Association of Plasma Biomarkers with Sleep Outcomes and Treatment Response After Mild Traumatic Brain Injury

Shawn R. Eagle,Ava M. Puccio,Denes V. Agoston,Michael Mancinelli,Rachel Nwafo,Peyton McIntyre,Allison Agnone,Savannah Tollefson,Michael Collins,Anthony P. Kontos,Walter Schneider,David O. Okonkwo,Ryan J. Soose
DOI: https://doi.org/10.1089/neur.2023.0002
2023-04-01
Neurotrauma Reports
Abstract:Sleep disturbances occur in up to 70% of patients with mild traumatic brain injury (mTBI). Modern mTBI management recommends targeted treatment for the patient's unique clinical manifestations (i.e., obstructive sleep apnea, insomnia). The purpose of this study was to evaluate the association of plasma biomarkers with symptom reports, overnight sleep evaluations, and response to treatment for sleep disturbances secondary to mTBI. This study is a secondary analysis of a prospective multiple interventional trial of patients with chronic issues related to mTBI. Pre- and post-intervention assessments were conducted, including overnight sleep apnea evaluation, the Pittsburgh Sleep Quality Index (PSQI), and blinded analysis of blood biomarkers. Bivariate Spearman correlations were conducted for pre-intervention plasma biomarker concentrations and 1) PSQI change scores and 2) pre-intervention sleep apnea outcomes (i.e., oxygen saturation measures). A backward logistic regression model was built to evaluate the association of pre-intervention plasma biomarkers with improvement in PSQI over the treatment period (<i>p</i> &lt; 0.05). Participants were 36.3 ± 8.6 years old and 6.1 ± 3.8 years from their index mTBI. Participants reported subjective improvements (PSQI = -3.7 ± 3.8), whereas 39.3% (<i>n</i> = 11) had improved PSQI scores beyond the minimum clinically important difference (MCID). PSQI change scores correlated with von Willebrand factor (vWF; ρ = -0.50; <i>p</i> = 0.02) and tau (ρ = -0.53; <i>p</i> = 0.01). Hyperphosphorylated tau correlated with average saturation (ρ = -0.29; <i>p</i> = 0.03), lowest desaturation (ρ = -0.27; <i>p</i> = 0.048), and baseline saturation (ρ = -0.31; <i>p</i> = 0.02). The multi-variate model (<i>R</i> <sup>2</sup> = 0.33; <i>p</i> = 0.001) retained only pre-intervention vWF as a predictor (odds ratio = 3.41; 95% confidence interval, 1.44-8.08; <i>p</i> = 0.005) of improving PSQI scores beyond the MCID. vWF had good discrimination (area under the curve = 0.83; <i>p</i> = 0.01), with an overall accuracy of 77%, sensitivity of 46.2%, and specificity of 90.0%. Validation of vWF as a potential predictive biomarker of sleep improvement post-mTBI could optimize personalized management and healthcare utilization.
What problem does this paper attempt to address?