JSTOR

Methods

The investigation occurred in an outpatient clinic for human immunodeficiency virus (HIV) infection and infectious diseases, located within a tertiary care hospital, that serves the urban Baltimore population and has 20,000 patient visits annually. Each HCW who provides patient care is assigned a room and is responsible for cleaning the room between patient visits, disposing of used gowns, and pulling a new length of paper across the examination table. No additional cleaning occurs between patient visits, although the clinic is cleaned daily. On the corresponding inpatient HIV service, active surveillance for MRSA is performed for all admitted patients by means of nares culture, and the prevalence of MRSA colonization and/or infection among patients was 33% for the quarter year that this investigation was conducted (August through November, 2004). This surveillance is not performed in the outpatient‐care setting.

To determine whether there were additional cases of MRSA colonization and/or infection, each HCW who works in the clinic was asked to complete a survey. Information about any current or past history of MRSA infection was asked. One‐time surveillance cultures to detect MRSA colonization were performed for HCWs at the clinic; specimens for culture were obtained from the anterior nares and hands. In addition, environmental cultures were performed, targeting surfaces that are frequently touched, such as patient chairs, examination tables, computer keyboards, door knobs, telephones, and medical equipment.

Specimens were cultured for detection of Staphylococcus aureus, and pulsed‐field gel electrophoresis (PFGE), PVL testing, and SCCmec type testing were performed as described elsewhere.⁹ MRSA isolates from patient 2, from 2 other HCWs identified by screening (HCWs 1 and 2), and from environmental cultures were characterized by strain typing with PFGE, determination of antibiotic susceptibility, detection of PVL, and SCCmec typing by PCR. The isolate from patient 1 was not available for analysis. The investigation was approved by the Johns Hopkins Hospital Institutional Review Board.

Results

Two healthcare workers (patients 1 and 2) developed skin and soft‐tissue infections due to MRSA during a 6‐week period. Patient 1 reported caring for 15 patients with CA‐MRSA infections during the preceding several months and reported incising and draining many of these patients' abscesses. Patient 1 developed furuncles on the chest, and cultures of the lesions grew MRSA. This HCW was treated successfully with clindamycin. Patient 2 works in an administrative position without direct patient contact. This HCW developed a buttock abscess that required incision and drainage. Patient 2 received clindamycin followed by trimethoprim‐sulfamethoxazole with resolution of symptoms.

A total of 138 HCWs are employed in the clinic, of whom 58 perform administrative duties and 80 have clinical responsibilities. One hundred eight HCWs (78.3%) returned a survey. No additional cases of CA‐MRSA infection were identified during case finding. On screening, 2 (3%) of 70 HCWs who provided patient care (HCW 1 and HCW 2) and 0 (0%) of 58 support staff were colonized with MRSA in both anterior nares and on the hands. One additional HCW had a culture of hand specimens that grew MRSA, but additional cultures of hand specimens were negative for MRSA. Neither patient 1 nor patient 2 was colonized with MRSA in their anterior nares.

Among survey respondents, 70.6% of patient‐care providers and 77.5% of support staff were female. Patient‐care providers reported having seen a median of 2 patients (range, 0‐20 patients) with MRSA infections in the past 2 months (53 patient‐care providers responded to this question). Thirty‐one (29%) of 108 HCWs reported having taken antibiotics in the past 6 months. Only 1 HCW reported having a household contact with a person who had a diagnosis of MRSA skin infection, though 1 additional HCW reported having a cat with MRSA skin infection. No additional HCWs reported having an infection due to MRSA. Neither patient 1 nor patient 2 had recently been hospitalized nor had household contacts with MRSA infections. Family members were not screened during the course of this investigation.

Seven (19%) of the 36 environmental cultures grew MRSA. Sites from which MRSA was cultured included a patient examination table, a computer keyboard, a pulse oximeter, and multiple patient chairs (located in the triage station, the waiting room, and the examination room). Other sites sampled that did not yield S. aureus on culture included door knobs, telephones, ophthalmoscopes, otoscopes, blood‐pressure cuffs, scales, thermometers, and sinks. Interestingly, no methicillin‐susceptible S. aureus strains were isolated from the clinic environment.

The antibiotic susceptibility profiles and the PVL and SCCmec‐IV testing results for the MRSA isolates are summarized in the Table. The MRSA isolate from patient 1 was not available for molecular analysis, though it was determined to be susceptible to clindamycin, tetracycline, and trimethoprim‐sulfamethoxazole by a commercial laboratory. The 7 environmental isolates of MRSA had variable susceptibility patterns. The MRSA isolates from patient 2 and the environment all contained the SCCmec‐IV element and were positive for the PVL genes, findings consistent with their being CA‐MRSA strains. Neither of the isolates from the asymptomatic carriers identified during screening (HCW 1 and HCW 2) produced PVL, though the isolate from HCW 1 carried the SCCmec‐IV gene.

Table  Comparison of the Antibiotic Susceptibility Profiles and Molecular Testing Results for Methicillin‐Resistant Staphylococcus aureus Isolates Recovered From Healthcare Workers (HCWs) and Environmental Samples in the Study

Open New Window

Isolates from patient 2 and the 7 isolates from environmental cultures were determined to be the same strain or related strains by PFGE (Figure). PFGE demonstrated that the isolates from the 2 asymptomatic carriers were different from the isolates from patient 2 and the environment and also that they were unrelated to each other, findings consistent with their being HA‐MRSA strains.

(67 KB)

Figure.  Pulsed‐field gel electrophoresis patterns of methicillin‐resistant Staphylococcus aureus (MRSA) isolates from study patients. Lanes 3, 4, 5 and 7 represent the MRSA isolates recovered from an examination table, chair A, a keyboard, and chair C, respectively, and are the same strain type, designated strain I. Lane 2 represents the isolate from patient 2; this isolate is related to strain I. Lanes 6 and 9 represent the isolates recovered from chair B and a oxygen saturation probe, respectively, which are the same strain type and are related to strain I and the isolate from patient 2. Lanes 10 and 11 represent the isolates recovered on screening from the nose and hands of healthcare worker (HCW) 1, respectively, and are the same strain type, which is not related to the strain types of the other isolates in this gel image. Lanes 12 and 13 represent the isolates recovered on screening from the nose and hands of HCW 2, respectively, and are the same strain type, which is not related to the strain types of the other isolates in this gel image. Lane 14 represents the isolate from the HCW transiently colonized with MRSA on the hands only, and this isolate is not related to the other isolates. Lanes 1, 8, and 15 represent molecular markers.

Open New Window

The hospital's environmental services department was notified of the contamination of surfaces with MRSA. The clinic was cleaned thoroughly, with particular attention to surfaces, using a quaternary ammonium disinfectant. Disinfectant wipes (containing ethyl alcohol and quaternary ammonium) were placed in each examination room, with recommendations to clean surfaces such as the examination table, patient chairs, and when appropriate, computer keyboards, telephones, and doorknobs. Use of keyboard covers was also recommended, to facilitate routine cleaning of keyboards. Additional dispensers of alcohol‐based hand disinfectant were installed in multiple locations throughout the clinic.

Both HCWs who were colonized with MRSA (HCW 1 and HCW 2) were contacted and treated with intranasal mupirocin 2% ointment (applied twice per day for 5 days), and results of follow‐up nasal cultures were negative for MRSA. All environmental cultures performed after extensive cleaning were negative for MRSA.

Discussion

The mechanisms of transmission of CA‐MRSA strains in hospitals has not been elucidated in outbreaks that have been reported to date. Our data show that the environment can be contaminated with toxin‐producing strains, and this is, to our knowledge, the first report of extensive environmental contamination of a hospital with CA‐MRSA strains. It also suggests that prevention and control measures must include cleaning of the hospital environment. It is unknown how long CA‐MRSA strains can survive on fomites, though we speculate it is at least several days, given the extent of the contamination found within the clinic. Thorough disinfection of the clinic environment on at least a daily basis is critical to prevent spread of MRSA.

We suspect the infections with CA‐MRSA strains that occurred in patients 1 and 2 were the result of occupational exposure. Outbreaks of infection with CA‐MRSA strains have been reported in healthcare settings, with probable nosocomial transmission from healthcare worker to patient.^10,11 As more patients infected with CA‐MRSA strains seek medical care in clinics and hospitals, HCWs will have increased exposure to this organism.

Currently at our hospital, inpatients on high‐risk floors are screened for MRSA colonization. However, there is no surveillance of outpatients. In addition, inpatients are placed under contact isolation precautions if they have a history of MRSA infection or colonization. In the outpatient setting, no special precautions are taken. These issues are particularly relevant in outpatient clinics with high volumes of patients with CA‐MRSA infection, such as emergency departments, urgent care clinics, and other primary care clinics. Further research is needed to assess the impact of CA‐MRSA strains on healthcare institutions and to guide infection control recommendations for outpatient settings.