Hypoxia-Dominant Positional Obstructive Sleep Apnea in Obese Bariatric Patients: A Distinct and Underrecognized Phenotype

Dr. Chelsie Rohrscheib, Ph.D.

 

Click here to download full article. 

Overview

Positional obstructive sleep apnea (pOSA) is traditionally defined by changes in apnea–hypopnea index (AHI) between supine and non-supine sleep [1-3]. While this approach captures differences in event frequency, it does not fully reflect physiological severity, particularly in patients with severe obesity. In bariatric populations, obesity-related reductions in lung volume and oxygen reserve may amplify the depth and duration of oxygen desaturation during obstructive events without proportionally increasing event counts [4, 5]. As a result, reliance on AHI alone may underestimate clinically meaningful positional vulnerability.

Hypoxic burden (HB), expressed in units of percent–minutes per hour of sleep (% min/hr), quantifies the cumulative depth and duration of oxygen desaturation during sleep [6]. Unlike AHI, which counts discrete respiratory events, HB integrates the severity of hypoxemia over time and has been shown to better reflect cardiometabolic risk in OSA [7]. Despite growing recognition of hypoxic burden as a clinically relevant metric, its positional behavior has not been well characterized, particularly in populations with severe obesity.

This report examines positional hypoxic burden (pHB) in a large obese bariatric clinic cohort using posture-resolved home sleep apnea testing (HSAT) data. We further identify and characterize a hypoxia-dominant positional OSA (HD-pOSA) phenotype, defined by marked position-dependent hypoxic stress in the absence of traditional AHI-based positional OSA.

Study Objectives

1. To quantify the prevalence of positional hypoxic burden (pHB) in a bariatric clinic population using posture-resolved Wesper Lab HSAT data.
2. To compare pHB with traditional AHI-based positional OSA definitions.
3. To identify a hypoxia-dominant positional OSA (HD-pOSA) phenotype that is not detected by AHI-based positional criteria.
4. To evaluate the clinical relevance of posture-time–aware positional phenotyping in patients with severe obesity.

Methods

Study Design and Population

We conducted a retrospective observational analysis of adults undergoing home sleep apnea testing at a bariatric clinic between January 1, 2025 and December 31, 2025 using the Wesper Lab system, which provides posture-resolved respiratory indices, HB, and sleep-duration metrics.

The initial dataset included 6,067 HSATs from 3,225 unique patients. Because valid positional comparisons require adequate sampling of sleep in different body positions, analyses were restricted to studies meeting a minimum posture-duration criterion, consistent with commonly used research standards for positional OSA.

Posture-Time Inclusion Criteria

To ensure stable and physiologically meaningful comparisons:

• Studies were required to include ≥30 minutes (≥1,800 seconds) of sleep in both:
  
  • Supine, and
  • Non-supine positions (defined as the combined time spent in left, right, and prone positions).

After applying this criterion:

• 4,141 HSATs met inclusion criteria
• Representing 2,500 unique patients

Positional Metrics and Phenotype Definitions

Hypoxic Burden (HB) was defined as the integrated area of oxygen desaturation below baseline oxygen saturation over time, normalized by total sleep time and expressed in % min/hr.

• Positional hypoxic burden (pHB) was defined as:
  • Supine HB ≥ 2× non-supine HB; where non-supine HB was calculated as the mean of left-, right-, and prone-position values.
  • Classic AHI-based positional OSA (pOSA) was defined analogously:
• Supine AHI ≥ 2× non-supine AHI
  
  

All positional classifications were performed at the patient level, using median values across all qualifying studies for each patient to reduce night-to-night variability.

Hypoxia-Dominant Positional OSA (HD-pOSA)

A hypoxia-dominant positional OSA phenotype was defined a priori as:

1. Presence of positional hypoxic burden (pHB),
2. Absence of AHI-based positional OSA,
3. An additional absolute HB difference threshold to ensure clinical relevance.

This definition was designed to identify patients in whom body position modulates hypoxic stress rather than event frequency.

Results

Cohort Characteristics

The final analytic cohort included 2,500 unique bariatric patients with posture-qualified studies. Patients had a mean BMI of 41.0 ± 7.1 kg/m², consistent with class III obesity. The cohort was predominantly female (74.4% female), reflecting typical bariatric clinic demographics.

Prevalence of Positional Hypoxic Burden and AHI-Based Positional OSA

After accounting for posture-specific sleep duration:

• Positional hypoxic burden was identified in 1,851 patients (74.0%)
• Classic AHI-based positional OSA was present in 1,587 patients (63.5%)

Thus, pHB was more prevalent than AHI-based positional OSA in this bariatric population.

Discordance Between Hypoxic and AHI-Based Positional Classifications

Cross-classification of patients based on positional hypoxic burden and AHI-based positional OSA identified four mutually exclusive groups: 1,375 patients (55.0%) met both definitions, 476 patients (19.0%) demonstrated positional hypoxic burden without AHI-based positional OSA, 212 patients (8.5%) met AHI-based positional OSA criteria without positional hypoxic burden, and 437 patients (17.5%) met neither definition.

Within the subgroup exhibiting positional hypoxic burden without AHI-based positional OSA, 222 patients (8.9% of the total cohort; 46.6% of this subgroup) met stricter criteria for hypoxia-dominant positional OSA (HD-pOSA), characterized by moderate overall AHI and markedly elevated absolute hypoxic burden during supine sleep (Table 1). In this subgroup, body position influenced desaturation severity and duration, rather than the number of scored respiratory events. Importantly, nearly one in five patients exhibited clinically meaningful positional hypoxic stress that would have been missed by AHI-based positional definitions alone.

Subgroup Analysis: Demographic Characteristics of Hypoxia-Dominant Positional OSA

To further characterize the hypoxia-dominant positional OSA (HD-pOSA) phenotype, demographic characteristics of patients meeting HD-pOSA criteria were compared with those of the remaining posture-qualified cohort (Table 1). Among the 2,500 patients included in positional analyses, 222 patients (8.9%) met criteria for HD-pOSA. Compared with patients who did not meet HD-pOSA criteria, this subgroup demonstrated a distinct demographic profile.

Patients with HD-pOSA had a higher mean body mass index (43.1 ± 7.8 kg/m²) compared with non-HD-pOSA patients (40.8 ± 7.0 kg/m²), indicating a greater degree of obesity despite similar overall AHI severity. HD-pOSA patients were also older on average (45.1 ± 9.7 vs. 40.9 ± 10.6 years). While the overall bariatric cohort was predominantly female, the HD-pOSA subgroup exhibited a relatively higher proportion of male patients (37.0% vs. 20.2%), suggesting that sex-related physiological differences may further modulate vulnerability to hypoxia-dominant positional disease.

Together, these findings indicate that HD-pOSA is not a random subset of positional hypoxic burden but rather a demographically enriched phenotype, associated with higher BMI and older age, in which body position disproportionately influences hypoxic severity rather than event frequency. This demographic pattern is consistent with obesity- and age-related reductions in pulmonary reserve and supports the concept that positional hypoxic burden captures clinically relevant physiological vulnerability not reflected by AHI alone.

Table 1: Demographic characteristics of patients with hypoxia-dominant positional OSA (HD-pOSA) compared with non-HD-pOSA patients. Values are reported as mean ± standard deviation for continuous variables and as percentages within each phenotype group for categorical variables. 

Clinical and Scientific Implications

These findings highlight important limitations of AHI-centric positional classifications in patients with severe obesity. In bariatric populations, obesity-related physiological factors, such as reduced functional residual capacity and diminished oxygen reserve, may magnify the hypoxic consequences of obstructive events in the supine position without increasing event frequency.

By explicitly accounting for posture-specific sleep duration and incorporating HB, this analysis identifies a previously underrecognized positional phenotype that may carry increased cardiometabolic relevance. Importantly, this phenotype is detectable using posture-resolved HSAT data and would be overlooked by conventional positional OSA definitions.

Conclusions

When posture-specific sleep duration is explicitly incorporated into analysis, positional hypoxic burden is highly prevalent in bariatric patients and exceeds the prevalence of classic AHI-based positional OSA. A distinct hypoxia-dominant positional OSA phenotype, affecting approximately 9% of patients, exhibits position-dependent hypoxic stress without corresponding positional changes in event frequency. These findings demonstrate that AHI alone incompletely characterizes positional OSA physiology in severe obesity and that posture-resolved hypoxic metrics provide clinically meaningful insight into disease heterogeneity. Incorporating positional hypoxic burden into OSA assessment may improve risk stratification and support more personalized treatment strategies in bariatric populations.

Reference

1. Duce BL, Kulkas AE, Leppänen TT, Oksenberg A, Töyräs J, Hukins CA. Community prevalence of positional obstructive sleep apnea. J Clin Sleep Med. 2025 May 1;21(5):875-882. doi: 10.5664/jcsm.11584. 
2. Romaker AM. Positional sleep apnea revisited. J Clin Sleep Med. 2025 May 1;21(5):751-752. doi: 10.5664/jcsm.11676. 
3. Levendowski DJ, Oksenberg A, Vicini C, Penzel T, Levi M, Westbrook PR. A systematic comparison of factors that could impact treatment recommendations for patients with Positional Obstructive Sleep Apnea (POSA). Sleep Med. 2018 Oct;50:145-151. doi: 10.1016/j.sleep.2018.05.012.
4. Peppard PE, Ward NR, Morrell MJ. The impact of obesity on oxygen desaturation during sleep-disordered breathing. Am J Respir Crit Care Med. 2009 Oct 15;180(8):788-93. doi: 10.1164/rccm.200905-0773OC. 
5. Littleton SW, Tulaimat A. The effects of obesity on lung volumes and oxygenation. Respir Med. 2017 Mar;124:15-20. doi: 10.1016/j.rmed.2017.01.004. 
6. Martinez-Garcia MA, Sánchez-de-la-Torre M, White DP, Azarbarzin A. Hypoxic Burden in Obstructive Sleep Apnea: Present and Future. Arch Bronconeumol. 2023 Jan;59(1):36-43. English, Spanish. doi: 10.1016/j.arbres.2022.08.005.
7. Azarbarzin A, Sands SA, Stone KL, Taranto-Montemurro L, Messineo L, Terrill PI, Ancoli-Israel S, Ensrud K, Purcell S, White DP, Redline S, Wellman A. The hypoxic burden of sleep apnoea predicts cardiovascular disease-related mortality: the Osteoporotic Fractures in Men Study and the Sleep Heart Health Study. Eur Heart J. 2019 Apr 7;40(14):1149-1157. doi: 10.1093/eurheartj/ehy624. Erratum in: Eur Heart J. 2019 Apr 7;40(14):1157. doi: 10.1093/eurheartj/ehz028.