Avoid injection risks

As the rate of intravitreal injections rises, so does the need for protecting the ocular surface.

The introduction of highly efficacious medications such as anti-vascular endothelial growth factors (anti-VEGF), antibiotics and steroids has resulted in a dramatic increase in the use of intravitreal injections. Although these are safe medications, they do come with the possibility of complications. Endophthalmitis (0.02% to 1.9% incidence per injection),1-6 ocular surface side effects and dry eye syndrome can occur and result in significant ocular morbidity and compromised visual acuity. Yet, no standardized protocol exists for the use of anesthetics, topical antibiotics and povidone-iodine (PI) solution prior to intravitreal injections. Although the use of peri-injection antibiotics, antiseptics and anesthetics can prevent infection and ocular pain, they also have the potential to contribute to antibiotic resistance, endophthalmitis and ocular surface changes, such as epithelial defects and corneal pathologies.1-3

In this article, we will evaluate the ocular surface side effects linked to these medications and provide recommendations for ophthalmologists who routinely perform intravitreal injections.


PI is the only preventive measure with some consensus and clinical trial support shown to decrease the risk of bacterial colonization and the risk of endophthalmitis.4,5 The possibility of reducing rates of catastrophic complications have prompted many physicians to investigate the use of this solution in combination with antibiotics in the pre- and post-injection setting. In the appropriate concentration, PI has a broad antimicrobial spectrum, including activity against bacteria, viruses and fungi.4,6,7

For preoperative antisepsis, a concentration of 1% to 5% PI is used to achieve adequate reduction in bacterial colonies. In a study by Ferguson et al., a 5% solution resulted in a 60% decrease in bacterial colony forming units (CFUs) after irrigation compared to 16.7% after treatment with 1% PI solution.5 However, prolonged contact of 5% PI solutions with the corneal epithelium has been shown to be cytotoxic, decreasing the epithelial integrity of the cornea, decreasing functional vision and increasing subjective patient complaints such as dry eye.6,7 Studies have documented that PI exposure of 30 seconds is sufficient to cause a significant decrease in bacterial CFUs.8

Hence, limiting contact of 5% PI to 60 seconds during antisepsis immediately followed by irrigation with a saline solution is encouraged to minimize adverse effects to the ocular surface. Furthermore, the use of 10% PI swabs to clean the eye and eyelashes is not encouraged, as the incidence of corneal toxicity is known to increase with higher concentrations of PI.


The use of peri-injection antibiotics in preparing the ocular surface remains controversial. In some studies, it is reported to increase the prevalence of resistant flora and risk for endophthalmitis.8-12 Many retina specialists may prescribe topical antibiotics for up to three days after an injection. However, several other studies have noted that the patients treated with post-injection antibiotic drops are more at risk of endophthalmitis than those who are not treated.8-11 Short-term use of antibiotics remains highly effective; on the other hand, chronic use and high concentrations can both damage the ocular surface from the toxicity of the drug formulation and increase the likelihood of secondary ocular infection.9

This is a 27-year-old female with a history of intravitreal injections who presented with diffuse epitheliopathy secondary to ocular surface disease.

Benzalkonium chloride (BAK) is featured in many ophthalmic antibiotic preparations. While it serves to maintain the integrity of the formulation and aide in corneal penetration, BAK is also a known allergen and cytotoxin in high concentrations. It induces apoptosis in conjunctival cells and corneal cells, which is secondary to its ability to disrupt the cellular membrane.10,11 Consequently, high concentrations and frequent use can disrupt the corneal epithelium and interfere with the lipid components in tears, causing tear film instability and eventually resulting in the development of dry eye syndrome.12

Preservative-free antibiotics, such as Vigamox (moxifloxacin, Alcon) 0.5%, are commonly used by cornea and refractive surgeons preoperatively and in treating bacterial keratitis and conjunctivitis. Studies evaluating it have shown decreases (93.12%)13 in bacteria CFUs, which are comparable to that achieved by 5% PI (91% decrease).14 Known to have excellent penetration in the cornea and anterior chamber,15 moxifloxacin is a safe alternative in the peri-injection period when ocular surface disease is present.

In the settings of compromised corneal epithelium, tear film instability and bacterial resistance secondary to repeated use of topical antibiotics with preservatives, the surgeon should work to promptly identify and address the presence of ocular surface disease. Therefore, in patients who are required to receive pre- or post-injection antibiotic drops, we encourage the use of preservative-free topical options.


Achieving adequate anesthesia remains an essential step in ensuring patients are comfortable during any injection or procedure. Commonly used medications, such as proparacaine, tetracaine, benoxinate and lidocaine, act by decreasing nerve cell membrane permeability to sodium ions, thereby preventing the ability of nerves to generate an action potential. At high concentrations or frequent use, topical anesthetics negatively affect the ocular surface via several mechanisms.

Ester- (tetracaine and proparacaine) and amide- (benzocaine and lidocaine) based anesthetics have toxic effects on stromal keratocytes that are related to temporal exposure and drug concentrations. Furthermore, these compounds also inhibit limbal corneal epithelial migration and adhesion. Combined with corneal epithelial cell detachment in prolonged use, they can result in delayed wound healing and persistent epithelial defects.16 This effect can be augmented in patients with neurotrophic corneas from underlying conditions (diabetes) by impairing ocular surface sensation and reducing reflex tear secretion, leading to increased corneal surface damage from desiccation, blinking and eye rubbing. Finally, the presence of resistant ocular flora secondary to topical antibiotic use can pose an additional risk of superimposed bacterial keratitis when the epithelial integrity is compromised from anesthetics.

Therefore, the surgeon’s choice of an agent should be based on concerns regarding corneal epithelial toxicity, allergy history and patient comfort prior to intravitreal injection.

Intravitreal injection guidelines

  • Recommend 5% povidone-iodine solution be applied immediately prior to intravitreal injection, allowing a maximum of 60 seconds for contact with the ocular surface.
  • If antibiotics are prescribed, the use of preservative-free solutions such as moxifloxacin is encouraged.
  • Avoid high strength topical anesthetics in patients suspected of having neurotrophic corneas.
  • Consider alternatives such as low strength povidone-iodine and topical anesthetics in cases of compromised ocular surface.


Intravitreal injections remain the treatment of choice for a variety of ophthalmic conditions. The clinician must determine the best combination of topical antisepsis and anesthesia for the individual patient while considering the long-term effects of each agent selected. Though no standard protocol exists, the selection of an appropriate topical anesthetic and povidone-iodine concentration is encouraged to minimize ocular surface damage. And, in cases where antibiotic use is indicated, avoid preserved solutions. OM


  1. Moss JM, Sanislo SR, Ta CN. A prospective randomized evaluation of topical gatifloxacin on conjunctival flora in patients undergoing intravitreal injections. Ophthalmology. 2009;116:1498-1501.
  2. Ta CN, Egbert PR, Singh K, Shriver EM, Blumenkranz MS, Miño De Kaspar H. Prospective randomized comparison of 3-day versus 1-hour preoperative ofloxacin prophylaxis for cataract surgery. OPHTHA. 2002;109(11):2036–40–discussion2040–1.
  3. Cheung CSY, Wong AWT, Lui A, Kertes PJ, Devenyi RG, Lam W-C. Incidence of Endophthalmitis and Use of Antibiotic Prophylaxis after Intravitreal Injections. Ophthalmology. 2012;119:1609-1614. doi:10.1016/j.ophtha.2012.02.014.
  4. Apt L, Isenberg SJ, Yoshimori R, Spierer A. Outpatient Topical Use of Povidone-iodine in Preparing the Eye for Surgery. Ophthalmology. 1989;96:289-292.
  5. Ferguson AW, Scott JA, McGavigan J, et al. Comparison of 5% povidone-iodine solution against 1% povidone-iodine solution in preoperative cataract surgery antisepsis: a prospective randomised double blind study. Br J Ophthalmol. 2003;87:163-167. doi:10.1136/bjo.87.2.163.
  6. Chou S-F, Lin C-H, Chang S-W. Povidone–iodine application induces corneal cell death through fixation. Br J Ophthalmol. 2010;95(2):bjo.2010.189407–283.
  7. Ridder WHI, Oquindo C, Dhamdhere K, Burke J. Effect of Povidone Iodine 5% on the Cornea, Vision, and Subjective Comfort. Optometry and Vision Science. 2017;94:732-741.
  8. Friedman DA, Mason JO III, Emond T, Mcgwin G Jr. Povidone–iodine contact ime and lid speculum use during intravitreal injection. Retina. 2013;33:975-981.
  9. Tu EY. Balancing antimicrobial efficacy and toxicity of currently available topical ophthalmic preservatives. Saudi Journal of Ophthalmology. 2014;28:182-187.
  10. Brasnu E, Brignole-Baudouin F, Riancho L, Warnet J-M, Baudouin C. Comparative study on the cytotoxic effects of benzalkonium chloride on the Wong-Kilbourne derivative of Chang conjunctival and IOBA-NHC cell lines. Mol Vis. 2008;14:394-402.
  11. Epstein SP, Ahdoot M, Marcus E, Asbell PA. Comparative Toxicity of Preservatives on Immortalized Corneal and Conjunctival Epithelial Cells. Journal of Ocular Pharmacology and Therapeutics. 2009;25:113.
  12. Sarkar J, Chaudhary S, Namavari A, et al. Corneal Neurotoxicity Due to Topical Benzalkonium Chloride. Invest Ophthalmol Vis Sci. 2012;53:1792.
  13. Kowalski RP, Romanowski EG, Shanks RMQ, Mah FS. The comparison of fluoroquinolones to nonfluoroquinolone antibacterial agents for the prevention of endophthalmitis in a rabbit model. Journal of Ocular Pharmacology and Therapeutics. August 2012:120827083836001.
  14. Ferguson AW, Scott JA, McGavigan J, et al. Comparison of 5% povidone-iodine solution against 1% povidone-iodine solution in preoperative cataract surgery antisepsis: a prospective randomised double blind study. Br J Ophthalmol. 2003;87:163-167.
  15. Silva GCM, Jabor VAP, Bonato PS, Martinez EZ, Faria-e-Sousa SJ. Penetration of 0.3% ciprofloxacin, 0.3% ofloxacin, and 0.5% moxifloxacin into the cornea and aqueous humor of enucleated human eyes. Braz J Med Biol Res. 2017;50:217.
  16. Tok OY, Tok L, Atay IM, Argun TC, Demirci N, Gunes A. Toxic keratopathy associated with abuse of topical anesthetics and amniotic membrane transplantation for treatment. International Journal of Ophthalmology. 2015;8:938-944.

About the Authors