Purchase this article with an account.
Gregory R. Jackson, Laura E. Walter, Ingrid U. Scott, David A. Quillen, Thomas W. Gardner; Measurement Error Of The Matrix FDT For Patients With Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2012;53(14):2862. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The purpose was to assess the measurement error (precision) of the Matrix FDT in patients with diabetic retinopathy. The Matrix FDT assesses contrast sensitivity to a flickering sinusoidal grating at 55 locations throughout the central visual field. Matrix FDT impairment has been shown to be highly sensitive marker of diabetic retinopathy. Basic performance data like measurement error will be useful for the design of future natural history and interventional studies.
All subjects in the study were experienced with FDT perimetry before participation. Each participant’s FDT field was measured twice. Two measurements were made with an enforced 5 minute rest break. FDT performance was characterized using two parameters: the central target sensitivity, and the mean field sensitivity. Parameters were selected because they are highly sensitive to the presence of diabetic retinopathy. Measurement error was calculated using a method based on within standard deviation.
The study sample consisted of 30 adults with diabetic retinopathy (mean age = 57 years, range = 29 to 76 years). The measurement error for the central test point was calculated to be 2.88 dB. The measurement error for the mean sensitivity of the field was calculated to be 1.44 dB. For both the central test point and mean field sensitivity, the variability of the measurement did not change as a function of the magnitude of the parameter. The rod-cone break and rod intercept parameters were highly correlated (r = 0.70, p < 0.0001).
The measurement error (precision) of the Matrix FDT in patients with diabetic retinopathy ranges from 1.44 dB to 2.88 dB depending upon the summary parameter. A change of more than 3 dB represents a real change. We propose that a 6 dB change should be considered as meaningful change in performance.
This PDF is available to Subscribers Only