Correlation of outer nuclear layer thickness with cone density values in patients with retinitis pigmentosa and healthy subjects

Abstract

Purpose: Outer nuclear layer thickness (ONL) has been widely used as a marker for photoreceptor numbers. However, some studies have not shown a consistent correlation between ONL thickness and cone density. Variability in ONL thickness measurements may be attributed to the inclusion of Henle’s fiber layer (HFL) when using standard optical coherence tomography (OCT) imaging techniques. The aim of our study is to compare ONL thickness using directional OCT (D-OCT) with cone density values at identical retinal locations in normals and retinitis pigmentosa (RP) patients.

Methods: Cross-sectional analysis of images acquired with D-OCT and adaptive optics scanning laser ophthalmoscopy (AOSLO) from normals and RP patients. The D-OCT data sets were obtained using a previously described method of varying the pupil entry position of the beam to reveal the HFL. Five retinal layers (ILM, OPL/HFL junction, HFL/ONL junction, ELM and outer RPE border) were segmented using Image J and MatLab® software. Multiple locations along the central horizontal meridian were selected on the AOSLO images and compared with ONL thicknesses at identical eccentricities. Cone density was analyzed in areas where cone mosaics could unambiguously be identified and expressed as cones/degree2.

Results: Mean age in our population was 40.5 in patients and 42.0 years in normals. Average central retinal thickness was 240±25.0um and 234±12.8um, with an average ONL thickness at 1° eccentricity of 47.2±8.8um and 46.0±3.7um in patients and in normals. Mean ONL thickness at 2° eccentricity was also comparable with 33.7±6.7um in RP patients and 33.7±7.1um in normals. Average cone densities measured at 1° eccentricity were 3718±1287 and 4313±541 for patients and normals, respectively. At more eccentric locations (2° eccentricity) the values were 2646±531 and 2646±245. Combining the normal and patient images, ONL thickness was significantly correlated with cone density (Pearson’s correlation coefficient 0.74, 95% CI 0.67, 0.82). The correlation between ONL+HFL values and cone density on the other hand was weaker (coefficient 0.59, 95% CI 0.46, 0.78).

Conclusions: ONL thickness measures correlated with cone density measures in normals and RP patients, as measured using D-OCT. ONL thickness could be an easily obtainable indirect parameter to represent cone density values both in diseased and healthy individuals.

Keywords: 696 retinal degenerations: hereditary • 550 imaging/image analysis: clinical • 552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)