INTRODUCTION|: To investigate potentially adverse effects of different topical glaucoma medications and preservatives on cultured ocular epithelial cells. METHODS|: Confluent cultures of human corneal (10.014 pRSV-T) and conjunctival cells (1-5c-4) were assayed with 100 μL of different glaucoma medications for 25 minutes at 37°C and 5% CO₂. We also tested the preservative sofZia® (Alcon Laboratories, Fort Worth, TX, USA), as well as a range of concentrations of the preservative benzalkonium chloride (BAK; 0.001% to 0.050%). Balanced salt solution was used as the "live" control and a solution containing 70% methanol and 0.2% saponin was used as a "dead" control. The LIVE/DEAD viability/cytotoxicity kit (Invitrogen, Carlsbad, CA, USA) was used to determine the percentage of dead and live cells via ethidium homodimer and calcein fluorescence, respectively. RESULTS|: The toxicity of the prostaglandin analogs latanoprost, tafluprost and travoprost preserved with BAK was similar to the toxicity observed in their respective BAK concentrations. The prostaglandin analog travoprost (0.004%) preserved with the oxidizing preservative sofZia had much greater corneal and conjunctival cell survival than travoprost preserved with BAK. Travoprost (0.004%) containing polyquad also performed statistically better than its BAK-preserved formulation. CONCLUSION|: Ocular surface side effects have previously been demonstrated with chronic, long-term exposure to intraocular pressure-lowering medications containing the common preservative BAK. BAK alone has significant in-vitro cytotoxicity to cultured ocular epithelial cells. Substitution of BAK with polyquad or sofZia resulted in significantly higher percentages of live conjunctival and corneal cells. Further studies are needed to understand the- clinical implications of these findings.