Three-year study of targeted screening for methicillin-resistant Staphylococcus aureus at hospital admission.

Abstract

Listen

Translate article

The problem of the spread of methicillin-resistant Staphylococcus aureus (MRSA) during hospital transfers, re-admission and admission from nursing homes is well documented. Since MRSA carriage can be occult, only to be detected later in clinical specimens, screening at hospital admission has been advocated for areas in which MRSA is endemic [1]. In a previous study Troillet et al. [2] found that without admission screening half of the MRSA carriers whose status was unknown at admission would have been missed or detected late. Compared with no screening, targeted screening has been shown to be effective in reducing the rates of MRSA acquisition in hospitals with a high level of endemic MRSA [3]. In Hong Kong, MRSA has been endemic since the mid-1980s [4]. Currently, screening for MRSA at admission is only practiced in a few selected units. The present report details the experience gained at an acutecare hospital in Hong Kong (Northern District Hospital) where screening for MRSA at admission is practiced routinely. The data was obtained by retrospectively reviewing the culture results in the hospital’s mainframe computer and the infection control records over a 3-year period (1 January 1999–31 December 2001). The 600bed hospital has all of the major specialty services, including an intensive care unit and a 24-h emergency department. The hospital’s screening policy called for cultures to be performed within 24 h for all patients with one or more of the following risk factors: a history of MRSA carriage, nursing home residence or any previous hospitalization (including inter-hospital transfer) during the preceding 6 months. Patients were grouped together and isolation precautions were implemented until the results of screening were available [1]. Screening cultures were also performed before the start of hospital duty for all new healthcare workers whose jobs involved close patient contact. During the study period, compliance with the MRSA screening policy was monitored regularly by ward visits and chart review, and it was estimated to exceed 90%. In order to obtain a background rate of MRSA carriage, 600 patients without the above risk factors for MRSA were prospectively screened at the emergency department for 3 months (August–November 2000) using the same protocol. An average of 10 patients per day was screened. To facilitate data retrieval and auditing, codes were entered into the laboratory records of all screened patients and healthcare workers. Initial and follow-up samples were coded separately. Individual risk factors were not coded separately until 1 June 2001. Hence, a breakdown of the MRSA isolation rate according to individual risk factors was available only for the period 1 June–31 December 2001. As part of the infection control policy, all patients and staff members who tested positive for MRSA during screening were treated topically for 5 days with a daily wash or bath using 4% chlorhexidine gluconate (Hibiscrub; Zeneca Pharma, France) and the twice-daily application of 2% mupirocin (Bactroban nasal; GlaxoSmithKline, Hong Kong) to both nares. Follow-up MRSA cultures were performed weekly for 3 weeks following T. L. Que · K. T. Yip · H. L. Ng New Territory North Pathology Service, Northern District Hospital, Hong Kong SAR, People’s Republic of China