Factors that predispose to OSA include obesity, gender, age, ethnic (including genetic) factors, and craniofacial structure, and OSA may be aggravated by use of certain drugs and smoking. It is pathophysiologically characterized by repetitive episodes during sleep of upper airway narrowing and/or closure, accompanied by increased breathing efforts in attempts to overcome such narrowing/closure, also by arousals and/or outright wakenings from sleep, as well as attendant respiratory and cardiovascular perturbations such as hypoxia, systemic and pulmonary hypertension and tachy- and bradycardia. The adverse effects of OSAS are well documented, and include poor sleep quality and consequent neurobehavioral dysfunction, reduced daytime vigilance and excessive daytime sleepiness, and risk for motor vehicle and other accidents, and cardiovascular morbidity and mortality [3-8]. A full description of the epidemiology, the diagnosis and clinical correlates of OSA has been presented recently [1, 2, 9-12].
Non-pharmacological therapies for OSA
The most effective non-pharmacological therapy currently available is (nasal) continuous positive airway pressure (nCPAP), but even so this therapy may be unacceptable or be used irregularly in over 50 % of patients prescribed nCPAP . Tracheostomy is an effective surgical treatment for OSA but not currently recommended except in the most extreme circumstances. Other surgical procedures have included uvu-lopalatopharyngoplasty but substantial evidence for benefit is lacking . Facial reconstructive surgery has a limited role in individuals with OSA secondary to facial dysmorphia. Oral appliances including mandibular advancement splints (MAS) may have a role in treating mild or moderate degrees of OSA, but long-term compliance is uncertain, and occasionally dental malocclusion and tempero-mandibular joint dysfunction may eventuate with use of MAS [16, 17]. These non-pharmacological therapies of OSA share a number of features, such as variable efficacy, significant side-effect profile, potentially high cost and reliance on skilled technical intervention, and lack of patient acceptance, which in concert argue strongly for the promotion of effective pharmacological therapies for OSA.
Individuals with OSA usually have structural narrowing of the upper airway but are able to maintain upper airway patency in wakefulness, albeit with increased levels of genioglossus muscle (the major pharyngeal dilator) activity compared to controls . Upper airway obstruction in sleep is ultimately caused by processes that affect the motor control of the pharyngeal muscle dilators (Fig. 1), and that control is mediated through the actions of relevant neurotransmitters impacting on motor neurons particularly of the hypoglossal nerve. Thus, an important prelude to consideration of specific pharmacological therapies in OSA is to outline current understanding of the neuropharmacology of the upper airway.
There are many neurotransmitters present in the motor nuclei of upper airway dilator motor neurons in the brainstem, and more centrally in the central nervous system (CNS), which are implicated in the neural control of upper airway patency [19, 20]. Glycine and 7-aminobutyric acid contribute inhibitory influences on upper airway motor neuronal activity . Other neurotransmitters such as acetylcholine, glutamate, noradrenaline, thyrotropin-releasing hormone, substance P, vasopressin, oxytocin and orexin also have roles, but the pre-eminent excitatory neurotransmitter is serotonin (5-hydroxytryptamine, 5-HT) . The relative importance and interplay between these neurotransmitters has been studied in vitro in reduced motor neuron preparations, and in vivo in healthy animals including cats, rats and a natural animal model of sleep-disordered breathing, the English bulldog. It is likely that there will be some differences between the results of these studies and the interplay of brainstem upper airway motor neuron neurotransmitters in human adults with established sleep-disordered breathing.
Support for serotonin having a prominent role in the neurochemical basis of upper airway patency is provided by excitation of brainstem dilator motor neurons by local administration of serotonin [23-26], and conversely by reduction of brainstem motor neuron activity by local administration of serotonin antagonists [25, 27, 28]. Identified serotonergic brainstem motor neurons increase activity linearly with respiratory motor challenges , and nucleus raphe pallidus serotonin-containing motor neurons that innervate brainstem motor neurons implicated in upper airway dilator muscle activity become less active in non-rapid eye movement (NREM) sleep and virtually absent in REM sleep [30-32]. Microperfusion of serotonin into brainstem hypoglossal motor nuclei protects against sleep-related suppression of upper airway dilator muscle activity in NREM sleep, and attenuates the suppression seen in REM sleep . Systemic administration of serotonin antagonists in the English bulldog produces obstructive breathing in wakefulness , attesting to the importance of this neurotransmitter to the maintenance of airway patency in the wake state in this model. Importantly, the administration of serotonergic drugs (L-tryptophan and trazadone) in the English bulldog produces a dose-dependent reduction in measures of sleep-disordered breathing, more markedly in NREM sleep .
The neurochemical control of upper airway motor neurons is complex, and that complexity is significantly contributed to by the existence of at least 18 serotonin
Fig. 2. Schema of the neuronal circuitry that is currently believed to be involved in the pontine regulation of rapid eye movement (REM) sleep progressively disinhibits pontine cholinergic neurons preceding and during REM sleep progressively disinhibits pontine cholinergic neutones of the laterodorsal and pedunclopontine tegmental nuclei (LDT/PPT) via withdrawal of serotonin (5-HT)-mediated and noradrenaline-mediated inhibitory inputs. Activation of these LDT/PPT neurones then leads to increased acetylcholine (ACh) release into the pontine reticular formation, resulting in activation of the neuronal systems that mediate ascending and descending signs of REM sleep (e.g. cortical desynchronization end motor atonia, respectively). Exogenous application of a cholinergic agonist (e.g. carbachol) by microinjection into the pontine reticular formation is used to mimic this process end trigger REM-like neural events in reduced preparations (e.g. anaesthetized or decerebrate animals). Postural motor atonia in REM sleep is produced by postsynaptic inhibition of motor neurones by 7-aminobutyric acid (GABA) and glycine. Neurones of the medullary reticular formation are thought to drive this inhibition, themselves being driven by neurones in the pontine reticular formation (the reticular structures are indicated by the boxes). Whether hypoglossal (XII) motor neurones are also postsynaptically inhibited in REM sleep by similar mechanisms is uncertain. Hypoglossal motor neurones also receive excitarory inputs from the locus coeruleus complex and medullary raphe that many also contribute to reduced genioglossus muscle activity in sleep, especially REM sleep. Corelease of thyrotropin-releasing hormone (TRH) and substance P from raphe neurones may contribute to this process. The influences of other neural systems that are potentially modulated by sleep states are not included for clarity. See text (of original article) for more details. +, excitation; —, inhibition; M, muscarinie. Reproduced from Fig. 2 of 
receptor subtypes . Such diversity of receptor subtypes is played out across both the CNS and peripheral nervous system (PNS). The predominant receptor subtype in hypoglossal motor neurons is 5-HT2a, but 5-HT2c is also present, and both postsynaptic subtypes being excitatory . Other receptor subtypes are present in smaller quantities when measured by a semi-quantitative technique, and receptor subtypes such as 5-HT4, 5-HT6 and 5-HT7 may also have an excitatory role in upper airway motor neurons  (Fig. 2). Stimulation of a presynaptic receptor, 5-HTib, is inhibitory to hypoglossal neuron activity , thereby providing a local negative feedback loop. Furthermore, although not present directly on hypoglossal motor neurons, stimulation of 5-HT3 receptors on interneurons connecting with hypoglossal neurons likely has inhibitory effects on hypoglossal motor output . There are excitatory serotonergic effects on respiratory neurons at other points in the brainstem CNS involving the 5-HT2 and 5-HTia receptor subtypes [41,42].
In the PNS, specifically at the nodose (inferior vagal) ganglion, stimulation of 5-HT2 and 5-HT3 receptor subtypes suppresses respiration . Administration of the 5-HT3 antagonist ondansetron reduces CSA in rats through this peripheral effect [44, 45], and reduces sleep-disordered breathing in REM sleep in the English bulldog, without influencing changes in NREM sleep  (see section 'Ondansetron' below for its effects in humans).
There is some evidence that long-term intermittent hypoxia analogous to the hypoxic exposure of human cases of OSAS, may predispose to oxidative injury to upper airway brainstem neurons, i.e. , hypoglossal motor neurons, and thereby diminish potential serotonergic excitatory responsiveness. At least in Sprague-Dawley rats exposed to 3 weeks of intermittent hypoxia, unilateral serotonin and glutamate agonist and antagonist microinjections, respectively, into the hypoglossal motor nuclei showed reduced hypoglossal nerve responsiveness (log EC50) for serotonin and N-methyl-D-aspartate (NMDA) . These results may explain at least in part the modest or absent responses to serotonergic drug therapy in OSA patients (see below).
Serendipity led to an early trial of L-tryptophan in 15 patients with sleep-disordered breathing, but the study used non-uniform dosage schedules, was unblinded and non-placebo controlled. Nevertheless, there were encouraging reductions in markers of sleep respiratory disturbance particularly in NREM sleep . Subsequent reports of eosinophilic myalgic syndrome and life-threatening pulmonary hypertension with use of L-tryptophan  led to withdrawal of this drug preparation, and stymied interest in this general class of drug therapy for several years.
Buspirone is a partial 5-HTia agonist, and systemically is a respiratory stimulant; it has been used as an anxiolytic. In a small trial of five OSAS patients there was an overall modest reduction of the apnea index (though worsening in one of these patients) .
Specific 5-HT2A/2C receptor agonists
The 5-HT2a/2c receptor agonist [±]-2,5-dimethoxy-4-iodoaminophentamine improved upper airway collapsibility in Zucker rats, but had complex other effects, including increasing upstream airways resistance, while maintaining unchanged maximal airflow . Studies in humans are not available at this time.
Fluoxetine is a selective serotonin-reuptake inhibitor (SSRI) that produces a net increase in (post-synaptic motor neuron) serotonin delivery after 4-6 weeks of use. A double-blind, randomized cross-over trial compared fluoxetine to the tricyclic antidepressant agent protriptyline and placebo in 12 patients with sleep-disordered breathing . The group apnea-hypopnea index (AHI) improved with fluoxetine compared to placebo, but there was great variability of response and other measures of disordered sleep did not change. These potentially beneficial results in a small number of patients need to be replicated in well-designed larger studies to support a useful role in clinical practice.
Paroxetine is another SSRI that has undergone a trial in a small number of patients with mild to moderate sleep-disordered breathing, and been shown to produce modest decrements of the apnea index in NREM sleep only . Trazodone is a weak SSRI, and its metabolite is a powerful 5-HT2C agonist that can cross the blood-brain barrier. Its beneficial effects (in combination with L-tryptophan) on sleep-disordered breathing have been noted in the English bulldog , and also documented in a case report of a patient with olivopontocerebellar degeneration manifesting both obstructive and CSA events .
Mirtazapine is an antidepressant that increases both serotonin and noradrenaline by blockade of central a2 auto- and heteroreceptors; mirtazapine also blocks 5-HT2 and 5-HT3 serotonin receptor subtypes, and that former property may induce slow-wave sleep. Systemic administration of mirtazapine has been shown to increase genioglossus muscle activity in anesthetized rats in a dose-dependent manner . In a randomized, double-blind, cross-over trial of ten patients with OSA, mirtazapine at a dose of 15 mg reduced the AHI by 50 %, and the arousal index by some 29 % . Side-effects with use of mirtazapine include somnolence and hyperphagia/weight gain.
As mentioned above, the 5-HT3 antagonist ondansetron has a salutary effect on CSA in rats and on REM-related sleep-disordered breathing in the English bulldog model of OSA. However, in the only human trial of this drug in ten patients with moderate OSA, compared to placebo, there was no effect on sleep architecture nor on any index of sleep-disordered breathing , although the postulated tissue levels of active drug in this trial were an order of magnitude below that in the English bulldog study.
Interest has accrued in recent years in the endogenous cannabinoid neuromodula-tory system, whose effects are mediated by two recognized receptors, CB1 and CB2. Both exogenous and endogenous cannabinoid ligands may impact sleep/wake and autonomic behaviors, in part at least through interactions with serotonin receptor function. Exogenous A9-tetrahydrocannabinol and the endogenous cannabinoid lig-and oleamide both stabilized respiration in all sleep stages in instrumented Sprague-Dawley rats, significantly reducing the apnea index in both NREM and REM stage sleep ; both agents blocked serotonin-induced exacerbation of apnea consistent with a coupling between cannabinoids and specific serotonin receptors (5-HT3) in the PNS. The potential exists for human trials of exogenous cannabinoids in the pharmacological treatment of OSA, but such trials are lacking to date. Interestingly, the CB 1 selective blocker rimonabant is currently undergoing trial as a weight-loss and smoking-cessation agent , and therefore may have potential use in OSA patients with co-existing obesity.
Was this article helpful?