advertisement
Heijl et al. (1021) evaluate the performance of visual field progression criteria that have been employed by three major multicenter trials in glaucoma, the Advanced Glaucoma Intervention Study (AGIS), the Collaborative Initial Glaucoma Treatment Study (CIGTS) and the Early Manifest Glaucoma Treatment Study (EMGT). The methods used by these investigations were applied to the same visual field dataset from the EMGT study. Results from one eye of 255 patients were evaluated, although data were available for only 245 patients due to several reasons for exclusion. A minimum of five visual fields were required, and all visual fields were also evaluated by a panel of three expert visual field graders. The graders were able to unanimously agree upon 77 eyes that had visual field progression and 95 eyes that had no visual field progression, which served as the basis for the AGIS, CIGTS and EMGT analyses. Agreement among the visual field graders was statistically superior (kappa values above 0.70). Progression, non-progression, sensitivity, specificity, time to progression and sustainability were evaluated. The authors found that the EMGT criteria identified progression earlier (time to progression) and more frequently (sensitivity) than the AGIS and CIGTS criteria, although specificity was lower than for AGIS and CIGTS. All three criteria demonstrated high sustainability, particularly the CIGTS analysis algorithm. There was only modest agreement among the three multicenter trial criteria, in contrast to the superior agreement among expert visual field graders.
EMGT criteria identified progression earlier (time to progression) and more frequently (sensitivity) than the AGIS and CIGTS criteria, although specificity was lower than for AGIS and CIGTSThis investigation was conducted by a group of glaucoma specialists, biostatisticians, epidemiologists and scientists using a thorough, careful and comprehensive procedure. It provides the reader with an outstanding view of the merits and shortcomings of each method of evaluating visual field progression, and attests to the strength and value of combining the expertise of specialists with varied but complementary backgrounds. The authors should be congratulated on providing a careful, easy to understand evaluation of an important issue regarding the management of glaucoma patients and in the critical analysis of multicenter trial results and interpretation.