Human obesity leads to an increase in respiratory demands. As obesity becomes more pronounced some individuals are unable to compensate, leading to elevated arterial carbon dioxide levels (PaCO2), alveolar hypoventilation, and increased cardiorespiratory morbidity and mortality (Pickwickian syndrome). The mechanisms that link obesity and hypoventilation are unknown, but thought to involve depression of central respiratory control mechanisms. Here we report that obese C57BL/6J-Lepob mice, which lack circulating leptin, also exhibit respiratory depression and elevated PaCO2 (> 10 mm Hg; p < 0. 0001). A role for leptin in restoring ventilation in these obese, mutant mice was investigated. Three days of leptin infusion (30 microg/d) markedly increased minute ventilation (V E) across all sleep/wake states, but particularly during rapid eye movement (REM) sleep when respiration was otherwise profoundly depressed. The effect of leptin was independent of food intake, weight, and CO2 production, indicating a reversal of hypoventilation by stimulation of central respiratory control centers. Furthermore, leptin replacement in mutant mice increased CO2 chemosensitivity during non-rapid eye movement (NREM) (4.0 +/- 0.5 to 5.6 +/- 0.4 ml/min/%CO2; p < 0.01) and REM (-0.1 +/- 0.5 to 3.0 +/- 0.8 ml/min/%CO2; p < 0.01) sleep. We also demonstrate in wild-type mice that ventilation is appropriately compensated when obesity is diet-induced and endogenous leptin levels are raised more than tenfold. These results suggest that leptin can prevent respiratory depression in obesity, but a deficiency in central nervous system (CNS) leptin levels or activity may induce hypoventilation and the Pickwickian syndrome in some obese subjects. O'Donnell CP, Schaub CD, Haines AS, Berkowitz DE, Tankersley CG, Schwartz AR, Smith PL. Leptin prevents respiratory depression in obesity.

Obstructive sleep apnea (OSA) acutely increases systemic (Psa) and pulmonary (Ppa) arterial pressures and decreases ventricular stroke volume (SV). In this study, we used a canine model of OSA (n = 6) to examine the role of hypoxia and the autonomic nervous system (ANS) in mediating these cardiovascular responses. Hyperoxia (40% oxygen) completely blocked any increase in Ppa in response to obstructive apnea but only attenuated the increase in Psa. In contrast, after blockade of the ANS (20 mg/kg iv hexamethonium), obstructive apnea produced a decrease in Psa (-5.9 mmHg; P < 0.05) but no change in Ppa, and the fall in SV was abolished. Both the fall in Psa and the rise in Ppa that persisted after ANS blockade were abolished when apneas were induced during hyperoxia. We conclude that 1) hypoxia can account for all of the Ppa and the majority of the Psa response to obstructive apnea, 2) the ANS increases Psa but not Ppa in obstructive apnea, 3) the local effects of hypoxia associated with obstructive apnea cause vasodilation in the systemic vasculature and vasoconstriction in the pulmonary vasculature, and 4) a rise in Psa acts as an afterload to the heart and decreases SV over the course of the apnea.

We examined the hemodynamic responses to normal breathing and induced upper airway obstructions during sleep in a canine model of obstructive sleep apnea. During normal breathing, cardiac output decreased (12.9 +/- 3.5%, P < 0.025) from wakefulness to non-rapid-eye-movement sleep (NREM) but did not change from NREM to rapid-eye-movement (REM) sleep. There was a decrease (P < 0.05) in systemic (7.2 +/- 2.1 mmHg) and pulmonary (2.0 +/- 0.6 mmHg) arterial pressures from wakefulness to NREM sleep. In contrast, systemic (8.1 +/- 1.0 mmHg, P < 0.025), but not pulmonary, arterial pressures decreased from NREM to REM sleep. During repetitive airway obstructions (56.0 +/- 4.7 events/h) in NREM sleep, cardiac output (17.9 +/- 3.1%) and heart rate (16.2 +/- 2.5%) increased (P < 0.05), without a change in stroke volume, compared with normal breathing during NREM sleep. During single obstructive events, left (7.8 +/- 3.0%, P < 0.05) and right (7.1 +/- 0.7%, P < 0.01) ventricular outputs decreased during the apneic period. However, left (20.7 +/- 1.6%, P < 0.01) and right (24.0 +/- 4.2%, P < 0.05) ventricular outputs increased in the post-apneic period because of an increase in heart rate. Thus 1) the systemic, but not the pulmonary, circulation vasodilates during REM sleep with normal breathing; 2) heart rate, rather than stroke volume, is the dominant factor modulating ventricular output in response to apnea; and 3) left and right ventricular outputs oscillate markedly and in phase throughout the apnea cycle.