Exp Physiol 92.1 pp 45–50 45

Experimental Physiology

Sleep apnoea and hypertension: role of chemoreflexesin humans

Michael L. Smith and Christina F. Pacchia

Department of Integrative Physiology, University of North Texas Health Science Center, Fort Worth, TX, USA

The link between sleep apnoea and systemic hypertension in humans is well documented.However, a direct causal association between the two diseases independent of comorbiditieshas been difficult to establish. Comorbidities clearly play an important role in this strongrelationship; however, new findings also suggest that sleep apnoea is an independent riskfactor for hypertension. This relationship appears to be at least in part a result of chronicallyelevated sympathetic activity, and therefore manifests as a neurally mediated hypertension.Although the mechanism(s) for this causal relationship of sleep apnoea to hypertension remainsill defined, a growing body of literature suggests that autonomic dysfunction, mediated byabnormal chemoreflex control of sympathetic activity, is a potential mechanism. Abnormalchemoreflex responses to both acute and chronic apnoea or hypoxia have been demonstrated.Hypothesized mechanisms by which chemoreflex dysfunction may contribute to chronicallyelevated sympathetic tone and ultimately hypertension are explored in this review. Thus,this review focuses on the current evidence linking chemoreflex function to obstructive sleepapnoea and systemic hypertension in humans and provides an analysis of these data and theirimplications.

(Received 26 September 2006; accepted after revision 9 November 2006; first published online 10 November 2006)Corresponding author M. L. Smith: Department of Integrative Physiology, University of North Texas Health ScienceCenter, 3500 Camp Bowie Boulevard, Fort Worth, TX 76107, USA. Email: msmith@hsc.unt.edu

The link between obstructive sleep apnoea (OSA)and cardiovascular morbidity, particularly systemichypertension, is well recognized. However, a true causalrelationship, independent of common comorbidities(such as obesity, metabolic disorders, etc.), has beendifficult to demonstrate. In 2000, Peppard and colleaguesdemonstrated in a longitudinal study a dose-dependentrelationship between the severity of sleep apnoeaand subsequent development of hypertension (Peppardet al. 2000). Moreover, these data suggest that thisrelationship was independent of comorbidity andother confounding factors. These findings are thefirst to directly implicate sleep apnoea as causativefor the occurrence of hypertension. The physiologicalmechanism(s) responsible for this direct causal associationof sleep apnoea to hypertension remains elusive. A growingbody of evidence points to autonomic dysfunction andabnormal vascular control as the primary culprits, andthis is the focus of this themed issue of ExperimentalPhysiology. In this review, a case is made for abnormalchemoreflex control of sympathetic activity as the primary

mechanism driving the increased risk of neurogenic-mediated hypertension accompanying sleep apnoea.

In patients with chronic OSA, there are significantelevations of production and circulation of catecholaminesduring sleep and wakefulness (Fletcher et al. 1987; Carlsonet al. 1993; Marrone et al. 1993; Somers et al. 1995;Garcia-Rio et al. 2000). Marrone et al. (1993) illustratedthat both noradrenaline and adrenaline were elevated inpatients with sleep apnoea and that urinary adrenalinesecretion decreased in the absence of apnoeas. Fletcheret al. (1987) and Carlson et al. (1993) both demonstratedelevated noradrenaline levels in OSA patients during thedaytime, suggesting a physiological adaptation to hypoxiaand apnoeas during sleep. Relative to a potential role of thechemoreceptors, it is not known whether patients werehypoxaemic during wakefulness. In most OSA patients,however, hypoxaemia during wakefulness is normallyuncommon and mild; therefore, these data would suggesteither that the chemoreflex control of sympathetic activitybecomes tonically active at rest (normoxia), or thatother mechanisms (e.g. sleep deprivation) contribute to

C© 2007 The Authors. Journal compilation C© 2007 The Physiological Society DOI: 10.1113/expphysiol.2006.033753

46 M. L. Smith and C. F. Pacchia Exp Physiol 92.1 pp 45–50

the chronic elevation of sympathetic tone. Similarly,Garcia-Rio et al. (2000) found elevated nocturnaladrenaline levels that were correlated with the pressorresponse to hypoxia. Together, these data suggest thatthe elevated catecholamine concentrations found in thispopulation may be mediated in part by the nighttimeapnoeas, but also by altered autonomic control throughoutthe day.

Numerous studies also have shown that directlymeasured muscle sympathetic nerve activity (MSNA) iselevated during normal wakeful breathing and duringepisodes of sleep apnoea in OSA patients (Carlson et al.1993; Leuenberger et al. 1995; Somers et al. 1995; Saitoet al. 1988). Narkiewicz & Somers (1997) compared MSNAin normal-weight subjects, obese subjects without OSAand obese subjects with OSA. During normal daytimebreathing, the obese OSA subjects had significantly higherresting MSNA than the normal-weight subjects and obesesubjects without OSA. These results strongly suggest thatthe increase in basal MSNA is independent of obesityin this population. Collectively, these studies imply thatthe high incidence of hypertension among OSA patientsis, in part, due to chronically elevated MSNA. In thislight, considerable attention has focused on the role ofchemoreflex function, or dysfunction, in both the acuteand chronic responses to sleep apnoea, and its potentialrole in imparting an increased risk of hypertension.

As noted, the majority of investigations intoOSA and sympathetic tone and its control focuson microneurographic measures of MSNA. Thismeasurement is thought to be reflective of normalreflex control of vascular beds which determine systemicvascular resistance and thus, arterial pressure. Thus,these studies address effects of apnoea and hypoxia onaltered reflex control of sympathetic tone which wouldbe expected to affect long-term arterial pressure and itscontrol.

Chemoreceptors and OSA

Chemoreflex control of sympathetic activity. Untilrecently, the investigation of the physiological role of thechemoreceptors has generally been focused on the controlof respiration. However, a growing body of evidence hasshown that disturbed blood gases, whether acute, chronicor intermittent, can affect autonomic and cardiovascularfunction as well. Activation of the chemoreceptors byacute exposure to either hypoxia or hypercapnia is knownto provoke a sympathoexcitatory response in humans(Xie et al. 2000; Shoemaker et al. 2002; Cutler et al.2004b; Leuenberger et al. 2005). The effects appear tobe more sensitive to hypoxia than to hypercapnia, andgrowing evidence suggests that the hypoxic stimulus isthe principal stimulus for sympathoexcitation duringepisodes of apnoea. When breathing a hypoxic gas,

these sympathoexcitatory responses occur in the face ofa modulatory effect of associated increased ventilationresponses (Somers et al. 1989a). Despite the significanceof hypoxia as a chemoreflex-mediated stimulus forsympathoexcitation during apnoea, the concomitanthypercapnia and apnoea itself also contribute to thisresponse (Somers et al. 1989b; Morgan et al. 1995;Tamisier et al. 2004). Hypercapnia and hypoxia are knownto act synergistically to augment both ventilatory andsympathetic neural responses (Somers et al. 1989b).

Chemoreflex control in OSA. It is well documented thatOSA is characterized by increased MSNA compared withbaseline during acute bouts of hypoxia during wakefulnesswhen subjects are breathing normally, and this abnormalresponse is independent of the ventilatory response(Hedner et al. 1992b; Leuenberger et al. 1995; Somerset al. 1995). In addition, it has been demonstrated thatpatients with sleep apnoea exhibit enhanced ventilatoryand pressor responses to hypoxia when compared withnormal subjects. Accumulating data suggest that theperipheral chemoreceptors (via sensitivity to hypoxia)are responsible for these augmented MSNA, pressor andventilatory responses observed in OSA (Narkiewicz et al.1999; Iturriaga et al. 2005). These data point to thehypoxia stimulus as the primary drive for these adaptedresponses in OSA patients; however, both the intermittentinterruption of respiration and hypercapnia are alsolikely to contribute to this adaptation of the reflex, asnoted above. Other studies further support the idea thatchemoreceptor control of MSNA is exaggerated in patientswith OSA. Leuenberger et al. (1995) studied OSA patientsduring episodes of hypoxia at night and observed markedincreases in muscle sympathetic nerve activity. Smithet al. (1996) reported that transient bouts of hypoxaemiaprovoked sympathetic stimulation in OSA patients duringwakefulness, although the response was less than thatobserved during episodes of hypoxaemia during sleep.They also demonstrated that this response was greater inOSA patients than in non-OSA weight-matched controlsubjects.

The increased MSNA observed during sleep apnoeicevents can be explained by the repetitive bouts ofhypoxia, hypercapnia and apnoea. In contrast, themechanism by which sympathetic outflow is increased insleep apnoea patients during wakefulness remains to beelucidated. It has been postulated that tonic peripheralchemoreceptor activation contributes to the increasedMSNA observed during wakefulness in this population(Narkiewicz et al. 1998; Cutler et al. 2004a). Moreover,several studies examining the effects of sustained andintermittent hypoxia in rats concluded that continuousexposure to hypoxia did not elicit an enhanced sensoryresponse by the chemoreceptors, whereas intermittenthypoxia produced an enhanced chemosenstivity (Fletcher

C© 2007 The Authors. Journal compilation C© 2007 The Physiological Society

Exp Physiol 92.1 pp 45–50 Sleep apnoea and hypertension in humans 47

et al. 1992b; Fletcher, 2001; Peng et al. 2003; Peng& Prabhakar, 2004). The seminal studies of Fletcherand colleagues demonstrated that chronic intermittenthypoxia can lead to development of hypertension, andthat this effect is prevented by denervation of thechemoreceptors (Fletcher et al. 1992a,b). These resultsstrongly suggest that in a rat model: (1) the intermittentnature of the stimulus is key to these pathology-induced changes; and (2) the chemoreceptors can mediatephysiological changes resulting in hypertension. However,the relative contribution of intermittent versus sustainedhypoxia to changes in autonomic function and thelink between chemoreflex dysfunction and hypertensionremain inconclusive in humans and merit furtherinvestigation.

Four studies addressed the hypothesis that short-terminterventions akin to sleep apnoea can evoke an adaptationof chemoreflex control of MSNA. Morgan et al. (1996)and Xie et al. (2000) observed a sustained elevation ofMSNA after 20 min of intermittent asphyxia (hypercapnichypoxia). The basal MSNA remained elevated for at least20 min after the asphyxic stimulus had subsided. Using asimilar study design, Cutler et al. (2004a,b) demonstratedin healthy subjects that 20 min of intermittent hypoxialead to sustained elevation in MSNA even after hypoxicbouts were terminated. These responses and the associatedtime course of adaptation were similar to 20 minof intermittent apnoea. Thus, they concluded thatthis prolonged elevation postconditioning stimulus wasmediated by hypoxia and that the chemoreceptors playedan integral role in this outcome. The fact that intermittenthypoxia or apnoea produces altered chemoreflex controlof MSNA that is sustained well beyond the period ofthe intervention supports the hypothesis that impairedchemoreflex function may have long-term effects onhaemodynamic control and thus, the risk of hypertension.

The hypothesis that chemoreflex input may contributeto the sustained elevation of MSNA during wakefulnessimplies that the chemoreceptors are tonically activeduring normal wakeful breathing. Studies in which theeffect of hyperoxia on MSNA was assessed supportthis premise. Leuenberger et al. (1995) showed thatbreathing 100% oxygen produced a modest reductionin MSNA in OSA subjects, thereby suggesting that thechemoreflex was actively controlling MSNA even undernormoxic conditions. Similarly, Narkiewicz et al. (1998)deactivated the chemoreflex by having subjects breathe100% oxygen and found that MSNA was attenuatedin obese subjects with OSA but not in obese subjectswithout OSA. In a preliminary study, we assessed theMSNA response to 60 s periods of gradations of hypoxiainduced by an initial breath of nitrogen followed by anhypoxic gas mixture. In nine OSA patients, the mildesthypoxia (producing O2 saturations of 93 ± 1%) eliciteda significant increase in MSNA (�28 ± 15%), whereas

weight-matched control subjects without OSA showedno change in MSNA (�−6 ± 11%) with similar hypoxia(92 ± 1%). Collectively, these studies support the tenetthat the chemoreflex control of MSNA is active undernormoxic conditions in OSA patients, but is not in healthycontrol subjects.

Cardiovascular responses to hypoxia. Ultimately, if weare to determine a mechanistic link between OSA andhypertension, the link must involve net effects on thecardiovascular system. The changes in chemoreflex controlof MSNA discussed above may be implicated; however,the question also remains whether those changes affectthe determinants of arterial pressure, namely, cardiacoutput and systemic vascular resistance. The majorityof the literature offers data suggesting that pressorresponses to hypoxia are exaggerated in subjects with OSA.Narkiewicz et al. (1999) found significant increases inheart rate, blood pressure and ventilation during hypoxicbreathing in OSA patients compared with control subjects.In addition, the authors demonstrated that the MSNAresponse to hypoxia that is normally attenuated by thecompensatory increases in heart rate, ventilation andblood pressure in healthy subjects was sustained in the OSApatients. Thus, the chemoreflex seems to be an importantmediator in the exaggerated cardiovascular responsesobserved in this population (Narkiewicz et al. 1999).Narkiewicz et al. (1998) demonstrated that hyperoxicconditions decrease blood pressure in normoxic OSApatients but not in normal control subjects. The authorspostulated that tonic activation of chemoreceptors couldpartly account for the increase in blood pressure, againconsistent with the observations of chemoreflex control ofMSNA noted above. Hedner et al. (1992a) demonstratedthat OSA patients have an exaggerated pressor andventilatory response to hypoxia compared with normalsubjects, and that both responses were not differentbetween normotensive and hypertensive OSA patients.These results suggest the possibility of greater peripheralchemosensitivity in the OSA patients; however, thereason for the lack of difference between normotensiveand hypertensive OSA patients remains unclear. Inaddition, the authors found that ventilatory and pressorincreases were positively correlated. These data suggest thepresence of an excessive vasoconstriction mediated by thechemoreflex during hypoxia. The relationship between theventilatory and pressor response is debatable. Not all OSApatients had high ventilatory responses that accompanya high pressor response; likewise, some normal controlsubjects with no pressor responses display exaggeratedventilatory responses (Hedner et al. 1992a). Beyond thesestudies, there is now growing evidence to suggest thatvascular function is adversely impacted by OSA. Theseeffects are discussed in the associated reviews in thisthemed issue (see Foster et al. and Weiss et al.).

C© 2007 The Authors. Journal compilation C© 2007 The Physiological Society

48 M. L. Smith and C. F. Pacchia Exp Physiol 92.1 pp 45–50

The chemoreceptor reflex and hypertension

Evidence suggests that in humans with early and mildhypertension, sympathetic neural activity is increased(Trzebski et al. 1982; Anderson et al. 1989; Matsukawaet al. 1991; Floras & Hara, 1993; Grassi et al. 2000). Inturn, it is hypothesized that this elevation of MSNA iscausally related to the hypertension. One hypothesizedmechanism by which this augmentation occurs involvesaltered peripheral chemoreceptor control of MSNA.

Carotid body morphology. Why might the chemoreflexcontribute to development of hypertension? One reasoncould be that changes in the chemoreceptor structure mayadversely impact upon receptor function. Morphologicalstudies that examined the anatomical and vascularcharacteristics of the carotid bodies in hypertensiondisclosed another possible explanation for the augmentedchemoreflex response observed in hypertension. Since thecarotid bifurcation (Howe & Neil, 1972) is very susceptibleto atherosclerosis (Friedman et al. 1975), several autopsystudies examined the arteries that feed into the carotidbodies in order to understand the possible relationshipof carotid body pathology to hypertension. For example,Habeck et al. (1983), Habeck (1986), Przybylski & Trzebski(1980) and Przybylski (1981) observed atherosclerosis ofthe glomic arteries and thus a decreased blood flow to thecarotid bodies of hypertensive subjects. Along these lines,Kluge (1985) examined the vascularization of dissectedcarotid bodies from deceased patients with hypertension.There was a significant difference in the vascularizationof the carotid bodies between the hypertensives and thecontrol group such that the patients with hypertension hada lower number of arteries supplying the carotid bodies.Kluge concluded that the difference in the number ofarteries may be genetically determined and that the carotidbodies may play an important role in the pathogenesis ofhypertension. It has been proposed that chronic ischaemiaof the carotid bodies leads to chronic stimulation of thechemoreceptors and subsequent hypertrophy (Przybylski& Trzebski, 1980; Przybylski, 1981). Morphological studiessuggest that there is in fact hypertrophy and hyperplasiaof the carotid bodies in hypertensive humans. One studyexamined the histology of the carotid bodies in deceasedcases with myocardial hypertrophy secondary to bothsystemic hypertension and hypoxaemia. The authorsfound a higher cell count, increased weight and a greaterdiameter of the carotid bodies in those with systemichypertension (Smith et al. 1982). In a follow-up study,the authors conclude that the type of cell that proliferatesand causes hyperplasia and hypertrophy is the elongatedsustentacular type II cell (Heath et al. 1982). Takentogether, these findings suggest an important relationshipbetween abnormalities of the carotid bodies and systemichypertension. However, caution is warranted when

interpreting any results from morphological studies of thecarotid bodies because it is likely that the subjects sufferedfrom several diseases; thus it is difficult to characterize therelationship between the anatomical changes found in thecarotid bodies and systemic hypertension independent ofimportant systemic vascular abnormalities, cardiac diseaseand other comorbidities. Whether hypertension precedescarotid body ischaemia and hyperplasia or whetherin fact ischaemia, chronic stimulation and subsequenthyperplasia of the carotid bodies causes hypertensionremains unclear. These changes are likely not to bethe principal cause in many aetiologies of essentialhypertension, but may be contributory to the facilitationof disease progression.

Chemoreflex resetting and chemoreceptor hyperactivity.Data that demonstrate the role of chemoreflex dysfunctionin OSA and its association with the development ofsystemic hypertension have only recently been uncoveredand are limited at best. Nevertheless, several human studieshave attempted to clarify this association. As postulatedabove, an increased chemoreflex gain and threshold foractivation of MSNA supports the view that the chemoreflexmay contribute to chronic elevation of MSNA. Two otherpossibilities that have been proposed are chemoreflexresetting and chemoreflex hyperactivity.

First, although not extensively researched, the conceptof chemoreflex resetting has been postulated by fewresearchers and warrants evaluation. Hedner et al. (1992a)suggested that as a result of chronic episodic hypoxia,the chemoreflex may reset to a higher level. Trzebski(1992) hypothesized that recurrent apnoeas as seen in OSAreset the chemoreflex to a higher level, thus increasingsympathetic outflow during normoxia or mild hypoxiaand initiating hypertension. Tafil-Klawe et al. (1991)found a decrease in ventilation after inactivation of thecarotid chemoreceptors in hypertensive OSA patients, thusindicating an increase in resting peripheral chemoreceptordrive in this population. Although these studies are limited,there is the suggestion that a process of resetting mayalso occur with sleep apnoea (Garcia-Rio et al. 2000).Whether this occurs in the chemoreflex control of MSNAremains to be determined, and this hypothesis needsfurther investigation.

Second, as mentioned earlier, patients with OSAexhibit elevated MSNA and catecholamine levels duringwakefulness, which may result in part from thehyperactivity of the chemoreceptors. Loredo et al. (2001)found that hypertensive OSA patients had higher tonicchemoreceptor activity than OSA normotensives. Theyconcluded that tonic chemoreceptor activity has somerole in the development of systemic hypertension in OSApatients. The mechanism that would lead to a chronichyperactivity of the chemoreceptors is unknown, and this

C© 2007 The Authors. Journal compilation C© 2007 The Physiological Society

Exp Physiol 92.1 pp 45–50 Sleep apnoea and hypertension in humans 49

outcome could be a manifestation of resetting and/or achange in the threshold or gain of the control system.

Conclusion

It is well established that there is a link between OSA andhypertension; however, the physiological mechanism(s)explaining this relationship remains unclear. A growingbody of evidence supports the premise that a chemoreflex-mediated elevation of basal sympathetic activity may playan important role in the pathogenesis of hypertensionamong OSA patients. A chronic elevation of sympatheticactivity would logically lead to vascular remodelling andsustained changes in vascular reactivity and tone, whichare hallmark manifestations of essential hypertension.Nevertheless, further investigation into these potentiallinks and mechanisms are needed, and many are addressedin the accompanying review papers in this journal.

References

Carlson JT, Hedner JA, Elam M, Ejnell H, Sellgren J & Wallin G(1993). Augmented resting sympathetic activity in awakepatients with obstructive sleep apnea. Chest 103, 1763–1768.

Cutler MJ, Swift NM, Keller DM, Wasmund WL, Burk JR &Smith ML (2004a). Periods of intermittent hypoxic apneacan alter chemoreflex control of sympathetic nerve activity inhumans. Am J Physiol Heart Circ Physiol 287, H2054–H2060.

Cutler MJ, Swift NM, Keller DM, Wasmund WL & Smith ML(2004b). Hypoxia-mediated prolonged elevation ofsympathetic nerve activity after periods of intermittenthypoxic apnea. J Appl Physiol 96, 754–761.

Fletcher EC (2001). Invited review: Physiological consequencesof intermittent hypoxia: systemic blood pressure. J ApplPhysiol 90, 1600–1605.

Fletcher EC, Lesske J, Culman J, Miller CC & Unger T (1992a).Sympathetic denervation blocks blood pressure elevation inepisodic hypoxia. Hypertension 20, 612–619.

Fletcher EC, Lesske J, Qian W, Miller CC III & Unger T(1992b). Repetitive, episodic hypoxia causes diurnalelevation of blood pressure in rats. Hypertension 19,555–561.

Fletcher EC, Miller J, Schaaf JW & Fletcher JG (1987). Urinarycatecholamines before and after tracheostomy in patientswith obstructive sleep apnea and hypertension. Sleep 10,35–44.

Floras JS & Hara K (1993). Sympathoneural andhaemodynamic characteristics of young subjects with mildessential hypertension. J Hypertens 11, 647–655.

Foster GE, Poulin MJ & Hanly PJ (2007). Intermittent hypoxiaand vascular function: implications for obstructive sleepapnoea. Exp Physiol 92, 51–65.

Friedman M, O’Brien V & Ehrlich LW (1975). Calculations ofpulsatile flow through a branch: implications for thehemodynamics of atherogenesis. Circ Res 36, 277–285.

Garcia-Rio F, Racionero MA, Pino JM, Martinez I, Ortuno F,Villasante C & Villamor J (2000). Sleep apnea andhypertension. Chest 117, 1417–1425.

Grassi G, Seravalle G, Bertinieri G, Turri C, Dell’Oro R, StellaML & Mancia G (2000). Sympathetic and reflex alterationsin systo-diastolic and systolic hypertension of the elderly.J Hypertens 18, 587–593.

Habeck JO (1986). Morphological findings at the carotidbodies of humans suffering from different types of systemichypertension or severe lung diseases. Anat Anz 162, 17–27.

Habeck JO, Waller H & Protze J (1983). Pathological alterationsof the arterial vessels of the carotid bodies in hypertensivehumans. Dt Gesundh Wesen 38, 1970–1972.

Heath D, Smith P & Jago R (1982). Hyperplasia of the carotidbody. J Pathol 138, 115–127.

Hedner JA, Wilcox I, Laks L, Grunstein RR & Sullivan CE(1992a). A specific and potent pressor effect of hypoxia inpatients with sleep apnea. Am Rev Respir Dis 146,1240–1245.

Hedner JA, Ejnell H, Sellgren J, Hedner T & Wallin G (1988). Ishigh and fluctuating muscle nerve sympathetic activity in thesleep apnoea syndrome of pathogenetic importance for thedevelopment of hypertension? J Hypertens Suppl 6,5529–5531.

Howe A & Neil E (1972). Arterial chemoreceptors. InHandbook of Sensory Physiology, ed. Neil E, pp. 47–80.Springer-Verlag, Berlin.

Iturriaga R, Rey S & Del Rio R (2005). Cardiovascular andventilatory acclimatization induced by chronic intermittenthypoxia: a role for the carotid body in the pathophysiologyof sleep apnea. Biol Res 38, 335–340.

Kluge P (1985). Vascularization and morphology of carotidbodies in patients with essential hypertension. Acta PhysiolPol 36, 76–82.

Leuenberger U, Jacob E, Sweer L, Waravdekar N, Zwillich C &Sinoway L (1995). Surges of muscle sympathetic nerveactivity during obstructive apnea are linked to hypoxemia.J Appl Physiol 79, 581–588.

Leuenberger UA, Brubaker D, Quraishi S, Hogeman CS,Imadojemu VA & Gray KS (2005). Effects of intermittenthypoxia on sympathetic activity and blood pressure inhumans. Auton Neurosci 121, 87–93.

Loredo JS, Clausen JL, Nelesen RA, Ancoli-Israel S, Ziegler MG& Dimsdale JE (2001). Obstructive sleep apnea andhypertension: are peripheral chemoreceptors involved? MedHypotheses 56, 17–19.

Marrone O, Riccobono L, Salvaggio A, Mirabella A, Bonanno A& Bonsignore MR (1993). Catecholamines and bloodpressure in obstructive sleep apnea syndrome. Chest 103,722–727.

Matsukawa T, Gotoh E, Uneda S, Miyajima E, Shionoiri H,Tochikubo O & Ishii M (1991). Augmented sympatheticnerve activity in response to stressors in young borderlinehypertensive men. Acta Physiol Scand 141, 157–165.

Morgan BJ (1996). Acute and chronic cardiovascular responsesto sleep disordered breathing. Sleep 19, S206–S209.

Morgan BJ, Crabtree DC, Palta M & Skatrud JB (1995).Combined hypoxia and hypercapnia evokes long-lastingsympathetic activation in humans. J Appl Physiol 79,205–213.

Narkiewicz K & Somers VK (1997). The sympathetic nervoussystem and obstructive sleep apnea: implications forhypertension. J Hypertens 15, 1613–1619.

C© 2007 The Authors. Journal compilation C© 2007 The Physiological Society

50 M. L. Smith and C. F. Pacchia Exp Physiol 92.1 pp 45–50

Narkiewicz K, van de Borne PJ, Montano N, Dyken ME,Phillips BG & Somers VK (1998). Contribution of tonicchemoreflex activation to sympathetic activity and bloodpressure in patients with obstructive sleep apnea. Circulation97, 943–945.

Narkiewicz K, van de Borne PJ, Pesek CA, Dyken ME, MontanoN & Somers VK (1999). Selective potentiation of peripheralchemoreflex sensitivity in obstructive sleep apnea.Circulation 99, 1183–1189.

Peng YJ, Overholt JL, Kline D, Kumar GK & Prabhakar NR(2003). Induction of sensory long-term facilitation in thecarotid body by intermittent hypoxia: implications forrecurrent apneas. Proc Natl Acad Sci U S A 100,10073–10078.

Peng YJ & Prabhakar NR (2004). Effect of two paradigms ofchronic intermittent hypoxia on carotid body sensoryactivity. J Appl Physiol 96, 1236–1242.

Peppard PE, Young T, Palta M & Skatrud J (2000).Prospective study of the association betweensleep-disordered breathing and hypertension. N Engl J Med342, 1378–1384.

Przybylski J (1981). Do arterial chemoreceptors play a role inthe pathogenesis of hypertension? Med Hypotheses 7,127–131.

Przybylski J & Trzebski A (1980). Hyperactivity ofthe arterial chemoreflex as a factor in the mechanismof early spontaneous hypertension. Fed Proc 39, 1186.

Saito M, Mano T, Iwase S, Koga K, Abe H & Yamazaki Y (1988).Responses in muscle sympathetic activity to acute hypoxia inhumans. J Appl Physiol 65, 1548–1552.

Shoemaker JK, Vovk A & Cunningham DA (2002). Peripheralchemoreceptor contributions to sympathetic andcardiovascular responses during hypercapnia. Can J PhysiolPharmacol 80, 1136–1144.

Smith P, Jago R & Heath D (1982). Anatomical variation andquantitative histology of the normal and enlarged carotidbody. J Pathol 137, 287–304.

Smith ML, Niedermaier ON, Hardy SM, Decker MJ & StrohlKP (1996). Role of hypoxemia in sleep apnea-inducedsympathoexcitation. J Auton Nerv Syst 56, 184–190.

Somers VK, Dyken ME, Clary MP & Abboud FM (1995).Sympathetic neural mechanisms in obstructive sleep apnea.J Clin Invest 96, 1897–1904.

Somers VK, Mark AL, Zavala DC & Abboud FM (1989a).Influence of ventilation and hypocapnia on sympatheticnerve responses to hypoxia in normal humans. J Appl Physiol67, 2095–2100.

Somers VK, Mark AL, Zavala DC & Abboud FM (1989b).Contrasting effects of hypoxia and hypercapnia onventilation and sympathetic activity in humans. J ApplPhysiol 67, 2101–2106.

Tafil-Klawe M, Thiele AE, Raschke F, Mayer J, Peter JH & vonWichert W (1991). Peripheral chemoreceptor reflex inobstructive sleep apnea patients; a relationship betweenventilatory response to hypoxia and nocturnal bradycardiaduring apnea events. Pneumologie 45 (Suppl. 1),309–311.

Tamisier R, Nieto L, Anand A, Cunnington D & Weiss JW(2004). Sustained muscle sympathetic activity afterhypercapnic but not hypocapnic hypoxia in normal humans.Respir Physiol Neurobiol 141, 145–155.

Trzebski A (1992). Arterial chemoreceptor reflex andhypertension. Hypertension 19, 562–566.

Trzebski A, Tafil M, Zoltowski M & Przybylski J (1982).Increased sensitivity of the arterial chemoreceptor drive inyoung men with mild hypertension. Cardiovasc Res 16,163–172.

Weiss JW, Liu MD & Huang J (2007). Physiological basis for acausal relationship of human obstructive sleep apnoea tohypertension. J Physiol 92, 21–26.

Xie A, Skatrud JB, Crabtree DC, Puleo DS, Goodman BM &Morgan BJ (2000). Neurocirculatory consequences ofintermittent asphyxia in humans. J Appl Physiol 89,1333–1339.

C© 2007 The Authors. Journal compilation C© 2007 The Physiological Society