Table 2.

Retrospective studies showing the association between prolonged early antibiotics and NEC.

First Author and Year	Study design	Study Population	Results
Cotten 2009	Multi-center retrospective cohort study	4,039 ELBW infants treated with early antibiotics despite sterile cultures. Infants who received ≥5 days early antibiotics were compared to infants who received <5 days.	Increased odds for death (1.46, 95% CI 1.19-1.78) and increased odds for NEC or death (1.30, 95% CI 1.10-1.54) associated with ≥5 days exposure to early antibiotics.
Alexander 2011	Single-center retrospective case-control study	124 NEC cases (stage 2 or 3) were compared to 248 controls that were matched by gestational age, birth weight, and birth year.	Cumulative duration of antibiotic exposure associated with increased risk of NEC (aOR 1.10, 95% CI 1.02-1.19).
Kuppala 2011	Multi-center retrospective cohort study	365 VLBW infants ≤ 32 weeks’ gestation exposed to early antibiotics despite sterile cultures. Infants were categorized into prolonged antibiotics (≥5 days), limited antibiotics (1–4 days) and no antibiotics (0 days).	Each day of early antibiotic treatment was associated with increased odds for composite outcome of NEC, LOS, and death (aOR 1.24, 95% CI 1.07-1.44).
Ghany 2012	Single-center retrospective cohort study	207 VLBW infants who received early antibiotics despite sterile cultures. Antibiotic treatment for ≥5 days were compared to <5 days.	Each day of early antibiotic treatment was associated with increased odds of NEC (aOR 1.32, 95% CI 1.05-1.65).
Cantey 2018	Single-center retrospective cohort study	374 VLBW infants with gestational age <33 weeks at birth. Infants with composite outcome of interest (NEC + LOS + death) were compared to infants without this composite outcome.	Each day of early antibiotic treatment in the first 14 days of life was associated with increased risk for the composite outcome of NEC + LOS + death (aOR 1.24, 95% CI 1.17-1.31).
Esmaeilizand 2018	Multi-center retrospective case-control study	224 NEC cases (stage 2 or 3) were compared with 447 controls that were matched by gestational age, birth weight, and gender.	Early antibiotic treatment for ≥5 days was associated with increased NEC (aOR 2.02, 95% CI 1.55-3.13) compared to antibiotic treatment for 0–4 days.
Raba 2019	Single-center retrospective case-control study	22 NEC cases (stage 2 or 3) were compared with 32 controls that were matched by gestational age, sex, maternal chorioamnionitis exposure, and mode of delivery.	Prolonged early antibiotics for >5 days associated with NEC (OR 3.6, 95% CI 1.13-11.47).
Chen 2022	Single-center retrospective cohort study	132 VLBW infants were investigated by multivariable logistic regression to determine the association of antibiotic treatment and NEC.	Each day of early antibiotic treatment in the first 14 days of life was associated with increased odds of NEC (aOR 1.28, 95% CI 1.03-1.59).
Zhu 2022	Single-center retrospective cohort study	51 NEC cases (stage 2 or 3) were compared with 516 with no NEC. Infants were all VLBW and <32 weeks’ gestation at birth.	Early antibiotic therapy duration was associated with increased odds of NEC (aOR 1.27, 95% CI 1.13-1.42).
Vatne 2022	Population-based retrospective study	4,932 VLBW infants were studied using nationwide registry of Norway. Association between empirical antibiotics and NEC was assessed using multivariable logistic regression models, adjusting for known confounders.	Antibiotics ≥ 5 days were associated with higher odds of NEC (aOR 2.27, 95% CI 1.02-5.06).

ELBW, extremely low birth weight; NEC, necrotizing enterocolitis; LOS, late-onset sepsis; VLBW, very low birth weight.