By Kai Lyn Goh, Dr Carla J. Abbott and Professor Robyn H. Guymer. Shared with permission from the authors and first published for Mivision, May 2022 issue.
One in seven Australians aged 50 years or older have the early stages of AMD. Over five years, 15% of these individuals with the usually asymptomatic, early stages will develop late, vision-threatening complications including geographic atrophy (GA) and/or choroidal neovascularisation (CNV). Drusen, the hallmark feature of the early stages of AMD, are seen as yellow/white deposits on colour fundus photography (CFP) and as areas of retinal pigment epithelium (RPE) elevation on OCT. The utilisation of OCT, which allows for cross-sectional viewing of retinal structures, has enabled the characterisation of additional features of drusen, such as their internal reflectivity and homogeneity.
AMD is defined by the presence of drusen, and staged by its size, yet the term ‘drusen’ encompasses many different phenotypes, which may each confer unique risks for progression. Drusen phenotypes include cuticular drusen, hyporeflective cores within drusen (HCD), large colloid drusen, and pachydrusen. High-resolution OCT has also allowed for easier differentiation of conventional drusen, which form under the RPE, from reticular pseudodrusen (RPD), which are located above the RPE. RPD are independently associated with an increased risk of progression to late AMD in individuals with CNV in the fellow eye. These deposits are also associated with an increased risk of progression to late AMD in those with the early signs of AMD, but this relationship with disease progression has not been observed in all longitudinal studies.
Phenotyping drusen and other deposits such as RPD is important, especially with emerging treatments for earlier stages of AMD, as treatment efficacy may depend on particular disease phenotypes. In a recent interventional trial of subthreshold nanosecond laser aiming to slow progression of intermediate AMD (iAMD), the co-existence of RPD modified the response. In a post-hoc analysis, subthreshold nanosecond laser appeared to have a possible beneficial treatment effect for those without RPD, but not so for those with RPD in individuals with bilateral large drusen. This indicates that there may be different and specific interventions required, dependent upon the type of deposit present in an eye.
Cuticular drusen and HCD are two drusen phenotypes that are of great interest to both clinicians and researchers, and as such, have been the subject of two recent studies conducted by our group. In this article, we will describe how they can be identified and in light of our findings, their prognostic significance in individuals with iAMD.
Cuticular Drusen
What Are They And How Can They Be Identified?
Cuticular drusen were first described by Gass in 1977, who noted numerous, small, round, yellow nodules scattered throughout the fundus that were best visualised on fluorescein angiography, showing a ‘stars-in-the-sky’ hyperfluorescence appearance. They are considered to be an early-onset drusen phenotype and can be seen in individuals younger than 50-years that do not fit the usual definition of AMD.
On CFP, cuticular drusen appear as multiple yellow/pale, small, uniform, round deposits corresponding to a series of RPE elevations on OCT, often in a sawtooth configuration. There is occasionally thinning of the overlying RPE at the peak of each elevation, which is associated with hypertransmission of the OCT signal into the choroid, producing a characteristic ‘barcoding’ appearance (i.e., a series of small regions of signal hypertransmission alternating with normal transmission). On fundus autofluorescence, cuticular drusen have been reported to correspond to focal spots of hypoautofluorescence, which can be surrounded by a rim of hyperautofluorescence.
Why Are They Important And What Did We Find?
Cuticular drusen have been reported to be associated with a high prevalence of CNV, one of the main late complications of AMD. Boon et al reported that 56% of 198 individuals with cuticular drusen seen at their tertiary referral centre had CNV, and Cohen et al suggested that those with cuticular drusen may require “close follow up” due to the “high prevalence of CNV”. However, individuals were included in these studies based solely on the presence of cuticular drusen, regardless of the co-existence of conventional drusen. These studies have also included participants as young as 23-years-of-age, who would therefore not fall into the diagnosis of typical AMD. As such, based on previous studies, it is not possible to draw any conclusions about the progression risk to vision-threatening complications in individuals with AMD who also have cuticular drusen.
In our study of cuticular drusen in AMD, recently published in Ophthalmology, we found that 25% of the 140 individuals in our cohort had cuticular drusen at baseline, which were always present bilaterally. Over a three-year study period, 48 out of 280 eyes progressed to late AMD, when defined on OCT. The presence of cuticular drusen at baseline was not associated with the rate of progression to late AMD, both without adjustment for the confounders of age, drusen volume and the presence of pigmentary abnormalities, and also when these confounders were included in an adjusted analysis.
We also investigated the impact of cuticular drusen on visual function (as assessed using microperimetry) and found that the presence of cuticular drusen was not associated with lower visual sensitivity at baseline, nor with a faster rate of visual sensitivity decline prior to the development of late AMD.
Hyporeflective Cores Within Drusen
What Are They And How Can They Be Identified?
HCD is another drusen phenotype of interest. The term ‘hyporeflective core’ is used to describe the internal contents of a druse that is more hyporeflective than a typical druse (i.e. the core has less signal and thus appears more black). HCD can only be determined on cross-sectional OCT imaging, as their en face fundus appearance is often indistinguishable from drusen without hyporeflective cores.
Why Are They Important And What Did We Find?
Several previous studies have identified HCD or lesions including HCD as a risk factor for AMD progression, although different terminologies have been used to describe this feature. Their presence has been associated with an increased risk of developing local RPE atrophy and late AMD. However, even though this association has been established, it is still not known whether the presence of HCD is an independent risk factor for individuals with iAMD. This is because key risk factors for progression, including drusen volume, the presence of pigmentary abnormalities, and late AMD in the fellow eye, were not accounted for in prior analyses.
In our study of HCD, published in Ophthalmology Retina, we found that in a cohort of 280 eyes from 140 individuals, 20 (7%) eyes from 12 (9%) individuals had HCD present at baseline, which was associated with a non-significantly increased rate of progression to late AMD in unadjusted analysis. In contrast, the extent of HCD was significantly associated with an increased rate of progression to late AMD and lower visual sensitivity at baseline, in unadjusted analyses. However, these associations regarding the extent of HCD on progression and visual function, were no longer significant after adjusting for known risk factors for AMD progression (age, presence of pigmentary abnormalities and drusen volume). HCD extent was also not associated with a faster rate of visual sensitivity decline prior to the development of late AMD, with and without adjustment.
Prognostic Significance of Cuticular Drusen and Hyporeflective Cores
We found that in a cohort of individuals with bilateral large drusen, the presence of cuticular drusen and the presence/extent of HCD were not independently associated with progression to late AMD over three years. Additionally, the presence of cuticular drusen and the presence/extent of HCD were both not associated with baseline visual sensitivity, or the rate of visual sensitivity decline prior to the development of late AMD, after adjusting for key risk factors of disease progression (age, drusen volume and pigmentary abnormalities).
This indicates the importance of considering all known risk factors in an eye when assessing progression risk. The results from our cohort suggest that individuals with iAMD, who also have concomitant cuticular drusen or HCD, have a similar prognosis for progression and level of visual function as those with conventional drusen. Thus, there need not be a differentiation in their management in clinical practice. As outlined in the RANZCO Referral Pathway for AMD Management and Optometry Australia’s chairside reference for the diagnosis and management of age-related macular degeneration, these individuals with iAMD should be reviewed by an optometrist every six to 12 months depending on clinical risk modifiers, such as the presence of hyperreflective foci, nascent GA, RPD, and non-exudative macular neovascularisation. While the presence of cuticular drusen or HCD in individuals with AMD should not change optometric management, the presence of clinical risk modifiers (including other drusen phenotypes and deposits such as RPD) may. As such, it is imperative that clinicians monitoring patients with AMD familiarise themselves with various features of anatomical change seen on OCT and other imaging modalities.
