Novel approaches to the clinical investigation of glaucoma
Hudson, Cameron James
The aim of the thesis was threefold. Firstly, to undertake a retrospective analysis for the presence of 'dissociating factors' in a longitudinal study designed to investigate the relationship between progressive structural and functional damage in open angle glaucoma (OAG). Secondly, to evaluate, in a prospective cross-sectional study, the diagnostic precision of current promising techniques for the detection of OAG. Namely: Optical Coherence Tomography (OCT). Confocal Scanning Laser Ophthalmoscopy (CSLO) and the Multifocal Visual Evoked Potential (mfVEP). Thirdly, to investigate the validity of RNFL thickness measurements obtained using novel software (OCTAnalyse) against those obtained using a commercially available OCT (StratusOCT). The longitudinal study (follow-up 78.4 +9.5 months) was concerned with the concordance between progressive damage identified using Heidelberg Retina Tomograph (HRT) Topographic Change Analysis (TCA) and the Humphrey Field Analyzer (HFA) Glaucoma Probability Analysis (GPA) in patients with OAG or 'high risk' ocular-hypertension (OHT). Seventeen out of 23 patients exhibited progressive damage by TCA and/or by GPA: however, only 4 patients exhibited progressive damage by both techniques. In a majority (70%) of the eyes progressing by only one technique, one of three 'dissociating factors' explained the discordance. The prospective cross-sectional study evaluated the diagnostic precision of retinal nerve fibre layer (RNFL) thickness using StratusOCT. neuroretinal rim area using the HRT and the mfVEP using the Visual Evoked Response Imaging System (VERIS) in a cohort of 20 normal individuals and 23 individuals with OAG. OAG was defined in terms of the subjective assessment of the digital stereoscopic image (DSI) of the ONH by each of 3 masked observers. The sensitivity and specificity of OCT. CSLO. mfVEP for OAG was 91.3% and 65.0%. 43.5% and 95.0%, 69.6% and 60.0%. respectively. In the same cohort. Standard Automated Perimetry (SAP) achieved sensitivity and specificity of 47.8% and 95%. respectively. The best combination of tests for the detection of OAG was OCT and CSLO. which yielded a sensitivity and specificity of 82.6% and 80%. respectively. The topographical correspondence of the ONH abnormality identified by DSI was greatest using CSLO (100%). The final study utilised custom-derived software. OCTAnalyse. designed for use in instances where StratusOCT yielded erroneous RNFL thickness measurements (such as with a detatched internal limiting membrane and/or in a retina exhibiting poor reflectivity). The validity of the RNFL thickness measurements collected in the cross-sectional study and analysed using OCTAnalyse were compared against those obtained using the standard StratusOCT analysis software. OCTAnalyse RNFL thickness measurements were greater (by approximately 9um) and exhibited poorer within-session repeatability than those obtained using the StratusOCT. The overall findings from the three studies provide important information for clinicians implementing the current promising techniques for the routine management of patients with OAG.
Online Research @ Cardiff (http://orca.cf.ac.uk/54707/1/U585096.pdf)