Follow-up blood cultures in Gram-negative bacteremia: How do they impact outcomes?

Abstract

Background: As opposed to Staphylococcus. aureus bacteremia, there are no guidelines to recommend repeating blood cultures in Gram-negative bacilli bacteremia (GNB). Several studies have questioned the utility of follow-up blood cultures (FUBCs) in GNB, but the impact of this practice on clinical outcomes is not fully understood. Our aim was to study the practice of obtaining FUBCs in GNB at our institution and to assess it's impact on clinical outcomes. Method(s): We conducted a retrospective, single-center study of adult patients, >= 18 years of age admitted with GNB between January 2017 and December 2018. We aimed to compare clinical outcomes in those with and without FUBCs. Data collected included demographics, comorbidities, presumed source of bacteremia and need for intensive care unit (ICU) admission. Presence of fever, hypotension /shock and white blood cell (WBC) count on the day of FUBC was recorded. The primary objective was to compare 30-day mortality between the two groups. Secondary objectives were to compare differences in 30-day readmission rate, hospital length of stay (LOS) and duration of antibiotic treatment. Mean and standard deviation were used for continuous variables, frequency and proportion were used for categorical variables. P-value < 0.05 was defined as statistically significant. Result(s): 482 patients were included, and of these, 321 (67%) had FUBCs. 96% of FUBCs were negative and 2.8% had persistent bacteremia. There was no significant difference in 30-day mortality between those with and without FUBCs (2.9% and 2.7% respectively), or in 30-day readmission rate (21.4% and 23.4% respectively). In patients with FUBCs compared to those without FUBCs, hospital LOS was longer (7 days vs 5 days, P < 0.001), and mean duration of antibiotic treatment was longer (14 days vs 11 days, P < 0.001). A higher number of patients with FUBCs needed ICU care compared to those without FUBCs (41.4% and 25.5% respectively, P < 0.001) Conclusion(s): Obtaining FUBCs in GNB had no impact on 30-day mortality or 30-day readmission rate. It was associated with longer LOS and antibiotic duration. Our findings suggest that FUBCs in GNB are low yield and may not be recommended in all patients. Prospective studies are needed to further examine the utility of this practice in GNB. (Table Presented).