Optic Disc Cupping and Cognitive Decline: Three-year Prospective Study Demonstrates Association
Abstract & Commentary
By Marc Dinkin, MD
Assistant Professor of Ophthalmology, Weill Cornell Medical College
Dr. Dinkin reports no financial relationships relevant to this field of study. This article originally appeared in the January 2014 issue of Neurology Alert.
Synopsis: This long-term epidemiological study shows an association between glaucoma and Alzheimer's disease. Cup to disc ratio, but not elevated intraocular pressure, is a risk factor for dementia.
Source: Helmer C, et al. Is there a link between open-angle glaucoma and dementia: The three-cityAlienor cohort. Ann Neurol 2013;74:171-179.
From a neurologist's perspective, glaucoma is the mir-ror image of pseudotumor cerebri, with both diseases leading to optic atrophy due to pressure on one side of the optic nerve head. However, since neuronal degeneration in glaucoma typically occurs over decades, and may occur even in the absence of elevation in intraocular pressure (IOP), the disease is perhaps most similar to Alzheimer's disease (AD). Indeed, multiple studies have shown areas of overlap between the two diseases — diffusion tensor imaging has shown a loss of cortical white matter along the optic radiations in glaucoma patients,1 while optical coherence tomography (OCT) has demonstrated retinal nerve fiber layer (RNFL) thinning in AD patients.2
The evidence linking one disease to the other has so far been conflicting, with two case-control studies demonstrating an increased risk of glaucoma in AD patients,3,4 while several studies based on review of hospitalization records were negative.5 The case-control studies were problematic in that the diagnosis of glaucoma could have been artifactually increased due to visual field defects secondary to the posterior cortical atrophy that may occur in some AD patients, while the negative studies, which depended on hospitalization records, may have missed either the diagnosis of AD, glaucoma, or both in some patients, since the rate of missed diagnosis of these conditions in the inpatient setting is high.
Helmer and colleagues aimed to settle the issue, with a prospective, epidemiological study, following a group of patients from one French city over a three-year period. Unlike prior studies addressing the AD-glaucoma link, patients were all actively screened for both AD and glaucoma at baseline and then again at 3 years. The results were impressive, with 17.5% of the 29 patients who developed AD at some point over the 3 years also receiving a diagnosis of glaucoma vs only 4.5% of the non-demented patients. Additionally, those patients who began the study with a diagnosis of glaucoma were four times as likely to develop AD, even when adjusting for age, sex, family history of glaucoma, education, and apolipoprotein ε4 status. Interestingly, an increased vertical cup to disc ratio (vCDR) of ≥ 0.65 and a small minimal rim to disc ratio (mRTDR), both predisposed to the development of dementia, while elevated IOP ≥ 21 and taking IOP-lowering drops did not.
This appears to be the first study in which a large cohort of patients were actively screened for both glaucoma and AD over several years, in search of an association. The results suggest that similar vulnerabilities of neuronal populations, perhaps determined by genetics or prior environmental exposures, may underlie the neurodegeneration in both glaucoma and AD, with the identity of the degenerating cell population determined by a secondary factor, such as elevated IOP in glaucoma patients. The finding that markers of neurodegeneration (such as a large vCDR) correlated with the risk of dementia, while elevated IOP did not, is consistent with that hypothesis. The lack of correlation of interocular cup to disc asymmetry (≥ 0.2) with dementia also supports a link through a factor predisposing to neurodegeneration, as cases with monocular or asymmetric cupping are more likely to result from a specific ocular factor.
A second interpretation of the results is that one disease actively predisposes to the other. AD, for example, has been shown to be associated with a low cerebrospinal fluid pressure,6 which in turn would cause a higher trans-laminar pressure differential across the optic nerve head. It may be that this difference in pressure (whether it results from high IOP or low intracranial pressure) predisposes to glaucoma.
The study is limited by the relatively low number of patients diagnosed with glaucoma in the cohort, and also by the lack of OCT data to corroborate and quantify the degree of glaucoma. Furthermore, although lumbar puncture is not necessary for the diagnosis of AD, it would have been interesting to measure the opening pressure in the AD patients to see if low measurements correlated with the incidence of glaucoma. Cerebrospinal fluid analysis for AD biomarkers such as β-amyloid1-42 (Aβ1-42), total tau protein (T-tau), and hyperphosphorylated tau would have been useful as well, as an association with glaucoma would have pointed to their role in both diseases.
While one pathological study that showed retinal degeneration in postmortem retinas of AD patients failed to show any neuritic plaques,7 more recently curcumin angiography has allowed in vivo visualization of retinal Aβ plaques in transgenic mouse models of AD,8 suggesting a primary degeneration of retinal neurons that could include the ganglion cells. This has been corroborated by OCT studies that have shown increased atrophy of the RNFL in AD patients. In light of this, it is possible that the results of the current study were confounded by a prior misdiagnosis of glaucoma based on AD-associated structural changes in the optic disc, especially since the definition of glaucoma in this study did not require an elevation in IOP. Of course, an alternative interpretation is that the association between glaucoma and AD demonstrated in this study explains the RNFL thinning previously demonstrated in AD patients.
Despite many remaining questions, this study makes an impressive statement on a powerful relationship between two of the most common neurodegenerative diseases in the world. The findings have significant implications for population screening of both diseases, and may even be relevant to future avenues of therapy. Further research will hopefully illuminate the cause and effect relationships subserving the observed association in this study, bringing us closer to a true understanding of both of these potentially devastating neurological conditions.
- Michelson G, et al. Graefes Arch Clin Exp Ophthalmol 2013;251:243-253.
- Kirbas S, et al. J Neuroophthalmol 2013;33:58-61.
- Bayer AU, et al. Eur Neurol 2002;47:165-168.
- Tamura H, et al. J Neurol Sci 2006;246:79-83.
- Kessing LV, et al. J Glaucoma 2007;16:47-51.
- Wostyn P, et al. Cerebrospinal Fluid Res 2009;6:14.
- Blanks JC, et al. Brain Res 1989;501:364-372.
- Koronyo Y, et al. Neurodegener Dis 2012;10:285-293.