Reproducibility of ventilatory measurements during sleep on different nights in patients with chronic obstructive pulmonary disease.

To evaluate the reproducibility of respiratory measurements between nights we performed studies in 20 outpatients with stable, moderately severe chronic obstructive pulmonary disease. All patients had symptoms from their lung disease but had no sleep complaints. Their mean age was 61 years, mean 1-second forced expiratory volume was 42% of predicted, and mean functional residual capacity 195% of predicted. Arterial Pco2 averaged 40 +/- 1 (SEM) mm Hg and mean Po2 64 +/- 1 mm Hg. Sleep was monitored for 7 hours by standard techniques on 2 nights 1 week apart. Breathing was assessed by measuring airflow at the nose and mouth with thermistors, and rib cage and abdominal respiratory movements with inductive plethysmography. Oxygen saturation was measured with an ear oximeter. Patients slept on the average 58% of the time in the first night and 63% in the second. Arousals were common but apneas uncommon in both nights. There was no significant difference in median nocturnal O2 saturation on the 2 nights. Tidal volume and minute ventilation, but not respiratory rate, were significantly lower and more variable in rapid eye movement (REM) sleep as compared with wakefulness and non-REM sleep; however, mean values and the variance for tidal volume, respiratory rate, or minute ventilation were similar on both nights.