advertisement

Editors Selection IGR 17-4

IOP, VF, Imaging and Electrophysiology: Novel VEP technique for diagnosis

Comment by Brad Fortune on:

22307 Novel electrophysiological instrument for rapid and objective assessment of magnocellular deficits associated with glaucoma, Zemon V; Tsai JC; Forbes M et al., Documenta Ophthalmologica, 2008; 117: 233-243

Find related abstracts

Following an impressive series of publications describing results of their basic research on a novel electrodiagnostic technique, Zemon et al. (1446) have recently published a study in which they applied a streamlined version of their technique to glaucoma diagnosis.¹ The technique is based on recording visual evoked potentials (VEP) to a low-contrast pattern of isolated checks that are sinusoidally modulated to appear and disappear at a relatively fast rate (10 Hz). The square checks are moderately sized (14 arc min per side) and the stimulus pattern subtends 11 degrees total diameter. It also employs two stimulus conditions, one in which the modulation rises above the mean luminance level (as the checks appear, they look brighter than the background, therefore called 'ON') and another where the modulation occurs below the mean luminance level (as the checks appear they look darker than the background, therefore called 'OFF'). By their choice of stimulus characteristics (low contrast and relatively fast modulation), the investigators argue that the test is designed to preferentially drive responses of magno-cellular projecting ganglion cells. This in turn drives the hypothesis that this technique should consequently offer powerful diagnostic capabilities and detection of early glaucomatous functional loss. The authors support this notion citing well-known findings of preferential loss of larger ganglion cells in glaucoma. And although they do mention that such preferential magno-cellular loss was not observed in experimental glaucoma in primates,² the authors do not consider numerous careful psychophysical studies whose findings contradict any notion of preferential magno-cellular deficits in human glaucoma (such as the work of Allison McKendrick and colleagues,^3-5 and others,^6-8 for example). Though some controversy over this issue persists,⁹ the technique developed by Zemon and co-workers nevertheless offers a rapid, objective method for assessment of vision function and their earlier pilot work showed promise for glaucoma diagnosis. Therefore it was interesting to see the results of this study that included 18 patients clinically diagnosed with glaucoma, and 16 controls. Eleven of the patients had early loss defined as mean deviation (MD) values above ‐6 dB on standard automated perimetry (SAP); three patients had MD values between ‐6 and ‐12 dB and four had MD values below ‐12 dB. The diagnostic accuracy of the 'ON' test at 15% contrast was 94%. Using the optimal signal-to-noise discriminant value, sensitivity was 78% and specificity was 100%. The 'OFF' condition did not perform quite as well and the 10% contrast 'ON' condition also had reduced specificity. The latter result suggests that the technique would not be as robust if cataract or other optical media opacities were present, an important caveat to consider. Another caveat is that the patient group was older than the control group, though the authors found no age dependence of the VEP response parameters and argued that the age difference was not important. Once the gold-cup electrodes were affixed to the scalp for the VEP recordings, the test required less than two minutes to complete both eyes. This is indeed rapid, though not necessarily faster than pattern electroretinography (PERG) techniques,^10,11 which also offer objective assessment and similar diagnostic performance for glaucoma. The multifocal VEP (mfVEP) also offers comparable performance,^e.g.,12-14 but requires test times of about six to eight minutes per eye after set-up (electrode placement).^e.g.,12-14 However, the mfVEP also enables topographic assessment, which most PERG techniques and this new isolated check VEP (icVEP) technique do not. Interestingly, Zemon and colleagues found that the signal-tonoise of icVEP responses was not correlated with SAP MD values, concluding that this was "additional evidence these techniques tap independent (central versus peripheral) mechanisms". However, they did not report results of any comparison with central SAP sensitivities (perhaps because only the foveal test point and four nearest neighbors would fall within the area of the icVEP stimulus) so it is difficult to be convinced of the two tests' independence until further studies address this question directly.

• ← Previous
• Next →