Polysomnographic Predictors of Neuropsychological Performance Two Years After Traumatic Brain Injury: A Multicenter Cohort Study

Research Objectives: To examine the relationship between sleep disturbance during inpatient rehabilitation and post-discharge neuropsychological test performance in persons with traumatic brain injury (TBI).

Design: Cohort study of participants co-enrolled in the TBI Model Systems (TBIMS) and Comparison of Sleep Apnea Screening and Diagnostic Tools (CSAS) studies. Participants underwent attended Level 1 polysomnography during acute neurorehabilitation for TBI. Follow-up interviews including neuropsychological testing were conducted two years from injury date.

Setting: Six TBIMS and CSAS centers.

Participants: Participants (N=123) were 81% male and 69% White, with median age of 39 years (IQR = 26, 53) at time of TBI. Posttraumatic amnesia duration ranged from 0 to 279 days (median = 23, IQR = 9, 40).

Interventions: n/a

Main Outcome Measures: Brief Test of Adult Cognition by Telephone (BTACT).

Results: After adjusting for age, sex, education, and posttraumatic amnesia duration, Stage 1 sleep (%N1) was negatively associated with the Verbal Memory composite (β = -0.026, p = 0.0107, η2 = 5.78%), Executive Function composite (β = -0.026, p = 0.0042, η2 = 7.28%) and overall BTACT composite (β = -0.029, p = 0.0009, η2 = 9.80%). Also, Wake After Sleep Onset (percent of time) had a significant negative relationship with only the Verbal Memory composite (β = -0.020, p = 0.0333, η2 = 4.05%). The Apnea Hypopnea Index, percent time spent in other sleep stages, and hypoxemia metrics were not significant predictors of any BTACT composite scores.

Conclusions: Results suggest that increased shallow sleep depth during predicts poorer neuropsychological functioning well after the acute TBI recovery period. Findings in this TBI sample are consistent with prior research showing an association between sleep disturbance and cognition in the general population. However, replication and external validation will be prudent.

Author(s) Disclosures: The views expressed in this abstract are those of the authors and do not necessarily reflect the official policy or position of the Department of Veterans Affairs or the U.S. Government. This work is supported with resources and use of facilities at the James A Haley Veterans’ Hospital. Funding was provided by PCORI (CER-1511-33005) and NIDILRR (90DPTB0007, 90DP0084, 90DPTB0013, 90DPTB0008, 90DPT80004, 90DPTB0001, 90DPTB0024).