advertisement
Editors Selection IGR 11-2
Structure and Function: Cup-to-disc ratio and RGC loss
Comment by George Lambrou on:
53513 The relationship between cup-to-disc ratio and estimated number of retinal ganglion cells, Tatham AJ; Weinreb RN; Zangwill LM et al., Investigative Ophthalmology and Visual Science, 2013; 54: 3205-3214
Ever since realizing that glaucomatous damage has both structural and functional implications, ophthalmologists have tried to quantify this damage either through its structural or its functional correlates. And, for several decades prior to the development of modern computer-assisted imaging technologies, the only way to quantify structural damage was through the clinical assessment of the (preferably vertical) cup-to-disc ratio ‐ CDR ‐ as observed ophthalmoscopically. Thus, medical records taken throughout the last third of the XXth century are replete with hand-drawings of cups and discs, while many clinical trials have been based on CDR, alone or in conjunction with perimetry, to quantify glaucoma stabilization or progression.
Yet we have always known that this is a very rough approximation of structural glaucomatous damage, due to the notorious variability of CDRs, as well as to the large inter- and intra-observer variability in assessing them. But how (in)accurate is this approximation? And what is the relationship between CDR and surviving retinal ganglion cells (RGCs)? This is what Tatham et al have investigated in a multicenter cross-sectional study of 336 eyes (156 healthy, 53 POAG-suspect and 127 with confirmed POAG) from 209 subjects. The study methodology was straightforward: each eye underwent a comprehensive ophthalmological examination, including a.o. disc stereophotographs, several 24-2 SITA visual fields and SD-OCT assessments of the disc (with the ONH algorithm) and the retinal nerve fiber layer.
CDR assessment is an insensitive method for evaluation of progressive neural losses in glaucoma
RGC count estimates were obtained using both functional (SAP) and structural (OCT) parameters, according to a model developed and previously published by part of the authors. Anonymized disc stereophotographs were assessed by two or more experienced graders, masked to the subjects' other results, to obtain CDR values. Finally, the authors analyzed the scatterplots between RGC estimates and vertical/average CDRs using LOWESS and polynomial curve-fitting.
The results confirmed that CDR increases with disease severity. The relationship between CDR and RGC loss, however, is non-linear and rather loose (as suggested by the wide scatterplots), influenced by disease-independent factors such as disk size and age. Modelling this relationship (as a 3rd-degree polynomial) suggests that in moderate disease stages, CDR changes may occur after a limited RGC loss, while in more advanced stages, extensive loss RGC may be required for a detectable change in CDR to occur. In the discussion, the authors consider various factors explanations this non-linearity and conclude that assessment of CDR is an insensitive method for evaluation of progressive neural losses in glaucoma.
The implications of the study are that while clinical assessment of the cup is likely to remain an integral part of glaucoma diagnosis and follow-up, we should be aware that this method is not optimal and that the computerized imaging methodologies available today may be better suited for quantifying progressive RGC losses due to glaucoma.
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One