Follow-up blood cultures in Gram-negative bacilli bacteremia: are they needed for critically ill patients?

Abstract

Gram-negative bacilli bacteremias (GNB-Bs) represent a major cause of morbidity and mortality in Intensive Care Unit (ICU) patients. Aim of this study was to investigate the role of follow-up blood cultures (FUBCs) and the clinical significance of persistent bacteremia (PB) in these settings. We retrospectively analyzed clinical data and outcome of GNB-Bs that occurred in ICU patients over a span of 1 year. In particular we sought information on development and clinical details of PB, defined as repeatedly positive FUBCs after ≥96 hours of appropriate antibiotic treatment and ≥48 hours after removal of endovascular devices. Among 307 ICU patients, 69 (22.4%) developed 107 GNB-Bs. Of these, 78 (73%) could be eventually analyzed: 50 of 78 (64.1%) were non-PBs from 26 patients and 28 of 78 (35.9%) were PBs from 23 patients. Duration of fever and bacteremia, time to procalcitonin normalization and weaning from vasopressors were longer in episodes of PBs than non-PBs (P=0.04, P<0.001, P=0.02 and P=0.004, respectively). Primary bacteremia was more frequent in non-PBs than in PBs (29 of 50, 58% vs. 3/28, 10.7%, P=0.0001), whereas septic thrombus infection (STI) was the source of infection in 14 of 28 (50%). Finally, clinical features and 30-day mortality did not differ between patients with PB and those who developed only non-PB episodes. Among our ICU patients, more than one third of GNB-Bs for which FUBCs were performed resulted PB. This condition is often associated with the presence of STI; therefore, FUBCs seem useful for the optimal management of GNB in this clinical setting.