Exercise capacity of patients with obstructive sleep apnea at moderate to high altitude.

BACKGROUND

It remains unknown to what extent exercise capacity is diminished at moderate to high altitudes in patients with obstructive sleep apnea (OSA).

PURPOSE: The primary aim was to compare peak oxygen uptake per kilogram body weight (VO2/kg) between OSA patients and healthy controls (HCs) at moderate-to-high altitude (~ 1,900 m) using cardiopulmonary exercise testing (CPET). Secondary aims were to evaluate the associations of OSA severity (apnea-hypopnea index, AHI) and nocturnal hypoxemia (lowest oxygen saturation, LSpO2; oxygen desaturation index, ODI) with CPET parameters, and to explore the potential discriminatory performance of CPET indices.

METHODS

This prospective study enrolled 38 OSA patients (AHI ≥ 5 events/h) and 36 healthy controls from Kunming, Yunnan Province, China (study site altitude: ~1,900 m; participants' residence altitude range: 1,800-2,800 m above sea level). All participants underwent overnight polysomnography and symptom-limited CPET using a cycle ergometer ramp protocol. Both groups were subgrouped by apnea-hypopnea index (AHI) severity and by lowest oxygen saturation (LSpO2) for CPET comparisons.

RESULTS

After adjustment for BMI, the OSA group exhibited significantly lower VO2/kg (P = 0.013) and a significantly higher VE/VCO2 slope (P = 0.038) compared with healthy controls. AHI and ODI were negatively correlated, and LSpO2 positively correlated, with VO2/kg; LSpO2 was positively correlated with VO2/WR but negatively correlated with VE/VCO2 slope; ODI was positively correlated with VE/VCO2 slope. Additionally, in an exploratory ROC analysis, VO2/kg showed a modest AUC (0.69), high sensitivity (97.4%) and limited specificity (38.9%), suggesting a supportive rather than standalone screening role.

CONCLUSIONS

OSA patients at moderate-to-high altitude show reduced exercise capacity (lower VO2/kg) and ventilatory inefficiency (higher VE/VCO2 slope), correlating with OSA severity and nocturnal hypoxemia; the VO2/kg and VE/VCO2 slope differences persisted after adjustment for BMI. These limitations may reflect superimposed chronic intermittent and environmental hypoxia, with skeletal muscle dysfunction, altered ventilatory control, and endothelial impairment as possible contributing mechanisms. CPET provides valuable functional assessment; in an exploratory analysis, CPET parameters may serve as supportive indicators to raise suspicion for OSA but cannot replace polysomnography. Validation in larger, multi-center cohorts is required.