Our group was impressed to read the editorial by Olson discussing the best approach for antibiotic prophylaxis of postoperative endophthalmitis (POE) and would like to raise several points.
To date, the only randomized controlled trial regarding intracameral (IC) cefuroxime, conducted by the European Society of Cataract and Refractive Surgeons (ESCRS), randomized patients into 1 of 4 treatment groups and showed a 4.92-fold increase in the risk of POE when IC cefuroxime was not used. Olson raises the concern that the ESCRS study had a high rate of endophthalmitis in the non-cefuroxime group compared with similar cohorts in that time period, possibly exaggerating the benefit of IC cefuroxime. A meta-analysis, however, has demonstrated comparable rates of endophthalmitis in the non-cefuroxime group found in the ESCRS study and in observational studies. In addition, in the recently published ESCRS guidelines there are acknowledged limitations to which Olson refers, in that the literature offers limited data for ophthalmologists to make informed decisions about the choice of prophylactic interventions for cataract surgery.
The studies referenced by Olson in favor of the Topicalists included retrospective cohort and animal studies. Topical antibiotics may reduce the number of conjunctival bacteria, lowering risk of intraocular contamination during surgery or through a leaking wound postoperatively. However, the ESCRS study did not demonstrate a lower rate of endophthalmitis with topical antibiotics.
It would seem that although a case has been made for IC cefuroxime, the most likely and not discussed confounder is nonclosure of wounds in a definitive fashion. This is because firstly, it is less likely to be organisms that reach the anterior chamber at surgery that cause POE, or there would have been the same incidence of POE in conjunctival closure or scleral closure cases. Historically, this is not what occurred. Secondly, in standard phacoemulsification surgery, there is an estimated washout effect of 1400 times per case, as the anterior chamber volume is 0.25 mL compared with the irrigation volume of 350 mL. Therefore, there is little likelihood of viable pathogens remaining in the anterior chamber at the end of the surgery.
The ongoing debate between Intracameralists and Topicalists remains dynamic. However, though we accept that IC cefuroxime or equivalent will probably become de rigueur with the passage of time, our group would like to propose a third approach to the prevention of POE. This approach would be supported by those who close corneal wounds definitively: The Suturists. The Suturists maintain that POE occurs following ingress of organisms into the anterior chamber in the hours and days following surgery. They would therefore maintain that they are doing precisely what needs to be done to treat the basic mechanism of POE.