Accuracy of scanning laser polarimetry, scanning laser tomography, and their combination in a glaucoma screening trial.

Purpose: To compare the usefulness in glaucoma screening of scanning laser polarimetry (GDx-VCC), scanning laser tomography [Heidelberg retina tomograph II (HRT II)], and their combined evaluation.

Methods: In a nonpopulation-based prepublicized trial, self-recruited white participants were screened for glaucoma with GDx-VCC, HRT II, and by independent clinical examination. Cases with possible glaucoma as found with any of the screening methods underwent a detailed clinical investigation to verify or exclude glaucoma.

Results: Of the 136 attendees 118 participants (218 eyes) successfully underwent the GDx-VCC and HRT II measurements. Twenty-three eyes (11%) of 13 participants had glaucoma. Seventeen of these glaucomatous eyes (74%) had early damage. Evaluated separately, the GDx-VCC screening test (borderline cases grouped with the normal cases) performed best with 96.8% specificity, 89.5% accuracy, 7.5 positive likelihood ratio (PLR), but with only 23.8% sensitivity. Accuracy and PLR for all HRT parameters were <86.4% and <3.7, respectively. Combining different threshold criteria, for GDx-VCC accuracy increased to 90.3% to 90.8% and PLR to 14.0 to 17.7; but for HRT no useful increase was seen (accuracy <86.4% and PLR<4.7 for all combinations). Combination of the best HRT and GDx-VCC criteria resulted in a PLR which was increased compared with the HRT combinations, but decreased compared with the GDx-VCC combinations (PLR<12.7 for all combinations).

Conclusions: In this white screening population with relatively high risk for mild glaucomatous damage, a combination of different GDx-VCC criteria was useful for glaucoma screening; but combinations of HRT criteria or combinations of GDx-VCC criteria with HRT criteria were less good for this purpose.