Laboratory evaluation for cushing's syndrome in psychiatric patients with cortisol nonsuppression following the overnight dexamethasone suppression test

1264

BIOL

PSYCHIATRY

1987;22:1264-1270

Laboratory Evaluation for Cushing’s Syndrome in Psychiatric Patients with Cortisol Nonsuppression following the Overnight ~examethasone Suppression Test Thomas H. Lampe, Bruce L. Fariss, Steven C. Risse, Murray A. Raskind, and Stephen R. Plymate

Luboratory tests used for the di~erentia~ d~agno.~~s of Gushing’s syndrome have infrequently been employed in jnvest~gations of psychiatric patients who demonstrate hypothalamic-pituitary-adrenul (HPA) overactivity, and these laboratory procedures have not previously been applied for the speciJLicpurpose qf further evaluating the endocrine function of psychiatric patients with serum cortisol nonsuppression following the standard I -mg overnight Dexamethasone Suppression Test (DST). Low-dose (4 mgi48 hr) and hig~~-do~~e(16 rng~4~ far) DSTs were administered to IO psyrhiatri~ patients who e.~~libited cortisol nonsuppression aj>er the overnight DST. Patients ail had normal suppres.s~on to both the low-dose und high-dose tests. HPA overactivity in these patients was thus not sufJicient to meet laboratory criteria for the diagnosis of Cushing’s syndrome. Stud? results suggest that psychiatric patients with abnormal cortisol suppression following the I-mg overnight DST are like/y to have normal responses when assessed by standard laborato~ protocols used for the diagnosis of Gushing’s syndrome,

Introduction Nonsuppression of serum cortisol after a single-dose overnight Dexamethasone Suppression Test (DST) may be indicative of Cushing’s syndrome (Pavlatos et al. 1965), depression (Carroll et al. 19811, or, in some instances, other psychiatric disorders (Hirschfeld et al. 1983). Cushing’s syndrome and depression, in particular, have clinical as well as biochemical features in common. In a recent study of Gushing’s syndrome, 25 of 30 patients (83%) met strict criteria for an episode of affective disorder during the course of their endocrine illness (Haskett 1985). Thus, patients with psychiatric symptoms and coexisting hy~thaiamic-pituita~-adrenal (HPA) overactivity as manifested by cortisol

From the Seattle/American MAR.):

Lake Veterans

the Department

Departments

of Psychiatry

Washington,

Seattle,

Address

reprintrequests

Supported Received

of Clinical

Q 1987 Society of Biological

30,

Investigation.

and Behavioral

Geriatric Madigan

Sciences

(T.H.L.,

Research, Army

Education

Medical

S.C.R..

and Clinical

Center,

M.A.R.)

Tacoma

and Medicine

Center (T.H.L., (B.L.F., (S.R.P.).

S.R.P.);

S.C.R., and the

University

WA. to Dr. T. H. Lampe.

in part by rhe Veterans September

Administration

GRECCfl82B.

1986; revised February

Psychiatry

American

Lake VA

Medical

Center,

Tacoma,

WA

98493.

Admjnis~tion, 14. 1987.

0006-322U871503.50

of

Gushing’s Syndrome

in Psychiatric Patients

BIOL PSYCHIATRY 1987;22:126&1270

1265

nonsuppression following the single-dose overnight DST will at times raise the issue of differentiating primary psychiatric disorders from Cushing’s syndrome. The issue of differentiating Cushing’s syndrome from primary psychiatric disease is appropriately raised in psychiatric patients when cortisol nonsuppression is accompanied by findings suggestive of Cushing’s syndrome. The issue of differentiating Cushing’s syndrome from psychiatric disorders associated with HPA overactivity may also be relevant to some psychiatric patients without clinical features of Cushing’s syndrome, as psychiatric symptoms may occasionally precede the development of physical signs in patients with Cushing’s syndrome (Haskett 1985). For example, it may be difficult to differentiate early Cushing’s disease with complicating depressive symptomatology but without the physical stigmata of the endocrinopathy from primary depressive disorder (Gold et al. 1984). Furthermore, it has been suggested that subsets of patients with Cushing’s disease and depression may exist on a pathophysiological continuum (Gold et al. 1984; Gold 1985). One potential adjunct to the differential diagnosis of psychiatric disorders and Cushing’s syndrome associated with nonsuppression of cortisol following the I-mg overnight DST is the application of the more extensive laboratory procedures used to evaluate individuals with suspected Cushing’s syndrome (Liddle 1960; Ashcraft et al. 1982). However. the responses of psychiatric patients with HPA overactivity to such procedures have infrequently been determined (Butler and Besser 1968; Stokes et al. 1973, and these laboratory procedures have not previously been applied for the specific purpose of further evaluating the endocrine function of psychiatric patients with serum cortisol nonsuppression following the 1-mg overnight DST. The present study was conducted to evaluate HPA overactivity in psychiatric patients with overnight DST nonsuppression by laboratory methods that would facilitate comparison with established patterns of HPA overactivity representative of Cushing’s syndrome. Specifically, we asked the following question: Do psychiatric patients with significant cortisol nonsuppression after an overnight I-mg DST have normal responses when challenged with the more extensive low-dose (4 mg/48 hr) and high-dose (I 6 mg/48 hr) Dexamethasone Suppression Tests (Liddle 1960; Ashcraft et al. 1982) used to evaluate individuals with suspected Cushing’s syndrome?

Methods This study was approved by the Human Subjects Committee of the American Lake Veterans Administration Medical Center, Tacoma, WA. Subjects were 10 randomly chosen inpatients of the Acute Psychiatry Service at the American Lake Veterans Administration Medical Center who had exhibited nonsuppression of serum cortisol after an overnight I-mg DST and who gave written informed consent for participation. Patients were eligible for inclusion if they exhibited serum cortisol levels greater than 5 pg/dl at 4:00 PM after an overnight DST with 1 mg of dexamethasone at I I:00 PM the previous evening and were likely to remain hospitalized for at least IO days. The initial overnight DST was administered to subjects during the first week of hospitalization, and nursing staff visually monitored oral intake of dexamethasone to ensure compliance. Patients were free of active medical disease and did not have manifestations of Cushing’s syndrome on physical examination. None of the patients had other medical conditions that have been reported to result in “false” responses to a I-mg overnight DST, and patients with a history of alcohol abuse before admission (or in whom such use was suspected) were excluded from participation (Carroll et al. I98 I). Patients were treated with psychotropic

1266

BtOL PSYCHIATRY I Y87;22:1261-1270

T.H. Lampe et al.

medications as clinically indicated during their participation in the study, but none was receiving psychotropic or any other medications reported to produce “false” nonsuppression to a DST (Carroll et al. 1981). The 4:00 PM cortisols following an overnight DST of the patients enrolled in the study ranged from 7 @g/d1 to 23 pgldl. Psychiatric diagnoses for each patient were determined by DSM-III criteria (APA 1980). Nine of i0 patients were men. These and other patient characteristics are presented in Table I. Low-dose and high-dose DSTs, as described by Liddle (1960) and modified by Ashcraft et al. (1982), were initiated at least 6 days after completion of the initial overnight DST and were administered over 4 consecutive days. Beginning at 6:00 AM on day I, patients received 0.5 mg of dexamethasone by mouth every 6 hr for 2 days. A blood sample for serum cortisol determination was drawn from each patient at 4:00 PM on day 2 (low-dose protocol). Beginning at 6:00 AM on day 3 of the study, patients received 2.0 mg of dexamethasone by mouth every 6 hr for 2 days. A blood sample for serum cortisol determination was drawn from each patient at 4:00 PM on day 4 (high-dose protocol). A single-dose, I-mg overnight DST was repeated 5-7 days after completion of the highdose dexamethasone protocol. Patients again received 1 mg of dexamethasone by mouth at I l:OO PM, and a blood sample for serum cortisol determination was obtained at 4:00 PM the following day (Carroll et al. 1981). Serum cortisol was measured by a double-antibody radioimmunoassay following ether extraction. The cortisol antibody was obtained from Accurate Scientific Company, Westbury, CT. The interassay and intraassay coefficients of variation were 14.2% and 10.7%. respectively. The lower limit of assay sensitivity was 0.5 pgidl.

Results Patient responses to the low-dose (4 mg148 hr) and high-dose (16 mg148 hr) dexamethasone suppression protocols are shown in Table 1. On the low-dose protocol, 10 of IO patients had serum cortisol levels at 4:00 PM on day 2 of 5 pg/dl or less. In response to the highdose protocol, all patients suppressed serum cortisol to 4 p.g/dl or less at 4:00 PM on day 4. As previously demonstrated, these responses to the low-dose and high-dose tests fall in the normal range (Ashcraft et al. 1982). Repeat overnight DST evaluation was obtainable on 7 of the 10 patients, and these results are presented in Tabie 1. Five of these 7 patients continued to demonstrate nonsuppression (serum cortisol >5 pgldl). Of the two patients with normalization of DST response, one demonstrated significant concurrent clinical improvement, and one was partially improved. Of the three patients unavailable for repeat overnight DST assessment, one had been discharged from the hospital in an improved condition, one patient required transfer to another institution for involunta~ hospit~ization because of severe psychotic symptoms and refusal to remain volunt~ly hospitalized, and one patient with significant ongoing depressive symptoms had been transferred to the medical service for management of an acute bronchitis.

Discussion In this study, IO psychiatric patients who demonstrated nonsuppression of serum cortisol following the I-mg overnight DST were systematically administered the more extensive dexamethasone laboratory tests used for the clinical evaluation of individuals with suspected Cushing’s syndrome (Liddle 1960; Ashcraft et al. 1982). Patients had normal

Cushing’s Syndrome

1267

BIOL PSYCHIATRY 1987;22:12641270

in Psychiatric Patients

Table 1. Subject Characteristics and Responses to Dexamethasone Suppression Protocols Subject no. I

2

3 4 5 6 I 8 9 IO

Diagnosis Major depression Major depression Major depression Schizoaffective disorder, depressed Bipolar disorder, depressed Major depression Dysthymic disorder Major depression Major depression Schizoaffective disorder, depressed

be

Sex

Initial overnight DST

Low-dose DST

High-dose DST”

Repeat overnight DST”

10.5

IS


33

23 8

3 I I

F

I7

5

2

II

M M M M M

8 I4 I9 7 IO

4 3 2 2 I

I 3 2 2 I

Not done Not done 4 I5 I

59 72 33 2x

M M M M

49 42 76 35 63 21

9

10


7 Not done

results following both low-dose (4 mg/48 hr) and high-dose (16 mg/48 hr) DST protocols and thus did not manifest biochemical evidence of Cushing’s syndrome (Ashcraft et al. 1982). Although our sample size was small, the consistency of study results suggests that psychiatric patients with nonsuppression of serum cortisol following the 1-mg overnight DST, but without physical findings of Cushing’s syndrome, are likely to have normal responses to these diagnostic tests for Cushing’s syndrome. Two of seven patients had normal cortisol suppression on the repeat overnight DST. Their normalization might have reflected an attenuation of preadmission stressors that contributed to initial overnight DST nonsuppression and then diminished after hospitalization, or alternatively, normalization might have reflected the success of treatment (Carroll 1985). Five of seven patients retested with the overnight I-mg DST showed continued cortisol nonsuppression, and two of three patients unavailable for this retesting had significant ongoing psychiatric symptoms. It is therefore unlikely that normalization of HPA overactivity due to effects of time or response to treatment accounted for the consistent and normal responses of patients to the multiple-dose (4 mg/48 hr and 16 mg/48 hr) DST protocols. Two previous investigations of HPA overactivity in depressed patients have incorporated a low-dose (4 mg/48 hr) DST strategy (Butler and Besser 1968; Stokes et al. 1975). These studies determined serum cortisol at 9:00 AM, following administration of 0.5 mg of dexamethasone every 6 hr for 48 hr. In the study of Stokes et al. (1975), 20 of 27 depressed patients had persistent nonsuppression (cortisol >5 kg/dl) at 9:00 AM. In an earlier investigation by Butler and Besser (1968), 3 of 3 depressed patients studied had nonsuppression of serum cortisol on the low-dose (4 mg/48 hr) DST protocol. The only previous report of the response of psychiatric patients to a high-dose (16 mg/48 hr) DST protocol is that of Butler and Besser (1968), discussed above, who also administered a high-dose (16 mg/48 hr) DST protocol to 2 of their 3 depressed patients. After 2.0 mg dexamethasone every 6 hr for 2 days, one patient had partial suppression of serum cortisol(10 kg/dl). Surprisingly, the other patient had continued nonsuppression of cortisol (- 18 kg/dl) on the high-dose DST of a magnitude suggestive of the presence

1268

BIOL PSYCHIATRY 1987;22: IX+ 1270

T.H. Lampe et al.

of an adrenocortical adenoma (Ashcraft et al. 1982). Butler and Besser (1968) noted that the responses of their patients were similar to those of patients with Cushing’s syndrome. Our findings of consistent cortisol suppression following the low-dose (4 mg148 hr) and high-dose (16 mg148 hr) DSTs are thus difficult to reconcile with the aforementioned studies and may reflect differences in study design. In the present study, responses to the low-dose DST were assessed by serum cortisol levels at 4:00 PM on day 2 of the test, whereas the aforementioned studies determined cortisol at 9:00 AM, a time that may overlap with the early morning major circadian peak of serum cortisol (Krieger 1975). Although female subjects predominated in these two previous investigations, 9 of 10 subjects in our study were men. Stokes et al. (1975) reported that a majority of their patients who participated in the 4 mg148 hr DST protocol received nighttime medication (predominantly barbiturates); after 48 hr and a cumulative dose of 4 mg dexamethasone, patients receiving no medications had significantly lower cortisol levels than medicated patients. It is now recognized that barbiturates may contribute to “false” nonsuppression of cortisol following a DST (Carroll et al. 1981). The three patients studied by Butler and Besser (1968) had unspecified medications that were discontinued only 2 days prior to investigation. Finally, criteria for subject selection were different in the present and these two previous studies. Selection criteria in these previous studies were designed to enroll patients with primary affective disorder of moderate to severe proportions. Psychiatric inpatients were selected for participation in our study on the basis of cortisol nonsuppression following a I-mg overnight DST, irrespective of severity of illness or specific psychiatric diagnosis. Thus, it is possible that the subjects studied by Stokes et a1. ( 1975) and Butler and Besser (1968) were more severely depressed than our subjects. Two recent investigations (Gold et al. 1986; Schlechte et al. 1986) indicate that the HPA overactivity associated with depression and Cushing’s disease may indeed be differentiated by biochemical criteria, and thus, they are of relevance to our finding that psychiatric patients with overnight DST nonsuppression consistently fail to exhibit responses characteristic of Cushing’s disease when challenged with a low-dose (4 mg/48 hr) DST. Schlechte et al. (1986) reported that depressed patients with cortisol nonsuppression on the I-mg overnight DST had significantly lower mean 24-hr plasma free cortisol than patients with Cushing’s disease. Moreover, only 1 of 10 patients with depression had a plasma free cortisol value that was in the range of values observed in the 6 patients with Cushing’s disease. Gold et al. (1986) have recently reported that depressed patients exhibiting basal hypercortisoiism displayed attenuated plasma adrenocorticotrophic hormone (ACTH) responses to ovine corticotropin-releasing hormone, whereas patients with Cushing’s disease had exaggerated plasma ACTH responses to corticotropin-releasing hormone, despite basal hypercortisolism. These findings were interpreted by the authors to suggest that the co~ico~oph cell in the pituitary of depressed patients responds appropriately to the negative feedback of high cortisol levels, and thus, in depression, the HPA axis retains the characteristics of a feedback-regulated system. This proposed preservation of negative feedback regulation of the HPA axis in hypercortisolemic depressed patients is in accord with the findings of the present study. Although the overnight I-mg DST may provide an insufficient stimulus for negative feedback inhibition of the HPA axis in some psychiatric patients, the present findings indicate that the integrity of HPA negative feedback mechanisms in such patients is demonstrable when more extensive dexamethasone suppression strategies (e.g., a 4 mg/48 hr low-dose DST) are employed. Given the extensive HPA overactivity manifested by some psychiatric patients (Butler

Cushing’s Syndrome

in Psychiatric

BIOL PSYCHIATRY 1987:22: 1264-1270

Patients

1269

and Besser 1968; Stokes et al. 1975; Carroll et al. 1976a,b), it is certainly possible that occasional psychiatric patients will demonstrate a “cushingoid” response when challenged by the low-dose (4 mg/48 hr) DST, as employed in the present study. Such psychiatric patients would constitute a heuristically important subgroup for further study of the possible relationship between affective disturbances and Cushing’s disease (Brown et al., 1985; Gold 1985). Our results suggest that such patients, if identified, will constitute a small subset of psychiatric patients evaluated by this procedure. Whether or not patients with early Cushing’s syndrome without physical signs of the disorder yield consistently abnormal results to such DST challenges also awaits verification, as suitable cohorts for such investigations are, by present criteria, very difficult to identify. Evaluating the potential of the low-dose DST (4 mg/48 hr), as employed in the present study, in differentiating psychiatric disorders from Cushing’s disease as the cause of HPA overactivity and in exploring the possibility of a pathophysiological continuum between psychiatric disorders with HPA overactivity and Cushing’s disease, will necessarily await the application of this procedure to larger groups of psychiatric patients and to patients with suspected early Cushing’s syndrome. In summary, 10 psychiatric patients with evidence of HPA overactivity on the standard overnight DST had normal noncushingoid responses to laboratory protocols used to evaluate individuals with suspected Cushing’s syndrome (Liddle 1960; Ashcraft et al. 1982). These results suggest that psychiatric patients with nonsuppression of serum cortisol after a standard overnight DST, who do not have physical stigmata of Cushing’s syndrome, are likely to have normal responses to the low-dose (4 mg/48 hr) DST when 4:00 PM serum cortisol on the second day of the test is used to assess response. The authors thank the ALVAMC Acute Psychiatry Service Staff; Mina Garrison and Lyndel Cubberley, R.N., for technical assistance; Drs. R. Veith, B. Roos, and P. Gkonos for helpful comments; and Debra Mills and Mary Mason for manuscript preparation.

References American Psychiatric Association Washington, DC: APA.

(1980): Diagnostic and Statistical Manual of Mental Disorders.

Ashcraft MW, Van Herle AJ, Vener SL, Geffner DL (1982): Serum cortisol levels in Cushing’s syndrome after low- and high-dose dexamethasone suppression. Ann Intern Med 97:21-26. Brown WA, Keither G, Qualls CB, Haier R (1985): The Dexamethasone pituitary adrenocortical function. Arch Gen Psychiatry 42: 12 l-123. Butler PWP, Besser GM (1968): Pituitary-adrenal i:1234-1236. Carroll BJ (1985): Dexamethasone Psychiatry 46: 13-24.

Suppression

Suppression

function in severe depressive

Test: A review of contemporary

Test and

illness. Lancer confusion,

Carroll BJ, Curtis GC, Mendels J (1976a): Neuroendocrine regulation in depression. system-adrenocortical dysfunction. Arch Gen Psychiatry 33: 1039-1044.

J Clin

I. Limbic

Carroll BJ. Curtis GC, Mendels J (1976b): Neuroendocrine regulation in depression. II. Discrimination of depressed from nondepressed patients. Arch Gen Psychiatry 33: 1051-1058. Carroll BJ, Feinberg M, Greden FJ, Tarika J, Albala AA, Haskett RF, James NM, Kronfol Z, Lohr N, Steiner M, de Vigne JP. Young E (1981): A specific laboratory test for the diagnosis of melancholia. Arch Gen Psychiarry 38: 15-22. Gold PW (1985): Corticotropin-releasing factor stimulation in hypercortisolemic In Chrousos GP (moderator), Clinical applications of corticotropin-releasing Med 102:352-355.

psychiatric states, factor. Ann Intern

1270

T.H. Lampe et al.

BIOL PSYCHIATRY 1987;22: 1264- 1270

Gold PW, Chrousos G, Kellner C, Post R, Roy A, Augerinos P, Schulte H, Oldfield E, Loriaux Dl_ (1984): Psychiatric implications of basic and clinical studies with corticotropin-releasing factor. Am J Psychiatry 141:619-627. Gold PW, Loriaux DL, Roy A, Kling MA, Calabrese JR, Keilner CH, Nieman LK. Post RM, Pickar D, Gailucci W, Avgerinos P, Paul S, Oldfield EH, Cutler GB, Chrousos GP (1986): Responses to corticotropin-releasing hormone in the hypercortisolism of depression and Cushing’s disease. N Engl J Med 314:1329-1335. Haskett RF (1985): Diagnostic categorization Am J Psyrhiutry 142:91 l-916.

of psychiatric

disturbance

in Gushing’s syndrome.

Hirschfetd RMA, Koslow SH, Kupfer DJ (1983): The clinical utility of the Dexamethasone sion Test in psychiatry. JAMA 250:2 172-2174.

Suppres-

Krieger UT (1975): Rhythms of ACTH and corticosteroid secretion in health and disease, and their experimental modification. J Steroid Biochem 6:785-791. Liddle GW (1960): Tests of pituitary-adrenal suppressibility drome. J Clin Endocrinol Metab 20: 1539- 1.560.

in the diagnosis

of Cushing’s

syn-

Pavlatos FC, Smiio RP, Forsham PH f 1965): A rapid screening test for Cushing’s syndrome. JAVA 193:96-99. Schlechtc JA, Sherman B, Pfohl B (1986): A comparison of adrenal cortical function in patients with depressive illness and Cushing’s disease. Hormone Res 23:1-8. Stokes PE. Pick GR, Stall PM, Nunn WD (1975): Pituitary adrenal function in depressed patients: resistance to dexamethasone suppression. J Psychiutr Res 1227 l-28 1,