Journal
of Hospital
Infection
(1992)
20, 247-255
Dynamics of coagulase-negative staphylococcal colonization in patients and employees in a surgical intensive care unit J. R. Thurn, Infectious
K. B. Crossley,
A. Gerdts
Disease Section, Department Center, 640-7ackson Street, Accepted
and
L. Baken
of Medicine, St. Paul-Ramsey St. Paul, Minnesota, USA
for publication
12 November
Medical
1991
Summary:
Because there is little information about the frequency of carriage of various species of coagulase-negative staphylococci (CNS) in hosnitalized natients and the nossible connection of these organisms with hospital staff, we prospectively investigated nasal CNS in patients and personnel in a Surgical Intensive Care Unit (SICU). The majority of CNS from personnel were Staphylococcus epidermidis. The CNS species from patients on admission were more diverse and included multiply antibioticresistant S. haemolvticus. Patients’ CNS became more like CNS colonizing personnel after admission with respect to both antimicrobial susceptibilit; and speciation. Plasmid and antibiotic sensitivity profiles of S. epidermidis resistant to multiple antibiotics from five patients were identified as those from one employee, but there was no evidence that this was of clinical significance. A varietv of factors influence nasal colonization bv CNS in SfCUs. The nasal CgS of patients change after admission and may become more resistant and less diverse. The factors influencing changes in the antibiotic susceptibility and the aetiology of CNS infection require further study. Keywords:
Coagulase-negative
staphylococci;
colonization;
nasal
carriage.
Introduction
Coagulase-negative staphylococci (CNS) have been increasingly recognized as causes of nosocomial infection and have become a leading cause of nosocomial bacteraemia.’ Sepsis caused by CNS has often been noted to involve organisms resistant to multiple antibiotics,* to prolong hospital stay and to increase mortality.3 Infections caused by CNS are frequently related to prosthetic devices4 and have been documented with increased frequency in compromised patients.5-7 The antimicrobial susceptibility of CNS has been shown to be species and vancomycin resistance has developed in species other dependents-” than Staphylococcus epidermidis.“,‘2 Many laboratories, however, have Correspondence Minneapolis 019556701/92/040247+
to: Joseph Veterans Affairs
R.
Thurn, Medical
M.D., Center,
Infectious Minneapolis,
09 $03.00/O
Disease MN
Section, Department 55417, USA. 0 1992 The Hospital
247
of Medicine,
Infection
Society
248
J. R. Thurn
et al.
identified these organisms as ‘Staphylococcus epidermidis’ and have not attempted to speciate them correctly. Because there is limited information available about the frequency of carriage of various CNS in hospitalized patients or personnel, and few studies have involved a single hospital unit or ward, we examined this in a Surgical Intensive Care Unit (SICU). In order to define better the epidemiology of ,CNS, antibiotic resistance patterns were determined and plasmid analysis was performed on the organisms isolated in this study.
Methods
and
materials
Employee nasal swabs Nasal cultures were self-collected using moistened Dacron swabs (Culturettes, Baxter Healthcare Corporation, McGaw Park, IL, USA) by employees in our SICU during the period between August 17 and August 20, 1987. Patient nasal swabs Patients in the SICU were studied in the period between August and October 1987. Cultures were done within 24 h of admission and again at 7 days, death or discharge from the SICU. They were collected by staff in a manner identical to the employee nasal swabs. Swabs were inoculated on to 5% sheep blood agar and mannitol salt agar within 4 h of collection and incubated for 36 h at 35°C. S. aUreUs was identified by tube coagulase testing or use of the API Staph-Ident strip (Analytab Products, Inc., Chicago, IL). Other staphylococci were speciated using the API Staph-Ident. Novobiocin resistance was used to differentiate S. hominis from S. saprophyticus. Isolates were tested for susceptibility to vancomycin, trimethoprim-sulphamethoxazole, penicillin, oxacillin, erythromycin, amikacin, clindamycin, cefazolin, gentamicin and tobramycin using the NCCLS standard disc diffusion method. Plasmid analysis Plasmid analysis employed a method similar to that of Birnboim and D01y.l~ Briefly, the bacterial cells were lysed with lysozyme and 1% sodium dodecyl sulphate in 1 M sodium hydroxide. This was followed by neutralization and precipitation of chromosomal DNA with sodium acetate and phenol extraction. The collected DNA was electrophoresed using 0.7% agarose gel with a duration of 16 min at 25 V. The resulting DNA bands were photographed with Polaroid film after treatment with ethidium bromide. Statistical analysis Differences in proportions were compared using Chi-square and Fisher’s exact tests when appropriate. A commercially available statistical package was utilized (A-stat, Epistat Services, Richardson, TX, USA).
Dynamics
of CNS in an SICU
249
Results
Nasal swab cultures were obtained from 9.5 health-care workers employed in the SICU. Cultures were obtained from employees on all shifts and included nurses, aides, physicians, trainees and ancillary personnel. Only one employee refused to participate. Cultures were obtaineed from the anterior nares of 125 patients within 24 h of admission to the SICU. During this time, there were 12 patients who were not cultured because of transfer, death or the inability to obtain informed consent. A second culture was obtained from 65 patients (52%) a mean of 3.2 days (range l-7 days) after the initial culture. A third culture was obtained from five patients a mean of 4.2 days (range l-8 days) after the second culture and an average of 10.8 days after admission to the SICU. There were 173 isolates of staphylococci from the cultures of employees, 35 of which were S. aureus. No methicillin-resistant S. aUYeUS (MRSA) were detected. Of the 138 coagulase-negative staphylococci (CNS) isolated from employees, the majority (123 out of 138, 89.1%) were S. epidermidis; the remainder included S. capitis (6), S. warneri (4), S. haemolyticus (3), S. hominis (1) and S. simulans (1) (Table I). Staphylococcus aureus was found to comprise a similar proportion of patient staphylococcal isolates (53 out of 359, 15%) as employees (35 out of 173, 20%; P=O.l4). As for employees, no patients were found to be colonized by MRSA. There was, however, a difference in the proportions of species of CNS isolated from the initial patient cultures (obtained less that 24 h after admission) when compared with the CNS isolated from employees (Table I). There were significantly fewer S. epidermidis (128 out of 215, 59.5% vs. 123 out of 138, 89.1%; P< 0.001); significantly more S. haemolyticus (50 out of 215, 23.3% vs. 3 out of 138, 2.2; P24h
% (A/)
and patients
--
< 0.001 < 0.036 < 0*002.f
P
Patients
employees
0.5(l) lOO(215)
from
59,5(128) 23.3 (SO) 6.5(14) 3.7(8) 3.7(S) 0.9(2) 0.9(2) 0.9 (2)
7 days
2 a rb
Y F
Dynamics
of CNS
in
an SICU
251
When comparing the antibiotic sensitivity profiles for S. epidermidis, the isolates obtained from the employees were significantly more often resistant to penicillin, erythromycin and clindamycin than the initial patient isolates. When compared to the second patient isolates, there was no difference in the proportion of employee isolates resistant to any of the antibiotics (Table II). There was no significant difference in the proportions of S. haemolyticus resistant to the antibiotics between the initial patient isolates and the employee or the second patient isolates (Table III). Because of the limited number of isolates of other species of CNS, no analysis of these data was performed. Analysis of plasmid profiles demonstrated no clear relation between the number of plasmids and the antibiogram or species of CNS for either the employee or patient isolates. The employee isolates contained an average of Table
II.
Rates
of resistance
(%)
among
isolates antibiotics
Isolates
obtained
of Staphylococcus
from
epidermidis
to seven
patients
at Antibiotic
Vancomycin Co-trimoxazole Penicillin Oxacillin Erythromycin Clindamycin Gentamicin
O(O) 18(14.6) 97(78.9) 34(27.6) 56(45.5) 49(39.8) 18(14.6)
* For employee t For employee
Table
of Hospital
Infection
(1992)
20, 247-255
Dynamics of coagulase-negative staphylococcal colonization in patients and employees in a surgical intensive care unit J. R. Thurn, Infectious
K. B. Crossley,
A. Gerdts
Disease Section, Department Center, 640-7ackson Street, Accepted
and
L. Baken
of Medicine, St. Paul-Ramsey St. Paul, Minnesota, USA
for publication
12 November
Medical
1991
Summary:
Because there is little information about the frequency of carriage of various species of coagulase-negative staphylococci (CNS) in hosnitalized natients and the nossible connection of these organisms with hospital staff, we prospectively investigated nasal CNS in patients and personnel in a Surgical Intensive Care Unit (SICU). The majority of CNS from personnel were Staphylococcus epidermidis. The CNS species from patients on admission were more diverse and included multiply antibioticresistant S. haemolvticus. Patients’ CNS became more like CNS colonizing personnel after admission with respect to both antimicrobial susceptibilit; and speciation. Plasmid and antibiotic sensitivity profiles of S. epidermidis resistant to multiple antibiotics from five patients were identified as those from one employee, but there was no evidence that this was of clinical significance. A varietv of factors influence nasal colonization bv CNS in SfCUs. The nasal CgS of patients change after admission and may become more resistant and less diverse. The factors influencing changes in the antibiotic susceptibility and the aetiology of CNS infection require further study. Keywords:
Coagulase-negative
staphylococci;
colonization;
nasal
carriage.
Introduction
Coagulase-negative staphylococci (CNS) have been increasingly recognized as causes of nosocomial infection and have become a leading cause of nosocomial bacteraemia.’ Sepsis caused by CNS has often been noted to involve organisms resistant to multiple antibiotics,* to prolong hospital stay and to increase mortality.3 Infections caused by CNS are frequently related to prosthetic devices4 and have been documented with increased frequency in compromised patients.5-7 The antimicrobial susceptibility of CNS has been shown to be species and vancomycin resistance has developed in species other dependents-” than Staphylococcus epidermidis.“,‘2 Many laboratories, however, have Correspondence Minneapolis 019556701/92/040247+
to: Joseph Veterans Affairs
R.
Thurn, Medical
M.D., Center,
Infectious Minneapolis,
09 $03.00/O
Disease MN
Section, Department 55417, USA. 0 1992 The Hospital
247
of Medicine,
Infection
Society
248
J. R. Thurn
et al.
identified these organisms as ‘Staphylococcus epidermidis’ and have not attempted to speciate them correctly. Because there is limited information available about the frequency of carriage of various CNS in hospitalized patients or personnel, and few studies have involved a single hospital unit or ward, we examined this in a Surgical Intensive Care Unit (SICU). In order to define better the epidemiology of ,CNS, antibiotic resistance patterns were determined and plasmid analysis was performed on the organisms isolated in this study.
Methods
and
materials
Employee nasal swabs Nasal cultures were self-collected using moistened Dacron swabs (Culturettes, Baxter Healthcare Corporation, McGaw Park, IL, USA) by employees in our SICU during the period between August 17 and August 20, 1987. Patient nasal swabs Patients in the SICU were studied in the period between August and October 1987. Cultures were done within 24 h of admission and again at 7 days, death or discharge from the SICU. They were collected by staff in a manner identical to the employee nasal swabs. Swabs were inoculated on to 5% sheep blood agar and mannitol salt agar within 4 h of collection and incubated for 36 h at 35°C. S. aUreUs was identified by tube coagulase testing or use of the API Staph-Ident strip (Analytab Products, Inc., Chicago, IL). Other staphylococci were speciated using the API Staph-Ident. Novobiocin resistance was used to differentiate S. hominis from S. saprophyticus. Isolates were tested for susceptibility to vancomycin, trimethoprim-sulphamethoxazole, penicillin, oxacillin, erythromycin, amikacin, clindamycin, cefazolin, gentamicin and tobramycin using the NCCLS standard disc diffusion method. Plasmid analysis Plasmid analysis employed a method similar to that of Birnboim and D01y.l~ Briefly, the bacterial cells were lysed with lysozyme and 1% sodium dodecyl sulphate in 1 M sodium hydroxide. This was followed by neutralization and precipitation of chromosomal DNA with sodium acetate and phenol extraction. The collected DNA was electrophoresed using 0.7% agarose gel with a duration of 16 min at 25 V. The resulting DNA bands were photographed with Polaroid film after treatment with ethidium bromide. Statistical analysis Differences in proportions were compared using Chi-square and Fisher’s exact tests when appropriate. A commercially available statistical package was utilized (A-stat, Epistat Services, Richardson, TX, USA).
Dynamics
of CNS in an SICU
249
Results
Nasal swab cultures were obtained from 9.5 health-care workers employed in the SICU. Cultures were obtained from employees on all shifts and included nurses, aides, physicians, trainees and ancillary personnel. Only one employee refused to participate. Cultures were obtaineed from the anterior nares of 125 patients within 24 h of admission to the SICU. During this time, there were 12 patients who were not cultured because of transfer, death or the inability to obtain informed consent. A second culture was obtained from 65 patients (52%) a mean of 3.2 days (range l-7 days) after the initial culture. A third culture was obtained from five patients a mean of 4.2 days (range l-8 days) after the second culture and an average of 10.8 days after admission to the SICU. There were 173 isolates of staphylococci from the cultures of employees, 35 of which were S. aureus. No methicillin-resistant S. aUYeUS (MRSA) were detected. Of the 138 coagulase-negative staphylococci (CNS) isolated from employees, the majority (123 out of 138, 89.1%) were S. epidermidis; the remainder included S. capitis (6), S. warneri (4), S. haemolyticus (3), S. hominis (1) and S. simulans (1) (Table I). Staphylococcus aureus was found to comprise a similar proportion of patient staphylococcal isolates (53 out of 359, 15%) as employees (35 out of 173, 20%; P=O.l4). As for employees, no patients were found to be colonized by MRSA. There was, however, a difference in the proportions of species of CNS isolated from the initial patient cultures (obtained less that 24 h after admission) when compared with the CNS isolated from employees (Table I). There were significantly fewer S. epidermidis (128 out of 215, 59.5% vs. 123 out of 138, 89.1%; P< 0.001); significantly more S. haemolyticus (50 out of 215, 23.3% vs. 3 out of 138, 2.2; P24h
% (A/)
and patients
--
< 0.001 < 0.036 < 0*002.f
P
Patients
employees
0.5(l) lOO(215)
from
59,5(128) 23.3 (SO) 6.5(14) 3.7(8) 3.7(S) 0.9(2) 0.9(2) 0.9 (2)
7 days
2 a rb
Y F
Dynamics
of CNS
in
an SICU
251
When comparing the antibiotic sensitivity profiles for S. epidermidis, the isolates obtained from the employees were significantly more often resistant to penicillin, erythromycin and clindamycin than the initial patient isolates. When compared to the second patient isolates, there was no difference in the proportion of employee isolates resistant to any of the antibiotics (Table II). There was no significant difference in the proportions of S. haemolyticus resistant to the antibiotics between the initial patient isolates and the employee or the second patient isolates (Table III). Because of the limited number of isolates of other species of CNS, no analysis of these data was performed. Analysis of plasmid profiles demonstrated no clear relation between the number of plasmids and the antibiogram or species of CNS for either the employee or patient isolates. The employee isolates contained an average of Table
II.
Rates
of resistance
(%)
among
isolates antibiotics
Isolates
obtained
of Staphylococcus
from
epidermidis
to seven
patients
at Antibiotic
Vancomycin Co-trimoxazole Penicillin Oxacillin Erythromycin Clindamycin Gentamicin
O(O) 18(14.6) 97(78.9) 34(27.6) 56(45.5) 49(39.8) 18(14.6)
* For employee t For employee
Table
