Study objectives: Home overnight pulse oximetry (OPO) is used to assess nocturnal desaturation in patients with COPD, but the current practice of relying on one recording has not been studied. We assessed the variability of nocturnal desaturation in patients with COPD between nights, as measured by home OPO.
Design: Study subjects attended for clinical evaluation, spirometry, and arterial blood gas analysis. OPO was prospectively completed at home on 2 consecutive nights (study night 1 [N1] and study night 2 [N2]) and repeated at 3 weeks (study night 3 [N3]).
Setting: Respiratory Services, Green Lane Hospital, Auckland, New Zealand.
Patients: Twenty-six patients with clinically stable COPD (mean age, 69.3 years [SD, 6.9]; FEV(1), 28.6% predicted [SD, 10.6]; PO(2), 71.3 mm Hg [SD, 9.8]). Patients with asthma or clinical evidence of obstructive sleep apnea were excluded.
Measurements and results: Mean nocturnal saturation (MNS) and time spent with saturation below 90% (TB90%) were calculated for N1, N2, and N3. Group mean recording length, MNS, and TB90% were similar for each night. Little variation in MNS was seen between nights (N1 and N2 mean difference, 1.31%; N2 and N3, 1.26%; N1 and N3, 1.25%). Larger variation was seen between nights for TB90% (N1 and N2 mean difference, 17.46%; N2 and N3, 9.95%; N1 and N3, 14.05%). No factors were identified that predicted increased variability of TB90%. Using the current definition of "significant nocturnal desaturation" (TB90% > or = 30% of the night), 9 of 26 patients (34.6%) changed category between "desaturator" and "nondesaturator" from N1 to N2.
Conclusion: Nocturnal desaturation in patients with COPD exhibits considerable night-to-night variability when measured by home OPO. A single home OPO recording may be insufficient for accurate assessment of nocturnal desaturation.