Age-dependent normative values for differential luminance sensitivity in automated static perimetry using the Octopus 101
(Valeur normal en rapport avec l'âge des seuils de sensibilités rétinienne en périmétrie automatique avec l'Octopus 101)

Hermann A, Paetzold J, Vonthein R, Krapp E, Rauscher S, Schiefer U.
Centre for Ophthalmology, University of Tübingen, Tübingen, Germany.

PURPOSE: To determine age-dependent normative differential threshold values for the Octopus 101 instrument and to create a smooth mathematical model characterizing the age-dependency and asymmetry of the hill of vision.

METHODS: Static automated perimetry within the central 30 degrees visual field (VF) was conducted with the Octopus 101 (background luminance 10 cd/m(2)) in 81 eyes of 81 ophthalmologically healthy subjects (11-12 per decade of age) aged 10-79 years. A 4-2-2 staircase strategy with three reversals was run. The test point grid consisted of 68 concentrically arranged points with test point condensation towards the VF centre, representing the approximately rotation-symmetrical 30 degrees hill of vision. Thresholds of differential luminance sensitivity (DLS) were estimated by the maximum likelihood method. A smooth mathematical model was fitted to the normative data.

RESULTS: The model fit was satisfactory (r(2) = 0.74). Covariables were: age, eccentricity, angle and subject. Total random standard deviation (SD) was 1.75 dB. The residual SD exceeded 1.75 dB in the border region, was 1.5 dB within the centre and fell below 1.25 dB in a ring around the centre. Average thresholds of DLS varied with age quadratically. It is close to constant for the 10-40-year-old age group and declines ever more steeply thereafter. The effect of age on DLS in the VF increased with eccentricity. The greatest drop was located in the peripheral superior hemifield: at 25 degrees eccentricity the superior DLS was estimated to be 5.5 dB higher in 10-year-olds than in 75-year-olds.

CONCLUSIONS: This new smooth model allows for the prediction of age-related normal threshold values for any stimulus location within the 30 degrees VF and thus for the calculation of global and local measures of defect such as mean defects or p-values for any type of stimulus.

 Retour à la bibliographie de Juin 2008

(Dernière mise à jour de cette page le 31/07/08)