AbstractAge-related macular degeneration (AMD) is the most common cause of irreversible visual impairment in the western world. Multimodal imaging has contributed a lot into disease detection, diagnosis and management. The availability of multimodal imaging is an unprecedented advantageous opportunity into relating structure with function in the presence of disease.
The Northern Ireland Cohort of Longitudinal study of Aging (NICOLA) is a population based study which was designed to understand what factors are related with healthy aging. The sampling was designed to recruit individuals aged over 50 years old and consisted of two parts, a home-based interview and a health assessment which included an ophthalmic component which comprised of multimodal retinal imaging. Those attending the health assessment were the pool from which participants to the current study were drawn.
The Northern Ireland Sensory Aging study (NISA) is a prospective, case-control add-on study, which invited back persons from the main NICOLA study in order to assess in detail those who had early and/or intermediate AMD. The NISA study, involved detailed functional and structural phenotyping of these participants with the use of state-of-the-art psychophysical tests and retinal imaging modalities.
Chapter 1 of this thesis is an overview of the knowledge acquired so far through research regarding the pathogenesis, risk factors, classification and effect of visual function of both AMD and its recently reported phenotype, reticular pseudodrusen/subretinal drusenoid deposits (RPD/SDD).
Chapter 2, is an overview of the methods used to carry out the functional tests and imaging in the NISA study, as well as the rationale for inclusion of those assessments. In the same chapter, the effect of the time chosen to dark-adapt an individual, on their performance on a novel piece of equipment, the scotopic MAcular Integrity Assessment (MAIA, CenterVue, Padova, Italy) microperimeter, as well as the repeatability of the device are tested.
In chapter 3, the rationale and methods of developing an optical coherence tomography (OCT)-based classification system for the NISA study are described, thus adhering to the need for an OCT-based classification for AMD. The colour fundus photography (CFP) and OCT characteristics, as well as the prevalence of AMD phenotypes, of the NISA study population are described. Based on CFP detection a frequency of 31.8% and 15.9% for early and intermediate AMD respectively, were found. Furthermore, RPD were found in 1.7% of the study population, a frequency which is in agreement with previous studies. Based on OCT imaging, 39.9% of the study eyes showed typical drusen, with 34.7% of those being of homogeneous internal reflectivity. Subretinal drusneoid deposits were present in 39.8%. This percentage is similar to previous rates reported from studies which used the OCT modality. Of those with SDD, 19.1% were stage 1, 16.8% stage 2 and 3.9% stage 3 SDD. Analysis of agreement between a well-established CFP with the OCT-based classification system developed for the NISA study, showed only slight agreement between the two, whereas substantial agreement between the standard and wide-field OCT lenses was found.
In chapter 4 an assessment of the potential impact of AMD-related lesions on retinal structure took place. With the use of OCT segmentation analysis, it was found that for both the CFP and OCT-based classification systems, the RPE, outer retina and photoreceptor thicknesses can be good indicators to distinguish AMD from control individuals. Additionally, factors that related with the structure of the layers of interest, in the eccentricities assessed, were age, gender, spherical equivalent and smoking. Finally, the results of this chapter agree with previously published work stating that the total retina is thicker in AMD individuals in the foveal area and that the RPE is thicker at all eccentricities assessed. The volume of the RPE is also a reliable metric to distinguish those with AMD from controls as it was found to increase from one AMD stage to another.
In chapter 5, the impact of the AMD-related lesions on visual function and the sensitivity of each test within the NISA test battery, to detect results outside a reference range of control participants were assessed. It was found in this chapter that only some of the tests could serve as good visual function metrics and distinguish AMD individuals from their counterparts in normal macular health, including low luminance visual acuity, low luminance deficit, contrast sensitivity, mesopic microperimetry and rod-intercept time. These tests were found to be able to make this distinction in both the CFP and OCT-based analysis. Finally, these were the tests that showed increased sensitivity as the percentage of patients falling outside the reference range, increased while moving from one group to another.
In chapter 6, the OCT groups created initially, were reclassified based on the previous results into a graded severity scale. In this chapter it was found that the structure of the RPE and photoreceptors is significantly affected and that this significance, is increased while moving from one group to another. The RPE and photoreceptor thicknesses, can also still distinguish AMD from control individuals. In terms of the effect on visual function, the results of this chapter, suggest that visual performance declined while moving from one OCT group to another. This was more profound for low luminance visual acuity, low luminance deficit, contrast sensitivity, mesopic microperimetry and rod-intercept time. Finally, these tests are those that can reliably distinguish AMD individuals from controls.
Chapter 7 concludes this thesis with a general discussion of the overall findings and makes suggestions for future directions of the research in general and how specific findings of the NISA study could be further extended.
|Date of Award||Jul 2020|
|Supervisor||Usha Chakravarthy (Supervisor) & Ruth Hogg (Supervisor)|
- Age-related macular degeneration
- Optical coherence tomography
- visual function
- structure-function relationship