11.7% for cumulative incidence (43,60) and between 7.9 and 9.9 per 1000 patient days for
incidence density (61,62). In one study, it was observed that HA-MRSA was acquired at about
1% per day in the first week after admission and then at 3% per day thereafter (45). In a more
recent study the risk per day for acquisition of MRSA was less than 1% at ICU admission and
then was greater than 2% by day 12 and then leveled out (63).
Sources of HA-MRSA.The sources of HA-MRSA include colonized or infected patients,
colonized or infected health care workers (HCWs), and contaminated environmental surfaces.
One of the best indications of the importance of colonized and infected patients as an
important source of HA-MRSA is the significant relationship between colonization pressure
and acquisition of HA-MRSA colonization, or infection by patients who have no colonization
or infection due to HA-MRSA at the time of admission to an ICU (60). Colonization pressure is
defined as the number of patient days for patients with cultures positive for HA-MRSA
divided by the number of total patient days (64). It can be calculated for any day or for a given
period of time. The most common site of MRSA colonization in adults is the external nares
(42,65,66). The second most common site of colonization is skin and soft tissue other than
surgical sites (34%) (65). Other sites of colonization include rectal (11% to 28.9%), respiratory
tract (11%), and urinary tract (6%) (42,65,66).
Another source of HA-MRSA is colonized or infected health care personnel. Acquisition
of HA-MRSA in an ICU from a respiratory therapist with chronic sinusitis due to HA-MRSA
has been reported, as well as surgical site infections due to colonization of the external nares
and an area of dermatitis on the hand of a surgeon (67,68). The surgical site infections caused
by the colonized surgeon were initiated at the time of surgery but became manifest
postoperatively in the ICU. HCWs often become colonized with HA-MRSA from contacts with
patients when providing health care but are not often implicated in transmission to patients.
To implicate a colonized HCW as a source for colonization or infection of patients, it is
first necessary to epidemiologically establish an association between contact with the colonized
or infected HCW and acquisition of HA-MRSA by patients. Then it is necessary to prove that
the strain from the HCW and the patient is the same using molecular techniques such as
pulsed-field gel electrophoresis (PFGE) after restriction endonuclease digestion of genomic
DNA.
Contaminated surfaces of equipment and environmental surfaces appear to make up
another source of HA-MRSA for transmission to patients (69,70). HA-MRSA has been
recovered from cultures of computer terminals, the floor next to the patient’s bed, bed linens,
patient gowns, over-bed tables, blood pressure cuffs, bedside rails, infusion pump buttons,
door handles, bedside commodes, stethoscopes, and window sills. In the latter study, 27% of
350 environmental surface cultures yielded HA-MRSA (70). It has also been shown in in vitro
studies that outbreak isolates of HA-MRSA survive at significantly higher concentrations and
for longer periods of time on an inanimate surface than do sporadic HA-MRSA isolates (71).
Thus, environmental contamination is likely another important source for transmission of
HA-MRSA to patients.
Mode of transmission of HA-MRSA.The most common mode of transmission of HA-MRSA to
patients is by indirect contact. Several studies have shown that HA-MRSA is frequently
transmitted to the hands and clothing of HCWs from colonized or infected patients. Two
studies have shown that HA-MRSA can be recovered from 14% to 17% of HCWs’ hands after
patient contact (72,73). Another study showed that 7 out of 12 (58%) nurses who cared for
patients with HA-MRSA in a wound or urine had HA-MRSA on their gloves, recoverable by
direct plating to solid media (70). Culture of 13 of 20 (65%) nurses’ uniforms or gowns who
cared for these same patients yielded HA-MRSA. When cultures were taken from gloves of 12
personnel who touched only environmental surfaces in the rooms of these patients, five (42%)
had HA-MRSA recovered on culture. Arbitrary-primed polymerase chain reaction (PCR)
typing demonstrated that isolates recovered from patients and environment had very similar
banding patterns (70). Although additional studies are needed, data continue to accumulate in
support of indirect transfer of HA-MRSA to patients from contaminated hands and clothing
of HCWs.
MRSA/VRE Colonization and Infection in the Critical Care Unit 105