Objective: To evaluate the efficacy of drug treatment and iontophoretic collagen crosslinking alone and in combination in the treatment for acanthamoeba keratitis in rabbits. Methods: Rabbit corneas were scraped with circular blades and 0.1 ml acanthamoeba suspension was injected into the central corneal stroma. The injection caused a-3.O-mm diameter white edematous area. A soft corneal contact lens was then applied. Forty-eight hours after acanthamoeba were injected into the comeal stroma, the rabbits were randomly divided into four groups: Group A, untreated (n=10). Group B, treated 7 times/day with eyedrops containing 0.02% chlorhexidine combined with 2% metronidazole (n=10). Group C, stromal crosslinking with 0.1% riboflavin penetration by 5 min of iontophoresis, followed with 370 nm ultraviolet irradiation(10 mW/cm2, 9-mm aperture diameter, 9 min, total energy 5.4 J) (n=10). Group D, drug+crosslinked treatment (n=10). The right eye was selected as the experimental eye, and the left eye was the control eye.After treatment for 90 days, slit-lamp microscopy, confocal microscopy, and corneal histopathology were performed. Confocal microscopy was used to select the images with lesion depths of 100, 200, and 300 lim.The density of the acanthamoeba in each image was measured by Image J software. The experimental data were analyzed by Kruskal-Wallis H tests and 2x2 analysis of variance analysis. Results: After treatment for
90 days, the number of comeal lesions was significantly different in four groups (X2=19.73 8, P<O.O01). The classification of corneal lesions in the Groups B and C was significantly lower than that in the Group A, and the difference was statistically signmcant (P<0.05). The large reduction in the number of corneal lesions in Group D showed that the drug treatment and crosslinking acted synergistically to control the infection.Confocal microscopy measurements of the acanthamoeba density were significantly lower in the Groups B, C and D compared with the Group A (P<0.05). The combination of drug and crosslinking treatment synergistically lowered the density of the parasites. Histopathological sections stained by hematoxylin and
eosin confirmed the reduced density of acanthamoeba in the three treated groups compared to the Group A (P<0.05). The combination of drug and crosslinking treatment synergistically lowered the density of the parasites. Conclusions: Iontophoretic comeal collagen crosslinking of acanthamoeba keratitis in rabbit eyes is effective. The combination of drug therapy and crosslinking is superior to local drug therapy alone or simple crosslinking alone.