Premium
P2‐279: COGNITIVE AND FUNCTIONAL DECLINE DURING USUAL CARE OF PATIENTS WITH MILD ALZHEIMER’S DISEASE DEMENTIA WITH AND WITHOUT DIABETES
Author(s) -
Lebrec Jeremie,
Chen Yun-Fei,
Ascher-Svanum Haya,
Reed Catherine C.,
Kahle-Wrobleski Kristin,
Hake Ann Marie,
Raskin Joel,
Naderali Ebrahim,
Schuster Dara P.,
Heine Robert J.
Publication year - 2016
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2016.06.1539
Subject(s) - dementia , medicine , diabetes mellitus , post hoc analysis , analysis of covariance , cognitive decline , confidence interval , activities of daily living , cognition , observational study , repeated measures design , alzheimer's disease , physical therapy , disease , psychiatry , endocrinology , statistics , mathematics , machine learning , computer science
Background:Multiple neuropathologies are common in autopsied older adults. How co-occurring neuropathologies may affect disease progression is unclear. We estimated rates of progression in a longitudinal data set and tested whether clinical progression associated with Alzheimer’s disease neuropathologic change (ADNC) was modified by Lewy bodies (LB) or vascular brain injury (VBI). Methods:Data came from the National Alzheimer’s Coordinating Center on 2,046 autopsied participants with a clinical evaluation at a U.S. NIA funded Alzheimer’s Disease Center within 2 years of death. Participants with frontotemporal lobar degeneration and other rare diseases were excluded. Linear mixed effects models evaluated longitudinal trends in the Clinical Dementia Rating “sum of boxes” (CDR-SB) and associations with ADNC (Braak III or higher and CERAD moderate or frequent neuritic plaques), LB (any location), and VBI (gross or microscopic infarcts), and the interactions of ADNC with LB and VBI, separately. Models included adjustment for age, sex, race, education, and interval between last visit and death, as well as inverse probability of autopsy weights to adjust for potential bias due to autopsy selection. Results:Most subjects were demented at last visit (77.6%) and had ADNC (70.4%). Co-occurring LB (26.5%) and VBI (20.9%) were common. The annual change in CDR-SB (higher score1⁄4worse impairment) for those with ADNC only (1.7 points, 95%CI: 1.6,1.8) was slightly slower compared to those with ADNC+LB (1.9 points, 95%CI: 1.7,2.0) but slightly higher than those with ADNC +VBI (1.5 points, 95%CI: 1.3,1.6). Interestingly, there were significant negative interactions of ADNC with LB (p1⁄40.002) and VBI (p1⁄40.002), such that the rate of progression was significantly slower in those with dual neuropathologies than if each neuropathology contributed independently to progression. In secondary models with semi-quantitative categorization of ADNC, there were significant negative interactions of high ADNC (Braak V/VI + moderate/frequent plaques) with LB (p1⁄40.002) and VBI (p1⁄40.01), but not intermediate ADNC (Braak Stage III/IV + moderate/frequent plaques) with LB (p1⁄40.25) or VBI (p1⁄40.18). Conclusions: Our results suggest that considering interactions is important in characterizing clinical progression associated with neuropathologies. The impact of co-occurring pathologies on progression may also depend on severity of ADNC, although there are other possible explanations for our findings.