Friend or foe – topical anesthetic in ophthalmology

Posted on by admin

CASE SERIES

Hippokratia 2024, 28(3): 120-125

Hristova R, Oscar A, Zdravkov Y
Department of Ophthalmology, Medical University Sofia, University Hospital Alexandrovska, Sofia, Bulgaria

Abstract

Background: Topical anesthesia is an invaluable tool in ophthalmic practice with vast diagnostic and therapeutic utilization – from Goldman tonometry to intraocular surgery. Topical anesthetic abuse (TAA) is a rare form of severe self-induced ocular disease, which can lead to life-long disability.

Methods: We conducted a prospective observational case series of fourteen eyes of eleven patients. Inclusion criteria were patients with self-prescribed topical anesthetic use, corneal infiltrates, epithelial defects, and ulcers. We obtained a detailed history and investigated with visual acuity, slit lamp examination, anterior segment optical coherence tomography densitometry, aesthesiometry, and specular microscopy. Patients were treated conservatively for seven days, including antibiotics, systemic corticosteroids, and autologous serum eye drops, followed by amniotic membrane transplantation (AMT), and four eyes underwent cataract surgery.

Results: All patients were male, with a mean age of 40.1 ± 7.5 years. Ten patients reported traumatic corneal injury and one painful exposure keratopathy as the reason for anesthetic use. Eleven eyes (78.57 %) achieved complete epithelialization with improvement in visual acuity and corneal transparency; two eyes had persistent epithelial defects, which necessitated a second AMT, and in one, a corneal abscess formed that necessitated evisceration. Visual acuity improved from 0.023 ± 0.009 to 0.16 ± 0.041 (p =0.003). Four eyes underwent successful cataract surgery. Optical coherence tomography densitometry values at presentation were, on average, 64.04 ± 10.45 standardized greyscale units (GSU) and changed to 61.28 ± 9.03 GSU (p =0.096). The mean pain score of the patients was 8.6 ± 0.9 upon admission and 2.36 ± 2.2 at the last follow-up visit (p =0.003). Aesthesiometry improved significantly from 192.14 ± 5.34 mg/S to 97.21 ± 10.86 mg/S (p =0.001). The mean endothelial cell density was 1409.18 ± 516.03 and changed significantly to 1385.73 ± 510.9 (p =0.05).

Conclusions: Topical anesthetic abuse should be kept in mind in the differential diagnosis of ring infiltrates and non-healing ulcers. Prevention of trauma and strict pharmaceutical regulations are key to avoiding this debilitating disease. A combined conservative and surgical approach, including autologous serum, amniotic membrane, and phacoemulsification, could result in visual rehabilitation in TAA patients. HIPPOKRATIA 2024, 28 (3):120-125.

Keywords: Anesthetic, topical, abuse, cornea, ulcer, cataract, hypopyon

Corresponding Author: Dr. Rozaliya Hristova, 72A Burel str., 1408, Sofia, Bulgaria, tel: +359887973514, e-mail: alleta@abv.bg

Introduction

Topical anesthesia is an invaluable tool in ophthalmic practice with vast diagnostic and therapeutic utilization – from Goldman tonometry to intraocular surgery. The most common anesthetics currently in use are tetracaine, propacaine, and oxybuprocaine, compounds that block sodium channels inside nerve cells, ultimately preventing the generation of action potential and, therefore, neural conduction. It has been proved that anesthetics inhibit epithelial cell proliferation and migration1,2. Adding preservatives in the combination of topical antibiotics and anesthetics has additional detrimental effects on corneal wound healing. Moreover, given the essential role of corneal innervation and sensitivity for ocular surface homeostasis, neural conduction block due to topical anesthetic abuse can easily lead to severe ocular surface disease and possibly lifelong disability.

Topical anesthetic abuse (TAA) is a form of self-induced ocular disease. It can be related to addictive behavior, but usually, it is a consequence of other factors, including iatrogenic prescription. Treatment includes cessation of topical anesthetic, lubricants, and, in advanced cases, amniotic membrane transplantation (AMT) and keratoplasty3.

This study aims to describe the causes, clinical features, and treatment outcomes of TAA in a tertiary referral center.

Materials and methods

We conducted a prospective observational case series from January 2018 to January 2024. The study was approved by the Ethical Committee of the University Hospital Alexandrovska, Medical University of Sofia (decision No 2280, date: 28/06/2023), and all patients signed an informed consent before participating. We perform all procedures involving human participants according to the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

In the study, we included fourteen eyes of eleven patients and collected data from detailed history and comprehensive ocular exams. Inclusion criteria were patients with self-reported topical anesthetic use, corneal infiltrates, epithelial defects, and ulcers. We excluded patients with a history of hereditary corneal disease in the study or fellow eye or family members, patients with active infectious corneal ulcers, patients with a history of herpetic ocular infections, and any conditions that could lead to non-healing corneal ulcers, irrespective of topical anesthetic abuse.

We evaluated visual acuity using the Snellen chart, and for statistical analysis, we converted it to the decimal system. Using the results described by Schulze-Bonsel et al4, we quantified data of patients with very low vision. Corneal scrapes and conjunctival swabs, including smears for Gram staining, were obtained for microbiological and molecular tests.

Anterior segment optical coherence tomography (Maestro2, Topcon, Japan) was used to obtain radial scans of the diseased area. Those were analyzed using ImageJ software to determine the mean grayscale of all pixels comprising the corneal tissue before and after treatment. The method has been previously described5 and resembles Scheimpflug densitometry. We used the Cochet-Bonnet aesthesiometry to track the corneal sensitivity changes, and the CellCheck 20&20 PLUS (Konan Medical, Irvine, CA, USA) specular microscope to measure the endothelial cell count.

We obtained from medical records data concerning patients’ demographic details (age, gender, occupation, source of obtaining the drug, length, and frequency of its use), clinical findings, initial best corrected visual acuity (BCVA), time from treatment initiation to the completion of epithelization, and post-treatment BCVA. We requested patients to rate their pain intensity from zero (no pain) to ten (worst possible pain) upon admission and six months post-treatment.

The therapeutic strategy included conservative measures as well as surgical reconstruction of the ocular surface. The former included fortified topical antibiotics (Vancomycin 5 %), topical dexpanthenol, topical cyanocobalamin, cycloplegic (Cyclopentolate 1 %), autologous serum eye drops (ASEDs), Doxycycline 100 mg once daily for four weeks and systemic corticosteroids (Methylprednisolone 1 mg/kg).

The surgical approach incorporated AMT, and the technique involved scraping off loose epithelium, 360-degree conjunctival peritomy, and either a single AM layer epithelial side up or multilayer inlay/onlay technique sutured with interrupted 10/0 Nylon for corneal sutures and continuous 8/0 Vicryl for conjunctival sutures.

Once the ocular surface was stable, we performed cataract surgery in cases with adequate corneal transparency. The technique included four corneal paracenteses, a 2.7 mm limbal incision, and filling of the anterior chamber with cohesive viscoelastic, synechiolysis, and pupil expanding device (iris hooks). We performed standard capsulorhexis, hydrodistention, and phacoemulsification and implanted a mono-focal three-piece IOL (HOYA i-sert PY60AD) in the capsular bag. We injected intracamerally a prophylactic antibiotic (cefuroxime 0.01 %), and the surgical wounds were hydrated, and no leakage was observed.

At the end of all surgeries, we injected dexamethasone subconjunctivally, and the postoperative regimen included a topical preservative-free antibiotic (Moxifloxacin) four times daily for a week and steroid drops (Dexamethasone) four times daily for four weeks in the operated eye, and continuous lubrication.

The primary outcome measures were postoperative BCVA at the last follow-up (including four eyes following cataract surgery), improved corneal transparency, and complete epithelialization of the cornea.

We performed all statistical analyses using the SPSS Statistics for Windows, Version 26.0 (IBM Corp., Armonk, NY, USA).

Results

Demographic data

All patients were male, with a mean age of 40.1 ± 7.5 (range: 28-54) years. Ten patients reported traumatic corneal injury and one painful exposure keratopathy secondary to thyroid-associated orbitopathy as the reason for anesthetic use. Among patients with trauma, the causative event happened in the setting of automobile repair (n =7), agricultural (n =2), or industrial (n =1) activity. The trauma reported most often was corneal metallic foreign body with subsequent removal (seven patients, eight eyes), followed by ultraviolet (UV) ocular damage (two patients, four eyes) and chemical burn (one patient, one eye). Patients were referred with the diagnosis of infectious keratitis (n =9), one with photokeratitis, and one with chemical burns. The mean time from onset to presentation to the referral center was 28 ± 15.5 days.

All patients reported using Alcaine (0.5 % proparacaine hydrochloride, Alcon, Inc, Switzerland) for various intervals, ranging from once every 15 min to once every hour for a mean of 21.7 ± 12.8 (range: 7-50) days. They obtained topical anesthetic as an over-the-counter drug at pharmacies without a doctor’s prescription. Five patients reported that they had previously used the drug repeatedly for corneal foreign bodies, three reported alcohol abuse, and one reported regular cocaine use.

Symptoms

The most common symptoms were visual loss, severe pain, photophobia, blepharospasm, redness, and tearing. Clinical characteristics are summarized in Table 1. The right eye was affected in 42.86 % of all eyes, and the condition was bilateral in three out of 11 patients.

Treatment

Upon hospitalization, patients stopped all previous medication, including Alcaine. Due to the referring diagnosis from primary physicians, conservative measures, as described in the Methods section, were initiated. Since complete epithelialization was not achieved with persistent epithelial defects, a tendency for stromal melting, and severe unresponsive pain on day seven of the treatment, we proceeded with AMT. We used the multilayer technique in five patients (seven out of 14 eyes; 50 %).

The patients were followed up for an average of 11.2 ± 11.6 (2-36) months. Following treatment, 11 eyes (78.57 %) achieved complete epithelialization with improvement in visual acuity and corneal transparency; two eyes had persistent epithelial defects that necessitated a second AMT, and in the patient with reported cocaine use, there was a relapse of non-healing ulcer with formation of a corneal abscess that necessitated evisceration (Figure 1). The mean time to complete epithelization after AMT was 18.64 ± 10.65 (rage: 6-35) days.

Figure 1: Clinical image showing severe corneal infiltration (corneal abscess) with perforation in a patient with known cocaine use and topical anesthetic abuse.

Visual acuity

Visual acuity at presentation ranged from 0.001 to 0.1 (mean: 0.023 ± 0.009) and improved in 12 eyes to a range of 0 to 0.4 (0.16 ± 0.041 on average). The improvement was statistically significant (p =0.003). Three patients (four eyes) underwent successful cataract surgery with significant visual acuity improvement (Figure 2).

Figure 2: Clinical image of the left eye of a patient following successful cataract surgery. A mild corneal haze is seen in the lower half where the previous epithelial defect used to be.

Microbiology

Microbiological cultures from corneal scrapes were negative in nine patients. Samples from two eyes demonstrated Staphylococcus epidermidis, interpreted as part of the normal bacterial flora on the ocular surface6,7.

Imaging

Optical coherence tomography densitometry values at presentation were, on average, 64.04 ± 10.45 standardized greyscale units (GSU). Densitometry values changed to 61.28 ± 9.03 GSU (p =0.096), forming a demarcation line similar to that observed in corneal crosslinking in five eyes. (Figure 3). Endothelial cell density was measured in 11 out of 14 eyes. The mean endothelial cell density (ECD) was 1409.18 ± 516.03 and changed significantly to 1385.73 ± 510.9 (p =0.05). Results are summarized in Table 2. 

Figure 3: Optical coherence tomography densitometry showing demarcation line in the cornea after treatment.

Pain and aesthesiometry

Upon admission, the mean pain score of the patients was 8.6 ± 0.9 (range 8-10) and 2.36 ± 2.2 (range 0-7) at the last follow-up visit (p =0.003). Aesthesiometry revealed severe dysfunction with a mean score of 192.14 ± 5.34 mg/S before treatment that improved to a mean of 97.21 ± 10.86 mg/S with five patients reporting a transient increase in pain during the early postoperative period (p =0.001).

Discussion

TAA is a severe form of self-inflicted ocular surface disease. It is rare in countries with strict regulations. However, factors that influence its occurrence include easier access through unregulated pharmacies, prescription policies of emergency doctors and ophthalmologists, and general socio-economic circumstances, mainly the ease of access to specialized medical help. A study among Iranian welders revealed a staggering 80.5 % unregulated use of topical anesthetic8. Most of the patients started using topical anesthesia following a traumatic event, which should bring attention to adhering to safety rules and always using protective goggles; however, this is also subject to regulation. Strict control of protocol adherence would minimize the risk of traumatic events in general and, therefore, prevent topical anesthetic abuse9.

In the setting of an Emergency Medicine Department Shipman et al concluded that “for acute corneal abrasions, short-term topical tetracaine is an efficacious analgesic, which compared to placebo is associated with less hydrocodone use, and found to be safe among 111 randomized patients to either tetracaine or placebo”. A retrospective cohort study regarding 444 emergency department patients administered tetracaine for 24 hours reported no severe complications or uncommon adverse events10. The authors of the two studies argue that short-term use of topical anesthesia is safe and could mitigate the risks of opioid analgesia. However, a strong argument against even short-term topical anesthetic use is the lack of control and the consideration of the patients’ addictive tendencies, i.e., if a patient has any addiction, topical anesthetic abuse could lead to debilitating morbidity as serious as opioid addiction.

An effective and far less risky approach to ocular surface trauma and pain includes lubrication, cycloplegia, and patching11-13. Although patching is no longer recommended by some, especially in cases of contact lens wear, it is still included in the treatment of persistent non-healing abrasions/ulcers, which merits further investigation with a randomized study on the effectiveness of these two strategies.

As in previous reports, all patients included in our study were male14. Perhaps due to its rarity, TAA could be a diagnostic challenge, unless there is self-reported history of anesthetic use. Previous studies have found that patients with psychiatric disorders were more likely to develop TAA15. In our study, we did not complete a psychiatric evaluation. Nevertheless, four patients did report addictive behavior.

Regarding clinical presentation, our findings support previous data. Clinical signs included conjunctival injection, persistent epithelial defects, non-healing ulcers, stromal edema and infiltrates (classic ring infiltrates), Descemet membrane folds, endothelial damage, and ocular inflammation16.

Even when the classical clinical signs and history point to TAA, microbiological and molecular tests are still warranted to exclude Acanthamoeba keratitis, Herpetic ocular infection, and bacterial/fungal keratitis as part of the differentialdiagnosis17, moreover, the presence of a non-healing epithelial defect/ulcer could lead to secondary infection which should be recognized and treated promptly and adequately.

We observed hypopyon as a frequent complication that could be explained by the theory that topical anesthetics induce antigen-antibody complex formation18. Ring infiltrates are observed when antibodies from the limbal vasculature react with foreign antigens, which triggers both the classical and alternative pathways of complement activation. Adding Benzalkonium chloride, a preservative in Alcaine, could worsen these processes. Benzalkonium chloride19 is known to have direct cell toxicity, damaging organelles, and cytoplasmic membranes and breaking down the epithelial barriers, facilitating deeper penetration of the active ingredient. Therefore, benzalkonium chloride could lead to lower ECD. The statistically significant loss of endothelial cells despite successful treatment of the corneal ulcers could be further explained by immune processes in the anterior segment.

Cataract formation could be induced or worsened by severe anterior segment inflammation, which could explain the need for cataract surgery in our relatively younger patients16. Our results indicate visual rehabilitation through cataract surgery once the ocular surface is stable enough. Steps could be taken to optimize surgical outcomes, i.e., avoiding corneal paracentesis in scarred regions with neovascularization, using pupil-expanding devices, choosing a three-piece lens suitable for implanting in the sulcus in case of capsular instability, etc. We consider the relatively recent cataract formation as a contributing factor to the success of the surgery, as the nucleus removal did not require vast amounts of ultrasound energy, which would further compromise the endothelium. Although the surgery was technically challenging, we achieved significant improvement in visual acuity. When the corneal scarring is too advanced, combined procedures could be performed, e.g., deep anterior lamellar keratoplasty (DALK) or penetrating keratoplasty (PK), and with cataract surgery.

The therapeutic approach in cases of TAA is to stop all topical medications; however, given the circumstances (trauma), conservative treatment should include a preservative-free topical antibiotic. Corticosteroids could prevent sympathetic ophthalmia and reduce anterior chamber reaction and ocular surface inflammation. However, topical preparations are not appropriate as they could lead to further stromal melting and slow down epithelialization20. Therefore, we chose to administer systemic intravenous corticosteroids.

Doxycycline is an antibiotic of the tetracycline class known to reduce inflammation and inhibit proteolytic enzymes. Doxycycline has numerous applications in ocular disease, from Meibomian gland dysfunction to infectious posterior uveitis21,22. We decided to include Doxycycline in our treatment strategy due to its inhibitory effects on matrix metalloproteinases in the cornea, its anti-inflammatory effects, and its relatively broad spectrum of antimicrobial activity. Its use would additionally improve any Meibomian gland dysfunction our patients could have previously had. All these characteristics lead to optimizing the corneal wound healing process.

Another element in our treatment was autologous serum eye drops (ASEDs). Many studies have proved ASEDs’ role in ocular surface restoration, especially in non-healing corneal ulcers. It is believed that the most significant components in ASEDs are transformer growth fibroblast β -factor (TGF-β), vitamin A, fibronectin, epithelial growth factor (EGF), albumin, α-2 macroglobulin, hepatocyte growth factor, platelet-derived growth factor (PDGF-AB), and neuropeptides such as substance P and the insulin-like growth factor23,24.  Autologous serum eye drops have been shown to improve the ocular surface in patients using preserved glaucoma eye drops, the most common preservative being Benzalkonium chloride, shared by Alcaine. The neuropeptides in autologous serum could additionally improve corneal nerves’ function.

The exact mechanisms of ocular surface damage in TAA remain unclear. Nonetheless, it is most likely a combination of several factors. First, research data suggest a direct cytotoxic effect of topical anesthetics on the epithelium, which is dose- and time-dependent, meaning repeated use in regular doses has cumulative detrimental effects. Topical anesthetics reduce the number of microvilli on the epithelial cells, decrease keratocyte viability, and impair key cellular metabolism elements such as glycolysis or Na+/K+ pump25 leading to the observed corneal edema. Another aspect of TAA is the disruption of the normal blink reflex, loss of corneal sensation, and impaired tear film secretion. As part of the corneal stem cell niche, corneal nerves also play an important role in limbal stem cell homeostasis. Interference in this area leads to impaired healing due to partial limbal stem cell deficiency26.

The limitations of our study include the small sample size, albeit TAA is a relatively rare disease, as well as the lack of more extensive imaging, especially corneal confocal in vivo microscopy, which would confirm the cell changes as well as detect the presence of any fungal or protozoan infections (i.e. Acanthamoeba).

Topical anesthetic abuse should be kept in mind in the differential diagnosis of ring infiltrates and non-healing ulcers. Prevention of trauma and strict pharmaceutical regulations are key to avoiding this debilitating disease. A combined conservative and surgical approach, including autologous serum, amniotic membrane, and phacoemulsification, could result in visual rehabilitation in TAA patients.

Conflict of interest

All authors declare that they have no relevant financial or non-financial interests to disclose regarding the publication of this paper.

Acknowledgement

This work is supported under Project No 8352.24/11/2022, Medical University Sofia. Amniotic membrane transplants were provided by Tissue Bank Bioregeneration.

References

  1. Rosenwasser GO, Holland S, Pflugfelder SC, Lugo M, Heidemann DG, Culbertson WW, Kattan H. Topical anesthetic abuse. Ophthalmology. 1990; 97: 967-972.
  2. Zhang E, Gupta S, Olson E, Sinha PR, Hesemann NP, Fraunfelder FW, et al. Effects of Regular/Dilute Proparacaine Anesthetic Eye Drops in Combination with Ophthalmic Antibiotics on Corneal Wound Healing. J Ocul Pharmacol Ther. 2022; 38: 232-239.
  3. 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. Int J Ophthalmol. 2015; 8: 938-944.
  4. Schulze-Bonsel K, Feltgen N, Burau H, Hansen L, Bach M. Visual acuities “hand motion” and “counting fingers” can be quantified with the freiburg visual acuity test. Invest Ophthalmol Vis Sci. 2006, 47: 1236-1240.
  5. Hristova R, Yankova P, Markov G, Oscar A, Zdravkov Y. Effect of autologous serum after amniotic membrane transplantation for persistent corneal ulcers. Int J Ophthalmol. 2024; 17: 1639-1644.
  6. Matysiak A, Kabza M, Karolak JA, Jaworska MM, Rydzanicz M, Ploski R, et al. Characterization of Ocular Surface Microbial Profiles Revealed Discrepancies between Conjunctival and Corneal Microbiota. Pathogens. 2021; 10: 405.
  7. Ueta M, Iida T, Sakamoto M, Sotozono C, Takahashi J, Kojima K, et al. Polyclonality of Staphylococcus epidermidis residing on the healthy ocular surface. J Med Microbiol. 2007; 56: 77-82.
  8. Sharifi A, Sharifi H, Karamouzian M, Mokhtari M, Esmaeili HH, Nejad AS, et al. Topical ocular anesthetic abuse among Iranian welders: time for action. Middle East Afr J Ophthalmol. 2013; 20: 336-340.
  9. Wong DT, Hillier RJ. Serious ocular trauma: prevention is key. Can J Ophthalmol. 2012; 47: 329-330.
  10. Waldman N, Winrow B, Densie I, Gray A, McMaster S, Giddings G, 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. 2018; 71: 767-778.
  11. Al-Saleh GS, Alfawaz AM. Management of traumatic corneal abrasion by a sample of practicing ophthalmologists in Saudi Arabia. Saudi J Ophthalmol. 2018; 32: 105-109.
  12. Kaiser PK. A comparison of pressure patching versus no patching for corneal abrasions due to trauma or foreign body removal. Corneal Abrasion Patching Study Group. Ophthalmology. 1995; 102: 1936-1942.
  13. Lim CH, Turner A, Lim BX. Patching for corneal abrasion. Cochrane Database Syst Rev. 2016; 7: CD004764.
  14. Yagci A, Bozkurt B, Egrilmez S, Palamar M, Ozturk BT, Pekel H. Topical anesthetic abuse keratopathy: a commonly overlooked health care problem. Cornea. 2011; 30: 571-575.
  15. Wu LT, Gersing K, Burchett B, Woody GE, Blazer DG. Substance use disorders and comorbid Axis I and II psychiatric disorders among young psychiatric patients: findings from a large electronic health records database. J Psychiatr Res. 2011; 45: 1453-1462.
  16. Pharmakakis NM, Katsimpris JM, Melachrinou MP, Koliopoulos JX. Corneal complications following abuse of topical anesthetics. Eur J Ophthalmol. 2002; 12: 373-378.
  17. Chern KC, Meisler DM, Wilhelmus KR, Jones DB, Stern GA, Lowder CY. Corneal anesthetic abuse and Candida keratitis. Ophthalmology. 1996; 103: 37-40.
  18. Chen HT, Chen KH, Hsu WM. Toxic keratopathy associated with abuse of low-dose anesthetic: a case report. Cornea. 2004; 23: 527-529.
  19. Eleftheriadis H, Cheong M, Sandeman S, Syam PP, Brittain P, Klintworth GK, et al. Corneal toxicity secondary to inadvertent use of benzalkonium chloride preserved viscoelastic material in cataract surgery. Br J Ophthalmol. 2002; 86: 299-305.
  20. Sulewski M, Leslie L, Liu SH, Ifantides C, Cho K, Kuo IC. Topical ophthalmic anesthetics for corneal abrasions. Cochrane Database Syst Rev. 2023; 8: CD015091.
  21. Smith VA, Cook SD. Doxycycline-a role in ocular surface repair. Br J Ophthalmol. 2004; 88: 619-625.
  22. Holmes NE, Charles PGP. Safety and Efficacy Review of Doxycycline. Clin Med Ther. 2009; 1.
  23. Quinto GG, Campos M, Behrens A. Autologous serum for ocular surface diseases. Arq Bras Oftalmol. 2008; 71: 47-54.
  24. So HR, Park HYL, Chung SH, Kim HS, Byun YS. Effect of Autologous Serum Eyedrops on Ocular Surface Disease Caused by Preserved Glaucoma Eyedrops. J Clin Med. 2020; 9: 3904.
  25. Penna EP, Tabbara KF. Oxybuprocaine keratopathy: a preventable disease. Br J Ophthalmol. 1986; 70: 202-204.
  26. Rosenwasser GO. Complications of topical ocular anesthetics. Int Ophthalmol Clin. 1989; 29: 153-158.