Objective: Reliable identification of cognitive impairment in hemodialysis patients is of utmost importance, as it is associated with poor outcomes including dialysis withdrawal and death. High prevalence of cognitive impairment has been demonstrated in several studies using brief screening instruments or neuropsychological test batteries. However, the relevance of cognitive impairment as well as the accuracy of screening procedures have never been studied in this patient population.
Methods: 151 chronic hemodialysis patients (mean age 65.78 ± 14.88 years, 73,5% male) underwent cognitive testing under standardized conditions by the Montreal Cognitive Assessment (MoCA) and, in a second step, the Clinical Dementia Rating scale (CDR), an international standard to measure the severity of dementia. For calculating MoCA cut-off values on the basis of the CDR global score, receiver operator characteristics (ROC) analysis and c-statistic were applied.
Results: 49.0% of patients were categorized as 0.5 in the CDR global with memory being the predominantly affected domain (47.7% of patients scored ≥ 0.5). Youden's Index led to a threshold of 23.5 points for the MoCA test for optimal differentiation between cognitively normal (CDR global < 0.5) and impaired patients (CDR global ≥ 0.5) based on a sensitivity of approximately 99% and a specificity of approximately 74%.
Conclusion: Interference of cognitive impairment with patients' independence and daily life was shown using the CDR for the first time in hemodialysis patients. A MoCA score of 23.5 points turned out as optimal threshold to differentiate between patients with and without functional impairment in the CDR, thereby paving the way for implementation of the MoCA test as a quick and thus highly feasible screening instrument for periodic testing in clinical routine.