Diffuse lamellar keratitis: Isolation of endotoxin and demonstration of the inflammatory potential in a rabbit laser in situ keratomileusis model

Purpose: To systematically examine sources of endotoxin contamination in eye centers as a potential cause of diffuse lamellar keratitis (DLK) and to demonstrate the inflammatory potential of endotoxin in a rabbit model of laser in situ keratomileusis (LASIK) surgery. Setting: University of Calgary, Calgary, Alberta, Canada. Methods: In this prospective study, all water sources that routinely come in contact with LASIK instruments, including sterilizer reservoirs, eyedrops, microkeratome blades, and cleaning solutions, were examined for endotoxins at 5 eye centers. Bacterial cultures were performed on water samples from 5 sterilizer reservoirs. A LASIK flap was created in 8 rabbit eyes using an Automated Corneal Shaper(R) microkeratome (Bausch & Comb). The flaps were reflected, and a dose of endotoxin at various concentrations was placed on the interface. After 1 minute, the flap was irrigated and repositioned. The rabbit eyes were examined daily with a slitlamp biomicroscope for 3 days for the development of DLK, which was classified on a scale from grade 1 to 4 (mild to severe). The rabbits were killed at the conclusion of the study, and the interfaces were stained to rule out infectious etiologies. Results: Endotoxin was detected in significant concentrations in tap water, filtered and distilled water, instrument washbasins, and sterilizer reservoirs at all 5 centers. The cultures of the water samples taken from the sterilizer reservoirs ranged from no growth to the presence of >100 colony-forming units of Flavobacterium and Pseudomonas aeruginosa. Endotoxins caused DLK-like interface inflammation in all eyes tested. Examination of stained scrapings showed no microorganisms in the interface of the rabbit eyes. Conclusion: Endotoxin contamination was detected in water sources that routinely come in contact with LASIK instruments. Endotoxins were capable of inducing interface inflammation in a rabbit model and may therefore be a significant factor in epidemic DLK. J Cataract Refract Surg 2001; 27:917-923 (C) 2001 ASCRS and ESCRS.