The authors would like to acknowledge the contribution of Drs. V. K. Sharma, P. Saunders, C. Sullivan, С McWilliam, L. M. Voruganti, and S. V. Manohar, who undertook the follow-up interviewing, and Dr. N. Wood, Miss M. Heary, Mrs. J. Silcock, Miss C. Hensey, Mrs. J. Wood, and Mrs. R. Searle, who undertook the interviewing of the initial sample. Our thanks are also due to Mr. M. Kayodi for his work on the computer and to Mrs. B. Ackerley for her skill in organizing the interviewing. This research was supported by grants from the Wellcome Trust and the Mersey Regional Health Authority, United Kingdom.
REFERENCES (more…)

The subjects in this study (4) were consecutive referrals to our Sleep Disorders Center located in a private setting. All but one were outpatients. The group of anxious depressives (« = 22) consisted of various DSM-III (6) anxiety disorders: 11 with generalized anxiety disorder, 5 with full-blown panic disorder, and 3 with agoraphobia with panic attacks; in 1 patient the anxiety state was best defined as a simple phobia, and in 2 patients the additional diagnosis of obsessive-compulsive disorder could be given. Low grade, typically subsyn-dromal, fluctuating depression occurred in most subjects. However, anxiety symptoms were the dominant clinical core and typically preceded the onset of the depressive symptoms. In view of the chronicity of the illness in the anxious depressive group, we chose depressed subjects with comparable chronicity (n = 20). They consisted of individuals with early onset and long-standing, fluctuating subsyndromal de­pressive symptoms in the absence of a diagnosable nonaffective illness from the Washington University list (9) of validated psychiatric disorders; exception was made for substance use disorders, which can complicate the early course of primary mood disorders before full syndromal episodes occur. These patients conformed to what we have elsewhere (1) defined as a subaffective dysthymic disorder, i.e., a form of dysthymia that is a subsyndromal but chronic manifes­tation of major depressive illness and that is referred to as primary dysthymia in the latest DSM-III revision. The nonpsychiatric control group (« = 11) consisted of patients with medical illness and asymptomatic volunteers without evidence of major psychiatric dis­order by DSM-III criteria. All of them were in some physical discomfort due to their medical condition or to the various sleep monitoring montages in which they had volunteered to participate. Such a comparison group was necessary in order to determine whether sleep EEG findings in the depressive and dysthymic groups could be distinguished from those with nonspecific effects of mild to moderate stress. The routine procedure in our sleep disorders center (5) consists of at least 14 days free of psychoactive drugs (except for "stable" nicotine and caffeine intake); the Minnesota Multiphasic Personality Inventory (MMPI) and the Beck Depres­sion Inventory (BDI); formal psychiatric evaluation completed blind to sleep polygraphy; at least two consecutive nights of sleep polysomnographic study consisting of continuous EEG, horizontal electrooculogram (EOG), submental electromyogram (EMG), and electrocardiogram (EKG). Categorical data were analyzed by chi-square (with Yates’ correction when appropriate) and continuous data by ANOVA. In all three groups, the mean age of subjects was around 40, and sex ratio nearly even. Insomnia was a prominent clinical feature of the anxious patients, which differentiated them from the dysthymics (P < 0.05). The anxious patients had moderately elevated BDI scores (15 ± 10) indicating a level of subjective dysphoria comparable with that of the dysthymics (17 ± 8). This finding underscores the similarity of cross-sectional dysphoric complaints in those two groups and explains the difficulties in clinical differential diagnosis. Further, on four MMPI scales measuring dysphoric and somatic symptoms, the only significant difference between the anxious and dysthymic groups was higher mean psychasthenia scores in the latter (P < 0.02). Finally, dysthymics appeared to be globally somewhat more symptomatic than the anxious patients as reflected in modestly higher mean numbers of elevations exceeding 2 SD on the 10 MMPI psychopathology scales (P < 0.05). Compared with dysthymic and control subjects, the anxious group experienced significantly greater difficulty adapting to the sleep clinic environment on night 1. This was manifested by lower mean (±SD) sleep efficiency percentage in the anxious versus the dysthymic and control groups (P < 0.001), and a higher mean (+SD) number of awakenings in the anxious versus the dysthymic and control groups (P < 0.001). In addition, the anxious group had significantly longer (P < 0.02) mean (±SD) REM latency (128 ± 173 min) compared with the dysthymics (66 ± 23 min) but not the controls (108 ± 39 min). Mean REM% (±SD) showed a similar pattern with anxious patients comparable with controls but significantly lower (P < 0.02) than the dysthymics. In brief, compared with the anxious group, dysthymics and controls had less trouble in sleep continuity on the night of adaptation, but dysthymics could be distinguished from the other two groups on the basis of shorter REM latency and higher REM%. Table 1 summarizes significant night 2 differences. The sleep efficiency of the anxious group had improved to a level comparable to that of the other two groups, but they continued to experience higher degrees of arousal as measured by greater mean number of gross awakenings and stage shifts; dysthymics were no worse than nonpsychiatric controls on measures of sleep continuity. The major differences between the groups were in REM measures: dysthymics had significantly shorter REM latencies and higher REM% compared with the other two groups, which were indistinguishable. Thus, although group differences are less prominent on night 2, anxious and dysthymic subjects can still be differ­entiated on selected measures of sleep continuity and REM sleep. Comparing the variability of the three groups across nights demonstrated that anxious subjects experienced greater change (improvement) than the other two groups in sleep latency (P < 0.02), sleep efficiency (P < 0.001), and number of awakenings (P < 0.005). Overall, these findings testify to the greater adaptational difficulties of the anxious patients in sleep continuity. By contrast, REM measures appear relatively robust across nights in all groups. To summarize, despite overall comparable levels of dysphoria and chronicity, anxious depressive and dysthymic patients could be distinguished neurophysi-ologically. The performance of the sleep EEG in differentiating the two groups should be measured against the background of the failure of psychometric mea­sures to set them apart. Shortened REM latency appears to be the most robust difference. This difference has now been shown (8) to be accentuated with the cholinomimetic drug arecoline that reduces the latency in primary mood dis­orders but not in primary anxiety disorders. From a clinical standpoint, the increased number of awakenings, especially in the early part of the night, in the anxious group are significant in that they underline the patients’ tendency to achieve normal sleep patterns only in the last part of the night, hence the irri­tability, fatigue, and somnolence—the "neurasthenic" triad—so characteristic of anxious neurotics.

This study (2,3) compared masked depressives (n = 25) with classical major depressives (n = 20) and anxious depressives (n = 22) with comparable age and sex (Table 2). Masked depressives were affectively ill patients who were referred from primary care settings with autonomic manifestations of anxiety and who denied subjective depressive complaints. Such patients, who constitute a large segment of the psychiatrically ill population seen in general medical practice, may be considered "crocks," "hysterics," or "hypochondriacs." They usually are treated with benzodiazepines or other sedative-hypnotics on which they often become dependent. In our sample, which was free of psychoactive medication for 2 weeks prior to sleep study, lethargy, insomnia, fear of choking in sleep, chest pain, and impotence were the most common presenting complaints; except for subjective depression, they also met the Washington University (9) and DSM-III (6) criteria for major depression. Compared with major depressives diagnosed using both Washington University criteria (9) and DSM-III (6), masked depressives denied psychological manifes­tations of the depressive syndrome and showed low BDI and MMPI D (depres­sion) scores; furthermore, as judged by the number of elevated scales above 2 standard deviation, they were less likely to volunteer psychopathology on all the scales of the MMPI. When compared with patients who met the full criteria for hysterical neurotic, and sociopathic and anxiety disorders, masked depressives had shorter REM latencies similar to those of primary major depressives. Such findings suggest that some primary depressions with a protracted course occur in the absence of subjective depression; autonomic nervous system manifestations of anxiety or other somatic complaints are the major presenting complaints.

The total scores of the scales, and individual clinical, personality, and bio­graphical items were subjected to Principal Component Analysis. A series of sequential analyses indicated the stability of the clusters derived. Figure 1 illus­trates a typical outcome of this analysis, with symptoms and personality scores included. It can be seen that the upper right-hand quadrant contains a cluster of items that are consistent with the entity depression; similarly in the upper left-hand quadrant is a cluster consistent with the term anxiety. These data were reanalyzed a number of times, leaving out a variety of items on each occasion and were also analyzed separately for sex. No matter which items were included or excluded, the basic patterns established remained invariant.
(more…)