Purpose: To evaluate the influence of optic disc size and disease severity on the diagnostic validity of optical coherence tomography (Stratus OCT), scanning laser polarimetry [GDx variable corneal compensator (VCC)], and confocal scanning laser ophthalmoscopy [Heidelberg retina tomograph II (HRT II)] in Indian eyes with glaucoma.
Methods: Ninety-five normal and 125 glaucoma patients underwent imaging with Stratus OCT, GDx VCC, and HRT II. One eye of each person was randomly selected for analysis. Using disc area determined on HRT II, discs were classified as small (<2 mm), moderate (2 to 3 mm), and large (>3 mm). The parameter with the best sensitivity for each device, at a fixed specificity, was compared for different disc sizes. Logistic marginal regression was used to study the effect of disc size and disease severity (mean deviation on standard automated perimetry) on the diagnostic performance of these imaging devices.
Results: At a fixed specificity of 84.2%, the sensitivity of HRT II was significantly different for varying disc sizes (P=0.0004). The sensitivities for dissimilar disc sizes were not significantly different for the GDx VCC (P=0.928) or Stratus OCT (P=0.381). Logistic marginal regression also showed that sensitivity of HRT II increased with increasing disc size, whereas sensitivity of OCT and GDx were independent of the disc size. The sensitivity of all 3 technologies increased with increasing disease severity (decreasing mean deviation).
Conclusions: Optic disc size affects the diagnostic capability of HRT II but not that of GDx VCC or Stratus OCT. The sensitivity of all 3 imaging technologies increased with increasing disease severity.