Abstract: TH-OR44
ANU-ADRI, BDSI, CAIDE, and LIBRA as Predictors of Mild Cognitive Impairment Dementia in CKD
Session Information
- CKD: Novel Risk Factors and Consequences
October 24, 2024 | Location: Room 24, Convention Center
Abstract Time: 05:40 PM - 05:50 PM
Category: CKD (Non-Dialysis)
- 2302 CKD (Non-Dialysis): Clinical, Outcomes, and Trials
Authors
- Sarwal, Amara, University of Utah Health, Salt Lake City, Utah, United States
- Boucher, Robert E., University of Utah Health, Salt Lake City, Utah, United States
- Hartsell, Sydney Elizabeth, University of Utah Health, Salt Lake City, Utah, United States
- Wei, Guo, University of Utah Health, Salt Lake City, Utah, United States
- Singh, Ravinder, University of Utah Health, Salt Lake City, Utah, United States
- Katkam, Niharika, University of Utah Health, Salt Lake City, Utah, United States
- Takyi, Augustine, University of Utah Health, Salt Lake City, Utah, United States
- Chakravartula, Akhil Ramanujam, University of Utah Health, Salt Lake City, Utah, United States
- Moghaddam, Farahnaz Akrami, University of Utah Health, Salt Lake City, Utah, United States
- Beddhu, Srinivasan, University of Utah Health, Salt Lake City, Utah, United States
Background
The four commonly used risk prediction scores for dementia in the general population include Australian National University Alzheimer's Disease Risk Index (ANU-ADRI), Brief Dementia Screening Indicator (BDSI), Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE), and LIfestyle for BRAin Health (LIBRA). There is a paucity of data on the applicability of these scores for predicting cognitive function decline in CKD.
Methods
In Systolic Blood Pressure Intervention Trial (SPRINT)- MIND Study participants with CKD at baseline (N=2105), we compared the predictive ability of age alone, ANU-ADRI, BDSI, CAIDE and LIBRA scores for adjudicated incident mild cognitive impairment (MCI) or probable dementia (PD) in separate multivariate Cox regression models adjusted for BP group in SPRINT, age, sex, race/ethnicity, education, comorbidities such as CVD, PVD, depression and tobacco use as well as BP, BMI and eGFR.
Results
The mean age was 72 ± 9 years, 40% female and 37% African American. Mean eGFR was 47 ± 10. There was a total of 418 MCI events/9156 person years, and 137 PD events/9649 person years. As shown in Table, age had the strongest association with MCI but with modest c-statistic (0.66). Age, ANU-ADRI and BDSI were strong predictors of PD (Table and Figure) with c-statistic of 0.71, 0.74 and 0.73, respectively. CAIDE and LIBRA fared poorly for both MCI and PD.
Conclusion
ANU-ADRI and BDSI offer modestly higher concordance for predicting PD than age alone in CKD. Both age and the existing scores are poor predictors of MCI in CKD. Better tools to predict MCI in CKD are warranted.
| MCI | PD | |||
| HR per 1 SD | C-statistic | HR per 1 SD | C-statistic | |
| Age | 1.75 (1.51, 2.03) | 0.66 | 2.55 (1.96, 3.32) | 0.71 |
| ANU-ADRI | 1.10 (0.83, 1.46) | 0.65 | 2.45 (1.48, 4.07) | 0.74 |
| BDSI | 1.23 (1.00, 1.51) | 0.66 | 1.80 (1.23, 2.62) | 0.73 |
| CAIDE | 1.10 (0.96, 1.27) | 0.51 | 1.19 (0.94, 1.50) | 0.54 |
| LIBRA | 1.01 (0.84, 1.21) | 0.52 | 1.48 (1.08, 2.04) | 0.51 |
Funding
- NIDDK Support