Corneal injury

The most often reported eye complications following general anesthesia are corneal abrasion and corneal exposure, both of which are very painful and blur vision. An abrasion is caused by trauma, with complete loss of corneal epithelium, whereas corneal exposure is caused by damage to (but not loss of) the corneal epithelium due to exposure and secondary loss of tear film protection, which is necessary for protecting the integrity of corneal epithelium. Most injuries are thought to be secondary to lagophthalmos, an incomplete closure of the eyelid, with an abrasion being the result of direct trauma to the cornea from facemasks, surgical drapes, fingers, or other foreign objects that inadvertently contact the cornea.

The prevalence of corneal injuries varies depending on the methods used to detect them, but the incidence—as defined by clinical symptoms in patients whose eyes were taped closed during a surgical procedure—ranges from 0.05% to 0.15%. A specific cause of injury can be determined in only approximately 20% of the cases. Because the mechanisms of corneal injury are poorly understood, it is difficult to formulate preventive strategies that will completely eliminate the risk of injury. However, a review of the literature reveals several recurring themes. The pulse oximeter probe should be placed on patients’ fourth or fifth finger because patients are less likely to rub their eyes with these fingers. Patients’ eyes should be taped closed immediately after induction of anesthesia (do not wait until after intubation). Studies show that, if patients’ eyes are taped closed, no extra protection is achieved by using eye ointment unless the surgical procedure is prolonged or if patients are placed in other than the supine position. Even when patients’ eyes are taped shut, the anesthesia provider must use caution to prevent foreign objects from coming into contact with the patients’ eyes during intubation (e.g., stethoscope draped around the clinician’s neck, identification badge clipped to the chest pocket, a loose watchband or bracelet). The patient should be checked periodically, particularly following repositioning, to ensure that movement, moisture, or tears have not loosened or repositioned the tape.

In patients with proptosis, in patients undergoing head and neck procedures, for procedures performed with the patient prone or in the lateral position, and for procedures that are going to last for more than 90 min, the use of ointment on the eyes in addition to taping should be considered. Ointment provides good protection if there is concern that the eyelid tape may come off, but patients who have ointment on their eyes may have some blurred vision at the end of the procedure. The use of petroleum-based ointment is recommended for longer procedures but, because it is flammable, petroleum-based ointment should not be used if electrocautery or electrosurgery is used around the patient’s head and neck. Methylcellulose ointment is also an option; because it dissipates more quickly than petroleum-based ointment, it can be used for shorter procedures, and its use is associated with less blurring of vision postoperatively, as compared with petroleum-based products.

Corneal injury should be suspected postoperatively if the patient has pain, photophobia, blurred vision, or the sensation of having a foreign body in the eye. All but the most serious injuries can be managed conservatively. Ophthalmic antibiotic ointment should be administered to the eye, and the patient should be reassured that the injury will resolve within 24 to 48 h without permanent sequelae. An eye patch can be taped in place, but, because it is so often done incorrectly, the use of an eye patch is associated with additional problems.

Postoperative visual loss

Postoperative visual loss (POVL) is a rare but devastating complication seen most commonly following spine, cardiac, and head and neck surgical procedures (Figure 242-1). During the 1990s, the incidence of POVL seemed to be increasing, so in 1999 the ASA Committee on Professional Liability established the ASA POVL Registry to tabulate data on POVL following nonocular operations. Seven years later, a review of the registry identified 93 cases of POVL associated with spine operations; most were caused by ischemic optic neuropathy (ION)—either anterior or posterior—and not by compression of the globe. Only 10 of the patients had a central retinal artery occlusion, whereas the remainder had ION. Patients with ION, as compared with those without ION, were relatively healthy, were more likely to have an associated blood loss of 1000 mL or greater, or were more likely to have had an anesthetic duration of 6 h or longer; such conditions were found in 96% of the patients.