4,068 posters, 219 videos, 101 topics, 69 sessions, 14,365 authors, 6,128 institutions
ePostersLive by SciGen Technologies S.A. All rights reserved.
Alzheimer's Association International Conference 2026
July 12-15, 2026 | London, United Kingdom, and Online
10407
Association Of Allostatic Load With Cerebral Free Water In Hispanic/latino Adults: Findings From The Sol-inca-mri Study (hchs/sol)
Poster Presenter
Authors
Cesar Moreno, Addison Berg, Mayra Estrella, Pauline Maillard, Ariana Stickel, Carmen Isasi, Fernando Testai, Martha Daviglus, Tatjana Rundek, Linda Gallo, Hector Gonzalez, Charles DeCarli, Shraddha Sapkota
Affiliations
University of Miami, San Diego State University, University of California, San Diego, Albert Einstein College of Medicine, Univeristy of California, Davis, Rush Alzheimer’s Disease Center, Rush University Medical Center, University of Illinois, Chicago
Part of Topic
•Allostatic load (AL) reflects cumulative multisystem physiological dysregulation that may reflect exposure to chronic stress. AL has been linked to cognitive decline and is disproportionately elevated in Hispanic/Latino adults relative to other racial/ethnic groups1,2.
•Cerebral free water (FW), derived from magnetic resonance (MRI) diffusion tensor imaging (DTI), is a sensitive biomarker of neuroinflammation and cerebrovascular white matter damage3.
•Goal: To investigate whether AL is associated with FW and cognitive performance in Hispanic/Latino older adults. We had two research aims:
•Aim I: Examine the association between AL and FW (~10 years later) in the overall group and by Hispanic/Latino heritage. We expect that higher AL will be associated with greater FW and this association will differ by heritage groups.
•Aim II: Test cross-sectional associations between AL and cognitive performance at visit 1 (baseline) and visit 2 (~7 years later). We expect that higher AL will be associated with poorer cognitive performance at both visits.
•Participants:
•We used Hispanic/Latino adults from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL)- Investigation of Neurocognitive Aging-MRI (INCA-MRI) (N = 2,442, baseline age = 52.0 ± 8.9 years, 67% female; Table 1 ). Participants were from four HCHS/SOL field centers: Bronx, Chicago, Miami, and San Diego.
•Allostatic Load (AL):
•We used a count-based index of 16 biological markers summed across five physiological domains (cardiometabolic/lipid metabolism, glucose metabolism, cardiopulmonary, parasympathetic, inflammation) (Table 2 ).
•Following prior AL calculation, each of the 16 biological markers across the five domains received 1 point if it fell in the top quartile or the participant was on related medication, and the points were summed (range 0-16).
•Cerebral free water (FW):
•DTI was used to measure extracellular water content.
•FW is the average white matter voxels of water diffusing freely.
•Cognition:
•We used two neuropsychological tests to examine cognition at both visits4,5.
•Brief Spanish-English Verbal Learning Test (B-SEVLT; trials 1-3 [immediate recall] and delayed recall): Assesses episodic memory (EM). Immediate recall captures initial encoding and learning capacity and delayed recall evaluates memory consolidation and retention.
•Word Fluency Test (WF; Letter A and Letter F): Assesses working memory (WM) through phonemic verbal fluency by asking participants to name as many words as they can in 60 seconds starting with A or F.
•Statistical Analysis:
•Aim I: Multivariable regression analyses
•Model: AL -> FW in the overall group and by heritage6.
•Aim II: Confirmatory factor analysis to test for latent cognitive domain at visits 1 (V1) and 2 (V2).
•Path analysis of AL and latent cognitive domain at visits 1 and 2.
•All analyses were performed in R version 4.5 and Mplus version 8.67.
•Covariates: Models were adjusted for age, sex, education, income, heritage, field center, language, years in the US, cigarette use, alcohol use, physical activity, diet quality, depressive symptoms, and doctor visit frequency as a measure of fluctuating health and access to healthcare. Models with FW were also adjusted for DTI b-values.
•Aim I:
•Overall group: Higher AL was associated with greater FW ~10 years later (ß=0.0014, SE = 0.00022, p < .001; Figure 1 ).
•By heritage group: Higher AL was associated with greater FW ~10 years later in Mexican, Dominican, Central/South American, and Cuban adults, but not in Puerto Rican adults (see Table 3 ).
•Post-hoc analyses: Mean AL differed across heritage groups (p < .001) and was highest among Puerto Rican adults of the defined heritage groups (Figure 2 ); Tukey-HSD confirmed significantly elevated AL in Puerto Rican relative to Mexican, Dominican, and Central/South American adults. Notably, Puerto Rican adults showed the weakest AL-FW coupling despite carrying the greatest allostatic burden.
•The two-factor model (Figure 3 ) provided the best fit at V1 ( X2 (31) = 119.32, p < .001; RMSEA = .037 (90% CI .030-.045); CFI = .975; TLI = .948; SRMR = .017) and V2 (X2 (31) = 99.05, p < .001; RMSEA = .033 (90% CI .026-.040); CFI = .982; TLI = .962; SRMR = .015).
•Higher baseline AL was associated with poorer WM at V1 (ß = -0.026, SE = 0.010, p = .008) and V2 (ß = -0.040, SE = 0.010, p < .008).
•We did not observe significant associations of AL with EM V1 (ß = -0.002, SE = 0.009, p =.835) and V2 (ß = -0.011, SE = 0.009, p =.207).
•Higher AL was significantly associated with greater FW (~10 years later), suggesting that AL is a sensitive early marker that may capture upstream, stress-related mechanisms and potentially predict early vascular changes.
•AL was significantly associated with poorer WM but not EM, consistent with previous work2. AL may act through vascular and inflammatory pathways affecting frontal-subcortical white matter2 supporting our WM findings, whereas EM depends more on medial-temporal regions vulnerable to AD-specific pathology, such as amyloid and tau, that AL may not capture.
•Future work will leverage longitudinal data to model AL trajectories and test FW as a mediator of future WM and EM.
1.Kaestner et al., Social Science Quarterly, 90(5), 2009.
2.Estrella et al., Journal of Alzheimer's Disease, 99(3), 2024.
3.Maillard et al., Neurology, 92(19), 2019.
4.González et al., Archives of Clinical Neuropsychology, 30, 2015.
5.Sapkota et al., Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 16, 2024.
6.Salazar et al., SSM-Population Health, 2, 2016.
7.Muthén & Muthén, Mplus User's Guide (8th ed.), 2017.
We acknowledge all participants in this study and staff who support this effort. This study was supported by grants RF1 AG054548,R56 AG048642,R01 AG075758,R01 AG083894.
camoreno@health.ucdavis.edu
Comments
Email is required but is not shown publicly.
Loading comments...