Article
Few well-designed studies evaluate endophthalmitis prophylaxis. Available evidence suggests a role for intracameral cefuroxime. The benefit must be weighed against the potential risk of errors in formulation, however.
Baltimore
-Protocols for prophylaxis of postoperative endophthalmitis remain controversial and vary widely among surgeons because of the paucity of Level I evidence to support specific interventions, said Oliver D. Schein, MD, MPH, at Current Concepts in Ophthalmology.
"While a large number of studies have been conducted to investigate enophthalmitis risk factors and prophylactic regimens, most are not rigorously designed, and the prophylaxis studies have been based primarily on surrogate outcomes," said Dr. Schein, the Burton E. Grossman Professor of Ophthalmology, Wilmer Eye Institute, Johns Hopkins University School of Medicine, Baltimore. "Therefore, it is possible by reviewing the literature to find some published study that will support virtually any conclusion one chooses about endophthalmitis prophylaxis."
Considering the best available studies, Dr. Schein said that good evidence implicates exposure to vitreous, older patient age, and wound integrity as endophthalmitis risk factors. For prophylaxis, strong evidence exists from Swedish studies using national registry data and from the European Society of Cataract and Refractive Surgeons (ESCRS)-sponsored randomized, prospective, multicenter study that intracameral cefuroxime is protective.
In the ESCRS study, use of intracameral cefuroxime reduced the rate of endophthalmitis five-fold. Intensive perioperative topical treatment with levofloxacin did not significantly reduce the risk of endophthalmitis, and patient age was not a significant risk factor. Intraoperative complications, use of a clear corneal incision, and silicone IOL use also independently predicted risk, however.
Further discussing the ESCRS results, Dr. Schein said that one argument against the purported benefit of intracameral cefuroxime centers on the high background rate of endophthalmitis in the control group, about 1 in 300 cases.
"This information could be used to suggest that intracameral cefuroxime may be beneficial only in settings where the risk of endophthalmitis is very high and where perhaps it is offsetting the effect of poor surgical habits," he said. "However, a counterargument emerges from the Swedish experience, where intracameral cefuroxime was used in 97% of cases and the rate of endophthalmitis was four-fold lower than the reported rates in the United States."
The role of corneal incisions as an endophthalmitis risk factor is controversial, Dr. Schein added. Although some surgeons have used this technique with excellent success and no cases of endophthalmitis, multiple studies have shown clear corneal incision is associated with an excess risk.
Cons of intracameral antibiotics
Although the Swedish experience and the ESCRS study may be considered to provide a sufficiently sound evidence base for justifying intracameral antibiotic use, a clinical role for this strategy must consider whether the benefits are worth the risks, he said. The latter include ocular toxicity resulting from using the wrong medication, concentration, or pH.
“With these types of administration errors, any benefit of intracameral antibiotic disappears because the complication it is trying to prevent, i.e., postoperative endophthalmitis, is relatively rare,” Dr. Schein said. “Overcoming these risks would require fail-safe measures, such as having commercially available, premixed, unit-dose preparations or very strict operative control.”
A number needed to treat (NNTT) analysis provides a method to quantify and compare the benefits of different interventions. The NNTT calculation identifies the number of patients who need to be treated to prevent one additional bad outcome. For endophthalmitis, the NNTT would be expected to vary geographically based on the local background rate of the complication. If the incidence of endophthalmitis is 1 in 1,000 cataract surgery cases and it can be cut in half to 1 in 2,000 surgeries with intracameral antibiotic use, however, then 2,000 patients would need to be treated to prevent one case of endophthalmitis. If the existing rate is 1 in 500 and it can be reduced to 1 in 2,000 with intervention, then only 667 patients would need to be treated to prevent an additional case.
“It is very difficult for the individual cataract surgeon to come to conclusions about endophthalmitis prophylaxis based on his/her own personal experience,” Dr. Schein said. “I have had one case in my own practice about 5 years ago, and none since in more than 1,500 subsequent cases.
“On the other hand, on a national basis, we have good evidence that the rate of endophthalmitis in the United States is on the order of one case per 500 to 600 cataract surgeries,” he said. “As a rough estimate, if we perform 2 million cataract surgeries annually in the United States, we would see about 4,000 cases of endophthalmitis annually. If intracameral cefuroxime were used in the United States as a standard of care, as it is in Sweden, then it is possible that as many as 3,000 of these cases might be prevented.”
He also said that intracameral antibiotic use, which provides an anterior chamber concentration exceeding the minimum inhibitor concentration for common endophthalmitis pathogens for only about 4 hours, likely is not the ideal prophylaxis. He said he still considers it an improvement over topical treatment alone, however.
“It should be of significant interest to cataract surgeons in the United States that in Sweden, where the rate of endophthalmitis is significantly lower, there is no routine use of topical antibiotics postoperatively,” Dr. Schein said.
The ideal agent for intracameral antibiotics would have bactericidal activity and provide adequate ocular concentrations at the close of surgery and for 24 to 48 hours postoperatively, he said.“Considering the millions of patients exposed per year, reasonable cost and an extremely low risk of side effects are also important criteria,” Dr. Schein said.OT