Abstract
Purpose: :
Eyes suffering from ocular hypertension or manifest glaucoma showed a pronounced autofluorescence in the parapapillary atrophic zone alpha. The aim of this study was to evaluate the reproducibility of parapapillary autofluorescence (PAF)-planimetry for further glaucoma diagnostics.
Methods: :
Thirty six eyes (randomised selection) of 22 consecutive patients (7 PSX glaucoma, 6 normal tension glaucoma, 9 primary open angle glaucoma) of the Department of Ophthalmology (Erlangen) were examined on three days with the HRA (Heidelberg-Retina-Angiograph II): The fundus was excited with an argon-laser (wavelength 488 nm) using confocal scanning laser ophthalmoscopy and the emitted light above 500 nm wavelength was detected. The reproducibility of 15 HRA images was analyzed (intraindividual analysis, 5 images per day on 3 days). Evaluating the intra-/interobserver-reliability, the PAF was measured by 2 experienced ophthalmologists (U1 and U2) and one beginner (U3) analyzing 5 images of the third examination day. A standardized glaucoma analysis was performed in each patient.
Results: :
The mean PAF was evaluated by the different observers: U1 0.154 ± 0.069 mm² [range: 0.04 - 0.29], U2 0.148 ± 0.094mm² [0.03 - 0.37], and U3 0.243 ± 0.135 mm² [0.05 - 0.68] on third day. Cronbach's alpha was 0.716 for the interobserver reliability. The intraobserver reliability for the PAF measurements on three different days was for all three investigators more than 0.91.
Conclusions: :
Manual PAF-analysis shows good intraobserver reproducibility in contrast to interobserver analysis. Only well experienced and trained ophthalmologists should be allowed to perform this measurement for clinical investigation. A computerized PAF detection should simplify and standardize the PAF-measurement.
Keywords: optic disc • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • retinal pigment epithelium