Association between initial empirical antibiotic therapy and necrotizing enterocolitis in very preterm infants at different risk for early-onset sepsis: a multicenter cohort study.

BACKGROUND

We aimed to evaluate the association between initial empirical antibiotic therapy (IEAT) and necrotizing enterocolitis (NEC) in very preterm infants at different early onset sepsis (EOS) risk.

METHODS

This prospective, observational, multi-center study enrolled very low birth weight (VLBW) infants of < 32 weeks' gestation without culture-proven EOS. Infants delivered by cesarean section, without rupture of membranes before delivery, and without maternal chorioamnionitis were categorized as low-risk for EOS. Competing risk Cox regression analysis and cumulative incidence function curves were used to analyse the association between the duration of IEAT (0 days, < 3 days, 3-5 days or > 5 days) and NEC cumulative incidence, and death was considered as a competing risk. Finally, restricted cubic spline (RCS) models were used to further examine the relationship between time-varying antibiotic exposure and the risk of NEC.

RESULTS

Of the 11678 included infants, 2767 infants were categorized as low risk for EOS. In the whole cohort and the infants with non-low-risk EOS, IEAT < 3 days showed a significantly lower risk of NEC (adjusted sHR 0.36, 95% CI [0.19,0.68]; (adjusted sHR 0.40, 95% CI [0.21, 0.79]) and IEAT for 3-5 days showed a lower risk of NEC (adjusted sHR 0.69, 95% CI [0.52, 0.94]; adjusted sHR 0.72, 95% CI [0.5, 1.03]), compared with those with No-IEAT. In the infants with low-risk EOS, there was no significant association between the duration of IEAT and the incidence of NEC. The RCS analysis showed a positive linear relationship between the duration of IEAT and the risk of NEC in very preterm infants at different EOS risk.

CONCLUSIONS

Competing-risk analysis showed that NEC risk may be inversely associated with short-term IEAT in the whole cohort and the non-low-risk EOS group. Notably, the risk of NEC tended to be higher with prolonged IEAT duration in very preterm infants at different EOS risk. These associational findings are exploratory and warrant further investigation.