ENDOPHTHALMITIS ASSOCIATED WITH INTRAVITREAL INJECTIONS Office-Based Setting and Operating Room Setting HOMAYOUN TABANDEH, MD, MS,* FRANCESCO BOSCIA, MD,† ALESSANDRA SBORGIA, MD,† LORENZA CIRACÌ, MD,† POUYA DAYANI, MD,* CESARE MARIOTTI, MD,‡ CLAUDIO FURINO, MD,† HARRY W. FLYNN, JR, MD§ Purpose: To report on the occurrence of endophthalmitis after intravitreal injections (IVI) in two different settings: office-based and operating room. Methods: Consecutive case series. Retrospective review of all patients who underwent IVI by 2 physicians between January 2009 and December 2011. Group A underwent IVI in the examination room in office-based setting and Group B underwent IVI in the operating room. Results: A total of 11,710 IVIs were performed during the study period. Group A: A total of 8,647 IVIs performed including 2,041 ranibizumab, 6,169 bevacizumab, and 437 triamcinolone acetonide. The diagnosis included neovascular age-related macular degeneration (5,376), diabetic macular edema (1,587), retinal vein occlusion (1,068), and miscellaneous diagnosis (616). Group B: A total of 3,063 IVIs performed including 683 ranibizumab, 2,364 bevacizumab, and 16 triamcinolone acetonide. The diagnosis included neovascular age-related macular degeneration (1,836), diabetic macular edema (771), retinal vein occlusion (189), and miscellaneous diagnosis (267). A total of 5 cases (0.043%) of clinically suspected endophthalmitis occurred in 11,710 injections. Three cases (0.035%) occurred in Group A, and 2 cases (0.065%) occurred in Group B. Conclusion: The rate of clinically suspected endophthalmitis after IVIs is low whether the procedure is performed in the office or operating room setting. The findings have implications in terms of the patient convenience, efficiency, and cost of administrating these treatments. RETINA 34:18–23, 2014
I
procedures. Endophthalmitis remains a sight threatening complication of intravitreal administration of drugs. A varying number of steps have been taken in an attempt to reduce the risk of IVI-associated endophthalmitis including the use of prophylactic topical antibiotics, sterile drapes, sterile gloves, face mask, eyelid speculum, and topical povidone-iodine. Performing the procedure in the operating room encompasses many of the above mentioned measures and has been mandated in many countries with the assumption that it decreases the rate of infection. These measures impact safety, patient convenience, efficiency, and cost of administrating these treatments. With the exception of povidone-iodine, there is little direct evidence that the above measures result in a decrease in rate of endophthalmitis. This study evaluated the rate of endophthalmitis associated with IVIs performed in two settings: officebased setting and operating room setting.
ntravitreal injection of anti-vascular endothelium growth factor (VEGF) agents has become the standard of care for the treatment of neovascular age-related macular degeneration and macular edema secondary to retinal vein occlusion.1–9 There is also widespread use for the treatment of retinal complications of diabetes. This has made intravitreal injection (IVI) of antiVEGF drugs one of the most commonly performed
From the *Retina-Vitreous Associates Medical Group, Los Angeles, California; †Department of Ophthalmology, University of Bari, Bari, Italy; ‡Department of Ophthalmology, Polytechnic University of Ancona, Ancona, Italy; and §Bascom Palmer Eye Institute, Miami, Florida. Presented at the Association for Research in Vision and Ophthalmology Annual Meeting, Fort Lauderdale, FL, May 2012. None of the authors have any financial/conflicting interests to disclose. Reprint requests: Homayoun Tabandeh, MD, MS, RetinaVitreous Associates Medical Group, 9001 Wilshire Boulevard, Suite 301, Beverly Hills, CA 90211; e-mail: [email protected]
18
ENDOPHTHALMITIS AND INTRAVITREAL INJECTION TABANDEH ET AL
Methods Consecutive patients undergoing IVIs of ranibizumab, bevacizumab, and triamcinolone between January 2009 and December 2011 were identified by the search of the diagnostic and billing databases, and review of medical records of practices of 2 retina physicians (H.T. and F.B.) at the Retina-Vitreous Associates Medical Group, Los Angeles, and Department of Ophthalmology, University of Bari. The study protocol was reviewed by the regional institutional review board who concluded that approval was not required for this study. The study was in accord with the principles outlined in the Declaration of Helsinki. Data was collected on type of injected medication, diagnosis, and occurrence of endophthalmitis. Medical records of patients with clinically suspected endophthalmitis were reviewed and further data was collected including the microbiology results, baseline diagnosis, and type and number of injections. All data were collected and entered onto a computerized database (Microsoft Excel 2007; Microsoft Corporation, Redmond, WA). Statistical analysis was performed using statistical software, SAS version 9.3 (Cary, NC). The difference in rate of endophthalmitis between various factors was compared using the Fisher exact test. P , 0.05 were considered as statistically significant. In the office-based setting, intravitreous injection was performed in the examination room. In a majority of the cases, anesthesia was achieved with 0.5% topical proparacaine hydrochloride (Akorn Pharmaceuticals, Lake Forest, IL) followed by application of a 4% lidocaine-soaked cotton tip applicator to the injection site. Subconjunctival injection of 2% lidocaine was used in a minority of patients. A sterile eyelid speculum was placed followed by administration of 5% Povidone–iodine, using cotton tip applicators, to prepare the ocular surface before injection. Nonsterile disposable gloves were used throughout the procedure. Surgical mask was not used and no restriction was placed on conversation during the procedure. Preoperative topical antibiotics, sterile gloves, and sterile drapes were not used. Ocular surface was irrigated with saline solution and topical antibiotic drops were used immediately after the injection. Patients were instructed to use topical antibiotic eye drops 4 times a day for 4 days postinjection. Bevacizumab was prepackaged by a compounding pharmacy (California Pharmacy and Compounding Center, Newport Beach, CA). Ranibizumab and triamcinolone were prepared by the treating physician immediately before injection as per package instruction.
19
In the operating room setting, anesthesia was achieved with 0.4% topical benoxinato cloridrato single use vials. Intravitreal injection protocol included the use of preoperative topical antibiotics, sterile gloves, surgical mask, sterile drape, and sterile eyelid speculum. The treating physician used a sterile scrub brush soaked with povidone-iodine before wearing sterile glove. Povidone-iodine (5%) was used to prep the ocular surface before injection. Restriction was placed on all conversation during procedure. The patients were instructed to use topical antibiotic eye drops 4 times a day for 4 days post injection. Bevacizumab, ranibizumab, and triamcinolone acetonide (Kenacort; Bristol-Myers Squibb, Anagni, Italy) were prepared by the treating physician immediately before the injection. Bevacizumab was prepared in the operating room from the stock bottle using “Securemix” (Eurospital SPA, Trieste, Italy), a self-contained system for drug reconstitution, using a 0.22-mm filter (Table 1).
Results From January 2009 to December 2011, a total of 11,710 IVIs including 2,724 ranibizumab, 8,533 bevacizumab were performed by 2 physicians (H.T. and F.B.). Eight thousand six hundred and forty-seven IVIs were performed in the office-based setting and included 2,041 ranibizumab, 6,169 bevacizumab, and 437 triamcinolone acetonide injections. The diagnosis included neovascular age-related macular degeneration (5,376), diabetic macular edema (1,587), retinal vein occlusion (1,068), and miscellaneous diagnosis (616). A total of 3,063 IVIs were performed in the operating room including 683 ranibizumab, 2,364 bevacizumab, and 16 triamcinolone acetonide. The diagnosis included neovascular age-related macular degeneration (1,836), diabetic macular edema (771), retinal vein occlusion (189), and miscellaneous diagnosis (267). Table 1. Procedure for Performing Intravitreal Injections In Office Operating Room Povidine Iodine 5% prep Surgical mask Sterile gloves Sterile drape Sterile eyelid speculum Restriction on conversation during procedure Pre-injection antibiotic eye drops Post-injection antibiotic eye drops
Yes No No* No Yes No
Yes Yes Yes Yes Yes Yes
No
Yes
Yes
Yes
*Nonsterile disposable gloves were used.
20
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 1
A total of 5 cases (0.043%) of clinically suspected infectious endophthalmitis occurred after IVIs. Three cases (0.035%) occurred in the office-based cohort and 2 cases (0.065%) occurred in the operating room cohort. Cases presented between 3 days and 20 days after the IVI with symptoms of eye pain, increased floaters, decreased vision and red eye. All cases had diagnostic vitreous tap for microbiologic studies and were treated with intravitreal antibiotics. One case was culture positive (Staphylococcus epidermidis) and 4 cases were culture negative. All cases were treated with IVIs of vancomycin (1 mg/0.1 mL) and ceftazidime (2.25 mg/0.1 mL). The eyes with endophthalmitis had received an average of 10 anti-VEGF injections (range, 1–34 injections) before presenting with endophthalmitis and received an average of 9.4 subsequent intravitreal anti-VEGF injections (range, 0–17 injections). The characteristics of the cases of endophthalmitis are outlined in Table 2. There was no significant difference in the rate of endophthalmitis between the office-based versus operating room cohorts (Fisher exact test, P = 0.6111), The odds ratio of endophthalmitis for office-based versus operating room procedures was 0.53 with a 95% confidence interval of 0.089 to 3.18. There was no significant difference in rate of various baseline diagnosis (Fisher exact test, P = 0.6594), and the type of medication that was injected (Fisher exact test, P = 1.0000). Discussion Intravitreal injection of anti-VEGF drugs has become one of the most commonly performed procedures in ophthalmology. Although considered generally safe,
rarely serious sight threatening complications such as retinal detachment or endophthalmitis may occur. The reported rates of clinically suspected endophthalmitis after IVI range from 0.018% to 1.4%.2,5,9–22 In 2004, a report by a panel of experts suggested guidelines to minimize the risk of endophthalmitis.23 Occurrence of endophthalmitis after IVI is likely to be multifactorial and include preoperative, operative, and postoperative factors. Potential risk factors for endophthalmitis include presence of significant ocular adnexal pathology, contamination of the ocular surface, needle, and the medication during the procedure, contamination of the medication during compounding process or storage, and postprocedure contamination. A number of steps to reduce the risk of IVIassociated endophthalmitis have been advocated by treating physicians. These include use of povidoneiodine, prophylactic topical antibiotics in the preinjection and postinjection periods, use of sterile drapes, sterile gloves, and eyelid speculum. A survey of retinal specialists in the United States performing IVI reported 99.6% use povidone-iodine, 92% use an eyelid speculum, 52% wear gloves, 20% wear sterile gloves, and 12% use a sterile drape.24 Recently, wearing of surgical face mask and avoidance of conversation during the procedure have been proposed.25 In many countries, the IVI is considered a “surgical procedure” and performed with full precautions in the operating room. Apart from use of povidone-iodine, there is no direct evidence that many of these precautions reduce the occurrence of endophthalmitis.26 This study evaluated the occurrence of endophthalmitis developing after IVIs in two settings: officebased and operating room settings. In the office-based group, the procedure was performed in the examination room without the use of sterile gloves, sterile
Table 2. Characteristics of Patients With Clinically Suspected Endophthalmitis
Setting Medication Baseline diagnosis Vitreous fluid culture Number of days from last IVIs to presentation with endophthalmitis Number of IVIs before endophthalmitis Number of IVIs after endophthalmitis BCVA before endophthalmitis BCVA at presentation with endophthalmitis BCVA at last follow-up Follow-up (months)
Case 1
Case 2
Case 3
Office
Office
Office
7
4
20
3
1 17 20/40 20/400
34 14 20/70 20/800
7 15 20/60 20/400
2 1 20/20 20/100
6 0 20/40 20/200
20/40 37
20/200 18
20/30 25
20/50 12
20/25 6
Ranibizumab nvAMD Staphylococcus epidermidis 6
Case 4
Case 5
Operating Operating room room Bevacizumab Bevacizumab Ranibizumab Ranibizumab nvAMD nvAMD Myopic CNV nvAMD Negative Negative Negative Negative
BCVA, best-corrected visual acuity; CNV, choroidal neovascularization; nvAMD, neovascular age-related macular degeneration.
ENDOPHTHALMITIS AND INTRAVITREAL INJECTION TABANDEH ET AL
drape, or face mask. In the operating room setting, sterile drapes, sterile gloves, and surgical scrubs and face mask were used. In both settings, 5% povidone-iodine was used to prepare the ocular surface, and sterile eyelid speculum was used. In both cohorts, the rates of postinjection endophthalmitis were low and similar to those reported in the existing literature. This finding suggests that the operating room protocol does not seem to significantly lower the rate of endophthalmitis compared with the officebased setting. In a meta-analysis of endophthalmitis after IVI of anti-VEGF agents, McCannel25 reported a disproportionately higher occurrence of Streptococcus species compared with endophthalmitis occurring after cataract surgery. A relatively higher occurrence of Streptococcal isolates were also reported by Moshfeghi et al,20 and Shah et al,22 although the most commonly reported organism remained coagulase negative Staphylococcus. It has been postulated that the disproportionately high occurrence of streptococcus species may be because of respiratory airborne droplet transmission from the health care provider or patient during the procedure. Wearing of face mask and a restriction on conversation during the procedure has been suggested. In this study, the rate of endophthalmitis in the office-based setting, where face masks were not used and no restrictions were placed on conversation, was similar to the operating room setting where face masks were routinely used. In a study of simulated IVI, bacterial dispersal was associated with 5 minutes of speech. Wearing a face mask or remaining silent decreased culture plate contamination.27 However, in a study comparing pretreatment with povidone-iodine, wearing of face mask, and avoidance of conversation, plates pretreated with povidone-iodine demonstrated the least bacterial growth.28 It is likely that in real life situation if organisms originate from the injecting physician or patient, they are neutralized on contact with povidone-iodine, that is, present on the ocular surface. The disproportionately higher occurrence of streptococcus species may represent the higher virulence of the organism and its ability to adhere to and penetrate the ocular surface tissue. Intravitreal injections are akin to transient “micro-trabeculectomies” where a residual microtract may be present for a short time after the injection. After multiple injections, the chances of transient or persistent “micro tracts” with or without vitreous incarceration under the conjunctival surface increases. The organisms that are able to adhere and penetrate the surface tissue would be expected to be more likely to cause endophthalmitis. Similarly, a higher prevalence of Streptococcus species has been observed in delayed-onset bleb-associated
21
endophthalmitis.29–32 It is possible that ocular surface colonization by the patient’s own upper respiratory tract flora including streptococcus species, and by airborne droplet transmission during the postinjection period may contribute to the disproportionate occurrence of streptococcus species among culture positive endophthalmitis occurring after IVIs. Efficacy of prophylactic topical antibiotics in preventing endophthalmitis is not established. There are suggestions that prophylactic use of topical antibiotics may be associated with development of resistance and subsequent colonization by more virulent organisms and therefore increased risk of endophthalmitis.33–36 Use of a broad spectrum antibiotic eye ointment in the immediate postinjection period may reduce the possibility of contamination during the postprocedure period by the antibacterial action, and also by providing a temporary protective layer against bacterial contamination while the injection tract undergoes the initial wound healing process. However, use of a postinjection prophylactic topical antibiotic with suboptimal coverage for the important organisms may be counterproductive by promoting colonization with more virulent pathogens that may be able to penetrate into the eye through the injection microtract. Indeed there is suggestion that prophylactic topical antibiotic use may be associated with a higher rate of endophthalmitis.14,37 In this study, the rate of endophthalmitis was slightly higher in the operating room setting (0.065%) compared with the office setting (0.035%). The rate of endophthalmitis was low and comparable to previously published series in both groups, implying that there was no “clinically” significant difference between the two groups. A statistically significant difference was not detected either. Even though a “statistically” significant difference was not detected, it does not mean that one does not exist. Given that the rate of endophthalmitis is low (0.0001–0.0005 per IVI), using a 2-sided Fisher exact test with a Type 1 error (alpha error) of 0.05, a study of at least 90,000 patients in each group is required to achieve 80% power. The findings of this study may allow for the possibility of a meta-analysis of other future studies of similar construct, evaluating larger groups of patients. To summarize, IVIs have become one of the most frequently performed procedures in ophthalmology. Endophthalmitis is a serious and potentially sight threatening complication. Concerns for development of endophthalmitis have influenced the technique for this commonly performed procedure. A varying number of technical precautions have been advocated in an attempt to reduce the risk. Although many of these
22
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 1
precautions make clinical sense, there is no direct evidence that precautions, such as use of a surgical drape, sterile gloves, face masks, preinjection antibiotics, nor postinjection antibiotics make a clinically significant difference to the rate of endophthalmitis. This study, although with limitations of a retrospective study, indicates that the rate of endophthalmitis after IVIs is low whether the procedure is performed in the office or operating room setting, where many of the above mentioned precautions are in place. The findings have implications in terms of safety, patient convenience, efficiency, and economics of administering these treatments. Key words: age-related macular degeneration, diabetic retinopathy, endophthalmitis, intravitreal injection, macular edema, retinal vein occlusion, ranibizumab, bevacizumab, triamcinolone. References 1. Brown DM, Campochiaro PA, Singh RP, et al. Ranibizumab for macular edema following central retinal vein occlusion: sixmonth primary end point results of a phase III study. Ophthalmology 2010;117:1124–1133.e1. 2. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med 2006;355:1432–1444. 3. Campochiaro PA, Heier JS, Feiner L, et al. Ranibizumab for macular edema following branch retinal vein occlusion: sixmonth primary end point results of a phase III study. Ophthalmology 2010;117:1102–1112.e1. 4. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An optical coherence tomography-guided, variable dosing regimen with intravitreal ranibizumab (Lucentis) for neovascular age-related macular degeneration. Am J Ophthalmol 2007;143:566–583. 5. Heier JS, Antoszyk AN, Pavan PR, et al. Ranibizumab for treatment of neovascular age-related macular degeneration: a phase I/II multicenter, controlled, multidose study. Ophthalmology 2006;113:633–642. 6. Martin DF, Maguire MG, Fine SL, et al. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology 2012;119: 1388–1398. 7. Martin DF, Maguire MG, Ying GS, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med 2011;364:1897–1908. 8. Regillo CD, Brown DM, Abraham P, et al. Randomized, double-masked, sham-controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER study year 1. Am J Ophthalmol 2008;145:239–248. 9. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006;355:1419–1431. 10. Antoszyk AN, Tuomi L, Chung CY, Singh A. Ranibizumab combined with verteporfin photodynamic therapy in neovascular age-related macular degeneration (FOCUS): year 2 results. Am J Ophthalmol 2008;145:862–874. 11. Artunay O, Yuzbasioglu E, Rasier R, et al. Incidence and management of acute endophthalmitis after intravitreal bevacizumab (Avastin) injection. Eye (Lond) 2009;23:2187–2193.
12. Bhatt SS, Stepien KE, Joshi K. Prophylactic antibiotic use after intravitreal injection: effect on endophthalmitis rate. Retina 2011;31:2032–2036. 13. Bhavsar AR, Googe JM Jr, Stockdale CR, et al. Risk of endophthalmitis after intravitreal drug injection when topical antibiotics are not required: the diabetic retinopathy clinical research network laser-ranibizumab-triamcinolone clinical trials. Arch Ophthalmol 2009;127:1581–1583. 14. Cheung CS, Wong AW, Lui A, et al. Incidence of endophthalmitis and use of antibiotic prophylaxis after intravitreal injections. Ophthalmology 2012;119:1609–1614. 15. Diago T, McCannel CA, Bakri SJ, et al. Infectious endophthalmitis after intravitreal injection of antiangiogenic agents. Retina 2009;29:601–605. 16. Fintak DR, Shah GK, Blinder KJ, et al. Incidence of endophthalmitis related to intravitreal injection of bevacizumab and ranibizumab. Retina 2008;28:1395–1399. 17. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med 2004;351:2805–2816. 18. Klein KS, Walsh MK, Hassan TS, et al. Endophthalmitis after anti-VEGF injections. Ophthalmology 2009;116:1225.e1. 19. Mason JO III, White MF, Feist RM, et al. Incidence of acute onset endophthalmitis following intravitreal bevacizumab (Avastin) injection. Retina 2008;28:564–567. 20. Moshfeghi AA, Rosenfeld PJ, Flynn HW Jr, et al. Endophthalmitis after intravitreal anti-vascular endothelial growth factor antagonists: a six-year experience at a university referral center. Retina 2011;31:662–668. 21. Pilli S, Kotsolis A, Spaide RF, et al. Endophthalmitis associated with intravitreal anti-vascular endothelial growth factor therapy injections in an office setting. Am J Ophthalmol 2008;145:879–882. 22. Shah CP, Garg SJ, Vander JF, et al. Outcomes and risk factors associated with endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor agents. Ophthalmology 2011;118:2028–2034. 23. Aiello LP, Brucker AJ, Chang S, et al. Evolving guidelines for intravitreous injections. Retina 2004;24:S3–S19. 24. Green-Simms AE, Ekdawi NS, Bakri SJ. Survey of intravitreal injection techniques among retinal specialists in the United States. Am J Ophthalmol 2011;151:329–332. 25. McCannel CA. Meta-analysis of endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor agents: causative organisms and possible prevention strategies. Retina 2011;31:654–661. 26. Speaker MG, Menikoff JA. Prophylaxis of endophthalmitis with topical povidone-iodine. Ophthalmology 1991;98:1769– 1775. 27. Wen JC, McCannel CA, Mochon AB, Garner OB. Bacterial dispersal associated with speech in the setting of intravitreous injections. Arch Ophthalmol 2011;129:1551–1554. 28. Doshi RR, Leng T, Fung AE. Reducing oral flora contamination of intravitreal injections with face mask or silence. Retina 2012;32:473–476. 29. Busbee BG, Recchia FM, Kaiser R, et al. Bleb-associated endophthalmitis: clinical characteristics and visual outcomes. Ophthalmology 2004;111:1495–1503; discussion 1503. 30. Song A, Scott IU, Flynn HW Jr, Budenz DL. Delayed-onset bleb-associated endophthalmitis: clinical features and visual acuity outcomes. Ophthalmology 2002;109:985–991. 31. Jacobs DJ, Leng T, Flynn HW Jr, et al. Delayed-onset blebassociated endophthalmitis: presentation and outcome by culture result. Clin Ophthalmol 2011;5:739–744.
ENDOPHTHALMITIS AND INTRAVITREAL INJECTION TABANDEH ET AL 32. Leng T, Miller D, Flynn HW Jr, et al. Delayed-onset blebassociated endophthalmitis (1996-2008): causative organisms and visual acuity outcomes. Retina 2011;31:344–352. 33. Alabiad CR, Miller D, Schiffman JC, Davis JL. Antimicrobial resistance profiles of ocular and nasal flora in patients undergoing intravitreal injections. Am J Ophthalmol 2011;152:999– 1004.e2. 34. Kim SJ, Toma HS. Antimicrobial resistance and ophthalmic antibiotics: 1-year results of a longitudinal controlled study of patients undergoing intravitreal injections. Arch Ophthalmol 2011;129:1180–1188.
23
35. Moss JM, Sanislo SR, Ta CN. Antibiotic susceptibility patterns of ocular bacterial flora in patients undergoing intravitreal injections. Ophthalmology 2010;117:2141–2145. 36. Milder E, Vander J, Shah C, Garg S. Changes in antibiotic resistance patterns of conjunctival flora due to repeated use of topical antibiotics after intravitreal injection. Ophthalmology 2012;119:1420–1424. 37. Bhavsar AR, Stockdale CR, Ferris FL III, et al. Update on risk of endophthalmitis after intravitreal drug injections and potential impact of elimination of topical antibiotics. Arch Ophthalmol 2012;130:809–810.
I
procedures. Endophthalmitis remains a sight threatening complication of intravitreal administration of drugs. A varying number of steps have been taken in an attempt to reduce the risk of IVI-associated endophthalmitis including the use of prophylactic topical antibiotics, sterile drapes, sterile gloves, face mask, eyelid speculum, and topical povidone-iodine. Performing the procedure in the operating room encompasses many of the above mentioned measures and has been mandated in many countries with the assumption that it decreases the rate of infection. These measures impact safety, patient convenience, efficiency, and cost of administrating these treatments. With the exception of povidone-iodine, there is little direct evidence that the above measures result in a decrease in rate of endophthalmitis. This study evaluated the rate of endophthalmitis associated with IVIs performed in two settings: officebased setting and operating room setting.
ntravitreal injection of anti-vascular endothelium growth factor (VEGF) agents has become the standard of care for the treatment of neovascular age-related macular degeneration and macular edema secondary to retinal vein occlusion.1–9 There is also widespread use for the treatment of retinal complications of diabetes. This has made intravitreal injection (IVI) of antiVEGF drugs one of the most commonly performed
From the *Retina-Vitreous Associates Medical Group, Los Angeles, California; †Department of Ophthalmology, University of Bari, Bari, Italy; ‡Department of Ophthalmology, Polytechnic University of Ancona, Ancona, Italy; and §Bascom Palmer Eye Institute, Miami, Florida. Presented at the Association for Research in Vision and Ophthalmology Annual Meeting, Fort Lauderdale, FL, May 2012. None of the authors have any financial/conflicting interests to disclose. Reprint requests: Homayoun Tabandeh, MD, MS, RetinaVitreous Associates Medical Group, 9001 Wilshire Boulevard, Suite 301, Beverly Hills, CA 90211; e-mail: [email protected]
18
ENDOPHTHALMITIS AND INTRAVITREAL INJECTION TABANDEH ET AL
Methods Consecutive patients undergoing IVIs of ranibizumab, bevacizumab, and triamcinolone between January 2009 and December 2011 were identified by the search of the diagnostic and billing databases, and review of medical records of practices of 2 retina physicians (H.T. and F.B.) at the Retina-Vitreous Associates Medical Group, Los Angeles, and Department of Ophthalmology, University of Bari. The study protocol was reviewed by the regional institutional review board who concluded that approval was not required for this study. The study was in accord with the principles outlined in the Declaration of Helsinki. Data was collected on type of injected medication, diagnosis, and occurrence of endophthalmitis. Medical records of patients with clinically suspected endophthalmitis were reviewed and further data was collected including the microbiology results, baseline diagnosis, and type and number of injections. All data were collected and entered onto a computerized database (Microsoft Excel 2007; Microsoft Corporation, Redmond, WA). Statistical analysis was performed using statistical software, SAS version 9.3 (Cary, NC). The difference in rate of endophthalmitis between various factors was compared using the Fisher exact test. P , 0.05 were considered as statistically significant. In the office-based setting, intravitreous injection was performed in the examination room. In a majority of the cases, anesthesia was achieved with 0.5% topical proparacaine hydrochloride (Akorn Pharmaceuticals, Lake Forest, IL) followed by application of a 4% lidocaine-soaked cotton tip applicator to the injection site. Subconjunctival injection of 2% lidocaine was used in a minority of patients. A sterile eyelid speculum was placed followed by administration of 5% Povidone–iodine, using cotton tip applicators, to prepare the ocular surface before injection. Nonsterile disposable gloves were used throughout the procedure. Surgical mask was not used and no restriction was placed on conversation during the procedure. Preoperative topical antibiotics, sterile gloves, and sterile drapes were not used. Ocular surface was irrigated with saline solution and topical antibiotic drops were used immediately after the injection. Patients were instructed to use topical antibiotic eye drops 4 times a day for 4 days postinjection. Bevacizumab was prepackaged by a compounding pharmacy (California Pharmacy and Compounding Center, Newport Beach, CA). Ranibizumab and triamcinolone were prepared by the treating physician immediately before injection as per package instruction.
19
In the operating room setting, anesthesia was achieved with 0.4% topical benoxinato cloridrato single use vials. Intravitreal injection protocol included the use of preoperative topical antibiotics, sterile gloves, surgical mask, sterile drape, and sterile eyelid speculum. The treating physician used a sterile scrub brush soaked with povidone-iodine before wearing sterile glove. Povidone-iodine (5%) was used to prep the ocular surface before injection. Restriction was placed on all conversation during procedure. The patients were instructed to use topical antibiotic eye drops 4 times a day for 4 days post injection. Bevacizumab, ranibizumab, and triamcinolone acetonide (Kenacort; Bristol-Myers Squibb, Anagni, Italy) were prepared by the treating physician immediately before the injection. Bevacizumab was prepared in the operating room from the stock bottle using “Securemix” (Eurospital SPA, Trieste, Italy), a self-contained system for drug reconstitution, using a 0.22-mm filter (Table 1).
Results From January 2009 to December 2011, a total of 11,710 IVIs including 2,724 ranibizumab, 8,533 bevacizumab were performed by 2 physicians (H.T. and F.B.). Eight thousand six hundred and forty-seven IVIs were performed in the office-based setting and included 2,041 ranibizumab, 6,169 bevacizumab, and 437 triamcinolone acetonide injections. The diagnosis included neovascular age-related macular degeneration (5,376), diabetic macular edema (1,587), retinal vein occlusion (1,068), and miscellaneous diagnosis (616). A total of 3,063 IVIs were performed in the operating room including 683 ranibizumab, 2,364 bevacizumab, and 16 triamcinolone acetonide. The diagnosis included neovascular age-related macular degeneration (1,836), diabetic macular edema (771), retinal vein occlusion (189), and miscellaneous diagnosis (267). Table 1. Procedure for Performing Intravitreal Injections In Office Operating Room Povidine Iodine 5% prep Surgical mask Sterile gloves Sterile drape Sterile eyelid speculum Restriction on conversation during procedure Pre-injection antibiotic eye drops Post-injection antibiotic eye drops
Yes No No* No Yes No
Yes Yes Yes Yes Yes Yes
No
Yes
Yes
Yes
*Nonsterile disposable gloves were used.
20
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 1
A total of 5 cases (0.043%) of clinically suspected infectious endophthalmitis occurred after IVIs. Three cases (0.035%) occurred in the office-based cohort and 2 cases (0.065%) occurred in the operating room cohort. Cases presented between 3 days and 20 days after the IVI with symptoms of eye pain, increased floaters, decreased vision and red eye. All cases had diagnostic vitreous tap for microbiologic studies and were treated with intravitreal antibiotics. One case was culture positive (Staphylococcus epidermidis) and 4 cases were culture negative. All cases were treated with IVIs of vancomycin (1 mg/0.1 mL) and ceftazidime (2.25 mg/0.1 mL). The eyes with endophthalmitis had received an average of 10 anti-VEGF injections (range, 1–34 injections) before presenting with endophthalmitis and received an average of 9.4 subsequent intravitreal anti-VEGF injections (range, 0–17 injections). The characteristics of the cases of endophthalmitis are outlined in Table 2. There was no significant difference in the rate of endophthalmitis between the office-based versus operating room cohorts (Fisher exact test, P = 0.6111), The odds ratio of endophthalmitis for office-based versus operating room procedures was 0.53 with a 95% confidence interval of 0.089 to 3.18. There was no significant difference in rate of various baseline diagnosis (Fisher exact test, P = 0.6594), and the type of medication that was injected (Fisher exact test, P = 1.0000). Discussion Intravitreal injection of anti-VEGF drugs has become one of the most commonly performed procedures in ophthalmology. Although considered generally safe,
rarely serious sight threatening complications such as retinal detachment or endophthalmitis may occur. The reported rates of clinically suspected endophthalmitis after IVI range from 0.018% to 1.4%.2,5,9–22 In 2004, a report by a panel of experts suggested guidelines to minimize the risk of endophthalmitis.23 Occurrence of endophthalmitis after IVI is likely to be multifactorial and include preoperative, operative, and postoperative factors. Potential risk factors for endophthalmitis include presence of significant ocular adnexal pathology, contamination of the ocular surface, needle, and the medication during the procedure, contamination of the medication during compounding process or storage, and postprocedure contamination. A number of steps to reduce the risk of IVIassociated endophthalmitis have been advocated by treating physicians. These include use of povidoneiodine, prophylactic topical antibiotics in the preinjection and postinjection periods, use of sterile drapes, sterile gloves, and eyelid speculum. A survey of retinal specialists in the United States performing IVI reported 99.6% use povidone-iodine, 92% use an eyelid speculum, 52% wear gloves, 20% wear sterile gloves, and 12% use a sterile drape.24 Recently, wearing of surgical face mask and avoidance of conversation during the procedure have been proposed.25 In many countries, the IVI is considered a “surgical procedure” and performed with full precautions in the operating room. Apart from use of povidone-iodine, there is no direct evidence that many of these precautions reduce the occurrence of endophthalmitis.26 This study evaluated the occurrence of endophthalmitis developing after IVIs in two settings: officebased and operating room settings. In the office-based group, the procedure was performed in the examination room without the use of sterile gloves, sterile
Table 2. Characteristics of Patients With Clinically Suspected Endophthalmitis
Setting Medication Baseline diagnosis Vitreous fluid culture Number of days from last IVIs to presentation with endophthalmitis Number of IVIs before endophthalmitis Number of IVIs after endophthalmitis BCVA before endophthalmitis BCVA at presentation with endophthalmitis BCVA at last follow-up Follow-up (months)
Case 1
Case 2
Case 3
Office
Office
Office
7
4
20
3
1 17 20/40 20/400
34 14 20/70 20/800
7 15 20/60 20/400
2 1 20/20 20/100
6 0 20/40 20/200
20/40 37
20/200 18
20/30 25
20/50 12
20/25 6
Ranibizumab nvAMD Staphylococcus epidermidis 6
Case 4
Case 5
Operating Operating room room Bevacizumab Bevacizumab Ranibizumab Ranibizumab nvAMD nvAMD Myopic CNV nvAMD Negative Negative Negative Negative
BCVA, best-corrected visual acuity; CNV, choroidal neovascularization; nvAMD, neovascular age-related macular degeneration.
ENDOPHTHALMITIS AND INTRAVITREAL INJECTION TABANDEH ET AL
drape, or face mask. In the operating room setting, sterile drapes, sterile gloves, and surgical scrubs and face mask were used. In both settings, 5% povidone-iodine was used to prepare the ocular surface, and sterile eyelid speculum was used. In both cohorts, the rates of postinjection endophthalmitis were low and similar to those reported in the existing literature. This finding suggests that the operating room protocol does not seem to significantly lower the rate of endophthalmitis compared with the officebased setting. In a meta-analysis of endophthalmitis after IVI of anti-VEGF agents, McCannel25 reported a disproportionately higher occurrence of Streptococcus species compared with endophthalmitis occurring after cataract surgery. A relatively higher occurrence of Streptococcal isolates were also reported by Moshfeghi et al,20 and Shah et al,22 although the most commonly reported organism remained coagulase negative Staphylococcus. It has been postulated that the disproportionately high occurrence of streptococcus species may be because of respiratory airborne droplet transmission from the health care provider or patient during the procedure. Wearing of face mask and a restriction on conversation during the procedure has been suggested. In this study, the rate of endophthalmitis in the office-based setting, where face masks were not used and no restrictions were placed on conversation, was similar to the operating room setting where face masks were routinely used. In a study of simulated IVI, bacterial dispersal was associated with 5 minutes of speech. Wearing a face mask or remaining silent decreased culture plate contamination.27 However, in a study comparing pretreatment with povidone-iodine, wearing of face mask, and avoidance of conversation, plates pretreated with povidone-iodine demonstrated the least bacterial growth.28 It is likely that in real life situation if organisms originate from the injecting physician or patient, they are neutralized on contact with povidone-iodine, that is, present on the ocular surface. The disproportionately higher occurrence of streptococcus species may represent the higher virulence of the organism and its ability to adhere to and penetrate the ocular surface tissue. Intravitreal injections are akin to transient “micro-trabeculectomies” where a residual microtract may be present for a short time after the injection. After multiple injections, the chances of transient or persistent “micro tracts” with or without vitreous incarceration under the conjunctival surface increases. The organisms that are able to adhere and penetrate the surface tissue would be expected to be more likely to cause endophthalmitis. Similarly, a higher prevalence of Streptococcus species has been observed in delayed-onset bleb-associated
21
endophthalmitis.29–32 It is possible that ocular surface colonization by the patient’s own upper respiratory tract flora including streptococcus species, and by airborne droplet transmission during the postinjection period may contribute to the disproportionate occurrence of streptococcus species among culture positive endophthalmitis occurring after IVIs. Efficacy of prophylactic topical antibiotics in preventing endophthalmitis is not established. There are suggestions that prophylactic use of topical antibiotics may be associated with development of resistance and subsequent colonization by more virulent organisms and therefore increased risk of endophthalmitis.33–36 Use of a broad spectrum antibiotic eye ointment in the immediate postinjection period may reduce the possibility of contamination during the postprocedure period by the antibacterial action, and also by providing a temporary protective layer against bacterial contamination while the injection tract undergoes the initial wound healing process. However, use of a postinjection prophylactic topical antibiotic with suboptimal coverage for the important organisms may be counterproductive by promoting colonization with more virulent pathogens that may be able to penetrate into the eye through the injection microtract. Indeed there is suggestion that prophylactic topical antibiotic use may be associated with a higher rate of endophthalmitis.14,37 In this study, the rate of endophthalmitis was slightly higher in the operating room setting (0.065%) compared with the office setting (0.035%). The rate of endophthalmitis was low and comparable to previously published series in both groups, implying that there was no “clinically” significant difference between the two groups. A statistically significant difference was not detected either. Even though a “statistically” significant difference was not detected, it does not mean that one does not exist. Given that the rate of endophthalmitis is low (0.0001–0.0005 per IVI), using a 2-sided Fisher exact test with a Type 1 error (alpha error) of 0.05, a study of at least 90,000 patients in each group is required to achieve 80% power. The findings of this study may allow for the possibility of a meta-analysis of other future studies of similar construct, evaluating larger groups of patients. To summarize, IVIs have become one of the most frequently performed procedures in ophthalmology. Endophthalmitis is a serious and potentially sight threatening complication. Concerns for development of endophthalmitis have influenced the technique for this commonly performed procedure. A varying number of technical precautions have been advocated in an attempt to reduce the risk. Although many of these
22
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 1
precautions make clinical sense, there is no direct evidence that precautions, such as use of a surgical drape, sterile gloves, face masks, preinjection antibiotics, nor postinjection antibiotics make a clinically significant difference to the rate of endophthalmitis. This study, although with limitations of a retrospective study, indicates that the rate of endophthalmitis after IVIs is low whether the procedure is performed in the office or operating room setting, where many of the above mentioned precautions are in place. The findings have implications in terms of safety, patient convenience, efficiency, and economics of administering these treatments. Key words: age-related macular degeneration, diabetic retinopathy, endophthalmitis, intravitreal injection, macular edema, retinal vein occlusion, ranibizumab, bevacizumab, triamcinolone. References 1. Brown DM, Campochiaro PA, Singh RP, et al. Ranibizumab for macular edema following central retinal vein occlusion: sixmonth primary end point results of a phase III study. Ophthalmology 2010;117:1124–1133.e1. 2. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med 2006;355:1432–1444. 3. Campochiaro PA, Heier JS, Feiner L, et al. Ranibizumab for macular edema following branch retinal vein occlusion: sixmonth primary end point results of a phase III study. Ophthalmology 2010;117:1102–1112.e1. 4. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An optical coherence tomography-guided, variable dosing regimen with intravitreal ranibizumab (Lucentis) for neovascular age-related macular degeneration. Am J Ophthalmol 2007;143:566–583. 5. Heier JS, Antoszyk AN, Pavan PR, et al. Ranibizumab for treatment of neovascular age-related macular degeneration: a phase I/II multicenter, controlled, multidose study. Ophthalmology 2006;113:633–642. 6. Martin DF, Maguire MG, Fine SL, et al. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology 2012;119: 1388–1398. 7. Martin DF, Maguire MG, Ying GS, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med 2011;364:1897–1908. 8. Regillo CD, Brown DM, Abraham P, et al. Randomized, double-masked, sham-controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER study year 1. Am J Ophthalmol 2008;145:239–248. 9. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006;355:1419–1431. 10. Antoszyk AN, Tuomi L, Chung CY, Singh A. Ranibizumab combined with verteporfin photodynamic therapy in neovascular age-related macular degeneration (FOCUS): year 2 results. Am J Ophthalmol 2008;145:862–874. 11. Artunay O, Yuzbasioglu E, Rasier R, et al. Incidence and management of acute endophthalmitis after intravitreal bevacizumab (Avastin) injection. Eye (Lond) 2009;23:2187–2193.
12. Bhatt SS, Stepien KE, Joshi K. Prophylactic antibiotic use after intravitreal injection: effect on endophthalmitis rate. Retina 2011;31:2032–2036. 13. Bhavsar AR, Googe JM Jr, Stockdale CR, et al. Risk of endophthalmitis after intravitreal drug injection when topical antibiotics are not required: the diabetic retinopathy clinical research network laser-ranibizumab-triamcinolone clinical trials. Arch Ophthalmol 2009;127:1581–1583. 14. Cheung CS, Wong AW, Lui A, et al. Incidence of endophthalmitis and use of antibiotic prophylaxis after intravitreal injections. Ophthalmology 2012;119:1609–1614. 15. Diago T, McCannel CA, Bakri SJ, et al. Infectious endophthalmitis after intravitreal injection of antiangiogenic agents. Retina 2009;29:601–605. 16. Fintak DR, Shah GK, Blinder KJ, et al. Incidence of endophthalmitis related to intravitreal injection of bevacizumab and ranibizumab. Retina 2008;28:1395–1399. 17. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med 2004;351:2805–2816. 18. Klein KS, Walsh MK, Hassan TS, et al. Endophthalmitis after anti-VEGF injections. Ophthalmology 2009;116:1225.e1. 19. Mason JO III, White MF, Feist RM, et al. Incidence of acute onset endophthalmitis following intravitreal bevacizumab (Avastin) injection. Retina 2008;28:564–567. 20. Moshfeghi AA, Rosenfeld PJ, Flynn HW Jr, et al. Endophthalmitis after intravitreal anti-vascular endothelial growth factor antagonists: a six-year experience at a university referral center. Retina 2011;31:662–668. 21. Pilli S, Kotsolis A, Spaide RF, et al. Endophthalmitis associated with intravitreal anti-vascular endothelial growth factor therapy injections in an office setting. Am J Ophthalmol 2008;145:879–882. 22. Shah CP, Garg SJ, Vander JF, et al. Outcomes and risk factors associated with endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor agents. Ophthalmology 2011;118:2028–2034. 23. Aiello LP, Brucker AJ, Chang S, et al. Evolving guidelines for intravitreous injections. Retina 2004;24:S3–S19. 24. Green-Simms AE, Ekdawi NS, Bakri SJ. Survey of intravitreal injection techniques among retinal specialists in the United States. Am J Ophthalmol 2011;151:329–332. 25. McCannel CA. Meta-analysis of endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor agents: causative organisms and possible prevention strategies. Retina 2011;31:654–661. 26. Speaker MG, Menikoff JA. Prophylaxis of endophthalmitis with topical povidone-iodine. Ophthalmology 1991;98:1769– 1775. 27. Wen JC, McCannel CA, Mochon AB, Garner OB. Bacterial dispersal associated with speech in the setting of intravitreous injections. Arch Ophthalmol 2011;129:1551–1554. 28. Doshi RR, Leng T, Fung AE. Reducing oral flora contamination of intravitreal injections with face mask or silence. Retina 2012;32:473–476. 29. Busbee BG, Recchia FM, Kaiser R, et al. Bleb-associated endophthalmitis: clinical characteristics and visual outcomes. Ophthalmology 2004;111:1495–1503; discussion 1503. 30. Song A, Scott IU, Flynn HW Jr, Budenz DL. Delayed-onset bleb-associated endophthalmitis: clinical features and visual acuity outcomes. Ophthalmology 2002;109:985–991. 31. Jacobs DJ, Leng T, Flynn HW Jr, et al. Delayed-onset blebassociated endophthalmitis: presentation and outcome by culture result. Clin Ophthalmol 2011;5:739–744.
ENDOPHTHALMITIS AND INTRAVITREAL INJECTION TABANDEH ET AL 32. Leng T, Miller D, Flynn HW Jr, et al. Delayed-onset blebassociated endophthalmitis (1996-2008): causative organisms and visual acuity outcomes. Retina 2011;31:344–352. 33. Alabiad CR, Miller D, Schiffman JC, Davis JL. Antimicrobial resistance profiles of ocular and nasal flora in patients undergoing intravitreal injections. Am J Ophthalmol 2011;152:999– 1004.e2. 34. Kim SJ, Toma HS. Antimicrobial resistance and ophthalmic antibiotics: 1-year results of a longitudinal controlled study of patients undergoing intravitreal injections. Arch Ophthalmol 2011;129:1180–1188.
23
35. Moss JM, Sanislo SR, Ta CN. Antibiotic susceptibility patterns of ocular bacterial flora in patients undergoing intravitreal injections. Ophthalmology 2010;117:2141–2145. 36. Milder E, Vander J, Shah C, Garg S. Changes in antibiotic resistance patterns of conjunctival flora due to repeated use of topical antibiotics after intravitreal injection. Ophthalmology 2012;119:1420–1424. 37. Bhavsar AR, Stockdale CR, Ferris FL III, et al. Update on risk of endophthalmitis after intravitreal drug injections and potential impact of elimination of topical antibiotics. Arch Ophthalmol 2012;130:809–810.
