Acute Intermittent Hypoxia Therapy for Depression

This multicenter, randomized, parallel-group, sham-controlled trial aims to evaluate the efficacy and safety of acute intermittent hypoxia (aIH) in patients with depressive disorders and to investigate its potential mechanisms of action.

A total of 156 participants will be recruited from one lead center (Beijing Anding Hospital) and three participating centers. Participants will be randomly assigned in a 1:1 ratio to aIH or sham intervention using block randomization stratified by study center. Both groups will complete 10 treatment sessions over 2 weeks, followed by a 6-week follow-up period. The aIH intervention will consist of alternating periods of hypoxia and normoxia. An individualized oxygen concentration for hypoxic exposure will be established before treatment based on tolerability and will remain fixed throughout the 10-session treatment course. The sham intervention will be administered under normoxic conditions.

Clinical assessments will be conducted at baseline and at weeks 1, 2, and 8. The primary outcome is the between-group difference in change in the Montgomery-Åsberg Depression Rating Scale (MADRS) total score from baseline to week 2 (After 10 Treatment Sessions). Exploratory assessments will include hypoxia tolerance testing and functional near-infrared spectroscopy (fNIRS). A subset of participants will undergo additional neurophysiological assessments, brain imaging, magnetoencephalography (MEG), and peripheral blood sampling at selected time points, including baseline, the end of the intervention, and the end of follow-up, to explore potential neurobiological mechanisms underlying the effects of aIH.

One interim analysis of efficacy and safety is planned after approximately half of the participants have completed the primary endpoint assessment. An independent Data Monitoring Committee will review unblinded interim data according to prespecified procedures.

Participants in the acute intermittent hypoxia (aIH) group will receive 10 sessions of acute intermittent hypoxia intervention within 2 weeks. Each session will include intermittent hypoxia-reoxygenation cycles, with a cumulative hypoxic exposure time of 40 minutes per session.The inspired oxygen fraction (FiO₂) used in the hypoxia group will be set according to the tolerated level determined at baseline and kept constant throughout the 10 treatment sessions.

Other: Acute Intermittent Hypoxia (AIH) · Device: Acute Intermittent Hypoxia Delivery Device

Participants in the normoxic sham intervention group will complete a sham oxygen concentration exploration session at baseline, followed by 10 sessions of sham normoxic intervention within 2 weeks. Each session will use the same device, intervention environment, cycle settings, treatment frequency, and session duration as the acute intermittent hypoxia group. Participants will receive normobaric normoxic air throughout the formal intervention, with the inspired oxygen fraction (FiO₂) kept constant at 0.21.

Other: Normoxic Sham Intervention · Device: Acute Intermittent Hypoxia Delivery Device

Participants will receive acute intermittent hypoxia using a study device with built-in safety protections. A brief FIO₂ exploration session will be conducted at baseline to determine an individualized tolerated oxygen level for each participant. This individualized FIO₂ will then be used and kept constant throughout all 10 formal treatment sessions.The 10 sessions will be completed within 2 weeks, with each session including eight hypoxia-reoxygenation cycles totaling 40 minutes of hypoxic exposure.

Participants in the normoxic sham intervention group will complete a brief sham oxygen concentration exploration session at baseline, followed by 10 sessions of sham normoxic intervention within 2 weeks. Each session will use the same device, intervention environment, cycle settings, treatment frequency, and session duration as the acute intermittent hypoxia group. Participants will receive normobaric normoxic air throughout the formal intervention, with the inspired oxygen fraction (FiO₂) kept constant at 0.21.

A computerized device used to deliver acute intermittent hypoxia by precisely controlling inspired oxygen concentration (FiO₂), hypoxia-reoxygenation cycles, treatment duration, and physiological safety monitoring. The device includes built-in safety protections and continuous monitoring throughout treatment.

The same computerized device will be used to deliver normobaric normoxic air (FiO₂ = 0.21) during sham intervention while maintaining identical operating procedures, treatment schedule, cycle settings, physiological monitoring, and safety protections as used in the experimental group

Plazo: Baseline to the end of 10 treatment sessions (Week 2).

Change in the Montgomery-Åsberg Depression Rating Scale (MADRS) total score from baseline to the end of 10 treatment sessions (Week 2) will be compared between the aIH and sham-control groups. The MADRS is a clinician-rated measure of depressive symptom severity, with total scores ranging from 0 to 60. Higher scores indicate more severe depressive symptoms, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Proportion of participants with a clinical response, defined as a 50% or greater reduction in Montgomery-Åsberg Depression Rating Scale (MADRS) total score from baseline.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Proportion of participants with a clinical response, defined as a 50% or greater reduction in the 17-item Hamilton Depression Rating Scale (HAMD-17) total score from baseline.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Proportion of participants achieving remission, defined as a Montgomery-Åsberg Depression Rating Scale (MADRS) total score of 10 or lower.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Proportion of participants achieving remission, defined as a 17-item Hamilton Depression Rating Scale (HAMD-17) total score of 7 or lower.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Montgomery-Åsberg Depression Rating Scale (MADRS) total score from baseline to each specified assessment time point. The MADRS total score ranges from 0 to 60, with higher scores indicating more severe depressive symptoms. A greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the 17-item Hamilton Depression Rating Scale (HAMD-17) total score from baseline to each specified assessment time point. The HAMD-17 is a clinician-rated measure of depressive symptom severity, with total scores ranging from 0 to 52. Higher scores indicate more severe depressive symptoms, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the 14-item Hamilton Anxiety Rating Scale (HAMA-14) total score from baseline to each specified assessment time point. The HAMA-14 is a clinician-rated measure of anxiety symptom severity, with total scores ranging from 0 to 56. Higher scores indicate more severe anxiety symptoms, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline ; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

The Clinical Global Impressions-Severity (CGI-S) and Improvement (CGI-I) scales are clinician-rated scales used to assess illness severity and change over time. CGI-S scores range from 1 to 7 (higher scores indicate greater severity), and CGI-I scores range from 1 to 7 (lower scores indicate greater improvement). CGI-S change from baseline will be calculated as baseline minus post-baseline scores, while CGI-I will be assessed at each post-baseline time point.

Plazo: Baseline; after 1 treatment session (Day 1); after 3 treatment sessions (Day 3); after 5 treatment sessions (Week 1); after 7 treatment sessions (Day 9); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the 16-item Quick Inventory of Depressive Symptomatology-Self-Report (QIDS-SR16) total score from baseline to each specified assessment time point. The QIDS-SR16 is a self-report measure of depressive symptom severity, with total scores ranging from 0 to 27. Higher scores indicate more severe depressive symptoms, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 1 treatment session (Day 1); after 3 treatment sessions (Day 3); after 5 treatment sessions (Week 1); after 7 treatment sessions (Day 9); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Generalized Anxiety Disorder-7 (GAD-7) total score from baseline to each specified assessment time point. The GAD-7 is a 7-item self-report measure of anxiety symptom severity, with total scores ranging from 0 to 21. Higher scores indicate more severe anxiety symptoms, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 1 treatment session (Day 1); after 3 treatment sessions (Day 3); after 5 treatment sessions (Week 1); after 7 treatment sessions (Day 9); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Immediate Mood Scale (IMS) total score from baseline to each specified assessment time point. The IMS is a 22-item self-report measure of current mood state, with total scores ranging from -66 to +66. Higher scores indicate better mood, and a greater increase from baseline indicates greater improvement.

Plazo: Baseline; after 1 treatment session (Day 1); after 3 treatment sessions (Day 3); after 5 treatment sessions (Week 1); after 7 treatment sessions (Day 9); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Lake Louise Score (LLS) total score from baseline to each specified assessment time point. The LLS is a 4-item self-report measure covering headache, gastrointestinal symptoms, fatigue/weakness, and dizziness/lightheadedness. Total scores range from 0 to 12, with higher scores indicating greater symptom severity. An increase from baseline indicates worsening symptoms.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the 10-item Perceived Stress Scale (PSS-10) total score from baseline to each specified assessment time point. The PSS-10 is a self-report measure of perceived stress, with total scores ranging from 0 to 40. Higher scores indicate greater perceived stress, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Sheehan Disability Scale (SDS) total score from baseline to each specified assessment time point. The SDS is a self-report measure of functional impairment in three domains: work/school, social life, and family life/home responsibilities. Total scores range from 0 to 30, with higher scores indicating greater functional impairment. A greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

The Digit Symbol Substitution Test (DSST) is an assessor-administered neuropsychological test used to assess cognitive processing speed, attention, and visuomotor coordination. The DSST score is defined as the number of correctly completed digit-symbol substitutions within 90 seconds. Change from baseline will be calculated as the DSST score at each post-baseline assessment minus the baseline DSST score. A positive change from baseline indicates improvement in cognitive processing speed, attention, and visuomotor performance Unit of Measure: Points

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Digit Span Test (DST) score from baseline to each specified assessment time point. The DST assesses attention and working memory through forward and backward digit recall tasks. Higher scores indicate better performance, and an increase from baseline indicates improvement.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8)

Change in the Insomnia Severity Index (ISI) total score from baseline to each specified assessment time point. The ISI is a 7-item self-report measure of insomnia severity, with total scores ranging from 0 to 28. Higher scores indicate more severe insomnia, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline; after each treatment session and Week 8.

All adverse events (AEs) and serious adverse event (SAE) occurring during the study will be collected and recorded, including their type, severity, duration, and relationship to acute intermittent hypoxia treatment.

Plazo: Baseline; after 5 treatment sessions (Week 1); after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in the Positive Expectancy and Negative Expectancy subscale scores of the Chinese version of the Stanford Expectations of Treatment Scale (C-SETS) from baseline to each specified assessment time point. The C-SETS is a 6-item self-report measure of treatment expectations. Each subscale comprises three items and has a summed score ranging from 3 to 21. Higher Positive Expectancy scores indicate greater positive treatment expectations, whereas higher Negative Expectancy scores indicate greater negative treatment expectations. Changes in the two subscales will be evaluated separately.

Plazo: Baseline; after 5 treatment sessions (Week 1) and after 10 treatment sessions (Week 2).

Change in the Biological Rhythms Interview of Assessment in Neuropsychiatry (BRIAN) total score from baseline to each specified assessment time point. The BRIAN is a clinician-administered measure of biological rhythm disturbances covering sleep, activity, social rhythms, and eating patterns. The 18-item total score ranges from 18 to 72. Higher scores indicate greater biological rhythm disturbances, and a greater reduction from baseline indicates greater improvement.

Plazo: Baseline (before treatment).

The Life Events Scale (LES) will be administered at baseline to assess exposure to stressful life events and their perceived impact. Higher scores indicate a greater burden of life event-related stress.

Plazo: Baseline; after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Resting-state electroencephalography (EEG) will be used to assess spectral power in predefined frequency bands, including delta (\<4 Hz), theta (4-7 Hz), alpha (8-13 Hz), beta (14-25 Hz), and low-gamma (25-50 Hz). The theta/beta power ratio will also be calculated. Changes from baseline in these EEG measures will be assessed at each post-baseline assessment.

Plazo: Baseline; after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Change in resting-state electroencephalography (EEG)-derived functional connectivity from baseline to the end of the intervention. Functional connectivity measures across prespecified frequency bands will be evaluated to explore treatment-related changes in brain functional networks.

Plazo: Baseline; after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Functional near-infrared spectroscopy (fNIRS) will be used to assess oxyhemoglobin (HbO), deoxyhemoglobin (HbR), and total hemoglobin (HbT) concentration changes in prespecified cortical regions. HbT will be calculated as the sum of HbO and HbR. Changes from baseline in these parameters and prespecified derived hemodynamic measures will be evaluated at each post-baseline assessment.

Plazo: Baseline; after 10 treatment sessions (Week 2); and at the end of follow-up (Week 8).

Functional near-infrared spectroscopy (fNIRS) will be used to assess task-related brain activation and resting-state functional connectivity. Task-related activation will be quantified using beta coefficients or activation amplitudes in prespecified channels or regions of interest. Resting-state functional connectivity will be quantified using Pearson correlation coefficients between channel pairs. Changes from baseline in these measures will be evaluated at each post-baseline assessment to explore the neurofunctional effects of aIH.

Plazo: Baseline and after 10 treatment sessions (Week 2).

Salivary cortisol will be assessed at baseline and after 10 treatment sessions. At each assessment, saliva samples will be collected at five time points: 4:00 PM, before bedtime, upon awakening, 30 minutes after awakening, and 60 minutes after awakening. Cortisol concentrations in the saliva samples will be analyzed. The outcome is the change in salivary cortisol levels from baseline to after 10 treatment sessions at each sampling time point.

Plazo: Baseline and after 10 treatment sessions (Week 2).

Peripheral blood samples will be analyzed to quantify a prespecified panel of metabolites involved in adenosine synthesis, metabolism, and degradation, including adenosine and related upstream and downstream metabolites in the purine metabolic pathway. Changes from baseline in these metabolites will be evaluated to characterize treatment-related alterations in peripheral adenosine metabolism.

Plazo: Baseline and after 10 treatment sessions (Week 2).

Peripheral blood samples will be analyzed to explore treatment-related molecular changes in biological pathways potentially involved in the effects of acute intermittent hypoxia therapy. Exploratory analyses may include metabolomic, proteomic, transcriptomic, and epigenomic profiling, including DNA methylation analysis, with particular attention to pathways related to adenosine and purine metabolism, neuroplasticity, inflammation, stress regulation, and other relevant biological processes. The primary molecular comparison will evaluate changes from baseline to the end of treatment after completion of 10 treatment sessions.

Plazo: Baseline and after 10 treatment sessions (Week 2).

Resting-state MEG data will be collected at baseline and after 10 treatment sessions using a 128-channel optically pumped magnetometer (OPM)-based MEG system (Marvel MEG 128; Unimagtech, Beijing, China) at a sampling rate of 1,000 Hz. Participants will lie supine with their heads positioned within the OPM sensor array and will be instructed to remain as still as possible and not fall asleep. Resting-state recordings will include two conditions: eyes closed and eyes open while fixating on a cross. Each condition will last 5 minutes. The total OPM-MEG acquisition time will be approximately 35 minutes, including resting-state and task-based recordings. The outcome is the change in the prespecified resting-state MEG measure(s) from baseline to after 10 treatment sessions.

Plazo: Baseline and after 10 treatment sessions (Week 2).

Functional magnetic resonance imaging (fMRI) will be used to assess treatment-related changes in resting-state brain function and task-evoked brain activation. Resting-state measures may include the amplitude or fractional amplitude of low-frequency fluctuations, regional homogeneity, and functional connectivity within and between predefined brain regions and functional networks. Task-evoked activation will be assessed during the \[exact task name\] and quantified using changes in blood oxygen level-dependent (BOLD) signals in predefined brain regions. Changes from baseline to the end of treatment will be evaluated to characterize alterations in spontaneous brain activity, functional network organization, and task-related cortical activation associated with acute intermittent hypoxia therapy.

Plazo: Baseline and after 10 treatment sessions (Week 2).

High-resolution structural magnetic resonance imaging will be used to assess treatment-related changes in brain structure. Prespecified measures may include cortical thickness, regional gray matter volume, and subcortical volume in predefined brain regions. Changes from baseline to the end of treatment will be evaluated to characterize structural brain alterations associated with acute intermittent hypoxia therapy.

Efficacy and Safety of Acute Intermittent Hypoxia Therapy for Major Depressive Disorder: A Randomized Controlled Trial