Purchase this article with an account.
Michael Kalloniatis, Jack Phu, Agnes Yiu Jeung Choi, Sieu Khuu, Nayuta Yoshioka, Lisa Nivison-Smith, Barbara Zangerl; Hemifield asymmetry analysis within the central 20 degrees in open-angle glaucoma: diagnostic utility of 10-2 visual fields and macula optical coherence tomography. Invest. Ophthalmol. Vis. Sci. 2018;59(9):4083.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Up to half of early open-angle glaucoma (OAG) patients may have visual field (VF) defects within the central 20o. However, there are no current asymmetry analyses across the horizontal midline available for central VF test grids or optical coherence tomography (OCT) scans. We aimed to determine the diagnostic utility of asymmetry analyses of instruments testing the central ~20o of the retina for OAG.
Thirty-one patients with OAG underwent VF testing on the Humphrey Field Analyzer 10-2 test grid (Goldmann size III, GIII) and had OCT scans performed within the posterior pole (Spectralis OCT 8x8 posterior pole scans). VF and OCT tests were performed at least twice for each patient, and OCT scans were manually corrected for errors due to automated segmentation. Data from normal subjects (n=56 for VF and n=253 for OCT scans) were used to generate normative distributions for comparison. Pattern recognition analysis was used to identify test locations sharing the same sensitivity – contrast sensitivity isocontours (for 10-2: Phu et al IOVS 2017), or areas displaying similar ganglion cell layer thickness – ganglion cell isocontours (for OCT scans: Yoshioka et al IOVS 2017). Resultant pattern recognition theme maps were used to identify the zones for asymmetry analysis. Outside normal limits (ONL) was defined as p < 0.01 in both superior and inferior hemifields, or p < 0.005 level of asymmetry when comparison across hemifields.
71.0% of OAG patients were flagged as ONL on asymmetry analysis by the 10-2 and 51.6% by the OCT. Fifteen patients had test locations within the Spectralis test grid which had poor segmentation and thus asymmetry analysis could not be reliably conducted, i.e. regarded as missed diagnoses. Of the 31 OAG patients, ONL results were found for VF only (n=9), OCT only (n=4) criteria, while 13 had ONL flagged on both. Five OAG subjects did not fail either asymmetry criteria.
Hemifield asymmetry analyses within the central 20o may play a role in the diagnosis of OAG. Using VF or OCT tests in isolation miss a significant proportion of OAG cases. Both VF and OCT have limitations, namely, the relative insensitivity of the GIII stimulus size and segmentation errors. Further studies are warranted to improve the sensitivity of both techniques for early OAG.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only