- Email: [email protected]
Practice patterns in the interdisciplinary management of corneal abrasions
Practice patterns in the interdisciplinary management of corneal abrasions Michael Ross, MD, Jean Deschênes, MD, FRCSC ABSTRACT ● Objective: To characterize the treatment and follow-up patterns of corneal abrasions at an academic health centre. Methods: This is a retrospective review of 90 cases of corneal abrasions over a 2-year period at a tertiary care academic hospital, of which 75 were seen by the ophthalmology department. All consultations primarily for corneal abrasion, as determined by the emergency department physician, were included in the study. Information on treatment regimen, corneal findings by emergency and ophthalmology physicians, time between follow-ups, and final outcomes was collected. Results: Seventy-five patients were seen by ophthalmology a median of 1 day after the emergency room visit. Twenty-five of these patients did not arrive for their subsequent follow-up appointment. Twenty-two of the abrasions were healed by the time of the ophthalmology examination, 51 patients had unhealed corneal abrasions, and 2 had corneal ulcers. Management was changed in 29 of the patients by ophthalmology. The most common management changes were hypertonic saline ointment for prophylaxis or treatment of recurrent erosion syndrome, followed by bandage contact lenses for comfort. Conclusions: Corneal abrasions are a common condition, and practice patterns for follow-up vary widely. Although the vast majority of patients do very well and likely would heal on their own without ophthalmology referral, it seems reasonable that patients with corneal abrasions are assessed once by an ophthalmologist immediately or possibly up to 1–2 days after the initial emergency visit, depending on the individual patient circumstances.
Corneal abrasions refer to the injury of the outermost layer of the cornea, the epithelium. They are one of the most common reasons for presentation to an emergency department.1 Typically, corneal abrasions heal quickly and patients make a full recovery.2 The healing process may take 24–72 hours, and healing takes the pattern of migrating epithelial sheets developing over the circumference of the defect, which progress toward the centre.3 While typically a benign, if extremely painful, condition, the complications of corneal abrasions, such as infectious keratitis or recurrent erosion syndrome, may be severe. For that reason, patients with corneal abrasions are typically seen and treated by medical personnel, ranging from primary health care providers such as family physicians or emergency doctors to optometrists and ophthalmologists.4 However, there are no standardized practices as to which practitioners initially treat patients with corneal abrasions. There is also wide variation in treatment and follow-up patterns. This study is a multicentre retrospective review of all the patients who presented to the 2 hospitals comprising the McGill University Health Center in Montreal, Canada, for a 2-year period from July 1, 2014, to June 30, 2016. The aim of this study was to characterize the treatment and follow-up patterns at a major academic health system to help guide treatment and referral patterns in corneal abrasions.
At the McGill University Health Center hospitals, most patients presenting to the emergency department with corneal abrasions are referred to ophthalmology. They may be seen by any of 20 residents or a similar number of staff physicians, providing an opportunity to examine different practice patterns and better determine the utility for initial ophthalmologic consultation and/or further follow-up.
METHODS Approval was obtained from the McGill University Health Centre Research Ethics Board. All new ophthalmology consults entered into the electronic medical record system of the Montreal General Hospital and Royal Victoria Hospital between July 1, 2014, and June 30, 2016, were reviewed. All patient charts from this search were reviewed by a senior ophthalmology resident (postgraduate year 4) and a cornea, anterior segment, and uveitis specialist. All consultations primarily for corneal abrasion, as determined by the emergency department physician, were included in the study. Patients who had previously been operated for corneal transplant in the affected eye or had undergone glaucoma filtration surgery (a bleb) were excluded. In addition, patients referred for a foreign body or for status post–foreign body removal by an emergency doctor were not included in the study.
& 2017 Canadian Ophthalmological Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcjo.2017.03.016 ISSN 0008-4182/16 CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
1
Patterns of corneal abrasions in health care—Ross, Deschênes Demographic data were collected from the electronic medical record. Information on treatment regimen, corneal findings by emergency and ophthalmology physicians, time between follow-ups, and final outcomes was collected. In addition, for patients who did see ophthalmology, the chart was reviewed and a determination was made as to whether the involvement of the ophthalmologist changed management for the patient, and if so, what that change was and why.
RESULTS A total of 1844 consults were made to ophthalmology from the 2 adult emergency departments of the McGill University Health Center (Montreal General and Royal Victoria Hospitals). Ninety-four patients were referred for corneal abrasion as assessed by the attending emergency physician. Four patients were excluded according to predetermined exclusion criteria (patients with a corneal transplant or glaucoma filtering surgery), leaving a total of 90 patients included in the study. This represented 4.9% of all consults to the ophthalmology department at both hospitals over the 2-year period in question, third in frequency behind posterior vitreous detachment symptoms and blunt orbital or facial traumas. The median age of the patients was 38 years (range 18– 83 years, standard deviation [SD] 17.4 years). Patients included 50 males and 40 females. Of the total of 90 patients, 75 (83%) saw an ophthalmologist after being initially assessed by the emergency doctor. Fifteen patients were given consults and told to follow-up with ophthalmology but did not arrive for their appointments. The 75 patients examined by ophthalmology were seen a median of 1 day after the emergency room (ER) visit (range 0–10 days, SD 1.9 days). Twenty-two of the abrasions were already healed by the time of the examination (median 2 days, SD 2.6 days, range 1–10 days). Fifty-one patients were diagnosed as having corneal abrasions with epithelial defects (median 0 days, SD 0.9 days, range 0–4 days). Overall, 2 patients were initially diagnosed incorrectly by the emergency physician: 1 had bacterial keratitis status post–fingernail trauma to the eye, and the other had a marginal sterile infiltrate. Management was changed in 5 of the 22 (23%) healed abrasions: in 3, hypertonic saline ointment (Muro 128) was prescribed for prophylaxis against future recurrent erosion syndrome, and in 2 the antibiotic drops were stopped earlier than prescribed by the emergency physician
(Table 1). Management was changed in 24 of the 51 active abrasions (47%), with 7 patients being given bandage contact lenses for comfort (typically for large abrasions), 10 patients given Muro 128 for prophylaxis against or treatment for recurrent erosion syndrome, 6 patients had their antibiotic regimen changed or stopped earlier, and 1 patient was followed and treated for a persistent epithelial defect. Thirty-five out of the 75 patients had at least a second follow-up visit with ophthalmology (46%), ranging from 1 to 30 days after (median 3 days, SD 6.25 days). In all cases, the patients recovered completely. Only for patients with infiltrates (2) and persistent epithelial defect (1) and for patients who had a subsequent episode of recurrent erosion (2) was management changed on any appointment after the initial follow-up. Treatment was started in a majority of cases (58/90 [64%]) by primary care physicians. The majority of patients were started on moxifloxacin alone (57) or moxifloxacin plus erythromycin (10). Three patients were started on an antibiotic plus topical diclofenac, all by emergency physicians, with the topical diclofenac being subsequently discontinued by ophthalmology at the first visit.
DISCUSSION Corneal abrasions are typically viewed as a minor if painful condition by ophthalmologists, but in a small proportion of cases they can have serious consequences.2 The American Association of Family Physician has followup guidelines for its members, suggesting follow-up by the primary care physician 24 hours after the first visit and again in 3–4 days if still not healed. They also suggest referral to ophthalmologists be limited to situations with deep eye injuries, suspicion of recurrent erosion syndrome, suspicion of an infiltrate, a foreign body that cannot be removed, worsening symptoms or symptoms not improving daily, or nonresolution after 3 days.2 The Wills Eye Manual, a common reference for ophthalmologists, suggests administering antibiotics, including antipseudomonal coverage for contact lens wearers; avoiding steroids with epithelial defects; debriding loose or hanging epithelium; and considering NSAIDs for pain control. Follow-up is suggested the next day for large or central abrasions and contact lens wearers or in 2–5 days for other cases.5 Emergency physicians at certain institutions also have similar guidelines.4 In an urban emergency department, however, arranging daily follow-up is not always practical.
Table 1—Management changes by ophthalmology
Patients Median days after primary care visit Management changed by ophthalmology visit
2
CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
Healed at First Ophthalmology Visit
Not Healed at First Ophthalmology Visit
22 2 (range 1–10) 5 (23%)
51 0 (range 0–4) 24 (47%)
Patterns of corneal abrasions in health care—Ross, Deschênes At our academic institution, emergency doctors tend to refer all corneal abrasions. This is, however, practitioner dependent, and one limitation of this study is that the nature of our records review does not allow us to be able to identify patients with corneal abrasions who were not referred to ophthalmology. Based on our experience, though, most ER doctors will refer all of their abrasions to our department. A significant portion of patients who were seen by ophthalmology did have their management changed, which was unexpected to the authors. One of the most common management changes initiated by ophthalmologists in this study was hypertonic saline ointment (Muro) for treatment or prevention of recurrent erosion syndrome. A number of patients were determined to be experiencing or at risk of experiencing recurrent erosion syndrome, and although there has been no conclusive evidence for the best therapy for either treatment or prevention, with one study finding no difference in hypertonic saline versus paraffin ointment, many ophthalmologists believe that the best clinical practice is to prescribe hypertonic saline ointment.6 Also, patients with large epithelial defects were often given bandage contact lenses for symptomatic relief. This necessitates additional follow-up, but some patients do benefit from the pain relief. The time course for referral also varied, and some patients were seen the same day by ER and ophthalmology, whereas others were seen by ER, treated, and seen in follow-up by ophthalmology. Patients seen later than 24– 48 hours tended to have healed abrasions. In this study, 2 patients who were referred as corneal abrasions turned out to have infiltrates, one sterile and one infectious. The 40-year-old patient who was ultimately diagnosed with infectious keratitis had been referred by ER for “? corneal ulceration/abrasion.” He had been poked in the eye by his son 6 days previously, and his symptoms of pain, redness, tearing, and photophobia had been worsening since then. The prolonged course and nonresolution were not typical for an uncomplicated corneal abrasion. Although the ER physician did not detect the infiltrates or 2þ cells found at the initial ophthalmology examination, the ER physician’s consult note suggests that he was nonetheless concerned for a more serious diagnosis. Although guidelines for primary care practitioners already suggest referral if there is any doubt about an infiltrate, this reinforces the guidelines indicating that primary care practitioners should have a low threshold to refer if there is any doubt about the diagnosis. In addition, as evidenced in our study, the practice patterns for follow-up once seen by ophthalmology vary widely. Some patients are seen only once and told to come back only if there is a problem, whereas other physicians will follow until the epithelial defect has closed and sometimes even after that. The low complication rate of our study suggests that although initial ophthalmology referral may be warranted, once seen by ophthalmology—
Table 2—Antibiotic and NSAID treatment of corneal abrasions Time from ER Arrival to Ophthalmology Examination Moxifloxacin drops Tobramycin drops Erythromycin ointment Moxifloxacin drops plus erythromycin ointment Ofloxacin drops Artificial tears alone Fusidin ointment plus diclofenac topical Ciprofloxacin drops plus diclofenac topical
Patients 57 2 10 10 5 4 1 1
and if there are no complicating factors such as infection, recurrent erosion syndrome, or patient discomfort requiring a bandage contact lens—subsequent follow-up may not be necessary or left on a PRN basis. However, this applies more specifically to tertiary care centres and urban areas where ophthalmologists are available. In centres with less ready access to ophthalmologists, primary care physicians may consider referring cases only if symptoms warrant, or patients with persistent ulcers or whose abrasions fail to resolve. Studies could be done in the future to correlate patients’ symptoms with outcomes and could provide more guidance in determining follow-up. With regard to antibiotics, we found that the vast majority (86/90) of patients were prescribed antibiotics for the prophylaxis of infection (Table 2). There is limited evidence of the utility of this2; however, given the potential serious complications of bacterial keratitis, the standard of care involves the use of antibiotics at least until the epithelial defect has healed. At our institution, we showed a high predilection for the use of a fourthgeneration fluoroquinolone, prescribed by both emergency physicians and ophthalmologists, despite only a small minority of patients being contact lens wearers and thus requiring the antipseudomonas protection conferred by these medications. Our findings suggest that primary care physicians and ophthalmologists could be more cognizant of antibiotic stewardship and of cheaper and less potent antibiotics such as erythromycin or ofloxacin (Ocuflox, second-generation fluoroquinolone) should be considered. More potent broad-spectrum antibiotics such as fourthgeneration fluoroquinolones can be reserved for patients with risk factors for pseudomonas such as contact lens wearers.
DISCUSSION: PATCHING
AND
NSAIDS
Pressure patching of corneal abrasions was once thought to improve healing and pain control. However, multiple studies, including a meta-analysis, have found no benefit to pressure patching, and it is no longer recommended.2,7,8 Notably, none of the patients in our study were patched, suggesting ophthalmologists and emergency physicians alike no longer accept this practice. A limited number of patients (3 out of 90) were started on NSAIDs for pain control by the emergency physician. Although there are publications supporting NSAID use as CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
3
Patterns of corneal abrasions in health care—Ross, Deschênes a pain control measure,2,5,9 many of these studies focused primarily on pain control as an outcome and did not follow patients for long-term adverse events. Ophthalmology texts state that topical NSAIDs should be used with caution,10 and in our ophthalmology clinical practice, NSAIDs are not used because of concerns of adverse effects. Although these are rare, NSAIDs can be toxic to corneal epithelium; can lead to severe complications such as keratitis, corneal and scleral melting, and even corneal perforation; and may also have systemic side effects.11,12 Some patients may be at higher risk, such as those with corneal denervation, corneal epithelial defects, diabetes mellitus, rheumatoid arthritis, and past ocular surgeries.13 Ophthalmologists at our academic centre believe that patients should not be treated with NSAIDs. If pain control is an issue, patients can be referred to ophthalmology on a more urgent basis for consideration of bandage contact lens, a measure frequently used in our study. Table 1 and 2 In summary, our study finds that corneal abrasions are a very common condition for which patients are referred, and the practice patterns for follow-up vary widely. Although the vast majority of patients do very well and likely would heal on their own without ophthalmology referral, it seems reasonable that if ophthalmology services are available, patients with corneal abrasions are assessed at least once by an ophthalmologist, which could be immediately or possibly up to 1–2 days after the initial emergency visit, depending on the individual patient circumstances.
4. Thyagarajan S, Sharma V, Austin S, Lasoye T, Hunter P. An audit of corneal abrasion management following the introduction of local guidelines in an accident and emergency department. Emerg Med J. 2006;23:526-9. 5. Gerstenblith A, Rabinowitz M. Wills Eye Manual. Philadelphia: Lippincott Williams & Wilkins; 2012. 6. Watson S, Barker N. Interventions for recurrent corneal erosions. Cochrane Database Syst Rev 2007:4:CD001861. 7. Kaiser P. A comparison of pressure patching versus no patching for corneal abrasions due to trauma or foreign body removal. Ophthalmology. 1995;102:1936-42. 8. Flyn C, D’Amico F, Smith G. Should we patch corneal abrasions? A meta-analysis J Fam Pract. 1998;47:264. 9. Weaver C, Terrell K. Update: Do ophthalmic nonsteroidal antiinflammatory drugs reduce the pain associated with simple corneal abrasion without delaying healing? Ann Emerg Med. 2003;41: 134-40. 10. Cantor L, Rapuano C, Cioffi G. Basic and Clinical Science Course— External Disease and Cornea San Francisco, CA: American Academy of Ophthalmology; 2014. 11. Flach AJ. Corneal melts associated with topically applied nonsteroidal anti-inflammatory drugs. Trans Am Ophthalmol Soc. 2001;99:205-12. 12. Burling Phillips L. Topical NSAIDs: Best Practices for Safe Use. San Francisco, CA: American Academy of Ophthalmology; 2013. www. aao.org/eyenet/article/topical-nsaids-best-practices-safe-use?july-2013 accessed 24 Aug. 2016. 13. Gaynes B, Onyekwuluje A. Topical ophthalmic NSAIDs: a discussion with focus on nepafenac ophthalmic suspension. Clin Ophthalmol. 2008;2:355-68.
REFERENCES
From the Department of Ophthalmology, McGill University Health Center, McGill University, Montreal, Que.
1. Edwards R. Ophthalmic emergencies in a district general hospital emergency department. Br J Ophthalmol. 1987;71:938-42. 2. Wilson S, Last A. Management of corneal abrasions. Am Fam Physician. 2004;70:123-8. 3. Dua H, Forrester J. Clinical patterns of corneal epithelial wound healing. Am J Ophthalmol. 2015;104:481-9.
4
CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
Footnotes and Disclosure: The authors have no proprietary or commercial interest in any materials discussed in this article.
Originally received Dec. 10, 2016. Final revision Mar. 15, 2017. Accepted Mar. 22, 2017. Correspondence to Michael Ross, MD, McGill Academic Eye Centre, 5252 de Maissoneuve Ouest, Montreal, Que. H4A3S9; [email protected]
Corneal abrasions refer to the injury of the outermost layer of the cornea, the epithelium. They are one of the most common reasons for presentation to an emergency department.1 Typically, corneal abrasions heal quickly and patients make a full recovery.2 The healing process may take 24–72 hours, and healing takes the pattern of migrating epithelial sheets developing over the circumference of the defect, which progress toward the centre.3 While typically a benign, if extremely painful, condition, the complications of corneal abrasions, such as infectious keratitis or recurrent erosion syndrome, may be severe. For that reason, patients with corneal abrasions are typically seen and treated by medical personnel, ranging from primary health care providers such as family physicians or emergency doctors to optometrists and ophthalmologists.4 However, there are no standardized practices as to which practitioners initially treat patients with corneal abrasions. There is also wide variation in treatment and follow-up patterns. This study is a multicentre retrospective review of all the patients who presented to the 2 hospitals comprising the McGill University Health Center in Montreal, Canada, for a 2-year period from July 1, 2014, to June 30, 2016. The aim of this study was to characterize the treatment and follow-up patterns at a major academic health system to help guide treatment and referral patterns in corneal abrasions.
At the McGill University Health Center hospitals, most patients presenting to the emergency department with corneal abrasions are referred to ophthalmology. They may be seen by any of 20 residents or a similar number of staff physicians, providing an opportunity to examine different practice patterns and better determine the utility for initial ophthalmologic consultation and/or further follow-up.
METHODS Approval was obtained from the McGill University Health Centre Research Ethics Board. All new ophthalmology consults entered into the electronic medical record system of the Montreal General Hospital and Royal Victoria Hospital between July 1, 2014, and June 30, 2016, were reviewed. All patient charts from this search were reviewed by a senior ophthalmology resident (postgraduate year 4) and a cornea, anterior segment, and uveitis specialist. All consultations primarily for corneal abrasion, as determined by the emergency department physician, were included in the study. Patients who had previously been operated for corneal transplant in the affected eye or had undergone glaucoma filtration surgery (a bleb) were excluded. In addition, patients referred for a foreign body or for status post–foreign body removal by an emergency doctor were not included in the study.
& 2017 Canadian Ophthalmological Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcjo.2017.03.016 ISSN 0008-4182/16 CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
1
Patterns of corneal abrasions in health care—Ross, Deschênes Demographic data were collected from the electronic medical record. Information on treatment regimen, corneal findings by emergency and ophthalmology physicians, time between follow-ups, and final outcomes was collected. In addition, for patients who did see ophthalmology, the chart was reviewed and a determination was made as to whether the involvement of the ophthalmologist changed management for the patient, and if so, what that change was and why.
RESULTS A total of 1844 consults were made to ophthalmology from the 2 adult emergency departments of the McGill University Health Center (Montreal General and Royal Victoria Hospitals). Ninety-four patients were referred for corneal abrasion as assessed by the attending emergency physician. Four patients were excluded according to predetermined exclusion criteria (patients with a corneal transplant or glaucoma filtering surgery), leaving a total of 90 patients included in the study. This represented 4.9% of all consults to the ophthalmology department at both hospitals over the 2-year period in question, third in frequency behind posterior vitreous detachment symptoms and blunt orbital or facial traumas. The median age of the patients was 38 years (range 18– 83 years, standard deviation [SD] 17.4 years). Patients included 50 males and 40 females. Of the total of 90 patients, 75 (83%) saw an ophthalmologist after being initially assessed by the emergency doctor. Fifteen patients were given consults and told to follow-up with ophthalmology but did not arrive for their appointments. The 75 patients examined by ophthalmology were seen a median of 1 day after the emergency room (ER) visit (range 0–10 days, SD 1.9 days). Twenty-two of the abrasions were already healed by the time of the examination (median 2 days, SD 2.6 days, range 1–10 days). Fifty-one patients were diagnosed as having corneal abrasions with epithelial defects (median 0 days, SD 0.9 days, range 0–4 days). Overall, 2 patients were initially diagnosed incorrectly by the emergency physician: 1 had bacterial keratitis status post–fingernail trauma to the eye, and the other had a marginal sterile infiltrate. Management was changed in 5 of the 22 (23%) healed abrasions: in 3, hypertonic saline ointment (Muro 128) was prescribed for prophylaxis against future recurrent erosion syndrome, and in 2 the antibiotic drops were stopped earlier than prescribed by the emergency physician
(Table 1). Management was changed in 24 of the 51 active abrasions (47%), with 7 patients being given bandage contact lenses for comfort (typically for large abrasions), 10 patients given Muro 128 for prophylaxis against or treatment for recurrent erosion syndrome, 6 patients had their antibiotic regimen changed or stopped earlier, and 1 patient was followed and treated for a persistent epithelial defect. Thirty-five out of the 75 patients had at least a second follow-up visit with ophthalmology (46%), ranging from 1 to 30 days after (median 3 days, SD 6.25 days). In all cases, the patients recovered completely. Only for patients with infiltrates (2) and persistent epithelial defect (1) and for patients who had a subsequent episode of recurrent erosion (2) was management changed on any appointment after the initial follow-up. Treatment was started in a majority of cases (58/90 [64%]) by primary care physicians. The majority of patients were started on moxifloxacin alone (57) or moxifloxacin plus erythromycin (10). Three patients were started on an antibiotic plus topical diclofenac, all by emergency physicians, with the topical diclofenac being subsequently discontinued by ophthalmology at the first visit.
DISCUSSION Corneal abrasions are typically viewed as a minor if painful condition by ophthalmologists, but in a small proportion of cases they can have serious consequences.2 The American Association of Family Physician has followup guidelines for its members, suggesting follow-up by the primary care physician 24 hours after the first visit and again in 3–4 days if still not healed. They also suggest referral to ophthalmologists be limited to situations with deep eye injuries, suspicion of recurrent erosion syndrome, suspicion of an infiltrate, a foreign body that cannot be removed, worsening symptoms or symptoms not improving daily, or nonresolution after 3 days.2 The Wills Eye Manual, a common reference for ophthalmologists, suggests administering antibiotics, including antipseudomonal coverage for contact lens wearers; avoiding steroids with epithelial defects; debriding loose or hanging epithelium; and considering NSAIDs for pain control. Follow-up is suggested the next day for large or central abrasions and contact lens wearers or in 2–5 days for other cases.5 Emergency physicians at certain institutions also have similar guidelines.4 In an urban emergency department, however, arranging daily follow-up is not always practical.
Table 1—Management changes by ophthalmology
Patients Median days after primary care visit Management changed by ophthalmology visit
2
CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
Healed at First Ophthalmology Visit
Not Healed at First Ophthalmology Visit
22 2 (range 1–10) 5 (23%)
51 0 (range 0–4) 24 (47%)
Patterns of corneal abrasions in health care—Ross, Deschênes At our academic institution, emergency doctors tend to refer all corneal abrasions. This is, however, practitioner dependent, and one limitation of this study is that the nature of our records review does not allow us to be able to identify patients with corneal abrasions who were not referred to ophthalmology. Based on our experience, though, most ER doctors will refer all of their abrasions to our department. A significant portion of patients who were seen by ophthalmology did have their management changed, which was unexpected to the authors. One of the most common management changes initiated by ophthalmologists in this study was hypertonic saline ointment (Muro) for treatment or prevention of recurrent erosion syndrome. A number of patients were determined to be experiencing or at risk of experiencing recurrent erosion syndrome, and although there has been no conclusive evidence for the best therapy for either treatment or prevention, with one study finding no difference in hypertonic saline versus paraffin ointment, many ophthalmologists believe that the best clinical practice is to prescribe hypertonic saline ointment.6 Also, patients with large epithelial defects were often given bandage contact lenses for symptomatic relief. This necessitates additional follow-up, but some patients do benefit from the pain relief. The time course for referral also varied, and some patients were seen the same day by ER and ophthalmology, whereas others were seen by ER, treated, and seen in follow-up by ophthalmology. Patients seen later than 24– 48 hours tended to have healed abrasions. In this study, 2 patients who were referred as corneal abrasions turned out to have infiltrates, one sterile and one infectious. The 40-year-old patient who was ultimately diagnosed with infectious keratitis had been referred by ER for “? corneal ulceration/abrasion.” He had been poked in the eye by his son 6 days previously, and his symptoms of pain, redness, tearing, and photophobia had been worsening since then. The prolonged course and nonresolution were not typical for an uncomplicated corneal abrasion. Although the ER physician did not detect the infiltrates or 2þ cells found at the initial ophthalmology examination, the ER physician’s consult note suggests that he was nonetheless concerned for a more serious diagnosis. Although guidelines for primary care practitioners already suggest referral if there is any doubt about an infiltrate, this reinforces the guidelines indicating that primary care practitioners should have a low threshold to refer if there is any doubt about the diagnosis. In addition, as evidenced in our study, the practice patterns for follow-up once seen by ophthalmology vary widely. Some patients are seen only once and told to come back only if there is a problem, whereas other physicians will follow until the epithelial defect has closed and sometimes even after that. The low complication rate of our study suggests that although initial ophthalmology referral may be warranted, once seen by ophthalmology—
Table 2—Antibiotic and NSAID treatment of corneal abrasions Time from ER Arrival to Ophthalmology Examination Moxifloxacin drops Tobramycin drops Erythromycin ointment Moxifloxacin drops plus erythromycin ointment Ofloxacin drops Artificial tears alone Fusidin ointment plus diclofenac topical Ciprofloxacin drops plus diclofenac topical
Patients 57 2 10 10 5 4 1 1
and if there are no complicating factors such as infection, recurrent erosion syndrome, or patient discomfort requiring a bandage contact lens—subsequent follow-up may not be necessary or left on a PRN basis. However, this applies more specifically to tertiary care centres and urban areas where ophthalmologists are available. In centres with less ready access to ophthalmologists, primary care physicians may consider referring cases only if symptoms warrant, or patients with persistent ulcers or whose abrasions fail to resolve. Studies could be done in the future to correlate patients’ symptoms with outcomes and could provide more guidance in determining follow-up. With regard to antibiotics, we found that the vast majority (86/90) of patients were prescribed antibiotics for the prophylaxis of infection (Table 2). There is limited evidence of the utility of this2; however, given the potential serious complications of bacterial keratitis, the standard of care involves the use of antibiotics at least until the epithelial defect has healed. At our institution, we showed a high predilection for the use of a fourthgeneration fluoroquinolone, prescribed by both emergency physicians and ophthalmologists, despite only a small minority of patients being contact lens wearers and thus requiring the antipseudomonas protection conferred by these medications. Our findings suggest that primary care physicians and ophthalmologists could be more cognizant of antibiotic stewardship and of cheaper and less potent antibiotics such as erythromycin or ofloxacin (Ocuflox, second-generation fluoroquinolone) should be considered. More potent broad-spectrum antibiotics such as fourthgeneration fluoroquinolones can be reserved for patients with risk factors for pseudomonas such as contact lens wearers.
DISCUSSION: PATCHING
AND
NSAIDS
Pressure patching of corneal abrasions was once thought to improve healing and pain control. However, multiple studies, including a meta-analysis, have found no benefit to pressure patching, and it is no longer recommended.2,7,8 Notably, none of the patients in our study were patched, suggesting ophthalmologists and emergency physicians alike no longer accept this practice. A limited number of patients (3 out of 90) were started on NSAIDs for pain control by the emergency physician. Although there are publications supporting NSAID use as CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
3
Patterns of corneal abrasions in health care—Ross, Deschênes a pain control measure,2,5,9 many of these studies focused primarily on pain control as an outcome and did not follow patients for long-term adverse events. Ophthalmology texts state that topical NSAIDs should be used with caution,10 and in our ophthalmology clinical practice, NSAIDs are not used because of concerns of adverse effects. Although these are rare, NSAIDs can be toxic to corneal epithelium; can lead to severe complications such as keratitis, corneal and scleral melting, and even corneal perforation; and may also have systemic side effects.11,12 Some patients may be at higher risk, such as those with corneal denervation, corneal epithelial defects, diabetes mellitus, rheumatoid arthritis, and past ocular surgeries.13 Ophthalmologists at our academic centre believe that patients should not be treated with NSAIDs. If pain control is an issue, patients can be referred to ophthalmology on a more urgent basis for consideration of bandage contact lens, a measure frequently used in our study. Table 1 and 2 In summary, our study finds that corneal abrasions are a very common condition for which patients are referred, and the practice patterns for follow-up vary widely. Although the vast majority of patients do very well and likely would heal on their own without ophthalmology referral, it seems reasonable that if ophthalmology services are available, patients with corneal abrasions are assessed at least once by an ophthalmologist, which could be immediately or possibly up to 1–2 days after the initial emergency visit, depending on the individual patient circumstances.
4. Thyagarajan S, Sharma V, Austin S, Lasoye T, Hunter P. An audit of corneal abrasion management following the introduction of local guidelines in an accident and emergency department. Emerg Med J. 2006;23:526-9. 5. Gerstenblith A, Rabinowitz M. Wills Eye Manual. Philadelphia: Lippincott Williams & Wilkins; 2012. 6. Watson S, Barker N. Interventions for recurrent corneal erosions. Cochrane Database Syst Rev 2007:4:CD001861. 7. Kaiser P. A comparison of pressure patching versus no patching for corneal abrasions due to trauma or foreign body removal. Ophthalmology. 1995;102:1936-42. 8. Flyn C, D’Amico F, Smith G. Should we patch corneal abrasions? A meta-analysis J Fam Pract. 1998;47:264. 9. Weaver C, Terrell K. Update: Do ophthalmic nonsteroidal antiinflammatory drugs reduce the pain associated with simple corneal abrasion without delaying healing? Ann Emerg Med. 2003;41: 134-40. 10. Cantor L, Rapuano C, Cioffi G. Basic and Clinical Science Course— External Disease and Cornea San Francisco, CA: American Academy of Ophthalmology; 2014. 11. Flach AJ. Corneal melts associated with topically applied nonsteroidal anti-inflammatory drugs. Trans Am Ophthalmol Soc. 2001;99:205-12. 12. Burling Phillips L. Topical NSAIDs: Best Practices for Safe Use. San Francisco, CA: American Academy of Ophthalmology; 2013. www. aao.org/eyenet/article/topical-nsaids-best-practices-safe-use?july-2013 accessed 24 Aug. 2016. 13. Gaynes B, Onyekwuluje A. Topical ophthalmic NSAIDs: a discussion with focus on nepafenac ophthalmic suspension. Clin Ophthalmol. 2008;2:355-68.
REFERENCES
From the Department of Ophthalmology, McGill University Health Center, McGill University, Montreal, Que.
1. Edwards R. Ophthalmic emergencies in a district general hospital emergency department. Br J Ophthalmol. 1987;71:938-42. 2. Wilson S, Last A. Management of corneal abrasions. Am Fam Physician. 2004;70:123-8. 3. Dua H, Forrester J. Clinical patterns of corneal epithelial wound healing. Am J Ophthalmol. 2015;104:481-9.
4
CAN J OPHTHALMOL — VOL. ], NO. ], ] 2017
Footnotes and Disclosure: The authors have no proprietary or commercial interest in any materials discussed in this article.
Originally received Dec. 10, 2016. Final revision Mar. 15, 2017. Accepted Mar. 22, 2017. Correspondence to Michael Ross, MD, McGill Academic Eye Centre, 5252 de Maissoneuve Ouest, Montreal, Que. H4A3S9; [email protected]