Diagnostic accuracy of structural and visual function tests to detect age-related eye disease

Global trends in population ageing will lead to an exponential increase in age-related eye disease. If undetected, and in the absence of timely treatment, these diseases could result in significant visual impairment. New technologies for the assessment of ocular structure and function could potentially improve disease identification, but it is paramount that their diagnostic performance is fully evaluated before they can be employed in a routine clinical setting. Chapter 2 describes the results from a systematic review of the diagnostic accuracy of five non-invasive tests to detect those at risk of primary angle closure glaucoma. The tests evaluated were the van Herick test of limbal anterior chamber depth, flashlight test, Scanning Peripheral Anterior Chamber Depth Analyser, Scheimpflug photography and Anterior Segment Optical Coherence Tomography. A meta-analysis was performed and summary estimates of sensitivity and specificity were calculated for each test. In addition, test comparisons were conducted based on the thresholds with the best performing diagnostic odds ratio. Overall, most tests performed well and showed equivalent accuracy. However, care should be taken in applying the summary estimates in clinical practice due to the observed heterogeneity and high risk of bias found in most studies. Chapter 3 describes a diagnostic accuracy study to investigate the performance of a prototype flicker perimeter (Accelerator 4-Alternative Forced-Choice Flicker Test prototype (A4FTp)) to detect primary open angle glaucoma (POAG). Participants with glaucoma (n=38) were compared to normal controls (n=40). The diagnosis of POAG was confirmed by a reference standard ophthalmic examination and the performance of the A4FTp was compared with two available screening tests: Frequency Doubling Technology perimeter and optical coherence tomography (OCT). The clinician performing these tests was masked to the results of the reference standard examination. Diagnostic accuracy of all three tests was equivalent in the detection of POAG. Time taken to complete the A4FTp was relatively short with good subject acceptability. Initial results are promising and with further development, the test could have a role in glaucoma detection. Chapter 4 reports on the diagnostic value of OCT to detect glaucomatous disc damage or retinal pathology using a clinical vignette methodology. A sample of 50 community optometrists undertook online training followed by completion of a computer-based vignette assessment, showing either a single fundus/disc photographic image (n=26) or a combination of a fundus/disc image with the corresponding OCT scan (n=26). Comparing the OCT combination to fundus imaging alone improved overall diagnostic performance by approximately 20%, with fewer false positives and false negatives recorded. These findings suggest that OCT could augment case-finding, but further research is needed to evaluate its value in a real-world setting.