Purchase this article with an account.
A. Ostrovsky, S. Liang, J. Serle, M. Basile, C. Marcus, A. Prywes, R. Rothman, J. Danias; Comparison of Methods Used to Predict Glaucoma Progression and Their Degree of Influence on the Clinician’s Diagnosis of Progression. Invest. Ophthalmol. Vis. Sci. 2008;49(13):736. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare concordance of structural and functional diagnostic tests in predicting glaucoma progression and their influence on the clinician’s impression of progression.
Charts of 201 glaucoma or glaucoma suspect patients followed for at least 3 years with a minimum of 5 Humphrey visual field tests (HVF) and 3 Heidelberg Retinal Tomograph HRT II examinations were reviewed. For each patient, the eye with the higher Glaucoma Staging System (GSS) score was used in the analysis. Evidence of glaucoma progression was considered to be (A) change in Glaucoma Progression Analysis (GPA) on HVF, (B) at least 1dB change in mean deviation (MD) on HVF, (C) change in Moorfields Regression Analysis (MRA) on HRT II, (D) change of >10% of at least 3 out of 5 stereometric analysis (SMA) parameters on HRT II. Pairwise concordance rates were determined for combinations of parameters above and overall agreement using Kappa testing. Logistic regression was used to measure the effect of each parameter on clinician’s impression.
Mean age of patients progressing by any parameter (n=79) was 70.4±15yrs. 18 patients had stage 0 disease, 31 had stage 1-2 disease, and 30 had stage 3-4 disease. Mean follow-up was 6.1±2.3 yrs (range 3.0 to 14.7 yrs). Agreement between pairs of individual parameters ranged between 0 (GPA and MRA) and 10.1% (GPA and MD). Kappa values were <0.7 for all parameters, indicating poor pairwise concordance. Overall concordance between parameters was also found to be low (kappa = 0.09). Concordance rates between clinical impression of progression and any individual parameter ranged between 5.1% (MRA) and 12.7% (SMA). For combinations of the studied parameters, the agreement with clinical impression of progression was 17.7% when compared to structural and 25.3% when compared to functional parameters. MRA changes were not related to clinician’s impression of progression (p=0.46) while all other parameters were (p values 0.012, 0.031 and 0.035 for GPA, MD and SMA respectively).
Overall and pairwise concordance rates of various structural and functional parameters for diagnosing progression are generally low. Clinical impression of glaucomatous progression correlates with both worsening visual field parameters (GPA, MD) and optic nerve parameters (SMA).
This PDF is available to Subscribers Only