A series of recent acute blindness cases following non–complicated retinal detachment surgery caused the release of several health alerts in Spain. The blindness was attributed to certain lots of perfluoro-octane (PFO; a volatile and transient medical device). Similar cases have been reported in other countries. This has raised questions regarding the validity of cytotoxicity test methods currently used to certify the safety of PFO lots. The tests were performed according to the International Organization for Standardization (ISO) norms, using the extract dilution method or the indirect contact method as applied to L929 cells, a line derived from mouse fibroblasts. The limitations of those methods have been resolved in this study by proposing a new cytotoxicity test method for volatile substances. The new method requires direct contact of the tested substance with cells that are similar to those exposed to the substance in the clinical setting. This approach includes a few new technical steps that are crucial for detecting cytotoxicity. Our new method detected toxic PFO lots that corresponded to the lots producing clinical blindness, which previous methods failed to detect. The study suggests applying this new method to avoid occurrence of such cases of blindness.

To describe a series of retinal acute toxicity cases with severe visual loss after intraocular use of a toxic perfluoro-octane (PFO). The clinical presentation is described, and the likely causes are analyzed. New biological methods for testing safety of intraocular medical devices are proposed. Information regarding a series of eyes suffering acute severe events after intraocular use of a toxic PFO was analyzed. Four types of spectroscopy, nuclear magnetic resonance, and chromatography were used to identify the potential PFO contaminants. Cultures of human retinal pigment epithelial cells (ARPE-19) and porcine neuroretina were used to quantify the toxicity of the suspect PFO lots. Of 117 cases of intraocular toxicity, 96 were considered clearly related to the use of PFO. Fifty-three cases had no light perception, and 97 had no measurable visual acuity. Retinal necrosis (n = 38) and vascular occlusion (n = 33) were the most characteristic findings. Two hydroxyl compounds, perfluorooctanoic acid and dodecafluoro-1-heptanol, and benzene derivatives were identified as the suspected toxic agents. While existing toxicity testing failed, we proposed new tests that demonstrated clear toxicity. Protocols to determine cytotoxicity of intraocular medical devices should be revised to assure safety. Acute toxic events should be reported to health authorities and scientific media.