Olfactory Dysfunction in the Elderly: Basic Circuitry and Alterations with Normal Aging and Alzheimer's Disease

With the increased efforts on developing therapies for Alzheimer's disease (AD), it has also become imperative to develop methods to detect AD at early or even preclinical stages at which potential treatments may be most effective. Smell (or olfactory) testing could provide a simple, noninvasive way to accomplish this. In this Current Geriatrics Reports review, we summarize the behavioral and pathophysiological evidence, including work by Dr. Devanand's group, that olfactory pathways are affected during Alzheimer's disease, even at early stages, and differentiate these findings from those found with normal aging. Both human and rodent model studies are presented. We also identify key questions that should be the focus of future work.

During aging, in both humans and rodent, there is a decline in odor sensitivity, meaning a higher concentration of odor is needed for reliable detection. This correlates with a decline in olfactory sensory neurons in the nasal epithelium that first interact with odors. Other brain areas involved in odor processing also change with age, but how this contributes to behavioral deficits in aging or susceptibility to olfactory decline in AD is not clear.

In Alzheimer's disease, odor identification is more specifically affected, and decline in this has been shown to correlate with conversion from cognitively normal to mild cognitive impairment (MCI), and MCI to dementia. Dysfunction in piriform (olfactory) cortex and the medial temporal lobe (hippocampus, entorhinal cortex) appear to be most related to these deficits. Primary olfactory structures such as the olfactory bulb can also show evidence of neurofibrillary tangle pathology at very early stages of disease. In rodent models of AD, a sequence of olfactory deficits also progresses with spread of amyloid pathology in the brain, but this does not correlate well with findings in humans. Work in rodent models has also been limited by few studies on the role of tangle pathology, which appears to be important in human olfactory deficits in AD.

We suggest that future work should aim to better correlate the location and spread of AD pathology with progression of olfactory deficits, for example with amyloid imaging in humans. While awaiting the development of tau PET ligands for humans, we also recommend further studies on the role of tau pathology in olfactory pathways in rodent models. Furthermore, larger scale investigations on the predictive utility of smell testing in normal at-risk individuals should also be pursued. Research in this area is critical not only for the development of preclinical detection methods, but also to our basic understanding of Alzheimer's disease.

Arjun V. Masurkar MD, PhD
avm2114@columbia.edu

Davangere P. Devanand, MBBS, MD
dpd3@columbia.edu