Abstract
Purpose::
To evaluate the sensitivity and specificity of grading age-related maculopathy (ARM) using slit-lamp biomicroscopy (SLB) and a 78 dioptre convex lens.
Methods::
ARM was graded in 150 participants of the Carotenoids in Age-related Maculopathy (CARMA) study, by SLB and by grading of standardised (colour and stereoscopic) fundus photographs (using the Wisconsin Age-Related Maculopathy Grading System) in an accredited reading centre (The Retina Image Reading Centre, Queen’s University, Belfast). The kappa (k) statistic, which corrects for chance agreement, was used to express agreement between these two techniques of ARM grading.
Results::
There was a fair degree of agreement for identifying the presence of any drusen, regardless of the size and type (k = 0.22), and the examining ophthalmologist failed to identify the presence of any drusen in 10% of cases. However, there was perfect agreement between SLB and grading of fundus photographs for estimating the total number of drusen in patients with early ARM (drusen number < 20, k = 0.81; drusen number > 20, k = 0.9). Furthermore, both techniques of ARM grading demonstrated a high degree of agreement for localizing drusen to either the inner or the outer zones (k = 0.8).In addition, the level of agreement between SLB and grading of fundus photographs for identifying hypo-pigmentation and hyper-pigmentation was moderate and high, respectively (hypo-pigmentation, k = 0.415; hyper-pigmentation, k = 0.650).
Conclusions::
Slit-lamp biomicroscopic grading of ARM, with the exception of soft indistinct drusen, exhibits comparable sensitivity to standardised and accredited grading of fundus photographs.
Keywords: clinical (human) or epidemiologic studies: systems/equipment/techniques • age-related macular degeneration • drusen