Acute exacerbation of sleep apnea by hyperoxia impairs cognitive flexibility in Brown-Norway rats.

STUDY OBJECTIVES

To determine whether learning deficits occur during acute exacerbation of spontaneous sleep related breathing disorder (SRBD) in rats with high (Brown Norway; BN) and low (Zucker Lean; ZL) apnea propensity.

DESIGN

Spatial acquisition (3 days) and reversal learning (3 days) in the Morris water maze (MWM) with polysomnography (12:00-08:00): (1) with acute SRBD exacerbation (by 20-h hyperoxia immediately preceding reversal learning) or (2) without SRBD exacerbation (room air throughout).

SETTING

Randomized, placebo-controlled, repeated-measures design.

PARTICIPANTS

14 BN rats; 16 ZL rats.

INTERVENTIONS

20-h hyperoxia.

MEASUREMENTS AND RESULTS

Apneas were detected as cessation of respiration ≥ 2 sec. Swim latency in MWM, apnea indices (AI; apneas/hour of sleep) and percentages of recording time for nonrapid eye movement (NREM), rapid eye movement (REM), and total sleep were assessed. Baseline AI in BN rats was more than double that of ZL rats (22.46 ± 2.27 versus 10.7 ± 0.9, P = 0.005). Hyperoxia increased AI in both BN (34.3 ± 7.4 versus 22.46 ± 2.27) and ZL rats (15.4 ± 2.7 versus 10.7 ± 0.9) without changes in sleep stage percentages. Control (room air) BN and ZL rats exhibited equivalent acquisition and reversal learning. Acute exacerbation of AI by hyperoxia produced a reversal learning performance deficit in BN but not ZL rats. In addition, the percentage of REM sleep and REM apnea index in BN rats during hyperoxia negatively correlated with reversal learning performance.

CONCLUSIONS

Acute exacerbation of sleep related breathing disorder by hyperoxia impairs reversal learning in a rat strain with high apnea propensity, but not a strain with a low apnea propensity. This suggests a non-linear threshold effect may contribute to the relationships between sleep apnea and cognitive dysfunctions, but strain-specific differences also may be important.