Preoperative oral antibiotics and surgical-site infections in colon surgery (ORALEV): a multicentre, single-blind, pragmatic, randomised controlled trial

Eloy Espin Basany*, Alejandro Solís-Peña, Gianluca Pellino, Esther Kreisler, Doménico Fraccalvieri, Manuel Muinelo-Lorenzo, Olga Maseda-Díaz, José María García-González, Marta Santamaría-Olabarrieta, Antonio Codina-Cazador, Sebastiano Biondo

*Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

52 Citations (Scopus)


Background: Previous studies have found that mechanical bowel preparation with oral antibiotics can reduce the incidence of surgical-site infections, but no randomised controlled trial has assessed oral antibiotics alone without mechanical bowel preparation. The aim of this study was to determine whether prophylaxis with oral antibiotics the day before elective colon surgery affects the incidence of postoperative surgical-site infections. Methods: In this multicentre, pragmatic, randomised controlled trial (ORALEV), patients undergoing colon surgery were recruited from five major hospitals in Spain and 47 colorectal surgeons at these hospitals participated. Patients were eligible for inclusion if they were diagnosed with neoplasia or diverticular disease and if a partial colon resection or total colectomy was indicated. Participants were randomly assigned (1:1) using online randomisation tables to either administration of oral antibiotics the day before surgery (experimental group) or no administration of oral antibiotics before surgery (control group). For the experimental group, ciprofloxacin 750 mg was given every 12 h (two doses at 1200 h and 0000 h) and metronidazole 250 mg every 8 h (three doses at 1200 h, 1800 h, and 0000 h) the day before surgery. All patients were given intravenous cefuroxime 1·5 g and metronidazole 1 g at the time of anaesthetic induction. The primary outcome was incidence of surgical-site infections. Patients were followed up for 1 month after surgery and all postsurgical complications were registered. This study was registered with EudraCT, 2014-002345-21, and, NCT02505581, and is closed to accrual. Findings: Between May 2, 2015, and April 15, 2017, we assessed 582 patients for eligibility, of whom 565 were eligible and randomly assigned to receive either no oral antibiotics (n=282) or oral antibiotics (n=282) before surgery. 13 participants in the control group and 16 in the experimental group were subsequently excluded; 269 participants in the control group and 267 in the experimental group received their assigned intervention. The incidence of surgical-site infections in the control group (30 [11%] of 269) was significantly higher than in the experimental group (13 [5%] of 267; χ2 test p=0·013). Oral antibiotics were associated with a significant reduction in the risk of surgical-site infections compared with no oral antibiotics (odds ratio 0·41, 95% CI 0·20–0·80; p=0·008). More complications (including surgical-site infections) were observed in the control group than in the experimental group (76 [28%] vs 51 [19%]; p=0·017), although there was no difference in severity as assessed by Clavien-Dindo score. No differences were noted between groups in terms of local complications, surgical complications, or medical complications that were not related to septic complications. Interpretation: The administration of oral antibiotics as prophylaxis the day before colon surgery significantly reduces the incidence of surgical-site infections without mechanical bowel preparation and should be routinely adopted before elective colon surgery. Funding: Fundación Asociación Española de Coloproctología.

Original languageEnglish
Pages (from-to)729-738
Number of pages10
JournalThe Lancet Gastroenterology and Hepatology
Issue number8
Publication statusPublished - Aug 2020


Dive into the research topics of 'Preoperative oral antibiotics and surgical-site infections in colon surgery (ORALEV): a multicentre, single-blind, pragmatic, randomised controlled trial'. Together they form a unique fingerprint.

Cite this