AIM: To find a stable, inexpensive, and reliable method to produce a rat meibomian gland dysfunction (MGD) model. METHODS: We inserted slim guidewires into the meibomian gland orifices of twelve Brown Norway rats and fulgurized every guidewire to destroy part of the meibomian gland. We then observed the morphological changes in the eyelid margin, and compared the data of tear breakup time (TBUT), Schirmer I test, and the corneal fluorescence staining scores at different times (1, 2, 4, and 6wk). We observed pathological changes of the cornea, conjunctiva and meibomian gland, and we used real-time polymerase chain reaction to analyze epithelial growth factor (EGF), interleukin-6 (IL-6), IL-8, tumor necrosis factor-α (TNF-α), and Ki67. RESULTS: In the fourth week, compared with the control group, the TBUT of the model group began to decreased (P<0.05). The tear secretion remained stable (P>0.05). The corneal dots were significantly increased in the fourth week when the fusion stain began to appear (P<0.05). In the fourth week, partial meibomian gland openings had hoary secretions blocked, orifices were expanded, and there was a partial convex deformation. In the sixth week, the tissue section showed that the number of conjunctival goblet cells was decreased, epithelial cells were irregular, the epithelium was detached and rough, and meibomian glands were lost. The expressions of EGF, IL-6, IL-8, and TNF-α in corneal, conjunctival, and meibomian tissues were highly increased (P<0.05), but no statistical difference was found in the expression of Ki67 in corneal and conjunctival tissues (P>0.05). CONCLUSION: The MGD rat model, produced via electrocauterization of meibomian gland orifices, matched clinical manifestations and cytokine levels. Our research provides a new method of achieving an MGD animal model.