Topical Anesthetic Use on Corneal Abrasions

Patients with corneal abrasions typically come to the emergency department for eye pain.  Most physicians treat these with topical antibiotics, oral analgesia, and for those who are lucky enough 48 – 72 hour follow up with ophthalmology. Oral analgesia does a poor job of controlling these patients pain. Tetracaine is an esterase type anesthetic with a onset of action of 10 – 20 seconds and a duration of action of 10 – 15 minutes.  Use of topical anesthetics are very effective at reducing pain, but there use is discouraged secondary to poor wound healing of the corneal epithelium.  So what is the evidence for topical anesthetic use on corneal abrasions?

Where does the evidence for NO topical anesthetic use on corneal abrasions come from?

Case ReportsThe evidence for the toxic effects of topical anesthetics comes from case reports, case series, and animal studies.  The case reports and series are listed above, and in all but one of the cases the patients were abusing topical anesthetics for weeks to months and/or concentrations higher than that typically used today.  So the statements that topical anesthetics have toxic effects to the cornea and decrease or inhibit corneal reepithelialization come from patients abusing these medications and not using them for short periods of time.

Two animal studies [1] and [2] were examining toxicity of topical anesthetics.  Topical anesthetics were being used at increasing concentrations until corneal epithelial damage was seen. Both studies were from the corneas of rabbits who have a more sensitive cornea and a reduced blinking rate when compared to humans, therefore it is hard to draw generalizations to humans.

What trials have evaluated the use of topical anesthetics on corneal abrasions?

Conclusion:  Topical anesthetics did a better job of controlling pain and in all but one of the studies no statistical difference in cornea epithelial healing at 72 hours.

Hot Off the Press

What is the most recent trial to evaluate topical anesthetic use on corneal abrasions? [3]

What they did:

  • Prospective, double-blind, randomized trial of  a convenience sample of 116 patients with uncomplicated corneal abrasions
  • 1% tetracaine vs saline q30 minutes prn pain for 24 hours
  • Largest RCT evaluating safety of topical anesthetics with uncomplicated corneal abrasions

Exclusion Criteria:

  • > 36 hours after initial injury
  • < 18 years of age
  • Wear contact lenses
  • Injury to both eyes
  • Grossly contaminated foreign body in eye
  • Current herpes keratitis
  • Allergy to tetracaine
  • Injury requiring urgent optho evaluation
  • Unable to make follow up at 48 hours

Main Safety Outcomes:

  • Complications specifically attributed to topical anesthetic defined as:
    • Delayed healing
    • Enlarged abrasion
    • Recurrent corneal ulceration
    • Toxic keratitis
    • surface keratopathy
    • Corneal stromal infiltration
    • Candida and bacterial keratitis
    • Uveitis
    • Hypopyon
  • Persistent fluorescein uptake at 48 hours for corneal evaluation
  • 1 week and 1 month telephone interviews
  • Every patient had at least one follow up evaluation

Secondary Outcomes

  • Pain control with a 100-mm visual analog scale (VAS)
  • Overall effectiveness with a numeric rating scale (NRS) of 0 – 10


  • No complications attributed to topical anesthetic (CI of 0 – 6.1%), but this is a wide CI
  • No statistically significant difference in corneal healing at 48 hours (23.9% vs 21.3% with p value of 0.761)
    • Retained rust ring in 23 patients: tetracaine 22.0% vs saline 17.5% (p value = 0.544)
    • Fluorescein uptake at 48 hours: tetracaine 23.9% vs saline 21.3% (p value = 0.761)
  • No persistent symptoms at 48 hours (21.7% vs 21.3% with p value of 0.957)
  • No difference in VAS pain scores at any time between groups
  • Patients rated tetracaine having a better overall effectiveness on the NRS (7.7 vs 3.9)


  • Excluded penetrating eye injuries, large or complicated corneal abrasions, and injuries causing a significant disruption of vision
  • Some patients discovered when using the tetracaine drops they experienced burning similar to the tetracaine they were treated with in the ED which may have unblinded the patient and researcher
  • Patient compliance with administering the tetracaine or saline was not recorded
  • The study was under powered to detect a difference in the efficacy between the two groups
  • There was poor follow up at 48 hours: Tetracaine group 69.0% vs Saline Group 64%

Conclusion: Topical tetracaine used for 24 hours was safe, patients felt they had a significantly higher overall effectiveness with tetracaine, but no significant difference in the VAS pain score.


There was a commentary to this study [4] in which the author states: “Many common practices are driven by dogma, untested conventional wisdom, and opinions of both actual and self-promoted experts. Working with residents and students on a day-to-day basis is often the only thing that slows down busy clinicians long enough to consider the lack of evidence supporting many common practices.” He goes on to say that “ED physicians have been taught that topical anesthetic use is associated with poor healing, but this teaching is grounded in laboratory-based studies and case reports that describe prolonged, unsupervised use (or abuse) of topical anesthetics.” He finishes up by saying, “Whether or not to start using tetracaine routinely in the treatment of simple corneal abrasions remains up to the reader.”

How Your Pharmacist can Make a 1:10 (0.05%) Dilution From 0.5% Anesthetic

Topical Anesthetic Dilution

Topical Anesthetic and 18G Needle

The inpatient pharmacy team should label the bottle with the following instructions for the patient:

Proparacaine 0.05% Ophthalmic
1 – 2 drops in affected eye q30 min prn pain
Disp: 2 – 3 mL
Lot#:                 Date of Disp:
**Not to exceed 24 – 48 hours of use

Take Home Message:

To date the only evidence that topical anesthetics in uncomplicated corneal abrasions causing more harm come from experimental animal studies, case reports, and case series. Based on available evidence, it is most likely reasonable to send patients home with dilute topical anesthetics for a period of no more than 24 – 48 hours without patients having complications, but larger studies will need to occur before this recommendation is officially made.


  1. Bisla K et al. Concentration-Dependent Effects of Lidocaine on Cornal Epithelial Wound Healing. Invest Ophthalmol Vis Sci 1992. PMID: 1399407
  2. Maurice DM et al. the Absence of Corneal Toxicity with Low-Level Topical Anesthesia. Am J Ophthalmol 1985. PMID: 2409803
  3. Waldman N et al. Topical Tetracaine Used for 24 Hours is Safe and Rated Highly Effective by Patients for the Treatment of Pain Caused by Corneal Abrasions: A Double-Blind, Randomized Clinical Trial. Acad Emerg Med 2014. PMID: 24730399
  4. Ufberg JW. Dogma Challenged: Tetracaine for corneal Abrasions? Acad Emerg Med 2014. PMID: 24730412
  5. Waldman N et al. An Observational Study to Determine Whether Routinely Sending Patients Home With a 24-Hour Supply of Topical Tetracaine From the Emergency Department for Simple Corneal Abrasion Pain is Potentially Safe. Ann Emerg Med 2017; S0196 – 0644(17): 30195 – 6. PMID: 28483289
  6. Epstein DL et al. Keratitis From Misuse of Corneal Anesthetics. NEJM 1968. PMID: 4232838
  7. Willis WE et al. Corneal Complications of Topical Anesthetic Abuse. Can J Opthalmol 1970. PMID: 5472832
  8. Duffin RM et al. Tetracaine Toxicity. Ann Ophthalmol 1984. PMID: 6508100
  9. Chern KC et al. Corneal Anesthetic Abuse and Candida Keratitis. Ophthalmology 1996. PMID: 8628558
  10. Wasserman BN et al. Recurrent Corneal Ulceration as Late Complication of Toxic Keratitis. Br J Ophthalmol 2002. PMID: 11815357
  11. Verma S et al. A Prospective, Randomized, Double-Masked Trial to Evaluate the Role of Topical Anesthetics in Controlling Pain After Photorefractive Keratectomy. Ophthalmology 1995. PMID: 9098296
  12. Verma S et al. A Comparative Study of the Duration and Efficacy of Tetracaine 1% and Bupivacaine 0.75% in Controlling Pain Following Photorefractive Keratectomy (PRK). Our J Ophthalmol 1997. PMID: 9457454
  13. Ting JY et al. Management of Ocular Trauma in Emergency (MOTE) Trial: A Pilot Randomized Double-Blinded Trial Comparing Topical Amethocaine with Saline in the Outpatient Management of Corneal Trauma. J Emerg Trauma Shock 2009. PMID: 19561949
  14. Ball IM et al. Dilute Proparacaine for the Management of Acute Corneal Injuries in the Emergency Department. CJEM 2010. PMID: 20880433

For More on This Topic Checkout:

  1. The SGEM: Episode #83 – In Your Eyes (Topical Tetracaine for Corneal Abrasions)
  2. Swaminathan A et al. The Safety of Topical Anesthetics in the Treatment of Corneal Abrasions: A Review. J Emerg Med 2015 [epub ahead of print] PMID: 26281814
  3. Puls HA et al. Safety and Effectiveness of Topical Anesthetics in Corneal Abrasions: Systematic Review and Meta-Analysis. J Emerg Med 2015 [epub ahead of print]
  4. The SGEM: Episode #145 – Topical Anesthetics for ED Patients with Corneal Abrasions
  5. EM Lit of Note: Counterpoint – Topical Anesthetics for Corneal AbrasionsCounterpoint – Topical Anesthetics for Corneal Abrasions
  6. The SGEM: Episode #315: Comfortably Numb with Topical Tetracaine for Corneal Abrasions
Cite this article as: Salim Rezaie, "Topical Anesthetic Use on Corneal Abrasions", REBEL EM blog, April 21, 2014. Available at:
The following two tabs change content below.

Salim Rezaie

Emergency Physician at Greater San Antonio Emergency Physicians (GSEP)
Creator & Founder of REBEL EM

Like this article?

Share on facebook
Share on Facebook
Share on twitter
Share on Twitter
Share on linkedin
Share on Linkdin
Share on email
Share via Email

Want to support rebelem?

35 thoughts on “Topical Anesthetic Use on Corneal Abrasions”

  1. Salim – Great post on the topic. There’s clearly far more dogma here than actual literature to defend the traditional stance of not using these agents for pain relief. In addition to the EM literature, there are optho studies on PRK with topical anesthetics. Clearly, there is a difference between a surgically induced injury versus a naturally occurring abrasion but the PRK data dispels the idea that these agents stunt healing.

    One major weakness of the AEM 2014 study was the poor follow up at 48 hours of only about 70%. I commend the authors for tracking them down by phone within a month. Additionally, I imagine that hose with ulceration, progression of abrasion or continued pain would represent at 48 hours.

    I’ve started using this in selected patients but I do dilute down to 1/10th strength. I think it’s also clearly vital to have a long conversation with the patient about what to look out for and the importance of follow up.

    • Thanks for reading Anand. You know how you and I love #dogmalysis. Interesting that you also dilute your topical anesthetic. I usually dilute in a 10cc syringe and send the patient home with 1 – 2 days max of drops, so far, had no bad outcomes.


  2. What kind of bottle/container are you putting the 1/10 dilution in. Sending pts home with the syringe seems like it would make it difficult to apply (high risk for squirt themselves in the eye).

    • Hello Ben,
      Great question. Once I make the 1:10 mixture I dump out what is in the topical anesthetic bottle and refill with 2 – 3 cc of the 1:10 mixture and send them home with that. Hope that helps.


  3. Dear Salim,

    Thanks for covering this topic. I was the lead researcher on the 2009 trial, which was actually done as a pilot in 2003 and as the proper study in 2004. It took months to shop around and find a hospital ethics committee and an ophthalmologist who would even consider giving approval for the trial to be done. Once the paper was written we were unable to find a publisher (“too controversial”) and I ended up giving up on publication (and on my EM training too). My boss, Joseph Ting, was more persistent and found a publisher some 5 years after we had originally written the manuscript. So REBEL EM indeed!

    I would like to point out that the two earlier trials by Verma (as I recall them) were not for corneal abrasion per se, but assessed the rate of healing of a laser-caused corneal defect under sterile conditions in theatre – they are surgical, not EM trials.

    The main difficulty in our study (as it was in the pilot) was getting patients to attend for follow-up; they all felt better and could not be convinced to return in person. We had a much larger number of patients who were followed-up by phone, but only 47 attended follow-up within the pre-defined window, and thus could count towards the primary outcome measures. It would be interesting to see in clinical practice, and in the other trials, how many comply with requested reviews.

    I have no problems with sending a patient home with usual strength topical anaesthetic. If they are using it constantly, they will run out within 12 hours, and re-present to the ED … which is not a bad thing.


    • Hello Ken,
      TY for reading the post. Great point about the Verma studies. The reason I included them was that if a laceration of a cornea can heal with topical anesthetics, certainly an uncomplicated corneal abrasion will, but point taken that this clinical scenario is different than what we see in the ER.
      I agree with you that one of the biggest issues with most of the studies was the follow up, but one could argue that these patients self select themselves…in other words, people who get better won’t come back, but those that don’t will return, which as you said is a good thing.
      Interesting to hear you send people home with the usual strength topical anesthetics. At my shop we have 0.5% tetracaine and proparacaine, is it the same concentration where you are at? Appreciate your comments and points.


  4. Dear Salim
    Great work chalending dogma. I recall feeling a little guilty sneaking away with amethocaine to treat my own keratitis but it did hurt a lot. Was also a little worried I would melt my cornea! I usually have access to minims the volume of which is unlike to cause harm.

    • Hello Mark
      Thanks for reading. I agree. I generally send folks home with only 2 – 3 cc’s max. That way I can ensure they will not have more than 48 hours of anesthetic.


  5. Hi Salim,
    Nice post, made for a good read. However, I would caution you in including compounding instructions for ophthalmic preparations at the bedside for patients to take home. First, ophthalmic preparations are recommended to be prepared in a sterile environment (see Making this product up in the ED would not comply with most of these recommendations, potentially putting the patient at an increased risk of infection. If this is a product you would find routinely useful in your ED, I’m sure your inpatient pharmacy would work with you to prepare an ophthalmic product under appropriate conditions. Second, there are very clear regulations on labeling prescription products for outpatient use (these may vary slightly from state to state), but often when ED providers try to make up their own “homemade” labels for take home products they leave out some of the required labeling information and thus present a liability to the dispensing institution. I would encourage you to check with the pharmacy at your institution to make sure that any labeling on products dispensed from the ED for patient home use are compliant with regulations in your state.

    • Hello Meghan,
      Thank you for reading. Your comments are very much appreciated. I have updated the post to specify the inpatient pharmacist should do this and not the ED physician, although when I make this mixture I do it with sterile gloves and equipment, that being said, you are absolutely correct, every person needs to make sure they comply with state regulations. As for the labels, Clear instructions with a lot number and dispense date should pretty much cover any prescription bottle and comply with requirements in every state, but that being said, I agree, be sure to always double check and if possible have the inpatient pharmacy mix and this for you. I am lucky, in our ED we have an ED pharmacist who helps us with all this, so not as much of an issue. Thanks again, and great points.


  6. Hello Salim!

    Thank you for your summary. I am a new residential in Ophthalmology with yet very little experience. In my clinic I’ve been taught this “dogma”, so today thought I needed to check it up and I found your page.

    I have read that topical NSAID could contribute with some pain relief in corneal abrasions . I haven’t tried to prescribe it yet and I don’t know wether my senior colleagues do either, they usually prescribe an antibiotic ointment and tell the patient that hell will end within one or two days.

    Do you have any experience on topical NSAID versus topical anaesthetic in relieving symptoms or have you seen any studies?

    I haven’t tried to prescribe cyclopentholate either, do you usually prescribe that against painful ciliary spasm together with the tetracaine?

    All the best

    Björn Lindström

    • Dear Bjorn,

      In the literature review done for our trial about 10 years ago, as I recall it, trials of topical NSAIDs were of similar frequency and quality to trials of topical LAs. That is, very few trials, with low numbers (and no comment on the percentage of patients who met inclusion criteria who were followed through until primary end-point data was collected – presumably the unreported drop-out rate was just as high as ours).

      My overall impression was that NSAIDs were better than placebo, but not as good as LAs for analgesia. All of NSAIDs, LAs, and placebo seemed to have similar safety data … and the quality of safety data was no better for the NSAIDs than the LAs.

      As of 2004, there had been no trials with a head-to-head comparison of NSAIDs with LAs.

      So it was a source of frustration that, with suggestive but not conclusive data for either group of medications, you can easily find ophthalmologists who will endorse the use of NSAIDs clinically, but very few who will even consider the utility of investigating LAs.

      As I recall it, none of the following provide statistically significant improvement in comfort after simple corneal abrasions/FB removal: antibiotic drops, cycloplegics, patching, or oral analgesia. However, it is a long time since I have been active in academic emergency medicine (or ophthalmology for that matter), so the general wisdom might have moved on since then … or still might not have changed since I was in medical school in the early 1990’s.

      My clinical practice nowadays would be to inform patients of the literature as I understand it. I provide antibiotic cover if the abrasion/FB was from organic material. However, because providing outpatient LA is still considered “fringe”, I would not send a patient home with LA. Most of the time, the discharged patient would be comfortable after the provision of LA in the ED. If they re-presented later the same evening with unbearable pain, then only at that point would LA (a single minim) be dispensed.

      All patients are advised to be reviewed every 24 hours until corneal reepithelialisation is confirmed. However, in the Australian context at least, we repeatedly found that, of patients with no residual symptoms, fewer than 25% would actually attend even the first scheduled review.


  7. I just finished seeing a 5 year old who squished a laundry detergent pod into his eyes last night. He was given two bottles of topical anesthetic by the ER and was told to use these as needed. I was surprised these were dispensed and thought I’d see if there was literature backing for this traditionally taboo practice. His parents said he hated the drops anyway because they burned so much when they were instilled. They were happy when I told them they didn’t have to use them anymore!
    I appreciate this site and the comments and the efforts to relieve pain after minor corneal trauma. I’m biased by my ophthalmology training, true. But here are my concerns:
    *Patients tend to put bottles in their med cabinets and use them whenever they have eye symptoms. If they are sent home with numbing drops they should be cautioned about prolonged use of these or use in un-examined future problems. They should be advised of the concern that their use might slow epithelial healing. When I have seen patients given topical anesthetic drops, they are told nothing about concerns with their use.
    *If you have seen someone abuse anesthetic drops and get in to real trouble as every ophthalmologist has, you understandably have a bias against the widespread practice of sending folks home with anesthetic drops. I see that you are suggesting formulating diluted drops, but I fear the ease of giving drops out of the ER eye kits will lead to problems that are not always published as case reports. As a private eye doc, I am not likely to take the time to do a case report on anesthetic related complications. Though I might be now after visiting this site.
    *Any preserved drop with BAK as a preservative used hourly will slow wound healing. This may not be significant for every patient, but for certain patients it leads to prolonged epithelial healing. Nonpreserved celluvisc provides a great cushion between the lid and the epi defect and can be used prn without slowing healing.
    *Topical NSAIDS don’t provide great relief of pain and generic diclofenac was associated with disasters. Use lubricants.
    *I can’t believe that topical cycloplegics have not been shown to relieve pain associated with ciliary spasm that results from keratitis and abrasions. My own experience over 20 years is that these are amazing in some patients at helping with the photophobia associated with such problems and they are underutilized in ER settings.
    *I totally agree that narcotics don’t help much for the pain associated with corneal trauma/keratitis. Since that is known, why do ERs still give narcotics so freely for this type of problem? Seems to me we have a problem with overuse of these agents and they should be used much more cautiously in this setting.
    I appreciate the discussions and data driven research. I’m concerned about small studies that are not large enough to catch disasters related to take home numbing drops that can happen.

    • Dear Barbara,

      A few of points:

      The abuse of local anaesthetic eye drops seems to be a consequence of providing patients with large quantities of local anaesthesia. The only topical eye anaesthetics I have seen in Australia are minims (0.5mL) of either 0.4% oxybuprocaine/benoxinate or 0.5%/1.0% amethocaine/tetracaine. Neither use preservatives. (Though on occasions, in isolated practice, I have used 1% lignocaine when naught else is available.)

      The provision of a single 0.5mL minim of either of these is highly unlikely to cause anaesthetic abuse. With continuous use, the patient would run out in under 24 hours. If they still have pain when they run out, that should be a trigger for them to come back to the ED for review.

      I agree that provision of large volumes of local anaesthetic eye drops to patients, if such large volume containers are available in your country, would be highly inappropriate.

      When we did our literature review in 2003 for the Ting JY, et al (2009) trial, we referenced a BestBET from 2001 that could only find a single 1996 trial looking at the use of cycloplegics in corneal abrasion. That trial demonstrated there was no difference in analgesia between lubricant and cycloplegic, and no difference between NSAID alone and NSAID + cycloplegic. [J Accid Emerg Med (1996); 13: 186-8]. There may have been further research since then, but more likely there hasn’t.

      I agree that a few small trials, such as the 3 published thus far, are not enough to catch infrequent disasters, and so are probably not enough to change standard practice. Hopefully though, with the EM blogosphere chatter around Waldman et al’s most recent study, there is now enough data and publicity to make this a valid topic for further, more conclusive, research. Hopefully, the days of having to defend against ophthalmologist claims that “it is inappropriate to even consider doing such research” are behind us.


  8. Hello Salim,
    Your post on the controversial use of topical anaesthetics use is interesting. For a few patients who had corneal abrasion and who were very sensitive to pain I used Lignocaine jelly 2%. On applying the stinging effect was for 10-20 seconds, very severe. This was done just before applying antibiotic eye ointment and pad. Have you used Lignocaine jelly 2% ? Its available here in India for topical cataract surgery. Please reply.
    Thank you
    Dr. S Luke

    • Hello Salim,
      Lignocaine jelly 2% is not something available to me in the US, also I would be concerned about the concentration being too high. None of the research we have discussed looked at this particular anesthetic or concentration. Hope this helps.


  9. thank you, I finally feel vindicated. I’ve been researching this topic since Emergency clinic refused to send me home with any medication after I sustained a chemical burn to eyeball and cornea. Refused to give me eye drops to take home, or Rx for anything. Said it would heal in about 24hrs. Went straight from there to regular ER, got more drops for the severe pain, and Rx for Percoset. That did help somewhat but not like numbing drops. Why is opioid Rx considered better than drops that seem to have no ill effects until multiple days of abuse??? Every doctor I asked repeated the mantra -oh, corneal melt. BS!

  10. My god I wish this would get figured out. I’m in the US. I have suffered over 70 RCEs, in both eyes, since I started keeping track in 2010, and surely dozens before then. It’s maddening to be told that something that could give me relief when it’s at its worst is going to damage my eye, and/or that I’ll become an eye numbing junkie. This dogma even exists in the RCE Facebook support group, where it is forbidden to even talk about whether or how one might obtain numbing drops. I’d like to post this article there, but I fear reprisal.

    I love the idea of a severely quantity limited, and even highly diluted, solution that would reduce my initial pain even by 20%. I can’t say how many times I’ve gone to my optometrist to have my RCE “checked out”, when all I really want is the relief of numbing drops, because god knows the various approaches at treatment have unreliable outcomes at best (this coming from someone whose RCE was just made 2x worse by a bandage contact lens — I’m typing this with one eye). Just let me suffer it out with at least a small reduction in pain.


Leave a Comment

Time limit is exhausted. Please reload CAPTCHA.