Urinary Incontinence in a Community-Based Autopsy Cohort Is Associated with Limbic Predominant Age-Related TDP-43 Encephalopathy Neuropathologic Changes

Abstract

Background: Dementia and urinary incontinence (UI) are etiologically complex clinical syndromes. Dementia and UI often occur in the same individuals, but underlying factors connecting them are incompletely understood.

Objective: Query data from a community-based autopsy series to assess pathologies that underlie UI.

Methods: Included research subjects came to autopsy from the University of Kentucky Alzheimer's Disease Research Center longitudinal cohort. A total of 368 research volunteers met inclusion criteria for this cross-sectional study. The average age at death was 85.3 years and the average number of annual clinic visits was 5.2 visits. Statistical models were run to evaluate which pathologies were associated with UI. Data included pathologies scored according to conventional stage-based systems, and these studies were complemented by quantitative digital neuropathology.

Results: Dementia was diagnosed at the final clinical visit in 208 (56.7% of the sample) and UI was documented in 156 (42.7%). UI was associated with depression and dementia (both p < 0.001). More women than men had a history of UI (p < 0.04), and women with UI had had more biological children than those without UI (p < 0.005). Participants with limbic predominant age-related TDP-43 encephalopathy neuropathologic changes (LATE-NC) were more likely to have UI than those without LATE-NC (p < 0.001). The presence of LATE-NC (Stage > 1) was associated with UI with or without severe Alzheimer's disease neuropathologic changes and/or Lewy body pathology.

Conclusion: In this community-based autopsy cohort, multiple factors were associated with UI, but the neuropathologic change most robustly associated with UI was LATE-NC.

Keywords: ADNC; ARTAG; ScanScope; clinical dementia rating; obstetric; oldest-old; sex; synuclein; urology.

Figure 1:

Final clinical visit Clinical Dementia Rating (CDR) score distributions, stratified by urinary incontinence (UI) status. CDR scores span 0-3 with 3 being most severely impaired. In these stacked-bar charts, the y-axis represents percentages of cases to convey the frequencies of various CDR scores. When dementia status was evaluated with any of the CDR rating domains, UI is associated with more severe dementia. Statistical comparisons were made using Extended Cochran–Armitage test. ***-P < 0.0001.

Figure 2:

Urinary Incontinence (UI): associations with neuropathology either without (CDR<1) or with (CDR >=1) dementia at the final clinic visit. In these analyses, neuropathologic features were dichotomized: LATE-NC Stages 0/1 vs 2/3 (A); Braak NFT stages 0-IV vs V/VI (B); and, neocortical Lewy bodies absent vs present (C). With the smaller sample sizes in each group, none of these comparisons were statistically significant using two-tailed Student t-tests.

Figure 3.. Analytic workflow schematic (A) for high-throughput quantitation of Alzheimer disease type neuropathologic changes, with examples of immunostains and analytic results for Aβ amyloid plaque (B, C) and pTau-immunoreactive neurofibrillary tangles (NFTs; D, E).

These methods were described in detail previously[44]. Digital analysis of Aβ amyloid plaques returned values of density per square millimeter via a modified nuclear algorithm. The raw immunohistochemical (IHC) results are shown in B and following analyses (C), amyloid plaques are false-colored yellow, orange, and red to convey the digital detection of individual amyloid plaques for quantitation purposes. Representative digital quantification of tau pathologies are depicted in (A). Analyses were performed on PHF-1immunostained sections (D). After using the crafted Genie NFT/NP algorithm to isolate the NFTs, the NFT density (NFTs per square millimeter) was determined by a modified nuclear algorithm, with NFTs here stained yellow (E). The same pTau-immunoreactive NFT recognized by the algorithm is indicated by a blue arrow in both D and E. Scale bar = 100 microns for B, C, and 150 microns for D, E.

Figure 4:

Urinary incontinence (UI): associations with quantitative Alzheimer’s disease neuropathologic changes (ADNC) assessed using digital neuropathologic evaluations in different brain regions. ADNC was operationalized by Aβ plaque (A) or pTau tangle (B) immunohistochemistry (IHC) using digital neuropathological methods that were previously described in detail [44]. There was a general tendency for both Aβ and pTau tangles to be higher in cases with UI across different anatomical regions. Looked at individually, some of the regions of interest showed nominally significant differences (UI+ vs UI− cases), where the ADNC was higher in UI cases than cases without UI. However, the statistical significance of individual regions (calculated with two-tailed Welch’s t-tests) did not survive at P<0.05 after correcting for multiple comparisons. Error bars indicate Standard error of the mean. Nominal p-values: * - P<0.05; ** - P<0.01.