A Clinical Trial of Povidone-Iodine for the Treatment of Bacterial Corneal Ulcers

Thrasher Research Fund, Research to Prevent Blindness, University of California, Los Angeles, Kolokotrones Family Foundation, University of the Philippines, L.V. Prasad Eye Institute, Joseph Eye Hospital

Bacterial corneal ulcers are a leading cause of pediatric blindness in underdeveloped countries due to a lack of antibiotic availability and affordability, among other reasons. Povidone-iodine, an inexpensive and readily available broad-spectrum antimicrobial agent, may be an effective and affordable treatment for corneal ulcers, allowing preservation of sight for those afflicted with this disease.

Hypothesis to be tested: Infectious corneal ulceration causes corneal scarring, opacification, and blindness in hundreds of thousands of children annually. In developing countries, it is the number one cause of avoidable blindness in children. While in some cases, the cornea was healthy prior to ulceration, in others; xerophthalmia, trachoma, rubeola, trauma, or some other cause previously compromised the cornea. Even when the ulceration is associated with one of the above etiologies, it is usually the bacterial corneal ulcer that directly leads to blindness in the short-term. In many countries, few of these infections are properly treated with appropriate antimicrobial agents due to a number of factors, including cost, availability, and effectiveness. If a topical antimicrobial agent were readily available in developing countries which is also inexpensive and effective, many of these ulcers could be cured before scarring and blindness result. The best candidate for such a medication is povidone-iodine ophthalmic solution. It has already been proven the best agent to sterilize the ocular surface before and after eye surgery and to prevent conjunctivitis in newborn babies. It was also shown an effective treatment against bacterial conjunctivitis in children. Povidone-iodine ophthalmic solution can be prepared in a local hospital or clinic by mixing the powder with an ophthalmic saline solution bringing the cost of a 10-ml bottle to less than US $0.10. By contrast, in the Philippines, a 10 ml. bottle of Neosporin ophthalmic solution cost about $4.00 and it may take a few bottles to cure a bacterial corneal ulcer. For far less than a U.S. dollar, a corneal ulcer could be fully cured. Povidone-iodine has the widest antimicrobial spectrum of any known topical medication. It has been shown to be effective against all bacteria, fungi, and viruses. No true resistance to povidone-iodine has ever been demonstrated. Allergic reactions are rare. Thus, if any medication could serve the purpose for this study, it would be povidone-iodine. The primary research question is whether povidone-iodine ophthalmic solution is effective in healing bacterial corneal ulcers, which, if proven, would have a dramatic effect in reducing pediatric blindness. The number of subjects was determined from the primary outcome measure of non-inferiority to achieve a minimum of a one-day difference between the control and investigative groups, with povidone-iodine being compared to the control medication at each study site. The control medications are Neosporin ophthalmic solution in Manila and ciprofloxacin ophthalmic solution in the two combined Indian sites (Hyderabad and Tiruchirapalli) for a total of four study arms. The number of recruits was calculated to achieve 90% power for detecting non-inferiority using t-tests with a level of significance of 0.05 for comparing mean time to cure. Cure is defined as closure of the epithelial defect with no inflammatory signs other than minimal injection. Subjects provided written consent. As appropriate, child assent was also obtained. At each study site, subjects were allocated to receive either povidone-iodine or the control medication. If corneal ulcers were present bilaterally, one eye was randomized to receive povidone-iodine, while the other received the control medication standard for that center. All subjects were hospitalized for at least 7 days for careful monitoring and appropriate treatment. To assure compliance, all subjects had their medications administered by medical personnel. Treatment: The study medication is 1.25% povidone-iodine and the control medications are Neosporin ophthalmic solution in Manila, Philippines and ciprofloxacin 0.3% ophthalmic solution in Hyderabad and Tiruchirapalli, India. These medications are the standard of care at each respective study site. The only eye medication permitted beside povidone-iodine 1.25% ophthalmic solution and the control medications (Neosporin and ciprofloxacin) is atropine ophthalmic solution to reduce intraocular inflammation and prevent synechiae. The atropine solution is administered to the affected eye(s) twice a day. Subjects less than one year old receive a 0.25% solution, those aged one to three years receive 0.5%, and those older than 3 years receive the 1% solution. The dosing schedule of povidone-iodine 1.25% or control ophthalmic solutions is as follows: For the first three days, one drop of the medication is applied every hour. Day 4 and thereafter: hourly while awake. When asleep (sleep not to exceed 9 hours), drops are administered every three hours. All cases are treated with this intense drop therapy for a minimum of 5 days unless criteria for change in therapy are met (see below). At 5 days: If there is no deterioration in any factor (see below) and improvement in at least one factor (other than epithelial defect size) on 2 consecutive examinations, dosing frequency is decreased to every 2 hours while awake (about 8 times/day) for 2 days, then 4 times/day until discharge. If status remained unchanged at Day 5, intense drop therapy continues for 5 more days. If at 10 days the status is unchanged, the subject exits the study. All patients are examined daily and have their visual acuity and ocular examination documented. The daily eye examination findings are scored and recorded for the following six factors: Visual acuity Inflammatory signs Stromal infiltrate dimensions Epithelial defect dimensions MELT Tissue firmness, including keratic precipitate (KP),anterior chamber(AC)cellular reaction, AC fibrin and hypopyon status. The daily assessment of the ulcer status is classified as cured, improved, persistent, worsening or failure. Criteria for Change in Therapy: Within the first 48 hours, treatment is changed for deterioration in all factors or appearance of a descemetocele. After 48 hours of treatment, treatment is changed for deterioration in one or more of the following factors on 2 consecutive examinations: stromal infiltrate, melt, or two or more signs or symptoms of inflammation. Epithelial defect size is used only for assessing status; not for change in medication. The new treatment is left to the discretion of the ophthalmologist. Culture Technique: All subjects undergo microbiological analysis of the corneal ulcer upon intake into the study. A spatula is applied to the cornea of the affected eye(s) and blood agar and chocolate agar plates and brain heart infusion broth (BHI) are streaked for each patient. Cultures are repeated daily until two negative cultures are obtained. If the cultures are positive for any organism, that organism is considered the infecting agent. Each culture plate is incubated for at least seven days and the colony forming units are differentiated and enumerated by standard bacteriological techniques, including the VPI technique for anaerobic bacteria. No culture growth at seven days constitutes a negative culture.

Days to cure, with corneal epithelial defect closed, and no inflammatory signs other than minimal injection.

Safety of study medication use compared to control medications. No untoward reactions or loss of visual acuity (VA) compared with pre-treatment VA findings.

Inclusion Criteria: Clinical diagnosis of untreated bacterial corneal ulcer that began within 14 days of presentation to the study center. Corneal ulcer size is 2 mm-8 mm. in diameter, and does not extend to the limbus. Exclusion Criteria: Topical or systemic antimicrobial or immunosuppressant therapy within 14 days of presentation to the study center. Corneal ulcer smear and culture fail to show the presence of bacteria. Allergic history to povidone-iodine, iodine or any components of Neosporin. Dacrocystitis, neurotropic or exposure keratitis, keratitis sicca, positive HIV status, and legal blindness in the unaffected eye.

Isenberg SJ, Apt L, Yoshimori R, Pham C, Lam NK. Efficacy of topical povidone-iodine during the first week after ophthalmic surgery. Am J Ophthalmol. 1997 Jul;124(1):31-5. doi: 10.1016/s0002-9394(14)71640-x.

Berkelman RL, Holland BW, Anderson RL. Increased bactericidal activity of dilute preparations of povidone-iodine solutions. J Clin Microbiol. 1982 Apr;15(4):635-9. doi: 10.1128/jcm.15.4.635-639.1982.

Benevento WJ, Murray P, Reed CA, Pepose JS. The sensitivity of Neisseria gonorrhoeae, Chlamydia trachomatis, and herpes simplex type II to disinfection with povidone-iodine. Am J Ophthalmol. 1990 Mar 15;109(3):329-33. doi: 10.1016/s0002-9394(14)74560-x.

Holderman LV, Cato EP, Moore WEC. Anaerobe Laboratory Manual. Blacksburg, Virginia. VPI Anaerobic Laboratory, 1977

Everett SL, Kowalski RP, Karenchak LM, Landsittel D, Day R, Gordon YJ. An in vitro comparison of the susceptibilities of bacterial isolates from patients with conjunctivitis and blepharitis to newer and established topical antibiotics. Cornea. 1995 Jul;14(4):382-7. doi: 10.1097/00003226-199507000-00006.

HOGAN MJ, KIMURA SJ, THYGESON P. Signs and symptoms of uveitis. I. Anterior uveitis. Am J Ophthalmol. 1959 May;47(5 Pt 2):155-70. doi: 10.1016/s0002-9394(14)78239-x. No abstract available.