The purpose of this study was to determine whether hypoventilation contributes to the sleep hypoxemia observed in chronic obstructive pulmonary disease (COPD) patients and to examine breathing pattern and respiratory muscle electromyographic (EMG) activity during these episodes. Seven COPD patients who experienced at least a 10% decrease in arterial O2 saturation (SaO2) during rapid-eye-movement sleep (REM) sleep, six COPD patients with a minimal fall in SaO2, and five healthy subjects were studied. An inductance vest was used to quantitate ventilation. Skin electrodes were used to estimate diaphragmatic and intercostal electromyographic activity. Minute ventilation and EMG activity decreased in all three groups during sleep. Ventilation was irregular during REM sleep in the patients. During REM sleep, desaturating patients had longer episodes of hypopneic breathing [30 +/- 8 s (SE)] than nondesaturating patients (13 +/- 1 s, P less than 0.01). Desaturating patients spent a greater proportion of REM time hypopneic (53 +/- 5 vs. 28 +/- 5%, P less than 0.01) and had a greater decrease in functional residual capacity during hypopnea (P less than 0.05). SaO2 followed the hypopneic and hyperpneic breathing in REM sleep so that desaturating patients had more time for desaturation to occur. Thus hypoventilation appears to be a primary factor in sleep O2 desaturation in these patients. Because of the fall in lung volume, maldistribution of ventilation may also contribute.

We analyzed the accuracy of the inductance vest in measuring several ventilatory parameters in five patients with chronic obstructive pulmonary disease (COPD). We assessed tidal volume (VT) accuracy at different respiratory frequencies in different lying body positions with different thoracic and abdominal contributions to breathing and the accuracy over a 4-h time span. Mean percent error was calculated without regard to direction of error. The mean error of vest VT estimation was 7.6% for all body positions studied and 5.6% for right and left lateral positions combined. Vest VT accuracy was unchanged after 4 h and with changes in thoracic and abdominal contributions to VT. The mean errors for inspiratory and expiratory times were 3.3 and 2.0%, respectively. Volume was differentiated to flow. For respiratory rates ranging from 12 to 30 breaths/min, the mean error of the vest and our differentiation circuit in duplicating peak flows measured at the mouth was 3.5%. The ability of the vest to estimate changes in end-expiratory position or functional residual capacity was not as good as with VT; the mean error was 30.7%. For estimation of VT, ventilatory timing, and airflow in COPD patients, the inductance vest performs well. For measurement of changes in lung volume, improvements in vest design need to be made.

No abstract

No abstract