arvo 2012 annual meeting abstracts

ARVO 2012 Annual Meeting Abstracts

Copyright 2012 by the Association for Research in Vision and Ophthalmology, Inc., all rights reserved. For permission to reproduce any abstract, contact the ARVO Office at [email protected].

234 Endophthalmitis Monday, May 7, 2012, 8:30 AM - 10:15 AM Hall B/C Poster Session Program #/Board # Range: 1674-1698/D1021-D1045 Organizing Section: Retina+

Program Number: 1674 Poster Board Number: D1021 Presentation Time: 8:30 AM - 10:15 AM Comparison Of Membrane Filter System And Blood Culture Bottles For Culturing Vitrectomy Samples Aleksandra V. Rachitskaya1, Harry W. Flynn, Jr.2, Darlene Miller3. 1Bascom Palmer Eye Inst, University of Miami, Miami, FL; 2Bascom Palmer Eye Inst, University of Miami, Miami, FL; 3Bascom Palmer Eye Institute, Univ of Miami Miller Sch of Med, Miami, FL. Purpose: To evaluate two methods of culturing vitreous samples obtained during vitrectomy: a membrane filter system and blood culture bottles Methods: This is a prospective study of vitreous samples from patients who underwent vitrectomy for clinically suspected infectious endophthalmitis or open globe injuries. The microbiological outcomes were compared between the vitrectomy samples from the same patient, which were cultured both on a membrane filter system and in a blood culture bottle. The culture results were compared at 24 hours, 48 hours, 1 week, and 2 weeks. Results: From September 2009 to April 2011, a total of 92 patients underwent vitrectomy for clinically suspected endophthalmitis or open globe injuries and had their vitrectomy samples processed by the Bascom Palmer Eye Institute microbiology laboratory. Out of 92 samples, 29% were positive in the membrane filter group and 25% were positive in the blood culture bottle group. In the culture positive samples, the rates of positivity comparing the membrane filter group versus blood culture bottles group were the following: at 24 hours (44% versus 30%), at 48 hours (41% versus 44%), at 1 week (15% versus 26%). No culture positivity was observed at 2 weeks. At 24 hours, 7 samples had identical microbiological outcomes between the membrane filter and the blood culture bottle groups, 5 grew on the filter paper but not in the bottle. By 48 hours, 16 samples matched (9 Staphylococcus, 3 Streptococcus, 2 Candida, 1 Gram-negative, 1 Gram-positive species), 7 organisms (2 Gram-negative, 2 Staphylococcus, 1 Streptococcus, 1 Mycobacterium, 1 Aspergillus species) grew on filter paper only, and 1 organism (Streptococcus species) grew in the bottle only. Out of 8 organisms that grew at 1 week, three were Propionibacterium acnes and two of those grew both on filter paper and in the bottle. Conclusions: The direct inoculation of blood culture bottles may be an acceptable adjunctive or alternative technique to more complicated membrane filter system for vitrectomy samples in infectious endophthalmitis. The outcomes of the two methods of vitrectomy sample culturing were similar at 48 hours. Blood culture bottle vitrectomy sample culturing is easy to perform as compared to membrane filter system that requires trained microbiology personnel to perform and evaluate. Commercial Relationships: Aleksandra V. Rachitskaya, None; Harry W. Flynn, Jr. , None; Darlene Miller , None Support: None

Program Number: 1675 Poster Board Number: D1022 Presentation Time: 8:30 AM - 10:15 AM Bacterial Susceptibility Profiles in Trauma Associated Endophthalmitis Duncan A. Friedman1, Mark L. Hill2, Sejal Amin2, Andrew Bartlett2, Richard M. Feist1, John O. Mason, III1, Martin L. Thomley1, Michael A. Albert, Jr.3, Jacob J. Yunker1, Tracy L. Emond1. 1Retina Consultants of Alabama, Birmingham, AL; 2Department of Ophthalmology, University of Alabama at Birmingham, Birmingham, AL; 3Retina Consultants of Alabama, PC, Birmingham, AL. Purpose: Previous studies have pointed out the need for broad antibiotic coverage in endophthalmitis associated with recent ocular trauma. No specific guidelines exist as to which antibiotics would serve as adequate coverage for the most common bacteria seen in traumatic endophthalmitis. Methods: A retrospective chart review of all cases of endophthalmitis associated with trauma was performed for all patients presenting to a Level 1 Eye Trauma hospital between 2006 and 2010. Data included patient demographics, nature of injury, source of culture, and susceptibility profiles for positive bacterial cultures. Positive cultures were analyzed for specifics of mechanism of injury to characterize suspected pathogens. An antibiogram was constructed from all positive cultures. Results: A total of 33 cases of trauma associated endophthalmitis occurred over the four year period observed. Fifty-six percent of all trauma associated endophthalmitis cultures grew specific bacteria. Of the positive cultures, 39% grew multiple microbes. Ninety-three percent of these organisms were gram-positive bacteria. All Gram positive organisms were susceptible to vancomycin while all Gram negative organisms were susceptible to 3rd generation cephalosporins. Conclusions: Endophthalmitis associated with recent trauma can be caused by both Gram positive and Gram negative organisms. Many times, these injuries result in multiple bacteria contributing to the overall inflammatory milieu. Susceptibility

testing reveals that most organisms are susceptible to vancomycin with no resistant strains found. Adequate broad spectrum coverage can be achieved with the use of intraocular vancomycin and ceftazidime. Commercial Relationships: Duncan A. Friedman, None; Mark L. Hill , None; Sejal Amin, None; Andrew Bartlett , None; Richard M. Feist, None; John O. Mason, III , None; Martin L. Thomley , None; Michael A. Albert, Jr. , None; Jacob J. Yunker, None; Tracy L. Emond, None Support: Research to Prevent Blindness

Program Number: 1676 Poster Board Number: D1023 Presentation Time: 8:30 AM - 10:15 AM Clinical outcomes and antibiotic susceptibilities of Staphylococcus aureus endophthalmitis Jonathan I. Huz1, Krishna Mukkamala2, Ivelisse Rodriguez Pagan1, David Ritterband3, Mahendra Shah3, Ronald C. Gentile4, Michael Engelbert2. 1New York University School of Medicine, New York, NY; 2Vitreous Retina Macula Consultants of New York/New York University School of Medicine/New York Eye and Ear Infirmary, New York, NY; 3New York Eye and Ear Infirmary, New York, NY; 4Ophthal-NY Medical College, New York Eye & Ear Infirmary, New York, NY. Purpose: To compare the antibiotic susceptibilities and final visual acuity of endophthalmitis caused by methicillin-resistant S. aureus (MRSA) versus methicillin-sensitive S. aureus (MSSA). Methods: Microbiological and clinical records of 35 eyes of 35 patients with S. aureus endophthalmitis treated at The New York Eye and Ear Infirmary since Jan. 1st, 1998 were reviewed. Data collected included antibiotic susceptibilities and visual acuity (VA) at presentation, and 3, 6, and ≥12 months after treatment. All eyes received intravitreal antibiotic treatment as a “tap and inject” or following a pars plana vitrectomy (PPV). Eyes were divided into two groups based on methicillin resistance and Fisher’s exact test was used to compare the MSSA and MRSA groups. Results: Of the 35 eyes, 15 (43%) were in the MRSA group and 20 (57%) were in the MSSA group. Mean age in years was 70 in the MRSA group and 69 in the MSSA group. There were 40% men in the MRSA group compared to 35% in the MSSA group (p=1.0). Antibiotic susceptibilities for both groups are shown in Table 1. The known sources of endophthalmitis (n=34) were cataract surgery in 18 (53%), intravitreal injection in 4 (12%), penetrating keratoplasty in 3 (9%), retinal surgery in 2 (6%), trauma in 2 (6%), endogenous in 2 (6%), bleb associated in 2 (6%) and glaucoma surgery in 1 (3%). PPV was performed on 12/15 (80%) of the MRSA and 16/19 (84%) of the MSSA eyes (p=1.0). Median presenting VA was hand motions for both the MRSA and MSSA groups. Mean LogMAR VA at presentation was 3.9 in the MRSA group and 3.5 in the MSSA group (p=0.40). There was no significant difference in VA between the MRSA and MSSA groups at any time point. A final VA of better than 20/200 at ≥ 12 months was achieved in 7/11 (64%) MRSA eyes compared to 6/12 (50%) MSSA eyes (p=0.68). A final VA of 20/50 or better at ≥ 12 months was achieved in 7/11 (64%) MRSA eyes compared to 4/12 (33%) MSSA eyes (p=0.22). Conclusions: There was no significant difference in presenting or final VA between the MRSA and MSSA groups. Resistance to fourth generation fluoroquinolones was present in over 85% of MRSA isolates and in no MSSA isolates. Resistance to the third generation fluoroquinolone levofloxacin was seen in 69% of MRSA strains and in only one MSSA strain. No strains were resistant to



vancomycin or gentamicin.

Commercial Relationships: Jonathan I. Huz, None; Krishna Mukkamala , None; Ivelisse Rodriguez Pagan, None; David Ritterband, None; Mahendra Shah, None; Ronald C. Gentile, None; Michael Engelbert, Genentech (C) Support: The Macula Foundation, Inc.

Program Number: 1677 Poster Board Number: D1024 Presentation Time: 8:30 AM - 10:15 AM Endogenous Endophthalmitis caused by Yeast: Risk Factors, Management Strategies, and Visual Acuity Outcomes Jayanth Sridhar, Harry W. Flynn, Jr., Darlene Miller, Thomas Albini. Ophthalmology, Bascom Palmer Eye Institute, Miami, FL. Purpose: To report the risk factors, management strategies, and visual outcomes in patients with culture-proven endogenous endophthalmitis caused by yeast. Methods: A non-randomized consecutive case series of all patients with culture-proven unilateral or bilateral endogenous endophthalmitis caused by yeast at the Bascom Palmer Eye Institute between January 1, 1990 and September 30, 2011 Results: Study criteria were met in 51 eyes of 39 patients. Patients had a mean follow-up of 16.7 months from presentation to last follow-up visit. Decreased vision was the most common presenting symptom (74.3%) followed by redness (46.1%) and pain (38.5%). 65.8% of patients had a current or recent (< 6 months) hospitalization at the time of diagnosis. Recent surgery was the most common identified risk factor (31.6%) followed by cancer (23.7%). None of the patients were diagnosed on routine screening for fungemia. Diffuse anterior and posterior inflammation was the most common finding on examination of affected eyes (64.7%) followed by focal posterior inflammation (23.5%). Hypopyon was noted in only 5.8% of affected eyes. Initial treatment consisted of a combination of medical and surgical treatment in 33.3% of eyes. 37.2% of eyes received medical treatment consisting of systemic antifungal therapy, intravitreal injection of antifungals, or both. 43.1% of eyes underwent vitrectomy with or without lensectomy, with or without intravitreal injection. The most common organism cultured by vitreous aspirate or vitrectomy specimen was Candida albicans (87.2%) followed by Candida tropicalis (7.7%) and Cryptococcus neoformans (5.1%). Blood cultures were positive in 17.9% of patients. Visual acuity on presentation was 20/200 or better in 41.3% of eyes. Visual acuity of 20/200 or better was present in 61.3% of eyes at last follow-up. Retinal detachment occurred in 29.4% of eyes. Conclusions: In this study of endogenous endophthalmitis caused by yeast, Candida albicans was the predominant etiology. Decreased vision was the most common presenting symptom. Endogenous endophthalmitis is associated with poor

visual acuity outcomes despite early and appropriate management. Retinal detachment is a frequent event in the clinical course. Commercial Relationships: Jayanth Sridhar, None; Harry W. Flynn, Jr. , None; Darlene Miller , None; Thomas Albini, None Support: None

Program Number: 1678 Poster Board Number: D1025 Presentation Time: 8:30 AM - 10:15 AM Infectious Keratitis Progressing To Endophthalmitis: A 15-year-study Of Risk Factors, Microbiology, And Clinical Outcomes Christopher R. Henry, Harry W. Flynn, Jr., Darlene Miller, Richard K. Forster, Eduardo C. Alfonso. Bascom Palmer Eye Institute, Miami, FL. Purpose: To describe the incidence, microbiology, risk factors and clinical outcomes of a consecutive series of patients with infectious keratitis progressing to endophthalmitis. Methods: Non-randomized, consecutive case series. Patients with positive corneal and intraocular cultures (anterior chamber and/or vitreous) between January 1, 1995 and December 31, 2009 were included in the study. Results: Over the period of the study, a total of 9934 corneal cultures were performed for suspected infectious keratitis. Only 52 of these patients (0.5%) progressed to culture-proven endophthalmitis. All eyes had documented keratitis before the development of endophthalmitis. Fungi (n=28) were the most common responsible organism followed by gram positive bacteria (n=14) and gram negative bacteria (n=10). All gram positive isolates were sensitive to vancomycin and all gram negative isolates were sensitive to amikacin, gentamicin, and ceftazidime. The use of topical steroids was common (39/52 [75%]) and represented the most common risk factor identified in the current study. Other risk factors included previous surgery (33/52 [63%]), dry eye syndrome (15/52 [29%]), relative immune suppression (11/52 [21%]), organic matter trauma (9/52 [17%]) and contact lens wear (5/52 [10%]). There were 27 patients in which a primary infectious keratitis developed into endophthalmitis, and 25 patients in which an infectious keratitis adjacent to a previous surgical wound developed into endophthalmitis. Patients in the primary keratitis group were more likely to be male (22/27 [81%] vs. 9/25 [36%], p<0.001), have a history of organic matter trauma (8/27 [30%] vs. 1/25 [4%]), p=0.016), and have a fungal etiology (21/27 [78%] vs. 7/25 [28%], p<0.001). Patients in the surgical-wound-associated group were more likely to have used topical steroids (22/25 [88%] vs. 17/27 [63%], p=0.037) and to have a bacterial causative organism (18/25 [72%] vs. 6/27 [22%], p<0.001). A visual acuity (VA) of ≥ 20/50 was achieved in 9/52 patients (17%) and a VA≥20/400 was achieved in 15/52 patients (29%), however, 35/52 patients (65%) had a VA < 5/200 at last follow up visit. Enucleation was ultimately performed in 15/52 patients (29%). Conclusions: Progression of infectious keratitis to endophthalmitis is relatively uncommon. The current study suggests that patients at higher risk for progression to endophthalmitis include patients using topical corticosteroids, patients with a fungal keratitis, and patients with infectious keratitis developing adjacent to a previous surgical wound. Patients with sequential keratitis and endophthalmitis have generally poor visual outcomes. Commercial Relationships: Christopher R. Henry, None; Harry W. Flynn, Jr. , None; Darlene Miller , None; Richard K. Forster, None; Eduardo C. Alfonso, None Support: None

Program Number: 1679 Poster Board Number: D1026 Presentation Time: 8:30 AM - 10:15 AM Histopathologic Correlation of Enucleated Globes following Aspergillus Endophthalmitis after Uncomplicated Cataract Surgery Luis J. Haddock, Sander R. Dubovy, Harry W. Flynn, Jr.. Ophthalmology, Bascom Palmer Eye Institute, Miami, FL. Purpose: To demonstrate the histopathological findings of viable fungal organisms in multiple intraocular structures following enucleation for Aspergillus endophthalmitis after uncomplicated cataract surgery. Methods: Consecutive case series of two patients with Aspergillus endophthalmitis after cataract surgery seen at Bascom Palmer Eye Institute. The clinical course, surgical history, and date of enucleation were recorded and the histological outcomes were reviewed. Results: Two patients presented with pain, redness, and decreased vision within one week after uncomplicated cataract surgery. In both patients, vitreous aspiration and intravitreal injections were the initial treatment. Because of worsening clinical condition pars plana vitrectomy was performed and Aspergillus was isolated from the vitreous samples. Despite repeated surgical and medical interventions, the clinical course in both patients was prolonged, unsuccessful and resulted in enucleation for a blind painful eye. In spite of prolonged local and systemic therapy the histological examination of the enucleated specimens using gomori methamine silver stain showed persistent diffuse infiltration of the anterior chamber, iris and ciliary body by a filamentous mold. In addition the histological specimens showed



retinal detachments, cyclitic membranes and fibrous plaques. Conclusions: Persistent fungal organisms were identified histologically in the intraocular structures of the enucleated globes using gomori methamine silver stain. These histological findings suggest that despite repeated treatment with intravitreal injections of amphotericin B and voriconazole and vitrectomy, viable fungal organisms can be identified in the intraocular structures of patients undergoing enucleation for Aspergillus endophthalmitis. Commercial Relationships: Luis J. Haddock, None; Sander R. Dubovy, None; Harry W. Flynn, Jr. , None Support: None

Program Number: 1680 Poster Board Number: D1027 Presentation Time: 8:30 AM - 10:15 AM The Endophthalmitis Kit (E-KIT) Taraprasad Das1, Ravindran D. Ravilla2, Sriam D. Ravilla3, Kannan Venkatesh3. 1Retina Vitreous Services, LV Prasad Eye Institute, Bhubaneswar, India; 2Aravind Eye Hospital, Madurai, India; 3Aurolab, Madurai, India. Purpose: To provide essential first- intervention material for treatment of bacterial and fungal endophthalmitis in a ready- to- use kit. Methods: We have developed E- kit that contains most essential devices (wire speculum, povidone iodine, syringes, and butter fly needle), antibiotics (ceftazidime 250 mg vial and vancomycin 100 mg vial) and antifungal (voriconazole 1 mg vial) for initial treatment of bacterial and fungal endophthalmitis. We used visual analogue scale (VAS) to score the health personnel comfort of each contents of the kit in 2 sites and 20 consecutive patients. Results: The antibiotics preparation was reduced to two-step instead of current four- step procedure. The overall comfort improved by 1.34 fold while using the E-kit in endophthalmitis patients. The comfort level improvement in antibiotic preparation was 2.70 fold for anti bacterial antibiotics and 3.02 fold for anti fungal antibiotic. Conclusions: The E-kit benefit included accurate and easy-to- prepare antibiotics and antifungal, safe vitreous biopsy and intravitreal injection, reduced ordering time and inventory cost. Commercial Relationships: Taraprasad Das, None; Ravindran D. Ravilla, None; Sriam D. Ravilla, Aravind Eye Hospital (E); Kannan Venkatesh, Aurolab (E) Support: Hyderabad Eye Research Foundation, Hyderabad, India

Program Number: 1681 Poster Board Number: D1028 Presentation Time: 8:30 AM - 10:15 AM Vitreous Isolates and Antibiotic Sensitivities: A 10-year Review of Culture-proven Endophthalmitis Cases Andrew M. Schimel1, Darlene Miller2, Harry W. Flynn, Jr.3. 1Ophthalmology, Bascom Palmer Eye Institute, Miami Beach, FL; 2Bascom Palmer Eye Institute, Univ of Miami Miller Sch of Med, Miami, FL; 3Bascom Palmer Eye Inst, University of Miami, Miami, FL. Purpose: To investigate the spectrum of organisms causing culture-proven endophthalmitis and their sensitivities to commonly used antimicrobial agents. Methods: Medical records were reviewed of all patients with culture-proven endophthalmitis at a single institution from 2002 through 2011. The outcome measures included intravitreal isolates identified as well as antibiotic sensitivities. Results: In all, 443 organisms were isolated during the study interval. The most common organisms identified were Staphylococcus epidermidis in 30.0% (133/443), Streptococcus viridians group in 11.1% (49/443), Staphylococcus aureus in 7.9% (35/443), Candida albicans in 5.6% (25/443), other coagulase-negative staphylococci in 5.8% (26/443), Propionibacterium acnes in 4.7% (21/443), and Pseudomonas aeruginosa in 3.2% (14/443). Overall, 323 of 443 (72.9%) isolates were gram-positive organisms, 48 (10.8%) were gram-negative organisms, 70 (15.8%) were fungi, and 2 (0.5%) were viruses. For gram-positive organisms, sensitivities were the following: vancomycin 100%, gentamicin 87.8%, sulfamethoxazole/trimethoprim 77.2%, levofloxacin 58.8%, oxacillin 54.5%, ciprofloxacin 51.2%, gatifloxacin 50.9%, and moxifloxacin 47.3%. For gram-negative organisms, sensitivities were the following: ceftazidime 100%, levofloxacin 100%, ciprofloxacin 95.0%, tobramycin 90.6%, gentamicin 80.6%, and sulfametoxazole/trimethoprim 59.4%. Conclusions: In the current study, no single antibiotic provided coverage for all of the microbes isolated from eyes with endophthalmitis. Combination therapy is generally the recommendation as the initial empiric treatment of suspected bacterial endophthalmitis. Appropriate history and characteristic clinical features may guide the use of initial antifungal agents. Commercial Relationships: Andrew M. Schimel, None; Darlene Miller , None; Harry W. Flynn, Jr. , None Support: NIH Grant P30-EY014801, Support from Research to Prevent Blindness

Program Number: 1682 Poster Board Number: D1029 Presentation Time: 8:30 AM - 10:15 AM

Ocular involvement among inpatients with fungemia: Analysis from a New York City tertiary care center Kristine K. Yin1, Benjamin M. Levine2. 1Ophthalmology, Weill Cornell Medical Center, New York, NY; 2Ophthalmology, Weill Cornell Medical College, New York, NY. Purpose: Fungal dissemination can lead to ocular infections in the form of chorioreitinitis and endophthalmitis, and is a common reason for ophthalmic consultation in hospitalized patients. Early studies have estimated the prevalence of ocular involvement at up to 45%.1 We sought to evaluate the current prevalence, microbial profile, and risk factors of fungal ocular disease among patients with positive fungal cultures at a tertiary care medical center in New York City. Methods: All consultation requests to evaluate for ocular involvement by fungal organisms at New York Presbyterian Hospital-Weill Cornell Medical Center from July 8, 2008 to November 7, 2011 were retrospectively reviewed. Ophthalmic examination, including indirect ophthalmoscopy, was performed by residents and confirmed by attending ophthalmologists. Results: 132 patients (14 pediatric, 118 adult) had consult requests indicating concern for ocular fungal disease in the setting of positive fungal cultures. 57% of patients were male, and 43% were female. Age ranged from 27 days to 95 years, with mean age of 53 years. All patients had at least one positive fungal culture; 120 (90.9%) had at least one positive blood culture. The most commonly identified organisms were Candida albicans (40%), Candida glabrata (26%), Candida parapsilosis (14%), Candida tropicalis (6%), Candida krusei (2%), Candida lusitaniae (1%), and Candida dubliniensis (1%). In addition, Aspergillus fumigatus, Histoplasma capsulatum, Trichosporon mucoides, and Rhodutula species were found. All patients were receiving systemic antifungal therapy at time of examination. Risk factors for fungal infection included: malignancy, immunosuppression, indwelling catheters, treatment in an intensive care unit, recent surgery, and parental nutrition. All patients had at least one risk factor, with the most common being malignancy. 5 adult and no pediatric patients (3.8%) had retinal findings consistent with fungal chorioretintis (4 cases) or endophthalmitis (1 case). Among these, one patient verbalized visual symptoms of new floaters. All patients improved with systemic antifungal therapy, and did not require intravitreal treatment. 39 patients had nonspecific findings inconsistent with fungal chorioretinitis and endophthalmitis. Visual symptoms were rare (4.0%) among patients with negative findings. Conclusions: Ocular involvement seems to occur infrequently in hospitalized patients with systemic fungal disease in the current era of widespread prophylactic antifungal therapy. Prospective and controlled studies may help to determine a specific subset of patients with fungemia that require routine ophthalmic screening. [1] Chen JY. Acta Paediatr Jpn 1994;36:261-5. Commercial Relationships: Kristine K. Yin , None; Benjamin M. Levine, None Support: None

Program Number: 1683 Poster Board Number: D1030 Presentation Time: 8:30 AM - 10:15 AM Vancomycin Resistant Gram Positive Bacterial Endophthalmitis in Southern India Avinash Pathengay. Retina, Bascom Palmer eye institute, Miami, FL. Purpose: To evaluate the microbiological profile and treatment outcomes of vancomycin resistant gram- positive bacterial endophthalmitis Methods: Medical records of consecutive patients with gram- positive bacterial endophthalmitis resistant to vancomycin (disc diffusion method) presenting between January 1, 2005 and December 31, 2010 were reviewed. Favourable outcome was defined as best corrected visual acuity greater or equal to 20/200. Results: Out of 682 culture positive endophthalmitis isolates , 448/682(65.6%) were associated with gram- positive bacteria. In vitro resistance to vancomycin was noted in 7/448 (1.56%). Three cases were post-traumatic ,three were postoperative and one was endogenous in origin. Resistance was commonly noted in Bacillus spp (4/7, 57.14%), followed by Staphylococcus spp (2/7, 28.6%) and Enterococcus sp (1/7,14.3%). . These isolates in vitro were most sensitive to ciprofloxacin in 6/7 (85.7%) of patients. Presenting visual acuity was only light perception in all 7 cases. Favourable outcome was achieved in only 1/7 (14.3%) cases. Conclusions: Vancomycin resistant endophthalmitis is rare and could be associated with poor visual outcome. Bacillus spp was the commonest gram positive bacteria resistant to vancomycin. Ciprofloxacin may be considered as an useful alternative in vancomycin Commercial Relationships: Avinash Pathengay, None Support: None

Program Number: 1684 Poster Board Number: D1031 Presentation Time: 8:30 AM - 10:15 AM Anaerobic Bacteria Causing Endophthalmitis: Spectrum, Frequency And Sensitivities Andres Emanuelli, Harry W. Flynn, Jr., Darlene Miller. Ophthalmology, Bascom Palmer Eye Inst-Univ of Miami, Miami, FL.



Purpose: To report the spectrum, frequency and sensitivities of anaerobic bacteria causing endophthalmitis over the last 20 years. Methods: From the Microbiology Laboratory Department records, all anaerobic bacterial vitreous isolates were reviewed between January 1, 1991 and September 30, 2011. All the vitreous isolates were tested for vancomycin, ceftazidime, gatifloxacin and moxifloxacin sensitivities at high and low dose. Results: Anaerobic bacteria accounted for 115 of 1386 (8.3%) in vitreous culture-positive isolates from 1991 to 2011. The most common bacteria for the isolates was Propionibacterium acnes (89.4%). Other etiologies included, Bacteroides species (5.3%), Peptostreptococcus species (2.6%), Propionibacterium granulosum (0.9%), Propionibacterium avidum (0.9%), and Clostridum subterminale (0.9%). The sensitivities were as follows: 92% of the isolates were sensitive to vancomycin (222 µg/ml & 22 µg/ml), 66.7% were sensitive to high dose ceftazidime (5 µg/ml) and 33.3% were sensitives to low dose ceftazidime (0.5 µg/ml). The sensitivities to fluoroquinolones, was as follows: 41.7% of the isolates were sensitive to high dose moxifloxacin (1.1 mg/ml), 25% were sensitive to low dose moxifloxacin (0.11 mg/ml), 50% of the isolates were sensitive to high dose gatifloxacin (0.67 mg/ml), and 8.3% were sensitive to low dose gatifloxacin (0.07 mg/ml). Conclusions: In the current study, anaerobic bacteria account for less than 10% of culture positive vitreous isolates. The most common etiology was Propionibacterium acnes. Regarding the sensitivities, vancomycin was the most effective antibiotic against anaerobes pathogens. The fluoroquinolones had a variable effect in the high dose group, but were not generally effective in the low dose group. Commercial Relationships: Andres Emanuelli, None; Harry W. Flynn, Jr. , Alcon (C), Alimera (C), Allergan (C), Lilly (C), Pfizer (C), Santen (C); Darlene Miller , None Support: P30EY014801, RPB Unrestricted Award

Program Number: 1685 Poster Board Number: D1032 Presentation Time: 8:30 AM - 10:15 AM Endophthalmitis Caused by Streptococcus Species: Frequency and Sensitivities Ajay E. Kuriyan, Harry W. Flynn, Jr., Eduardo C. Alfonso, Darlene Miller. Bascom Palmer Eye Institute, University of Miami/Miller School of Medicine, Miami, FL. Purpose: To examine the different species of Streptococcus causing endophthalmitis and to report their sensitivities to commonly used antibiotics. Methods: This study is a non-comparative case series. Microbiology laboratory data was analyzed to identify all streptococcal vitreous isolates and their sensitivities from January 1, 1990 to December 31, 2010, at a single institution. Results: In total, 194 streptococcal organisms were isolated from endophthalmitis cases during the study period. The species identified include Streptococcus viridans (56%), Enterococcus (21%), Streptococcus pneumoniae (14%), Group B Streptococcus (GBS, 5%), and miscellaneous Streptococcus species (4%). Overall, in vitro, the streptococcal isolates were 95% sensitive to levofloxacin, 92% sensitive to cephalothin, 79% sensitive to ceftazidime, 52% sensitive to ciprofloxacin, and 42% sensitive to gentamicin sulfate. All isolates were sensitive to vancomycin, except one Enterococcus strain, which showed intermediate resistance. Less resistance was found to gentamicin among Enterococcus isolates (29%), than S. pneumoniae (92%), GBS (80%), and S. viridans (62%) isolates. Conversely, much greater resistance was found to ceftazidime and cephalothin among Enterococcus isolates (100%), than isolates from other streptococcus species (0-2%). Conclusions: Streptococcal endophthalmitis is most commonly due to S.viridans and Enterococcus. In spite of reports of increasing resistance of Streptococcus species to vancomycin, all isolates were sensitive to vancomycin in the current study and only one isolate had intermediate resistance. Cephalothin and ceftazidime were variably effective, with 98-100% sensitivity observed in non-Enterococcus Streptococcus isolates and complete resistance in Enterococcus isolates. While the second generation fluoroquinolone, ciprofloxacin, had high rates of resistance (48%) among all Streptococcus isolates, the third generation fluoroquinolone, levofloxacin, had very low rates of resistance (5%). Commercial Relationships: Ajay E. Kuriyan , None; Harry W. Flynn, Jr. , None; Eduardo C. Alfonso, None; Darlene Miller , None Support: None

Program Number: 1686 Poster Board Number: D1033 Presentation Time: 8:30 AM - 10:15 AM Endophthalmitis associated with Baerveldt Glaucoma Drainage Devices Avnish A. Deobhakta1, Ryan Young1, Michael Banitt1, Thomas A. Albini1, William E. Smiddy2, Harry W. Flynn, Jr.3. 1Ophthalmology, Bascom Palmer Eye Institute, Miami, FL; 2Ophthalmology, Retina Group of Florida, Miami, FL; 3Bascom Palmer Eye Inst, University of Miami, Miami, FL. Purpose: To identify the risk factors, microbes, treatment modalities, and outcomes of patients with endophthalmitis associated with Baerveldt Glaucoma

Drainage Devices (GDDs). Methods: A non-comparative consecutive case series of patients at the Bascom Palmer Eye Institute diagnosed with endophthalmitis associated with Baerveldt GDDs between January 1, 1999 and July 31, 2011 was done. Results: 9 patients were identified. The average time elapsed from GDD surgery to endophthalmitis was 20 months with a range of 1 week to 4 years. 2 patients grew Staphylococcus epidermidis. 2 patients grew Staphylococcus aureus. 2 patients grew Serratia marcescens. 1 patient grew Streptococcus sanguis. 1 patient grew both Proteus mirabilis and Staphylococcus warneri. 1 patient grew both Staphylococcus epidermidis and Mycobacterium fortuitum. All patients were injected with intravitreal vancomycin and ceftazidime at the time of infection. 5 of 9 patients had tube exposure. 2 patients underwent evisceration. 3 patients underwent tube removal. Median pre-infection visual acuity was 20/100 with a range of 20/50 to HM. Median post-infection visual acuity was 20/400 with a range of 20/60 to NLP. Average loss of vision was 3.4 lines with one patient progressing to LP and two patients to NLP. Conclusions: Endophthalmitis associated with Baerveldt GDDs is uncommon. In the current study, tube exposure was a common risk factor for endophthalmitis. Infections were caused by a broad spectrum of microbes. Visual outcomes were generally poor. Commercial Relationships: Avnish A. Deobhakta, None; Ryan Young, None; Michael Banitt, None; Thomas A. Albini, None; William E. Smiddy, None; Harry W. Flynn, Jr. , None Support: None

Program Number: 1687 Poster Board Number: D1034 Presentation Time: 8:30 AM - 10:15 AM Acute endophthalmitis: a 222 cases multicentric retrospective study boris rysanek1A, jean jacques parienti1B, frederic mouriaux1A. Aophthalmology, Bclinical research and biostatistic, 1Caen university hospital, Caen, France. Purpose: Acute endophthalmitis is a dramatic event after cataract surgery. The aim of this multicentric retrospective study is to analyse initial signs and management and then to compare our results to the guidelines of Endophthalmitis Vitrectomy Study. Methods: All cases of endophthalmitis after cataract surgery from 3 french ophthalmological university centers between 1997 to 2007 were reviewed. Only patients with an acute endophthalmitis were recorded. Initial visual acuity and clinical signs, microbiological results, medical or surgical management, ocular complications and final visual acuity were recorded. All visual acuity score were modified in logmar scale. Results: 222 patients were included in this study. The mean age was 73.3 year. Vitreous biopsy or vitrectomy was realised in 117 patients. The choice of surgical treatment was statistically associated with lower initial visual acuity. Cyclitic membrane in the pupillary area could explain some low visual acuity. There was no significant difference for final visual acuity between medical treatment and surgical treatment. Use of ocular ultrasound was informative to analyse the vitreous body. Intraocular samples were taken in 95% of the patients. Pathogens were identified in cultures in 55.5% of cases. 40% of "aggressive pathogens” like staphylococcus aureus, streptococcus species and bacilles gram-negative were found. Final visual acuity 12 months after treatment was 20/40 or better in 55.5%. Loss of red reflex at the initial examination was statistically associated with a lower final visual acuity (p<0,001). Positive bacteriological culture was statistically associated with a lower final visual acuity (p<0,001). “Aggressive pathogens” were statistically associated with ocular complications (p<0,001). Conclusions: The EVS guidelines are not always valid in our experience, especially with the cyclitic membranes. The initial visual acuity and slip lamp examination is essential for optimized management. The most predictive criteria for final acuity in our study were the red reflex and presence and type of bacteria in acute postoperative endophthalmitis. Commercial Relationships: boris rysanek, None; jean jacques parienti, None; frederic mouriaux, None Support: None

Program Number: 1688 Poster Board Number: D1035 Presentation Time: 8:30 AM - 10:15 AM Specific Pcr Detection Of Staphylococcus And Streptococcus In Endophthalmitis Pierre-Loic Cornut1, Sandrine BOISSET2, Jérôme ETIENNE2, Carole BURILLON1, Yvonne BENITO3, Max MAURIN4, Christophe Chiquet5, François VANDENESCH2. 1Ophthalmology, Herriot Hospital, Lyon, France; 2Microbiology, LYON I UNIVERSITY, Lyon, France; 3Microbiology, Hospices Civils de Lyon, Lyon, France; 4Microbiology, GRENOBLE UNIVERSITY, Grenoble, France; 5Department of Ophthalmology, Grenoble University Hospital, Grenoble, France. Purpose: Gram-positive cocci, mainly Staphylococcus and Streptococcus, are causal agents in the majority of endophthalmitis cases. The microbiological diagnostic yield of these infections, which has been improved by the use of



molecular biology tools (mainly panbacterial PCR) in addition to conventional culture methods, remains imperfect. This study reports the contribution of specific conventional PCR of the Staphylococcus and Streptococcus genera followed by sequencing for the microbiological diagnosis of endophthalmitis. Methods: Specific PCR assays targeting the tuf gene of the Staphylococcus and Streptococcus genera were performed in addition to the reference techniques (conventional culture and panbacterial PCR) on samples of aqueous humor and/or vitreous in patients with endophthalmitis when Staphylococcus or Streptococcus had been identified using reference methods or no identification had been possible. Results: Out of the 91 samples analyzed, the culture was positive in 38 cases (30 Staphylococcus and 8 Streptococcus) and panbacterial PCR was positive in 59 cases (42 Staphylococcus and 17 Streptococcus). Staphylococcus-specific PCR was positive in 49 cases. Staphylococcus PCR was positive while panbacterial PCR was negative in 7 cases (5 of which were also negative in culture). Conversely, in accordance with culture, panbacterial PCR found a Staphylococcus infection in 1 case, whereas the Staphylococcus PCR was negative. The Streptococcus-specific PCR was positive in 13 cases. The Streptococcus-specific PCR provided no additional diagnosis to the pan-bacterial PCR. Conversely, the pan-bacterial PCR found a Streptococcus infection in 4 cases, whereas the Streptococcus-specific PCR was negative (as was the culture). The specific PCR provided better identification of the species than pan-bacterial in sequencing. In 14 cases, all the analyses remained negative. Conclusions: The Staphylococcus-specific PCR seemed more sensitive than panbacterial PCR in identifying Staphylococcus causing endophthalmitis. These superior results of the specific technique were not repeated for the identification of the Streptococcus in our experiment. The sensitivity tests performed on different bacterial dilutions confirmed these data. The Staphylococcus- and Streptococcus-specific techniques used in this study were complementary to the panbacterial PCR techniques and improved sensitivity in the identification of Staphylococcus and better identification of the species with sequencing. Commercial Relationships: Pierre-Loic Cornut , None; Sandrine Boisset, None; Jérôme Etienne, None; Carole Burillon , None; Yvonne Benito, None; Max Maurin , None; Christophe Chiquet, None; François Vandenesch, None Support: None

Program Number: 1689 Poster Board Number: D1036 Presentation Time: 8:30 AM - 10:15 AM Staphylococcus aureus Endophthalmitis: Antibiotic Susceptibilities, Methicillin Resistance, and Clinical Outcomes Ashkan M. Abbey, Darlene Miller, Harry W. Flynn, Jr.. Ophthalmology, Bascom Palmer Eye Institute, Miami, FL. Purpose: To investigate the antibiotic susceptibility and clinical outcomes of endophthalmitis caused by methicillin-sensitive (MSSA) versus methicillin-resistant (MRSA) Staphylococcus aureus. Methods: A consecutive case series of 7 patients with culture-proven S. aureus endophthalmitis seen at the Bascom Palmer Eye Institute from January 1, 2008, through October 1, 2011, was performed. Antibiotic susceptibility profiles, identified using standard microbiologic protocols, and visual acuity at 1 and 3 months were the main outcome measures. Historical published outcomes (1995-2007) were also compared. Results: MSSA was recovered from 4/7 patients (57%) and MRSA was recovered from 3/7 patients (43%). The etiologies included the following: cataract surgery in 3/7 (43%), intravitreal injection in 2/7 (29%), endogenous in 1/7 (14%), and trauma/open globe injury in 1/7 (14%). All isolates were sensitive to vancomycin. Fluoroquinolone (FQ) resistance was 2/4 (50%) in the MSSA group, and 3/3 (100%) in the MRSA group. The median presenting visual acuity was hand movements for both MSSA and MRSA eyes. All eyes received intravitreal antibiotics. Pars plana vitrectomy was performed on 25% of MSSA and 33% of MRSA patients. At 3 months, a visual acuity of 20/400 or better was achieved in 3/7 (43%). In a previously published case series from the same institution evaluating 32 cases of S. aureus endophthalmitis from 1995 to 2007, 41% of cases involved MRSA. In the previous study, all MSSA and MRSA isolates were sensitive to vancomycin. However, MRSA isolates from the current study exhibited 100% FQ resistance (38% resistant in previous study), and 50% of the MSSA isolates demonstrated FQ resistance (5% resistant in previous study). A visual acuity of 20/400 or better at 3 months was achieved in 43% of the current series versus 51% of the previous series. Conclusions: The frequency of MRSA has remained stable in the current study compared to a previous study. All isolates were sensitive to Vancomycin, but a higher percentage of isolates was resistant to fluoroquinolones in the current study. A similar percentage of these patients achieved 20/400 or better visual acuity in both studies. Commercial Relationships: Ashkan M. Abbey, None; Darlene Miller , None; Harry W. Flynn, Jr. , None Support: None

Program Number: 1690 Poster Board Number: D1037 Presentation Time: 8:30 AM - 10:15 AM Staphylococcus aureus Isolated from Endophthalmitis May Be Hospital-Acquired Based on PVL and Antibiotic Susceptibility Testing Regis P. Kowalski, Tyler A. Kowalski, Kristin A. Rarey, Robert M. Shanks, Eric G. Romanowski, Francis S. Mah. Ophthalmology/Microbiology, University of Pittsburgh, Pittsburgh, PA, Pittsburgh, PA. Purpose: Staphylococcus aureus (SA) endophthalmitis is generally a post-surgical infection with an undefined source of entry. Hospital-acquired (HA) infections are associated with multi-antibiotic resistance and not possessing the Panton Valentine Leukocidin (PVL) toxin. Community-acquired (CA) infections are not associated with multi-antibiotic resistance and possess PVL toxin. We hypothesize that CA infection is more common than HA for Staphylococcus aureus endophthalmitis. Methods: Twenty de-identified SA isolates, collected from the vitreous and/or aqueous of clinical endophthalmitis, were tested for the presence of PVL toxin and antibiotic susceptibility. PVL testing was performed using a kit to detect the staphylococcal toxin by reversed passive latex agglutination (PVL-RPLA “Seiken”, Denka Seiken Co., LTD, Japan) and PCR (targets lukF-PV and lukS-PV genes). SA isolates were tested for antibiotic susceptibility using disk diffusion at the time of isolation. Multi-antibiotic resistance was defined as resistance to at least 3 classes of antibiotics. Results: Of the 20 isolates (PVL-RPLA testing), 15 were multi-antibiotic resistant and PVL-negative consistent with HA, and 1 was not multi-antibiotic resistant and PVL-positive, consistent with CA. Only two isolates tested positive for PVL with one demonstrating methicillin (MR) and fluoroquinolone (FQ) resistance. Of the 18 PVL-negative SA isolates, 15 (83%) were multi-antibiotic resistant (12 MRSA, 14 FQ resistant). All SA isolates tested PVL negative by PCR. Conclusions: Our results reject the hypothesis that Staphylococcus aureus isolated from endophthalmitis is consistent with CA sources due to the lack of the PVL toxin and multiple resistant patterns of the SA. PVL does not appear to be a key virulence factor for the development of SA endophthalmitis. Commercial Relationships: Regis P. Kowalski, None; Tyler A. Kowalski , None; Kristin A. Rarey , None; Robert M. Shanks, None; Eric G. Romanowski, None; Francis S. Mah, None Support: Pennsylvania Lions Club

Program Number: 1691 Poster Board Number: D1038 Presentation Time: 8:30 AM - 10:15 AM Antibacterial Properties of 2% Lidocaine on Staphylococcus Aureus, Staphylococcus Epidermidis, and Streptococcus Viridans and Reduced Rate of Post-injection Endophthalmitis Stephen J. Kim1, Aaron Tustin1, Amy Chomsky1, G. Baker Hubbard, III2, Jinsong Sheng1. 1Ophthalmology, Vanderbilt University, Nashville, TN; 2Ophthalmology, Emory University, Atlanta, GA. Purpose: To determine whether application of subconjunctival 2% lidocaine for anesthesia reduces rates of endophthalmitis after intravitreal injection. Methods: We performed in vitro experiments to confirm the antibacterial properties of commercially available 2% lidocaine with 0.1% methylparaben (Hospira, Inc., Lake Forest, IL) against causative organisms of endophthalmitis. Isolates of Staphylococcus aureus, Staphylococcus epidermidis, and Streptococcus viridans from patients with endophthalmitis were incubated in trypticase soy broth with either 2% lidocaine/0.1%methylparaben or equal volume distilled water (control). Aliquots (100 µl) were plated on Mueller-Hinton (S. aureus and S. epidermidis) or blood agar plates (S. viridans) at 0, 10, 30, 120, and 240 minutes and colonies were counted after 24 hours. We performed a retrospective chart review from January 2005 to February 2011 at the Vanderbilt University, Nashville, TN, the Nashville Veteran’s Administration Hospital, and Emory University, Atlanta, GA to determine the rate of intravitreal injection-related endophthalmitis with or without the pre-application of subconjunctival 2% lidocaine/0.1% methylparaben (lidocaine) for anesthesia. Total number of intravitreal injections was determined by procedure code and endophthalmitis cases were identified using ICD-9 codes. Endophthalmitis cases were only included if they occurred within 7 days after an injection and either had a positive culture or hypopyon. All suspected cases related to triamcinolone without a positive culture were excluded. Intravitreal injections were then categorized by providers who strictly adhered to always using or never using subconjunctival lidocaine. Results: In vitro, 2% lidocaine/0.1% methylparaben demonstrated rapid bactericidal effects against all 3 organisms. After 10 minutes of exposure, there was approximately a 90% (P < 0.01), 95% (P < 0.001), and 92% (P < 0.001) reduction in colony forming units (CFU) when compared to time 0 for S. aureus, S. epidermidis, and S. viridans respectively. Complete elimination of CFU occurred at subsequent time points for each organism in contrast to logarithmic increase for control plates. There were a total of 8 cases (7 culture positive) of post-injection endophthalmitis out of 7969 (0.1%) intravitreal injections performed without the use of subconjunctival lidocaine. In contrast, there were 0 cases out of 7224 injections performed with subconjunctival lidocaine (P = 0.02).



Conclusions: Application of subconjunctival 2% lidocaine for anesthesia may reduce the incidence of post-injection endophthalmitis. Commercial Relationships: Stephen J. Kim, None; Aaron Tustin, None; Amy Chomsky, None; G. Baker Hubbard, III , None; Jinsong Sheng, None Support: Supported in part by an unrestricted grant from Research to Prevent Blindness to the Vanderbilt University School of Medicine Department of Ophthalmology and Visual Sciences.

Program Number: 1692 Poster Board Number: D1039 Presentation Time: 8:30 AM - 10:15 AM Endophthalmitis Following Intravitreal Injection Ve rsus Endophthalmitis Following Cataract Surgery Andrew A. Chang1, Matthew Simunovic2, Ryan Rush2. 1Sydney Eye Hospital. Sydney Retina Clinic, Sydney, Australia; 2Ophthalmology, Sydney Eye Hospital, Sydney, Australia. Purpose: To describe and compare the causative organisms, clinical features and visual outcomes of endophthalmitis following intravitreal injection (IVI) to endophthalmitis following cataract surgery (CS). Methods: Retrospective case series.Patient Population & Setting: Patients with acute endophthalmitis following either CS or IVI presenting to a tertiary referral centre - Sydney Eye Hospital - between 2007 and 2010. Main outcome measures: 1. identification of the causative organism; 2. time to presentation; 3. odds of improvement in visual acuity following treatment; 4. odds of final visual acuity of counting fingers or less and 5. odds of enucleation. Results: Of the 101 patients in this study, 48 had preceding CS and 53 had preceding IVI. There was an increased incidence of Streptococcus spp. endophthalmitis in post IVI cases (24.53% versus 6.25%; odds ratio (OR) 5.85; P = 0.022). Endophthalmitis following IVI had increased likelihood of a final visual acuity of counting fingers or less (OR = 6.0; P < 0.01), decreased likelihood of any improvement in acuity following treatment (OR = 0.13; P <0.01) and an increased likelihood of presenting within a week of the procedure (OR = 3.93; P <0.01). Streptococcus spp. endophthalmitis was associated with increased likelihood of a final visual acuity of counting fingers or less (OR = 10.2; P < 0.01), decreased likelihood of any improvement in acuity following treatment (odds ratio = 0.06; P < 0.01) and increased likelihood of enucleation (OR = 17.11; P < 0.01). Conclusions: Post IVI endophthalmitis is associated with an increased incidence of Streptococcus spp. infection, earlier presentation and poorer visual outcomes when compared to post CS endophthalmitis. Commercial Relationships: Andrew A. Chang, None; Matthew Simunovic, None; Ryan Rush, None Support: None

Program Number: 1693 Poster Board Number: D1040 Presentation Time: 8:30 AM - 10:15 AM Endophthalmitis After Intravitreal Vascular Endothe lial Growth Factor Antagonists: A Seven-year Experience At A University Referral Center.endophthalmitis After Intravitreal Vascular Endothelial Growth Factor Antagonists: A Seven-year Experience At A University Referral Center Andrew A. Moshfeghi1, Philip J. Rosenfeld1, Harry W. Flynn, Jr.1, Stephen G. Schwartz2, Janet L. Davis3A, Timothy G. Murray3A, William E. Smiddy4, Audina M. Berrocal3B, Thomas A. Albini3A. 1Vitreoretinal Surgery & Diseases, Bascom Palmer Eye Institute, Palm Beach Gardens, FL; 2Bascom Palmer Eye Institute, Univ of Miami Miller Sch of Medicine, Naples, FL; AOphthalmology, BRetina Department, 3Bascom Palmer Eye Institute, Miami, FL; 4Ophthalmology, Retina Group of Florida, Miami, FL. Purpose: To assess the rate of infectious endophthalmitis and to describe the clinical and microbiological features of eyes that develop clinically suspected endophthalmitis after an intravitreal injection of vascular endothelial growth factor (VEGF) antagonists. Methods: The medical records of patients undergoing intravitreal injections of anti-vascular endothelial growth factor agents from January 1, 2005, through November 30, 2011, at a single university referral center and associated satellite clinics were retrospectively analyzed to determine the rate of infectious endophthalmitis after intravitreal anti-vascular endothelial growth factor injections. Results: Twelve cases (11 patients) of clinically suspected endophthalmitis were identified after a total of 75,018 injections (0.016%, 95% confidence interval, 0.0092% to 0.0280%). Of the 12 cases, 11 presented within 3 days of the injection. Of the 7 culture-positive cases (0.009% rate of culture positive endophthalmitis; 95% confidence interval, 0.0045% to 0.0193%), 5 were because of Streptococcus species. In 4 of the 5 Streptococcus cases, final visual acuity was hand motions or worse. There was no statistically significant difference noted in the rates of infection encountered following different types of anti-VEGF injections (pegaptanib, ranibizumab, bevacizumab, or aflibercept). All cases of endophthalmitis occurred with patients who were receiving post-injection antibiotics. In the last three years, the majority of physicians have not used antibiotics after intravitreal injections and no cases of endophthalmitis have been

encountered in this cohort. Conclusions: A very low rate of endophthalmitis after intravitreal injections of anti-vascular endothelial growth factor agents was observed. Patients typically presented within 3 days of injection. Streptococcus species was the most common bacteria isolated, and it was generally associated with poor visual outcomes. The use of topical antibiotics does not appear to mitigate the risk of endophthalmitis after intravitreal injections. Commercial Relationships: Andrew A. Moshfeghi, Alcon, Inc. (C), Allergan, Inc. (C), Eyetech, Inc. (C), Genentech/Roche, Inc. (C); Philip J. Rosenfeld, None; Harry W. Flynn, Jr. , Alcon (C), Allergan (C); Stephen G. Schwartz, None; Janet L. Davis, None; Timothy G. Murray , Alcon (C); William E. Smiddy, None; Audina M. Berrocal , None; Thomas A. Albini, Allergan (C), Bausch & Lomb (C) Support: The Palm Beach Community Trust Fund, NIH Center Grant P30-Ey014801, and an unrestricted grant from the Research to Prevent Blindness foundation, NY, NY.

Program Number: 1694 Poster Board Number: D1041 Presentation Time: 8:30 AM - 10:15 AM A Comparison of Rates of Infectious Endophthalmitis after Intravitreal Injection Martin Wilkes, Christopher D. Riemann, Michael R. Petersen, Daniel M. Miller, Robert E. Foster. Retina, Cincinnati Eye Institute, Cincinnati, OH. Purpose: To determine whether there is a significant difference in the rate of infectious endophthalmitis following intravitreal injections with different therapies, at a high-volume, outpatient setting. Methods: All patient encounters (42) with a coded diagnosis of endophthalmitis with preceeding intravitreal injection from the time period of 2006-2010 at Cincinnati Eye Institute were compiled. The diagnosis (true infectious endophthalmitis versus pseudo-endophthalmitis), agent administered (bevacizumab, ranibizumab, triamcinolone), indication for treatment, culture results, and final visual outcome were reviewed in detail. Results: There were a total of 11 cases of true, culture-confirmed infectious endophthalmitis. A total of 4 cases were associated with bevacizumab, 4 with ranibizumab, and 3 with triamcinolone. Conclusions: Compounded bevicizumab has a much lower incidence of infectious endophthalmitis than ranibizumab. This may be attributable to the process in which ranibizumab is drawn up prior to injection. When taken into account the much higher volume of bevacizumab administered at our institution (3:1, bevicizumab vs ranibizumab), there is a relatively lower rate of endophthalmitis associated with this medication. Commercial Relationships: Martin Wilkes , None; Christopher D. Riemann, None; Michael R. Petersen, None; Daniel M. Miller , None; Robert E. Foster, None Support: None

Program Number: 1695 Poster Board Number: D1042 Presentation Time: 8:30 AM - 10:15 AM Endophthalmitis Associated with Intravitreal Anti-V ascular Endothelial Growth Factor Injections in a Single Large Retinal Therapy Unit of Moorfields Eye Hospital, United Kingdom Ranjit Sandhu, Robin Hamilton. Ophthalmology, Moorfields Eye Hospital, London, United Kingdom. Purpose: Intravitreal anti-vascular endothelial growth factor agents have become the standard treatment for wet age-related macular degeneration. However, there is a risk of endophthalmitis as a serious adverse effect with devastating implications for vision in the affected eye. The aim was to evaluate the incidence of endophthalmitis following intravitreal ranibizumab or bevacizumab over a 12 month period and compare it with the published rates of endophthalmitis. Methods: The injections were performed as an outpatient procedure in a dedicated Retinal Therapy Unit (RTU) at Moorfields Eye Hospital, London, United Kingdom. The injections were performed in a clean room, under aseptic technique, using sterile disposable instruments, povidone-iodine as a pre-injection preparation, eye drape, gloves and post-injection antibiotics. A total of 9977 intravitreal injections were administered over a 12 month period between January and December 2010, with 9185 ranibizumab 0.5mg and 792 bevacizumab 1.25mg injections. Patients with endophthalmitis had aqueous and vitreous sampling for microscopy, culture and sensitivity and were treated with intravitreal amikacin and vancomycin, oral moxifloxacin, oral prednisolone and intensive topical steroids. Results: A total of 3 endophthalmitis cases were reported between January and December 2010 following intravitreal ranibizumab injections. All 3 eyes had non-infectious endophthalmitis. Two of the patients were female and one male, age-range between 70-77 years, with poor visual acuity of hand movements at the time of diagnosis. Conclusions: The endophthalmitis rate after an intravitreal injection of ranibizumab 0.5 mg was approximately 1:3325 or 0.03% and is similar to the



published data. There were no cases of endophthalmitis following bevacizumab but it can be argued that this could be due to the small number of bevacizumab injections administered.

Commercial Relationships: Ranjit Sandhu, None; Robin Hamilton, None Support: None

Program Number: 1696 Poster Board Number: D1043 Presentation Time: 8:30 AM - 10:15 AM Endophthalmitis After Intravitreal Injections: In-o ffice Setting Vs Operating Room Setting Alessandra Sborgia1, Homayoun Tabandeh2, Lorenza Ciracì1, Cesare Mariotti3, Claudio Furino1, Francesco Boscia1. 1Department of Ophthalmology, University of Bari - Italy, University of Bari - ITALY, Italy; 2Retina Vitreous Assoc Med Group, Los Angeles, CA; 3Clinica Oculistica, Universita Politecnica delle Marche, Ancona, Italy. Purpose: To report on the occurrence of endophthalmitis following intravitreal injections (IVI) in two different settings: in-office and in operating room. Methods: Retrospective review of all patients who underwent intravitreal injections in two retina practices. Group (A) underwent intravitreal injection in the examination room in office setting and Group (B) underwent intravitreal injection in the operating room. Results: Group( A): A total of 5687 IVI performed including 1895 ranibizumab, 3494 bevacizumab, and 298 triamcinolone acetonide. The diagnosis included neovascular age-related macular degeneration (3877 eyes), diabetic macular edema (999), retinal vein occlusion (722), and miscellaneous diagnosis (89) eyes. Group( B): A total of 3063 IVI performed including683 ranibizumab, 2364 bevacizumab, and 16 triamcinolone acetonide. The diagnosis included neovascular age-related macular degeneration (1836 eyes), diabetic macular edema (771), retinal vein occlusion (189), and miscellaneous diagnosis (267) eyes. In group (A) two (0.035%) cases of endophthalmitis occurred in 5687 intravitreal injections. In group (B) two (0.065%) cases of endophthalmitis occurred in 3063 intravitreal injections. Conclusions: The rate of endophthalmitis after intravitreal injections is low whether the procedure is performed in the office or in the operating room setting. The findings have implications in terms of the patient convenience, efficiency, and cost of administrating these treatments Commercial Relationships: Alessandra Sborgia, None; Homayoun Tabandeh, None; Lorenza Ciracì, None; Cesare Mariotti, None; Claudio Furino, None; Francesco Boscia, Alcon, Allergan, Bayer, Novartis (C) Support: None

Program Number: 1697 Poster Board Number: D1044 Presentation Time: 8:30 AM - 10:15 AM

Endophthalmitis Following Intravitreal Injection Of Anti-vascular Endothelial Growth Factor (anti-vegf) Agents Pavel Nemec, Jan Ernest, Leos Rejmont, Zaneta Benesova, Libor Hejsek. Ophthalmology Dept., UVN Prague, Prague, Czech Republic. Purpose: To determine the incidence of endophthalmitis following intravitreal injection of anti-vascular endothelial growth factor (VEGF) agents and to comment results of laboratory specification and treatment. Methods: A total of 2,274 intravitreal anti-VEGF injections (1,213 pegaptanib, 1,061 ranibizumab) were performed from 2007 to 2010. The injections were performed as an out-patient procedure with use of povidone-iodine as a part of preinjection preparation. Preinjection topical antibiotics, sterile environment, sterile gloves and postinjection topical antibiotics were used. Results: There were eight cases of endophthalmitis, six cases following pegaptanib injection and two cases following ranibizumab injection. Only one patient regained her preinjection visual acuity. The incidence of endophthalmitis was 0,35%. There was no difference in the incidence of endophthalmitis between pegaptanib and ranibizumab (P = 0.6). Conclusions: Although there is no consensus regarding the intravitreal injection procedure technique, the incidence of endophthalmitis was low in a large series of injected patients in a comparison with results reported in clinical trials where much more extensive preinjection preparation was mandated. The most frequent pathogen was found Streptococcus epidermidis (75% cases). We found no difference in the endophthalmitis risk of patients receiving pegaptanib as compared with ranibizumab. Commercial Relationships: Pavel Nemec, None; Jan Ernest, None; Leos Rejmont, None; Zaneta Benesova, None; Libor Hejsek, None Support: None

Program Number: 1698 Poster Board Number: D1045 Presentation Time: 8:30 AM - 10:15 AM Trends In The Incidence & Outcomes Of Endophthalmitis After Anti-vascular Endothelial Growth Factor (anti-vegf) Injections In The Last Seven Years Anna Gabrielian1, Colin T. Flood2, Timothy P. Flood1. 1Ophthalmology, Illinois Retina Assoc, Rush Univ Med Ctr, Chicago, IL; 2Dept of Preventive Medicine, The Graduate School Program in Public Health, Northwestern Univ Feinberg School of Medicine, Chicago, IL. Purpose: To determine overall and yearly incidences of post-injection endophthalmitis in the last seven years. Methods: Retrospective chart review of patients undergoing intravitreal injections of anti-VEGF agents from 06/2005 - 11/2011 in a multi-surgeon, multi-office retina practice. Post-surgical cases and those secondary to trauma and endogenous sources were excluded. Main outcomes were overall and yearly rates of post-injection endophthalmitis. Secondary outcome was final visual acuity (VA). Results: 17 cases of clinically suspected endophthalmitis were identified in a total of 53,264 injections of anti-VEGF agents, with incidence of 0.03% [95% CI 0.02-0.05%]. Yearly incidence of endophthalmitis from 2005- 2007 was 0.04%, 0.05% and 0.4% in each year respectively. Yearly incidence from 2008 - 2010 was 0.03%, 0.04% and 0.03% respectively, and 0% in 2011. Average length of follow-up was 23.6 months (range 1-72.5 mo), with 2,201 intraocular injections of Macugen, 15,455 injections of Avastin, and 35,608 injections of Lucentis. No difference was noted in endophthalmitis rates between any of the anti-VEGF agents (p > 0.1 for all groups). Average time from injection to presentation was 52.4hrs (range 5 hrs - 11 days). Incidence of culture-positive endophthalmitis was 0.02% [95% CI 0.01-0.03%]. Mean time to presentation was 27hrs in culture-positive and 106hrs in culture-negative cases (p=0.03). Majority (67%) were streptococcal infections. LogMAR of baseline VA was 0.79 (range 0 - 2.3) on average. LogMAR of presenting VA was 1.96 (range 0.2 - 2.7), and logMAR of final VA was 1.39 (range 0.3 - 3) on average. Only 29.5% of patients returned to baseline VA. Conclusions: The overall incidence of post-injection endophthalmitis was 0.03%, and the incidence of culture-positive endophthalmitis was 0.02%. Choice of anti-VEGF agent did not affect these rates. Culture-positive cases presented sooner. Fortunately, yearly incidence of endophthalmitis has been dropping over the past seven years. This could be due to the increased awareness of oral pathogen virulence, and undertaking of special precautions to minimize oropharyngeal droplet transmission. Since only 29.5% of patients returned to the baseline VA, post-injection endophthalmitis remains a vision-threatening entity in the age of anti-VEGF injections. Commercial Relationships: Anna Gabrielian, None; Colin T. Flood, None; Timothy P. Flood, None Support: None

arvo 2012 annual meeting abstracts

Documents