Objective: To assess the burden of invasive infection following surgery (surgery-associated infections, SAI) among extremely premature infants.
Study design: This was an observational, prospective study of infants born at gestational age 22-28 weeks hospitalized for >3 days, between April 1, 2011-March 31, 2015 in academic centers of the NICHD Neonatal Research Network. SAI was defined by culture-confirmed bacteremia, fungemia, or meningitis ≤14 days following a surgical procedure.
Results: Of 6573 infants, 1154 (18%) who underwent surgery were of lower GA (mean [SD]: 25.5 [1.6] vs. 26.2 [1.6], p<0.001), lower BW (803  vs 886 , P < .001), and more likely to have a major birth defect (10% vs. 3%, p<0.001); 64% had one surgery (range 1-10 per infant). Most underwent gastrointestinal (GI) procedures (873, 76%) followed by central nervous system (CNS) procedures (150, 13%). Eighty-five (7%) infants had 90 SAI (78 bacteremia, 5 fungemia, 1 bacteremia and meningitis, 6 meningitis alone). Coagulase-negative staphylococci (CoNS) were isolated in 36 (40%) SAI and were isolated with another organism in 5 episodes. Risk of SAI or death ≤14 days after surgery was higher after GI compared with CNS procedures [16% vs 7%, adjusted RR (aRR) (95% CI): 1.95 (1.15-3.29), p=0.01]. Death ≤14 days after surgery occurred in 141 of the 1154 infants; 128 deaths occurred after GI surgeries.
Conclusions: Surgical procedures were associated with bacteremia, fungemia, or meningitis in 7% of infants. The epidemiology of invasive post-operative infections as described in this report may inform the selection of empiric antimicrobial therapy and post-operative preventive care