Towards optimizing the diagnosis of Lewy body dementia: Lessons from the NACC

Abstract

Introduction: The diverse presentations and co-pathologies of Lewy body dementia (LBD) present a diagnostic challenge. Utilizing the National Alzheimer's Coordinating Center (NACC) dataset, this study aimed to evaluate the concordance of premorbid diagnoses of LBD in individuals with autopsy-confirmed neocortical Lewy body disease.

Methods: Demographics, clinical and neuropsychological presentations, Lewy body pathology, and Alzheimer's disease (AD) co-pathology were related to clinical diagnosis.

Results: Diagnosis of LBD had high specificity but low sensitivity, with fewer than half of autopsy-confirmed cases having received an LBD diagnosis. AD was the most common misdiagnosis. Participants diagnosed with AD were more likely to be female, to have a more amnestic phenotype, and to harbor a higher burden of AD co-pathology, and were less likely to have documented clinical features characteristic of LBD.

Discussion: These results highlight the need to improve LBD diagnosis in research and clinical settings, a need that emerging biomarkers may help address.

Highlights: Less than half of cases with neocortical Lewy bodies were diagnosed with Lewy body dementia (LBD) Alzheimer's disease (AD) was the most common misdiagnosis Cases misdiagnosed were more likely to be female Cases diagnosed with AD presented with a more amnestic phenotype Cases diagnosed with AD had more AD co-pathology.

Keywords: LBD; Lewy bodies; Lewy body dementia; NACC; National Alzheimer's Disease Coordinating Center.

FIGURE 1

Clinical characteristics of participants with autopsy‐confirmed neocortical Lewy body disease: The percentage of cases with each clinical characteristic is shown for three diagnostic groups: primary Lewy body dementia (LBD), primary Alzheimer's disease (AD) with contributing LBD (AD with LBD), and primary AD without contributing LBD (AD no LBD). Numbers of cases with available data for each test are indicated. * denotes statistically significant difference in diagnosis based on presence or absence of the clinical characteristic (* p < 0.0167 [Bonferroni adjusted p‐value], ** p < 0.001 *** p < 0.0001).

FIGURE 2

Neuropsychological presentations of participants with autopsy‐confirmed neocortical Lewy body disease: Neuropsychological battery test scores in participants with autopsy‐confirmed neocortical Lewy body pathology who were diagnosed with either primary Lewy body dementia (LBD), primary Alzheimer's disease (AD) with contributing LBD (AD with LBD), or primary AD without contributing LBD (AD no LBD) are shown. (A) MMSE (Mini‐Mental State Examination) reflects global cognition (of 30, lower score worse). (B) Logical Memory IIA Delayed Recall evaluates the memory domain (of 25, lower score worse). (C) Boston Naming Test evaluates the language domain (of 30, lower score worse). (D) Trail Making Test Part A (Trails A) evaluates attention (number of seconds taken to complete, up to 150, higher score worse). Numbers of cases with available data for each test are indicated. Box represents median and interquartile range; + represents mean, whiskers represent minimum and maximum. *p < 0.05 difference between the primary LBD and AD no LBD group.

FIGURE 3

Alzheimer's co‐pathology of participants with autopsy‐confirmed neocortical Lewy body disease: Percentage of cases with (A) CERAD scores of amyloid co‐pathology and (B) Braak stages of tau co‐pathology for the following diagnoses: primary Lewy body dementia (LBD), primary Alzheimer's disease (AD) with contributing LBD (AD with LBD), and AD without contributing LBD (AD no LBD)).