Timing of perioperative antibiotics for cesarean delivery: a metaanalysis

Maged M. Costantine, Mahbubur Rahman, Labib Ghulmiyah, Benjamin D. Byers, Monica Longo, Tony Wen, Gary D.V. Hankins, George R. Saade

Research output: Contribution to journalArticlepeer-review

151 Scopus citations


Objective: The purpose of this study was to summarize the available evidence on timing of perioperative antibiotics for cesarean delivery. Study Design: We searched the literature for studies that compare prophylactic antibiotics for cesarean delivery that are given before the procedure vs at cord clamping. Only randomized controlled trials were included. Results: Preoperative administration significantly reduced the risk of postpartum endometritis (relative risk [RR], 0.47; 95% CI, 0.26-0.85; P = .012) and total infectious morbidity (RR, 0.50; 95% CI, 0.33-0.78; P = .002). There was a trend toward lower risk of wound infection (RR, 0.60; 95% CI, 0.30-1.21; P = .15). Preoperative administration of antibiotics did not significantly affect suspected neonatal sepsis that requires a workup (RR, 1; 95% CI, 0.70-1.42), proven sepsis (RR, 0.93; 95% CI, 0.45-1.96), or neonatal intensive care unit admissions (RR, 1.07 95% CI, 0.51-2.24). There was no significant heterogeneity between the randomized controlled trials. Conclusion: There is strong evidence that antibiotic prophylaxis for cesarean delivery that is given before skin incision, rather than after cord clamping, decreases the incidence of postpartum endometritis and total infectious morbidities, without affecting neonatal outcomes.

Original languageEnglish (US)
Pages (from-to)301.e1-301.e6
JournalAmerican Journal of Obstetrics and Gynecology
Issue number3
StatePublished - Sep 2008


  • antibiotics
  • cesarean delivery
  • endometritis
  • infectious morbidity

ASJC Scopus subject areas

  • Medicine(all)
  • Obstetrics and Gynecology


Dive into the research topics of 'Timing of perioperative antibiotics for cesarean delivery: a metaanalysis'. Together they form a unique fingerprint.

Cite this