Randomized controlled trial of 7-Day vs. 14-Day antibiotics for neonatal sepsis

Abstract

There are no evidence-based guidelines available regarding the duration of antibiotics in neonatal septicemia. We compared the effectiveness of a 7-day intravenous antibiotic regimen with the standard 14-day regime in blood-culture-proven sepsis in neonates. This was a controlled, blinded, randomized trial with stratification (for birth weight). Blood-culture-positive septic babies > or =32 weeks and/or > or =1500 g were enrolled if meningitis and other deep-seated focal infections were ruled out. Parental consent was obtained. Randomization to either 7-day or 14-day therapy was done on day 7 of antibiotics if the baby had clinically remitted by day 5. Blood culture was repeated 24 h after antibiotic completion. Subjects were observed in the hospital for at least 72 h, and followed-up for 28 days by weekly visits and telephonic contacts. The primary outcome was treatment failure within 28 days defined as a positive blood culture, or clinical signs accompanied by either positive CRP or adjudicated to be a relapse by an expert committee. A total of 120 babies were eligible, 51 were excluded (no consent: 12; non-remission: 39), and 69 were randomized to receive either a 7-day course (n = 34) or a 14-day course (n = 35) of antibiotics. Baselines variables were comparable in the two groups. Primary outcome assessment could be done in 33 cases in either group. There was a trend to greater treatment failures in the 7-day group compared with 14-day group (5 vs. 1, respectively; P = 0.19). On subgroup analysis of subjects with Staphylococcus aureus infection, those who received 7-day therapy (n = 7) had significantly more treatment failure than 14-day therapy (n = 7) (four and zero, respectively; P = 0.022), whereas on sub-group analysis of babies with non-S. aureus infections, treatment failure rates were identical (3.8% in both groups). On comparing the organisms isolated in the group of subjects which was not randomized by virtue of being symptomatic (n = 39) vs. the group which was randomized (n = 69), it was found that S. aureus infections were significantly commoner in the former group (61.5 vs. 21.3%, respectively; P < 0.001). Neonates > or =32 weeks and/or > or =1500 g with S. aureus sepsis require 14 days of antibiotics. S. aureus infection is also associated with failure to achieve clinical remission by the 5th day of antibiotic therapy. Larger trials are required to confirm whether neonates with non-S. aureus sepsis, whose symptoms remit by 5 days, can be treated with 7 days of antibiotics.