0504 Neural Basis of Improved Sustained Attention Following Sleep Apnea Treatment

Abstract Introduction Obstructive sleep apnea (OSA) is associated with sustained attention deficits and lapses in vigilance. Although OSA treatment via continuous positive airway pressure (CPAP) is thought to normalize these attentional deficits, the normalization of presumed neural dysfunction is not yet understood. Here we sought to assess the neural underpinnings of sustained attention improvements associated with CPAP. Given the role of thalamus in attention regulation, we hypothesized that thalamic activity will be positively associated with attention/vigilance. Methods Twenty adults with OSA (13M/7F, Age:57±10, BMI:33±7) who were adherent to AutoPAP (≥4 hr/night for ≥70% of nights) completed polysomnograms (on-CPAP, off-CPAP; counterbalanced). The following morning, participants performed a psychomotor vigilance task (PVT) during fMRI. Lapses are defined as reaction time (RT)≥500ms, and participants are categorized as attentive ( 7.1 lapses) or nonattentive (7.1 lapses) by median split. Functional MRI is preprocessed by AFNI using a general linear model and group-level mixed-effects analysis to identify thalamic activity in (1) on-CPAP vs. off-CPAP (2) attentive vs. nonattentive on-CPAP and (3) an exploratory moderation by anxiety (N=6). Significant clusters were ≥20 voxels (3D units). Results Two thalamic clusters (contiguous voxel group) showed higher activity on-CPAP compared to off-CPAP: a left-thalamic cluster (57 voxels; CM = -7.1, 19.7, 6.1 center-of-mass coordinates; mean beta estimate = 0.145) and a right-thalamic cluster (39 voxels; CM = 10.8, 12.2, 5.3; mean = 0.146). The attentive-nonattentive analysis identifies two left-thalamic clusters (51 and 29 voxels; means = -0.193 and -0.1652), suggesting lower activation in high-lapse subjects. PVT performance shows a trend of fewer lapses on-CPAP, and more lapses in the nonattentive group. An exploratory analysis reveals three additional thalamic clusters (326, 124, and 50 voxels; means = 0.18, 0.16, 0.14) in which anxious subjects demonstrate larger on-off activation differences compared with non-anxious subjects. Conclusion Our findings suggest that thalamic activity may differ across treatment and attentional performance, where patients with good vigilance show stronger thalamic engagement. Patterns observed within anxious subjects suggest that some subgroups may be more sensitive to changes in treatment. These findings identify the thalamus as a key region for investigating variability in treated OSA patients. Support (if any) NIH K25HL151912, NIH R01HL171813, NIH R21HL165320