PURPOSE: Evaluate toxicity of acai fruit (Euterpe oleracea) dye concentrations in a rabbit model. METHODS: Rabbits were injected intravitreously with 10%, 25%, and 35% acai dye concentrations. Control eyes received balanced salt solution (BSS). Electroretinogram (ERG), fundus imaging, fluorescein angiography (FA), optical coherence tomography (OCT), and light and transmission electron microscopy (LM/TEM) were performed. RESULTS: Fundus imaging showed increased vitreous opacity with increased dye concentrations. FA and OCT showed normality with all concentrations. Comparisons between BSS and dye concentrations were analyzed using Kruskal-Wallis and Mood's median test (p < 0.05). At 24 h, ERGs showed reduced amplitudes from baseline in all eyes. Median b-wave amplitudes nonsignificantly decreased and latency increased with 10% and 25%; findings were significant (p < 0.05) for 35%. LM and TEM showed no abnormalities for 10% and 25%. With 35%, TEM showed ganglion cell edema at 24 h that resolved after 7 days. Vacuolization, multilamellar bodies, and nerve bundle damage occurred at 24 h/7 days in the inner nuclear layer. Mitochondrial cristae disruption occurred in the inner photoreceptor segment at 24 h that decreased by 7 days. CONCLUSION: Ten and twenty-five percent concentrations were safe and may improve identification of the posterior hyaloid and internal limiting membrane during chromovitrectomy in humans.
PURPOSE: Evaluate toxicity of acai fruit (Euterpe oleracea) dye concentrations in a rabbit model. METHODS:Rabbits were injected intravitreously with 10%, 25%, and 35% acai dye concentrations. Control eyes received balanced salt solution (BSS). Electroretinogram (ERG), fundus imaging, fluorescein angiography (FA), optical coherence tomography (OCT), and light and transmission electron microscopy (LM/TEM) were performed. RESULTS: Fundus imaging showed increased vitreous opacity with increased dye concentrations. FA and OCT showed normality with all concentrations. Comparisons between BSS and dye concentrations were analyzed using Kruskal-Wallis and Mood's median test (p < 0.05). At 24 h, ERGs showed reduced amplitudes from baseline in all eyes. Median b-wave amplitudes nonsignificantly decreased and latency increased with 10% and 25%; findings were significant (p < 0.05) for 35%. LM and TEM showed no abnormalities for 10% and 25%. With 35%, TEM showed ganglion cell edema at 24 h that resolved after 7 days. Vacuolization, multilamellar bodies, and nerve bundle damage occurred at 24 h/7 days in the inner nuclear layer. Mitochondrial cristae disruption occurred in the inner photoreceptor segment at 24 h that decreased by 7 days. CONCLUSION: Ten and twenty-five percent concentrations were safe and may improve identification of the posterior hyaloid and internal limiting membrane during chromovitrectomy in humans.