INTRACAMERAL ENDOPHTHALMITIS PROPHYLAXIS: WHERE WE STAND

INTRACAMERAL ENDOPHTHALMITIS PROPHYLAXIS: WHERE WE STAND Surgeon dicu the practical implication of a recent meta-analyi. BY LISA BROTHERS ARBISSER, MD; FRANCIS S. MAH, MD; DAVID F. CHANG, MD; RICHARD KENT STIVERSON, MD; AND STEVE A. ARSHINOFF, MD, FRCSC Routine endophthalmiti prophylaxi with an intracameral (IC) antibiotic for cataract urgery i vatly underued in the United State compared with internationally. Since the ESCRS tudy, in 2007, 1 and it dicontinuation on ethical ground, owing to the tartling decreae in endophthalmiti in the treatment arm, many countrie have made IC antibiotic available, and ome have even made them mandatory. At the end of that landmark year, I began following the lead of Steve A. Arhinoff, MD, FRCSC, regarding the off-label ue of moxifloxacin, and I never again aw endophthalmiti after cataract urgery in thouand of patient. The uggeted moxifloxacin doage ha been adjuted upward to a minimum inhibitory concentration to overcome laboratory-meaured reitance of organim mot likely implicated in endophthalmiti. Firt, do no harm. There i ample evidence of the afety of IC moxifloxacin; I conducted one of many tudie upporting thi aertion. 2 Significant rik have come to light for other antibiotic choice but not for elf-preerved Vigamox (moxifloxacin HCl ophthalmic olution 0.5%, Alcon), depite the manufacturer legal labeling added to the bottle that read, not for intraocular ue. In addition to the enlightening meta-analyi by Bowen and colleague that i the focu of thi article, 3 I would like to draw reader attention to an editorial publihed in Ophthalmology in 2016 that detailed a poible way to urmount the main barrier to IC prophylaxi in the United State: the lack of an FDA-approved preparation. 4 Given that 3 million cataract urgerie are performed yearly in thi country, an etimated 2,000 eye could be aved by IC antibiotic prophylactic ue. Big data from the Aravind Eye Hopital Sytem 5 and elewhere are confirmatory for me and may be enough to convince other. Kaier Permanente acted on it own big data 6 prior to the evidence from India. It i worth remembering that topical antibiotic prophylaxi i alo an off-label practice and hardly without economic and biome-altering conequence. For urgeon who wih to ue IC prophylaxi and do not chooe a compounding route, Dr. Arhinoff ha provided a detailed how-to guide; it i a template that I believe ha benefited my patient (ee Dr. Arhinoff Method of Preparation). ASCRS ha announced it intention to perform a multicenter randomized propective trial that will require more than 75,000 patient to reach ignificance. Thi will hopefully provide a definitive anwer and lead to uniform adoption of intracameral antibiotic prophylaxi. I hope mot urgeon will not wait for the reult far into the future. CRST and I invited everal well-known expert in our field to interpret the inight provided by the recent meta-analyi and to comment on their own practice. Lia Brother Arbier, MD FRANCIS S. MAH, MD Nobody i going to perform a clinical trial comparing IC cefuroxime, moxifloxacin, and vancomycin to ee if one of thee agent i uperior to the other. A meta-analyi of the literature i the next bet thing; it help to decreae the biae preent in individual article to uncover general finding. The main take-away meage of the meta-analyi by Bowen and colleague are a follow. 3 In term of the volume of tudie and the volume of patient, IC cefuroxime ha the greatet amount of upport in the literature and vancomycin the lowet. Of the three agent, cefuroxime wa the leat effective, although it wa more efficaciou than topical drop. Moxifloxacin wa more effective than cefuroxime and wa aociated with le toxicity. Mot effective of the three agent wa vancomycin, but it ha been aociated with hemorrhagic occluive retinal vaculiti (HORV). Interetingly, topical medication did not appear to be of benefit in the meta-analyi. My preferred IC agent i moxifloxacin. I am not uing IC vancomycin for two reaon. Firt, HORV i a devatating complication, potentially wore than the endophthalmiti the antibiotic wa meant to prevent. Second, the AAO and the Center for Dieae Control iued guidance againt uing thi medication a a prophylactic agent. ESTABLISHING TRUE ALLERGY IC anaphylaxi ha been aociated with cephaloporin uch a cefuroxime 16 CATARACT & REFRACTIVE SURGERY TODAY MAY 2018

and cefazolin. At mot, approximately 10% of patient who are allergic to penicillin are alo allergic to cephaloporin. A hitory of penicillin allergy would therefore raie my concern over the ue of IC cefuroxime. Compared with beta-lactam antibiotic uch a the cephaloporin, an advantage of fluoroquinolone i that they are generally aociated with fewer allergie. The firt tep to etablihing true allergy i the hitory. To my mind, patient who tate that ome pill upet their tomach in the pat do not have a true allergy. If they ay they developed hive or a rah, then I am more cautiou. Becaue I work in a clinic, urgent care i nearby. In thee cae, I will therefore intill a ingle drop of a fluoroquinolone in the patient eye and oberve him or her for a reaction in the clinic. If nothing happen, I will plan to ue IC moxifloxacin on the day of urgery. Obviouly, a hitory of anaphylaxi i a clear contraindication to a medication. If a patient ha an allergy to fluoroquinolone but not to beta-lactam antibiotic, I will ue cefazolin 2.5 mg in 0.1 ml intead of moxifloxacin. DR. ARSHINOFF S METHOD OF PREPARATION SUPPLIED Moxifloxacin HCl ophthalmic olution 0.5% (Vigamox, Alcon) = 500 µg/0.1 ml GOAL 150 µg/0.1 ml (dilution: 3 part Vigamox + 7 part BSS [Alcon]) In other word, to achieve 150 µg/0.1 ml, imply dilute the eye drop to 30% concentration of upplied Vigamox METHOD Inject 0.3 to 0.4 ml Vigamox 150 µg/0.1 ml at the end of the cae = 450 to 600 µg 1.0 to 1.2 mg/ml in the anterior chamber (AC). Eentially, thi i an exchange of mot of the newly peudophakic AC volume (0.5 ml) with the Vigamox olution. The volume indicated i what i likely left in the AC at the end of urgery. DETAILED INSTRUCTIONS 1. From a new bottle of Vigamox, 3 ml i withdrawn into a 12-mL yringe with a terile needle. 2. From a new 15-mL bottle of BSS, 7 ml i withdrawn into the ame yringe. The two ubtance will be mixed by the turbulence of apiration and rolling the yringe. The circulating nure inject 0.8 ml of the Vigamox olution into a medicine cup on the urgical tray. 3. The crub nure draw up 0.6 ml of the Vigamox olution into a tuberculin yringe to hand to the urgeon. 4. The urgeon expel 0.1 ml to be ure there are no bubble and then inject 0.3 to 0.4 ml via the ideport inciion, a the lat tep of cataract urgery, under the dital edge of the capulorrhexi. Then, a the eye i exited, a final purt of injection i ued at the inciion to hydrate the inciion and enure the AC i left preurized (Figure 1 and 2). Thi i a planned exchange of mot of the AC content and i therefore eay to do. PERSONAL EXPERIENCE Dr. Arhinoff ha ued variation of thi method in more than 8,700 cae and oberved no toxicity to date. ACCESSING IC ANTIBIOTICS More US cataract urgeon would ue IC antibiotic if an FDA-approved drug were available. Becaue that i not the cae, one option i to obtain an IC formulation from a compounding pharmacy uch a Leiter, a hopital pharmacy, or a company uch a Imprimi Pharmaceutical or Ocular Science. A econd option i either to dilute commercially available moxifloxacin or to ue it traight out of the bottle. The literature decribe uing anything from 50 to 500 mg in 0.1 ml at the end of urgery. A far a I know, there have been no publihed report about uing dilute generic moxifloxacin, and Moxeza (moxifloxacin HCl ophthalmic olution 0.5%, Alcon) hould definitely not be injected intracamerally. Figure 1. Via the ideport inciion, Dr. Arhinoff adminiter an intracameral injection of moxifloxacin. He favor a hockey tick cannula becaue he can perform the injection lowly, thu deepening the capular bag and allowing him to rotate the IOL lightly to the deired alignment. He continue the injection a the intrument exit the eye, enuring an exchange of mot of the aqueou with the moxifloxacin olution and preurization of the AC. If concerned that the ideport inciion will leak, Dr. Arhinoff will hydrate it prior to injecting the moxifloxacin, but uually he imply hydrate the inciion a the cannula exit the eye. Figure 2. During the injection, the cannula (1, green) depree the IOL lightly o that the moxifloxacin olution bathe the IOL within the capular bag. A a equetered pace, the capular bag i an excellent location for bacterial growth. The urgeon perform final hydration of the inciion (2, purple) while removing the cannula from the eye. MAY 2018 CATARACT & REFRACTIVE SURGERY TODAY 17

DAVID F. CHANG, MD The recent meta-analyi by Bowen and colleague pool data from 17 tudie publihed over the pat 2 decade. Apart from the ESCRS randomized clinical trial, 1 thee were all obervational tudie. When pooled, however, more than 925,000 eye can be analyzed, providing a large amount of data upporting the afety and efficacy of routine IC antibiotic prophylaxi for phacoemulification. One of the more intereting apect i the comparion of different IC antibiotic agent. The meta-analyi include a large obervational tudy of IC moxifloxacin from the Aravind Eye Hopital Sytem that my colleague and I publihed in 2017, which ignificantly expand the pooled data for thi drug. 5 The comparative endophthalmiti rate were 0.033% for cefuroxime, 0.015% for moxifloxacin, and 0.011% for vancomycin. Thi ugget that both moxifloxacin and vancomycin provide better coverage than cefuroxime. The data alo indicate that adding topical antibiotic prophylaxi may provide no benefit over IC antibiotic prophylaxi alone. A joint tak force formed by the ASCRS and the American Society of Retina Specialit, which I cochaired, concluded that HORV wa a type III hyperenitivity to vancomycin. 7 Although we uggeted that the ue of IC vancomycin hould be left to the individual urgeon dicretion, the AAO Cataract Preferred Practice Pattern and the FDA have ince dicouraged uing IC vancomycin becaue of the rik of HORV. I peronally ued IC vancomycin routinely for 18 year without any known complication. Should a cae of HORV ever occur, however, thee trong pronouncement would make it difficult to defend the off-label ue of IC vancomycin when alternative antibiotic were available. Baed on our Aravind tudie, I have witched to IC moxifloxacin upplied by a 503B compounding pharmacy. Imprimi Pharmaceutical and Leiter are the larget 503B outourcing facilitie that can upply moxifloxacin for IC injection. RICHARD KENT STIVERSON, MD The meta-analyi by Bowen and colleague confirm the efficacy of IC antibiotic in preventing potcataract urgery endophthalmiti, 3 a hown in the very large ESCRS, 1 Kaier, 6 and Aravind tudie. 5 Although I am not qualified to comment on the tatitical rigor of thi meta-analyi, rigor i crucial and eem to be well documented in thi tudy. The number of metaanalye in medicine ha increaed markedly from 334 worldwide in 1991 to 5,000 in 2012 and 10,000 in 2017. It i important to note that metaanalye are only a good a the trial included; in other word, including large number of low-quality tudie doe not make a meta-analyi more valid. Bowen and colleague excluded everal IC tudie from their metaanalyi, which give me more confidence in their finding. Propective randomized clinical trial remain the gold tandard for evidencebaed medicine, but their prohibitive cot i uch that high-quality, reliable meta-analye will aume increaing importance in anwering doctor big quetion. Meta-analyi gave rie to big data. In my opinion, changing tatitical ignificance to P <.005 would make a lot of publihed reearch more valid and worthier of conideration. Although a higher threhold for tatitical ignificance might increae the chance of a fale negative, multiple tudie even with maller ample ize would reduce the chance of fale poitive. I highly recommend that reader watch the video, I Mot Publihed Reearch Wrong? by Derek Muller, BSc, PhD (bit.ly/2bbi6of). Even randomized, propective tudie hould be regarded with kepticim when the ample ize i mall and ignificance i baed on a probability of le than 0.05. I do not believe the work by Bowen and colleague will change US cataract urgeon IC choice. Cefuroxime will not be widely adopted in the United State becaue of the rik of anaphylaxi and bacterial reitance profile that are ignificantly different from thoe in Europe (in part becaue of antibiotic abue in the United State). Kaier tarted with cefuroxime becaue that wa the antibiotic of the preexiting tudie, but many Kaier doctor have witched to moxifloxacin. The pecter of HORV i uch that mot ophthalmologit have abandoned or will abandon IC vancomycin. Moxifloxacin will be the mot commonly ued IC antibiotic in the United State. I topped uing topical antibiotic a year ago. Although I undertand the belt-and-upender approach of uing both topical and IC antibiotic that many urgeon advocate, there i imply no evidence to upport it. Moreover, a large number of cataract urgery patient receive anti-vegf injection or will in the future. The retina literature ugget that repeated prophylactic topical antibiotic might be contraindicated. 8 A week of topical antibiotic prophylaxi after cataract urgery would certainly fall under thi concern. I ue the ame formulation a Dr. Arhinoff (ee Dr. Arhinoff Method of Preparation). Like him, I perform immediately equential bilateral urgery. The importance of IC antibiotic prophylaxi cannot be overtated. Hi formulation allow more volume at a deired concentration o that 18 CATARACT & REFRACTIVE SURGERY TODAY MAY 2018

mot of the aqueou can be replaced with moxifloxacin olution. There i uually enough left over that I can alo perform ome tromal hydration of the inciion. That aid, the ucce of taggering number of patient in the Aravind tudy, 5 with uch mall volume of moxifloxacin, ugget that my concern are unfounded. AT A GLANCE A meta-analyi compared the afety and efficacy of intracameral cefuroxime, moxifloxacin, and vancomycin in preventing endophthalmiti after phacoemulification cataract urgery. The meta-analyi pooled data from 17 tudie publihed over the pat 2 decade and included more than 925,000 eye. Thi large amount of data upported the afety and efficacy of routine IC antibiotic prophylaxi. STEVE A. ARSHINOFF, MD, FRCSC One of the great trength of the reearch by Bowen and colleague i that it included obervational tudie a well a the ole randomized controlled tudy in thi field, 1 yielding a much broader depth of information than i available from the recent Cochrane review on the ame ubject. 9 Bowen and colleague concluded that, whether or not potoperative topical antibiotic were ued, IC antibiotic all achieved tatitically ignificant decreae in rate of infection, with no agent being tatitically ignificantly better than the other two. (Statitical analyi alo failed to demontrate any additional benefit of potoperative topical antibiotic.) The tudy by Bowen and colleague i excellent, but the problem with meta-analye i that they mut, by their methodology, group data from different tudie conducted at different time and analyze thi aggregate data. Although thi method produce ueful reult, the bet undertanding and concluion come from tudying the article ucceively and rereading previou one to dicover what error might have been made in each tudy and corrected by a later author or group. For example, two tudie from Japan failed to demontrate a tatitically ignificant benefit of IC moxifloxacin, but when they are read carefully, it i becaue ubtherapeutic doe of the drug were ued. 10,11 The equential analyi approach thu gradually clarifie the anwer to the quetion all of u really want to ak and i more ueful but much more time-conuming for the reader than meta-analye. I will addre ome of thee quetion here. DO IC ANTIBIOTICS WORK? Adding up tudie on the IC adminitration of cefuroxime, moxifloxacin, and vancomycin for cataract urgery endophthalmiti prophylaxi yield an aggregate of more than 6 million urgerie in the databae. Every tudy (except two in which too low a drug doe wa ued 10,11 ) demontrated a reduction in potoperative endophthalmiti cae of about 80% when IC prophylaxi wa ued. The efficacy of IC antibiotic i no longer in quetion. WHICH IC ANTIBIOTIC IS BEST? In term of efficacy, cefuroxime, moxifloxacin, and vancomycin all produce imilar reduction in the incidence of potoperative endophthalmiti. A the databae grow, I expect moxifloxacin to turn out to be the bet but only when the ample ize become huge becaue the difference in efficacy i mall. The ue of cefuroxime i of concern becaue dilution error have been hown to caue evere ocular damage. 12 Moxifloxacin i much eaier to dilute, and no complication from a dilution error have ever been reported. Another drawback to cefuroxime i that, a a relative of penicillin, it carrie a much higher rik of allergy than the other two drug. Vancomycin ha recently been aociated with a mall rik of the devatating allergic complication of HORV. With vancomycin, it thu appear that we are trading one rare devatating rik for another. Moxifloxacin i therefore the afet of the three drug. Alo worth noting i that data from the Swedih national databae recently revealed that failure with IC cefuroxime are often Enterobacter cae, which are difficult to treat and uually reult in devatating blindne. 13 In contrat, failure with moxifloxacin are generally eay-to-treat cae of Staphylococcu, and thee patient generally retain good viion. Furthermore, Libre and Mathew recently demontrated that only moxifloxacin i effective againt all 18 train of endophthalmiti iolate from the Bacom Palmer Eye Intitute. 14 SHOULD IC ANTIBIOTICS BE USED FOR IMMEDIATELY SEQUENTIAL BILATERAL? Abolutely. In 2009, the International Society of Bilateral Cataract Surgeon iued a recommendation that IC antibiotic be ued for immediately equential bilateral cataract urgery. The ociety tudy of endophthalmiti after bilateral cataract urgery found that, in more than 125,000 eye, the infection rate when IC antibiotic were MAY 2018 CATARACT & REFRACTIVE SURGERY TODAY 19

ued wa 1:16,890, the lowet infection rate reported in any tudy to date. 15 I have ued IC moxifloxacin for almot 9,000 conecutive eye with no negative ocular ide effect of the drug. I wa the firt in the world to ue IC moxifloxacin and to propoe it ue generally. Early on, when I wa uing a low doe (100 µg in 0.1 ml), one patient developed a reitant train of Staphylococcu epidermidi endophthalmiti, but with routine endophthalmiti treatment, viual acuity promptly improved to 20/25. I have ince increaed the doe I ue and recommend to cover even the mot reitant train (ee Dr. Arhinoff Method of Preparation). About 80% of my lat 14,000 cataract urgerie have been immediately equential bilateral cataract urgery. n 1. Endophthalmiti Study Group, European Society of Cataract & Refractive Surgeon. Prophylaxi of potoperative endophthalmiti following cataract urgery: reult of the ESCRS multicenter tudy and identification of rik factor. J Cataract Refract Surg. 2007;33(6):978-988. 2. Arbier LB. Safety of intracameral moxifloxacin for prophylaxi of endophthalmiti after cataract urgery. J Cataract Refract Surg. 2008;34(7):1114-1120. 3. Bowen RC, Zhou AX, Bondalapati S, et al. Comparative analyi of the afety and efficacy of intracameral cefuroxime, moxifloxacin and vancomycin at the end of cataract urgery: a meta-analyi [publihed online ahead of print January 11, 2018]. Br J Ophthalmol. doi:10.1136/bjophthalmol-2017-311051. 4. Javitt JC. Intracameral antibiotic reduce the rik of endophthalmiti after cataract urgery: doe the preponderance of the evidence mandate a global change in practice? Ophthalmology. 2016;123(2):226-231. 5. Haripriya A, Chang DF, Ravindran RD. Endophthalmiti reduction with intracameral moxifloxacin prophylaxi: analyi of 600,000 urgerie. Ophthalmology. 2017;124(6):768-775. 6. Shortein NH, Winthrop KL, Herrinton LJ. Decreaed potoperative endophthalmiti rate after intitution of intracameral antibiotic in a Northern California eye department. J Cataract Refract Surg. 2013;39(1):8-14. 7. Witkin AJ, Chang DF, Jumper JM, et al. Vancomycin-aociated hemorrhagic occluive retinal vaculiti: clinical characteritic of 36 eye. Ophthalmology. 2017;124(5):583-595. 8. Bande MF, Manilla R, Pata MP, et al. Intravitreal injection of anti-vegf agent and antibiotic prophylaxi for endophthalmiti: a ytematic review and meta-analyi. Sci Rep. 2017;7(1):18088. 9. Gower EW, Lindley K, Tulenko SE, et al. Perioperative antibiotic for prevention of acute endophthalmiti after cataract urgery. Cochrane Databae Syt Rev. 2017;2:CD006364. 10. Matuura K, Suto C, Akura J, Inoue Y. Bag and chamber fluhing: a new method of uing intracameral moxifloxacin to irrigate the anterior chamber and the area behind the intraocular len. Graefe Arch Clin Exp Ophthalmol. 2013;251(1):81-87. 11. Matuura K, Miyohi T, Suto C, et al. Efficacy and afety of prophylactic intracameral moxifloxacin injection in Japan. J Cataract Refract Surg. 2013;39(11):1702-1706. 12. Olavi P. Ocular toxicity in cataract urgery becaue of inaccurate preparation and erroneou ue of 50mg/ml intracameral cefuroxime. Acta Ophthalmol. 2012;90(2):e153-154. 13. Monta P. Antibiotic reviited. Paper preented at: XXXV Congre of the ESCRS; October 8, 2017; Libon, Portugal. 14. Libre PE, Mathew S. Endophthalmiti prophylaxi by intracameral antibiotic: in vitro model comparing vancomycin, cefuroxime, and moxifloxacin. J Cataract Refract Surg. 2017;43(6):833-838. 15. Arhinoff SA, Batianelli PA. Incidence of potoperative endophthalmiti after immediate equential bilateral cataract urgery. J Cataract Refract Surg. 2011;37(12):2105-2114. LISA BROTHERS ARBISSER, MD n Emeritu poition, Eye Surgeon Aociate, the Iowa and Illinoi Quad Citie n Adjunct Profeor, John A. Moran Eye Center, Univerity of Utah, Salt Lake City n Member, CRST Adviory Board n drlia@arbier.com n Financial dicloure: None STEVE A. ARSHINOFF, MD, FRCSC n Directing Partner, York Finch Eye Aociate, Toronto n Aociate Profeor, Department of Ophthalmology and Viion Science, Univerity of Toronto n Ifix2i@gmail.com n Financial dicloure: Paid conultant (Alcon) DAVID F. CHANG, MD n Clinical Profeor, Univerity of California, San Francico n Private practice, Lo Alto, California n CRST Editor Emeritu n dceye@earthlink.net n Financial dicloure: None FRANCIS S. MAH, MD n Scripp Health, La Jolla and San Diego, California n mah.franci@cripphealth.org n Financial dicloure: Conultant (Alcon); Medical adviory board (Ocular Science) RICHARD KENT STIVERSON, MD n Colorado Permanente Medical Group, Lone Tree, Colorado n richard.tiveron@kp.org n Financial dicloure: None 20 CATARACT & REFRACTIVE SURGERY TODAY MAY 2018