Eyeworld

EW NEWS & OPINION 18 A s an ophthalmology resi- dent, I live in uncertainty, always having more ques- tions than answers. At first I assumed this was a unique part of the resident condi- tion, but I soon realized that there are times when even the authorities (my attendings, textbooks, and the peer-reviewed literature) cannot pro- vide answers, and these situations intrigued me most of all. Such a situ- ation arose when I began to perform intravitreal injections and was taught to respectfully fear the specter of endophthalmitis during each step. Caution was necessary with every small movement, and yet the degree of risk attributable to each movement was far from clear. The process of questioning one's every move is an uncomfortable yet integral part of ophthalmic training, and I decided to embark on this path for the injection procedure. During the course of listing each step, identifying the variables in- volved in the step, and consulting the literature for advice on proper control of these variables, I came upon several areas where no consen- sus exists. While there was much to learn, I found myself reading most about the facet of the procedure that was of most concern to my patients: pain. Barring my secret fear of a re- flexive punch from a patient un- happy with his needle stick, it had not occurred to me that controlling my patients' pain would in any way affect safety. I knew that 2% lido- caine gel was gaining popularity as a topical anesthetic and offered com- parable anesthesia to subconjuncti- val lidocaine with less risk of subconjunctival hemorrhage and chemosis. 1 However, a retrospective series of all patients with clinically diagnosed endophthalmitis after cataract surgery at the Bascom Palmer Eye Institute from 2000-2004 suggested that ocular surface prepa- ration with 2% lidocaine gel prior to povidone-iodine was a risk factor for endophthalmitis, based on the use of this protocol in 2 of 7 patients de- scribed. 2 A laboratory experiment by Boden and colleagues later demon- strated that blood agar plates inocu- lated with standardized suspensions of bacteria grew as many colony forming units (CFUs) if lidocaine gel was used to coat the plates prior to povidone-iodine as they did if no antiseptic was used at all. 3 Despite these indications in the literature of the potential dangers of using lidocaine gel prior to antisep- sis, there was no suggestion that practice patterns had changed. One review highlighted the increasing popularity of gel anesthesia, 4 and one retina specialist told me that every retina surgeon in his institu- tion used lidocaine gel prior to povi- done-iodine for injections. This was also the case for most cataract sur- geries at a nearby surgery center. It was clear that either most physicians were not convinced or that the word had simply not spread—and in fact, I was unsure if I was convinced enough to begin spreading the word to my fellow residents. The most ef- fective way to combat these issues seemed to be to design a study to focus on the most practical ques- tions that one faces in considering the literature on lidocaine gel and endophthalmitis: 1) will the prob- lem be solved by using povidone-io- dine first, 2) can I prove it, and 3) for how long do I need to use the antiseptic before the gel in order to make it work? I began in a place most residents do, with lofty ideals and minimal practicality, but it didn't take long for the reality of the overall situa- tion to become apparent: The large patient numbers and resources re- quired for a randomized controlled trial were well beyond my means. The simplest way to exert the most control over my variables of choice would be in the laboratory. Since I'm no microbiologist and usually han- dle culture media only when needed to plate material scraped from a corneal ulcer, I sought help from two of my mentors, Dr. Anne Fung and Dr. Theodore Leng. They were happy to hear my ideas and help me figure out a feasible way to get from the bedside to the bench. We chose to use Boden and col- leagues' protocol as a starting point and modify it to answer our ques- tions. Since Staphylococcus epider- midis is the most commonly implicated organism in post-surgical endophthalmitis, 5 we inoculated blood agar plates with this bac- terium prior to applying 2% lido- caine gel and 5% povidone-iodine to the plates in various application se- quences. In addition, we included groups of plates in which the anti- septic had been present on the plate for 30 seconds before the gel (the length of time I had been taught to wait for povidone to dry) and in which it had only been present for 5 seconds (the length of time I had seen busy clinicians actually wait). After this, we incubated the plates for 24 hours at 37 degrees Celsius and documented the growth. December 2011 by Rishi Doshi, M.D. Povidone-iodine and lidocaine gel: Timing issues Fortunately, in this case, the vitrec- tomy probe could be visualized, reducing the risk of a "blind vitrec- tomy," as was needed for case 1. Her post-op course was routine. Although very rarely indicated, limited, single-port pars plana auto- mated vitrectomy may be necessary to safely perform cataract surgery in a crowded anterior segment. Con- comitant with or just afterward, the anterior chamber may be deepened with OVD and hypotony averted. Only small amounts of vitreous need to be removed in this tech- nique. The overall concept is to maintain normal IOP while expand- ing the anterior chamber by reduc- ing posterior segment volume. EW References 1. Machemer R. Symposium: pars plana vit- rectomy. Introduction. Trans Am Acad Ophthal- mol Otolaryngol 1976; 81:350-351. 2. Chang, DF. Pars plana vitreous tap for phacoemulsification in the crowded eye. J Cataract Refract Surg 2001; 27:1911-1914. Editors' note: Drs. Ceran, Fram, and Masket have no financial interests related to their comments. Contact information Ceran: basakbostanci@hotmail.com Fram: nicfram@yahoo.com Masket: sammasket@aol.com Vitrectomy continued from page 17 W ith three ophthalmology residency programs in the Bay Area, I often have residents observing surgery in my OR. As anyone who attends resident surgery knows, it is always fun to see how surgeons at the earliest stage of their careers think and what kinds of ques- tions they ask. I got into a discussion about povidone-iodine with one recent resident visitor, Rishi Doshi, M.D., from California Pacific Medical Center. I was struck by Rishi's knowledge and intellectual curiosity, and he referenced a recent study that he had initiated and published. For my column this month, I asked Rishi to describe the study and his motivation to conduct it for two reasons. First, there are practical les- sons from his study that should be of inter- est to every anterior segment surgeon. Secondly, Rishi reminds us that no one should assume that he cannot contribute meaningful work just because he is either a resident or practicing outside of an aca- demic setting. The essence of being a clini- cian scientist is to identify a question relating to patient care and then to try to answer or solve the problem with scientific study. It is great to see how this resident was able to answer some important ques- tions. David F. Chang, M.D., chief medical editor Chief medical editor's corner of the world continued on page 20

• Cover