Letters
Infectious diseases
Volume 197 - Issue 3

Testing the golden rules of infectious diseases: staphylococcal bacteriuria is predictive of staphylococcal bacteraemia

Authors:  Janine M Trevillyan and Denis Spelman

Med J Aust 2012; 197 (3): 149-150. || doi: 10.5694/mja12.10610
Published online: 6 August 2012
To the Editor: Medical students around the world are taught that staphylococcal bacteriuria may be indicative of concurrent bacteraemia.1 Yet there is little published evidence to support this “golden rule” of infectious diseases, and a retrospective review has suggested that no such relationship exists.2 To determine the relationship between Staphylococcus aureus (SA) bacteriuria and bacteraemia, we performed a retrospective review of microbiology results at the Alfred ...
Download PDF (201 KB)
Share: Share on Blue Sky Share on Facebook Share on LinkedIn Share by email

To the Editor: Medical students around the world are taught that staphylococcal bacteriuria may be indicative of concurrent bacteraemia.1 Yet there is little published evidence to support this “golden rule” of infectious diseases, and a retrospective review has suggested that no such relationship exists.2

To determine the relationship between Staphylococcus aureus (SA) bacteriuria and bacteraemia, we performed a retrospective review of microbiology results at the Alfred Hospital from 1 January 2000 to 31 December 2010. We included urine samples in which SA had been isolated, collected from 776 inpatients and outpatients over this period. The controls were 1003 patients with urine samples in which coagulase-negative Staphylococcus (CNS) had been identified. Samples were then cross-referenced with blood culture isolates. Catheter specimens were excluded because of the potential for contamination.

Eight per cent of patients (64 patients) with SA bacteriuria had concurrent staphylococcal bacteraemia. None of the patients with CNS bacteriuria had blood cultures that tested positive for SA. Only 48% of patients (374 patients) had undergone blood culture sampling within 2 months of the positive urine sample. Thus, 17% of patients with SA bacteriuria who had undergone blood culture sampling had bacteraemia. Methicillin-sensitive isolates were more likely to be associated with bacteraemia than methicillin-resistant isolates. Of the 339 patients with methicillin-sensitive SA in their urine, 12% (41 patients) had bacteraemia. By comparison, of the 437 patients with methicillin-resistant SA, only 5% (23 patients) had bacteraemia. Positive results were observed in urine culture on average 2.75 days earlier than in blood culture (median, 0 days; interquartile range, − 3 to 0), reflecting our hospital’s practice of recommending that blood cultures be taken when SA bacteriuria is identified.

While SA bacteriuria may be seen during isolated urinary tract infection, it also occurs through haematogenous seeding during bacteraemia, and may be indicative of complicated bacteraemia (bacteraemia with a deep focus, such as infective endocarditis or osteomyelitis).3,4 Animal models provide support for this observation, having shown that intravenous injection of SA results in the formation of renal microabscesses (which in turn leads to bacteriuria) and that the frequency with which abscesses develop is a function of the number of organisms injected.5 Our study supports the golden rule of infectious diseases, and we recommend that all patients with SA isolated from a non-catheter urine sample have blood cultures taken. If the results of these are positive, further investigation should be undertaken to rule out complicated disease.

Authors

  • Janine M Trevillyan1
  • Denis Spelman1
  • 1. Alfred Hospital, Melbourne, VIC.

Correspondence: j.trevillyan@alfred.org.au

Competing interests

No relevant disclosures.

References

View this article on Wiley Online Library