CAUSATIVE PATHOGENS

In order to select an appropriate antimicrobial agent, it is imperative to identify which organisms should be targeted. Pathogens causing postoperative infectious endophthalmitis are thought to originate mostly from the endogenous lid and conjunctival flora, but they may also arise from airborne contaminants, tainted intraocular solutions, infected lenses or surgical instruments, and OR personnel.3 The Endophthalmitis Vitrectomy Study4 reported the growth of different organisms cultured from 291 culture-positive intraocular specimens, the vast majority of which were found to be gram-positive, coagulase-negative micrococci (Tables 1 and 2). Of note, all gram-positive cultures were found to be sensitive to vancomycin. Although occurring much less frequently, gram-negative organisms have been found to be more virulent and to result in poorer visual outcomes.5

CHOICE OF INTRACAMERAL DRUGS

If the purpose of antibiotic prophylaxis is to clear an inoculation of organisms introduced during surgery, then the antibiotic should be in the anterior chamber at the time of introduction or shortly thereafter.3 The intracameral administration of antibiotics provides the greatest intraocular bioavailability. Given the findings of the Endophthalmitis Vitrectomy Study, the relative frequency of gram-positive organisms, and their 100% susceptibility to vancomycin, the drug is an attractive and viable option for intracameral antibiotic prophylaxis. Vancomycin is a bacteriocidal agent that works by binding to the target of a peptidogylan precursor and blocks both transglycolation and tranpeptidation. This action prevents polymerization of peptidoglycans, causes the loss of cell wall integrity, and finally results in autolysis of the bacterium.6 Vancomycin is highly effective against all gram-positive bacteria, but it is ineffective against gram-negatives. Vancomycin has been administered via either infusion (concentrations of 10 to 50µg/mL)6-8 or in the capsular bag (up to 1mg/mL).1

More recently, cefuroxime has been proposed and used as an intracameral prophylactic antibiotic. Cefuroxime is a second-generation cephalosporin that works by bacteriocidal, beta-lactam action—inhibiting penicillin-binding proteins, which prevent cross-linking of peptidoglycan strands normally needed for the cell wall's integrity and leads to osmotic lysis of the bacterium. Cefuroxime provides less gram-positive coverage than vancomycin, while providing some gram-negative coverage, notably against Haemophilus influenzae, Escherichia coli, Klebsiella, and Proteus. It has only been described for intracameral injection (1mg).2,5,9

REVIEW OF EVIDENCE
Safety
Because vancomycin and cefuroxime are not labeled for intracameral use, the burden of demonstrating their safety rests with the treating surgeons.9 A review of the literature finds no formal studies that have evaluated the safety of intracameral vancomycin in humans. There are empirical reports1 of doses greater than 2000µg/mL being used safely in thousands of patients, which is much higher dose than what is found in other studies. A retrospective analysis by Gimbal et al1 demonstrated that the average endothelial cell count loss in these patients was equivalent to that in patients not receiving intracameral antibiotics reported in other studies. One clinical trial demonstrated an almost threefold increase in the rate of cystoid macular edema in eyes infused with 10µg/mL compared with the control group; however, the results of this study were confounded by posterior capsular rupture, diabetes, and lengthy operating times.8

In contrast to vancomycin, the safety profile of cefuroxime has been studied formally in an observer-masked, controlled trial. Montan et al9 found that the intracameral instillation of 1mg of cefuroxime at the completion of surgery did not significantly affect visual acuity, endothelial cell counts, flare, or the postoperative rates of cystoid macular edema. The last has also been confirmed by optical coherence tomography of the macula in a separate study.10 Furthermore, Montan et al9 concluded that IgE hypersensitivity to cefuroxime was not a contraindication for its use intracamerally if patients were pretreated with oral antihistamines.

Efficacy
The most convincing data to support the use of vancomycin infusion come from Mendivil and Mendivil,11 who reported a reduction in the culture-positive intraocular aspirates from 13% to 5%, 2 hours after cataract surgery. All bacteria cultured were sensitive to vancomycin. Furthermore, the investigators determined that 47% of the initial concentration of vancomycin remained in the anterior chamber, significantly greater than the approximate 0.5- to 2.0-µg/mL minimum inhibitory concentration of most bacteria that are responsible for endophthalmitis. The pharmacokinetics derived from this study were used to generate an in vitro model by Libre et al7 that simulated decreasing vancomycin levels during an 8-hour period, approximately four half-lives; demonstrating a 1,000-fold reduction in the amount of methicillin-resistant Staphylococcus aureus colony-forming units as compared with controls. In a retrospective study, Gimbel et al1 reported on almost 12,000 surgeries without a single case of postoperative infectious endophthalmitis while utilizing a gentamycin infusion and in-the-bag injection of vancomycin.

Despite the apparent reduction in culture-positive isolates, the fact that vancomycin-sensitive organisms were still present in the Mendivil and Mendivil11 study is indicative of vancomycin's inability to achieve a complete bacteriocidal effect, and yet there were no cases of postoperative infectious endophthalmitis. This strongly suggests that the pathogenicity of postoperative infectious endophthalmitis is far more complicated than this single variable.6 In fact, Dickey et al12 reported that 43% of postoperative anterior chamber aspirates were culture-positive, yet there was not a single case of postoperative infectious endophthalmitis in the study; these results are indicative of eyes' inherent ability to defend themselves against bacterial infections. Furthermore, the sole use of vancomycin leaves a small but significant gap in its lack of coverage of gram-negative organisms.

Evaluating cefuroxime, Montan et al9 reported cefuroxime levels that exceeded the minimum inhibitory concentration for several relevant species for up to 5 hours after surgery, while acknowledging that it could not be automatically concluded that this presence was sufficient to eradicate all the target contaminants. However, clinically, Montan et al9 reported a drop in postoperative infectious endophthalmitis from 0.26% to 0.06% (P<.001) in 34,100 and 32,180 consecutive cases, respectively, after introducing intracameral cefuroxime injections. When only phacoemulsification was evaluated, the difference was even more impressive; the rate of postoperative infectious endophthalmitis dropped from 0.48% to 0.09% (P<.001).

These findings were consistent with a national prospective study conducted in Sweden, in which 95% of the total number of cataracts in Sweden were tracked by the Swedish National Cataract Register, which found rates of postoperative infectious endophthalmitis to be 0.22% and 0.053% in patients not receiving and receiving intracameral antibiotics (98.5% cefuroxime), respectively (P<0.001). In both the Montan et al9 study and the Swedish survey,5 the proportion of enterococcus responsible for postoperative infectious endophthalmitis in culture-proven cases was significantly higher (38.0% and 29.9%, respectively [Tables 3 and 4]) than that found in the Endophthalmitis Vitrectomy Study.4 These findings constitute a serious gap in the coverage of cefuroxime against gram-positive bacteria.

Resistance
The greatest opposition to the prophylactic use of intracameral vancomycin stems from concerns regarding increasing drug resistance. Vancomycin is typically used in bacterial infections that are life threatening or recalcitrant to other antimicrobial therapies. In the wake of increasing vancomycin resistance, the Centers for Disease Control and Prevention and the AAO jointly issued a statement in 1999 discouraging the use of vancomycin for surgical prophylaxis in cataract surgery. Critics have pointed out that an intraocular dose of 1mg, distributed over 40L of body water, is equivalent to a systemic dosage of 0.025µg/mL, which is much too low to affect an exposed organism's survival and reproduction. Thus, a 1-mg dose of vancomycin creates no pressure on an organism to develop resistance.7 Gordon6 surmised that, of the 12,000kg of vancomycin used in the US in 2000, only approximately 7.5kg were used for prophylaxis in cataract surgery, representing 0.07% of the total amount. He therefore concluded that ophthalmic usage would constitute an insignificant selection pressure for promoting vancomycin-resistant bacteria. No commentary regarding cefuroxime resistance was noted.

BOTTOM LINE

Given the relative infrequency of postoperative infectious endophthalmitis, it is unlikely that a proper, evidence-based algorithm will ever be developed to end the debate about antibiotic prophylaxis in cataract surgery. With respect to vancomycin, the evidence that is available is conflicting and inconclusive. The most notable opinion in this debate comes from the joint statement from the Centers for Disease Control and Prevention and the AAO, which discouraged the use of vancomycin for prophylaxis. Unlike vancomycin, cefuroxime has considerably more evidence to support its usage as a prophylactic agent, in the form of a prospective survey from Sweden that included more than 180,000 cases during a 3-year period. Whether or not these findings could be duplicated in North America is undetermined. Until then, cataract surgeons will have to continue to generate their own practice patterns based on their own experiences and extrapolations of the available literature. 

Reviewer
Baseer Khan, MD, states that he holds no financial interest in any product mentioned herein. Dr. Khan may be reached at (415) 258-8211; bob.khan@utoronto.ca.

Panel Members
Y. Ralph Chu, MD, is Medical Director, Chu Vision Institute in Edina, Minnesota. He states that he holds no financial interest in any product mentioned herein. Dr. Chu may be reached at (952) 835-1235; yrchu@chuvision.com.
Nina Goyal, MD, is a resident in ophthalmology at the Rush University Medical Center in Chicago. She states that she holds no financial interest in any product mentioned herein. Dr. Goyal may be reached at (312) 942-5315; ninagoyal@yahoo.com.
Wei Jiang, MD, is a resident in ophthalmology at the Jules Stein Eye Institute in Los Angeles. She states that she holds no financial interest in any product mentioned herein. Dr. Jiang may be reached at (310) 825-5000; wjiang70@yahoo.com.
Patty Lin, MD, MBA, is a resident in ophthalmology at the Jules Stein Eye Institute in Los Angeles. She states that she holds no financial interest in any product mentioned herein. Dr. Lin may be reached at (310) 825-5000; lin@jsei.ucla.edu.
Gregory J. McCormick, MD, is a cornea and refractive fellow at the University of Rochester Eye Institute in New York. He states that he holds no financial interest in any product mentioned herein. Dr. McCormick may be reached at (585) 256-2569; mccormick_greg@hotmail.com.
Lav Panchal, MD, is a fellow at the Wang Vision Institute in Nashville, Tennessee. He states that he holds no financial interest in any product mentioned herein. Dr. Panchal may be reached at (917) 751-8651; drpanchal@wangvisioninstitute.com.
Jay S. Pepose, MD, PhD, is Professor of Clinical Ophthalmology & Visual Sciences, Washington University School of Medicine, St. Louis. He states that he holds no financial interest in any product mentioned herein. Dr. Pepose may be reached at (636) 728-0111; jpepose@peposevision.com.
Paul Sanghera, MD, is senior resident in ophthalmology in the Department of Ophthalmology and Vision Sciences at the University of Toronto. He states that he holds no financial interest in any product mentioned herein. Dr. Sanghera may be reached at (416) 666-7115; sanghera@rogers.com.
Renée Solomon, MD, is an ophthalmology fellow at Ophthalmic Consultants of Long Island in New York. She states that she holds no financial interest in any product mentioned herein. Dr. Solomon may be reached at reneeoph@yahoo.com.
Tracy Swartz, OD, MS, states that she holds no financial interest in any product mentioned herein. She may be reached at (615) 321-8881; drswartz@wangvisioninstitute.com.
Ming Wang, MD, PhD, states that he holds no financial interest in any product mentioned herein. He may be reached at (615) 321-8881; drwang@wangvisioninstitute.com.