Reducing Endophthalmitis Risk Following Intravitreal Injections: What Really Works?

Techniques to reduce risk before, during, and after injection.


The use of intravitreal injections has increased1 as newer treatments for diseases such as exudative macular degeneration, retinal vein occlusions, choroidal neovascularization, and diabetic retinopathy now rely on the use of intravitreal agents.2-7 Despite a low incidence of endophthalmitis following intravitreal injections,8 the prevalence of endophthalmitis is rising due to the sheer volume of intravitreal injections being performed today compared to 10 years ago. There is an increasing need for clinicians to be aware of factors that decrease the risk of developing endophthalmitis following an intravitreal injection. Several proposed mechanisms can decrease postinjection endophthalmitis, such as eyelid scrubbing, preoperative antibiotics, eyelid speculum, and povidone-iodine drops and flushes. But what really works, and what has been shown not to work? The purpose of this report is to discuss evidence-based practices to reduce endophthalmitis following intravitreal injections. As per previous guidelines9 we will discuss the steps of an intravitreal injection in terms of what happens before, during the injection, and after the injection.

Matthew R. Starr, MD, is an ophthalmology resident and Sophie J. Bakri, MD, is a professor of ophthalmology at the Mayo Clinic in Rochester, Minnesota. Dr. Starr reports no disclosures. Dr. Bakri reports consultancy to Allergan, Novartis, and Roche. Reach Dr. Bakri at

Editor’s note: This article is featured in a journal club episode of “Straight From the Cutter’s Mouth: A Retina Podcast.” Listen at .


Antibiotic Use

Managing the infectious risk following an intravitreal injection begins well before injection occurs. Evidence now supports that the use of topical antibiotics before an intravitreal injection does not decrease the incidence of endophthalmitis following an injection10 and that this practice may foster drug resistance in the host flora. In fact, in a recent large retrospective review, there was no difference in the rates of endophthalmitis in patients who received topical antibiotics before an injection vs those who did not; and more so, patients who did receive topical antibiotics before the intravitreal injection had a trend toward higher rates of endophthalmitis, but this difference was not significant (odds ratio, 1.54; 95% confidence interval, 0.77-3.10).11 The use of topical antibiotics is thus discouraged prior to undergoing an intravitreal injection.

Injection Setting

The setting in which the injection occurs is also a point of discussion among experts. A recent meta-analysis found the rate of endophthalmitis following intravitreal injections in a clinical setting to be 0.049%,12 and in a recent European study of intravitreal injections performed in an operating room, the rate of endophthalmitis was 0.0075%.13 In a retrospective analysis comparing an office setting vs an operating room, the rate of endophthalmitis in an office setting was 0.04% and in an operating room was 0.07% with no significant difference (P=.6111).14

Certainly, for most physicians and patients there is more convenience in performing an injection in the office setting, as well as a lower cost vs an operating room setting. Given the similar rates of endophthalmitis in both settings, performing injections in an office setting is as safe as performing injections in an operating room and potentially more convenient for patients.

Povidone Allergy

Some patients list povidone-iodine as an allergy, however, a true allergy to povidone-iodine is very rare.15 There has never been a reported case of anaphylaxis following conjunctival application of povidone-iodine before an intravitreal injection, although cases have been reported following the use of povidone-iodine on other mucocutaneous surfaces.15 Patients routinely mistake irritation and contact dermatitis as an allergy, so clinicians need to be astute in determining whether patients have a true povidone-iodine allergy, as povidone-iodine is an extremely important factor in preventing endophthalmitis following intravitreal injections. In a retrospective review of endophthalmitis following intravitreal injections at 2 academic centers, 11.9% of cases of endophthalmitis were in patients who did not receive povidone-iodine antisepsis prior to the injection.16

If a patient is unwilling to receive a povidone-iodine drop prior to an injection, the next best option is possibly aqueous chlorhexidine in the form of chlorhexidine gluconate 0.05% or 0.1%. In a large retrospective review of more than 40,000 intravitreal injections for which patients received aqueous chlorhexidine as a primary means of antisepsis rather than povidone-iodine, the rate of endophthalmitis was 0.0074%, similar to rates with povidone-iodine.17 There are limited data, though, comparing conjunctival antisepsis using povidone-iodine and aqueous chlorhexidine. In a prospective study of 20 patients examining microbial growth in vitro following conjunctival application of povidone-iodine in one eye and aqueous chlorhexidine in the contralateral eye, there was a single case of microbial growth following application of the drop for both aqueous chlorhexidine and povidone-iodine.18 Of note, the patients receiving aqueous chlorhexidine had significantly less pain following antisepsis application. This could affect patient compliance with intravitreal injections. There is a prospective study that compared chlorhexidine vs povidone-iodine skin preparations in clean-contaminated surgical skin incisions that showed chlorhexidine was superior to povidone-iodine in preventing surgical site skin infections.19 Unfortunately, this does not translate to intravitreal injections as this study used alcohol-based preparations of chlorhexidine which would cause significant corneal and conjunctival toxicity. Only low-concentration aqueous chlorhexidine can be used for intravitreal injection prep (not alcohol-based). Further prospective studies comparing the rates of endophthalmitis between these 2 agents could prove beneficial for those patients who cannot receive topical povidone-iodine.

Skin Conditions

Examining the surrounding skin and conjunctival tissue of the eye undergoing an intravitreal injection should also be performed prior to injection. Although there is no absolute contraindication to performing an intravitreal injection, it is generally in the best interest of the patient to defer an injection if the surrounding skin is erythematous and edematous or if there is conjunctival inflammation. There is no data to support this guideline, but it is good medical practice to treat a potential cellulitis or episcleritis and defer the injection until the patient is healed.

Blepharitis is a significant risk factor for the development of endophthalmitis following intravitreal injections.20 Patients with profound, active blepharitis should postpone their injection until after the blepharitis is treated and cleared.


Sterile Technique

During the intravitreal injection, some aspects of the sterile technique are not absolutely required. There is limited evidence for the use of sterile gloves.21 In a recent survey of retina specialists, just over half of the respondents use gloves and only half of those use sterile gloves.22 Regardless, many providers use gloves, whether sterile or not, as this is a standard concept in good medical practice and thus it is generally recommended that providers utilize gloves, whether sterile or not, while performing an intravitreal injection.

Another common practice while performing intravitreal injections is the use of a mask. A large retrospective study of more than 15,000 injections had no cases of endophthalmitis using a protocol in which providers used povidone-iodine, applied a sterile drape, and wore a mask during the injection.23 Other studies have shown that several potential pathogens causing endophthalmitis likely came from oral flora rather than skin flora.12

Some providers, rather than wear a mask, require everyone participating in the injection not to talk during the procedure. Decreased conversation during an injection has shown a decrease in microbial growth in vitro compared to talking while performing a procedure.24 Placing a sterile drape is a less common practice among retinal specialists, with only about 12% using them in the aforementioned study.22 It likely does not decrease infection risk, and it causes more patient discomfort and anxiety, which can make routine injections much more challenging.

Antibiotic Use

Just as in the pre-injection setting, the use of antibiotics during an intravitreal injection is not warranted. There is no reduced risk of endophthalmitis with topical antibiotics during or after an intravitreal injection, according to the Diabetic Retinopathy Clinical Research Network, which reported similar rates of endophthalmitis between 2 cohorts with and without topical antibiotics both during and after an intravitreal injection.25

Povidone-Iodine Preparation

Ocular-surface preparation with povidone-iodine has significantly reduced the risk of endophthalmitis in cataract surgery,26 and although there are no prospective, controlled studies that have examined intravitreal injections with and without povidone, there is a consensus among providers that this is the most important step in preventing endophthalmitis following an intravitreal injection. Compared to topical antibiotics, the application of conjunctival povidone-iodine does not confer bacterial resistance and greatly reduces microbial growth.27 The guidelines call for the use of 5% povidone-iodine to be placed over the site of the injection.

Some authors believe the whole ocular surface should be bathed, as one prospective study showed a decrease in bacterial colonization following flushing the conjunctiva with 10 mL of 5% povidone-iodine compared to 2-3 drops of 5% povidone-iodine,28 but there is no consensus on which method of povidone-iodine application is superior.

The standard protocol calls for waiting at least 30 seconds after povidone-iodine application to the conjunctival surface before proceeding with the intravitreal injection. In vivo studies have shown a decrease in bacterial colony formation after 30 seconds of povidone-iodine exposure to the conjunctival surface, but not after 15 seconds of exposure.29 After the application of povidone-iodine, nothing is to touch the injection site, such as lashes or the lid speculum. If the lashes or lid speculum touch the injection site, then the povidone-iodine preparation process should be repeated. If in addition to topical anesthetic, providers elect to use an anesthetic gel, povidone-iodine should be placed before the application of the gel and then after, because the gel can inhibit the povidone-iodine contact with the ocular surface. Lid scrubs are not recommended during the intravitreal injection process as these are considered a way for bacteria and other secretions from meibomian glands and other accessory glands to be excreted onto the ocular surface, greatly increasing the risk of endophthalmitis. This is not the same as applying povidone-iodine to the lids, which is up to the discretion of the clinician performing the injection and is typically not associated with increased meibomian gland secretions.


It is imperative to prevent the eyelids and lashes from touching the injection site after the final povidone application, typically done with the use of a lid speculum. However, in one prospective study, there was no decrease in bacterial colony forming units with the use of a lid speculum,29 and thus several providers do not use lid speculums in their everyday practice. Rather, other techniques are used to prevent lid touch, such as manual lid retraction. Several randomized trials have found no decrease in the rates of endophthalmitis when comparing manual lid retraction vs speculum use.30

Speculums have also been associated with an increased risk of drug reflux following an intravitreal injection. There is no clear consensus on the use of a lid speculum vs manual lid retraction, and it is up to the discretion and comfort level of the provider performing the injection, as well as the comfort level of the patient. The placement of a lid speculum can be associated with increased patient anxiety and this should also be taken into consideration before placing a lid speculum.


Some providers will displace conjunctiva with the use of a cotton tip applicator or other sterile second instrument during the intravitreal injection to prevent drug reflux from the vitreous cavity. The safety of this technique has not been widely studied, but in a retrospective series of more than 27,000 intravitreal injections with 23 patients developing endophthalmitis following an intravitreal injection, there was no difference in the rates of endophthalmitis between the cohort of patients with and without conjunctival displacement (0.011% and 0.0076%, respectively).31 These limited data do suggest that displacing conjunctiva during intravitreal injections does not confer added risk, but future prospective studies are warranted to confirm this finding.


Applying topical antibiotics following an intravitreal injection confers no decreased risk of endophthalmitis.25 Patients should also be instructed not to extensively rub the eye following an injection as this also can increase secretions from the meibomian glands. Postinjection follow-up is also important given the low but serious risk of endophthalmitis. There is no specific timeframe that is agreed upon by all providers, as long as a patient has scheduled follow-up. Perhaps as important as the use of povidone-iodine is discussing with the patients the warning signs of potential complications and the need to know when to come in to be examined before scheduled follow-up. Patients should be alerted to look for worsening vision, eye pain, photophobia, or any ocular discomfort that may suggest endophthalmitis, and they should know the emergency contact information for the injecting provider. These patients should be examined promptly to rule out endophthalmitis following an intravitreal injection.


An expert panel was convened in 2014 regarding intravitreal injections and from that panel came a series of guidelines for performing intravitreal injections (See Table).9 Among them were guidelines on how to decrease the risk of endophthalmitis following intravitreal injections. This panel came to the consensus that some of the most important aspects of the antiseptic technique include the use of conjunctival 5% povidone-iodine, avoiding lash or lid touch to the site of injection following the povidone-iodine, and the use of surgical masks or decreased talking during the procedure. Clearly, there is some leeway with the recommendations as not all providers could agree on the methodology behind sterilizing the conjunctiva with either the “drop” or ocular surface “flush” technique, but that it is vital to place povidone-iodine onto the conjunctival surface prior to injection.

Table. Recommendations From 2014 Expert Panel on Intravitreal Injection Technique and Endophthalmitis Prevention9
Povidone-iodine Recommended for all intravitreal injections. No clear consensus on drop technique vs conjunctival flush.
Topical antibiotics Not recommended.
Sterile drape Not recommended.
Lid scrubs Not recommended. Povidone-iodine application to the eyelids may be beneficial, but there is no consensus.
Lid speculum No consensus on recommendation.
Avoidance of lash and lid touch Recommended, either with manual lid retraction or lid speculum.
Sterile gloves Gloves are recommended, but they do not necessarily have to be sterile.

The paper did not mention alternatives to povidone-iodine when patients may have an allergy. Chlorhexidine may offer a potential alternative therapy as the rates of endophthalmitis were also very low in a large retrospective study, but no prospective studies directly compare povidone-iodine and chlorhexidine on conjunctival surfaces. The availability of chlorhexidine is also a potential problem as not all centers are able to keep aqueous chlorhexidine available for conjunctival use as other preparations contain alcohol, which is toxic to the cornea and conjunctiva.

There was also no agreement on the use of lid speculum vs manual lid retraction, as both arrive at the same end goal of keeping lids and lashes away from the injection site and conferred no difference in the risk of developing endophthalmitis following an injection. The panel also agreed that topical antibiotics and a sterile drape were not necessary during intravitreal injections, because neither decreased the endophthalmitis risk.

Endophthalmitis following intravitreal injections is uncommon, particularly when following these antiseptic protocols. However, there are still several steps during the procedure that not all providers can agree upon as the best medical practice. Regardless, there are basic antiseptic practices that should be universal to all intravitreal injections. RP


  1. RBRVS DataManager Online. Chicago, IL: American Medical Association.
  2. Ashraf M, Souka AA, Singh RP. Central retinal vein occlusion: modifying current treatment protocols. Eye (Lond). 2016;30(4):505-514.
  3. Bandello F, Cunha-Vaz J, Chong NV, et al. New approaches for the treatment of diabetic macular oedema: recommendations by an expert panel. Eye (Lond). 2012;26(4):485-493.
  4. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006;355(14):1432-1444.
  5. Group CR, Martin DF, Maguire MG, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;364(20):1897-1908.
  6. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006;355(14):1419-1431.
  7. Simo R, Sundstrom JM, Antonetti DA. Ocular Anti-VEGF therapy for diabetic retinopathy: the role of VEGF in the pathogenesis of diabetic retinopathy. Diabetes Care. 2014;37(4):893-899.
  8. Klein KS, Walsh MK, Hassan TS, et al. Endophthalmitis after anti-VEGF injections. Ophthalmology. 2009;116(6):1225 e1221.
  9. Avery RL, Bakri SJ, Blumenkranz MS, et al. Intravitreal injection technique and monitoring: updated guidelines of an expert panel. Retina. 2014;34 Suppl 12:S1-S18.
  10. Park Y, Kim KS, Park YH. Acute endophthalmitis after intravitreal injection and preventive effect of preoperative topical antibiotics. J Ocul Pharmacol Ther. 2013;29(10):900-905.
  11. Storey P, Dollin M, Pitcher J, et al. The role of topical antibiotic prophylaxis to prevent endophthalmitis after intravitreal injection. Ophthalmology. 2014;121(1):283-289.
  12. 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(4):654-661.
  13. Casparis H, Wolfensberger TJ, Becker M, et al. Incidence of presumed endophthalmitis after intravitreal injection performed in the operating room: a retrospective multicenter study. Retina. 2014;34(1):12-17.
  14. Tabandeh H, Boscia F, Sborgia A, et al. Endophthalmitis associated with intravitreal injections: office-based setting and operating room setting. Retina. 2014;34(1):18-23.
  15. Wykoff CC, Flynn HW, Jr., Han DP. Allergy to povidone-iodine and cephalosporins: the clinical dilemma in ophthalmic use. Am J Ophthalmol. 2011;151(1):4-6.
  16. Modjtahedi BS, van Zyl T, Pandya HK, Leonard RE, 2nd, Eliott D. Endophthalmitis After Intravitreal Injections in Patients With Self-reported Iodine Allergy. Am J Ophthalmol. 2016;170:68-74.
  17. Merani R, McPherson ZE, Luckie AP, et al. Aqueous Chlorhexidine for Intravitreal Injection Antisepsis: A Case Series and Review of the Literature. Ophthalmology. 2016;123(12):2588-2594.
  18. Oakley C, Allen P, Hooshmand J, Vote BJT. Pain and Antisepsis after Ocular Administration of Povidone-Iodine Versus Chlorhexidine. Retina. 2017.
  19. Darouiche RO, Wall MJ, Jr., Itani KM, et al. Chlorhexidine-Alcohol versus Povidone-Iodine for Surgical-Site Antisepsis. N Engl J Med. 2010;362(1):18-26.
  20. Lyall DA, Tey A, Foot B, et al. Post-intravitreal anti-VEGF endophthalmitis in the United Kingdom: incidence, features, risk factors, and outcomes. Eye (Lond). 2012;26(12):1517-1526.
  21. Fagan XJ, Al-Qureshi S. Intravitreal injections: a review of the evidence for best practice. Clin Exp Ophthalmol. 2013;41(5):500-507.
  22. Green-Simms AE, Ekdawi NS, Bakri SJ. Survey of intravitreal injection techniques among retinal specialists in the United States. Am J Ophthalmol. 2011;151(2):329-332.
  23. Shimada H, Hattori T, Mori R, Nakashizuka H, Fujita K, Yuzawa M. Minimizing the endophthalmitis rate following intravitreal injections using 0.25% povidone-iodine irrigation and surgical mask. Graefes Arch Clin Exp Ophthalmol. 2013;251(8):1885-1890.
  24. Friedman DA, Lindquist TP, Mason JO, 3rd, McGwin G. Needle contamination in the setting of intravitreal injections. Retina. 2014;34(5):929-934.
  25. Bhavsar AR, Stockdale CR, Ferris FL, 3rd, et al. Update on risk of endophthalmitis after intravitreal drug injections and potential impact of elimination of topical antibiotics. Arch Ophthalmol. 2012;130(6):809-810.
  26. Speaker MG, Menikoff JA. Prophylaxis of endophthalmitis with topical povidone-iodine. Ophthalmology. 1991;98(12):1769-1775.
  27. Hsu J, Gerstenblith AT, Garg SJ, Vander JF. Conjunctival flora antibiotic resistance patterns after serial intravitreal injections without postinjection topical antibiotics. Am J Ophthalmol. 2014;157(3):514-518.
  28. Safar A, Dellimore MC. The effect of povidone iodine flush versus drops on conjunctival colonization before intravitreal injections. Int Ophthalmol. 2007;27(5):307-312.
  29. Friedman DA, Mason JO, 3rd, Emond T, McGwin G, Jr. Povidone-iodine contact time and lid speculum use during intravitreal injection. Retina. 2013;33(5):975-981.
  30. Fineman MS, Hsu J, Spirn MJ, Kaiser RS. Bimanual assisted eyelid retraction technique for intravitreal injections. Retina. 2013;33(9):1968-1970.
  31. 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(10):2028-2034.