Abstract
Purpose :
To compare the association of IOP parameters measured by DCT and GAT with the progression of glaucoma in eyes with a history of laser refractive surgery (LRS).
Methods :
Participants: Patients with open-angle glaucoma and a LRS history who visited the glaucoma clinic at the Department of Ophthalmology, Severance Hospital, Yonsei University School of Medicine, between January 2005 and June 2017 were identified by review of the medical records. Ninety-eight eyes (54 patients) were included in the study.
Interventions: IOP was measured by both GAT and DCT during follow-up. Two additional equations were used to convert the results obtained by GAT to corrected IOP parameters. Progression of glaucoma was defined by structural and/or functional alterations.
Main Outcomes and Measures: The main outcomes were the association between progression of glaucoma and each IOP parameter measured by the different tonometry methods and formulae and whether the DCT parameters were more relevant to progression of glaucoma than the other IOP parameters. Baseline, mean, and peak IOP, fluctuation of IOP, and reduction of IOP were measured by each tonometry method. Clustered logistic regression was used to identify variables correlated with the progression of glaucoma. The areas under the curve (AUCs) for correlated variables were compared using Delong’s method.
Results :
Twenty-two eyes showed progression of glaucoma. The mean DCT value (odds ratio [OR] 1.36, P = .024), peak DCT value (OR 1.19, P = .02), and pattern standard deviation (OR 1.10, P = .016) were significant risk factors for progression. There was a significant difference in the predictive ability of the mean DCT and GAT values (AUC 0.63 and 0.514, respectively; P = .01) and of the peak DCT and GAT values (0.646 and 0.503, respectively, P = .009). The AUC values for corrected IOP did not exceed those of DCT.
Conclusions :
IOP measurements were more associated with progression of glaucoma when measurements were obtained by DCT than GAT in eyes with an LRS history.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.