[The impact of different definitions of hypoxemia on the relation between awake pulmonary pressure and hypoxemia during sleep in patients with COPD]

Abstract

Background: Sleep related hypoxemia (SRH) in chronic obstructive pulmonary disease (COPD) can be easily detected by pulse-oximetry and may contribute to the development of pulmonary hypertension (PH). Since several parameters for the quantification of SRH are in use, we investigated which of these parameters has the strongest relation to the awake pulmonary arterial pressure (PAP) and is able to distinguish between patients without and with PH.

Patients and methods: 44 COPD-patients (awake PaO2 > or = 60 mm Hg) were investigated. PAP at rest (PAP; pathological threshold > 20 mm Hg) and under physical exercise (PAPB; p.t. > 28 mm Hg) were determined during daytime by Swan-Ganz-catheter. To quantify the degree of SRH the following parameters of nocturnal pulse-oximetry were used: mean nocturnal oxygen saturation (SaO2 m; p.t. < 90%), nadir SaO2 (SaO2 min; p.t. < 85%), and mean time of SaO2 < or = 90% in relation to total time of registration (t90; p.t. > 30%). Linear correlations and regressions as Chi 2-respectively Fisher-test were used for statistical analysis (p < 0.05).

Results: Generally there was only a weak relation between PAP and SRH. The best linear correlation at rest respectively under physical exercise was found between PAP and SaO2 min (r = -0.529 resp. -0.541, p < 0.001). Using the above defined thresholds for PAP and SaO2 patients could be most precisely separated into those without and with PH using SaO2 min with a threshold for the pathological range of < 85% (p = 0.030 resp. 0.002). t90 with a threshold > 30%, however, had a much worse selectivity (p = 0.487 resp. 0.057).

Conclusions: In COPD-patients with SRH the closest relation can be found between nadir SaO2 and PAP resp. PAPB. Furthermore nadir SaO2 (< 85%) could more precisely separate patients into those without and with pulmonary hypertension than t90. The overall weak relation between nocturnal oxygenation and pulmonary hypertension shows, however, that other factors such as daytime PaO2, hypercapnia or emphysema are involved in the development of pulmonary hypertension in COPD.