Cardiovascular and ventilatory acclimatization induced by chronic intermittent hypoxia : A role for the carotid body in the pathophysiology of sleep apnea

Patients with obstructive sleep apnea (OSA) show augmented ventilatory, sympathetic and cardiovascular responses to hypoxia. The facilitatory effect of chronic intermittent hypoxia (CIH) on the hypoxic ventilatory response has been attributed to a potentiation of the carotid body (CB) chemosensory response to hypoxia. However, it is a matter of debate whether the effects induced by CIH on ventilatory responses to hypoxia are due to an enhanced CB activity. Recently, we studied the effects of short cyclic hypoxic episodes on cat cardiorespiratory reflexes, heart rate variability, and CB chemosensory activity. Cats were exposed to cyclic hypoxic episodes repeated during 8 hours for 4 days. Our results showed that CIH selectively enhanced ventilatory and carotid chemosensory responses to acute hypoxia. Exposure to CIH did not increase basal arterial pressure, heart rate, or their changes induced by acute hypoxia. However, the spectral analysis of heart rate variability of CIH cats showed a marked increase of the low/high frequency ratio and an increased variability in the low frequency band of heart rate variability, similar to what is observed in OSA patients. Thus, it is likely that the enhanced CB reactivity to hypoxia may contribute to the augmented ventilatory response to hypoxia. Key terms: ventilatory acclimatization, carotid body, chemosensory activity, intermittent hypoxia, obstructive sleep apnea. Corresponding author: Dr. Rodrigo Iturriaga, Laboratorio de Neurobiología, Facultad de Ciencias Biológicas, P. Universidad Católica de Chile, Casilla 114D, Santiago, Chile. Tel.: (56-2) 686-2852, Fax: (56-2) 222-5515, E-mail:

The carotid body (CB) chemoreceptor senses the arterial levels of PO 2 , PCO 2, and pH, and contributes to the ventilatory and cardiovascular regulation (Eyzaguirre and Zapata, 1984;Iturriaga et al., 1994a;Zapata and Iturriaga, 1997;Zapata and Larraín, 2005).In response to hypoxia, hypercapnia, and acidosis, the firing rate of chemosensory impulses increases in the carotid sinus nerve.The current model of CB chemoreception states that stimuli act upon chemoreceptor (glomus) cells, which are synaptically apposed to the sensory endings of petrosal ganglion neurons.It is accepted that hypoxia releases one -or more-excitatory transmitter(s) from glomus cells, which in turn increases the firing rate of chemosensory impulses in the nerve terminals of petrosal neurons (Iturriaga 2001;Iturriaga and Alcayaga, 2004).In addition, Eyzaguirre and colleagues proposed that electrical transmission plays an important role in CB chemoreception.Indeed, there are chemical and electrical connections between glomus cells and between glomus cells and the nerve endings of petrosal neurons (Eyzaguirre, 2005).In humans and mammals, the CB initiates the systemic ventilatory and cardiovascular adjustments to hypoxia.It is well known that exposure to sustained hypoxia (i.e.high altitude) augments the ventilatory response to acute hypoxia in humans and animals, a phenomenon known as ventilatory acclimatization (Bisgard, 2000).Ventilatory acclimatization induced by sustained hypoxia requires the presence of functional CBs, because it does not occur when the CBs are removed or the carotid sinus nerve is sectioned (Smith et al., 1986).

CHRONIC INTERMITTENT HYPOXIA ENHANCES THE CAROTID BODY CHEMOSENSORY RESPONSE
TO HYPOXIA Exposure of humans or animals to sustained hypoxia for hours or days is normally known as chronic hypoxia.However, the most common pattern of hypoxic exposure in humans is chronic intermittent hypoxia (CIH), characterized by cyclic hypoxic episodes of short duration followed by normoxia.This hypoxic pattern is a feature of pathological conditions such as obstructive sleep apnea (OSA) and chronic obstructive pulmonary disease.The OSA syndrome is characterized by repeated episodes of partial or completes obstruction of the upper airway during sleep, which disturbs both ventilation and sleep architecture.The incidence of OSA (more than 10 apneas and hypopneas per hour) in middle-aged men and women is 5% and 3%, respectively, but 24% of men and 9% of woman present 5 apneas per hour of sleep (Hla et al., 1994).During the collapse of the upper airway, the resulting hypoxia and hypercapnia stimulate the CB chemoreceptors, increasing the activity of respiratory muscles and enhancing vascular sympathetic tone and arterial pressure.Finally, the stimulation of CB chemoreceptors and pulmonary mechanoreceptors produces arousal and restores ventilation.The alternation of hypoxia and normoxia produces sleep fragmentation and drowsiness.In addition to the acute cardiovascular responses, approximately half of OSA patients develop arterial hypertension.The OSA syndrome is linked to metabolic, vascular, and genetic markers associated with an increased risk of cardiovascular disease.Endothelial dysfunctions, inflammation, disorders in lipid metabolism, and sympathetic activation have been proposed as potential pathogenic phenomena in OSA-linked hypertension (Quan and Gersh, 2004).Most of the information on effects of CIH on peripheral chemoreflex control of cardiovascular and respiratory systems has been obtained from studies performed on OSA patients.Several studies have shown that patients with OSA had enhanced ventilatory and cardiovascular reflex responses to acute hypoxia (Cistulli and Sullivan, 1994;Narkiewicz et al., 1998a;1998b;1999).Therefore, the facilitator effect of CIH on the hypoxic ventilatory response has been attributed to a potentiation of CB chemosensory responses to hypoxia (Narkiewicz et al., 1999;Loredo et al., 2001).Experiments performed in animal models show that CIH enhances the hypoxic ventilatory response (Ling et al., 2001) and produces long-term facilitation of respiratory motor activity (Peng and Prabhakar, 2003).Recently, Peng et al. (2001;2003) found that basal CB discharges and chemosensory responses to acute hypoxia were larger in rats exposed to a CIH pattern consisting of short cyclic hypoxic episodes followed by normoxia, applied for 8 hours during 10 days.Using a protocol of short hypoxic episodes, we studied the early effects of CIH on cat chemosensory responses to hypoxia (Rey et al., 2004a;Del Río et al., 2004).We found that cats exposed to this CIH pattern (PO 2 75 Torr followed by normoxia during 8 hours for 4 days) showed an enhanced CB response to acute hypoxia.Moreover, we found that basal and hypoxic-induced CB chemosensory discharges were higher in CIH cats.Thus, present evidence supports that CIH enhances the CB chemosensory response to acute hypoxia.

HYPOXIA
Long-term exposure to sustained hypoxia for weeks or months produces CB hypertrophy and hyperplasia of glomus cells, also inducing angiogenesis (Heath and Smith, 1992;Wang and Bisgard, 2002).
However, short-term exposure to CIH for 4-10 days does not produce hypertrophy and hyperplasia of glomus cells.Thus, it is likely that CIH may potentiate CB reactivity to acute hypoxia by changing the balance between excitatory and inhibitory transmitters or modulators within the chemoreceptor organ (Iturriaga and Alcayaga, 2004).There is evidence that sustained hypoxia for 2 weeks increases the levels of the CB excitatory peptide endothelin-1 and the expression of its receptor ET A in rat glomus cells (Chen et al., 2002).In addition, plasma concentration of endothelin-1 has been found to be elevated in OSA patients (Cistulli and Sullivan, 1994) and in rats exposed to CIH for a few days (Kanagy et al., 2001), suggesting that endothelin-1 may be involved in the enhanced cardiovascular response to hypoxia.In contrast, other inhibitory modulators of CB chemoreception, such as nitric oxide (Iturriaga, 2001;Valdés et al., 2003) and dopamine (Iturriaga et al., 1994b;1996) also are up-regulated in the first weeks of chronic sustained hypoxia (Ye et al., 2002;Wang and Bisgard, 2002).Sustained hypoxia is a strong stimulus to increase catecolaminergic pathways in glomus cells.Prolonged and short exposure to sustained hypoxia increases the activity of tyrosine hydroxylase and the levels of dopamine in the CB (Hanbauer, 1977;Czyzyk-Krzeska et al., 1992).However, effects of CIH on the CB catecholaminergic pathway are smaller and delayed as compared to those induced by sustained hypoxia (Hui et al., 2003).Thus, effects of CIH on the expression of catecholaminergic enzymes in the glomus cells are transient and small compared with the effects of sustained hypoxia, suggesting that reduced dopaminergic induction by intermittent hypoxia in CB may contribute to the enhanced chemosensory response to hypoxia.Recently, Peng et al., (2003) found evidence involving oxygen radicals in the CB chemosensory potentiation induced by CIH.They found that treatment with a superoxide anion scavenger prevents rat CB chemosensory potentiation induced by CIH.In addition, the activity of the aconitase enzyme decreased in glomus cells compared with controls, and the activity of the complex I of the mitochondrial electron transport chain is inhibited, suggesting that mitochondria is one of the sources of reactive oxygen species.These observations indicate that the increased generation of these molecules may contribute to CB chemosensory potentiation induced by CIH.

EFFECTS OF CHRONIC INTERMITTENT HYPOXIA ON THE VENTILATORY SYSTEM
Exposure to sustained hypoxia progressively increases both chemosensory and ventilatory responses to acute hypoxia in 4 hours in goats to 10 days in rats (Nielsen et al., 1988;Bisgard, 2000).In humans, mild hypoxia maintained for 8 hours produces hyperventilation and increases the reactivity of the reflex ventilatory response to acute hypoxia (Fatemian et al., 2001).In contrast, less is known about the effects of CIH on ventilatory control and CB function.CIH is associated with an augmented hypoxic ventilatory response in OSA patients and in animal models.Clearly, OSA patients show enhanced hypoxic ventilatory responses, however, in some studies a normal or even attenuated hypoxic ventilatory response also has been observed in patients with a prolonged OSA history (Costes et al., 1995;Osanai et al., 1999).Nonetheless, a main problem with these studies is that peripheral CB reflexes may be impaired by comorbidities such as obesity, aging, hypertension, and application of continuous positive airway pressure to treat the apneas.In a controlled study, performed in patients with recently diagnosed OSA without treatment, Narkiewicz et al., (1999) found that reflex ventilatory, sympathetic and cardiovascular responses to acute hypoxia were enhanced.More recently, Loredo et al., (2001) found that hypertensive patients with OSA present higher basal tidal volume, suggesting that the peripheral CB chemosensory drive is enhanced.Studies performed in animal models have shown dissimilar effects of CIH on the hypoxic ventilatory response.Ling et al. (2001) reported that phrenic response to acute hypoxia is enhanced in rats exposed to 5 min of ~10% O 2 , followed by 5 min of air, 12 hours/night during 7 nights.Similarly, Reeves et al. (2003) found that rats exposed to a CIH pattern consisting of alternating episodes of room air and ~10% O 2 for 90 s for 30 days, showed an enhanced hypoxic ventilatory response.However, Greenberg et al. (1999) did not find any difference in ventilatory responses to hypoxia in rats exposed 30 s to 7% O 2 followed by 30 s of normoxia for 8 hours/day during 10 days.In a recent study, we (Rey et al., 2004a) found that four days of CIH increased cat basal tidal volume and inspiratory minute volume and potentiated the increase of these variables to acute hypoxia (PO 2 < 100 Torr).Thus, our results show that the enhanced hypoxic ventilatory response correlates with the potentiation of CB chemosensory responses to acute hypoxia.Therefore, experimental and clinical studies indicate that the initial response to CIH is an augmented ventilatory response to hypoxia.However, after long-term exposure to CIH, the ventilatory response to hypoxia seems to be attenuated.Indeed, in an animal model of OSA produced by the telemetric occlusion of the upper airway in dogs, Kimmof et al. (1997) found that after 3 months of CIH dogs showed blunted ventilatory responses to hypoxia.This pattern of response to intermittent hypoxia is similar to what is observed in humans and animals exposed to high altitude hypoxia for long periods (Moore, 2000;Santolaya et al., 1989).Thus, one of the initial physiological responses to CIH is an enhanced ventilatory response that would allow a compensation for the cyclic hypoxic challenge, therefore improving survival.The attenuation of the enhanced ventilatory response observed in long-term exposure to CIH resembles the physiological acclimatization to sustained hypoxia.Since the hypoxic stimulation of the CB contributes to the arousal, the attenuation of the capacity to hyperventilate may reduce the strength of the response to terminate the apnea (Lahiri, 1994).García-Río et al. (2000) classified patients with OSA in three groups: normotensive, nocturnal hypertensive, and diurnal hypertensive.The ventilatory response to acute hypoxia was larger in the group of diurnal hypertensive patients than the nocturnal hypertensives and normotensive OSA patients.These results show that the magnitude of the enhanced ventilatory response to hypoxia is correlated with the increase in blood pressure, which is compatible with different degrees of CB responsiveness to hypoxia in OSA patients.

EFFECTS OF CHRONIC INTERMITTENT HYPOXIA ON THE CARDIOVASCULAR SYSTEM
Patients with OSA and animals exposed to CIH show enhanced renal sympathetic output, vascular reactivity, and systemic hypertension (Greenberg et al., 1999;Fletcher et al., 1995;Smith and Muenter, 2000;Somers et al., 1995).The hypertension has been attributed to an increased sympathetic outflow due to the repetitive hypoxic CB stimulation.Indeed, the denervation of the CB prevents the hypertension induced by CIH (Fletcher et al., 1992) component (Shiomi et al., 1996;Narkiewicz et al., 1998a).In our model of CIH, we observed that the exposure to cyclic hypoxic episodes increases the LF component with a simultaneous reduction of the HF component (Rey et al., 2004a).Thus, cats exposed to CIH, like OSA patients, show an increased LF/HF ratio suggesting the existence of early changes in sympathetic control of heart rate.Nevertheless, these changes appear in cats after 4 days of exposure to CIH, compared to OSA patients that normally have a longer history of hypoxic exposure during sleep.It is widely accepted that the LF/HF ratio is a measure of autonomic influences on heart rate variability (Task Force, 1996).The LF and HF components are related to autonomic influences on heart rate; the HF component has been related to cardiac parasympathetic efferent activity; and the LF component has been related to sympathetic cardiac efferences (Task Force, 1996).

CONCLUSION
The available experimental and clinical evidence suggests that the CIH pattern of short episodes of hypoxia followed by normoxia selectively enhances the CB chemosensory and ventilatory responses to hypoxia, suggesting that the CB plays an essential role in the enhanced ventilatory and cardiovascular responses observed in animals and OSA patients.