Study of Antibiotic Prophylaxis for Endophthalmitis Following Cataract Surgery

Cataract extraction with intra-ocular lens implantation is the most commonly performed surgical procedure in the elderly population in Europe. The frequency varies in different European Union (EU) countries involving 2 to 7 per 1000 population per annum. A population with an ever-increasing proportion of the elderly is advancing this figure with numbers requiring surgery expected to increase by 70% by 2006. While technical advances (phacoemulsification) have enhanced the efficacy of the procedure, the possibility of serious post-operative infection with loss of vision remains the most important unsolved problem.

In the absence of scientific evidence, the European Society of Cataract and Refractive Surgeons (ESCRS) wishes to determine whether one currently developed method of delivering antibiotics intraocularly in Sweden is of benefit compared to surgery without the use of perioperative antibiotics, as currently practised in many European centres, or to the use of frequent application of topical antibiotic drops perioperatively. The ESCRS also wishes to assess possible risk factors for later endophthalmitis.

This study requires four groups of 8,750 patients - 35,000 in all - to demonstrate reductions of currently reported rates of endophthalmitis (approximately 0.3%) in patients where no intraocular antibiotics are used, or where they are used by the subconjunctival route. Other studies using intraocular vancomycin have reported results as low as 0.05%, but these studies were not standardised or controlled and the results were anecdotal. In addition, vancomycin should not be used for routine prophylaxis and should only be used as the antibiotic of 'last resort'.

This multi-centre randomised study involves 24 operating units in 8 EU countries (Austria, Belgium, England, Germany, Italy, Poland, Portugal and Spain) and Turkey. Results in 8,750 patients receiving an intracameral injection of antibiotic (cefuroxime) at the end of surgery will be compared to 8,750 patients receiving topical antibiotic (levofloxacin) prophylaxis before and at the end of surgery, 8,750 patients receiving the combination and 8,750 patients receiving neither regime; in addition, all patients will receive povidone iodine antiseptic prophylaxis prior to surgery and post-operative levofloxacin from days 1 to 6 to prevent wound infection.

This group size will be sufficiently large to ensure at least 80% statistical power to detect a reduction of incidence from 0.25% to 0.08% using 5% significance level tests.

Surgical data from all patients will be collected directly into computers based within each operating theatre and transmitted to a central server in Glasgow (University of Strathclyde). Follow-up data for the presence or absence of endophthalmitis will be collected from all patients at post-operative visits and similarly recorded on computers for transmission to the central server. Sophisticated techniques will be used to check all data as it is uploaded.