Comparing central corneal thickness measured using ultrasound pachymetry and the Pentacam in healthy subjects and patients with primary open-angle glaucoma

Purpose. - To determine the capacity for measurement of central corneal thickness by ultrasound pachymetry and by Pentacam (pachymetry at the pupillary axis and minimum corneal pachymetry) to discriminate between healthy controls and patients with primary open-angle glaucoma. Methods. - Measurements of the left eyes of 123 control subjects and 128 glaucomatous patients were taken. Differences between the two data sets for each variable were identified using a t-test for independent samples. Three univariate binary logistic regression models (to discriminate between glaucoma and controls) were constructed, in which the predictive variables were the determinations of corneal thickness. Results. - All variables showed a normal distribution. Controls and patients failed to differ significantly in terms of ultrasound pachymetry but did differ in terms of pupillary axis thickness (mean difference 22.01 mu m in favour of controls; 95% CI: 5.61-34.15) and minimum corneal thickness (mean difference 21.65 mu m in favour of controls; 95% CI: 3.05-32.25). The logistic regression model for ultrasound pachymetry was not significant (P = 0.051) but significant discriminatory capacity was shown for pupillary axis thickness (expB = 0.987; 95% CI: 0.99-0.993; sensitivity: 58.5%; specificity: 64.8%) and minimum corneal thickness (expB = 0.987; 95% CI: 0.981-0.994; sensitivity: 61.9%; specificity: 63.1%). Conclusions. - Central corneal thickness measurement by Pentacam, while not an ad hoc diagnostic test for glaucoma, showed a similar yet not inappreciable capacity to discriminate between glaucoma patients and controls. (C) 2011 Elsevier Masson SAS. All rights reserved.