By Makoto Ishii, MD, PhD

Assistant Professor of Neuroscience and Neurology, Feil Family Brain and Mind Research Institute, Department of Neurology, Weill Cornell Medical College

Dr. Ishii reports no financial relationships relevant to this field of study.

SYNOPSIS: In this cross-sectional, population-based, cohort study of 829 cognitively normal participants, abnormal neuroimaging biomarkers known to be associated with Alzheimer’s disease were associated with severe olfactory impairment.

SOURCE: Vassilaki M, Christianson TJ, Mielke MM, et al. Neuroimaging biomarkers and impaired olfaction in cognitively normal individuals. Ann Neurol 2017;81:871-882.

Since the pathological abnormalities that eventually lead to Alzheimer’s disease (AD) begin decades prior to the cognitive decline during the preclinical stage of AD, identifying cognitively normal individuals with the highest risk for developing AD is essential for the success of AD prevention trials. Currently, biomarkers used to identify those individuals are too expensive, invasive, or not widely available. Therefore, there is a need for an inexpensive, noninvasive test to screen and identify older healthy individuals potentially at risk for AD. Recent studies have found an association between olfactory impairment and cognitive decline, including mild cognitive impairment and dementia from AD. Importantly, in cognitively normal older individuals, worse odor identification has been associated with AD markers of brain pathology, suggesting that olfactory impairment could be a symptom of preclinical AD. In this study, the investigators tested the hypothesis that in vivo neuroimaging biomarkers of AD pathology are associated with olfactory impairment among cognitively normal older individuals.

Participants of this cross-sectional study were from the Mayo Clinic Study of Aging, an ongoing population-based study initiated on Oct. 1, 2004. Between 2004 and 2010, 829 cognitively normal participants were recruited and completed an in-person evaluation and the Brief Smell Identification Test (B-SIT) for assessment of olfactory impairment, and underwent magnetic resonance imaging (MRI) to evaluate AD signature cortical thickness (i.e., averaging cortical thickness for entorhinal, inferior temporal, middle temporal, and fusiform cortices) and hippocampal volumes. Of these participants, 306 had 11C-Pittsburgh compound B (11C-PiB) positron emission tomography (PET) scans for evaluation of brain amyloid load, and 305 had 18fluorodeoxyglucose (18F-FDG) PET scans for evaluation of brain glucose metabolism.

Of the 829 participants (mean age = 79.2 years, 51.5% men), anosmia was identified in 78 (9.4%) participants, hyposmia was identified in 503 (60.7%) participants, and the rest (248, 29.9%) had normal olfactory sensation (normosmia). Abnormal cortical thickness was found in 249 (30%) participants and abnormal hippocampal volume in 128 (15.5%) participants. For those participants who underwent PET scans, 119 (38.9%) participants had abnormal 11C-PiB PET scans, and 90 (29.5%) participants had abnormal 18F-FDG PET scans. In addition, 51 (16.9%) participants were positive for both amyloid-beta and neurodegeneration biomarkers. Abnormalities in brain amyloid accumulation, cortical thickness, and hippocampal volumes increased in frequency among those participants with olfactory impairment compared to those without olfactory impairment.

After adjusting for age, sex, and education, individuals with abnormal 11C-PiB PET scans, abnormal cortical thickness, and abnormal hippocampal volumes had significantly increased odds of having anosmia compared to normosmia (odds ratio [OR] = 2.74, 95% confidence interval [CI] = 1.12-6.66; OR = 2.20, 95% CI = 1.25-3.86; OR = 2.45, 95% CI = 1.21-4.94, respectively). None of the biomarkers were associated significantly with hyposmia. Abnormal cortical thickness and abnormal hippocampal volumes were significantly associated with lower B-SIT score (slope = -0.43, 95% CI = -0.76 to -0.09, P = 0.01; slope = 0.72, 95% CI = -1.15 to -0.28, P < 0.01, respectively). When participants were separated between those who had abnormal 11C-PiB or amyloid abnormalities and those who did not, abnormal hippocampal volume was associated with B-SIT score only in those with abnormal 11C-PiB PET scans. Furthermore, having both amyloid and neurodegeneration abnormalities was associated with a 3.84-fold increased odds of having anosmia (OR, 3.84; 95% CI, 1.14-12.97; P = 0.03) and significantly lower B-SIT score (slope, -0.99; 95% CI, -1.71 to -0.27) compared to those negative for both markers.

Finally, MRI and 18F-FDG PET ROI measurements representing areas of primary olfactory cortex and secondary olfactory regions were significantly associated with B-SIT score. There were no statistically significant associations between any of the 11C-PiB PET ROIs and B-SIT score.


These study results are intriguing and add to our expanding knowledge of the non-cognitive manifestations of AD, particularly in the preclinical stage. Although this study has significant strengths, including a relatively large sample size obtained from a well-characterized study population and the use of well-established multimodal neuroimaging markers of AD, there are notable limitations to this study, including a lack of ethnic and racial diversity in the study population and the cross-sectional study design, which precludes the ability to assess causality. If the results from this study can be verified in other populations and, importantly, in longitudinal studies, olfactory function could be used potentially as a noninvasive and inexpensive screen to identify those individuals at risk for AD. Because of the high number of older individuals with olfactory impairment, it is unlikely that screening olfactory function alone would be sufficient to identify accurately all those at risk. However, combining an initial inexpensive and noninvasive screen (e.g., olfactory function) followed by selectively screening with more sensitive and specific AD biomarkers (e.g., neuroimaging) potentially could increase the ability to identify individuals who have the highest risk of developing AD.