INFECTION PROPHYLAXIS

Risk MAN A CME MONOGRAPH GEMENT Update in INFECTION PROPHYLAXIS for Ocular Surgery Highlights from a Roundtable Discussion ORIGINAL RELEASE: SEPTEMBER 1, 2010 LAST REVIEW: AUGUST 17, 2010 EXPIRATION: SEPTEMBER 30, 2011 Sponsored by The New York Eye and Ear Infirmary Institute for Continuing Medical Education PROGRAM CHAIR AND MODERATOR Marguerite B. McDonald, MD, FACS In joint sponsorship with MedEdicus LLC In association with Ophthalmology Times FACULTY Esen Akpek, MD Peter McDonnell, MD Terrence P. O Brien, MD This continuing medical education activity is supported through an unrestricted educational grant from Bausch + Lomb Incorporated

PROGRAM CHAIR AND MODERATOR Marguerite B. McDonald, MD, FACS Clinical Professor of Ophthalmology NYU Langone Medical Center New York, New York Adjunct Clinical Professor of Ophthalmology Tulane University School of Medicine New Orleans, Louisiana Ophthalmic Consultants of Long Island Lynbrook, New York FACULTY Esen Akpek, MD Associate Professor of Ophthalmology Director Ocular Surface Disease and Dry Eye Clinic Wilmer Eye Institute The Johns Hopkins University Baltimore, Maryland Peter McDonnell, MD William Holland Wilmer Professor of Ophthalmology Chairman, Department of Ophthalmology Wilmer Eye Institute The Johns Hopkins University Baltimore, Maryland Terrence P. O Brien, MD Professor of Ophthalmology Charlotte Breyer Rodgers Distinguished Chair in Ophthalmology Director of the Refractive Surgery Service Bascom Palmer Eye Institute Ocular Microbiology Laboratory University of Miami Miller School of Medicine Palm Beach Gardens, Florida LEARNING METHOD AND MEDIUM This educational activity consists of a supplement and ten (10) study questions. The participant should, in order, read the learning objectives contained at the beginning of this supplement, read the supplement, answer all questions in the post test, and complete the evaluation form. To receive credit for this activity, please follow the instructions provided on the post test and evaluation form. This educational activity should take a maximum of 1.5 hours to complete. CONTENT SOURCE This continuing medical education (CME) activity captures content from a CME roundtable discussion held on Friday, April 9, 2010, during the American Society of Cataract and Refractive Surgery Meeting in Boston, Massachusetts. TARGET AUDIENCE This educational activity is intended for comprehensive ophthalmologists. OVERVIEW The successful prevention of a surgically related infection is important for ensuring a good outcome for the ophthalmic patient. The number of procedures has increased, and techniques have evolved, but infection rates, though low, have been rising. It is important for ophthalmic surgeons to understand that infection risk continues to be present. Understanding and effectively using surrogate parameters and drug profiles can assist the surgeon with antibiotic selection and administration. Learning about recent data and current best-practice perioperative techniques and regimens will help optimize infection prophylaxis. LEARNING OBJECTIVES After successfully completing this activity, you will have improved your ability to: 1. Review the causes of ocular infections related to common surgical procedures 2. Compare and contrast the relevant efficacy and safety profiles of the currently available topical antibiotics for ophthalmic surgical procedures 3. Discuss best-practice surgical techniques and perioperative regimens for infection prophylaxis ACCREDITATION STATEMENT This activity has been planned and implemented in accordance with the Essential Areas and Policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of The New York Eye and Ear Infirmary and MedEdicus LLC. The New York Eye and Ear Infirmary is accredited by the ACCME to provide continuing medical education for physicians. DESIGNATION STATEMENT The NewYork Eye and Ear Infirmary designates this educational activity for a maximum of 1.5 AMA PRA Category 1 Credits. Physicians should only claim credit commensurate with the extent of their participation in the activity. MISSION STATEMENT It is The New York Eye and Ear Infirmary Institute for Continuing Medical Education s stated mission to create medical education activities that will serve to increase the knowledge, skills, professional performance, and relationships that a physician uses to provide services for patients, the public, or the chosen profession. DISCLOSURE POLICY STATEMENT The New York Eye and Ear Infirmary requires that each teacher/contributor or individual in a position to control the content of a CME activity accredited by The New York Eye and Ear Infirmary disclose the existence of any relevant financial interests or other relationships (eg, paid speaker, employee, paid consultant on a board and/or committee for a commercial company) that would potentially affect the objectivity of activity content. Teachers/Contributors are also asked to make a disclosure that a product is still investigational when an unlabeled use of a commercial product or an investigational use, not yet approved for any purpose, is discussed during an educational activity. The disclosed information in no way presumes to assess the contributor s qualifications or suitability. The intention is to provide full disclosure of any potential conflict of interest, real or apparent, that is related to a specific educational activity. Individuals who neglect to provide information about relevant financial relationships will be disqualified from serving as a planning committee member, teacher, speaker, moderator, or author of the educational activity. In addition, such individuals will be prohibited from having control of, or the responsibility for, the development, management, presentation, or evaluation of the CME activity. Full disclosure of faculty and commercial relationships, if any, follows. DISCLOSURES Esen Akpek, MD: Dr Akpek had a financial agreement or affiliation during the past year with the following commercial interests in the form of Consultant/Advisory Board: Lux Biosciences, Inc, and Pfizer Inc. Contracted Research: Allergan, Inc; Inspire Pharmaceuticals; and Pfizer Inc. Marguerite B. McDonald, MD: Dr McDonald had a financial agreement or affiliation during the past year with the following commercial interests in the form of Consultant/ Advisory Board: Abbott Medical Optics (AMO/VISX); Allergan, Inc; Aton Pharma, Inc; Bausch + Lomb Incorporated; FOCUS Laboratories; ForSight VISION3, Inc; Inspire Pharmaceuticals; Ocularis Pharma, Inc; Santen Pharmaceutical Co, Ltd; and Vistakon Pharmaceuticals, Inc. Contracted Research: Alcon, Inc, and Pfizer Inc. Peter McDonnell, MD: Dr McDonnell had a financial agreement or affiliation during the past year with the following commercial interests in the form of Consultant/ Advisory Board: Alcon Research Institute, and Allergan, Inc. Terrence P. O'Brien, MD: Dr O Brien had a financial agreement or affiliation during the past year with the following commercial interestsinthe form of Consultant/Advisory Board: Abbott Medical Optics (AMO/VISX); Alcon, Inc; Allergan, Inc; Bausch + Lomb Incorporated; Inspire Pharmaceuticals; ISTA Pharmaceuticals, Inc; Santen Pharmaceutical Co, Ltd; and Vistakon Pharmaceuticals, Inc. Robert Latkany, MD, Peer Reviewer: Dr Latkany had a financial agreement or affiliation during the past year with the following commercial interests in the form of Royalties: Sonomed, Inc. Speakers Bureau: Alcon, Inc, and Allergan, Inc. Contracted Research: ISTA Pharmaceuticals, Inc; Sirion Therapeutics, Inc; and Rapid Pathogen Screening. EDITORIAL SUPPORT DISCLOSURES Derek Dore, PharmD, and Cynthia Tornallyay, RD, have no relevant commercial relationships to disclose. DISCLOSURE ATTESTATION The contributing physicians listed above have attested to the following: 1. that the relationships/affiliations noted will not bias or otherwise influence their involvement in this activity; 2. that practice recommendations given relevant to the companies with which they have relationships/affiliations will be supported by the best available evidence or, absent evidence, will be consistent with generally accepted medical practice; and 3. that all reasonable clinical alternatives will be discussed when making practice recommendations. OFF-LABEL DISCUSSION This activity includes off-label discussion of all the antibiotics used in ophthalmic surgery, as they are not indicated for infection prophylaxis for surgical procedures. GRANTOR STATEMENT This continuing medical education activity is supported through an unrestricted educational grant from Bausch + Lomb Incorporated. TO OBTAIN CME CREDITS To obtain CME credit for this activity, read the material in its entirety and consult referenced sources as necessary. Complete the evaluation form along with the post test answer box within this supplement. Remove the Activity Evaluation page from printed supplement or print the Activity Evaluation page from Digital Edition. Return via mail or fax to Kim Corbin, Director, ICME, The New York Eye and Ear Infirmary, 310 East 14th Street, New York, NY 10003 or fax to (212) 353-5703. Your certificate will be mailed to the address that you provide on the evaluation form. Please allow 3 weeks for mailed/faxed forms to process. Note: You must score a 70% or higher to receive credit for this activity. DISCLAIMER The views and opinions expressed in this educational activity are those of the faculty and do not necessarily represent the views of The New York Eye and Ear Infirmary, MedEdicus LLC, Ophthalmology Times, or Bausch + Lomb Incorporated. Please refer to the official prescribing information for each product for discussion of approved indications, contraindications, and warnings. 2 2010 MedEdicus LLC

Risk Update in MANAGEMENT INFECTION PROPHYLAXIS for Ocular Surgery Highlights from a Roundtable Discussion Anterior segment procedures are expected to increase dramatically, concurrently with the aging population. It is projected that by the year 2020, the number of Americans with cataracts will increase to 30 million, while those with pseudophakia/aphakia will number 9 million 50% and 60% increase, respectively, from the year 2000. 1 Approximately 700,000 LASIK (laser-assisted in situ keratomileusis) surgeries are performed annually in the United States. 2 In addition, intravitreal injections are becoming an increasingly common ophthalmic procedure as treatment for macular degeneration, diabetic macular edema, and macular edema of retinal vein occlusive diseases. With the increased frequency of such procedures has come the sense that these surgeries are essentially routine; they are not, however, risk-free. Recently, 4 experts in anterior segment surgery convened to discuss the consequences of the occurrence of infection in common ocular procedures and to offer insight on new management approaches for preventing suboptimal outcomes. Marguerite B. McDonald, MD: Among the potential complications of ocular surgery, the most highly feared and devastating is endophthalmitis. Although at the present time the estimated incidence rate of this vision-threatening complication is low, it has been increasing; and we expend tremendous effort to prevent its occurrence. 3,4,5 Several of our practices can potentially contribute to the incidence of endophthalmitis. It has been linked not conclusively, but strongly to the use of clear corneal incisions. 3,4 Peter McDonnell, MD: We examined the potential for extraocular microorganisms to infiltrate the anterior chamber and thereby increase the risk of endophthalmitis after clear corneal incisions. With the use of India ink, we determined the degree of ocular surface fluid ingress of 4 cadaveric human globes with clear corneal wounds at various controlled intraocular pressures that simulated normal intraocular pressure fluctuations: that is, upon blinking or eye squeezing. We found that the 3 globes with sutureless clear corneal wounds allowed penetration of India ink, whereas the 1 globe that was sutured did not show any ingress of India ink. Thus, we concluded that there is a potential for microorganisms to infiltrate the intraocular compartment following a sutureless clear corneal cataract surgery. 6 Because there is potential for inoculation of the anterior chamber during the postoperative period perhaps the first day or 2 before the epithelium is healed and the incision sealed I believe it is important to prepare the ocular surface at the beginning of surgery and, probably, to maintain effective levels of antibiotic in the anterior chamber for some period of time after surgery. 3 Terrence P. O Brien, MD: There are multiple factors that may increase the risk for postoperative infection. The surgeon must consider factors related to the organisms that can potentially cause infections as well as factors related to the host. There are intraoperative and extraoperative considerations. Some factors may not be readily apparent. For example, a potential risk factor for endophthalmitis is application of lidocaine, 2% gel prior to povidone-iodine preparation. 3 Lidocaine gel is a viscous lubricant and, if applied prior to the preparation of the ocular surface with an antiseptic or an antibacterial agent, the efficacy of those agents may be relatively blocked. 7 Esen Akpek, MD: Inferior incision location, particularly without the use of sutures, may also increase the risk of endophthalmitis. The inferior incision is made close to the lower tear meniscus likely the location of the potential microbial sources of endophthalmitis. 3,8 Dr McDonnell: This is consistent with data showing a greater risk of bleb-related endophthalmitis associated with inferior trabeculectomies compared with superior and temporal blebs. 8,9 Dr McDonald: We have begun seeing an increase in infections with LASIK surgery as well as with cataract surgery bacterial infections and fungal infections have been documented to be on the rise as interface infections. 10 Dr O Brien: Antifibrotic agents used as adjuncts in glaucoma surgery to modulate wound healing are a major potential risk factor for infection. When these antiproliferative agents are applied, there is a dramatic effect on the regional cells, with reduction in the local vasculature, even obliteration of the local blood vessels, which causes a vulnerability due to loss of neutrophils and other vesselderived leukocytes along with a thin area where there are really only a few epithelial cells as a barrier between the extraocular environment and the intraocular milieu. Dr McDonald: Also, it is important to stress adequate prophylaxis with the use of intravitreal injections for age-related macular degeneration, retinal vein occlusion, and diabetic macular edema. It has been estimated that endophthalmitis after intravitreal injections occurs at a rate of approximately 1 per 1000 procedures 11 ; that figure, however, certainly underestimates the per-patient risk because each patient typically receives numerous injections. CAUSES/SOURCES OF OCULAR INFECTION Approximately 2 decades ago an important paper by Speaker and colleagues provided insight into the source of the infecting organism in cases of acute postoperative endophthalmitis namely, the patient s own external flora. In patients with acute postoperative endophthalmitis, investigators studied the relationship of bacterial organisms isolated from intraocular (ie, vitreous) cultures with

those isolated from their own external ocular and nasal tissues at the time of admission for treatment. In 82% (14/17) of cases the genetic identity of the intraocular isolate was bacteriologically indistinguishable from external cultures (Figure 1). Thus, the investigators concluded that attention should be focused on the external tissues, specifically reduction in the colonization of the microbial flora in the prophylaxis and prevention of postoperative bacterial endophthalmitis. 12 In a 4-year multicenter prospective study of Endophthalmitis Vitrectomy Study patients, Bannerman and colleagues found that 67.7% (71/105) of eyelid isolates were indistinguishable from intraocular isolates. It was concluded that periocular microbiota are the source of postoperative endophthalmitis in the majority of cases, stressing the importance of aggressively eradicating as many periocular bacterial organisms as possible before initiating any procedure. 13 82% FIGURE 1. Percentage of organisms isolated from vitreous that are genetically indistinguishable from isolates from patient s eyelid, conjunctiva, or nose. 12 Dr McDonald: It is well established that the sources of ocular infection are the patient s own flora the lids, the lashes, and the nose. Dr O Brien: While a majority of cases may originate with the patient s own flora, exogenous sources of microbial inoculation during various stages of the procedure may also be a factor in causing postoperative endophthalmitis. Despite the best efforts of the surgeon and operating room team, contamination occasionally occurs. Dr Akpek: The take-home message is that there will always be contamination; thus, it is most important to decrease the microorganism numbers prior to surgery rather than attempting to cure the infection postsurgery. Ocular Pathogens Dr O Brien: Traditionally the normal ocular flora the microbiota surrounding the lids, eyelashes, conjunctiva, and ocular surface are the most likely source of infection. Specifically, Staphylococcus epidermidis is the most common pathogen, with Staphylococcus aureus also being a significant source. 13,14 Unfortunately, we have been observing a change in the prevalence of potentially dangerous, highly virulent pathogens. That is, in the past, we typically observed methicillin-resistant S aureus (MRSA) only in hospital settings via nosocomial transmission; however, community-acquired MRSA has become increasingly prevalent and also increasingly virulent, and it behaves in a microbiological way that is more likely to cause infection. Dr McDonald: Dr O Brien, your laboratory recently made an interesting finding regarding ocular pathogens. Dr O Brien: Yes, our group at the Bascom Palmer Eye Institute recently utilized genomic analysis to examine the diversity of the ocular surface microbiota. We found that the ocular surface is inhabited by a wide array of microorganisms: 40% of the recovered species in 1 sample had never been reported as ocular commensals or pathogens. Thus, it is possible that unusual pathogens may escape the normal defense mechanisms and find their way into the eye. 15 Ocular surface as a line of defense Dr Akpek: The eye, much like the body in general, has a built-in defense mechanism. Innate immunity, the first line of defense, as well as the acquired immune response, plays a defensive role. Immune defense of the eye obviously starts from the eye surface. Tears transport antimicrobial proteins, such as lysozyme and immunoglobulins (eg, IgA) to the ocular surface to prevent infections. Epithelial cells of the cornea and keratocytes secrete the cytokine interleukin (IL)-1 α, activating the immune response to protect against microbial invasion. Keratocytes, in response to IL-1 α and tumor necrosis factor (TNF)-α, synthesize IL-6 and -defensins, which can interact with IL-1 and TNF-α in a synergistic way. These all play a defensive role; however, when there is an imbalance of the inflammation, then destruction of the cornea can occur, such as happens in autoimmune diseases. 16 Dr McDonnell: The human eye has evolved, resulting in the emergence of a number of defense mechanisms against pathogens, which explains why so often it is possible for a person to have sustained trauma, corneal abrasions, intravitreal drug administration, and other injuries to the eye and not have infectious complications. Antibacterial peptides are also able to kill or neutralize some pathogens that enter the anterior chamber. Thus, when clinical infection does develop, it is either a particularly virulent organism or the quantity of these organisms that overwhelms the innate ocular defense mechanisms. 16 Dr O Brien: In addition to the endogenous peptides, a number of surface defense mechanisms have been characterized, including toll-like receptors that are present on the ocular surface as well as intracellular adhesion molecules, such as ICAM 1, and others, that may be involved in the upregulation during an inflammatory response after surgery. These substances can lead to the recruitment of inflammatory cells, including polymorphonuclear neutrophils, macrophages, and other immune surveillance cells that enhance the host defense mechanisms. 16 Dr McDonnell: With the development of improved surgical technology and techniques that lead to less traumatic surgery, the eye is, to some extent, deceived. Less surgical trauma may evoke a less elaborate defense mechanism, leading to a reduction in the eye s ability to handle these offending microorganisms. Dr Akpek: Painless surgery is not necessarily ideal: substance P is secreted by the keratocytes in response to pain, thus stimulating an influx of neutrophils into the anterior chamber. 16,17 PROPHYLAXIS Reduction of bacteria before and after surgery Dr Akpek: Prepping and draping the surgical area in a sterile fashion is very important. It is a critical step in decreasing the bacterial load prior to the surgery. Once the microbes are introduced into the anterior chamber, they are difficult to treat. The use of povidone-iodine, 5% is key in preventing endophthalmitis. 18 Dr O Brien: I agree that antiseptic agents, which possess a contact mechanism of action and exert an immediate direct effect, are pivotal and critical for application presurgery. 18-20 Studies have shown that the combination of an antiseptic agent and an antibacterial agent is more effective than either agent alone. An antiseptic can achieve approximately 90% reduction in surface colonization of bacteria. The addition of a topical antibacterial agent leads to a more pronounced reduction in surface colonization of bacteria. 20 4

The Pathogen Antibacterial Battle Endophthalmitis-causing bacteria are evolving to a more resistant group. 21-23 Ocular Tracking Resistance in U.S. Today (TRUST) is a nationwide longitudinal annual surveillance program that monitors in vitro antimicrobial susceptibility of pathogens isolated from ocular infections. In the inaugural study (2005 to 2006), ocular isolates were prospectively collected from a wide array of institutions. The isolates were tested by an independent, central laboratory for susceptibility to ciprofloxacin, gatifloxacin, levofloxacin, moxifloxacin, penicillin, azithromycin, tobramycin, trimethoprim, and polymyxin B. Mean minimum inhibitory concentrations that would inhibit growth of 90% (MIC 90 ) of the tested isolates were interpreted as susceptible, intermediate, or resistant, according to standardized breakpoints for systemic treatment. The fluoroquinolones tested (ciprofloxacin, gatifloxacin, levofloxacin, moxifloxacin) were generally active in methicillin-susceptible S aureus (MSSA), Streptococcus pneumoniae, and Haemophilus influenzae; however, MRSA susceptibility was low (15.2%). Trimethoprim was found to have high activity against MRSA(Table 1). 24 Updated data for 2007 to 2009 showed that susceptibility profiles were very similar to the previous period and remained stable over the surveillance period. 25 The authors concluded that resistance to MRSA by the fluoroquinolones tested suggests the need to consider alternative therapies when MRSA is the likely pathogen. 24 Recently, a study used The Surveillance Network (TSN) data to assess fluoroquinolone (ciprofloxacin, gatifloxacin, levofloxacin, moxifloxacin) resistance trends in the United States over the past 10 years among ocular isolates of S aureus and S epidermidis. The evaluation TABLE 1. Antimicrobial Susceptibility of Staphylococcus aureus According to Methicillin Status in Ocular TRUST 1 24 Susceptible Intermediate Resistant Antibiotic Status MIC90 No. % No. % No. % Ciprofloxacin MSSA >8 131 79.9 1 0.6 32 19.5 MRSA >8 5 15.2 0 0.0 28 84.8 Levofloxacin MSSA 16 133 81.1 0 0.0 31 18.9 MRSA >16 5 15.2 2 6.1 26 78.8 Gatifloxacin MSSA 4 133 81.1 0 0.0 31 18.9 MRSA >8 5 15.2 1 3.0 27 81.8 Moxifloxacin MSSA 4 133 81.1 5 3.0 26 15.9 MRSA >8 5 15.2 3 9.1 25 75.8 Azithromycin MSSA >16 89 54.3 0 0.0 75 45.7 MRSA >16 2 6.1 1 3.0 30 90.9 Penicillin MSSA >1 16 9.8 0 0.0 148 90.2 MRSA >1 0 0.0 0 0.0 33 100 Polymyxin B* MSSA >8 0 0.0 164 100 MRSA >8 0 0.0 33 100 Tobramycin MSSA 1 152 92.7 2 1.2 10 6.1 MRSA >32 12 36.4 0 0.0 21 63.6 Trimethoprim MSSA 2 160 97.6 4 2.4 MRSA 2 31 93.9 2 6.1 MIC90=minimum inhibitory concentration that inhibits growth of 90% of the tested isolates; MRSA=methicillinresistant S. aureus; MSSA=methicillin-susceptible S. aureus; TRUST=Tracking Resistance in U.S. Today. MSSA, n=164; MRSA, n=33. *Clinical and Laboratory Standards Institute breakpoints for Pseudomonas aeruginosa were used for interpretation of susceptible ( 2 µg/ml), intermediate (not available), and resistant ( 4 µg/ml). Clinical and Laboratory Standards Institute breakpoints unavailable for interpretation of intermediate. Package insert for Tobramycin Injection, USP, used for interpretation of susceptible ( 4 µg/ml). Reprinted with permission from Elsevier. found a steady increase in MRSA ocular isolates and a large proportion of methicillin-resistant S epidermidis (MRSE) ocular isolates, with frequent resistance to the fluoroquinolones evaluated. 26 Dr McDonnell: Ocular TRUST shows that the trends to increasing resistance continue. The old standby antibiotics that we have come to appreciate as being very effective in the past are increasingly less effective, at least in the in vitro environment. 24 Dr O Brien: It appears that some of the older ocular antibacterial agents, which are no longer used frequently or are not used systemically, and thus are under less selection pressure, such as bacitracin or trimethoprim-sulfamethoxazole, may have a renaissance role as prophylaxis agents. These agents may have some efficacy against the emerging resistant ocular strains. How is clinical efficacy defined? Dr McDonald: What is the accepted definition of clinical efficacy in regard to prophylaxis? What is the standard? And what other factors do we need to consider for optimal prophylaxis? Dr O Brien: There is no clinical trial that provides us with absolute guidance; therefore, we are relying on surrogate evidence from laboratory testing principally drug concentration in the relevant ocular tissue and potency of the individual agent as determined by in vitro susceptibility testing. The unfortunate issue is that in ocular microbiology there is no standardization for the breakpoints that determine susceptibility of the individual ocular isolates. Instead, we must extrapolate from achievable concentrations in the serum, which leads to controversy and confusion for the clinician who is attempting to prevent an infection in the best manner through optimal available agent selection. Caution should be used in interpreting laboratory data, and the clinician must understand the comparisons. Thus, we rely on MIC values, specifically the MIC 90, which is the lowest concentration that will inhibit the growth of 90% of isolates. The concentration max (C max ), the maximum concentration in the relevant tissue, is important, because this value can be used as the numerator over the denominator of MIC, computation of which provides the inhibitory quotient, an important pharmacodynamic parameter derived by Neu and colleagues. 27 For the fluoroquinolones, a C max :MIC 90 ratio of at least 10 is predictive of clinical efficacy. 28-31 Perhaps even more important values are the minimal bactericidal concentration (MBC) and the mutant prevention concentration (MPC). MBC and MPC determinations are more labor-intensive for the laboratory technician and more costly, which is a challenge. The MBC is the concentration that is 99.9% bactericidal to the organism. 32 Reaching the MPC is the optimal outcome; it is the concentration, typically 10-fold or more the MIC, that eliminates all organisms, so that there are none surviving to mutate and become resistant. 30,33 All the parameters used to evaluate antibiotics are important and they are intrinsic to the efficacy and safety of an individual agent. It is preferable to have a broad spectrum, bactericidal antibiotic. We want it 5

to be potent having a low MIC 90. We want bioavailability with favorable pharmacodynamics, indicating a high area under the curve (AUC) and sustained concentrations preferably in the tissue; even a postantibiotic effect is preferred. 29 Dr Akpek: Penetration into the tissues is also important in terms of antimicrobial effectiveness. The ideal antibiotic also would maintain antibacterial levels in the anterior chamber in addition to maintaining antibacterial levels on the surface. Dr O Brien: Penetration, however, is not the only factor. We have a dynamic integration of concentration in the tissue with potency, so it probably is more relevant to look at those pharmacodynamic parameters. Dr Akpek: I also consider the role of the vehicle, particularly because one of the goals is to optimize the ocular surface prior to the surgery. Maintaining sustained levels of the antibiotic on the eye surface, including the conjunctival tissues, tear film, and lids is very important. Residence time, that is, antibiotic association with epithelial cells, may play a key role in protecting the ocular surface against bacterial infection. 34 Dr O Brien: The vehicle is extremely important and directly impacts the pharmacodynamics of the antibiotic. For example, the polycarbophil polymer in the vehicle used in commercial formulations of azithromycin and besifloxacin helps provide a suspension matrix of drug on the ocular surface that increases contact time with the ocular surface. This vehicle has been shown to be a safe and effective drug delivery system with other ocular pharmaceuticals. 35 Dr McDonnell: It is ideal to use an agent that has good tolerability and low toxicity, specifically a gentle antibiotic with a vehicle and a preservative that are less disturbing. Dr O Brien: An optimal agent would fulfill the 4 killer B s. Bactericidal Broad spectrum of activity, including the most likely organisms that cause infection in cataract and LASIK surgery Biocompatible that is, less toxic to the ocular surface, perhaps also having some favorable immunomodulatory properties Bioavailability a favorable pharmacokinetic/pharmacodynamic profile being both potent and achieving high concentrations on the ocular surface, to eliminate organisms or reduce contamination, thus reducing the risk of an infection in cataract and refractive surgeries The Agents Dr McDonnell: There are no US Food and Drug Administration (FDA)-approved agents for prophylaxis in ophthalmic surgery. And, it is difficult to study agents because infectious complications after refractive surgery, cataract surgery, or intravitreal injections are rare; thus it is difficult to obtain large samples of patients and prove efficacy in terms of the end point of endophthalmitis. Most studies involve rabbits and use a surrogate end point, such as conjunctival cultures or cultures of the anterior chamber at the end of surgery. Large, expensive human clinical trials are needed to develop evidence-based guidance. In the absence of such guidelines, the prevailing practices in the United States involve the use of advanced-generation topical fluoroquinolones. These agents are administered on the day of surgery or multiple days prior to surgery to minimize the number of organisms on the surface. Dr O Brien: We can garner some useful information from these nonhuman models. The set-up and the laboratory methodology, however, is somewhat contrived relative to the human situation; therefore, caution is urged with extrapolation to the clinical condition of endophthalmitis in humans. Fluoroquinolones Dr O Brien: The advanced group of fluoroquinolones gatifloxacin, moxifloxacin, besifloxacin are more potent against gram-positive and anaerobic species. 36-38 Interestingly, there has been less focus on gram-negative activity; therefore, ciprofloxacin remains the most potent fluoroquinolone in terms of gram-negative activity, based on laboratory susceptibility testing. 39,40 The goal has been to obtain greater antistaphylococcal and antistreptococcal activity over time by changing the substituents on the basic quinolone nucleus. There is a fluorine at the sixth position of the quinolone nucleus, and at the eighth and seventh positions there have been substitutions to improve performance. The 8-methoxy substitution increased potency against gram-positives. 39 The chloride substituent at the C-8 position has led to even greater in vitro potency. 36-38,41 Currently, besifloxacin developed for ocular use is most potent against gram-positive organisms, including methicillin-resistant staphylococci. Moxifloxacin and gatifloxacin follow in potency, based on in vitro susceptibility studies (Figures 2A and 2B). 36-38 Besifloxacin has demonstrated similar potency to vancomycin. 42 Fluoroquinolones target 2 enzymes involved in bacterial replication, namely bacterial DNA gyrase and topoisomerase IV. Balanced activity against each of these enzymes in terms of affinity and avidity to the receptors theoretically allows greater potency and less resistance. 39 Besifloxacin has demonstrated more balanced activity compared with ciprofloxacin and moxifloxacin. 43 Dr McDonald: A stronger formulation of gatifloxacin, 0.5%, developed to address the resistance issue of the older formulation, was just approved by the FDA. We hope data on its efficacy against resistant strains of ocular pathogens will soon be available in the literature. Dr O Brien: The higher concentration of the gatifloxacin formulation is generally desirable, given the concentration-dependent mechanism of action. Higher concentration, however, does not always connote stronger activity Macrolides Dr Akpek: Macrolides erythromycin and azithromycin are bacteriostatic; however, they can be bactericidal at high concentrations. Azithromycin penetrates into the tissues quite well 44 and concentrates within the phagocytic cells and fibroblasts. Dr O Brien: The solubility characteristics of azithromycin, its lipophilicity, allow it to achieve extraordinarily high concentrations in human tissues, including ocular tissues. This characteristic applies to topical dosing as well, particularly in the conjunctiva, the preocular tear film, and the eyelids. High levels of azithromycin are achieved in the tissues around the eye. 44 Dr McDonald: The macrolides are used topically in preparation for surgery, that is, in the treatment of blepharitis; they are not used in the immediate postoperative period as a substitute for the fluoroquinolones, however, as they do not penetrate well into the anterior chamber. 45,46 6

Ratio of AUC 0-24 h to MIC 90 2 1.5 1 0.5 0 Besifloxacin Moxifloxacin Gatifloxacin Ratio of AUC 0-24 h to MIC 90 2 1.5 1 0.5 0 Besifloxacin Moxifloxacin Gatifloxacin FIGURE 2A. Ratio of fluoroquinolone AUC 0 24 h in conjunctiva to the MIC 90 for methicillin-resistant S aureus (MRSA) (mitt population). 37 FIGURE 2B. Ratio of fluoroquinolone AUC 0 24 h in conjunctiva to the MIC 90 for methicillin-resistant S epidermidis (MRSE) (mitt population). 37 MIC 90 values were obtained from susceptibility testing of 81 methicillin-resistant, ciprofloxacin-resistant S aureus isolates. The values were 4 µg/ml for besifloxacin, 32 µg/ml for moxifloxacin, and 64 µg/ml for gatifloxacin. MIC 90 values were obtained from susceptibility testing of 29 methicillin-resistant, ciprofloxacin-resistant S epidermidis isolates. The values were 4 µg/ml for besifloxacin, 64 µg/ml for moxifloxacin, and 128 µg/ml for gatifloxacin. AUC 0 24 h, area-under-the-curve, or the total amount of drug delivered to conjunctival tissue in 1 day following 1 dose; mitt, modified intent-to-treat population (N=108). Reprinted from Torkildsen G, Proksch JW, Shapiro A, Lynch SK, Comstock TL. Concentrations of besifloxacin, gatifloxacin, and moxifloxacin in human conjunctiva after topical ocular administration. Clin Ophthalmol. 2010;4:331-341. Aminoglycosides Dr O Brien: Aminoglycosides principally have gram-negative activity; however, with topical application, ocular tissue concentrations can be achieved that are statistically high enough to have some grampositive activity. The disadvantage of aminoglycosides is their low therapeutic-to-toxic ratio 47 ; thus, as the concentration of an aminoglycoside increases, more toxicity appears quite rapidly a particular problem during frequent exposure to the ocular surface. Additionally, aminoglycosides do not penetrate very well and have little activity against resistant organisms. 24 Trimethoprim Dr McDonald: Topical trimethoprim is a synthetic antibiotic that interferes with the production of tetrahydrofolic acid, a compound that is required in order for bacteria and human cells to produce proteins. Trimethoprim accomplishes this by inhibiting the enzyme responsible for making tetrahydrofolic acid from dihydrofolic acid. Polymyxin B is a peptide antibiotic that disrupts the bacterial cytoplasmic membrane. Trimethoprim is often paired with polymyxin B so that the latter s coverage of gram-negative organisms, particularly Pseudomonas aeruginosa, can supplement the former s gram-positive coverage. The combination of trimethoprim and polymyxin B has not often been used for surgical prophylaxis, however, because there is evidence to suggest that it does not penetrate into the anterior chamber as well as the fluoroquinolones, and on a relative basis MIC 90 values indicate that the fluoroquinolones are at least 4 times more active against S epidermidis, S aureus, and S pneumoniae. 48 Vancomycin Dr O Brien: Vancomycin is a glycopeptide and is still the agent of choice to treat MRSA infections. 49 Vancomycin does not penetrate well into ocular tissue, and so must be formulated at a higher concentration 25 to 50 mg/ml. Vancomycin is also toxic with frequent administration to the ocular surface. 50 The downside of the glycopeptides, including vancomycin and teicoplanin, is that they have a slow kill curve kinetic characteristic; it might be 8 to 10 hours out before a significant effect occurs, whereas the fluoroquinolones have a much more rapid effect within 15 to 30 minutes. Dr McDonnell: Some high-volume cataract surgeons cite their large clinical experience as supporting the use of intraoperative vancomycin that is, vancomycin instilled or infused during thousands of consecutive cases with close to a zero incidence of postoperative infection. 51 In the absence of perfect evidence-based data to the contrary, it is a challenge to refute these large case series that suggest efficacy of intracameral vancomycin. Dr O Brien: Factors in these observations for which there were no controls were the evolving changes in cataract surgical techniques of these prolific surgeons, including smaller incisions, advanced phacoemulsification technology, foldable intraocular lenses that were no-touch delivered into the eye. Thus, it is difficult to equate the infusion of the antibiotic alone with the direct causative reduction in the clinical occurrence of endophthalmitis. Infusion of an antibiotic into phacoemulsification fluids is an irrational mechanism of delivery because of the uncertain amount of fluid that will be infused in a given case. It is preferred to deliver a specified amount of an agent intraocularly at the end of the case, an agent that has been shown to be safe and effective, and at a concentration that is likely to be protective. Dr McDonald: Furthermore, you must rely on a pharmacy to ensure sterility and proper concentration. Intracameral antibiotics Dr McDonnell: A European Society of Cataract & Refractive Surgeons (ESCRS) multicenter study found that the absence of an intracameral cefuroxime prophylactic regimen at 1 mg in 0.1 ml normal saline was associated with a 4.92-fold increase (95% confidence interval, 1.87-12.9) in the risk of total postoperative endophthalmitis. 52 But the evaluation had methodological flaws and there was a high rate of endophthalmitis in the control group. In a recent retrospective clinical study, the use of topical povidoneiodine and gentamicin-irrigation led to a low rate of postoperative endophthalmitis, which was not much different from the rate recently reported in the ESCRS study. Thus, the authors speculate that intracameral cefuroxime was not needed to minimize postoperative infectious complications following cataract surgery. 53 7

Comparison and Contrast of Agents Update of the Evidence Note to readers: Current agents compared in recent head-to-head studies are included here Fluoroquinolone resistance in staphylococcal endophthalmitis isolates Declining in vitro susceptibility and increasing resistance to the fluoroquinolones (ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, moxifloxacin) were observed among coagulase-negative staphylococci (CNS) isolates recovered from patients with clinical endophthalmitis. Overall, in vitro efficacy was less than 80%. From 1990 to 1994, 96.6% of CNS isolates were sensitive to gatifloxacin and moxifloxacin (MIC 90 : 0.19 µg/ml and 0.12 µg/ml, respectively); whereas from 2000 to 2004 a significant (P=.02) decline occurred only 65.4% of CNS isolates were sensitive to gatifloxacin and moxifloxacin (MIC 90 : 32 µg/ml for both agents). The authors concluded that declining in vitro susceptibility to gatifloxacin and moxifloxacin may have important implications for the prevention and treatment of postoperative endophthalmitis. 30 Fluoroquinolone activity against ciprofloxacin-resistant MSSA, MRSA, MSSE, MRSE A recent study showed that the in vitro potency of advancedgeneration fluoroquinolones varies against ciprofloxacin-resistant (CR) S aureus (MSSA and MRSA) and CR S epidermidis (methicillin-sensitive S epidermidis [MSSE] and MRSE) ocular isolates. Specifically, MIC 90 values (µg/ml) against all CR S aureus (MSSA and MRSA) isolates were 4 for besifloxacin, 32 for gatifloxacin, 32 for moxifloxacin, 256 for ciprofloxacin, and 256 for levofloxacin. MIC 90 values (µg/ml) against all CR S epidermidis (MSSE and MRSE) isolates were 4 for besifloxacin, 32 for gatifloxacin, 32 for moxifloxacin, 64 for ciprofloxacin, and 256 for levofloxacin. 38 Antibiotic activity against methicillin-resistant and fluoroquinolone-resistant isolates The in vitro potency of different antibiotics against various ocular pathogens was evaluated in a recent prospective, multicenter surveillance study. Against MRSA and ciprofloxacin nonsusceptible isolates, the MIC 90 values (µg/ml) were as follows: besifloxacin, 4; moxifloxacin, 16; ciprofloxacin, 256. Against MR-CNS and ciprofloxacin nonsusceptible isolates, the MIC 90 values (µg/ml) were as follows: besifloxacin, 4; moxifloxacin, 32; ciprofloxacin, 64. 36 Fluoroquinolones versus S epidermidis To determine in vitro activity of besifloxacin, vancomycin, moxifloxacin, and ciprofloxacin against S epidermidis isolates, Chang and colleagues randomly recovered 42 isolates from vitreous (30), anterior chamber (4), and surface disease (8). Against various methicillin-resistant, moxifloxacin-resistant, and ciprofloxacinresistant S epidermidis isolates, the MIC 90 values (ug/ml) were vancomycin, 2; besifloxacin, 4; moxifloxacin, 64; and ciprofloxacin, 64. MBC 90 values (the concentration at which 90% of isolates were killed) were vancomycin, 2; besifloxacin, 4; moxifloxacin, 128; and ciprofloxacin, 256. 54 Ocular penetration of 2 advanced-generation fluoroquinolones An in vivo (rabbit) model found cornea and aqueous humor concentrations of moxifloxacin to be higher than those of besifloxacin following topical instillation of a single drop. The C max (at 15 minutes) and AUC values in the cornea were (mean ± SE) 44 ± 11 µg/ml and 35 µg*h/ml for moxifloxacin and 13 ± 2 µg/ml (at 30 minutes) and 30 µg*h/ml for besifloxacin. In the aqueous humor, moxifloxacin C max (at 30 minutes and 1 hour) and AUC values were 2.0 ± 0.2 µg/ml and 5 µg*h/ml, respectively, and 0.2 ± 0.1 µg/ml (at 2 hours) and 1 µg*h/ml for besifloxacin. 55 Fluoroquinolones with activity against emerging penicillinase-producing, penicillin-resistant streptococcal species Topical administration of gatifloxacin, moxifloxacin, and besifloxacin prior to and after penicillin-resistant S pneumoniae endophthalmitis infection in rabbits comparably reduced clinical scores and bacterial load in the aqueous humor. 56 Aqueous humor concentrations following cataract surgery The aqueous humor concentration of antibiotic was detectable in 100% (23/23) of human subjects who received moxifloxacin compared with 40% (10/25) of subjects who received besifloxacin (P<.0001, Pearson s chi-square test) following routine preoperative topical dosing in cataract surgery. 57 Aqueous humor fluoroquinolone bioavailability using a dosing regimen that simulated prophylactic use after cataract surgery In an in vivo study by Peroni and colleagues, the newer-generation fluoroquinolones failed to demonstrate a significant aqueous humor bioavailability using a dosing regimen that simulated prophylactic use after cataract surgery. Rabbit eyes were topically treated with ciprofloxacin, levofloxacin, gatifloxacin, and moxifloxacin. Although not significant, moxifloxacin showed an initial (30 minutes and 1 hour posttreatment) trend toward superior aqueous bioactivity compared with all other tested formulations. However, at and following the second hour, the aqueous humor withdrawn from all treated eyes failed to demonstrate any bacteria inhibitory potential for the 4 tested formulations. 58 Fluoroquinolones and vancomycin for treating endophthalmitis caused by MSSA or MRSA In patients with culture-proven S aureus endophthalmitis, all MRSA isolates were sensitive to vancomycin. However, frequent MRSA resistance occurred with moxifloxacin, 38% (5/13), and gatifloxacin, 38% (5/13). 23 Fluoroquinolone concentration in conjunctiva The comparative pharmacokinetics of 1 drop of the advancedgeneration fluoroquinolones in human conjunctiva has been evaluated. The peak mean concentrations (15 minutes after dosing) were besifloxacin, 2.30 ± 1.42 µg/g; gatifloxacin, 4.03 ± 3.84 µg/g; and moxifloxacin, 10.7 ± 5.89 µg/g. All 3 fluoroquinolones were well tolerated and reached levels in the conjunctiva above the MIC 90 values of MRSA and MRSE strains for at least 2 hours. Mean residence time favored besifloxacin (4.7 hours), an outcome that was attributed to its vehicle, a polymeric mucoadhesive delivery system. Besifloxacin had the greatest AUC to MIC 90 ratio among MRSA and MRSE strains, a pharmacokinetic/pharmacodynamic ratio deemed to be clinically relevant to predict clinical outcome, microbiological success, and decreased potential for antibacterial resistance. 37 8

Experts regimens for perioperative topical antibiotic prophylaxis Dr McDonald: At present, there is no compelling evidence to support a certain pattern of preoperative topical antibiotic prophylaxis. But in the face of that lack of information, it would be interesting to find out what our panelists do. Dr O Brien: Antiseptics and antibiotics are utilized to decrease the likelihood of infection. An effective regimen is to apply an antibacterial agent, such as an advanced-generation fluoroquinolone, that is potent, with a rapid kill curve, on the day of surgery, in combination with an antiseptic. Day-of-surgery dosing eliminates the chance that the patient may contaminate the tip of the eye drop medication, which may be an additional source of infection. With topical anesthesia, the patient has the advantage to initiate dosing of the antibiotic immediately after surgery and continue dosing for a duration of approximately 7 days. Dr Akpek: I see many patients with poor ocular surface in the Dry Eye Clinic at Wilmer. In those with chronic blepharitis and/or significant dry eye, I treat the aqueous tear secretion as well as the meibum layer separately, approximately 6 to 8 weeks prior to surgery; thus, I aim to improve the tear film and optimize the ocular surface as an innate defense line to prevent infections and also to achieve the best visual outcomes after surgery. In the majority of patients with clinically significant dry eye condition, aqueous tear deficiency and meibomian gland dysfunction co-exist. 59 In patients with diabetes with a poor ocular surface or in patients with actively ongoing inflammatory ocular surface disorder such as graft-versushost disease, or mucous membrane pemphigoid, I initiate antibiotics approximately 1 week prior to the surgery to ensure that the surface is optimum prior to surgery. In patients with no other underlying issues or associated conditions, I dose on the day of surgery with an advanced-generation fluoroquinolone, 1 drop 3 or 4 times, about 5 to 10 minutes apart. I use a Vicryl suture to close the wound following the clear corneal incision. And I use a new bottle of the fluoroquinolone after the surgery for approximately 7 days. Dr McDonnell: I routinely treat the day of surgery and 1 week following. Dr McDonald: My regimen is the same, dosing 5 minutes apart, for a total of 4 doses, right before the surgery in an otherwise uncomplicated case. CONCLUSION Endophthalmitis is a highly feared and devastating complication of ophthalmic anterior segment procedures. The infection is most often caused by the patient s own external flora. Therefore, in addition to using effective sterile techniques, much effort is set forth to reduce the bacteria load on the ocular surface minimizing the risk of endophthalmitis-producing bacteria from overwhelming the eye s innate line of defense, entering the aqueous humor, and causing a sight-threatening infection. Regrettably, ocular pathogens are becoming increasingly resistant to commonly used antibiotics; thus, there is an ongoing need for evolving agents. Topical povidone-iodine is important in antisepsis. Surgeons must rely on surrogate end points from laboratory data because evidence-based guidance for antibiotic use in infection prophylaxis is lacking. Potent agents with the ability to obtain high and enduring concentrations at the infecting site may provide the greatest likelihood of preventing infection. Currently, the advanced-generation fluoroquinolones gatifloxacin, moxifloxacin, and the newer agent besifloxacin appear to possess the essential characteristics for effective use in infection prophylaxis for ocular procedures. Please turn to pages 11 and 12 for the CME Post Test and Activity Evaluation. 9