Five-Year Visual Field Outcomes of the HORIZON Trial

Abstract

To compare visual field (VF) progression between glaucoma patients receiving cataract surgery alone (CS) or with a Hydrus microstent (CS-HMS). Post hoc analysis of VF data from the HORIZON multicenter randomized controlled trial. A total of 556 patients with glaucoma and cataract were randomized 2:1 to either CS-HMS (369) or CS (187) and followed up for 5 years. VF was performed at 6 months and then every year after surgery. We analyzed data for all participants with at least 3 reliable VFs (false positives < 15%). Average between-group difference in rate of progression (RoP) was tested using a Bayesian mixed model and a 2-sided Bayesian P value <.05 (main outcome). A multivariable model measured the effect of intraocular pressure (IOP). A survival analysis compared the probability of global VF sensitivity dropping by predefined cutoffs (2.5, 3.5, 4.5, and 5.5 dB) from baseline. Data from 352 eyes in the CS-HMS arm and 165 in the CS arm were analyzed (2966 VFs). The mean RoP was -0.26 dB/y (95% credible interval -0.36, -0.16) for CS-HMS and -0.49 dB/y (95% credible interval -0.63, -0.34) for CS. This difference was significant (P=.0138). The difference in IOP only explained 17% of the effect (P < .0001). Five-year survival analysis showed an increased probability of VF worsening by 5.5 dB (P=.0170), indicating a greater proportion of fast progressors in the CS arm. CS-HMS has a significant effect on VF preservation in glaucoma patients compared with CS alone, reducing the proportion of fast progressors.