The effects of triamcinolone crystals on retinal function in a model of isolated perfused vertebrate retina.

A good clinical experience of intravitreal triamcinolone acetonide (TA) has been reported in several studies, but there are growing indications that epiretinal crystals of TA exhibit retinal toxicity. To investigate the effects of TA on retinal function we used a model of an electrophysiological in vitro technique for testing retinal toxicity. Isolated bovine retinas were perfused with an oxygen saturated nutrient solution. The electroretinogram (ERG) was recorded as a transretinal potential using Ag/AgCl electrodes. After reaching stable ERG-amplitudes TA at the maximum solubility equilibrium (36μg/ml) was either applied to the nutrient solution for 45min or TA was administered epiretinally at concentrations (1mg/ml, 4mg/ml, 8mg/ml, 20mg/ml and 40mg/ml) above the maximum solubility equilibrium to assure direct contact of the TA crystals with the isolated perfused retinas. After that the retinas were reperfused for 75min with the standard nutrient solution. The percentage of a- and b-wave reduction directly after the application and at the washout was calculated. To assess the effects of TA at the level of the ganglion cell layer a Viability/Cytotoxicity Kit for mammalian cells was used. No changes of the ERG-amplitudes were detected during the exposure to 36μg/ml TA for 45min (b-wave: 9.6μV±2.1 vs. 8μV±2.1 (p=0.135); a-wave: −11μV±2.7 vs. −10.6μV±2.3 (p=0.889)) and at the washout (b-wave: 8μV±2.1 vs. 8.3μV±2.4 (p=0.18); a-wave: −10.6μV±2.3 vs. −12μV±2.6 (p=0.225)). At concentrations higher than 1mg/ml TA induced a decrease of the a- and b-wave in a concentration dependent manner. These changes were reversible for concentrations of TA up to 20mg/ml (b-wave: 9μV±2.4 vs. 6.6μV±2.5 (p=0.08); a-wave: −11.4μV±2.0 vs. −11.2μV±2.2 (p=0.37)), but irreversible at 40mg/ml even at the end of the washout (b-wave: 9.8μV±1.9 vs. 3μV±1.7 (p=0.009); a-wave: −9.8μV±2.1 vs. −2.6μV±2.1 (p=0.001)). Histological examination of the preparations revealed a dramatic ganglion cell death, in which an application of 20mg/ml and 40mg/ml TA led to a 60.53% (p=0.013) and 82.35% (p=0.002) ganglion cell death, respectively. The epiretinal application of 4mg/ml TA and higher resulted in distinct effects on the ERG of the isolated perfused retinas. Ganglion cell death was induced at a concentration of 20mg/ml and higher. TA shows an asymmetric and partly high concentrated distribution after intravitreal application. Therefore, we consider concentrations of 4mg/ml and higher might be toxic and should be avoided in clinical use.