and the simultaneous administration of other oral antimicrobial agents. The strengths included
eradication of gastrointestinal carriage of MRSA and the careful monitoring of vancomycin
resistance in MRSA and enterococci. No resistance was detected in many isolates of MRSA and
enterococci tested for vancomycin susceptibility during the study. The authors also noted that by
eradicating rectal carriage with vancomycin and preventing infection, they administered only
25% as much vancomycin to the group given oral vancomycin prophylaxis as was needed to treat
the infections in the control group. A four-year prospective observational study in a pediatric
intensive care unit in which cultures for MRSA were taken from throat and rectum of patients on
admission and then twice weekly thereafter assessed the effect of enteral vancomycin on the
prevention of primary and secondary endogenous infections due to MRSA (111). Patients with
colonization or infection were treated for five days with enteral vancomycin. The MRSA carrier
state was eradicated in 11 (79%) of 14 patients in a median of six days (IQR 3.5–9.75). Five primary
endogenous infections occurred in patients who came into the hospital colonized with MRSA.
There were no secondary endogenous infections in patients who acquired MRSA in the hospital.
Over the four-year period of the study, no VRE or vancomycin-intermediateS. aureus(VISA) was
isolated from 1611 cultures taken from infected and colonized sites in these patients (111).
Another approach to decolonize patients who are MRSA carriers is application of a topical
antimicrobial agent to the nares and total body bathing with chlorhexidine (112,113). Both of the
latter studies observed a significant decrease in infections due to MRSA.
Decolonization of patients in NICUs is similar to that in adult ICUs but has not been as
well studied. In one report of an MRSA outbreak, four patients were treated with nasal
mupirocin three times a day for five days and bathed with diluted (1:10) 4% chlorhexidine
gluconate once daily for three days (114). Two of the four neonates were successfully
decolonized and two remained colonized with MRSA. The latter two were decolonized after
the regimen was repeated. In a report of a second outbreak, colonized neonates were treated
with mupirocin twice daily to the anterior nares and the umbilical area for seven days (115).
The authors did not report the results of their decolonization regimen.
In an account of an MRSA outbreak in an NICU, one control measure was application of
triple dye to the umbilical area of the patients (47). This was one of several control measures
implemented. Other control measures instituted included reducing overcrowding and
understaffing, and placing an infection control nurse in the NICU. Because all of these control
measures were implemented at the same time, it was not possible to determine what effect the
triple dye had in controlling the outbreak.
Decolonization of HCWs Who Are MRSA Carriers
Decolonization of HCWs is necessary when they have been epidemiologically implicated in
the transmission of MRSA to patients from a colonized body site, which is most often the nose.
Eradication of MRSA carriage from HCWs has been shown to help control outbreaks (68). For
MRSA, mupirocin will decolonize the external nares effectively 91% of the time, although
recolonization may occur in about one quarter of treated individuals within four weeks (116). It
has also been shown that decolonization of HCWs with nasal carriage of MSSA results in a
substantial decrease in hand carriage (117). Temporary decolonization of most of the colonized
HCWs in an ICU for a few weeks may help control an outbreak. Although there are few data
on decolonization of HCWs carrying MRSA, it is likely that mupirocin will eradicate MRSA
from the nares of HCWs.
A second area where HCWs may be colonized with MRSA is at the site of dermatitis on
their hands or forearms. It is important that hands and forearms of HCWs be examined and
areas of dermatitis be cultured during an outbreak investigation. Other sites of colonization or
infection are less common but may have to be sought if epidemiologically indicated. Table 3
lists the control measures for MRSA in ICUs.
Cost Effectiveness of MRSA Control
One study of the cost-effectiveness of MRSA control in a medical intensive care unit (MICU)
has concluded that identification of patients who are carriers of MRSA on admission and
during hospitalization and isolating of these carriers is cost-effective (41). In spite of an
110 Mayhall