Biocompatibility assessment of liquid artificial vitreous replacements: relevance of in vitro studies.

The biocompatibility of liquid artificial vitreous replacements is generally assessed by performing tests in animal models before their clinical use, whereas in vitro experimentation is seldom carried out due to their physico-chemical characteristics. Since their introduction in vitreoretinal surgery, however, the use of some certified vitreous replacements has been discouraged after clinical trials, because of the occurrence of serious side effects. This observation suggests that the tests currently performed for biocompatibility assessment cannot fully guarantee their safety when they are used in humans. Here we review the available literature on in vitro biocompatibility testing of liquid artificial vitreous replacements and survey our own experience on the subject, obtained by using primary retinal cell cultures, seeded on micro-porous inserts. We suggest that in vitro biocompatibility assessment, conducted before experiments in animal models, could improve the required safety evaluation and decrease the risk of undesired side effects, as well as providing a beneficial reduction of animal experimentation.

Biocompatibility assessment of silicone oil and perfluorocarbon liquids used in retinal reattachment surgery in rat retinal cultures.

The effects of silicone oil and perfluorocarbon liquids used in retinal reattachment surgery were studied in vitro using rat retinal cultures seeded on microporous inserts. These inserts allow the cell layer to be in contact with the material to be tested on the apical side and with the nutrient medium on the basal side. The materials tested were silicone oil, the perfluorocarbons perfluorophenanthrene and perfluoroctane, and hydroxypropylmethylcellulose. Perfluorophenanthrene, the heaviest of the compounds, induced a very precocious detachment of the cell layer. All the other tested biomaterials were compatible with cell survival and did not alter the structural organization of the retinal cultures, as revealed by scanning electron microscopy. By immunocytochemical techniques we evaluated the cell composition and the differentiation state of each of the cultures. In both control and treated samples, neuronal cells were well preserved. The expression of microtubule-associated protein 2, a marker of differentiated neuronal cytoskeleton, was not affected. Amacrine neurons, immunolabeled for gamma-aminobutyric acid, still were detectable after treatment. Synapses, marked by immunoreactivity for synapthophysin, were equally preserved. Vimentin-positive glial cells did not show modifications. The apoptotic rate, as determined by the terminal transferase-mediated dUTP-biotin nick end-labeling assay, was similar in treated and control samples. The results confirm that the use of biomaterials with a specific gravity close to intraocular fluids is compatible with retinal cell survival and differentiation in vitro.