Efficacy and Safety of Glucagon-like Peptide-1 Receptor Agonists for Treatment of Obstructive Sleep Apnea: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Abstract

Objective: To review and synthesize the current literature of clinical trials that investigated the efficacy and safety of glucagon-like peptide-1 receptor agonists (GLP-1RAs) in people with obstructive sleep apnea (OSA). Method: MEDLINE, EMBASE, Cochrane Library, and PsycINFO were searched for randomized controlled trials (RCTs) in which GLP-1RAs were used to treat people diagnosed with OSA. This systematic review and meta-analysis complied with PRISMA 2020 guidelines and was registered on PROSPERO (CRD42024537280). A random effects model was used for meta-analysis to assess changes in OSA as measured by the apnea–hypopnea index (AHI) compared to continuous positive airway pressure (CPAP) or placebo controls. The standardized mean difference (SMD) and risk ratio (RR) were computed for continuous and binary outcomes. Variability between studies, risk of bias, subgroup analysis, and leave-one-out analysis were completed. Results: Five studies were included (N = 1023; 511 GLP-1RA and 512 control). Two trials used tirzepatide and four studies used liraglutide as the GLP-1RA. Six studies showed a decrease in AHI with an SMD of −14.5 events per hour (95%CI = −24.73 to −4.21; I² = 96.3%). When compared to placebo, GLP-1RA treatment had a significant reduction in AHI (SMD = −0.69; 95%CI = −1.10 to −0.26; p = 0.001; I² = 88.0%). When compared to CPAP, no significant difference in the reduction of AHI was found. No evidence of publication bias was found. Compared to control, there was no significant difference in serious adverse events (RR = 0.89; 95%CI = 0.50 to 1.57; p = 0.68; I² = 20.93%). Conclusions: People with psychiatric disorders may also experience comorbid OSA that can impact their quality of life, which may perpetuate psychiatric symptoms of depression. GLP-1RAs may provide therapeutic potential in the treatment of OSA in addition to their cardioprotective effects. Current studies are limited by small sample sizes, lack of blinding, and short duration. Future studies will require further investigation in long-term efficacy and safety.

1. Introduction

Obstructive sleep apnea (OSA) is a common sleep-related breathing disorder that affects approximately 1.36 billion people globally as of 2019 (936 million with mild to severe OSA and 425 million with moderate to severe OSA) [1]. This is further perpetuated by risk factors such as increasing age, higher BMI, and male sex [2]. Symptomatically, OSA manifests with or without symptoms and is associated with serious health consequences, such as cardiovascular disease, cerebrovascular disease, effects on cognition, and an increased risk of all-cause mortality [3,4,5].

Furthermore, although the direct economic costs of OSA have not been calculated, it is estimated to have economic effects in the billions [6]. The excessive daytime sleepiness associated with it contributes to indirect economic effects, such as motor vehicle collisions, accident-related near misses, low mood, decreased work productivity, and quality of life [7]. In the United States alone, diagnosing and treating OSA costs approximately USD 12.4 billion in 2015 [8].

Given the significant healthcare effects as sequelae of OSA, treatment is of utmost importance. Currently, the gold standard treatment option for OSA includes positive airway pressure (PAP), mainly continuous PAP (CPAP) [9]. However, PAP is not always well tolerated due to various reasons including patient preference, claustrophobia, and socioeconomic factors [10,11,12,13]. Non-adherence to CPAP is substantial, estimated at over 34% [14]. Other treatments for those who do not tolerate PAP include oral appliance therapy (OAT), lifestyle changes (such as weight loss), or surgery (i.e., bariatric surgery or airway surgery for abnormal airways) [15,16,17]. Although OAT, lifestyle changes, and surgery interventions can be recommended as treatment or adjuncts in the treatment of OSA, the role of medications remains unclear. Medications are thought to treat OSA by causing weight loss, which is a standard recommendation for people with OSA. Currently, there is no recommended pharmacotherapy for OSA.

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are being investigated in people with OSA to help achieve and maintain weight loss without bariatric surgery [18]. GLP-1RAs, originally a class of medications used in the treatment of type 2 diabetes mellitus (T2DM), work by mimicking the actions of GLP-1, which stimulates insulin secretion. In addition to improving glycemic control, they contribute to weight loss in individuals with and without T2DM by reducing food intake without affecting energy expenditure [19]. As a result, in 2021, the Food and Drug Administration (FDA) approved GLP-1RAs (e.g., semaglutide, tirzepatide, and liraglutide) for chronic weight management in adults with obesity or who are overweight with at least one weight-related condition [20]. Given this development in pharmacotherapies for weight management, there is a need for evidence synthesis of the literature on GLP-1RAs and their effects in people with OSA.

To our knowledge, there is no quantitative evidence synthesis of the effects of GLP-1RAs on OSA–related symptoms. This systematic review and meta-analysis aimed to address this knowledge gap by synthesizing the current evidence of GLP-1RAs for adults with OSA.

2. Methods

This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [21] (Table S1). An a priori protocol was registered on PROSPERO (CRD42024537280).

2.1. Searches and Inclusion Criteria

The databases MEDLINE, EMBASE, PubMED, Cochrane Library, and PsycINFO were searched (on 13 July 2024) for RCTs investigating the effects of GLP-1RAs on sleep apnea symptoms in people diagnosed with OSA (Table S2). A health sciences librarian (RS) and relevant content experts in sleep medicine (MM) and evidence synthesis (MS) were involved in the creation and optimization of the search strategy. There were no restrictions on publication date or language of publication.

The primary outcome was the effect of GLP-1RAs on objective measurements of OSA such as the apnea–hypopnea Index (AHI). Secondary outcomes include scales of OSA severity including the Respiratory Disturbance Index (RDI), Oxygen Desaturation Index (ODI), and measures of symptoms of OSA such as the Epworth Sleepiness Scale (ESS) in people with a diagnosis of OSA. Secondary outcomes also included author-defined serious adverse events (SAEs).

Studies included were as follows: (1) RCTs, (2) investigations of the effects of GLP-1RAs in people diagnosed with OSA using polysomnography (PSG) and with severity determined by AHI, RDI, ODI, or symptoms using ESS, (3) comparisons of the usage of GLP-1RAs to all possible OSA treatment comparators including but not limited to treatment as usual (i.e., CPAP), other OSA treatments (OAT, non-GLP-1RA medications, and surgery), and placebo.

2.2. Study Screening

Covidence was used to conduct title, abstract, and full-text screening [22]. Screening was completed in duplicate by a pair of blinded investigators (SW, CZ, and NF) who independently screened for eligible studies. Discrepancies were resolved by a third individual investigator to reach a consensus. The reference list of relevant studies and the grey literature were also searched, using the same systematic approach.

2.3. Data Extraction

Data extraction was completed in duplicate by two independent investigators (CZ and SS) using a Microsoft Excel spreadsheet designed a priori. Discrepancies between investigators during this phase were resolved through consensus. The following information was extracted: bibliographic details such as standard identifiers (DOI or PMID), author’s last name, year of publication, clinical trial registry number, clinical trial sponsor, country of study, and demographic information (including sex, age, weight, BMI, and underlying medical and psychiatric comorbidities). OSA outcomes including AHI, ODI, RDI, and ESS scores were extracted as well as serious adverse events.

2.4. Quality Assessment

The Cochrane Risk of Bias Tool 2 (RoB2) was used by the investigators (CZ and SS) in duplicate to assess the methodological quality and risk of bias of included RCTs [23]. Discrepancies in quality assessment were resolved through consensus.

2.5. Statistical Analysis

Statistical analysis and meta-analysis were completed using Comprehensive Meta-Analysis (CMA) v4 [24]. The random effects model was employed to account for anticipated heterogeneity. The Hartung–Knapp–Sidik–Jonkman (HKSJ) method was used since there were less than 10 trials [25]. Standardized mean difference (SMD) with a 95% confidence interval (95% CI) was used as the measure of effect size in which there were continuous variables. Dichotomous variables were analyzed separately. Effect size was calculated by Cohen’s d [26] for continuous variables and by risk ratio (RR) for dichotomous outcomes. The inverse variance method was used to weigh the pooled effect size in the random effects model [27].

Heterogeneity was assessed using I² where >50% was used as the threshold for significant heterogeneity [28,29].

Leave-one-out analysis was conducted to assess consistency of the effect size.

Publication bias was assessed using a visual inspection of the funnel plot and the Egger’s regression test (p ≤ 0.10) [30].

Subgroup analysis was conducted if 3 or more studies with relevant data were available for the following groups: industry sponsoring, study quality, comparators, and comorbid diabetes.

3. Results

3.1. Search Results and Baseline Characteristics of Included Trials

There were 318 records retrieved from the initial search (Figure S1). After removal of duplicates, 247 studies were included for title/abstract screening in which 19 studies underwent full-text screening. After full-text screening, five studies were included for data extraction. Malhotra et al. (2024) [31] was one study with two trials. Four studies investigated the usage of liraglutide and one study used tirzepatide. In total there were 1023 participants with 511 receiving treatment with a GLP-1RA and 512 in the control group with either placebo or CPAP therapy. Among all the participants, 281 were female (27.46%). The mean age of the treatment group was 49.5 ± 10.1 and the control group was 50.1 ± 10.3. In terms of outcomes, only AHI and ODI were reported.

Blackman et al. (2016) [32], Malhotra et al. (2024) [31], and O’Donnell et al. (2023) [33] studied populations with OSA without a diagnosis of type II diabetes mellitus. Jiang et al. (2023) [34] and Rawcliffe et al. (2020) [35] studied OSA populations with a comorbid diagnosis of type II diabetes mellitus. In terms of medical comorbidities, Blackman et al. (2016) [32] reported the following comorbidities: prediabetes (63.2% of all participants), dyslipidemia (33.4%), hypertension (42.3%), and depression or a mood disorder (16.4%). Jiang et al. (2023) reported hypertension in 86.5% of participants. Malhotra et al. (2024) reported prediabetes (61.7%), dyslipidemia (83.4%), and hypertension (76.3%) in participants. O’Donnell et al. (2023) reported hypertension (23.3%) and ischemic heart disease (6.67%) in participants. Rawcliffe et al. (2020) [35] did not report any comorbidities. Additional study characteristics and participant details can be found in

Table 1. Study characteristics of included studies.

Study (Country, Trial #, and Sponsor)	Study Design and Duration of Trial	Treatment Participants (n, Mean Age, Sex)	GLP-1RA Investigated, Dosing, and Duration	Control Participants (n, Mean Age, Sex)	Type of Control
Blackman 2016 [32] (Canada and USA, NCT01557166, Novo Nordisk A/S)	Double-blind, placebo-controlled parallel-group (32 week and 2 weeks post-trial)	n = 180 Mean age: 48.6 ± 9.9 Female: 51	Liraglutide starting at 0.6 mg/day and increased over 4 weeks to 3 mg, which was maintained for 28 weeks with diet and exercise counseling	n = 179 Mean age: 48.4 ± 9.5 Female: 50	Placebo dose-volume equivalent with diet and exercise counseling
Jiang 2023 [34] (China, 20180329, unknown)	Non-blinded randomized controlled trial (12 weeks)	n = 44 Mean age: 55.7  ± 7.4 Female: 10   Participants all had comorbid diabetes mellitus type 2	CPAP + liraglutide starting at 0.6 mg/day up to 1.2–1.8 mg/day at week two.	n = 45 Mean age: 54.8  ±  5.5 Female: 16   Participants all had comorbid diabetes mellitus type 2	CPAP therapy standalone
Malhotra 2024 [31] (USA, NCT05412004. Eli Lilly)	Double-blind, randomized placebo-controlled, parallel-group (52 weeks)	Trial 1: n = 114 Mean age: 47.2 ± 11.0 Female: 36 Trial 2: n = 120 Mean age: 50.8 ± 10.7 Female: 33	Tirzepatide starting at 2.5 mg/week and was increased by 2.5 mg every 4 weeks until the participant reached the maximum tolerated dose of 10 mg or 15 mg in week 20.	Trial 1: n = 120 Mean age: 48.4 ± 11.9 Female: 41 Trial 2: n = 115 Mean age: 52.7 ± 11.3 Female: 32	Subcutaneous placebo
O’Donnell 2023 [33] (Ireland, NCT04186494, St Vincent’s University Hospital)	Open-label, randomized placebo-controlled, parallel-group (24 weeks)	n = 10 Mean age: 50  ±  7 Female: 6 CPAP + liraglutide: n = 10 Mean age: 50  ± 5 Female: 3	Liraglutide standalone starting at 0.6 mg/day with a dose increase each week up to 3.0 mg at week 5. CPAP + liraglutide is the same as above.	n = 10 Mean age: 51  ± 8 Female: 1	CPAP therapy standalone
Rawcliffe 2020 [35] (United Kingdom, EudraCT # 2014-000988-41, University of Liverpool) [Trial completed with data available but not currently published]	Non-blinded, randomized, placebo-controlled (26 weeks)	Arm C: n = 33 Median age: 55 (45 to 61) Female: 19 Arm D: n = 33 Median age: 54 (46 to 61) Female: 11 Participants all had comorbid diabetes mellitus type 2	Arm C: liraglutide standalone starting at 0.6 mg/day with a dose increase each week up to 1.8 mg at week 3. Arm D: CPAP + liraglutide is the same as above	Arm A: n = 33 Median age: 52 (48 to 57) Female: 17 Arm B: n = 33 Median age: 57 (49 to 60) Female: 20 Participants all had comorbid diabetes mellitus type 2	Arm A: No treatment Arm B: CPAP therapy standalone

. Common reasons for exclusion between the included studies include prior usage of GLP-1RAs, type I diabetes, history of chronic or acute pancreatitis, thyroid dysfunction, and medications that affect weight or sleep. Table S5 highlights the exclusion criteria of the included studies.

3.2. GLP-1RAs Overall

Four trials compared the usage of GLP-1RA on OSA, and when compared to placebo, there was a medium effect size in the reduction of OSA as measured with AHI (k = 4; SMD = −0.68; 95%CI = −1.10 to −0.26; p < 0.01; I² = 87.96%; Figure 1). Six trials showed a decrease in AHI with an estimated mean difference of −14.5 events per hour (95%CI = −24.73 to −4.21; I² = 96.3%). Leave-one-out analysis showed a consistent effect size that was statistically significant regardless of the removal of any single study. Subgroup analysis showed no significant differences based on risk of bias (p = 0.60) and industry sponsorship (p = 0.60). Funnel plot analysis showed a relatively symmetrical distribution of points that represents each study (Figure S2), which would indicate no major publication bias. All studies fell within the pseudo 95% limits in the plot, which are represented by the diagonal lines in the plot. Egger’s p revealed no evidence of publication bias (Egger’s p = 0.33).

3.3. Risk of Bias

Three of the studies were at high risk of bias while two were low risk (Figure S8). The area of highest risk of bias in those three studies were in domain 2 (bias due to deviations from intended interventions). A lack of blinding was the main reason for the high risk of bias in these three studies within domain 2.

3.4. Liraglutide

Four studies were used in the meta-analysis of the effects of liraglutide on OSA with 554 participants (310 receiving liraglutide and 244 in the control group). When compared to placebo, liraglutide had a small but significant reduction in AHI (k = 2; SMD = −0.31; 95%CI = −0.51 to −0.10; p < 0.01; I² = 4.87%; Figure S3). Four studies showed a decrease in AHI with an estimated mean difference of −7.75 events per hour (95%CI = −10.62 to −4.89; I² = 0%). When compared to CPAP therapy, liraglutide had a medium increase but not a significant effect size (k = 3; SMD = 0.47; 95%CI = −1.04 to 1.97, p = 0.54; I² = 91.05%; Figure S4). The ODI was not changed by liraglutide treatment when compared to control (k = 3; SMD = −0.08; 95%CI = −0.51 to 0.68; p = 0.79; I² = 77.48%; Figure S5). Due to the limited number of studies, sensitivity analysis and publication bias could not be conducted.

3.5. Tirzepatide

Two studies were used in the meta-analysis with 462 participants (234 receiving tirzepatide and 235 receiving placebo). Compared to placebo, tirzepatide had a large and significant effect size reduction in AHI (k = 2; SMD = −0.966; 95%CI = −1.16 to −0.78; p < 0.01; I² = 0.01%; Figure S6). Two studies showed a decrease in AHI with an estimated mean difference of −27.34 events per hour (95%CI = −31.26 to −23.42; I² = 48.6%). However, it should be noted that these two studies were by the same investigator (Malhotra, 2024) [31]. Due to the limited number of studies, sensitivity analysis and publication bias could not be conducted. There were no ODI outcomes reported.

3.6. Serious Adverse Events

No deaths were reported by any of the studies included. In terms of serious adverse events (SAEs), four studies reported serious adverse events. In total there were 54 SAEs (24 from the GLP-1RA group and 30 from control). When compared to control groups, there was no reduction in SAEs (k = 6; RR = 0.89; 95%CI = 0.50 to 1.57; p = 0.68; I² = 20.93%; Figure 2). Leave-one-out analysis showed an effect size that was not statistically significant regardless of the removal of any single study. The funnel plot and analysis were not significant for publication bias (Egger’s p = 0.12; Figure S7). Subgroup analysis showed no significant differences based on risk of bias (p = 0.14), control type (p = 0.90), comorbid diabetes (p = 0.14), and industry sponsorship (p = 0.118). The most frequently reported adverse events in GLP-1RA groups were nausea (n = 114), diarrhea (n = 85), constipation (n = 57), vomiting (n = 44), dyspepsia (n = 31), nasopharyngitis (n = 31), upper respiratory tract infection (n = 30), headache (n = 25), and gastroesophageal reflux (n = 25). In terms of discontinuation, Blackman et al. (2016) [32] reported 46 participants in the liraglutide arm that discontinued (20 were due to adverse events and 1 withdrew due to a treatment emergent adverse event) compared to 37 in the control arm (6 due to adverse events). Malhotra et al. (2024) [31] reported 9 withdrawing from the tirzepatide arm due to treatment emergent adverse events and 10 in the control group. The other three studies did not report on discontinuation from their studies.

4. Discussion

This systematic review and meta-analysis investigated the efficacy and safety of GLP-1RAs for the treatment of OSA using synthesized evidence from five RCTs. It was found that compared to placebo, GLP-1RAs (liraglutide and tirzepatide) led to a reduction in OSA symptoms as measured by AHI that was statistically significant and medium in terms of effect size (k = 4; SMD = −0.69; 95%CI = −1.10 to −0.26; p < 0.01; I² = 87.96%). The six total trials showed a decrease in AHI with an estimated mean difference of −14.5 events per hour (95%CI = −24.73 to −4.21; I² = 96.3%). Moreover, the difference in AHI reduction was found to not be significant when compared to CPAP usage.

This finding of no significant difference in AHI reduction compared to CPAP should be highlighted because it reveals that GLP-1RAs can potentially improve OSA in those with obesity. Given the importance of the role of weight loss in the management of OSA, it is expected that the weight loss achieved from GLP-1RA usage would potentially improve symptoms of OSA. Although, GLP-1RAs would not address the apnea seen in structural or central causes. In addition to the improvements to OSA symptoms, usage of GLP-1RA usage, regardless of formulation, has been associated with a significant reduction in major adverse cardiovascular events, worsening kidney function, and all-cause mortality in people with T2DM [36,37]. Therefore, GLP-1RAs introduce a new therapeutic potential option for the treatment of OSA in obese or overweight individuals with or without T2DM. Currently, CPAP is a standard treatment option for OSA given the strong evidence available [10]. Despite the benefits of CPAP treatment, including the fact that it is nonpharmacological and nonprocedural, there are limitations to CPAP treatment including potential mask leaks, issues with adherence, technical issues with the machine, patient preference, and impact on quality of life of the patient and their partners [38,39]. Combination therapy of CPAP and GLP-1RA was investigated in Rawcliffe et al. (2020) [35] and O’Donnell et al. (2023) [34], but only Rawcliffe et al. (2020) [35] found a significant difference between combination therapy and GLP-1RAs. Further head-to-head trials could be investigated to assess a potential effect of combination therapy compared to GLP-1RAs alone. When considering treatment options for OSA, GLP-1RAs are not to replace CPAP treatment, rather they are a potential new avenue of treatment for people suffering from OSA who cannot tolerate CPAP and other standard forms of treatment with the added benefit of weight loss and cardiovascular benefits. Furthermore, GLP-1RAs can also potentially be used to allow for discontinuation from CPAP treatment for the aforementioned reasons. However, the interpretation of this finding must be interpreted with caution as it is a meta-analysis of three studies that were not trials of non-inferiority. Future RCTs should consider comparing GLP-1RAs to CPAP treatment directly and assess potential discontinuation from CPAP after starting GLP-1RAs.

From a safety perspective, our meta-analysis found no significant difference in SAEs between GLP-1RA treatment groups and control. Usage of GLP-1RAs has been found to have a significant increase in gastrointestinal (GI)-related adverse events (nausea, vomiting, and abdominal pain) [40,41]. In a meta-analysis that specifically investigated liraglutide and tirzepatide, there was an increase in GI–related adverse events [42,43]. However, there was no significant increase in SAEs when compared to control groups [42,43]. It should be noted for clinicians that GLP-1RA usage is cautioned in people with gastroparesis or severe gastroesophageal reflux disease, which requires dose adjustments and monitoring [44]. Additionally, GLP-1RAs (with the exception of exenatide and lixisenatide) are contraindicated in people with multiple endocrine neoplasia syndrome type 2, or a personal or family history of medullary thyroid carcinoma [45]. Furthermore, long-term safety data is needed in OSA populations. Although the safety of GLP-1RAs is well studied in type II diabetes mellitus populations, it is not well studied in OSA populations. Long-term safety can be extrapolated; however, it would not be disease specific. Future studies would benefit from longer term follow up in OSA populations to detect any safety issues in OSA–specific populations that may not necessarily have type II diabetes mellitus. As more clinical trials are conducted and the usage of GLP-1RAs increase, there will be more emerging evidence on the safety profiles of this class of medications.

There are multiple strengths to this systematic review and meta-analysis. First, to the best of the author’s knowledge, there is currently no systematic review and meta-analysis of all GLP-1RAs for the treatment of OSA adhering to a stringent methodology. Furthermore, this review used a robust and up-to-date search of the current literature across multiple databases to ensure that studies were not missed. Third, we used data and evidence from RCTs, which are the highest tier of evidence. Lastly, there was minimal heterogeneity in our meta-analysis in addition to sensitivity analysis that was conducted, which improves the quality of the meta-analysis and our findings.

However, the results of this review need to be interpreted with caution given the limitations of the current evidence. First, the small sample size of participants receiving GLP-1RA treatment, which can exaggerate the effect sizes of GLP-1RA treatment. Only two studies in this review had a sample size of over 100 participants. Studies with smaller sample sizes may allow for the assessment of the feasibility of the treatment, but it would not be appropriate to draw conclusions about efficacy. Additionally, smaller sample sizes may lead to underpowered studies in which the efficacy of GLP-1RA treatment for OSA cannot be accurately assessed. Future studies would benefit from larger sample sizes that are adequately powered with different population groups to allow for more generalizability and robustness of their findings. Secondly, three of the studies were deemed to be at high risk of bias due to a lack of blinding. The lack of blinding can introduce a risk of expectancy bias in participants; it can negatively impact participant retention and motivation to stay in study, and this also introduces observer bias as well. Future studies would benefit from blinding participants and assessing the robustness of their blinding procedures to minimize the risk of bias introduced from no blinding. Future studies can also consider blinding with active comparators such as stimulants or other medications that could produce similar appetite-suppressing effects of GLP-1RAs as there are no current studies doing so. Third, there is limited long-term follow-up data on GLP-1RA in OSA populations over an extended period of time [46,47,48,49]. Given the novelty of certain GLP-1RAs in the treatment of OSA, there is very limited long-term data available in this specific population. However, long-term data could be extrapolated from well-studied populations such as those in type II diabetes mellitus. Additionally, the longer follow-up data would allow for assessments of feasibility, acceptability, and assessment of ceiling effects of GLP-1RAs in OSA. Longer periods of observation in OSA populations would allow for investigating aspects such as adherence, weight gain associated with stopping treatment, and observing more rare adverse events such as pancreatitis, thyroid cancer, and pancreatic cancer. Furthermore, it would allow for gathering data on quality of life and cost-effectiveness of medications compared to CPAP. Future studies would benefit from extended periods of observation to investigate the aforementioned aspects.

5. Conclusions

GLP-1RAs for the treatment of OSA have shown promising initial results from clinical trials, showing a decrease in AHI. This systematic review and meta-analysis add to the current literature, reporting a significant improvement in OSA symptoms as measured by a reduction in AHI when compared to placebo. Additionally, there was no significant difference found when compared to CPAP. These emerging results provide a pharmacological avenue of treatment for people suffering from OSA who may encounter difficulties with conventional CPAP treatment. Future studies will benefit from larger and more diverse populations, more robust blinding of participants and using active comparators, and long-term follow-up data including quality of life and cost-effectiveness to provide more holistic data in the usage of GLP-1RAs for the treatment of OSA.