Duration of Antibiotic Therapy in Neonatal Gram-negative Bacterial Sepsis-10 Days Versus 14 Days: A Randomized Controlled Trial.

Abstract

Optimal duration of antibiotic therapy in Gram-negative bacterial (GNB) sepsis in non-VLBW infants has not been specifically evaluated in previous studies. This was an open labeled noninferiority randomized controlled trial. Non-VLBW infants with GNB sepsis without meningitis whose blood culture were sterile after day 7 of treatment and who were in clinical remission on day 9 of appropriate antibiotic were randomized to short duration (SDR) group and long duration (LDR) group. Infants in SDR group and LDR group received antibiotic therapy for 10 days and 14 days respectively. Primary objective was to compare treatment failure. Secondary objectives were to compare duration of hospitalization, complications of intravenous (IV) therapy and its duration, episodes of new-onset sepsis and all-cause mortality. Of 222 infants with GNB sepsis, 58 eligible infants were randomized in each group and 113 of these were analyzed. There was no difference in proportion of infants with multidrug-resistant (MDR) organism in SDR versus LDR group [33(60%) versus 32(55.1%) (P = 0.84)]. There were no treatment failures in either group. Median (IQR) duration of hospital stay was higher in LDR group as compared with SDR group: 20(18, 23) versus 16(13, 20) days (P < 0.001). Infants in LDR group required IV therapy for a longer duration as compared with SDR group mean (SD): 15.2(1.2) versus 10.9(0.8) days (P < 0.001). Median (IQR) episodes of extravasation were higher in LDR group: 5(4.7) versus 3(2.3) (P < 0.001). There was no difference in episodes of phlebitis and hematoma. No infants had died on follow up. In suitably selected non-VLBW infants with Gram-negative sepsis, 10 days therapy is noninferior to 14 days therapy.