Abstract
Purpose: :
To investigate predictive risk factors for development of atrophy in areas of drusen and drusenoid pigment epithelium detachment (DPED) in eyes with non-neovascular age-related macular degeneration (NNVAMD) by using serial morphological observations from spectral domain-optical coherence tomography (SD-OCT).
Methods: :
Patients with NNVAMD who underwent SD-OCT imaging (Spectralis HRA+OCT, Heidelberg Engineering) at our institute between January 2008 and October 2011 were retrospectively reviewed. Eyes with registered OCT images over a minimum follow-up of 6 months without other retinal diseases were selected. Based on the tracking and registration functionality of the Spectralis HRA+OCT, individual drusenoid lesions (drusen and DPEDs) were recognized from visit to visit throughout their full follow-up and given unique labels for subsequent consecutive gradings, Features of each drusenoid lesion (DL) were independently evaluated using detailed OCT grading, including maximum DL height, numbers of B-scans subtended by the individual DL (a measure of drusen diameter), choroidal thickness (CT) directly below the DL, presence of hyper reflective foci (HRF) above the DL, innermost extent of HRF, presence of IS-OS junction integrity, external limiting membrane integrity, RPE integrity, internal reflectivity of the DL, and finally presence of atrophy (defined as loss of the RPE band with increased choroidal reflectivity). A multivariate binomial logistic regression analysis was performed.
Results: :
41 eyes (29 patients) with a mean follow-up time of 21 months (range 6-44 months) were included in the study. A total of 1950 DL-specific visits from 608 DLs (583 DLs identified at baseline and 25 newly developed during follow-up) were graded. Baseline maximum DL height was 73.7±34.5 (range from 0 to 246) microns. 3.9% (24/608) of DLs progressed to atrophy within 17±9.6 (range 5-27) months of the initial visit. Logistic analyses demonstrated that at 6 months after baseline, HRF (odds ratio (OR) =3.21, 95% confidence interval (CI) =1.47-7.06) and RPE integrity (OR=0.23, CI=0.07-0.74) were together predictive of new atrophy onset. These same factors were also predictors for new atrophy at the final visit (OR=2.02, CI= 1.18-3.47 for HRF and OR=0.31, CI=0.18-0.55 for RPE integrity).
Conclusions: :
In this study, HRF and RPE integrity were significant predictors for drusenoid lesions which progressed to atrophy over the ensuing months. These findings may help to further the elucidate pathogenesis and natural disease progression of eyes with NNVAMD.
Keywords: age-related macular degeneration • imaging/image analysis: clinical • clinical (human) or epidemiologic studies: risk factor assessment