Endophthalmitis Prophylaxis For Intravitreal Injections

Endophthalmitis Prophylaxis For Intravitreal Injections


With the widespread use of intravitreal injections, there has been increased concern regarding the risk of endophthalmitis (Figures 1 and 2). The incidence of endophthalmitis in studies evaluating intravitreal anti-vascular endothelial growth factor (VEGF) agents indicates that, when a standardized injection protocol is followed, the risk of endophthalmitis after intravitreal injection may be comparable to the risk of endophthalmitis following cataract surgery. Of the 7545 injections performed on 890 patients during the first year of the VISION trial of pegaptanib sodium (Macugen, OSI/Pfizer), the rate of endophthalmitis was 0.16% (12 cases of endophthalmitis).1


Protocol deviations (primarily nonuse of an eyelid speculum) occurred in 8 out of 12 (75%) of reported endophthalmitis cases, and the incidence of endophthalmitis decreased in the second year of the trial, after adherence to the injection protocol improved. In the MARINA trial, which evaluated ranibizumab (Lucentis, Genenetech), 5 of 477 (1%) patients developed endophthalmitis after a cumulative 2 years, yielding an endophthalmitis rate of 0.05% per injection (5 of 10 443 total injections).2 In the ANCHOR trial evaluating ranibizumab, 2 of 277 (0.07%) patients developed endophthalmitis at 1 year follow-up.3 Thus, prophylactic strategies that may be important in reducing the risk of endophthalmitis include attention to issues before, during, and after the injection.

In 2004, a group of experienced investigators met to create a document containing "Best Practices" guidelines for intravitreal injections in an attempt to minimize complications and optimize outcomes associated with intravitreal injections.4 Protocols from various clinical studies were reviewed, including anti-VEGF therapies for exudative agerelated macular degeneration (AMD) and antiviral therapies for cytomegalovirus retinitis. In addition, published manuscripts and abstracts were reviewed regarding the risks of endophthalmitis associated with intravitreal injections. Based on information available at that time, a set of guidelines (with varying levels of agreement among panel members) for minimizing the risk of endophthalmitis was created.


Figure 1. Endophthalmitis 2 days after intravitreal ranibizumab injection.

Guidelines supported by a high level of agreement (Table 1) among the panel members include: the (1) use of povidone-iodine for the ocular surface, eyelids, and eyelashes, (2) use of an eyelid speculum, and (3) avoidance of needle contact with eyelashes or eyelid margins. Guidelines with no clear consensus (Table 2) included the use of pre- or postinjection topical antibiotics, the use of a povidone-iodine flush, and the timing for follow-up examinations by the treating physician.

Table 1. Guidelines with More Agreement
Apply povidone-iodine to ocular surface, eyelids, and eyelashes
Use lid speculum
Avoid contamination of the needle with eyelashes or eyelid margin
Avoid extensive massage of eyelids either pre- or post-injection (to avoid expressing meibomian glands)
Avoid injecting patients who have active eyelid or ocular adnexal infection
Use adequate anesthetic for a given patient (topical drops and/or subconjunctival injection)
Avoid prophylactic or post-injection anterior chamber paracentesis

Table 2. Areas with No Clear Consensus
Most did not want to use a povidone-iodine flush, and preferred drops; (no benefits attributed to allowing the povidone-iodine to dry)
Most did not use a sterile drape
Most advocated use of gloves
Use of pre- or post-injection topical antibiotics (little published scientific data to support reduction in endophthalmitis rates)
No consesus regarding the need for clinic follow-up examination vs telephone interchange with physician or nurse



Ocular surface bacteria may be the most common sources of bacteria causing postoperative endophthalmitis.5-8 A universal strategy to reduce the risk of endophthalmitis is to reduce or eliminate the bacteria on the patient's ocular surface and eyelids. While this may be accomplished in various ways (povidone-iodine, topical antibiotics, eyelid hygiene, and sterile isolation of the surgical site), povidone-iodine is the only agent that has been demonstrated (level 2 evidence) to reduce the risk of postoperative endophthalmitis in a prospective study of patients undergoing cataract surgery.9 It is unknown whether the application of povidone-iodine in the form of drops, a soaked cotton-tipped applicator, or flush affects the ability of this agent to decrease the risk of endophthalmitis. Excessive eyelid manipulation should generally be avoided since this may be associated with an increase in bacterial flora on the operative field (although the efficacy of lid scrubs in combination with povidone-iodine has not been reported).


The value of prophylactic topical antibiotics is controversial, given the cost and lack of efficacy data to support their use. Topical antibiotics have been demonstrated to significantly reduce ocular surface bacteria but have not been proved to have a significant impact on reducing the risk of endophthalmitis.10-13 In an evidence-based review of bacterial endophthalmitis prophylaxis for cataract surgery published in 2002, the effectiveness of topical antibiotics received a grade C (possibly relevant, but not definitely related to clinical outcome).14 Fourth-generation fluoroquinolones are among the most frequently used prophylactic topical antibiotics. Deramo and associates recently reported a series of 42 eyes of 42 patients with acute-onset endophthalmitis occurring within 6 weeks after cataract surgery; 31 (74%) of these eyes were treated with perioperative gatifloxacin (Zymar, Allergan) or moxifloxacin (Vigamox, Alcon) and 24 (54%) were continuously taking one of these antibiotics at the time of endophthalmitis diagnosis.15 This report emphasizes that endophthalmitis can develop after cataract surgery despite the use of prophylactic antibiotics. Further, Miller and associates reported that gatifloxacin and moxifloxacin demonstrated an in vitro efficacy of less than 80% for 111 coagulase-negative staphylococci isolates recovered from patients with clinical endophthalmitis during a 15-year period (1990-2004).16 In addition, sensitivity of the isolates decreased during the study period, with 96.6% of the isolates sensitive to gatifloxacin and moxifloxacin during the initial 5 years of the study period compared with only 65.4% during the last 5-year period of the study (P=.02).16 As of May 1, 2007, a total of 792 injections (348 eyes) were administered in the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study, which is evaluating the safety and efficacy of intravitreal triamcinolone acetonide (Kenalog, Bristol-Myers Squibb) for macular edema associated with retinal vein occlusion, and 1550 injections (545 eyes) were administered in protocol B of the Diabetic Retinopathy Clinical Research Network, which is evaluating the safety and efficacy of intravitreal triamcinolone for diabetic macular edema; culture-proven endophthalmitis occurred following 1 of these 2342 injections for a per-injection incidence rate of 0.04%. Topical antibiotics prior to the day of injection were not a part of the ocular surface preparation used in these studies.

Figure 2. Dense fibrin filling the anterior chamber on slit-lamp examination of the same patient as Figure 1

A sterile topical anesthetic is administered as the first step in the intravitreal injection procedure. The use of a sterile eyelid speculum is recommended to avoid contact of the needle with eyelids and eyelashes. The use of a sterile drape is optional, and gloves, part of universal precautions, are appropriate. Subconjunctival anesthesia may be considered, but this requires additional instrumentation and involves manipulation that may be associated with increased surface flora. If subconjunctival anesthesia is used, it should be noted that the needle used for intravitreal injection passes through the subconjunctival space filled with anesthetic and that surface bacteria may be introduced beneath the conjunctiva. Application of povidone-iodine is recommended. Recent studies report very low endophthalmitis rates (3/10254 or 0.029%; and 1/5233 or 0.019%)17-18 when following these guidelines.


The outcomes of treatment depend not only on the safety and efficacy of the pharmacotherapy being delivered, but also on the safety and potential adverse events associated with the procedure itself. In order to optimize the outcomes associated with intravitreal injection, careful attention should be paid to minimize the risk of postinjection endophthalmitis. RP

Ingrid U. Scott, MD, MPH, is professor of ophthalmology and health evaluation sciences at the Milton S. Hershey Medical Center College of Medicine at Pennsylvania State University. Harry W. Flynn, Jr, MD, is professor and J. Donald M. Gass Distinguished Chair of Ophthalmology at the Bascom Palmer Eye Institute at the Leonard M. Miller School of Medicine at the University of Miami. Both authors reports no financial interest in information presented in this article.


  1. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med. 2004;351:2805-2816.
  2. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular agerelated macular degeneration. N Eng J Med. 2006;355:1419-1431.
  3. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006;335:1432-1444.
  4. Aiello LP, Brucker AJ, Chang S, et al. Evolving guidelines for intravitreous injection. Retina. 2004;24:S3-S19.
  5. Speaker MG, Milch FA, Shah MK, et al. Role of external bacterial flora in the pathogenesis of acute postoperative endophthalmitis. Ophthalmology. 1991;98:639-649; discussion 650.
  6. Bannerman TL, Rhoden DL, McAllister SK, et al. The source of coagulase-negative staphylococci in the Endophthalmitis Vitrectomy Study. A comparison of eyelid and intraocular isolates using pulsed-field gel electrophoresis. Arch Ophthalmol. 1997;115:357-361.
  7. Han DP, Wisniewski SR, Wilson LA, et al. Spectrum and susceptibilities of microbiologic isolates in the Endophthalmitis Vitrectomy Study. Am J Ophthalmol. 1996;122:1-17.
  8. Leong JK, Shah R, McCluskey PJ, et al. Bacterial contamination of the anterior chamber during phacoemulsification cataract surgery. J Cataract Refract Surg. 2002;28:826-833.
  9. Speaker MG, Menikoff JA. Prophylaxis of endophthalmitis with topical povidone-iodine. Ophthalmology. 1991;98:1769-1775.
  10. Isenberg S, Apt L, Yoshimori R, Khwarg S. Chemical preparation of the eye in ophthalmic surgery IV: comparison of povidone-iodine on the conjunctiva with a prophylactic antibiotic. Arch Ophthalmol. 1985;103:1340-1342.
  11. Grimes S, Mein C, Trevino S. Preoperative antibiotic and povidone-iodine preparation of the eye. Ann Ophthalmol. 1991;23:263-266.
  12. Ta CN, Egbert PR, Singh K, et al. Prospective randomized comparison of 3-day versus 1-hour preoperative ofloxacin prophylaxis for cataract surgery. Ophthalmology. 2002;109:2036-2041.
  13. Osher RH, Amdahl LD, Cheetham JK. Antimicrobial efficacy and aqueous humor concentration of preoperative and postoperative topical trimethoprim/polymyxin B sulfate versus Tobramycin. J Cataract Refract Surg. 1994;20:3-8.
  14. Ciulla TA, Starr MB, Basket S. Bacterial endophthalmitis prophylaxis for cataract surgery. An evidence-based update. Ophthalmology. 2002;109:13-26.
  15. Deramo VA, Lai JC, Fastenberg DM, Udell IJ. Acute endophthalmitis in eyes treated prophylactically with gatifloxacin and moxifloxacin. Am J Ophthalmol. 2006; 142:721-725.
  16. Miller D, Flynn PM, Scott IU, et al. In vitro fluoroquinolone resistance in staphylococcal endophthalmitis isolates. Arch Ophthalmol. 2006;124:479-483.
  17. Pillis 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.
  18. Mason JO 3rd, White MF, Feist RM, et al. Incidence of acute onset endophthalmitis following intravitreal bevacizumab (Avastin) injection. Retina. 2008;28:564-567.