Purchase this article with an account.
M. W. Dul, W. H. Swanson; Clinical Validity of Contrast Sensitivity Perimetry: Assessing Eyes With Severe Glaucomatous Visual Loss. Invest. Ophthalmol. Vis. Sci. 2007;48(13):1608.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To assess the clinical utility of Contrast Sensitivity Perimetry (CSP) for assessing visual function in eyes with severe glaucomatous damage.
Twenty-one patients with stable glaucoma were recruited for whom stereoscopic structural assessment of the optic nerve head and neural retinal rim was judged as cupped to the rim in quadrants corresponding to dense scotomas (sensitivities of 10 dB to < 0 dB) on conventional automated perimetry (CAP). One eye was chosen for each patient, and tested with CSP using 0.4 cycle/deg sinusoidal patches at 26 locations in the central visual field. CAP was performed with the Humphrey Field Analyzer II (stimulus size III, 24-2 SITA Standard algorithm). Test-retest variability was assessed for CSP with testing sessions one week apart. CAP variability was computed from recent 24-2 fields.
CSP values were highly repeatable in all quadrants with dense defects, with 95% confidence limits of +/- 3 dB. HFA values were highly variable in these same quadrants, with 95% confidence limits of +/- 15 dB. For the locations where the maximum CSP stimulus was below threshold on one test, on the other test it was below threshold for 86% of locations and for the remaining locations it was a never more than 3 dB above threshold (median 1 dB above threshold). For CAP, when the maximum stimulus was below threshold on one test it was below threshold for 81% of locations and for the remaining locations it was as much as 20 dB above threshold (median 8 dB above threshold).
Contrast sensitivity perimetry (CSP) provided reliable measures of visual sensitivity with low variability in quadrants with dense scotomas where CAP measures had high variability and clinical optic nerve assessment found little if any visible neuroretinal rim. CSP may prove to be an effective means of following patients for continued progression when the lower extremes have been reached for the dynamic ranges of conventional functional and structural measures.
This PDF is available to Subscribers Only