The Emerging Role of Pharmacotherapy in Obstructive Sleep Apnea

Abstract

Obstructive sleep apnea (OSA) is a prevalent pathology with current modalities of treatment including continuous positive airway pressure (CPAP), surgery, weight loss, hypoglossal nerve stimulation, and pharmacotherapy. While CPAP is the current standard treatment for OSA, lack of tolerance and side effects necessitate alternative modalities of treatment. Various pharmacologic agents exist with mechanisms that may target OSA. Early trials have demonstrated efficacy of noradrenergic-antimuscarinic combinations to stimulate the airway, promote pharyngeal muscle tone, and prevent airway collapse. These agents, which we discuss in detail, have demonstrated significant reductions in apnea-hypopnea index (AHI) and lowest oxygen saturations based on preliminary studies. Glucagon-like peptide 1 receptor agonists (GLP-1RA), which stimulate endogenous insulin, reducing glucagon release, and decreasing gastric emptying, have shown positive results for OSA patients through weight loss with reductions in AHI. In this narrative review article, we highlight the mechanisms, current data, and future potential for multiple drug classes, including respiratory stimulants and GLP-1RAs.

1. Introduction

Obstructive sleep apnea (OSA) is a highly prevalent condition. Although discrepancies in OSA prevalence have been attributed to diagnostic criteria, equipment, and subject criteria, 11 studies from 1993 to 2013 have demonstrated that OSA with an apnea-hypopnea index (AHI) ≥ 5 has a prevalence of 22% in men and 17% in women, and the prevalence of OSA has increased over time [1]. Current prevalence of OSA is estimated to be over 1 billion worldwide [2].

The first-line treatment modality for OSA is Continuous Positive Airway Pressure (CPAP). CPAP mechanically maintains patency of the upper airway [3]. Despite the validated effectiveness of CPAP, select patients continue to struggle with adherence [3]. For patients that struggle to adhere to CPAP, physicians have begun to explore alternative and adjunctive treatment modalities, including weight loss and exercise, positional therapy, oral appliances, surgery, hypoglossal nerve stimulation, myofunctional therapy, and pharmacotherapy [3,4]. Due to the variable presentation of OSA severity with a range of comorbidities and side effects from treatments, multiple treatment modalities are essential for comprehensive and effective management of OSA. While several studies have thoroughly assessed the efficacy of other modes of treatment, existing literature regarding pharmacological interventions for OSA is currently limited and conflicting in some cases. In this narrative review article, randomized control trials, retrospective studies, and systematic reviews were selected for analysis from PubMed and Cochrane databases pertaining to various categories of OSA pharmacotherapy; these medications were subsequently summarized based on their categorized mechanism of action. This literature search involved manuscripts and published abstracts up to June 2024 with the aforementioned study designs with key words including “obstructive sleep apnea”, “obstructive sleep apnea therapy”, “pharmacotherapy”, “glucagon-like-peptide-1 receptor agonists”, and “noradrenergic-antimuscarinic agents” and exclusion of duplicate publications. Ultimately, in this review, we will discuss the mechanisms, initial data, indications, and future potential for OSA pharmacotherapy, including combined noradrenergic and antimuscarinic agents, glucagon-like-peptide-1 agonists, and other categories of medications. Pharmacotherapy for OSA is an evolving field, and with continuously emerging data on the array of medications for OSA, there is a fundamental need for a current review article to synthesize and evaluate the numerous studies focused on treating this complex pathology.

2. Noradrenergic and Antimuscarinic Agents for Obstructive Sleep Apnea Treatment

Early clinical applications of noradrenergic and antimuscarinic agents combined for patients with OSA have yielded mixed results [5]. Several selected studies examining the role of noradrenergic and antimuscarinic agents for OSA and their results and conclusions are illustrated in Table 1. These agents are used as respiratory stimulants to prevent pharyngeal tissue collapse. Since pharyngeal dilator muscle activity has shown to be decreased during sleep with airway collapse, respiratory pharmacotherapy may stimulate motor neurons to enhance genioglossus and tensor veli palatini tone [5,6,7]. Moreover, use of a combined antimuscarinic and noradrenergic regiment has yielded a significant reduction in OSA severity compared to use of noradrenergic agents alone [5,8,9]. Side effects from respiratory stimulant pharmacotherapy include those similar to sympathomimetics such as xerostomia, nausea, fatigue, decreased appetite, and urinary retention [5].

Table 1. Methodologies, interventions, and conclusions of selected noradrenergic and antimuscarinic agent studies [5,9].

Study	Methodology	Pharmacologic Intervention	Sample Size	Conclusion Summary
Aishah et al. [10]	Double-blind, randomized control trial over 1 month	80/5 mg atomoxetine-oxybutynin, 40/5 mg atomoxetine-oxybutynin, 40/2.5 mg atomoxetine-oxybutynin	39 participants	In individuals with predominantly severe OSA, nightly noradrenergic/antimuscarinic combination therapy is well-tolerated with mild side-effects aligning with known side effects of the individual medication, including increased heart rate. AHI and hypoxic burden reduction by ~50% in the 80/5 mg group. No significant effect on blood pressure and measures of alertness were found.
Messineo et al. [11]	Double-blind, crossover randomized control trial on 2 nights with 1-week washout	10 mg zolpidem plus atomoxetine-oxybutynin (80/5 mg) compared to atomoxetine-oxybutynin plus placebo	12 participants	Zolpidem addition to use of atomoxetine-oxybutynin in OSA patients increased sleep efficiency and respiratory arousal threshold but did not significantly impact OSA severity or next-day sleepiness. Thus, zolpidem may mitigate the wake-promoting characteristics induced by atomoxetine to improve sleep efficiency.
Taranto-Montemurro et al. [12]	Double-blind, crossover randomized control trial for 1 night with 1-week washout	80/5 mg atomoxetine-oxybutynin	20 participants	Oral noradrenergic and antimuscarinic agent combinations prior to sleeping on one night significantly reduced OSA severity. Atomoxetine-oxybutynin reduced AHI and increased genioglossus responsiveness.
Taranto-Montemurro [13]	Double-blind crossover randomized control trial for 1 night with 1-week washout	80/4 mg atomoxetine-fesoterodine	12 participants	In individuals with OSA, although there was no effect on AHI in the whole group, in the six participants with mild-to-moderate airway collapsibility, there was a complete response. Atomoxetine-fesoterodine improved airway collapsibility compared to placebo when administered for one night.
Aishah et al. [14]	Double-blind, crossover randomized control trial for 2 nights with 1-week washout	80/5 mg atomoxetine-solifenacin succinate or 80/2 mg atomoxetine-biperiden hydrochloride	10 participants	Amongst individuals with predominantly severe OSA, neither drug combination altered AHI compared to placebo. Both noradrenergic regimens decreased loop gain by ~10%, while arousal threshold and sleep efficiency were unchanged. Atomoxetine-solifenacin was associated with milder respiratory events with reduced OSA frequency, increased hypopnea during nonrapid eye movement, and improved nadir oxygen compared to placebo. Atomoxetine-biperiden hydrochloride decreased sleepiness, while atomoxetine-solifenacin improved upper airway function.
Perger et al. [15]	Double-blind, crossover randomized control trial for 7 nights with 7–10-day washout	4/5 mg reboxetine-oxybutynin	16 participants	In individuals with OSA, reboxetine-oxybutynin reduced AHI from 49 events per hour to 18 events per hour in comparison to 39 events per hour for placebo. Additionally, home oximetry demonstrated improved oxygen desaturation index with reboxetine-oxybutynin compared to placebo, indicating decreased OSA severity.
Lim et al. [16]	Double-blind crossover randomized control trial with 1-week washout	4/20 mg reboxetine-hyoscine	12 participants	In individuals with OSA, reboxetine-hyoscine reduced AHI, increased nadir oxygen saturation, increased genioglossus muscle responsiveness during non-rapid eye movement sleep, and improved airway collapsibility with no alteration in arousal threshold.

A systematic review performed by Lee et al. focused on trials utilizing atomoxetine-oxybutynin, reboxetine-hyoscine, reboxetine-oxybutynin, atomoxetine-fesoterodine, and milnacipran-oxybutynin-duloxetine for OSA therapy [5]. The 8 trials examined in this analysis demonstrated significant mean differences between OSA patients taking these combined regimens and placebo with regard to AHI and nadir oxygen saturation, but no effect on sleep efficiency or arousal index [5]. Lee et al. concluded that these combined stimulatory agents provided a positive yet modest effect on OSA severity in a clinical context, indicating a potential use as a substitute or supplementary OSA intervention [5].

A second systematic review by Zha et al. examined an alternative set of trials of noradrenergic and antimuscarinic agent combinations, including atomoxetine-solifenacin succinate, atomoxetine-biperiden hydrochloride, reboxetine-hyoscine butylbromide, atomoxetine-oxybutynin, atomoxetine-oxybutynin-zolpidem, reboxetine-oxybutynin, atomoxetine-fesoterodine [9]. Similar to the findings by Le et al., this study found the data from these seven trials to significantly reduce AHI and nadir oxygen saturation as well as arousal index, with no significant difference in sleep efficiency [9]. Thus, these drugs achieved improved OSA through reduced airway collapsibility and loop gain with increased genioglossus muscle stimulation (assessed through genioglossus electromyographic activity), resulting in decreased hypoxic burden [9]. Ultimately, through maintained patency of pharyngeal dilator muscles, especially the genioglossus, stimulant therapy is felt to have a direct mechanism in also relieving airway obstruction. Although numerous noradrenergic-antimuscarinic combinations have been evaluated in clinical trials with some differing results, these agents have overall demonstrated some promise as an alternative OSA pharmacotherapy agent.

3. Glucagon-like Peptide-1 Agonists for Obstructive Sleep Apnea Treatment

Glucagon-like peptide 1 receptor agonists (GLP-1RA) have been increasingly used in type 2 diabetes and obesity for weight control and blood glucose management [17]. This class of agents, including liraglutide, exenatide, semaglutide, and lixisenatide, have received FDA approval for diabetes due to their role in regulating glucose homeostasis through reducing glucagon release and promoting weight loss through appetite suppression and slowed gastric emptying [18,19]. Due to the high overlap in concurrence and pathophysiology between obesity and OSA, GLP-1RAs have been investigated in recent years as a potent pharmacologic intervention for OSA [17]. The mechanism of this category of drugs on OSA is most likely via weight loss, although additional mechanisms cannot be excluded without further study. Since metabolic surgery for weight loss has been shown to significantly reduce OSA severity, the assumed mechanism of GLP-1RAs is through weight loss rather than the direct effect on OSA, which is inherently different from the mechanism of atomoxetine-oxybutynin. Several selected studies examining the role of GLP-1RAs for OSA with their results and conclusions are illustrated in Table 2. A novel scoping review by Le et al. identified 9 articles that examined the use of GLP-1RAs to improve OSA and hypopnea syndromes. This review has offered promising preliminary evidence that OSA may reduce OSA severity with well-tolerated efficacy and only minor gastrointestinal side-effects [17].

Table 2. Methodologies, interventions, and conclusions of selected GLP-1 agonist studies [17].

Study	Methodology	Pharmacologic Intervention	Sample Size	Conclusion Summary
Blackman et al. [20]	Randomized, double-blind trial for 32 weeks across 40 sites	Liraglutide (3.0 mg)	276 participants completed the trial (out of 359 randomized participants)	Liraglutide (as a supplement to exercise and diet) yielded larger reductions compared to placebo in AHI, HbA1c, systolic blood pressure, and body weight in obese patients with moderative-severe obstructive sleep apnea.
Amin et al. [21]	Randomized control trial for 4 weeks	GLP-1RAs (unspecified, 0.6 mg daily initially increased by 0.6 mg weekly until a maximum dose of 1.8 mg is reached)	27 participants	Among patients with OSA, treatment of GLP-1 receptor agonists for 4 weeks is associated with significant OSA severity improvement, as measured by reduction in AHI.
Liu et al. [22]	Retrospective trial for 6 months	Liraglutide compared to traditional glucose-lowering medications	92 participants	In patients with types 2 diabetes mellitus and OSA, liraglutide may improve diabetic neuropathy and OSA severity with no impact on diabetic retinopathy or nephropathy.
O’Donnell et al. [23]	Randomized proof-of-concept study for 24 weeks	Liraglutide compared to CPAP and liraglutide-CPAP combination	30 participants	In individuals with OSA, CPAP, rather than weight loss from GLP1-agonist decreases unstable coronary plaque volume and improves vascular inflammation with improved endothelial function and decreased C-reactive protein in the CPAP therapy group.
Jiang et al. [24]	Two-center prospective randomized control trial	CPAP and drug treatment including liraglutide compared to CPAP and drug treatment without liraglutide (Liraglutide initially 0.6 mg/day and increased to 1.2 mg/day after week 1 and 1.8 mg/day after week 2)	90 participants	Liraglutide combined with CPAP improved OSA severity, decreased BMI, reduced systolic blood pressure, and improved hypoxia/AHI scores in patients with severe OSA and type 2 diabetes mellitus.
Malhotra et al. [25]	Two phase 3, double-blind randomized, controlled trials for 52 weeks across 60 sites	Tirzepatide (10 mg or 15 mg)	469 participants across both trials	Tirzepatide significantly decreased hypoxic burden, AHI, body weight, systolic blood pressure, and high-sensitivity C-reactive protein concentration with improved sleep outcomes in moderate-severe OSA patients.

The prospective study by Blackman et al. performed a randomized, double-blind trial in non-diabetic participants with obesity and moderate or severe OSA who were unwilling or unable to receive CPAP treatment [20]. For a 32-week period, participants were randomized to the GLP-1RA liraglutide (3.0 mg) or placebo as a supplement to diet and exercise [20]. The 3.0 dose of liraglutide was selected due to approval of 3.0 mg liraglutide for weight management purposes [20]. Over the course of the trial, the liraglutide regimen was well-tolerated and resulted in significantly greater decreases in AHI as body weight, hemoglobin A1c, and systolic blood pressure [20]. These results are corroborated by the 4-week trial performed by Amin et al., which utilized GLP-1RAs on 27 participants to determine that 70% of the participants demonstrated a decline in AHI by 44% and significant improvement in OSA severity over this duration [21].

Liu et al. evaluated the efficacy of GL-1RAs on OSA and diabetic microangiopathy specifically in 239 patients with type 2 diabetes mellitus with OSA [22]. This trial compared the use of liraglutide with conventional hypoglycemic drugs in the control group [22]. After 6 months of treatment, measurements of body mass index, HbA1c, AHI, systolic blood pressure, and HbA1c were significantly decreased, and diabetic microangiopathy was improved in the treatment group [22]. Moreover, the AHI was directly correlated with body mass index (BMI), waist circumference, and HbA1c measurements over the course of the study, indicating the complex interplay between these variables and the potential use of liraglutide for improvement of sleep-disordered breathing and diabetic microangiopathy [22].

Two prospective studies have evaluated the efficacy of GLP-1RAs in comparison to CPAP with differing results [23,24]. O’Donnell et al. performed a trial comparing CPAP and liraglutide isolated and combined for 24 weeks in 30 non-diabetic patients with OSA [23]. This study found that CPAP alone and combined with liraglutide provided significantly greater reduction in AHI and vascular inflammation with reduced C-reactive protein levels and improved endothelial health than liraglutide alone, thereby providing evidence for the traditional CPAP therapy [23]. In contrast, Jiang et al. conducted a prospective randomized controlled study amongst patients with type 2 diabetes mellitus with severe OSA assigned to either CPAP with or without liraglutide [24]. In contrast to the trial by O’Donnell, this study demonstrated significantly lower AHI, hypoxia, BMI, and systolic blood pressure in the liraglutide treatment group than the CPAP-only group, with no differences in side effects [24].

Lastly, the most recent SURMOUNT-OSA trials have yielded some of the most extensive and longitudinal data regarding GLP-1 agonist use for OSA [25]. This study involved two 52-week, phase 3 randomized, controlled trials across 60 sites to assess the effectiveness of maximum dosages of tirzepatide in adults with moderate-to-severe OSA compared to individuals receiving placebo [25]. Malhotra et al. concluded that across patients with OSA and obesity, tirzepatide significantly decreased hypoxic burden, AHI, body weight, systolic blood pressure, and high-sensitivity C-reactive protein concentration with improved sleep outcomes [25]. Notably, the mean AHI change at week 52 was −25.3 events per hour with tirzepatide and −5.3 events per hour with placebo during trial 1; in trial 2, mean AHI at week 52 was −29.3 events per hour with tirzepatide and −5.5 events per hour with placebo [25]. Ultimately, the SURMOUNT-OSA trials have yielded concrete results due to the multicenter and longitudinal nature of the study and offer promising initial results for the potential clinical use of GLP-1 agonists for OSA [25].

4. Alternative Medications for Obstructive Sleep Apnea Use

Various other medications have been explored to a limited extent with various targets and mechanisms of action in treating OSA. Recent studies have examined medications that alter serotonergic and cholinergic function to enhance pharyngeal muscle tone, including antiemetics (ondansetron) and antidepressants (fluoxetine, mirtazapine, desipramine, trazodone, paroxetine, and protriptyline), which affect the serotonin pathway with some positive outcomes with fluoxetine on AHI and increased respiratory arousal threshold with trazodone [26,27]. Trials exploring the effect of the parasympathomimetics physostigmine and donepezil to regulate muscle tension by motor end-plate depolarization have demonstrated some reduction in AHI [26]. Additionally, certain OSA treatment studies have focused on the use of respiratory stimulants with a central effect on the medullary respiratory center, including theophylline, almitrine, doxapram, and aminophylline, as well as acetazolamide through metabolic acidosis, resulting in negative or minor AHI effects due to decreased sleep quality [26]. However, these studies demonstrated null or minor impacts on AHI [26]. Lastly, some studies have examined the use of the parasympathomimetic donepezil or hypnotic/sedative agents, such as eszopiclone, zopiclone, tiagabine, and temazepam to control the arousal-threshold [26,27]. While some studies have yielded an increased arousal threshold, the overall literature is conflicting [26]. The specific mechanisms of these agents as well as the noradrenergic-antimuscarinic agents and GLP-1RAs are summarized in Table 3.

Table 3. Mechanisms of selected obstructive sleep apnea pharmacotherapy.

Drug Category	Notable Examples	Pharmacologic Effect	Clinical Effect on Obstructive Sleep Apnea
Noradrenergic Agents [5,9]	Atomoxetine Reboxetine	Selective norepinephrine reuptake inhibitors (SNRIs)- inhibits the norepinephrine transporter protein to promote increased synaptic norepinephrine concentration	Potentially promotes pharyngeal dilator muscle activity to provide patency and reduce collapse of airway
Antimuscarinic Agents [5,9]	Oxybutynin Fesoterodine, Solifenacin succinate Hyoscine Biperiden hydrochloride	Competitively antagonizes muscarinic acetylcholine receptors	Potentially promotes pharyngeal dilator muscle activity to provide patency and reduce collapse of airway
Glucagon-like-peptide-1 Agonists [17]	Liraglutide Tirzepatide Exenatide Semaglutide Lixisenatide	Agonizes the GLP-1 receptor, leading to appetite suppression/slowed gastric emptying, decreased glucagon release, and increased endogenous insulin release	Potentially promote weight loss to alleviate airway collapse due to correlation between OSA and obesity
Acetylcholinesterase inhibitors [25,26]	Donepezil Physostigmine	Inhibits acetylcholinesterase, leading to increased synaptic acetylcholine concentration	Potentially influence pharyngeal muscle depolarization and contraction at neuromuscular junction
Serotonergics [25,26]	Fluoxetine Paroxetine Trazodone	Inhibit serotonin reuptake to promote increased synaptic serotonin concentration	Potentially increase pharyngeal motor contraction and tone
Central respiratory stimulants [25,26]	Theophylline (bronchodilator) Aminophylline (bronchodilator) Doxapram (respiratory stimulant) Almitrine (repiratory stimulant)	Stimulates respiration centrally through activation of the medullary respiration center	Potentially improve respiratory patterns during sleep through continuous central stimulation to reduce apnea and hypoxia
Hypnotics/Sedatives [25,26]	Eszopiclone (nonbenzodiazepine hypnotic) Zopiclone (nonbenzodiazepine hypnotic) Tiagabine (GABA agonist) Temazepam (GABA agonist)	Promotes inhibitory neurotransmission effects to reduce arousal, especially during sleep	Potentially increase arousal threshold during sleep since arousal can disrupt sleep breathing patterns and contribute to obstruction

5. Discussion: The Promise of Pharmacotherapy in OSA

Ultimately, numerous pharmacologic agents exist for the management of OSA, with some of the most studied being noradrenergic-cholinergic drugs as well as GLP-1 agonists. Likely through modulating pharyngeal dilator muscle tone, noradrenergic-cholinergic agents help in improving airway patency, with evidence that certain combinations improve OSA. GLP1-agonists have also been shown to have positive effects on OSA, via weight loss, and other measures of diabetic and cardiovascular health through effects on glucagon secretion, gastric emptying, and endogenous insulin secretion. However, the therapeutic targets for OSA are not limited to these agents, and proposed agents that have been studied include serotonergic drugs, parasympathomimetics, central respiratory stimulants, and hypnotics/sedatives. In essence, the current exploration of pharmacotherapy is highly promising, notably GLP-1 agonists as an adjunctive therapy, yet further evaluation is critical in determining their efficacy as monotherapy, obtaining Food and Drug Administration Approval for OSA use, and developing a consensus guideline recommendation.

A multi-modal approach to treating OSA is logical considering the various endo and phenotypes of OSA and their overlap in a single individual at different ages and comorbidities. The wide array and severity of OSA symptoms has often been categorized into Disturbed Sleep Subtype, Excessive Sleepy Subtype, and Minimally Symptomatic Subtype, each of which has different implications for management [28,29]. For instance, the Minimally Symptomatic Subtype has limited clinical urgency, resulting in low CPAP adherence, so less invasive interventions may be implemented, such as weight loss and altered sleep positioning, to potentially promote reduced long-term cardiovascular risk [28,29,30]. Disturbed Sleep Subtype patients, who typically present with interrupted sleep and insomnia, may experience persistent symptoms despite CPAP and cognitive behavioral therapy [28,29,30]. The Excessively Sleepy Subtype of OSA consists of patients with high daytime sleepiness scores and significant cardiovascular burden [28,29,30]. Therefore, due to high prevalence of symptoms and cardiovascular disease risk, both Disturbed Sleep Subtype and Excessively Sleepy Subtype patients may benefit from alternative therapy when resistant to CPAP, including pharmacotherapy [28,29,30]. Both noradrenergic-cholinergic drugs and GLP-1 agonists are favorable candidates for future studies, and through further evaluation, there is a potential for developing personalized adjunctive or alternative treatment for refractory OSA. The frontier for OSA pharmacotherapy is still largely unexplored, and greater investigation into long-term outcomes and side effects with clinical trials of greater sample size is critical to better understanding the efficacy of these drugs in a clinical context.