The dogma: Patients with corneal abrasions should not be discharged home with topical anesthetic agents because their use will lead to increased complications, including ulceration, opacification, and vision loss.
But we call this dogma for a reason: It's no longer supported by the literature.
It's clear to anyone who has suffered from a corneal abrasion that the condition is quite painful and that systemic analgesia is typically inadequate. We have an effective agent for analgesia — topical anesthetics like tetracaine and proparacaine — but its use is discouraged by major emergency medicine textbooks for fear that the abrasion becomes an ulceration and, in the extreme case, leads to vision loss.
Physiologically, there's an argument against the use of topical anesthetics. Studies involving cell preparations and animal studies on enucleated rat and rabbit eyes show inhibition of mitosis, cellular migration, and delayed wound healing. (Curr Eye Res 1988;7:947; J Ocul Pharmacol 1989;5:241; Br J Ophthalmol 1994;78:138; Ophthalmologica 2002;216:113.)
Fortunately, our patients don't have enucleated rat and rabbit eyes.
The argument against using these agents for analgesia persists, however, resting on two beliefs: Application of topical anesthetics decreases corneal sensitivity, so the patient may have a worsening abrasion without knowing it, and application causes direct toxicity, leading to increased corneal thickness, opacification, stromal infiltration, and epithelial defects. These are interesting hypotheses, but what about the actual evidence of harm in humans?
‘Evidence’ of Harm
The “evidence” of harm comes from case reports and case series, the two largest in 1968 and 1970. Epstein, et al., reported five cases of patients who developed keratitis after topical anesthetic use. (N Engl J Med 1968;279:396.) None of the patients in the report, however, had ocular exams prior to using the topical agents, and all of the patients used the agents for prolonged periods of time (five days to three months) with inappropriate frequencies (six minutes to an hour). Most of the patients also used incorrect concentrations (i.e., 5% tetracaine), and one patient even used veterinary-strength tetracaine!
Willis, et al., used a larger sample (n=9) but had similar problems. (Can J Ophthalmol 1970;5:239.) The patients all developed corneal complications after prolonged, inappropriately frequent topical anesthetic use with a single exception (one patient used proparacaine 0.5% for hours for three days). The results of the inappropriate application of a medication in 14 patients do not an evidentiary basis against its use make.
The arguments of delayed wound healing and loss of sensation have also been debunked by literature in ophthalmology journals on patients who received photorefractive keratectomy for vision correction. (J Refract Surg 2000;16:444; Ophthalmology 1997;104:1327; Eur J Ophthalmol 1997;7:327.) Multiple studies demonstrate no difference in surgical wound healing. Even though the corneal abrasions we see in the ED are not the same as surgically created lesions, it's reasonable to extrapolate these data to some degree.
It's fine to say that the data supporting an approach are poor, but we can't argue that the alternate approach is superior or safe without better evidence. Fortunately, two recently published studies lend some help.
Ball, et al., prospectively enrolled 33 patients and randomized them to receive dilute proparacaine (0.05%) or placebo drops. (CJEM 2010;12:389.) Patients were instructed to apply two to four drops as needed for up to seven days. All of the patients had follow-up visits on days one, three, and five after enrollment with an ophthalmologist who would assess the patients for complications. Patients were also given Tylenol with codeine for breakthrough pain and topical antibiotics. The researchers found that pain scores improved in the proparacaine group, and no patients in either group had any ocular complications. Though small, this study represented the best evidence to date on the topic and argued that topical anesthetics could be used in patients with corneal abrasions with close follow-up.
Another study by Waldman, et al., represents the largest, best-quality study to date addressing the use of topical anesthetics. (Acad Emerg Med 2014;21:374.) The study enrolled 116 patients in a prospective, double-blind, randomized trial of tetracaine versus saline. Participants were instructed to use the drops every 30 minutes as needed for 24 hours. The study excluded a number of groups, including those with contact lenses and those presenting more than 36 hours after injury.
Like the Ball study, Waldman, et al., did not find any difference in abrasion healing between the two groups. Surprisingly, there was not a significant difference in pain relief between the two groups, though the tetracaine group did rate their treatment as more effective. This study has weaknesses (loss to follow-up, unable to confirm use of drops), but it can help guide our decision on whether to use topical anesthetics.
The bottom line: The available data supporting the use of topical anesthetics in simple corneal abrasions are far superior to the case reports and case series that discourage it. It would be ideal to have larger multicenter studies, but we can use this information to help guide our practice. This evidence finds it reasonable to give patients with simple corneal abrasions and no contact lenses a 24-hour supply of dilute topical anesthetic (one-tenth of the standard concentration) and ophthalmology follow-up in one to two days.