Many physicians rely on differential time to positivity (DTP) when diagnosing catheter-related bloodstream infection (CRBSI). We evaluated whether DTP from routine blood cultures can predict catheter-related Staphylococcus aureus bloodstream infection.

From 2006 to 2011 adult patients with monomicrobial S. aureus bloodstream infection and matched pairs of central and peripheral blood cultures were prospectively followed. CRBSI was defined by the absence of other infective foci and recovery of S. aureus from the catheter tip or catheter exit-site, or local signs of infection at the catheter exit site. A DTP of more than two hours (cut-off) was used to define test positivity.

CRBSI was present in 30 (34%, prevalence) of 87 patients. In 24 (28%) patients a DTP of more than two hours was measured: eleven patients had CRBSI, ten deep-seated infections, and in three patients an infective focus was not identified. DTP showed a positive predictive value of 0.46 [95% confidence interval (CI) 0.28 to 0.65], and a negative predictive value of 0.70 [95% CI 0.58 to 0.80].

The low test performance suggests that DTP is not useful in diagnosing CRBSI in routinely obtained blood cultures. Therefore, physicians should not solely rely on DTP and rather promote catheter removal and culture.