Perioperative Corneal Abrasions: Etiology, Prevention, and Management
By Jonathan Anson, M.D., Instructor in Anesthesia, Penn State University College of Medicine
How corneal abrasions occur
Confirming and treating corneal abrasion
Corneal abrasions are the most frequent ocular complications following general anesthesia, and are a painful burden to the recovering postoperative patient. The most recent (1992) ASA closed claim analysis showed that eye injury occurred in 3 percent of all claims in the database (1)
. Of these claims, 35 percent represented corneal abrasions with a 16 percent incidence of permanent eye injury (1)
Corneal abrasions can occur during general anesthesia, monitored anesthesia care, and regional anesthesia. Several strategies are widely used to try and prevent corneal abrasions, although there is a paucity of recent studies to support one method over another. This review will discuss common causes of peri-operative corneal abrasions and review the literature supporting various approaches to prevention. The basic management of this painful condition will also be discussed.
In order to understand the causes of perioperative corneal abrasions some pertinent anatomy must first be reviewed. The cornea is an avascular structure composed of five histologically distinct layers. It is protected by a precorneal tear film composed of three layers: Lipid, aqueous, and mucin. The outermost lipid layer prevents evaporation of the aqueous layer and acts as a lubricant. The aqueous layer oxygenates the corneal epithelium, while the main function of the mucin layer is to create a hydrophilic surface on the corneal epithelium. The precorneal tear film is regenerated by blinking, thus the absence of blinking during general anesthesia renders the cornea vulnerable to injury.
The cornea is extremely sensitive to hypoxia and the partial pressure of oxygen in the cornea can decrease dramatically with as little as 30 seconds of hypoxia (2)
. Corneal hypoxia leads to edema and potential for loss of the epithelial layer, causing an abrasion. Therefore, physiologic factors that alter corneal blood flow can predispose patients to corneal injury. This includes conditions that decrease arterial blood flow such as elevated intra-ocular pressure, head malpositioning, or pressure from an incorrectly applied face mask. Decreased venous return can similarly lead to corneal edema and subsequent abrasion.
How corneal abrasions occur
Up to 20 percent of post-operative corneal abrasions occur from direct trauma or chemical injury to the eye, but the majority is caused by lagophthalmos (failure of eyelid to close properly) (2)
. Normally lid closure while sleeping is maintained by the orbicularis muscles, but under anesthesia up to 59 percent of patients fail to have complete eye closure (2)
. The risks of lagopthalmos are increased by the abolishment of both blinking and Bell’s phenomenon (normal upward turning of the eye while asleep) during general anesthesia.
General anesthesia reduces tear production which leads to corneal drying (3)
. One small study demonstrated a significant decrease in basal tear production under general anesthesia. After one hour basal tear production decreased from a baseline of 13.6 ml per five minutes to just 0.9 ml per five minutes (P < 0.001) (3)
. Corneal drying in the presence of lagophthalmos can lead to dry patches and corneal abrasion. The effect of anesthetic duration on tear production and risk of corneal abrasion was further demonstrated in a study utilizing fluorescein and a slit lamp to monitor the corneas of patients undergoing general anesthesia. The study concluded that corneal changes are evident after 100 minutes of anesthesia and eye erosions after two hours (2)
. Other studies have similarly demonstrated that these changes peak around two hours and are not seen in anesthetics with less than a one hour duration.
While the majority of corneal abrasions are caused by lagophthalmos and changes in tear production induced by general anesthesia, traumatic injuries occur as well. These injuries are often caused by face-masks, dangling name badges, laryngoscopes, or “stethoscope necklaces” during airway management. Chemical injuries can occur from surgical prep cleaning solutions or application of benzoin without adequate eye protection. Post-operative eye injuries are most often attributable to patients rubbing their eyes, pulse oximeters, and bed linens. Applying the pulse oximeter probe to the 5th digit rather than the index finger may alleviate some of the corneal injuries that occur post-operatively.
A multitude of eye protection strategies have been proposed and tested over the years. In response to the evidence demonstrating decreased tear production and break down of tear films as a mechanism of corneal abrasion, a variety of ointments and lubricants have been tested over the last two decades. The majority of these studies occurred in the 1980s and 90s, and few recent studies were found during a literature review.
Most of the ointments studied were either paraffin based or methylcellulose based. They range in duration of action from 45 minutes to two hours, and thus need to be reapplied during long procedures. When compared head to head in several studies, there was little difference in effectiveness between paraffin based and methylcellulose based ointments (2)
. Parrafin based products can cause erythema and edema during halothane anesthetics as halothane is highly soluble in paraffin and high concentrations can result in inflammation, but this is not seen with newer volatile anesthetics.
Cucchiara, et al (4)
compared the effectiveness of eye ointment plus eye tape versus eye tape with no ointment. They looked at 4,652 neurosurgical patients, about half of which received ointment and eye tape while the rest had their eyes taped without ointment. Four patients in each group were found to have corneal abrasions post-operatively. Five of the eight patients were in the prone position. Thus, they were unable to show any protective effect of routine use of eye ointment in their neurosurgical patient population (4)
Attempts have also been made to identify risk factors that predispose patients to corneal abrasion. A group in China performed a retrospective analysis of 75,120 patients who underwent non-ocular surgery with general anesthesia. Corneal abrasions represented 59 percent of all their eye injuries (5)
. These were found to be more common in patients undergoing surgery in the prone or lateral position. Head and neck surgery, sustained hypotension, and anemia were also found to be risk factors (5)
. The conclusions made by this group were supported in another retrospective study examining 60,965 at the University of Chicago. They similarly reported lateral position and head/neck procedures as risk factors for corneal injury. In addition, this study found that increased length of surgery was an independent risk factor (6)
Confirming and treating corneal abrasion
Diagnosis of corneal abrasion is confirmed with a cobalt-blue filtered light and the application of flourescein (7)
. Indications for referral to a specialist include: history of significant trauma, worsening of symptoms despite treatment, erosion, infiltrate around the edges of the abrasion (suggestive of infection) (7)
. Corneal abrasions generally do not lead to long term complications, however, in rare instances the healed epithelium may be poorly adhered to underlying layers leading to recurrent corneal erosions.
Historically, eye patching has been utilized in the management of corneal abrasion. Recently, several studies have shown that patching is not helpful and may in fact delay healing (7
). Small abrasions often do not need treatment and patients should be reassured that they heal within 24-72 hours. Topical non-steroidal anti-inflammatory drugs such as diclofenac or keterolac can be considered as they have been shown to reduce pain. The use of antibiotics is more controversial. The incidence of infection following corneal abrasion is <1 percent, however, some clinicians use prophylactic antibiotics because a concomitant infection will slow healing. A prospective cohort study of prophylactic topical antibiotics (chloramphenicol) did demonstrate a decrease in ulcer formation when started within 18 hours of injury (9)
. Thus, antibiotic drops may be indicated to prevent ulcers. On the other hand, topical anesthetics should never be used as they can hinder healing, mask worsening symptoms, and lead to keratitis.
In summary, corneal abrasions are a concern for any anesthesia provider given the severity of patient discomfort and the potential for legal issues. Eye injuries account for 3 percent of the ASA closed claims database, however, the newest data is from 1992.
The etiology and pathophysiology of corneal abrasions in the perioperative period has been well defined. Risk factors have also been identified and verified in several large trials. Most notably, prone or lateral position and long procedure length are known risk factors.
Taping of the eyes is an easy, cost efficient measure to reduce eye injury from lagophthalmos. Prophylactic lubrication or ointment has not been shown to significantly reduce the incidence of corneal abrasions. Several studies have confirmed that longer procedures pose the greatest risk to patients, so practicioners may want to consider prophylactic lubrication in long cases, particularly in the prone or lateral position. If corneal abrasion is suspected, the diagnosis can be made using flourescein and a blue light. Treatment options include topical NSAIDS for analgesia, lubricants, and potentially antibiotic eye drops. Small corneal abrasions often do not need treatment and patients should be reassured these will usually heal within 24-72 hours.
1. Gild WM, Posner KL, Caplan RA, Cheney FW: Eye injuries associated with anesthesia. Anesthesiology 76:204-208, 1992
2. White E, Crosse M.M: The aetiology and prevention of peri-operative corneal abrasions. Anaesthesia 53:157-161, 1998.
3. Krupin T, Cross DA, Becker B. Decreased basal tear production associated with general anesthesia. Archives of Ophthalmology 1977; 95: 107-8
4. Cucchiara RF, Black S. Corneal abrasions during anesthesia and surgery. Anesthesiology 1988. 69: 978-979
5. Yu HD, Chou AH, Yang MW, Chang CJ. An analysis of peri-operative eye injuries after non-ocular surgery. Acta Anaesthesiol Taiwan. 2010 Sep;48(3):122-9.
6. Roth S, Thisted RA, Erickson JP, Black S, Schreider BD. Eye injuries after nonocular surgery. A study of 60.095 Anesthetics from 1988-1992. Anesthesiology 1996 Nov;85(5):1020-7.
7. Fraser, S. Corneal Abrasion. Clinical Ophthalmology. 2010 May 6;4:387-90.
8 .Wilson, SA, Last, A. Management of corneal abrasions. American Family Physician. 2004 Jul 1;70(1):123-8.
9. Upadhyay MP, Karmacharya PC, Koirala S, Shah DN, Shakya S, Shrestha JK et al. the Bhaktapur eye study: ocular trauma and antibiotic prophylaxis for the prevention of corneal ulceration in Nepal. British Journal of Ophthalmology. 2001:85:388-92.