Results from the European Society of Cataract and Refractive Surgeons (ESCRS) endophthalmitis study, released by Peter Barry, FRCS, and colleagues, at the ESCRS congress in London, is the largest study of an antibiotic in medical history. Results showed that the risk of endophthalmitis following phacoemulsification cataract surgery could be significantly reduced to a base incidence rate of 0.05%, by using an intracameral injection of cefuroxime at the end of surgery. Sixteen thousand patients were recruited to the study, held between September 2003 and January 2006. Although the groundbreaking endophthalmitis study answered many questions regarding endophthalmitis outbreaks, it has also raised questions and debate among ophthalmologists. Below are responses from surgeons on the results of the ESCRS endophthalmitis study.
PAUL URSELL, MD, MBBS, FRCOphth
The ESCRS endophthalmitis study showed that intracameral cefuroxime reduced the base rate for endophthalmitis to approximately 0.05%. If the drug was not used, the rate was significantly higher, despite using preoperative levofloxacin drops. The rate of endophthalmitis was shown to be higher for silicone IOLs compared with acrylic lenses; and higher for clear corneal incisions compared with scleral tunnels. These secondary observations require further analysis. IOL and incision choices are primarily dependent on the surgeon, and the rate of endophthalmitis may be affected by factors related to that institution's practice, leading to an abnormal rate of endophthalmitis. Presently, surgeons must wait for further statistical analysis before deciding whether clear corneal incisions and silicone IOLs are still safe to use.
The majority of surgeons use subconjunctival cefuroxime at the end of surgery, and not the studied technique (ie, intracameral cefuroxime). This trial has not addressed the question of whether it is advisable to change from this form of administration to intracameral cefuroxime. There are pros and cons to using both techniques.
There may be a difference in endophthalmitis rates between these two techniques, but previously published data of endophthalmitis rates between intracameral and subconjunctival cefuroxime groups is not greatly different. Subconjunctival cefuroxime administration may be painful, especially when the patient has received topical anesthesia for surgery. It is also slightly messy, however, many surgeons will choose to administer a simultaneous subconjunctival steroid injection. If intracameral cefuroxime is used by itself, there is no pain. Unless the surgeon elects not to use subconjunctival steroids, however, this advantage may be lost.
In the ESCRS study, Dr. Barry alluded to the risks of producing the appropriate concentration of antibiotic, in a so-called kitchen pharmacy. The procedure for drawing up the dosage in an appropriate concentration by the theater staff is complex, and incorrect dosing is possible. Errors are very rare, but if the incorrect dose is administered intraocularly, it may have catastrophic consequences for the eye. This cannot occur if an incorrect dose is administered subconjunctivally. It is hoped that commercial prepackaged doses of cefuroxime can be produced, in response to surgeons demanding this form of cefuroxime for their patients.
SAMUEL MASKET, MD
All cataract surgeons and their patients owe a debt of gratitude to the ESCRS and to Peter Barry, FRCS, for organizing, performing, and reporting the ESCRS prospective endophthalmitis study.1 While some critical analyses of the study design have suggested certain flaws, and the study was terminated prior to completion of the original plan for total patient enrollment, the fact remains that a statistically valid, fivefold reduction in the incidence of postoperative endophthalmitis was associated with the intracameral injection of cefuroxime, a second-generation cephalosporin, at the close of surgery. That particular antibiotic had been successfully employed in an earlier, but nonrandomized investigation reported by Montan.2
Interestingly, in the ESCRS study, the use of topical antibiotics did not seem to offer prophylactic benefit. Topical agents were not administered postoperatively until the day after surgery, and only one agent, levofloxacin, a third-generation fluoroquinolone, was employed in the investigation. Would a fourth-generation fluoroquinolone topical agent have been more beneficial, or would frequent installation of topical antibiotics on the day of surgery been protective in the ESCRS study? Unanswered questions of this nature have fueled some degree of controversy regarding the investigation.
Perhaps it is most interesting to speculate on how US practice patterns will change as a result of the ESCRS investigation. Will surgeons routinely adopt a policy of using prophylactic intracameral antibiotics in general, or, more specifically, cefuroxime? Will topical antibiotics be utilized less? Regarding the former question, and in accord with survey data, roughly 30% of American Society of Cataract and Refractive Surgeons members have used prophylactic intracameral antibiotics in the past, although there were no controlled investigations of a proven benefit. Typically American surgeons have used vancomycin or gentamycin (Garamycin; Shearing, Berlin, Germany) alone or in combination for intracameral use. It should be recognized that amino-glycosides (eg, gentamycin) demonstrate particular retinotoxicity. Moreover, policy statements from the American Academy of Ophthalmology and the Centers for Disease Control have advised against the routine use of Vancomycin for surgical prophylaxis.3
Cefuroxime is unfamiliar to most US surgeons,and whether its antibacterial spectrum will cover local endophthalmitis pathogens remains unanswered. One concern is that neither cefuroxime nor any other antibiotic is available in a unit dose for intraocular injection. As a result, surgeons must be responsible for dilution and preparation of antibiotics. That said, there is always a concern regarding dilutional errors and contamination. It has been suggested that more eyes will be harmed by improper antibiotic preparation than would be at risk for infection, absent the antibiotic. It is hoped that in time, and with appropriate investigations, that the US Food and Drug Administration will allow manufacturers to supply specific agents in sterile prepackaged unit doses. An alternative strategy at this moment is to use small amounts of undiluted 0.5% moxifloxacin (Vigamox; Alcon Laboratories, Inc., Fort Worth, Texas) directly from the commercial eye drop bottle, as it is nonpreserved and sterile.4
Particularly with respect to the routine use of intracameral antibiotic prophylaxis, I have no doubt that the ESCRS endophthalmitis investigation will have a significant impact on practice patterns. Hopefully, in return, there will be a corresponding reduction in the overall incidence of postoperative infection.
PER MONTAN, MD, PhD
It can safely be stated that the ESCRS endophthalmitis study has added tremendously to our knowledge of prophylactic regimens in cataract surgery. With its prospective, randomized, controlled, semiblinded design, it is as close as one can get to the highest ranking in evidence-based medicine. The study provides a strong case for intracameral antibiotics, and it also clearly shows that povidone-iodine—alone or in combination with levofloxacin eye drops—does not achieve sufficient protection against postoperative infection. It also shows that the addition of topical levofloxacin offers little, if any, extra protection against postoperative endophthalmitis.
One issue that arises, however, is how much these results can be generalized. In terms of topical treatments, our experience at St. Erik's Eye Hospital (Stockholm, Sweden) has been just as disheartening as the results from the nonintracameral groups of the ESCRS study. At St. Erik's, we tried rinsing the conjunctiva with chlorhexidine solution in combination with gentamicin eye drop administration immediately before the operation. From this regimen, we found an endophthalmitis rate of 0.25% in 12,800 procedures.2 When we switched to intracameral cefuroxime—which we first tried in 1996—we found a marked reduction in the postoperative endophthalmitis rate to approximately 0.05%.6
That rate has been reproduced in all Swedish prospective surveys through the National Cataract Register.6 Intracameral ceforoxime is currently used in 99% of cataract operations (ie, 75,000 to 80,000, in Sweden per year). Antibiotic eye drops, either before surgery, after surgery, or both, are used in fewer than 10% of procedures without any positive effect on the postoperative endophthalmitis incidence. Therefore, the Swedish experience is in full accord with the ESCRS study's findings, regarding the preventive effects of antibiotics.
Other data from the ESCRS investigation revealed the use of clear corneal incisions and silicone implants as particularly hazardous. I, however, wish to warn against implementing alternative practices based on these results. In our recent prospective Swedish survey analyzing 225,000 procedures,7 there was only a slightly increased risk for postoperative endophthalmitis with clear corneal incisions or temporal incisions. Silicone IOLs were not found to be more permissive to infection, however. Additionally, the present off-label use of cefuroxime has spurred concern for legal implications of in-house diluting. It is preferable to have a commercialized and monitored formula. Currently, we intend to continue preparing the solution as we have in the past—especially since it has worked for Swedish doctors in practice for well over 400,000 procedures. It is apparent that the protective potency of intracameral antibiotics was demonstrated in the ESCRS study and in several Swedish investigations. In the future, regimens aimed at preventing postoperative endophthalmitis following cataract surgery will have to be supported by solid data generated in large-scale studies, and compete with a now-established benchmark incidence of 0.05%.
DAVID SPALTON, FRCS, FRCP, FRCOphth
The first thing I must do is congratulate Peter Barry, FRCS, David V. Seal, MD, FRCOphth, and their team, for conceiving this study and seeing it to the end. The logistical and bureaucratic difficulties have been immense, and the financial investment from the ESCRS enormous. In the end, their foresight has been rewarded. The ESCRS study confirms earlier Swedish work, showing that intracameral cefuroxime is extremely effective in reducing postoperative bacterial infection. The statistical analysis looks solid, so I think surgeons must then accept this data and change to this regimen. If a patient develops postoperative endophthalmitis, it would be medically and legally difficult to defend any other antibiotic regimen in the face of the proven scientific evidence.
There is already an enormous body of clinical experience with intracameral cefuroxime confirming the outcome and reliability of this technique. Safety concerns are, therefore, not an issue. Still, logistical problems do remain. Cefuroxime is sold in much higher concentrations than what is required for this procedure, and the drug may be unstable in a solution form. Hopefully, these issues can be quickly overcome. In the meantime, I believe we should not be alarmists and use these dilution problems to thwart a change in practice.
Like all good studies, the ESCRS study poses questions that need further thought or possibly more data analysis. For example, is cefuroxime the best antibiotic in this situation? Or, would using an antibiotic with a broader spectrum be better? The responsibility here lies with the manufacturer or proponent to prove any conclusion.
If corneal wounds do have a higher incidence of infection, should we not consider pursuing better corneal wound construction, rather than returning to scleral incision surgery? I would consider this a retrograde step. Another intriguing point of the ESCRS study is the apparent higher incidence of infection with silicone IOLs. To further understand this data, we must look for confounding factors including (1) whether injecting or folding the IOL is a factor; (2) if greater contact with the ocular surface influences the outcome; and (3) what the endophthalmitis rates are for hydrophilic or hydrophobic IOL materials—and not just acrylics. If any of that information were confirmed, it would be poor judgment to continue using an IOL material with a potentially higher risk of infection. In the meantime, surgeons must continue to base their decisions on scientific evidence, and not merely what has worked in the past. It is important for all ophthalmologists to read the fully published study report, as I am sure it will change routine cataract surgery for the better.
Paul Ursell, MD, MBBS, FRCOphth, is a Consultant Ophthalmologist, in Surrey, UK. He may be reached at enquires@cataract-doctor.com.
Samuel Masket, MD, is Clinical Professor of Ophthalmology at the Jules Stein Eye Institute, UCLA Center for Health Sciences and former Chairman of the Cataract Special Interest Committee of the American Society of Cataract and Refractive Surgery. He may be reached at avcmasket@aol.com.
Per Montan, MD, PhD, is an Anterior Segment Surgeon at St. Eriks' Hospital, in Stockholm, Sweden. Dr. Montan may be reached at + 46 8 672 32 49; per.montan@sankterik.se.
David Spalton, FRCS, FRCP, FRCOphth, is a Consultant Ophthalmic Surgeon at St. Thomas' Hospital, in London. Dr. Spalton may be reached at spalton@eyepractice.fsnet.co.uk.
Up Front | Jan 2007
ESCRS Study: What are the Implications?
This endophthalmitis study raised questions about the use of intracameral cefuroxime.
Paul Ursell, MD, MBBS, FRCOphth; Samuel Masket, MD; Per Montan MD, PhD; and DaviD Spalton, FRCS, FRCP, FRCOphth
