Chlorhexidine Bath Admission Protocol For Prevention of Healthcare-associated Methicillin-Resistant Staphylococcus aureus Infection: Results of a Hospital Wide Initiative

Background: Healthcare-associated methicillin-resistant Staphylococcus aureus (HA-MRSA) infections are estimated to occur in over 1.7 million patients each year, with over 100,000 fatalities. While randomized controlled trials (RCTs) support the use of local chlorhexidine (CHG) for preventing infection following placement of intravascular catheters, the use of CHG baths pre-operatively or in ICU patients has had limited and variable support from mostly non-randomized cohort studies and several small RCTs. The CHG bath is currently being applied to all adult patients on admission to our hospital with the goal of reducing the incidence of HA-MRSA infection to less than 0.1%, a national safety standard.  

Objective: This study evaluates the effectiveness of the CHG bathing protocol at reducing the incidence of HA-MRSA infection.

Methods: A retrospective cohort analysis of HA-MRSA infection was performed in a community teaching hospital before and after the December 2008 CHG protocol implementation. HA-MRSA infection was defined by CDC criteria and positive pertinent cultures. Patients admitted between June and November 2008 (pre-implementation) were compared to those admitted between February and July 2009 (post-implementation). Patients admitted during the implementation stage from December 2008 to January 2009 are omitted from this study. This was an intent-to-treat analysis.

Results: Forty-five patients out of nearly 23,000 admissions developed HA-MRSA infection during the study period. In the pre-implementation group, 20 patients out of 11,373 admissions (0.18%), developed HA-MRSA infection; this compares to 25 out of 11,599 (0.22%) after full implementation of the CHG protocol (p<0.49). The groups were similar in regards to age, ethnicity, length of stay preceding infection, and type of HA-MRSA infection acquired. The proportion of male patients with HA-MRSA infection increased from pre-implementation to post-implementation (40% to 64%), although this difference is not statistically significant (p<0.11). Adherence to the CHG bathing protocol was a considerable issue; a random chart review of 308 patients showed only 52% documentation of adherence (95% CI, 47% to 57%).

Conclusions: Our analysis suggests that a hospital-wide protocol to administer full body CHG baths to all patients on admission did not reduce HA-MRSA infection. Poor adherence limits the interpretation of the efficacy of CHG bathing but needs to be considered given the time involved and its inconvenience to the patient. In order for the protocol to have had a statistically significant (p<0.05) effect on the rate of HA-MRSA infection, the CHG bath would have had to reduce the number of infections by 50%. A larger study incorporating RCTs and methods to improve adherence is needed to determine the efficacy of the CHG baths.