Daytime oxygen saturation does not predict nocturnal oxygen desaturation in patients with chronic obstructive pulmonary disease

Nocturnal hypoxemia occurs commonly in patients with chronic obstructive pulmonary disease (COPD). Because pulmonary hypertension and cardiac arrhythmias are associated with this phenomenon, the detection and treatment of nocturnal hypoxemia should be part of the management of COPD patients. The ability to predict nocturnal hypoxemia by evaluating an awake patient would be desirable economically and logistically because continuous nocturnal oximetry is not widely available and because it is costly and labor intensive. We sought to determine whether awake oximetry would be precise enough to be clinically useful in predicting the degree of nocturnal oxygen desaturation in patients with COPD. We studied 71 patients with COPD. During sleep, the arterial oxygen saturation (SaO₂) decreased by an average of 9% with a maximum decrease of 21% (awake SaO₂ 93.0 ± 0.4% vs nocturnal lowest SaO₂ 84.0 ± 0.7%, p = 0.0001). The nocturnal oxygen desaturation in each patient, however, was poorly predicted from awake SaO₂. The standard error of estimate was large with a value of 5.3%. These data suggest that awake SaO₂ is not a good predictor of nocturnal oxygen desaturation in individual patients. The lack of a simple relationship between awake SaO₂ and nocturnal SaO₂ is due to a complex interplay of various physiologic and pathologic mechanisms involved in the control of breathing and oxygenation during sleep.