0519 Association of Ventilatory and Hypoxic Burden with AI-Derived Obstructive Sleep Apnea Symptoms and Risk Factors

Abstract Introduction Ventilatory Burden (VB) and Hypoxic Burden (HB) have emerged as physiologic metrics that may improve obstructive sleep apnea (OSA) diagnosis and risk stratification beyond the apnea–hypopnea index (AHI). However, the relationship between VB/HB and OSA symptoms/risk factors, which are used to estimate pre-test clinical probability, remains poorly characterized. Therefore, this study aimed to evaluate the distribution of VB and HB, measured by in-lab polysomnography (PSG), in patients with OSA symptoms/risk factors. Methods Retrospective data from suspected sleep apnea patients seen across Mount Sinai Health System in New York were analyzed. OSA symptoms and risk factors, including snoring, witnessed sleep apnea, tiredness/fatigue, age, and male sex, were extracted from clinical notes using large language models. VB and HB were calculated from PSG and categorized as low/high VB (cutoff = 25, Parekh et.al., 2023) or quartiles for HB. Ordinal logistic regression was used to assess associations of VB/HB with OSA symptoms/risk factors, and Spearman correlation evaluated relationships among VB, HB, and AHI. Results This study included 2,346 participants with 2,589 PSGs (mean age 55.4±16.0; 49.8% male). Among participants with ≥ 3 symptoms/risk factors, 40.4% had high VB, and the distribution of HB was Q1 (20.2%), Q2 (25.0%), Q3 (27.7%), and Q4 (27.1%). Compared with the lowest HB quartile (Q1), participants in Q2, Q3, and Q4 had greater sleep apnea symptoms/risk factors (ORs 1.82, 2.65, and 2.75; all p< 0.001). High VB (≥25) was also associated with greater symptoms/risk factors (OR 1.84; p< 0.001). HB and VB were moderately correlated (ρ = 0.55) and showed similar correlations with AHI (HB ρ = 0.39; VB ρ = 0.41). Notably, among participants with AHI < 5, 22.4% had high HB and 28.1% had high VB. Conclusion Higher VB and HB are strongly associated with greater sleep apnea symptoms/risk factors, supporting their relevance as diagnostic markers. The presence of high VB/HB among individuals with low AHI underscores the limitations of relying on AHI alone. These findings indicate the potential value of integrating VB and HB into diagnostic evaluation, particularly for patients with high clinical suspicion of OSA. Support (if any) R01HL171813, K25HL151912.

Authors

Institutions

Publication Details

Journal
SLEEP
Published
2026-05-01
DOI
https://doi.org/10.1093/sleep/zsag091.0518
Primary Topic
Obstructive Sleep Apnea Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

0519 Association of Ventilatory and Hypoxic Burden with AI-Derived Obstructive Sleep Apnea Symptoms and Risk Factors

Sri Saranya Ravi, Ankita Kumar, Korey Kam, Pavel Boulgakov et al.
SLEEP
Obstructive Sleep Apnea Research
article

0519 Association of Ventilatory and Hypoxic Burden with AI-Derived Obstructive Sleep Apnea Symptoms and Risk Factors

Sri Saranya Ravi, Ankita Kumar, Korey Kam, Pavel Boulgakov, I. Ayappa, Li Zhou, Sajila Wickramaratne, Katarina Martillo, Zachary Roberts, David Rapoport, Nagharjun Mathi Mariappan, Ankit Parekh, Sarah Chu, Thomas Tolbert, Andrew Varga
article en

Abstract

Abstract Introduction Ventilatory Burden (VB) and Hypoxic Burden (HB) have emerged as physiologic metrics that may improve obstructive sleep apnea (OSA) diagnosis and risk stratification beyond the apnea–hypopnea index (AHI). However, the relationship between VB/HB and OSA symptoms/risk factors, which are used to estimate pre-test clinical probability, remains poorly characterized. Therefore, this study aimed to evaluate the distribution of VB and HB, measured by in-lab polysomnography (PSG), in patients with OSA symptoms/risk factors. Methods Retrospective data from suspected sleep apnea patients seen across Mount Sinai Health System in New York were analyzed. OSA symptoms and risk factors, including snoring, witnessed sleep apnea, tiredness/fatigue, age, and male sex, were extracted from clinical notes using large language models. VB and HB were calculated from PSG and categorized as low/high VB (cutoff = 25, Parekh et.al., 2023) or quartiles for HB. Ordinal logistic regression was used to assess associations of VB/HB with OSA symptoms/risk factors, and Spearman correlation evaluated relationships among VB, HB, and AHI. Results This study included 2,346 participants with 2,589 PSGs (mean age 55.4±16.0; 49.8% male). Among participants with ≥ 3 symptoms/risk factors, 40.4% had high VB, and the distribution of HB was Q1 (20.2%), Q2 (25.0%), Q3 (27.7%), and Q4 (27.1%). Compared with the lowest HB quartile (Q1), participants in Q2, Q3, and Q4 had greater sleep apnea symptoms/risk factors (ORs 1.82, 2.65, and 2.75; all p< 0.001). High VB (≥25) was also associated with greater symptoms/risk factors (OR 1.84; p< 0.001). HB and VB were moderately correlated (ρ = 0.55) and showed similar correlations with AHI (HB ρ = 0.39; VB ρ = 0.41). Notably, among participants with AHI < 5, 22.4% had high HB and 28.1% had high VB. Conclusion Higher VB and HB are strongly associated with greater sleep apnea symptoms/risk factors, supporting their relevance as diagnostic markers. The presence of high VB/HB among individuals with low AHI underscores the limitations of relying on AHI alone. These findings indicate the potential value of integrating VB and HB into diagnostic evaluation, particularly for patients with high clinical suspicion of OSA. Support (if any) R01HL171813, K25HL151912.

SLEEPVol. 49(Supplement_1)
Icahn School of Medicine at Mount Sinai (US)
Quality Education
Openalex Percentile: Top 5%
Obstructive Sleep Apnea Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.