Abstract
Purpose :
To report the intra-session repeatability and reliability of central retinal thickness in the macular subfield in patients with non-central diabetic maculopathy.
Methods :
Participants in the randomized clinical trial that assessed the clinical efficacy and safety of a light mask for prevention of dark adaptation in treating and preventing progression of non-central diabetic macular oedema (CLEOPATRA) study had two OCT performed by blinded observers at the same session at baseline and 12 months. Heidelberg Spectralis (Heidelberg, Germany), Cirrus (Carl Zeiss Meditec, UK), Topcon 2000 (Topcon Corporation, Japan) and RS3000 (Nidek Co Ltd, USA) OCT machines were used.
Bland Altman methods of analysis and interclass Correlation of Coefficient were used to assess repeatability and reliability
Results :
A sample size of 601 repeated OCT scans was obtained and analysed.
The 95% coefficient of repeatability (CR) was 20um for retinal thickness in the central 1mm macular subfield. The test-retest reliability using interclass correlation coefficient was 0.89, 95% CI (0.88 to 0.90) for single measures and 0.94, CI (0.93-0.95) for average measures. The reliability of OCT measurements of central macular thickness in patients with non-central diabetic maculopathy is high with an interclass correlation coefficient was 0.89.
Conclusions :
This study suggests that a retinal thickness change of greater than 20um in the central 1mm subfields is required to distinguish true clinical change from measurement variability.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.