#1490 Long-term health outcomes in non-dialysis dependent chronic kidney disease patients with cognitive impairment or dementia

Abstract

Abstract Background and Aims A decline in cognitive function, with a spectrum ranging from mild cognitive impairment to dementia (CI-D), is common in patients with chronic kidney diseases (CKD), rendering them frailer. Whether the coexistence of CKD and CI-D increases the health risk profile remains unclear. Our work aims to clarify how this association increases adverse outcomes for CKD patients (stages 3-5). Method This retrospective observational cohort study was conducted on CKD patients (stages 3-5) from the Global Collaborative Network (GCN) of the TriNetX research platform between 2004 and 2024. TriNetX provides access to anonymized electronic medical records from over half a million CKD patients in large healthcare organizations (HCOs) across the world. Our report includes patients from 110 HCOs with different demographic backgrounds. The study compared outcomes of patients with CKD stages 3-5, aged 18-70 years, with and without pre-existent CI-D (diagnosed within 5 years). The two cohorts were propensity score-matched (PSM) for age, sex, ethnicity, comorbidities (neurological, haematological, musculoskeletal, respiratory including smoking history, oncological, cardiovascular, gastroenterological and endocrine conditions) and CKD stage. The proportional hazard assumption was tested using the generalized Schoenfeld approach built in the TriNetX platform. A 95% confidence interval (95% CI) was considered evidence of statistical significance throughout the analyses. Kaplan- Meier (KM) was used for survival probability. Statistical significance was defined as p-value < 0.05. Outcome events were included from 1 day after the index event (CKD diagnosis) until 3650 days after. Results Of the total of 503,298 CKD patients, 7,891 had a co-existent diagnosis of CI-D made within 5 years (prevalence rate 1.56%). A PSM generated a matched cohort of 7,890 patients each. The mean age of the cohort was 60.7 ± 7.0 years, with a predominance of white ethnicity (53.7%) and an equal male-to-female ratio. Standardized difference (Std diff) was used to evaluate the balance of baseline characteristics in the PSM populations. Generally, Std diff < 0.1 is considered a small difference and all comorbidities fell below this threshold after PSM (Table 1). The comparison of the outcomes at follow-up between the matched cohorts is illustrated in Table 2. CKD patients with concomitant CI-D had higher all-cause mortality (18.4% vs 12.1%, p < 0.0001), a higher risk of cerebrovascular disease (10.8% vs 7.8%), malnutrition (6.7% vs 4.3%), mood disorders (13.7% vs 8.9%), anxiety (13.2% vs 10.9%), encephalopathy (9.6% vs 4.5%), epilepsy (4.3% vs 1.3%). All findings were statistically significant with a p < 0.0001. Proportional hazard models showed that CI-D was a strong risk factor associated with all-cause mortality (HR: 1.761; 95% CI (1.621-1.913, p < 0.0001)) and other cerebrovascular and mental health illnesses. Conclusion Our “real-world data” demonstrate that a concomitant CI-D diagnosis is a strong risk factor for negative health outcomes and mortality in patients with CKD stages 3-5. These results highlight the need for routine cognitive assessments in patients with advanced CKD to deliver personalised care.