Eye Tests May Predict Alzheimer's Risk

Deborah Brauser

March 01, 2013

Abnormalities in retinal vascular parameters (RVPs) may indicate increased amyloid plaque in the brain and can serve as biomarkers for preclinical Alzheimer's disease (AD), new research suggests.

Findings from the Australian Imaging, Biomarkers, and Lifestyle (AIBL) Flagship Study of Ageing showed that participants with AD had several significantly different RVPs, including narrower veins and a significantly higher arteriole-to-venule ratio (AVR), than their peers without AD.

In additional analysis of just the participants without an AD diagnosis, Pittsburgh compound B positron emission tomography (PiB-PET) imaging showed that participants who had high neocortical plaque burden also had some of these RVP changes, possibly representing preclinical AD.

"This is the first study to investigate retinal blood vessel changes with respect to amyloid plaque burden in the brain," write the investigative team, led by Shaun Frost, a PhD candidate at the Commonwealth Scientific and Industrial Research Organization in Perth, Australia.

"Models combining RVPs perform well at distinguishing diagnosed AD patients from healthy controls," they add.

The researchers note that because PET and magnetic resonance imaging scans can be expensive to administer and are not widely available, these eye tests may represent a simpler, noninvasive option — although not as a stand-alone measure.

"Retinal photographic analysis shows potential as an adjunct for early detection of AD or monitoring of AD-progression or response to treatments," they write.

The study was published online February 26 in Translational Psychiatry.

Difficult Diagnosis

"The primary neuropathological hallmark of [AD] is the presence of cerebral amyloid deposits (plaques)," write the investigators.

"Although post-mortem examination of the brain is required for confirmation of AD, a diagnosis of 'probable AD' can be made in patients, fulfilling the criteria set down by the National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer's Disease and Related Disorders Association," they add.

However, a conclusive diagnosis of probable AD only comes after neurologic damage has already occurred in these patients. Therefore, there is a great need for ways to detect the condition before irreversible damage occurs.

Because the retina "is a developmental outgrowth of the brain," the investigators sought to assess whether pathologic retinal vascular changes could be used as a potential screening measure for AD.

For this analysis, the investigators evaluated a cohort of the AIBL study participants, consisting of 25 with probable AD (52% women; mean age, 72.4 years) and 123 of their healthy peers (55% women; mean age, 71.6 years).

Digital retinal color photographs for these individuals were taken, and semiautomatic software was used to evaluate for 19 RVPs, such as width and branching of retina vessels. In addition, AVR was calculated for each participant on the basis of central retinal arterial to venular equivalent thickness.

Additional analysis of RVPs in relation to neocortical plaque burden, as shown with PiB-PET scans, was conducted for 45 of the healthy peers.

Significant Abnormalities

Results showed significant between-group differences in 13 of the RVPs measured.

These included lower central retinal arteriolar and venular equivalent calibers (P = .01 and P < .001, respectively), fractal dimension of both the arteriolar and venular networks (P = .008 and P < .001, respectively), venular curvature tortuosity (P = .02), and number of first branching arterioles and venules (P = .007 and .006, respectively) for those with probable AD vs those without.

In addition, the participants with probable AD had significantly elevated zone B arteriolar and venular standard deviations (P = .001 and .002, respectively), venular branching coefficient (P = .02), arteriolar and venular branching asymmetry factor (P = .02 and .03, respectively), and arteriolar length-to-diameter ratio (P = .03).

In the imaging data analysis, the subgroup of healthy volunteers with high levels of amyloid plaque (as shown by PET-PiB-measured standardized uptake value ratio greater than 1.5) had higher venular branching asymmetry factor (P = .01) and arteriolar length-to-diameter ratio (P = .02) compared with those who had lower levels of amyloid plaque.

These 2 parameters were also larger in patients with AD compared with those with high levels of amyloid plaque, "hence these results are consistent with the hypothesis that RVP changes may precede AD diagnosis," write the study authors.

There were also significantly more APOE ε4 carriers in the AD group vs the non-AD group (P = .02) and in those with high levels of amyloid plaque vs those with lower levels (P = .04).

An Even Better Biomarker?

Dr. Gilbert Feke

"I think this is a good study, but I wish they had looked at another subgroup of patients," Gilbert T. Feke, PhD, senior research associate in the Department of Ophthalmology at Massachusetts Eye and Ear Infirmary in Boston, told Medscape Medical News.

"They looked at control subjects and those with Alzheimer's, but they didn't have an in-between group."

Dr. Feke, who was not involved with this research, presented preliminary findings from a similar study in a poster presentation at the 2011 Annual Meeting of the Association for Research in Vision and Ophthalmology (ARVO).

His investigative team examined 7 elderly participants with probable AD, 10 with mild cognitive impairment (MCI), and 17 with normal cognition.

Results showed that the AD group had significantly narrower retinal veins than the other 2 groups, as well as decreased retinal blood flow. In addition, the MCI group had significantly lower blood flow and blood speed than those with normal cognition.

"This interim report suggests that hemodynamic abnormalities may precede neural loss in the retina in cognitively impaired subjects," said the researchers in the presentation.

"We found that decrease in vessel diameter doesn't happen until people actually have [AD]. Those with MCI didn't show it. Instead, they showed the reduction in blood speed and flow. So that puts a bit of a damper on relying on vessel diameter measurements to be the sole biomarker," added Dr. Feke.

He noted that diameters in his study and in the current study were measured when the retinal blood vessels were rather large. However, the AD process "starts in the tiniest vessels. And when things happen in those vessels, that slows down blood speed."

"There aren't a whole lot of people pursuing the idea of examining the retina in this way. But the idea itself is excellent because the retina is part of the brain," he concluded.

The AIBL study was funded by a grant from the National Institute for Mental Health, and portions were presented at the 2011 Alzheimer's Association International Conference in Paris, France. Dr. Zeke's research, abstract 2132/poster A102 , was presented at ARVO in Fort Lauderdale, Florida, May 2, 2011. The study authors and Dr. Feke have reported no relevant financial relationships.

Tansl Psychiatry. Published online February 26, 2013. Full article