Assessment of the eye at risk of neovascularisation

Abstract

The causal linkage between retinal ischaemia and ocular neovascularisation has been recog­ nised and widely accepted by the ophthalmic community since the early pioneering work of Michaelson/ Ashton,2 Patz,3 Wise,4 and many others. In the initial stages of this research, the presence of retinal vascular occlusion in eyes with neovascular retinopath­ ies was surmised primarily on the basis of his­ tological specimens of the retinal vessels obtained from autopsy eyes of patients with diabetic retinopathy. Trypsin digestion prep­ aration of the retinal vasculature showed focal areas of capillary acellularity, which were interpreted as being nonperfused capillaries in life. 5.6 Subsequently, the development of fluorescein angiography of the retinal vessels allowed assessment of their perfusion status in the living eye. In the case of diabetic ret­ inopathy, focal nonperfusion of the retinal capillary bed, as well as of arterioles in the more advanced stages of the disease, was demonstrated angiographically. In a few cases in which both in vivo fluorescein angiography and post mortem trypsin digestion prep­ arations were available, clinicopathological correlation studies showed a topographical correlation between capillary acellularity and angiographic nonperfusion. 7,8,9 Numerous clinical studies over the years support the notion that retinal ischaemia pro­ vokes ocular neovascularisation. Both cross­ sectional and longitudinal studies have demonstrated the following: (1) retinal vascular growth in diabetic retino­ pathy and in sickle cell retinopathy pref­ erentially occur at the border between perfused and nonperfused retinal zones,lO,11 (2) disc neovascularisation in diabetic retino­ pathy occurs mainly in eyes with exten­ sive mid-peripheral or generalised capillary nonperfusion,12 (3) iris neovascularisation in central retinal vein occlusion (CRVO) is found primarily in eyes with widespread retinal vascular nonperfusion. 13,14,15,16 In a recent large-scale longitudinal study of eyes with non proliferative diabetic retino­ pathy (NPDR), the extent of capillary non­ perfusion, arteriolar abnormalities, and fluorescein leakage present at the baseline visit, (as measured by a standardised angio. graphic grading system) were significant risk factors for the subsequent development of proliferative retinopathy (PDR). 17 The differ­ ence in rates of developing new vessels within one year between eyes with and without these abnormalities ranged from two to threefold, depending upon the subgroup of NPDR sev­ erity considered. On the basis of this study, it is clear that fluorescein angiography is a useful tool in assessing the risk for neovascularisa­ tion in eyes with NPDR. Besides determining the anatomic exten� of nonperfusion in eyes with vascular retina· pathy, evaluating the effects of ischaemia on various functions of the eye can also be useful. The major test, in this respect, has been elec­ troretinography, although other measures such as perimetrl8 and pupillary lighf response18 have been used. Conventional electroretinography (ERG) is a non-invasive test that reflects the light-induced electrical activity of the outer and middle layers of the, retina throughout its entire extent; therefore, it can relatively easily detect extensive 01 generalised ischaemia. As example, ampli.