Collection of Data
We reviewed drug intake and sleep/wake schedules at the clinical interview and medical evaluation that took place 15 days before the research polygraphic monitoring. Patients were asked to abstain from drugs known to affect sleep latency or rapid eye movement (REM) sleep latency until after polygraphic evaluation and MSLT and were told to observe their usual times for bedtime and morning arising for one week before the monitoring. They reported to the recording facilities by 7:30 pm on the day of polygraphic monitoring. Lights were turned off at the usual bedtime, and patients who awoke early were encouraged to sleep longer or at least to stay in bed in the dark until the time of their normal wakening. Any patient still sleeping at 8 am was to be awakened, but this proved unnecessary. After arising, patients underwent a urine drug screen to rule out the presence of any substance that could affect the MSLT.
The MSLT consisted of five 20-min test periods given at two-hour intervals, beginning at 9:30 am Nocturnal polygraphic variables monitored were the electroencephalogram (EEG, C3/A2-C4/A1), electromyogram (EMG, submental, anterior tibialis), electro-oculogram (EOG, outer canthi), and electrocardiogram (ECG, modified V2 lead). Respiration was monitored by strain gauge or uncalibrated inductive respiratory plethysmography, air flow was measured by thermistors, and Sa02 by ear oximetry (Biox). Apneas and hypopneas were scored according to standard definitions, based upon respiratory, airflow, oximetric and other findings. We defined hypopnea as (1) reduction in maximal thermistor output by at least 50 percent compared with baseline, and (2) decrease in Sa02 to <92 percent from a baseline of at least 94 percent, or drop in Sa02 of at least 3 percent if baseline was below 90 percent. An MSLT or polygraphic “epoch” was defined as a period of 30 s.
Analysis of Nocturnal Polygraphic Recordings
To obtain the RDI, we calculated the number of apneas + hypopneas X 60/total sleep time (TST). The percentage of TST spent with an Sa02 <90 percent (percent TOO percent SaOJ, the mean nocturnal sleep Sa02 and an arbitrarily defined 02 desaturation index (“oxygen-80-index” or 02-80-I) supplied information on oxygenation during sleep.