Literature Update - December, 2007
Compiled by Dr Leon Worth, ID Physician
Association between Hospital Exposure and Bacteraemia: UK Experience
Wyllie DH et al (J Hosp Infect 2007; 67:301-307) estimated the impact of hospital admission within the previous 12 months on risk for developing
bacteraemia due to S. aureus, S. pneumoniae or E. coli.
Population-based anonymised data were used to determine microbiological isolates and hospital contacts during the period April 1999 to March 2005
for 4 primary care trusts in the UK.
Overall, about 8% of the total studied population had hospital contact during the previous 12 month period. This group of ‘hospital exposed’
individuals accounted for 54% of all hospital admissions, highest in specialty units characterised by regular attendance
(e.g. >90% for nephrology and haematology/oncology units). Between 15-33% of patients with bacteraemia died within 30 days. Rates of bacteraemia
were higher in the hospital-exposed, compared to the hospital-naïve population. Incidence rate ratios (hospital-exposed relative to hospital-naïve)
were 43 for MRSA, 20 for S. aureus other than MRSA, 7.3 for S. pneumoniae, and 14 for
E. coli. The rates of bacteraemia in the hospital-exposed population were approximately 10-fold greater
than hospital-naïve population.
This study confirms the fact that healthcare exposure is a risk factor for bacteraemia due to particular organisms. For S. aureus,
this finding is not surprising, but it is interesting that this relationship was also observed for E.coli and
S. pneumoniae bloodstream infections, where an association with healthcare-acquisition has not been so clearly
demonstrated. The study is limited by the fact that an insensitive definition of healthcare exposure was used. Also, to better understand
healthcare-associated infections, it would be advantageous for healthcare-exposed subpopulations to be studied (e.g. ambulatory-care, dialysis,
same-day admissions), to allow ongoing study of prevention and risk minimisation strategies in these groups.
Infrequent Serious MRSA Infections in Smaller Victorian Hospitals
Bennett NJ et al (Am J Infect Control 2007; 35:697-699) reported VICNISS surveillance findings for methicillin-resistant
S. aureus (MRSA) in 85 smaller hospitals (< 100 acute care beds) across the state of Victoria between
May 2004 and December 2005.
MRSA infection was defined as an event associated with a sterile site isolate or an event associated with a nonsterile site isolate during
which MRSA-specific antibiotic therapy was administered by a clinician. Rates were expressed per 10,000 acute care occupied bed days (OBD).
MRSA colonisations were excluded from analysis, given the varied intensity of screening undertaken at participating hospitals.
Overall, the statewide aggregate rate for all MRSA infections was 1.5 per 10,000 OBDs. Small (1-14 acute care beds), medium (15-49 acute care beds)
and large (50-99 acute care beds) hospitals had similar rates of infection. Approximately one third of infections were detected >48 hours
after admission or previously detected at the reporting hospital. The remaining infections were acquired in the community, or were ‘inherited’
from another healthcare facility.
Although direct comparison with similar hospital types is not possible, crude comparisons with other Australian states suggest that these
Victorian rates are relatively low.