Abstract
Purpose :
To use our published multivariate model of retinal nerve fiber layer (RNFL) variation to generate patient specific individualized expected values (EV) of the RNFL thickness, and to investigate whether normalized individual deviations from the EV (DoEV) of RNFL thickness improve the discrimination between healthy and glaucomatous subjects.
Methods :
This prospective study compares 69 cases of early to moderate glaucoma with existing data of 202 healthy subjects. All subjects underwent optical coherence tomography (OCT) (Cirrus®, Carl Zeiss Meditec) in macula and circumpapillary areas. The 2D projection images were used to determine retinal vessel density (RVD) at a 3.4mm diameter circle, as published. The regression model was recalculated using 281 healthy subjects of the Cirrus normative database (CND) considering 8 factors: RVD, fovea distance and angle, optic disk (OD) size, ratio and angle and refractive error. The EVs were calculated twice: 1) age corrected EV (AEV) based on CND using intercept and slope of age; 2) individual EV (IEV), using intercept and slopes of our multivariate model. The DoEV for the AEV and the IEV were calculated for 12 sectors and normalized by dividing through the standard deviations (nDoEV). The minimum nDoEV for the 12 sectors was determined for each subject using the AEV and the IEV. The diagnostic value of the minimum nDoEV of AEV and the IEV was analyzed using AROC. Sensitivities at fixed specificities were compared between methods with McNemar test.
Results :
For the minimum nDoEV of AEV and IEV the AROCs were 0.855 (0.796-0.914) and 0.893 (0.840-0.946). Although this was not significantly different (Fig. 1), there was an improved sensitivity at high specificities for nDoEV of IEV compared to the AEV (Tab. 1).
Conclusions :
Application of patient specific IEV improved sensitivity at high specificity level above 95% compared to standard AEV.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.