Abstract

Background: The proper duration for treating patients with non-fermenting gram-negative bacteremia is not yet defined; we attempt to find an appropriate course of treatment.

Methods: A retrospective multicenter study in three hospitals, Amman-Jordan. Medical records were reviewed for patients with Lactose Non-Fermenting Gram-Negative (LNF) bacteremia. Information on blood cultures was extracted from the microbiology logbook and records. For adults >18 years, primary bacteremia and a known source were included. Patients who needed prolonged antibiotics treatment due to the nature of their infections; neutropenic cancer patients, organs with abscesses/empyema, CVC retention, polymicrobial septicemia, and expected survival ≤ 48 hours. Continuous variables were analyzed by (χ²), Mann-Whitney test, ANOVA for means, and the Bonferroni for pairwise comparisons for P-value <0.05. SPSS version-25 was used in the analysis.

Results: Included patients were 115 with LNF gram-negative growth on blood cultures, distributed as follows: patients with one-week treatment duration were 45, two-week 43, and three-week duration was 27. Characteristics were balanced (P>0.05) except for chronic lung disease, and a few antibiotics were more in the three-week duration (P<0.05). There was a significant difference for 28-day all-cause mortality (P=0.019), but relapse and reinfection did not significantly differ among the three treatment durations (P>0.05). The relapse rate was 3%, and a new infection was 7%.

Conclusion: There was no significant difference for one-week, two-week, and three-week antibiotics treatment durations in the 90- day all-cause mortality, relapse, and reinfection rates, but increased 28-day mortality in the three-week duration.

Keywords: Bloodstream infection; Acinetobacter relapse; Mortality; Antimicrobial therapy