OCT angiography and retinal vein occlusions
Reviewed by Efrosini Papagiannuli

In this retrospective, observational case series, the authors set out to evaluate the morphology of the superficial and deep capillary plexa of eyes that suffered a retinal vein occlusion (RVO) on OCT angiography (OCTA) and compare the findings with fluorescein angiography (FA) and spectral domain optical coherence tomography (SD-OCT). The case notes were reviewed of 54 consecutive patients with unilateral RVO that had occurred between October 2014 and March 2015. Of these 29 had CRVO and 25 branch retinal vein occlusion (BRVO). Three patients with RVO had to be excluded due to poor quality of scan images. Excluded were eyes with diabetic retinopathy, previous retinal surgery, pathologic myopia or trauma. The parameters used to assess any capillary network abnormalities included disruption of the capillary network, capillary dilation and presence of shunting vessels, and well defined roundish black foci on OCTA (likely intraretinal cystoid spaces). The presence of any perifoveal capillary arcade disruption was also assessed in the superficial plexus (only visible here). All these OCTA findings were documented both for the superficial and deep capillary plexus and compared with FA and SD-OCT findings. The perifoveal capillary arcade was visible in 96% on OCTA and 83% on FA. It was disrupted in 92% on OCTA and 72% on FA (p=0.002). This disruption was correlated with peripheral retinal ischaemia (p=0.025). Intraretinal cystoid spaces were seen in 68% on FA, 76% on SD-OCT and 90% on OCTA. Retinal capillary network disruptions were seen in all patients in both superficial and deep plexa on OCTA. However, non-perfused (grayish) areas were more common in the deep plexus (84%) than the superficial (59%) (P<0.001). Best corrected visual acuity was correlated with the presence of macular ischaemia on FA (P=0.018), but not with any of the OCTA abnormalities. The authors point out that their study has for the first time highlighted the involvement of the deep capillary plexus in RVO and that the level of hypo-perfusion is more extensive here than in the superficial plexus. They argue that OCTA is a better imaging modality to capture the capillary plexa and arcade, as well as detecting retinal cystoid spaces. They speculate that the poorer visualisation of these structures on FA, may be due to the missing of early frames, focusing problems in the presence of extensive macular oedema, media opacities, and capillary dye leakage. Furthermore, FA is an invasive investigation with various potential side-effects. Whilst they advocate the use of OCTA as a non-invasive technology, which can easily be repeated, they are aware of its own limitations, such as the small scanning window (3x3 mm), possible artefacts that can hinder the view of the vascular bed, the softwares inability for accurate distinction of the capillary plexa (because of structural disorganisation after macular oedema) and the fact that OCTA can mask laser scars, because they are interpreted as no-flow areas by the software. Nevertheless the authors are in favour of the potential of this new imaging modality to enhance our management of RVOs, and at present are undertaking further studies with OCTA evaluating the macular oedema and perfusion in treated RVO.

Optical coherence tomography angiography in retinal vein occlusion: evaluation of superficial and deep capillary plexa.
Coscas F, Glacet-Bernard A, Miere A, et al.
AMERICAN JOURNAL OF OPHTHALMOLOGY
2016;161:160-71.