We can only speculate that this is due to the fact that the fraction of REM sleep was subnormal already during placebo sleep. Observations on healthy subjects may not be directly applicable to patients with diseases affecting sleep quality.
Another factor contributing to the short total sleeping time may be that all our patients took theophylline during the study. Theophylline, as many other xanthine derivates, affects sleep and may, even in therapeutic concentrations, cause deterioration in the quality of sleep. Acute administration of aminophylline may reverse diazepam-induced sedation, but from our study it seems unlikely that maintenance therapy with theophylline significantly inhibits the hypnotic effects of oral benzodiazepines.
Most patients reported that it was easier to fall asleep and that they slept better during the treatment nights. This was confirmed by our quantitative data which showed shorter sleep latencies, longer total sleeping times and higher sleep efficiency indices during treatment nights. Whether this improvement is due to antagonism of the effect of theophylline on sleep or to a reduction of the sleep disturbances produced by the disease itself, or to both together, cannot be ascertained from this study. review
Since our only patient with an abnormal number of sleep apneas denied daytime hypersomnolence we cannot call his condition a sleep apnea syndrome. The unexpectedly small effects of benzodiazepines on apneas in this patient may therefore not be representative of the situation in patients with this syndrome. His sleep was considerably improved by benzodiazepines, and he continues to use nitrazepam regularly at home (against our advice).
The sleep-induced increase in tcPco2 after active treatment was significantly larger than after placebo but the difference was small and the increase still within normal limits.