Skip to content

Endophthalmitis Study Group, European Society of Cataract & Refractive Surgeons. Prophylaxis of postoperative endophthalmitis following cataract surgery: results of the ESCRS multicenter study and identification of risk factors. Journal of cataract and refractive surgery. 2007;33:978-88. PMID 17531690

One-paragraph summary

A prospective randomised, partially masked, multicentre study across 24 ophthalmology units in Austria, Belgium, Germany, Italy, Poland, Portugal, Spain, Turkey and the United Kingdom recruited 16,603 patients into a 2×2 factorial design testing intracameral cefuroxime (1 mg in 0.1 mL normal saline) and topical perioperative levofloxacin. Twenty-nine patients presented with endophthalmitis, of whom 20 were classified as proven infective. Multivariable logistic regression showed that the absence of an intracameral cefuroxime regimen was associated with a 4.92-fold increase in total postoperative endophthalmitis (95% CI 1.87–12.9). Three other risk factors emerged: clear-corneal incisions versus scleral tunnels (5.88-fold, 1.34–25.9), silicone versus acrylic IOL optic material (3.13-fold, 1.47–6.67), and the presence of surgical complications (4.95-fold, 1.68–14.6). Among proven infective cases, absence of cefuroxime and silicone optic material remained significant, and men were more predisposed (OR 2.70, 1.07–6.8).

Key findings

  • Omitting intracameral cefuroxime: OR 4.92 (95% CI 1.87–12.9) for total postoperative endophthalmitis.
  • Clear-corneal incision vs scleral tunnel: OR 5.88 (1.34–25.9).
  • Silicone vs acrylic IOL optic: OR 3.13 (1.47–6.67).
  • Any surgical complication: OR 4.95 (1.68–14.6).
  • Male sex among proven infective cases: OR 2.70 (1.07–6.8).
  • More experienced surgeons were more likely to be associated with endophthalmitis cases — a case-mix signal, not a skill signal.
  • 29 events in 16,603 patients: the trial was large and still yielded very wide confidence intervals.

Limitations

  • Only 20 proven infective cases underpin the definitive analysis; every confidence interval is wide.
  • Partially masked; the topical levofloxacin arm was not blinded in the same way as the intracameral arm.
  • Tests one agent at one dose — it cannot rank agents, and the subsequent agent-level literature is observational (Kato 2022, PMID 36258003; Bowen 2018, PMID 29326317).
  • 2007 surgical practice: incision architecture, IOL materials and viscoelastics have all changed since.
  • Does not address prophylaxis after posterior capsule rupture, which required separate later synthesis (Arteche 2026, PMID 40958158).

Why it matters

This is the only large randomised trial of endophthalmitis prophylaxis in cataract surgery, and it converted intracameral antibiotic use from a regional habit into a standard of care. National guidance now instructs use of intracameral cefuroxime with commercially or pharmacy-prepared solutions (NICE NG77 recommendations 1.8.5–1.8.6, accessed 2026-08-31), and every subsequent synthesis is anchored to it — including the 6.8-million-eye network meta-analysis in which only the intracameral route achieved statistical significance (PMID 36258003) and the meta-analysis reporting endophthalmitis in 1 of 485 surgeries without versus 1 of 2,855 with intracameral antibiotics (Kessel 2015, PMID 25779209). Its risk-factor findings on incision type and IOL material also shaped surgical technique independently of the prophylaxis question.

Cited by wiki pages

  • complications
  • red-flags-and-safety-concerns
  • anaesthesia-and-perioperative-care
  • surgical-technique
  • guidelines
  • overview