Physicians frequently order repeat blood cultures for hospitalized patients on antibiotic therapy who have a persistent fever. The rationale for this step is to identify any new pathogen that may have arisen or one that was not identified on the initial blood culture performed before starting antibiotic therapy. Grace and colleagues present the results of a retrospective analysis of repeat blood cultures obtained from patients within 72 hours of antibiotic administration at a tertiary-care, university-based hospital.

Charts were reviewed for 1,446 patients with positive blood cultures, of which 139 (9 percent) qualified for analysis because pre-antibiotic and repeat blood cultures were available. Neutropenic patients, those in the intensive care unit, and any patients with antibiotic use before hospitalization were excluded.

In the 83 patients (60 percent) with initially negative blood cultures, none of the repeat blood cultures revealed any pathogen. The remaining 56 patients with initially positive blood cultures had varying likelihoods of positive repeat blood cultures, depending on the initial pathogen identified. In all patients except one, the pathogen seen on repeat culture was identical to the one present in the initial pre-antibiotic culture. The likelihood of a repeat positive blood culture was not increased with advanced age, higher fever, or higher white blood cell count. The strongest predictors of a repeat positive blood culture were identification of Staphylococcus aureus on the initial culture or presence of endocarditis.

The authors conclude that repeating blood cultures within 72 hours of administering antibiotic therapy is highly unlikely to isolate any new or missed pathogen. They suggest that physicians wait for the results of the initial pre-antibiotic blood culture rather than order repeat blood cultures in response to a persistent fever.