Respiratory effort during sleep predicts mortality in patients with suspected obstructive sleep apnea
Hamza Nahoui, Henrik Schirmer, Gunnar Einvik, Tarjei Øvrebotten, Kristian Berge, Peder L Myhre, Haldor Husby, Harald Hrubos-StrømAbstract
Rationale
Obstructive sleep apnea (OSA) confers increased risk of morbidity and mortality through mechanisms like exaggerated negative intrathoracic pressure and hypoxemia.
Objectives
Evaluate the association between negative esophageal pressure (PES), a surrogate of respiratory effort, during sleep and long-term all-cause mortality in suspected OSA patients.
Methods
Patients referred with suspected OSA (2003-2016) underwent overnight respiratory polygraphy with PES monitoring. PES, mean oxygen saturation (mSpO2), apnea/hypopnea index (AHI), total time in bed (TiB), time in bed spent above different PES thresholds (10 − 60 cmH2O) were recorded. Clinical data and mortality until June 2024 were obtained from medical records. Mortality was analyzed using multivariable Cox regression.
Results
In 16,083 patients (26% female; median age 48 years), median PES was 12.8 cmH2O (IQR 8.7-18.4). Median mSpO2 was 94% (93%−95%), and median AHI was 10 (4.3-21.1); 16% had AHI ≥ 30. During median follow-up of 15 years (IQR 12 − 18), 1,497 (9.3%) died. Patients in 3rd and 4th quartiles of median PES had increased mortality risk compared to 1st quartile. After adjustment for age, sex, BMI, comorbidities, AHI and mSpO2, median PES remained associated with mortality with the 3rd quartile showing increased risk (HR 1.21; 95% CI: 1.03-1.43; P = .024). Longer time spent above higher PES thresholds was associated with increased mortality. Risk rose across quartiles of %TiB at each threshold and increased further with more negative thresholds in the highest quartile.
Conclusion
Respiratory effort, measured by PES during sleep, independently predicts all-cause mortality, and provides prognostic information beyond traditional OSA metrics including AHI and hypoxemia. This supports respiratory effort as a useful component of comprehensive clinical evaluation.