Session P32.8

Chemical Instability, State Instability and Arousals in the Pathogenesis of Periodic Breathing in Heart Failure Patients

GD Pinna*, R Maestri, E Robbi, MT La Rovere

S. Maugeri Foundation
Montescano, Italy

Background: Periodic breathing (PB, waxing and waning of ventilation with cyclic apneas or hypopneas) is very common in heart failure (HF) patients both during nighttime and daytime, and is associated with increased morbidity and mortality. The mechanisms underlying this breathing disorder are complex and still not fully elucidated. Experimental and modeling studies support the hypothesis that PB results from an instability in the chemical feedback loop controlling ventilation. Ventilatory oscillations, however, may also result from repeated fluctuations in the EEG level of wakefulness (i.e., state instability) that commonly occur at sleep onset. Moreover, spontaneous arousals during N-REM sleep have also been proposed as a key mechanisms for the development of PB.
Aim: In this study we investigated whether PB can occur in HF patients in the absence of state instability and/or arousals.
Methods: Standard polysomnographic recordings were performed in 9 stable HF patients with PB during daytime supine rest (N=5) or nighttime (N=4). EEG analysis was carried out using an hybrid approach that included standard visual scoring and automated state transition analysis based on spectral analysis and statistical discrimination.
Results: During daytime recordings, the onset and end of 86% of cyclic apneas and hypopneas were synchronous with, respectively, transitions from alpha to theta and from theta to alpha. During nighttime recordings, one subject developed PB while fully awake. At sleep onset, repetitive state transitions occurred in all subjects but they were synchronous with the cyclic pattern of PB only intermittently. During Non-REM stage 2, 30% to 70% of hyperventilation phases were associated with an arousal. Sustained ventilatory oscillations without concurrent fluctuations in state or spontaneous arousals were observed in 3/4 patients during Non-REM stage 3. PB never occurred during REM sleep.
Conclusions: These preliminary results clearly indicate that, although state instability and arousals are often synchronized with ventilatory oscillations, they are not necessary for the occurrence of PB. This suggests that chemical instability might represent the underlying basic mechanisms of PB in HF patients, and that state instability and arousals can intermittently interact with such mechanism, likely promoting a mutual reinforcement.

(Abstract Control Number: 127)