Journal of Ophthalmology (Dec 2016)
Primary open-angle glaucoma progression depending on clinical indices at presentation
Abstract
Background: Given the importance of developing simple and efficient methods for the diagnosis and prognosis of primary open-angle glaucoma (POAG), we have previously proposed an index of progression of glaucoma stage (IPGS). Purpose: To investigate associations between progression of POAG and clinical indices at presentation. Materials and Methods: The study group comprised 172 patients (78 men (45%) and 94 (55 %) women; mean age at presentation, 57.3±1.1 years) diagnosed with POAG. POAG stage was graded by the Nesterov and Bunin classification of 1975 well as by the Hodapp classification based on the results of the ophthalmological observation of patients over two years. IPGS was calculated as a percentage ratio of (a) a sum of ranked POAG stages and IOP values (categories of IOP) at baseline and study time points and (b) difference between age and disease duration, expressed in years. Sex, age, duration of disease, stage of POAG at presentation, and category of IOP at presentation (IOP0) were input variables used in multiple logistic regression model building. Results: The IPGS index clearly represented both the change in POAG stages over time and progression of the disease. Age (r = -0.468), disease duration (r = 0.695), stage of POAG (r = +0.805) and IOP category (r= +0.735) correlated more significantly than did other variables with IPGS. Patient sex was not correlated with IPGS (r=-0.169; p = 0.027). The equation for predicting IPGS estimates based on characteristics at presentation was developed through multiple regression: "IPGS=35.311-0.521?A+2.512?D+3.913?S+4.191?C" where IPGS is index of progression of glaucoma stage; A, patient age; D, disease duration; S, POAG stage; C, IOP category. In terms of the relative impact of the predictors on the IPGS, IOP category had the strongest influence (? = 4.191±0.446), followed by stage of POAG at presentation (? = 3.913 ± 0.356); disease duration (? = 2.512 ± 0.313) and patient age (? = -0.521 ± 0.022). The developed equation is characterized by a significant effect of the combination of independent variables on the dependent variable (R=+0.965 (F=562.35, P<.0001); adjusted R2=+0.931). Conclusion: Amongst predictors, IOP category had the strongest influence, while patient age had the weakest influence on the IPGS. The equation for prediction of IPGS can be proposed as early as at presentation and will allow the clinician to conduct individualized treatment of patients.
Keywords