Visuospatial and Executive Dysfunction in Patients With Acute Kidney Injury, Chronic Kidney Disease, and Kidney Failure: A Multilevel Modeling Analysis

Background: Neurocognitive impairment is a common finding across the spectrum of kidney disease and carries important consequences for quality of life. We previously demonstrated that robotic technology can identify neurocognitive impairments not readily detectable by traditional testing in patients with acute kidney injury (AKI) and chronic kidney disease (CKD).

Objective: The present study aimed to assess whether these quantifiable deficits in neurocognition differ based on a diagnosis of AKI, CKD, or kidney failure.

Design: This was a cross-sectional analysis of participants previously enrolled in an observational study.

Setting: Patients were enrolled at a tertiary academic hospital, Kingston Health Sciences Centre, Kingston, ON, Canada.

Patients: Adults with AKI, CKD, or kidney failure.

Measurements: Each participant underwent robotic neurocognitive assessment using the Kinarm: an interactive robotic device that uses a series of behavioral tasks involving movement of the upper limbs to precisely quantify neurocognitive impairment across a variety of neurocognitive domains.

Methods: Multilevel modeling was used to determine the effect of Kinarm task type, kidney diagnostic group (AKI vs CKD vs kidney failure), and the interaction between the two, on neurocognitive performance.

Results: A total of 104 participants within 1 year of an AKI event or with CKD category G3-5 were enrolled. We found that across all of the kidney diagnostic groups, participants performed worst on the Kinarm tasks of Reverse Visually Guided Reaching ( = 0.64 [95% confidence interval = 0.42, 0.85]), Visually Guided Reaching ( = 0.28 [0.07, 0.49]), and Trail Making ( = 0.50 [0.28, 0.72]), relative to all other tasks. There were no significant differences in average performance across tasks based on kidney diagnostic group. However, diagnostic group and neurocognitive task type interacted to determine performance, such that patients with AKI performed worse than those with either CKD or kidney failure on the Reverse Visually Guided Reaching task.

Limitations: Kinarm assessment was performed at a single time point, and the sample size itself was small, which may lead to the risk of a false-positive association despite the use of multilevel modeling. Our sample size also did not permit inclusion of the underlying etiology of kidney impairment as a covariate in our analyses, which may have also influenced neurocognitive function.

Conclusions: In this study that utilized the Kinarm to assess neurocognitive function, patients with AKI demonstrated significantly worse neurocognitive functioning than patients with CKD or kidney failure on a task measuring executive function and visuomotor control.