Spectral domain optical coherence tomography findings in Endogenous Candida Endophthalmitis and their clinical relevance

PURPOSE:To describe vitreal, retinal, and choroidal features of eyes affected by Endogenous candida endophthalmitis (ECE) analyzed by spectral domain optical coherence tomography (SD-OCT) and to evaluate their clinical impact. METHODS: Medical records and SD-OCT images from eyes diagnosed with ECE at four retina and uveitis tertiary referral centers were retrospectively evaluated. Spectral domain optical coherence tomography images were analyzed to evaluate the structural changes occurring in the vitreous, the retina, and the choroid in areas involved by ECE. Baseline and final best-corrected visual acuity were correlated with SD-OCT findings. RESULTS: Fifteen eyes from nine patients were enrolled. Vitreous involvement (vitreous cells, posterior hyaloid thickening) was detected in 13/15 eyes. Peculiar hyper-reflective preretinal aggregates obscuring the underlying retina because of a shadowing effect ("rain-cloud" sign) were noticed in all eyes with vitreous involvement. Two patterns of retinal and choroidal findings were identified: the first (six eyes) confined within the inner retinal layers (Intraretinal Pattern), the second (nine eyes) involving both retina (full thickness) and choriocapillaris (Chorioretinal Pattern). None of the eyes showed both patterns. None of the eyes, regardless the pattern, showed choroidal thickening. Baseline best-corrected visual acuity was not associated with OCT pattern (P = 0.09). On the contrary, final best-corrected visual acuity was significantly higher in patients showing Intraretinal Pattern (∼20/25, 0.06 ± 0.08 LogMAR) than in subjects with Chorioretinal Pattern (∼20/50, 0.44 ± 0.30 LogMAR) (P = 0.01). CONCLUSION: Endogenous candida endophthalmitis showed peculiar features on SD-OCT. Two distinct patterns of chorioretinal involvement influencing the final best-corrected visual acuity were identified. Spectral domain optical coherence tomography could be useful in the diagnosis, management, and outcome prediction in ECE.