Incidence and outcome data on Staphylococcus aureus bacteraemia (SAB) in Australia and New Zealand are currently very limited. In 2007, the Australian Society for Antimicrobials established a co-operative in these two countries to prospectively examine the outcomes of patients diagnosed with SAB in secondary and tertiary hospitals. An agreed set of data fields addressing demographic, treatment and outcome variables was established by consensus, and a web-based data entry form created for facilitate real-time data collection. The key outcomes were 7- and 30-day all-cause mortality. Most participants commenced data collection by June 2007.
By May 6 2008, 2008 cases had been entered by 27 hospital laboratory services and had completed 30 day follow-up. 134 of these were from New Zealand. 64% of cases were male, and 15% were ≤ 30 years old. Aboriginal/TSI groups made up 3.1% of cases, while 4.0% were of Maori or Pacific Islander origin. In 60.1% of cases SAB had its onset in the community; and 18.0% of these were MRSA. By contrast MRSA was seen in 34% of hospital-onset infections. MRSA collectively caused 24.2% of cases; 14.8% were multi-resistant, 6.6% were of the non-multi-resistant community-associated type and 2.8% were EMRSA-15-like. 10.1% of SAB cases were caused by penicillin-susceptible strains.
Infections associated with indwelling devices were the commonest recorded clinical manifestation (20.2%), followed closely by skin and skin structure infections, bone and joint infections, and bacteraemia without a focus. Endocarditis comprised 6% of cases and pneumonia/empyema 8%, while sepsis syndrome was the principal clinical manifestation in 11%.
The 7-day all-cause mortality was 10.9% and the 30-day mortality was 20.7% (n=1879). On univariate analysis 30-day mortality was significantly higher if the patient was > 70 years old (p<0.0001), was of European (Caucasian) origin (p=0.015), had sepsis syndrome (p<0.005), had pneumonia/empyema (p<0.0001), if the infection was not device-related (p=0.004), if the infection was caused by MRSA (p=0.0001) as opposed to strains with other susceptibility profiles, and if the patient was treated mainly with vancomycin (p=0.0001, 26% mortality versus 18.7%), even for infections caused by methicillin-susceptible strains (p=0.002). On multivariate analysis the independent predictors of increased mortality remaining were age, sepsis, pneumonia, MRSA and treatment with vancomycin.
In conclusion, SAB is a common and serious problem in Australia and New Zealand. The number of deaths associated with SAB arising in the community is almost certainly substantially higher than with other invasive bacterial infections traditionally considered of greater public health importance.