Use of amyloid PET scans in the clinical management of patients with cognitive decline in a naturalistic setting at a tertiary memory disorders clinic

Abstract

AbstractBackgroundAmyloid PET(aPET) is used to aid differential diagnosis, increase diagnostic confidence and assist with management of patients with cognitive decline. In the United States, aPETs are not often used because they are costly and not covered by insurance. VA Boston Healthcare System provides a unique opportunity where aPETs are available for clinical purposes and are ordered to assist in the differential diagnosis of patients with cognitive decline when clinically relevant. This study aims to determine which factors are associated with the use of aPET for diagnostic confirmation of AD in a naturalistic setting at a tertiary memory disorders clinic and to determine which factors predict an abnormal result.MethodWe conducted a retrospective review of all patients that were evaluated for cognitive decline at the Memory Disorders Clinic at Boston VA Medical Center from 9/2016 to 1/2020. Clinical information was extracted, including demographics, neuropsychological testing, and imaging studies. We compared patients' characteristics for whom an aPET was ordered and those for who a diagnosis was made without an aPET. We reviewed the characteristics of patients with positive vs. negative aPETResult565 patients were evaluated. 202 patients (35.75%) had an aPET as part of their workup; of these, 72 patients (36.5%) had an abnormal result. Factors associated with clinicians ordering the aPET were: younger age, longer clinical course, absence of amyloid angiopathy on MRI, and a diagnosis of mild cognitive impairment (MCI) and mild dementia. When performing multivariate analysis of the clinical characteristics, imaging, and neuropsychological tests of patients, only a positive family history of dementia (p=0.004) and presence of false‐positives on recognition in the CERAD list learning task (p=0.004) were associated with an abnormal result.ConclusionIn this cohort, aPETs were ordered in younger patients who fulfilled the criteria for MCI and mild dementia, and had a longer clinical course. Only a family history of dementia and false‐positives on recognition during a list learning task were significantly associated with an abnormal result. The results of this study may be informative both regarding how aPET may be used if covered by insurance and regarding clues to determining underlying Alzheimer’s pathology without biomarkers.