Simultaneous panic and depressive disorders: Clinical and sleep EEG correlates

251

Psychiatry Research. 17, 25 l-259 Elsevier

Simultaneous Panic and Depressive and Sleep EEG Correlates Leon Grunhaus, David Rabin, Yossi H.arel, John Michael Feinberg, and Rita Hermann

Disorders:

Clinical

F. Greden,

Received August 7, 1985; revised version received December 5,1985; accepted January, 13, 1986. Abstract. Panic and depressive symptoms occur simultaneously in many depressed patients. To study the frequency of this association and to determine whether patients with simultaneous panic and major depression differed from those with only major depressive disorder (MDD) in clinical features and in sleep electroencephalographic (EEG) variables, we evaluated a total sample of 336 patients with M DD. Fifty-eight (17%) had both panic and MDD; 50 had complete data and were

matched for age and severity of illness with other patients having only MDD. Patients with simultaneous panic and depression had significantly higher ratings for psychic and somatic anxiety, and rapid eye movement (REM) latencies approximating normal values. Patients with only MDD (without panic disorder) rated significantly higher in guilt feelings and had shorter REM latencies. Our results suggest that the simultaneous occurrence of panic and depression is relatively frequent, is accompanied by differences in sleep EEG variables, and may

have implications for treatment. Key Words. Major depressive electroencephalogram.

disorder,

panic disorder,

sleep

Anxious and depressive symptoms occur simultaneously in many psychiatric patients. This combination of symptoms may create problems in the differential diagnosis and formulation of treatment of the primary condition. Considerable nosological research has been done in this area without consensus. Some authors (Mapother, 1926; Lewis, 1934, 1966) support the Unitarian view and categorize the anxiety symptoms as part of the affective illness. Others advocate a dualistic approach (Gurney et al., 1972; Roth et al., 1972; Schapira et al., 1972; Kerr et al., 1974; Mountjoy and Roth, 1982~) in which two syndromes can be identified on the basis of clinical features and treatment outcome. Finally, an intermediate position is promoted in the Research and Diagnostic Criteria (RDC) (Spitzer et al., 1978) and the DSM-III (American Psychiatric Association, 1980). These nosologies indicate that anxiety and depressive

Leon Grunhaus, M.D., is Assistant Professor of Psychiatry and Director, Inpatient Service. Clinical Studies Unit; David Rabin, B.A., is Research Assistant, Clinical Studies Unit; Yossi Harel, M.A., is Research Associate, Institute for Social Research; John F. Greden, M.D.. is Professor and Chairman. Department of Psychiatry; Michael Feinberg, M.D., Ph.D., is Associate Professor. Department of Psychiatry and Mental Health Research Institute; and Rita Hermann, A.C.S.W., is Senior Social Worker. Clinical Studies Unit for Affective Disorders, Department of Psychiatry, University of Michigan Medical Center, Ann Arbor. Ml. (Reprint requests to Dr. L. Grunhaus, Clinical Studies Unit. Department of Psychiatry. University of Mich. Med. Center, UHD 9702-0118, 1500 E. Medical Center Dr.. Ann Arbor. MI 48109-0118. USA.) 0165-1781:

86;$03.50

0 1986 Elsevier Science Publishers

B.V.

252 conditions can exist separately, but when they.occurjointly, the predominant diathesis is the affective illness. The acceptance of two separate anxiety syndromes in DSM-IIIand RDC, i.e., panic disorder and generalized anxiety disorder, seems to have promoted progress in this area. It appears that when panic disorder accompanies major depressive disorder (MDD), the resulting clinical syndrome may differ in a variety of parameters from either condition alone. For example, the prognosis of patients with panic disorder is poorer if they also manifest depressive features (Noyes and Clancy, 1976; Clancy et al., 1978). The response to antidepressant medication is less favorable in patients who have joint panic disorder and MDD (Van Valkenburg et al., 1982; Grunhaus et al., in press). In patients with anxiety disorders, MDD is associated more frequently with panic disorder than with generalized anxiety disorder (Dealy et al., 1981) and the incidence of psychopathology in the relatives of patients with MDD is higher when the proband also meets the diagnosis of panic disorder (Leckman et al., 1983a, 19836; Weissman et al., 1984). These preliminary reports suggest that the joint occurrence of panic symptomatology with MDD leads to a clinical syndrome with a more ominous prognosis. It is unknown whether the syndrome of MDD with simultaneous panic disorder is characterized by biological variables that differ from those found when either syndrome occurs in isolation. In this communciation, we address this question by extending our previous preliminary report (Grunhaus et al. 1984) on the sleep polysomnographic (sleep electroencephalographic) recordings in patients with simultaneous panic disorder and MDD compared to control patients with MDD alone. Our findings continue to indicate that patients with the joint syndrome have sleep electroencephalographic (EEG) tracings that are different from those of patients with MDD alone. We also report the frequency of this condition in our clinical setting. Methods The Clinical Studies Unit (CSU) of the University of Michigan is a 12-bed inpatient unit with an accompanying outpatient program that specializes in the research and treatment of affective disorders. Patients studied include both inpatients and outpatients. Our CSU diagnostic evaluation, described in greater detail elsewhere (Carroll et al., 1981) consists of a 14-day drug-free period (patients are completely free of psychotropic medications during this period): several clinical interviews; a structured diagnostic interview, the Schedule for Affective Disorders and Schizophrenia (SADS and SADS-Lifetime Version) (Spitzer and Endicott, 1975); a thorough medical evaluation; and a diagnostic family interview whenever feasible. Comprehensive diagnoses are formulated based on RDC and clinical criteria (Carroll et al., 1980). The study was designed to compare two groups of patients: those with simultaneous panic disorder and MDD (simultaneous subgroup) and those with only MDD (depressed subgroup). Information concerning panic disorder was extracted from the ratings of the current episode of illness in the SADS interview. For inclusion of patients into the simultaneous panic and MDD group, our first step was to determine whether they met RDC/ DSM-III criteria for MDD. For this we reviewed SADS interviews of those meeting MDD who also had a Hamilton Rating Scale for Depression score > 12. Our next step was to determine whether subjects met RDC/ DSM-III criteria for panic disorder; if so, we diagnosed both conditions regardless of whether subjects had MDD (this is a modification of written RDC/ DSM-IIIcriteria). None of the patients with only MDD (depressed group) met diagnostic criteria for panic disorder, either

253 for the current episode or in their life history (as reported by the SADS-L). Other inclusion criteria for both groups were: (1) Hamilton Rating Scale for Depression (HRSD) (Hamilton, 1960) score > 12; (2) no research exclusion criteria known to invalidate biological correlates (Carroll et al., 1981; Privitera et al., 1982); (3) no other illness that could account for the panic attacks; (4) absence of pregnancy; and (5) informed consent. Investigators and clinical raters were blind to research variables. Our initial sample consisted of 336 patients who met MDD criteria. Within this total sample, subjects meeting the joint diagnosis of MDD and panic disorder were matched for age (within a range of 5 years) and severity of illness (within a range of 5). To determine the “adjusted HRSD,” anxiety items - somatic and psychic anxiety - were extracted to give an adjusted score. Sleep recordings were performed following 2 weeks of drug-free status. Inpatients were recorded on 2 consecutive nights and while sleeping in their own hospital beds. Most outpatients came to the sleep laboratory 3 consecutive nights, with data for the first night being discarded to avoid the “first night” effect. The sleep variables were averaged for the “good” nights. Some outpatients had only 2 nights of recordings; in these, we include only the second night. These patients were studied during the initial stages of our sleep laboratory, hence, these methodological differences. The following definitions were applied when studying the sleep recordings: awake: time (minutes) awake between sleep onset and end of sleep; deltn sleep: stages III and IV; early morning awakenings (EMA): time (minutes) awake between end of sleep and arising (end of recording); REM activity (RA): the sum of the eye movement scores during REM sleep for the entire night, based on a score of O-8 for each minute of REM sleep; REM density: REM activity/ REM time (RA/ RT); REM latency: time (minutes) from sleep onset to onset of stage REM; REM latency minus awake (RLMA): REM latency minus any time awake between sleep onset and REM onset (minutes); REMpercent (REM%): RT/(TS-A) x lOO’& REM time(RT): time (minutes) spent in stage REM; sleep efficiency: (TS-A)/TRP x 100%; sleep latency: time (minutes) from beginning of recording to sleep onset; time sleeping (TS-A): time (minutes) from sleep onset to end of sleep minus awake; total recording period (TRP): time (minutes) from beginning of recording (“lights out”) to end of recording; REMperiod: a period containing at least three l-minute epochs of REM sleep, with not more than 30 minutes of non-REM sleep between any two epochs of REM (Feinberg et al., 1982); sleep onset: 2 3 minutes of continuous REM sleep, or 10 consecutive minutes of sleep with the first minute being stage 2,3,4, or REM sleep and at most 1 minute of time awake during this lo-minute period; sleep offset: last minute of stage 1, 2, 3, 4, or REM preceded by 9 consecutive minutes of sleep with 1 minute awake allowed. Ordinal data were analyzed with the x2 or the Fisher’s exact test. Matched clinical ratings were studied with paired Student’s t test, while the sleep variables and nonpaired measurements required the two-sample t test. Whenever variances were significantly different between the two groups, we used t tests adapted for unequal variances. Results Frequency of Simultaneous Panic and Depressive original 336 patients (17%) met the diagnostic criteria

Disorder.

Fifty-eight

of the

for MDD with simultaneous panic disorder. Another 23 patients (7%) had reported clinical complaints consistent with panic attacks, but the severity and frequency of the attacks failed to reach DSM-III/ RDC criteria; these patients were excluded from later data analyses. Clinical Features. To evaluate clinical features and sleep EEG variables in more detail, we then focused on those 58 subjects who had simultaneous MDD and panic features. Eight subjects had to be excluded because of missing clinical data. The remaining 50 were matched on age and severity of illness. Thus, our two subsamples

254

consisted of 50 subjects with simultaneous MDD and panic, and 50 matched subjects with only MDD. Table 1 presents the demographic and clinical data for both groups of patients. The sex distribution was similar in both groups (~2 = 0.184, NS). Samples were adequately matched for age and severity. The small differences in the HRSD ratings disappeared when the adjusted HRSDs were analyzed. The HRSD ratings for psychic and somatic anxiety were consistently higher in the group with simultaneous panic and depression, while the item of guilt was significantly higher in the group with MDD alone. We also analyzed other clinical features, including number of previous episodes of endogenous depression, age at first episode of depression, number of previous suicide attempts, number of previous hospitalizations and number of weeks since current episode. The only significant difference occurred in the number of previous episodes of endogenous depression; patients with only MDD had a significantly higher number of episodes (t = 10.23, p < 0.01). Table 1. HRSD mean ratings for depression patients with or without panic attacks (pirs) Age HRSD

total

Panic/ depression

(total and some items) for MDD

Depression

SD

50

35.26

35.50

2.83

NS

50

21.34

20.13

3.91

0.03

P

50

16.85

17.04

2.66

NS

psychic

anxiety

50

2.52

1.95

1.22

0.002

HRSD

somatic

anxiety

50

1.96

1.13

1.52

0.0008

HRSD

depressed

50

2.41

2.59

0.96

NS

HRSD

guilt

50

1.14

1.63

2.05

< 0.04

Adjusted HRSD

HRSD

mood

HRSD = Hamilton Rating Scale for Depression MDD = major depressive disorder.

Sleep EEG Results. Seventeen of the 50 patients with simultaneous panic disorder and MDD, and 16 of the 50 with only MDD had sleep EEG tracings. Of the patients with panic and MDD, only four had secondary MDD. Results are presented in Table 2. Age and severity (HRSD and adjusted HRSD) were similar in both groups. Patients with simultaneous panic and MDD had significantly longer sleep latencies @ < 0.02), their sleep was less efficient (p < 0.05), their RLMA was longer (p = 0.05), and their time awake 0, = 0.06) and REM latency (p 0.06) approached significance. Other sleep variables failed to differentiate between the two groups. The distribution of the REM latencies is shown in Fig. 1. Patients with simultaneous panic and MDD are distributed much less evenly than those with MDD alone. For all patients, REM latency grand mean was 68 minutes. To test if the distribution of the latencies around the grand mean was different between the groups, we performed a Fisher’s exact test, which was significant at the 0.03 level. These various analyses suggest that patients with simultaneous panic and MDD are less uniform in their REM latency findings. q

255 Table 2. Clinical variables and sleep measurements in MDD patients with and without panic attacks (mean + SD) Panic/depression In = 17)

Depression In = 161

Age HRSD

total

Adjusted

HRSD

EMA

37.13 +

8.66

32.94 k

a.32

NS

19.26 +

4.25

21.88 +

4.27

NS

16.06 2

3.75

16.68 f

3.92

NS

a.5

SLEEP

LAT

25

AWAKE

P

k 12

6.74 + 10.38

i 13.54

NS

44.88 f 32.17

0.021 0.061

17.31 + 14.02

30.65 zk 31.28

357.88 k 59.59

343.41 f 60.37

RL

54.78 2 24.83

81.76 t 63.57

0.061

RLMA

53.23 k 24.01

77.34 Ik 51.74

0.051

REM %

25.38 f

23.18 f

DELTA

17

TOT

S/A

RD SLEEP

EFF

5.13

k 21.40

1.56+

0.44

87.58 +

4.54

1.

t

for unequal

9.31

NS

17.76 i 18.40

NS

1.53 *

EMA= Early morning awakening; SLEEP LAT= RL = REM latency; RLMA = REM latency-awake;

NS

0.47

NS

81.71 k 11.56

0.051

sleep latency; TOT S/A= RD = REM density.

time sleeping;

variances.

Fig. 1. REM latencies in patients with MDD alone and simultaneous panic disorder.(PD)

MDD and

023s 018850

240

0161

g

go-

5

so-

4 l-

70-

3

60 50-

=

40-

l l

0 I

I .

MOD+ PO The distribution around exact, p = 0.031.

the grand

mean (66 minutes)

is significantly

different

between

the groups

(Fisher’s

256

Discussion Seventeen percent of our sample of 336 patients with MDD also met DSM-III or RDC criteria for simultaneous panic disorder. These data replicate and extend other reports indicating that the coexistence of the two syndromes is relatively frequent (Lewis, 1934, 1966; Cassidy et al., 1957; Woodruff et al., 1967). Although this association between panic attacks and MDD has been known for over 30 years, only recently have authors begun to explore its clinical, therapeutic, prognostic, and psychobiological significance. In their clinical profiles, patients with simultaneous panic and MDD scored consistently higher in HRSD ratings of somatic and psychic anxiety, and lower on guilt. Although patients with panic attacks might be expected to have high anxiety ratings, it should be remembered that depressed patients often report anxiety as a complaint. In our patients, we found no differences in age of the first episode of depressive illness, number of previous hospitalizations, number of previous suicide attempts, and number of weeks since episode began. The frequency of previous episodes of endogenous depression did differentiate between the groups; those without panic attacks had more episodes. It is unclear whether this difference suggests that patients with the joint syndrome have a higher frequency of chronic depression, as reported by some authors (Van Valkenburg et al., 1982). Our sleep EEG recordings suggest that patients with simultaneous panic disorder and MDD are less homogeneous than those with MDD alone. Patients with panic and MDD had more difficulty in both falling asleep and remaining asleep, and the overall efficiency of their sleep was poorer; their REM latencies varied considerably. These data extend those reported in a preliminary communication (Grunhaus et al., 1984). Dube et al. (1985) and Sitaram et al. (1984) reported recently on sleep EEG and arecoline infusion findings of patients with MDD and anxiety disorders. The patients were diagnosed with SADS/RDC criteria and divided into four groups: primary MDD, primary MDD with secondary anxiety disorder, primary anxiety disorder, and primary anxiety disorder with secondary MDD. The authors described significantly lower REM latencies and arecoline induction responses for both MDD groups, while the anxiety groups did not differ from normal controls. Their findings confirm previous suggestions that patients with primary depression (with or without anxiety) are significantly different from patients with secondary depressions (Kupfer, 1976; Kupfer et al., 1978). Unfortunately the authors included in the anxiety diagnosis patients with generalized anxiety disorder, panic disorder, and phobias, introducing confounding diagnostic variables whose impact on sleep recordings is difficult to assess. Their data also suggest that patients with MDD and anxiety, either primary or secondary, seem to constitute an “in-between” category, with REM latency in the primary MDD subjects being lower and in the primary anxiety subjects being higher. Akiskal et al. (1984) similarly reported that chronic dysthymics had lower REM latencies and higher REM To, while anxious depressives had more disruptions in their sleep. Thus, lower REM latencies again were associated with the affective diagnosis. These studies collectively appear to indicate that patients with primary depression who have a secondary diagnosis of panic disorder have higher REM latencies than

257 patients with only MDD. Perhaps the pathophysiology associated with the panic component may have a “normalizing” effect on REM patterns. Support for this suggestion comes from sleep studies in patients with generalized anxiety disorder (Reynolds et al., 1983) and panic disorder (Uhde et al., 1984). In both studies, anxious patients had REM latencies of 75 minutes and above. Further studies are indicated, ideally with all three groups. If findings remain consistent, this should be noted in future nosologic criteria for psychiatric research. Differences in clinical profile may be of relevance in the treatment of anxious/ panic/depressed patients. It has been reported that MDD patients with high anxiety ratings have poorer treatment outcome both with tricyclic antidepressant drugs and electroconvulsive therapy when compared to patients with lower anxiety ratings (Katz et al., 1982; Grunhaus et al., 1985). These effects of anxiety on treatment outcome of MDD patients also seem to apply to those with only panic disorder, suggesting that the associated panic disorder may be responsible for the poorer outcome (Van Valkenburg et al., 1982; Grunhaus et al., in press). Liebowitz et al. (1984) recently reported results of a placebo-controlled study on atypical depression; they found that patients with a history of panic attacks responded better to monoamine oxidase inhibitors. These data collectively suggest that patients with simultaneous panic and depressive disorders may prove to be more resistant to conventional tricyclic antidepressant regimens and may require more vigorous treatments or the use of monoamine oxidase inhibitors as drugs of preference. The authors thank Drs. Athanasios Zis, Roger Haskett, and A. Ariav Albala for generously allowing data from several of their patients to be incorporated in this report. The research reported was supported in part by Public Health Service Grant MH-28294; by the Mental Health Research Institute, University of Michigan, Ann Arbor; and by the State of Michigan Department of Mental Health, Lansing.

Acknowledgments.

References Akiskal, H.S., Lemmi, H., Dickson, H., King, D., Yerevanian, B., and Van Valkenburg, C.H. Chronic depressions. Journal of Affective Disorders, 6, 287 (1984). American Psychiatric Association. DSM-III: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. APA, Washington, DC (1980). Carroll, B.J., Feinberg, M., &eden, J.F., Haskett, R.F., James, N.McI., Steiner, M., and Tarika, J. Diagnosis of endogenous depression. Journal of Affective Disorders, 2, 177 (1980). Carroll, B.J., Feinberg, M., Greden, J.F., Tarika, J., Albala, A.A., Haskett, R.F., James, N.McI., Kronfol, Z., Lohr, N., Steiner, M., de Vigne, J.P., and Young, E. A specific laboratory test for the diagnosis of melancholia. Standardization, validation, and clinical utility. Archives of General Psychiatry, 38, 15 (1981). Cassidy, W.L., Flanagan, N.B., Spellman, N., and Cohen, M. Clinical observations in manic-depressive disease. Journal of the American Medical Association, 164, 1537 (1957). Clancy, J., Noyes, R.J., Hoenk, P.A., and Slymen, D.L. Secondary depression in anxiety neurosis. Journal of Nervous and Mental Disease, 166,845 (1978). Dealy, R.S., Ishiki, D.M., Avery, D.H., Wilson, L.C., and Dunner, D.L. Secondary depression in anxiety neurosis. Comprehensive Psychiatry. 22, 612 (198 I). Dube, S., Kumar, N., Ettedgui, E., Pohl, R., Jones, D., and Sitaram, N. Cholinergic REM induction response: Separation of anxiety and depression. Biological Psychiatry, 20,408 (1985).

258 Feinberg, M., Gillin, J.C.. Carroll, B.J., &eden, J.F., and Zis, A.P. EEG studies of sleep in the diagnosis of depression. Biological Psychiatry, 17, 305 (1982). Grunhaus. L., Alcser, K., Haskett, R.F., Greden, J.F., and Zelnik, T. Clinical and DST correlates of ECT response. Presented at the Annual Meeting of the American Psychiatric Association, Dallas, TX, May 18-24 (1985). Grunhaus, L., Rabin, D., Feinberg, M., Greden, J.F., and Hermann, R. Simultaneous panic and depressive disorder: Neuroendocrine and sleep-EEG correlates. Presented at the Annual Meeting of the Society for Biological Psychiatry, Los Angeles (1984). Grunhaus, L., Rabin, D., and Greden, J.F. Simultaneous panic and major depression. Journal of Clinical Psychiatry (in press). Gurney, C., Roth, M., Garside, R.F., Kerr, T.A., and Schapira, K. Studies in the classification of affective disorders: II. British Journal of Psychiatry, 121, 162 (1972). Hamilton, M. A rating scale for depression. Journal of Neurology, Neurosurgery, and Psychiatry,

23, 56 (1960).

Katz, M.M., Robins, E., Croughan, J., Secunda, S., and Swann, A. Behavioral measurement and drug response characteristics of unipolar and bipolar depression. Psychological Medicine, 12, 26 (1982).

Kerr, T.A., Roth M., and Schapira, K. Prediction of outcome in anxiety states and depressive illness. British Journal of Psychiatry, 124, 125 (1974). Kupfer, D.J. REM latency: A psychobiologic marker for primary affective disease. Biological Psychiatry, 11, 159 (1976). Kupfer, D.J., Foster, F.G., Coble, P., McPartland, R.J., and Ulrich, R. The application of EEG sleep for the differential diagnosis of affective disorders. American JournalofPsychiatry, 135,69

(1978).

Leckman, J.F., Merikangas, K.R., Pauls, D.L., Prusoff, B.A., and Weissman, M.M. Anxiety disorder and depression. American Journal of Psychiatry, 140, 880 (1983a). Leckman, J.F., Weissman, M.M., Merikangas, K.R., Pauls, D.L., and Prusoff, B.A. Panic disorder and major depression. Archives of General Psychiatry, 40, 1055 (1983b). Lewis, A.J. Melancholia: A clinical survey of depressive states. Journal of Mental Science, 80, 277 (1934).

A.J. Affective disorder. In: Bodley, S., ed. Price’s Textbook of the Practice of 10th ed. Oxford University Press, London (1966). Liebowitz, M.R., Quitkin, F.M., Stewart, J.W., McGrath, P.J., Harrison, W., Rabkin, J., Tricamo, E., Markowitz, J.S., and Klein, D.F. Phenelzine vs. imipramine in atypical depression. Archives of General Psychiatry, 41, 669 (1984). Mapother, E. Discussion on manic-depressive psychosis. British Medical Journal, II, 872 Lewis,

Medicine.

(1926).

Mountjoy, anxiety states. Mountjoy, anxiety states. Noyes, R.,

C.Q., and Roth, M. Studies in the I. Journal of Affective Disorders, C.Q., and Roth, M. Studies in the II. Journal of Affective Disorders, and Clancy, J. Anxiety neurosis:

,Wenral Disease,

relationship

between depressive

disorders

and

relationship between depressive disorders 4, 149 (19826). A 5-year follow-up. Journal of Nervous

and

4, 127 (19824).

and

162, 200 (1976).

Privitera, M.R., Greden, J.F., Gardner, R. W., Ritchie, J.C., and Carroll, B.J. Interference by carbamazepine with the dexamethasone suppression test. Biological Psychiatry, 17,611(1982). Reynolds, C.F. III, Shaw, D.H., Newton, T.F., Coble, P.A., and Kupfer, D.J. EEG sleep in outpatients with generalized anxiety: A preliminary comparison with depressed outpatients. P.s_vchiutry Research,

8, 8 1 (1983).

Roth, M., Gurney, C., Garside, R.F., and Kerr, T.A. Studies in the classification of affective disorders: I. British Journal of Psychiatry, 121, 147 (1972). Schapira, K., Roth, M., Kerr, T.A., and Gurney, C. Prognosis of affective disorders: The differentiation of anxiety states from depressive illness. British Journal of Psychiatry, 121, 175 (1972).

259

Sitaram, N., Dube, S., Jones, D., Pohl, R., and Gershon, S. Acetylcholine and alpha, adrenergic sensitivity in the separation of anxiety and depression. Psychopathology, 17 (Suppl. 3), 24 (1984). Spitzer, R.L., and Endicott, J. Schedule for Affective Disorders and Schizophrenia. New York State Psychiatric Institute, New York (1975). Spitzer, R.L., Endicott, J., and Robins, E. Research Diagnostic Criteriafor a Selected Group of Functional Disorders. New York State Psychiatric Institute, New York (1978). Uhde, T.W., Roy-Byrne, P., Gillin, J.C., Mendelson, W.B., Boulenger, J., Vittone, B.J., and Post, R.M. The sleep of patients with panic disorder. Psychiatry Research, 12, 251 (1984). Van Valkenburg, C.H., Akiskal, H.S., Puzantian, V., and Rosenthal, T. Anxious depressions. Journal of Affective Disorders, 6, 67 (1982). Weissman, M.M., Leckman, J.F., Merikangas, K.R., Gammon, G.D., and Prusoff, B.A. Depression and anxiety disorders in parents and children. Archives of General Psychiatry, 41, 845 (1984). Woodruff, R.A., Jr., Murphy, G.E., and Herjanic, M. The natural history of affective disorders. Journal of Psychiatric Research, 5,255 (1967).