Impact of Early Oral Antibiotic Therapy in Infants With Bacteremic Urinary Tract Infections.

Abstract

To determine the effect of early conversion to oral antibiotics in infants ≤90 days with gram-negative bacteremia and concomitant urinary tract infection on hospital length of stay (LOS) and 30-day revisits for urinary tract infection. This retrospective cohort study included infants ≤ 90 days who had concomitant positive blood and urine cultures with gram-negative bacteremia at 22 hospitals from January 1, 2002 through January 31, 2020. Early oral conversion was defined as a maximum intravenous (IV) duration of 4 days. We conducted a noninferiority test to compare early oral conversion to long IV therapy. A propensity score framework was used to assess comparisons in survey-weighted regression models using matched weights. Among 174 infants, 73 infants received early oral conversion. The LOS was shorter in infants receiving early oral conversion compared to long IV therapy (median 2 days [interquartile range 2, 3] vs 4 days [3, 5]), descriptively (P < .001) and in noninferiority analysis (ratio = 0.43; 90% CI 0.35, 0.53, P < .001). Thirty-day revisits occurred in 5 of 174 (2.8%) of total patients. Early oral conversion did not meet our noninferiority criteria for the 30-day revisits (odds ratio: 4.22, 90%; confidence interval: 0.83, 21.34, P = .91). Early oral conversion was associated with shorter LOS. The rate of 30-day revisit rates overall was low.