Sensitivity of fluorescein angiography alone or with SD-OCT for the diagnosis of myopic choroidal neovascularization

Abstract

Background: Myopic choroidal neovascularization (mCNV) has certain characteristics and features that distinguish it from choroidal neovascularization secondary to age-related macular degeneration. There may be angiographic diagnostic difficulties even when using the scanning laser ophthalmoscope, which gives more contrast and better definition than traditional angiography. The aim of the study is to compare the sensitivity of fluorescein angiography (FA) alone or combined with Spectral Domain Optical Coherence Tomography (SD-OCT) for assessing the incidence of mCNV.

Methods: In this retrospective study, two authors reviewed the charts and images of patients with recent (<30 days) vision deterioration, pathologic myopia, axial length >26 mm, documentation or suspicion of mCNV or macular exudative pathologies at FA and OCT. They only examined the images at first presentation obtained by the multi-modal imaging system that combines Infrared reflectance, FA, and SD-OCT, (Spectralis, Heidelberg Engineering, Germany). The images selected were then evaluated by three other investigators in blinded, independent conditions, in order to make their diagnosis, which was noted or rated as doubtful if it could not be decided on the basis of FA alone. SD-OCT images were then shown and compared to IR and FA by each of the three investigators individually to formulate a conclusive diagnosis.

Results: A total of 71 eyes of 69 patients were suitable for the study, mean age 65.97±14.57 years, spherical equivalent refraction -8.82 ± 2.51 diopters. Concordance between the three examiners' interpretations of FA features and FA-guided SD-OCT was 50/71 (70.4 %) and 67/71 (94 %) respectively. Total agreement on diagnosis between the three examiners was achieved in 55 % of cases for FA (κ = 0.53, p < 0.001), and 94 % for FA-guided SD-OCT (k = -0.01, p = 0.5). The final diagnosis with FA and FA-guided SD-OCT differed in 29 cases (40 %; 95 % C.I. 29-42 %), whereas 12 (17 %) mCNV were overlooked at FA, and in 11 (15 %) cases none of the examiners reached a diagnosis based on FA alone.

Conclusions: On the basis of FA alone, active mCNV can be misdiagnosed. The use of SD-OCT combined with FA should therefore be strongly considered.