Changes in Corneal Biomechanical Properties after Long-Term Topical Prostaglandin Therapy

OBJECTIVE: To compare corneal biomechanical properties, measured by a newly developed tonometer (Corneal Visualization Scheimpflug Technology, Corvis ST), in untreated primary open angle glaucoma (POAG) patients, POAG patients with long-term topical prostaglandin analog (PGA) therapy and in normal controls. Further is to investigate the potential effects of PGA on corneal biomechanics. METHODS: In this case-control study, 35 consecutive medication naïve eyes with POAG, 34 POAG eyes with at least 2 years treatment by PGA and 19 normal eyes were included. Intraocular pressure (IOP), central corneal thickness (CCT) and corneal biomechanical parameters, including deformation amplitude (DA), applanation time (AT1 and AT2), applanation length (AL1 and AL2), applanation velocity (AV1 and AV2), and peak distance and radius were measured using Corvis ST. Axial length and corneal curvature were measured with partial coherence interferometry (IOLMaster, Zeiss, Germany). General linear model analysis was performed to investigate the corneal biomechanical property changes among the normal controls, newly diagnosed POAG patients and POAG patients with long-term PGA treatment, and among the subgroups of different types of PGA treatment, including bimatoprost, latanoprost and travoprost. Furthermore, pairwise comparisons using Bonferroni correction for least squares means were employed. RESULTS: AT1 (p<0.0001), AV1 (p<0.0001), AT2 (p = 0.0001), AV2 (p<0.0001) and DA (p = 0.0004) in newly diagnosed glaucoma patients were significantly different from those in normal subjects and in patients underwent at least 2 years topical PGA therapy after adjusting for age and gender. After adjusting for age, gender, IOP, CCT, axial length and corneal curvature, a significant difference was detected for DA between glaucoma patients without PGA treatment and patients with long-term PGA therapy (p = 0.0387). Furthermore, there were no statistical significant differences in all of the corneal biomechanical parameters among the 3 types of PGA therapy subgroups, namely bimatoprost, latanoprost and travoprost. CONCLUSIONS: Significant changes in corneal deformation parameters were found among untreated POAG patients, POAG patients with long-term topical PGA therapy and normal controls. Long-term topical PGA treatment might have a direct effect on corneal biomechanical properties in addition to the indirect effect owing to the PGA-induced IOP reduction and CCT decrease on corneal dynamic properties.