Combination Drug-Therapy for Patients With Untreated Obstructive Sleep Apnea

Study participants will undergo three 3-day drug regimens. On days 1 and 2 of each drug regimen, subjects will take the study drugs at home; on day 3 of each drug regimen subjects will take the study drugs as part of an overnight inlab sleep study (including assessments of sleepiness/alertness, sleep quality and blood pressure). Initially subjects will take dual-therapy (acetazolamide+eszopiclone) vs placebo in random order; if sleep apnea resolved with dual-therapy, then subjects will undergo an open-label single-drug regimen (acetazolamide), else an open-label triple-drug regimen (acetazolamide + eszopiclone + venlafaxine).

• Acetazolamide (Drug), also known as Diamox: Acetazolamide tablet (encapsulated)
• Eszopiclone (Drug), also known as Lunesta: Eszopiclone tablet (encapsulated)
• Placebo (Drug): Sugar capsule manufactured to match encapsulated Acetazolamide/Eszopiclone
• Venlafaxine (Drug), also known as Effexor: Venlafaxine capsule

Primary outcomes

• Apnea Hypopnea Index (AHI) [3 nights]: The AHI is a measure of sleep apnea severity and based on the American Academy of Sleep Medicine (AASM)-recommended criteria is defined as the number of apneas (no breathing for 10+ seconds) and hypopneas (reduced breathing for 10+ seconds associated with a >=3% desaturation or cortical arousal) per hour of sleep. To avoid confounding by sleep stages and positions across study nights the primary focus was on the AHI during supine non rapid eye movement (NREM) sleep. For comparability with other studies, we also explored the AASM-acceptable "AHI4", which defines hypopneas as reduced breathing for 10+ seconds associated with a >=4% desaturation.

Secondary outcomes

• SpO2 Nadir [3 nights]: The lowest measured blood oxygen saturation during the overnight sleep study measured in percent.
• Pathophysiological Traits: Vpassive, Vactive, Arousal Threshold [3 nights]: Pathophysiological traits were quantified as %Veupnea from polysomnography data using a validated algorithm.
• Pathophysiological Trait: Loop Gain [3 nights]: Loop Gain 1 was quantified from polysomnography data using a validated algorithm. This metric measures the increase in respiratory drive relative to a preceding drop in ventilation and thus is dimensionless (typical range is approximately 0.2 to 1.5, with values >0.7 being considered high loop gain, indicating ventilatory instability)
• Percent Responders [3 nights]: Full responders were defined as a drop in AHI>50% to <10/h.
• Blood Pressure [3 nights]: Systolic/Diastolic Blood Pressure (measured at rest in the morning following the overnight sleep study).
• Subjective Sleepiness: Stanford Sleepiness Scale (SSS) [3 nights]: Subjective sleepiness was assessed using the Stanford Sleepiness Scale (SSS) in the morning following the overnight sleep study. The score ranges from 1 to 7, with greater values indicating more sleepiness.
• Sleep Quality: PROMIS (Patient-Reported Outcomes Measurement Information System) Sleep Disturbance [3 nights]: Sleep quality was assessed based on a modified 8-question PROMIS Sleep Disturbance (SDA 8b) questionnaire in the morning following the overnight sleep study. The raw score ranges from 8 to 40 and is translated into a T-score, a standardized score with a mean of 50 and a standard deviation of 10. Greater T-scores indicate greater sleep disturbance.
• Psychomotor Vigilance: Response Speed [3 nights]: Vigilance was assessed using the 10-minute psychomotor vigilance test (PVT) in the morning following the overnight sleep study. Primary focus was on "response speed" 1/reaction time (1/RT) and lapses (reaction time >500ms).
• Psychomotor Vigilance: Lapses [3 nights]: Vigilance was assessed using the 10-minute psychomotor vigilance test (PVT) in the morning following the overnight sleep study. Primary focus was on "response speed" 1/reaction time (1/RT) and lapses (reaction time >500ms).

Sex: All, ages: 18 Years to 65 Years

Inclusion Criteria:

* BMI 18-40 kg/m2
* Untreated Moderate or Severe OSA (AHI during supine NREM sleep >15/h) with a fraction of hypopneas >25% of all events

Exclusion Criteria:

* Pregnancy
* Breastfeeding
* Prisoners
* Adherent with effective therapy for OSA
* Other known untreated sleep fragmenting disorder, such as periodic limb movement disorder, or narcolepsy
* Inability to sleep supine for overnight sleep studies
* Circadian rhythm disorder
* Unrevascularized coronary artery disease, angina, prior heart attack or stroke, congestive heart failure
* Uncontrolled hypertension (systolic blood pressure >160mmHg, diastolic blood pressure >95mmHg)
* Presence of tracheostomy
* Hospitalization within the past 90 days
* Prior peptic ulcer disease, esophageal varices, or gastrointestinal bleeding (< 5 years)
* Prior gastric bypass surgery
* Chronic liver disease or end-stage kidney disease
* Active illicit substance use or >2 oz daily alcohol use (i.e. >2 12 oz bottles of beers, >2 5 oz glasses of wine, >2 1.5 oz glasses of hard liquor such as spirits, gin, whiskey, etc.)
* Psychiatric disease, other than well controlled depression/anxiety
* Cognitive impairment, inability to provide consent, or inability to complete research procedures (e.g. questionnaires that are only available/validated in English)
* Chronically using study drugs or drugs with similar pharmacodynamic effects (acetazolamide - carbonic anhydrase inhibitors, eszopiclone - benzodiazepine receptor agonists, venlafaxine - serotonin/norepinephrine reuptake inhibitors and other antidepressants)
* Regular use of medications known to affect control of breathing (opioids, benzodiazepines, theophylline)
* Contraindications to taking study drugs, including allergies to any of the drugs or sulfa allergy; concomitant use of antidepressants, opioids, sedatives/hypnotics, thiazide diuretics or angiotensin-receptor blockers; or severe nocturnal hypoxia (SpO2 nadir <70% on diagnostic sleep study).

Participant flow

Outcome results

• Apnea Hypopnea Index (AHI) (Primary), Median, events/hour of sleep: Dual-Therapy 17.1 (9.4 to 37.7); Placebo 32.7 (23.8 to 51.8); Triple-Therapy 26 (9.2 to 40)
• SpO2 Nadir (Secondary), Median, Percent Oxygen Saturation: Dual-Therapy 83.5 (79.8 to 87.2); Placebo 84 (82 to 85.2); Triple-Therapy 86 (82.5 to 87.5)
• Pathophysiological Traits: Vpassive, Vactive, Arousal Threshold (Secondary), Median, % Veupnea: Dual-Therapy 69.5 (58.7 to 78.2); Placebo 67.4 (59.4 to 73.2); Triple-Therapy 70.4 (61.7 to 75.3)
• Pathophysiological Trait: Loop Gain (Secondary), Median, dimensionless: Dual-Therapy 0.54 (0.49 to 0.59); Placebo 0.57 (0.51 to 0.69); Triple-Therapy 0.49 (0.44 to 0.6)
• Percent Responders (Secondary), Count of participants, Participants: Dual-Therapy 5; Placebo 0; Triple-Therapy 4
• Blood Pressure (Secondary), Median, mmHg: Dual-Therapy 112.5 (107.8 to 119.8); Placebo 116.5 (109.8 to 124.5); Triple-Therapy 116 (110.2 to 120.8)
• Subjective Sleepiness: Stanford Sleepiness Scale (SSS) (Secondary), Median, units on a scale: Dual-Therapy 2 (1 to 2.2); Placebo 2 (1 to 3); Triple-Therapy 2 (1 to 3.8)
• Sleep Quality: PROMIS (Patient-Reported Outcomes Measurement Information System) Sleep Disturbance (Secondary), Median, t-score: Dual-Therapy 53 (49 to 56.3); Placebo 53.9 (49.9 to 57.8); Triple-Therapy 53 (48.8 to 55.6)
• Psychomotor Vigilance: Response Speed (Secondary), Median, 1/seconds: Dual-Therapy 3 (2.8 to 3.3); Placebo 3.1 (2.8 to 3.4); Triple-Therapy 2.9 (2.8 to 3.3)
• Psychomotor Vigilance: Lapses (Secondary), Median, Number of lapses: Dual-Therapy 3 (0 to 8.2); Placebo 3 (1 to 6.2); Triple-Therapy 5 (2.2 to 9.8)

Adverse events

Time frame: During the trial subjects were asked at each visit whether any adverse events had occurred during that 3-day study phase.

Other adverse events by term

• Results publication: Schmickl CN, Orr JE, Alex RM, Gruenberg E, Parra G, White S, Spenceley A, DeSarkar T, Kong M, DeYoung PN, Sands SA, Owens RL, Malhotra A. Combination Drug Therapy with Acetazolamide, Eszopiclone +/- Venlafaxine for Obstructive Sleep Apnea (RESCUE-Combo): A Randomized, Double-Blind, Placebo-controlled Clinical Trial. Ann Am Thorac Soc. 2025 Feb;22(2):263-273. doi: 10.1513/AnnalsATS.202407-736OC. PMID 39514000

• Study Protocol and Statistical Analysis Plan (PDF, 2021-06-17)
• Informed Consent Form (PDF, 2021-08-17)

• Christopher Schmickl, MD, PhD, Principal investigator, Associate Physician and Postdoctoral Fellow