surveillance of MRSA infection is obviously important. Routine electronic data transmission from microbiological laboratories to a central
Survey
of MRSA:
Ireland
1995
183
surveillance centre would facilitate collection of numerator data. Bloodstream isolates of MRSA could also be used as a sentinel of infection trends; use of such a limited sentinel would simplify data collection. The need for standardized guidelines on MRSA control is apparent. The Department of Health has issued guidelines”-” on control of MRSA following recommendations of the Departmental Committee on MRSA;18 this study should serve as a useful baseline against which to measure their effect, We wish to acknowledge the support of the following people for this study: Mr Frank Aherne, Principal Officer, Department of Health, for funding this study; Mr Brian O’Boyce, Health Information Unit, Eastern Health Board, for data entry services; Dr Eric Mulvihill, consultant microbiologist, St James’ Hospital; Dr Mary Cafferkey, consultant microbiologist, The Children’s Hospital, Temple Street; Professor Conor Keane, consultant microbiologist, St James’ Hospital; Dr Rosemary Hone, consultant microbiologist, Mater Misericordiae Hospital; the members of the National MRSA Committee; the consultant microbiologists who responded to the survey; the staff of microbiology laboratories who assisted with data collection; the hospital administrative staff who provided discharge data.
References 1. Emori TG, Gaynes RP. An overview of nosocomial infections, including the role of the microbiology laboratory. Clin MicrobioE Rev 1993; 6: 428442. 2. Layton MC, Hierholzer WJ, Patterson JE. The evolving epidemiology of methicillinresistant Staphylococcus aureus. Am J Med 1991; 91 (Suppl. 3B): 228S-2328. 3. Brumfit W, Hamilton-Miller J. Methicillin resistant Staphylococcus aureus. N Engl J Med 1989; 320: 1188-1196. 4. Keane CT. Methicillin-resistant Staphylococcus aureus (MRSA). IMJ 1993; 86: 151. 5. Cox RA, Mallaghan C, Conquest C, King J. Epidemic methicillin-resistant Staphylococcus aureus: controlling the spread outside hospital. J Hasp Infect 1995; 29: 107-119. 6. Hone R. The epidemiology of methicillin-resistant Staphylococcus aureus. In: Cafferkey MT, Ed. Methicillin-resistant Staphylococcus aureus. Clinical management and laboratory aspects. New York: Marcel Dekker, 1992: 91-105. 7. Cox RA, Conquest C, Mallaghan C, Marples RR. A major outbreak of methicillinresistant Staphylococcus aureus caused by a new phage-type (EMRSA-16). J Hosp Infect 1995; 29: 87-106. 8. Murphy S, Denman S, Bennett RG, Greenough WB, Lindsay J, Zelesnick LB. Methicillin-resistant Staphylococcus aureus colonization in a long-term care facility. J Am Geriatr Sot 1992; 40: 213-217. 9. Cheong I, Tan SC, Wong YH, Zainudin BM, Rahman MZ. Methicillin-resistant m hospital. Med J Malaysia 1994; 49: Staphylococcus aureus (MRSA) . a Malaysian 24-28. 10. Tomasz A. Multiple-antibiotic-resistant pathogenic bacteria-a report on the Rockefeller University workshop. N Engl J Med 1994; 330: 1247-1251. 11. Murray BE. Can antibiotic resistance be controlled? N EnglJ Med 1994; 330: 1229-1230. 12. Chaudhuri AK. Infection control in hospitals: has its quality-enhancing and costeffective role been appreciated? J Hasp Infect 1993; 25: l-6. 13. Rosenberg J. Methicillin-resistant Staphylococcus aureus (MRSA) in the community: who’s watching? Lancet 1994; 346: 132-133. 14. Embil J, Ramotar K, Romance L. Methicillin-resistant Staphylococcus aureus in tertiary care institutions on the Canadian prairies 1990-1992. Infect Control Hasp Epidemiol 1994; 15: 646-657.
184 15.
16. 17.
18.
Z. Johnson
et al.
Department of Health. Guidelines for the Control of Methicillin Resistant Staphylococcus aureus (MRSA ) in Acute Hosbital Wards. Includine - Sbecialist Units. Dublin: De* partment of Health, 1995. a Department of Health. Guidelines for the Control of Methicillin Resistant Staphylococcus aureus (MRSA) in the Irish Healthcare Setting. Dublin: Department of Health, 1995. Department of Health. Guidelines for the Control of Methicillin Resistant Staphylococcus aureus (MRSA) in DistrictfCommunity Hospitals, Special Hospitals and Nursing/Residential Homes. Dublin: Department of Health, 1995. Committee established by the Department of Health. Control of Methicillin Resistant Staphylococcus aureus in the Irish Healthcare Setting. Dublin: Department of Health, 1994.