DISCUSSION AND CONCLUSIONS
Though the design of the present study does not conform with classical rules for clinical psychopharmacological trials, i.e., double-blind, placebo-controlled, and fixed-dosage protocols with larger patient groups, some interesting remarks
on anxiety, anxiolytics, and daytime sleepiness can nonetheless be made in the light of recent publications on the subject. Generalized anxiety is well known to produce a state of CNS arousal that does not favor sleep in spite of the fact that anxiety and sleepiness can, at times, coexist. Our patients were moderately anxious (total mean HARS score: 22) and complained of some sleep difficulty at night, although unable to compensate for whatever sleep loss might have occurred during the daytime. As compared with patients suffering from persistent psychophysiological insomnia who are also anxious but experience their daytime peak of anxiety just before going to bed, generalized anxiety disorder patients have a persistently elevated level of arousal throughout the daytime (Table 1). Seidel et al. (14) have found a minor percentage of such subjects (14%) in a large group of chronic insomniacs. Whether this tendency to high CNS arousal interferes with nighttime sleep onset or sleep maintenance cannot be determined by our limited observations, but the timing of anxiety in the 24-hr sleep/wake cycle is now being considered as one relevant factor in managing appropriate anxiolytic therapy. The fact that subjective nighttime sleep quality was significantly improved after 2 weeks of treatment and remained stable after 1 month only through daytime sedation suggests that conditioned insomnia is not a major factor in generalized anxiety. Daytime anxiety was also effectively reduced by treatment, since no clinically significant differences were noted between a single daily administration of a long half-life compound such as CDDZ or a twice-daily regime of an intermediate half-life compound such as APZ. Starting doses (2 mg CDDZ and 0.5 + 0.5 mg APZ), however, had to be tailored after the first week due to different responsiveness of individual patients on the anxiety-sleepiness continuum. During the treatment period, accumulation of both compounds occurred to a significant extent as proved by the chronic plasma levels that were, on average, five to seven times higher than those after acute administration. This is consistent with previous multiple-dose pharmacokinetic findings of CDDZ and APZ (4). The lack of significant correlation between individual SSS scores and objective values across the five sleep latency tests of each condition was not unexpected. Others have also shown that the across-subjects reliability of actual sleepiness ratings is low (7) and might be even less in patients than in normals, since the former may tend to confuse the sleepiness-alertness polarity with the anxiety-sedation one. The fact that the drugs’ plasma levels also did not correlate with MSLT values taken at the same time points supports the notion that daytime sleepiness profile is under the influence of several determinants (13). It has been demonstrated, however, that the buildup of plasma levels of benzodiazepines with chronic use is associated with a progressive partial tolerance to some of their clinical effects. Seidel et al. (15) gave 0.5 mg of APZ at 10 A.M. and at 2 P.M., to 9 healthy younger subjects for 7 consecutive days and found that their mean daily sleep latencies were significantly reduced as compared to baseline on both the first and the seventh day of administration. At the latter point, however, some tolerance to the hypnotic effect was already evident. We have been able to show that the same occurs after 1 month of treatment in a small group of target subjects. In fact, mean MSLT value after acute APZ was significantly lower than at baseline (9.2 vs. 18.7 min), but was less so after 1 month of treatment (14.4 min), in spite of a greater mean daily dosage and a significant plasmatic buildup. Moreover, our MSLT data indicate that a single dose of 2 mg of CDDZ is also able to increase the mean daytime sleep tendency of anxious middle-aged subjects and that this effect is maintained after 1 month in spite of a reduction of daily mean dosage. Seidel et al. (15) obtained quite similar results after short and intermediate diazepam administration. On the first experimental day, 10 mg of diazepam produced less objective sleepiness than 1 mg of APZ, but subjects did not develop any tolerance to the sedative effect after 1 week of drug ingestion. Besides confirming Seidel’s conclusion that APZ’s initial higher sedative effect decreases within a short time, our data do suggest that this tolerance might selectively spare the anxiolytic effect that, according to HARS scores, is well maintained after 1 month of treatment. Hence, treatment with APZ would be indicated more in patients for whom a more selective effect on pure anxiety is desirable, whereas treatment with CDDZ would better favor patients who like feeling sedated in the daytime (16). Although appealing, such conclusions are still speculative and warrant more controlled trials with larger patient groups. One point, however, seems particularly relevant: Taking daytime sleepiness into account can help improve the effectiveness and standing of an anxiolytic drug therapy.