AIM: To choose appropriate concentration of sodium hydroxide (NaOH) solution to establish a stable and consistent corneal alkali burn mouse model in grade II. METHODS: The mice (n=60) were randomly divided into four groups and 15 mice each group. Corneal alkali burns were induced by placing circle filter paper soaked with NaOH solutions on the right central cornea for 30s. The concentrations of NaOH solutions of groups A, B, C, and D were 0.1 mol/L, 0.15 mol/L , 0.2 mol/L, and 1.0 mol/L respectively. Then these corneas were irrigated with 20 mL physiological saline (0.9% NaCl). On day 7 postburn, slit lamp microscope was used to observe corneal opacity, corneal epithelial sodium fluorescein staining positive rate, incidence of corneal ulcer and corneal neovascularization, meanwhile pictures of the anterior eyes were taken. Cirrus spectral domain optical coherence tomography was used to scan cornea to observe corneal epithelial defect and corneal ulcer. RESULTS: Corneal opacity scores (
) were not significantly different between the group A and group B (P=0.097). Incidence of corneal ulcer in group B was significantly higher than that in group A (P=0.035). Incidence of corneal ulcer and perforation rate in group B was lower than that in group C. Group C and D had corneal neovascularization, and incidence of corneal neovascularization in group D was significantly higher than that in group C (P=0.000). CONCLUSION: Using 0.15 mol/L NaOH can establish grade II mouse model of corneal alkali burns.