Iatrogenic Corneal and Conjunctival Toxic Reaction From Hydrogen Peroxide Disinfection FREE

Goldmann applanation tonometer tips and contact fundus laser lenses can be disinfected using various means. Tonometer tips are commonly cleansed by rubbing briskly with a 70% isopropyl alcohol wipe. Alternative methods of disinfection include soaking in either 3% hydrogen peroxide, a 1:10 dilution of sodium hypochlorite (household bleach), or 70% isopropyl alcohol, followed by rinsing and thorough drying (see "Comment" section). Proper use of these solutions requires soaking for 5 to 10 minutes. At our institution, tonometer tips and fundus laser lenses are routinely soaked in hydrogen peroxide, rinsed with sterile water or an isotonic sodium chloride solution, then left to air dry. We report 2 cases of toxic reactions from exposure to instruments disinfected in a 3% hydrogen peroxide solution.

A 30-year-old woman with newly diagnosed proliferative diabetic retinopathy was scheduled for a planned session of panretinal photocoagulation. The patient was not receiving any ocular medications. Visual acuity of the involved eye was 20/20 OS. There was no history of diabetic keratopathy in either eye. One to 2 minutes after placing the fundus laser lens on the patient's eye, the physician noted a short-term onset of corneal edema with the presence of discrete, subepithelial bubbles measuring 0.1 to 0.3 mm in diameter and extending into the conjunctiva (Figure 1).There was a large epithelial defect on removal of the lens. The patient did not note significant pain. She was referred to the cornea service, where she was instructed to use frequent lubrication and to return the next day to her referring specialist. She returned to our clinic 3 weeks later with a visual acuity of 20/30 OS. She had moderate diffuse punctate epithelial erosions with faint subepithelial and anterior stromal corneal haze superior to the visual axis. Her epitheliopathy resolved over the next 6 weeks, and her visual acuity remained stable with persistent anterior corneal haze over the ensuing 3-month follow-up period.

A 32-year-old woman was seen in follow-up by the cornea service for chronic graft rejection. She had an ocular history of 3 penetrating keratoplasties in the left eye, first due to penetrating ocular trauma, then subsequently due to graft rejections. She also had a history of aphakic glaucoma. The patient was not receiving any ocular medications in the involved right eye. Visual acuity was 20/15 OD. Applanation tonometry was performed without the patient experiencing pain, first in the right eye, followed by the left eye. During applanation of the right eye, bubbles were observed within the area of the mires. Applanation progressed without incidence in the left eye. On ophthalmic examination by the staff physician 10 minutes later, the right eye exhibited moderate hyperemia along with few subepithelial corneal and conjunctival bubbles(Figure 2). The patient was instructed to use lubrication in the right eye and to call with worsening symptoms. Her symptoms worsened later that night despite using artificial tears every hour. She, therefore, was treated by her local optometrist who irrigated her right eye and started therapy with a 0.3% ciprofloxacin hydrochloride solution. Her symptoms resolved after another 24 hours, and findings from the examination with the local optometrist were normal at the 72-hour follow-up. On return to our clinic 2 weeks later, her visual acuity remained 20/15 OD; the results of the slitlamp examination were normal.

At the time of the first case, fundus laser lenses were disinfected at our institution by soaking in a hydrogen peroxide solution for at least 5 to 10 minutes, then rinsing with sterile water. The second case occurred after contact with a Goldmann applanation tonometer tip that had been soaked in hydrogen peroxide and rinsed with an isotonic sodium chloride solution, then left to air dry. In both cases, standard disinfecting technique was reported after questioning personnel. In the second case, the bulk hydrogen peroxide solution was analyzed (Institute for Rural and Environmental Health, Department of Preventive Medicine, Iowa City, Iowa), and no impurities were found. The applanation tip did not contain any residue or material that could be analyzed.

Corneal and conjunctival bubbles from hydrogen peroxide exposure have been described before.^1- 2 Pogrebniak and Sugar² reported a case of a reversible toxic reaction resulting from a tip that was soaked in hydrogen peroxide overnight, then dried, but not rinsed prior to use. They suggest that the act of drying overnight might lead to concentration of the solution to a toxic level. Levenson³ reported a case of permanent corneal scar after tonometry from a tip that was still moist from 3% hydrogen peroxide when applanation was performed. The development of gas bubbles results from the breakdown of hydrogen peroxide into water and oxygen. This reaction causes release of free oxygen radicals, which can cause damage to tissues. Hydrogen peroxide is directly toxic to corneal epithelial cells, decreasing cell proliferation and causing breaks in DNA.⁴ This damage has been demonstrated in vitro to cause death of corneal epithelial cells.⁵

Hydrogen peroxide solution is widely used for disinfecting tonometer tips and fundus laser lenses. The Centers for Disease Control and Prevention recommend the use of a soaking solution—either hydrogen peroxide, dilute sodium hypochlorite, or isopropyl alcohol—instead of wiping tonometer tips with alcohol.⁶ The American Academy of Ophthalmology guidelines allow for the use of a 5-second isopropyl alcohol wipe in place of the soaking solutions.⁶ The alcohol wipe, however, may not be adequate for complete removal of certain viruses, including hepatitis B and C viruses.^6- 7 The disadvantage of using soaking solutions is the damage that can be caused to the glue and the prism in the tonometer tip.⁶ Careful use of hydrogen peroxide should include rinsing and drying of the surface of the tonometer tip or fundus laser lens by the physician or technician just prior to use, rather than relying on adequate rinsing and drying to have been performed.

Corresponding author: John E. Sutphin, MD, Department of Ophthalmology and Visual Sciences, The University of Iowa Hospitals and Clinics, 200 Hawkins Dr, Iowa City, IA 52240 (e-mail: john-sutphin@uiowa.edu).