Adrenal weight and morphology in victims of completed suicide

Adrenal Weight and Morphology in Victims of Completed Suicide Eva Szigethy, Yeates Conwell, Nicholas T. Forbes, Christopher Cox, and Eric D. Caine

Several lines of evidence have implicated hypothalamic-pindtar)~-adrenal ( HPA ) axis dysfunction in major depression amt suicidal behavior. In the present study, the weight and morphology of postmortem adrenal glands were compared between suicide victims and sudden death, nonpsychiatric controls. The mean adrenal weight of the convbined left and right glands was significantly higher h~ the suicide group: this difference was accounted for specifically by increases in left adrenal weight of suictdes compared with the control group. There was a positive correlation between adrenal weight and total cortical thickness in both left and right glands, providing direct evidence that increased adrenal weight in suicide victims is due to cortical hypertrophy. The finding of left-right adrenal weight as.wnmeto' in suicides is consistent with the hypotheszs of abnormal lateralized b~putfrom higher control centers of the HPA ca'is. Key Words: Adrenal, weight, morphology, suicide, depression, HPA axis

Introduction Numerous studies have demonstrated a range of abnormalities in hypothalamic-pitmtary-adrenal (HPA) axis function in patients with major depression. The replicated findings it, t o h l ~~ . . . . . .

. .

htrrt,arr, r~rt~c~l~,~-,~ altkl ,~.-I ¢lCVdtlL-aLl ~1 . . . . . . .-t d.~'l'-Ill '~A t.._ u [ ' i n a i - y e x ,,:t~:~v~uav=~ltLia

of cortisol (Gibbons 1964; Sachar et al ! 973; Rubin et al 1987); a resistance to suppression by dexamethasone of co~icotropin, beta-endorphin, and cortisol (MeadorWoodruff et al 1987: Reus et ai 1982; Carroll et al 1976;

cretion

From the Um~er~tty of Rochester Schot~l o! Methcme (ES. YC. NTF, CC, EDC), the Department, of P~ychtatr3 (YC EDCL and Bio~tatt,ttcs (CC), the UR-NIMH Cltmcal Research Center for the Study of P,yehopathology of the Elderly IYC. NTF. CC EDC) and the Office of the Medtcal Examiner of Monroe County I NTF) 4,ddres.s repnnt requests to Yeates Cony, ell. MD. Umver~lty of Rochester Medlc',.d Center, 300 Cnttenden Bl~d. Rochester. NY 14642 Tht~ v, ork ~a.s supported m part by NIMH Grant MH4fi'~81 (UR-NIMH Chmcal Research Center for the Study of Psychopathologyof the Elderly). Dr Szlgethy Is now ~ tth Case Western Rese~,e Umverslty Dr E~a Szageth~', current addre~_, ts Case Western Reserve Umverstt~. Columbus. OH Recet~ ed August 4. 1993. re~-tsed January I 0. 1994

© 1994 Society of Biological Psychmtry

Carroll 1982; Jaeckle et al 1987) augmented secretion of endogenous cortisol in response to exogenously administered corticotropin [adrenocorticotrophic hormone (ACTH)] (Amsterdam et al 1983). Others have failed to find significant differences in the HPA axis functioning of depressed and normal individuals (Zimmerman et al 1986; Shopsin and Gershon 1971; Carpenter 197 !; Krishnan et al 1990; Fang et al 1981). Differences in age, timing and frequency of samples, characterization of depression, and failure to distinguish early escape from absolute resistance have been postulated to explain these discrepancies. A blunted ACTH response in ,lepressives to intravenously infused o~ine (Gold et al 1986; Kathol et al 1989; Amsterdam et al 1987a) and human (Holsboer el al 1985; Rupprecht et al 1989) corticotropin releasing factor (CRF) has been reported as well. These findings suggest an abnormality in the regulation of HPA ~xis function in major depressives above the level of ff,~: pituitary, with hypersecretion of CRF, down regulau~a of corticotroph CRF receptors, and hypertrophy of the adrenal gland. Increased 0006-3223/94/$07 00

Adrenal Gland in CompletedSuicides

tttoLPSYCF',I.~,~rRY

375

ltttlt4=~6~37*-3tm

concentrations of CRF in the cerebrospinal fluid (CSF) of depressed patients in some (Banki et al 1987; France et al 1988), but not all studies (Roy et al 1987; Geracioti et al 1992), support this model, as do recent reports by Amsterdam and colleagues (I 987b) and Nemeroff et al (1992) of increased adrenal gland volumes in major depressives, as compared with controls measured by computed tomography. HPA axis dysfunction has also been implicated in patients with suicidal behavior although conflicting data, perhaps due to diagnostic and behavioral heterogeneity of subjects, have made interpretation of the results difficult. Elevated urinary excretion of co.'tisol (Ostroff et al 1982) and 17-hydroxycorficosteroid (Bunney et al 1969) have been found in patients who attempted or completed suicide during prospective follow-up. In postmortem studies, CSF CRF levels were significantly higher (Arato et al 1989), and CRF binding sites in the frontal cortex significantlyreduced (Nemeroff et al 1988) in suicides compared with sudden death controls. Others, however, have failed to show any differences between depressives with and without suicide attempts in plasma (Kocsis et al 1986; Roy 1992) and CSF levels of cortisoi (Kocsis et al 1986: Traskman et al 1980; Arato et al 1986), or CSF levels ofCRF (Arato et al 1986: Roy 1992). A number of groups have found that escape from dexamethasone suppression and failure of Dexamethasone Suppression Test (DST) to normalize predict suicidality (Yerevanian et al 1983; Greden et al 1980; Banki and Arato ! 983; Coryell and Schlesser 1981; Carroll et al 198 ! ), whereas others have failed to find that association (Krieger 1979; Zimmerman et al 1986; Schmidke et al 1989: AyusoGutierrez et al 1987; Brown et al 1986). No significant differences in predexamethasone plasma cortisol levels have been noted between suicide attempters and nonsuicidal comparison samples (Ayuso-Gutierrez et al 1987: Brown et al 1986; Roy 1992). The few studies that have examined the HPA axis in victims of completed suicide through the analysis of postmortem tissue have yielded mixed results. Nemeroff et al (1988) found significantly reduced CRF receptor binding sites in the frontal cortex of 26 suicide victims compared with controls, a finding that they concluded was consistent with chronic CRF hypersecretion in depression. McCrank and colleagues (1993) found no statistically significant differences in adrenal weight among violent suicides, nonviolent suicides and sudden death controls. Dorovini-Zis and Zis (1987), however, found that the adrenal glands of suicide victims were significantly heavier than the glands of controls, a finding they interpreted to reflect progressive functional and anatomical hypertrophy of the gland in response to chrome h)~rstimulation by ACTH. Limitations include the lack of ilffonnation concerning toxicologic screening or exclusion cri,*eria; clinical data regarding psy-

chiatric illness~ stressful life cixcurnstar~-~s, and medical illness burden; or examination of adrenal morphology. The present study was designed with the objectives of replicating the findings of Dorovini-Zis and Zis f 1987), and extending them by the inclusion of morphometric analyses to directly test the hypothesis that increased adrenal weight in suicide victims is due to hypertrophy of the cortex in a pattern consistent with chronic stress, Methods The adrenal glands employed for this study were a sample of convenience taken from autopsies performed on 12 suicide victims and 22 sudden death controls by the Office of the Medical Examiner of Monroe County (ME) between January 1990 and December 1991. During this 2-year period there were 134 suicide deaths investigated by the ME, of which 107 were men and 27 were women, with a mean age of 44. | years (range 14--91). The suicides included in this study were representative of the larger group: nine wen and three women with mean age ( _ SEM) of 45.4 (_ 3.4) years (range 21-71) years. Methods of suicide were shooting (n = 4), hanging (n = 4), drug overdose (n = 3), and hypothermia from cold exposure (n = 1). The mean postmortem delay was 14.8 hr; range 2-25 hr. The control group consisted of adrenal glands taken from 22 individuals who had died suddenly from violent or natural causes over the same period of time. They included 17 men and 5 women with a mean (_ SEM) age of 36.4 (_ 3.2) years and postmortem delay of 12.2 hr (range 2-22). Their causes of death were shooting (n = 8), stabbing (n = 2), other acute trauma (n = 8), acute coronary insufficiency(n = 3), and drowning (n = 1). Results of the forensic investigation, including the physical autopsy, were used to characterize a su~ect's physical health status. The height and weight of each victim were recorded. We quantified total burden of each subject's physical dlnesses, unrelated to his/her immediate cause of death, using the Cumulative Illness Rating Scale (CIRS) (Linnet a~ 1968), a measure of pathology in each of 13 organ systems shown to be highly correlated with physical assessments made antemortem (Conwell et al 1993). Mean C1RS scores for suicides and controls were 3.0 and 2.1 respectively, levels that connote little focal or systenaic tlssue pamoiogy. Data on handedness were not available. Thirty-two of 34 cases were subjected to comprehensive toxicological screening that included gas chromatography (G-C) screening for volatile alcohols, barbiturates, acid/neutral drags, benzodiazepines, and chloral hydrate: immunoassays for cocaine, opiates, amphetamines, and marijuana: and thin layer chromatography screens for basic drugs and chemicals. Two cases received limited toxicology screening for alcohol, or cocaine and alcohol only. Positive findings were confirmed and quantitated by gas chromatogra-

376

E. Szigethy et ai

~aOLFS¥c,naTRY

1~4,~,374-380

phy, liquid chromatography, or GC-mass spectrometry where appropriate. We excluded subjects with a known past history of a drug or alcohol abuse disorder, but not if the presence of alcohol or illicit drugs in toxicological analyses was the sole finding suggesting substance abuse. In addition, subjects were excluded from the control group if the investigation yielded any suspicion that the death was intentionally self-inflicted, or there was any suggestion of diagnosable psychopathology. (The ME's investigators have been trained by our group to collect a variety of data that may point to the presence of clinically significant psychiatric disturbances. Though insufficient for diagnostic purposes, this process has proven to be valid for screening purposes.) At autopsy, the adrenal glands of suicide victims and controls were d~ssected from the body by a forensic pathologist, and the surrounding fat carefully removed from the capsule. Adrenals were excluded from study if autolysis was grossly evident at autopsy; if investigation revealed that a control or suicide victim had died during pregnancy or following severe, protracted or debilitating illness (i.e., malnutrition, cirrhosis, severe hypertension, cancer, or sepsis); or i f a subject had been prescribed exogenous steroids. The dissected glands were immediately fixed in a 10% buffered formalin solution. After a minimum of 16 months fixation each gland was measured (height, weight, and thickness) and weighed separately by an investigator blind to the manner of death. Each gland was tfi sectioned perpendicular to the main axis through the head, body, and tail. Eight p,m thick sections were cut through each of these three regions and histologically stained with haematoxylon and eosin. Under 10x microscopic magnification, ten measurements of width of the total cortex as well as the three cortical zones [glomerulosa (zG), fasciculata (zF), and reticularis (zR)] were made from each of three sections per region defined above. Only those sections qualified for analysis that were at right aneles to the g, rfaeo nf tho olnncl hacl an "'arcade arrangement" to the columns of fasciculata cells, and had medulla visible as a guarantee of the presence of the entire reticularis. Adrenals from seven suicides and 13 controls yielded sufficient sections for analysis of cortical thickness. Zonal widths were measured directly on the microscope stage by means of a translucent eyepiece micrometer calibrated against a stage micrometer. All microscopic measurements were performed by a single investigator (E.S.), blind to subject group, who re-measured a random selection of sections on each measurement day to monitor for technical variances. Statistical analysis consisted of the two-tailed t-test (assuming unequal variance, with approximate degrees of freedom) in comparing mean age, body weight and height, adrenal weight, cortical thickness, and respective zonal thicknesses between suicide and control groups. Analysis of

covariance (ANCOVA) was performed to examine the difference in adrenal weight between suicides and controls, corrected for body weight and, in other analyses, for cortical thickness. Each analysis included an examination of residuals as a check on the required assumptions of normally distributed errors with constant variance, p values less than 0.05 were considered significant.

Results There was no significant difference between suicides and controls in mean age (p = 0.066, t = 1.91, d f = 27), body weight (p = 0.777, t = 0.29, d f = 26), body height (p = 0.643, t = 0.47, df = 29), or sex distribution (Fisher's Exact Test, p = 1.000). Although there was no association between adrenal weight and age or gender distribution, adrenal weight and body weight were significantly correlated (p = 0.01). Therefore, age and gender distribution were not included in subsequent analyses, whereas adrenal weight and cortical thickness measurements were adjusted for body weight by ANCOVA. When adjusted for body ,~ eight, the mean weight ( + SEM) of left and fight adrenal glands combined was significantly greater for the suicide group (11.3 -4- 0.9 g) compared to controls (9.3 ___0.4) (p = 0.016, t = 2.56, d f = 31). The elimination of a single outlier (adrenal weight 17.7 g) did not affect the statistical significance. The difference between groups was accounted for in large measure by increases in weight of the suicides" left glands compared to controls (6.2 _ 0.5 g versus 4.6 ___0.3 g;p = 0.0030, t = 3.22, df = 31; see Figure 1). Although fight adrenal weight was slightly greater in suicides than controls, the difference was not statistically significant (5.09 + 0.4 g versus 4.72 _ 0.2 g; p = 0.37, t = 0.91, d f = 31). The mean +_ SEM left and fight adrenal cortical thickness measurements in the suicide group were 117.7 + 10.3 mm mm and 98.6 -+- 3.4 mm in the controls. Although both the left and fight mean adrenal cortical thickness measurements were greater in the suicide than control group, in neither case did the difference reach statistical significance when adjusted for body weight (p = 0.084, t = 1.8 I, d f = 22 for left adrenal; p = 0.26, t = 1.15, df = 26 for fight adrenal). The mean percentage of left adrenal cortex occupied by each of the three zones zG, zF, and zR in the suicides was 8.6, 59.1, and 29.9, respectively, compared to 7.5%, 63.3%, and 29.5% in controls. In the fight adrenal cortex, the zonal distribution was 9.5%, 60.2%, and 30.6% in the suicide group and 8.0%, 62.3%, and 29.9% for controls. The percentage of adrenal cortex occupied by each of the three zones was not significantly different between the two groups for either fight or left glands. To determine whether the differences between groups in

Adrenal Gland in CompletedSuicides

BK)L PSYCmATRY

377

m

E V

.C m

m

to

I.=

"tO

4

I

I

Suicide

I

control Left Adrenal

I

Suicide control Rlght Adrenal

Figure i. Mean (_+ SEM) left and right adrenal weights of suicide victims and controls. total and left adrenal weight were accounted for by the difference in cortical thickness, we adjusted adrenal weights for cortical thickness using ANCOVA. The adjusted mean weights were no longer different, and there was a significant association between adrenal weight and cortical thickness in total (p = 0.0011), left (p = 0.010), and fight (p = 0.015) glands.

Discussion Suicides and controls were well matched in terms of gender composition, weight, and height. Although suicide victims were older than controls, the difference was not statistically significant. Our findings are consistent with previous stud_.__ l e S m a t s n u w c u adrenal w e i g h t w a s n u t ~ z o r t r c t a t ~ u w l u n a g ~ , gender, or height (Dorovini-Zis and Zis 1987; Gelfman 1964). The significant association between adrenal weight and body weight had not been noted in previous studies, and requires that body weight be matched or statistically controlled in comparing measures of adrenal weight and morphology between groups. Although measurements were taken from fixed rather than fresh adrenal tissue, our results were in good agreement with previous studies that found the average weight of the normal human adrenal to be approximately 4 g (Symington 1969; Neville and O'Hare 1982). As well, tissue from subjects in both groups was handled identically. It is, therefore, unlikely that any of these variables accounted for the observed difference in weights between the two groups. In demonstrating that the mean combined weight of adrenals from suicide victims was siguificanfly greater than that l_

I-

~

I-

•

21

l_

of glands from sudden death controls, our results replicate the findings of Dorovini-Zis and Zis (1987). Neither that study nor investigations showing greater computed tomography (CT) scan adrenal volume in depressives compared to controls (Amsterdam et al. 1987b; Nemeroff et al. 1992) reported the sizes of left and right adrenal glands separately, however. We found that the statistically significant difference in our sample was explained by the greater mean left adrenal weight. This curious finding must be considered tentative and requires replication, perhaps with existing data sets. Any explanation would be highly speculative. Although no significant difference in the mean weights of left and right glands has been reported, the left adrenal gland of normal individuals has been documented to be sfightly l~,,I[~t~l

Ill

Vg.JIILll|I~

Ifffl[~lll I.]J.~ I I ~ l l . l l g

ql~l~ll

170iLl,

JL~-tLIILIr.LLIL I~1[1u

Phillips 1991). It is possible, therefore, that the exaggerated difference we observed in suicides was simply due to a proportionately equal amount of hypertrophy in each adrenal, yielding a relatively greater weight increase in the (larger) left gland. Because we found no such difference between left and fight adrenal gland weights in our control sample, this explanation is unlikely. Alternatively, there may be an enhancement (and/or dysfunction) of lateralized central nervous system input to the glands in suicide victims. Although direct evidence for such asymmetry in central glucocorticoid control has not been recorded, Arato and colleagues ( 1991 ) found markedly different, lateralized patterns of frontal cortex imipramine binding between suicides and controls. Although others have failed to replicate this finding (Lawrence et al 1990; Arora and Meltzer 1991), well-documented regulatory interactions between the sero-

378

alOLPSYCHIATRY 1994',36:374~380

tonin and glucocorticoid systems (Fuller 1981; Gibbs and Vale 1983; Meltzer et al 1984) support the potential validity of this hypothesis. We were unable to demonstrate a statistically significant difference between the mean cortical thicknesses of glands in the suicide and control groups, although a trend was evident for the left gland, possibly as a result of the smaller sample suitable for this morphometric analysis. Nonetheless, the significant positive correlation between total cortical thickness and adrenal weight for each gland suggested that the greater mass of adrenals from suicide victims was explained by cortical hyperplasia. Such cortical hypertrophy is most likely secondary to increased pituitary ACTH, as has been demonstrated in rats (Rivier and Vale 1985), dogs (Kemppainen et al 1985), and humans (Symington 1969; Malendowicz and Dembinska 1990). Human peptides have also been implicated in adrenal hypertrophy, supporting the hypothesis that ACTH acts as an indirect mitogen (see Estivariz et al 1992 for review). Whether increased adrenal cortical thickness correlates with in vivo adrenocortical hypersecretion remains uncertain. ACTH administration has been shown to stimulate protein and RNA synthesis in vivo (Farere and ReAdy 1963) and eficit adrenal mitogenesis in some studies (Dallman 1984; Malendewicz 1986) but not in others (Estivariz et al i 982). If cortical hypertrophy results in hypercortisolen~a, then our findings suggest that suicidal individuals may be hypersensitive to stress-induced activation of their HPA axis. There was no significant difference in the proportions of the three cortical zones between the suicide and sudden death controls, which suggests that the adrenal hypertrophy seen in our suicide victims resulted from chronic (months to years) rather than ac~tte (days to weeks) stimulation. Acute stressors, such as sepsis or severe burns, produce focal depletion of lipids from zF cells in humans (Syrnington 1969; Gelfman 1964), with conversion of these lipiddepleted fasciculata cells to compact reticulosis-like cells that actively secrete glucocorticoids (see Symington 1969 for review). Such histological changes were not observed in either group in the present study, as evidenced by absence of zonal redistribution. In contrast, chronic disease states (i.e., hypertension, bronchiogenic carcinoma) have been shown to result in increases in both adrenal cortical cell number and cell size (Gelfman 1964; Sholiton et al 1961), leading to greater weight and cortical thickness without alterations in relative zonal distribution. This type of response is similar

E. Szigethy et al

to our findings. Although indirect evidence argues against the role of the adrenal medulla in the increased adrenal weight (Neville and O'Hare 1982; Landsberg and Young 1985), the synergistic interaction between cortisol and adrenaline during stress is a well-documented phenomenon (Wumnan and Axelrod 1965; Vogt 1944). Direr measurements of medullary volume would settle this issue most directly; rapid postmortem autolysis and subsequent anatomical distortion of the medulla made this determination impossible for us to make. In future studies, much more could be learned about the associations between HPA axis activation, affective illness, and suicide by the simultaneous conduct of detailed clinical assessments of all subjects using postmortem psychological autopsies. In the absence of systematic data concerning subjects' psychiatric histories, we are unable to distinguish whether increased adrenal gland weight is associated with the suicidal state or with affective disorder. Although the greatest number of suicide victims suffer from a depressive disorder, up to 70% may have other psychiatric illnesses (Robins et al 1959; Dorpat and Ripley 1960). If changes in the adrenal cortex are found only in victims with affective illness, findings may be obscurred in studies with a higher proportion of victims with other diagnoses. Furthermore, detailed information about the victims' symptomatology and course of illness would enable one to explore whether HPA axis activation is associated with specific diagnostic subgroups. Knowledge of the life stressors to which a victim was subject, and of the temporal course of their symptomatology in relation to those stressors, would help define the characteristics of normal and abnormal stress responses. Definition of enduring character traits or early life experiences that are associated with adrenal hyperplasia may help identify individuals at risk for dysfunctional responses to stress. In addition, collection of a larger sample may allow comparison across the life course, thus providing oppo.m.jnity to examine age, diagnosis, adrenal weight interactions, a potentially important avenue of research, given our understanding about age-specific differences among those who complete suicide (Dorpat and Ripley 1960; Rich et al 1986).

The authors thank the staffof the Office of the Medical Examiner for their generous support in the conduct of this research, and Marge Roberts for manuscript preparation.

References Amsterdam JD, Winokur A, Abelman E, Lucid I, Rickels K (1983): Cosyna'opin (ACTH oq-2,) stimulation test in depressed patients and healthy subjects. Am J Psychiatry 140:907-909. Amsterdam JD, Maislin G, Winokur A, Kling M, Gold P (1987a): Pituitary and adrenocortical responses to the ovine cortico~o-

pin releasing hormone in depressed patients and healthy volunteers. Arch Gun Psychiatry 44:775-781. Amsterdam JD, Marinelli DL, Arger P, Winokur A (1987b): Assessment of adrenal gland volume by computed tomography in depressed patients and healthy volunteers: A pilot study. Psychiatry Res 21:189-197.

Adrenal Gland in Completed Suicides

Arat6 M, Banki CM, Nememff CB, Bissette G (1986): Hypothalamic-pimitary-adrenal axis and suicide. Ann NY Acad Sci 487:263-270. Afar6 M, Banki CM, Bisseue G, Nememff CB (1989): Elevated CSF CRF in suicide victims. Biol Psychiatry 25:355-359. Arat6 M, Tekes K, T6thfalusi L, et al (1991 ): Reversed hemispbetic asymmetry of imipramine binding in suicide victims. Biol Psychiatry 29:699-702. Atom RC, Meltzer HY (1991): Laterality and 3H-imipramine binding: Studies in the frontal cortex of normal controls and suicide victims. Biol Psychiatry 29:1016-1022. Ayuso-Gutierrez JL, Cabranes JA, Garcia-Camba E, Almoguera I (1987~: Pituitary-adrenal disinlfibition and suicide" *_~empters in depressed patients. Biol Psychiatry 22:1409-1412. Banki CB, Bissette G, Arato M, O'Connor L, Nemeroff CB (1987): CSF corticotropin-releasing factur-like immunoreactivity in depression and schizophrenia. Am J Psychiatry 144:873-877. Banki CM, Arato M (1983): Amine metabolites and neuroendocrine response related to depression and suicide. J Affective Disord 5:223-232. Brown RP, Mason B, Stoll P, et al (1986): Adrenocortical function and suicidal behavior in depressive disorders. Psychiatry Res 17:317-323. Bmmey WE, Fawcett JA, Davis JM, Gifford S (1969,~: Further evaluation of urinary 17-hydroxy-corticosteroid in suicidal patients. Arch Gen Psychiatry 21:138-150. Carpenter WT (1971): Adrenal cortical activity in depressive illness. Am J Psychiatry 128:31-40. Carroll BJ (1982): The dexamethasone suppression test for melancholia. Br J Psychiatry 140:292-304. Carroll BJ, Curtis GC, Mendels J (1976): Neuroendocrine regulation in depression - II. Discrimination of depressed from nondepressed patients. Arch Gen Psychiatry 33:1051-1058. Carroll BJ, Greden JF, Feinberg M (1981): Suicide, neuroendocrine dysfunction and CSF 5-HIAA concentrations in depression. In Angrist B, Burrows G, al Lader ME, ed. Recent Advances in Neuropsychopharmacology. Ehnsford: Pergamon Press, pp 307-313. Conweii Y, Forbes NT, t=ox u, Caine ED (i993): Vaii~tation of a measure of physical illness burden at autopsy: The Cumulative Illness Rating Scale. JAm GeriatrSoc41:38-41. Coryell W, Schlesser MA (1981): Suicide and the dexamethasone suppression test in unipolar depression. Am J Psychiatry 138:1120-1121. Dallman M (1984): Control of adrenocortical growth in vivo. Endocr Res 10:213-242. Dorovini-Zis K, Zis AP (1987): Increased adrenal weight in victims of violent suicide. Am J Psychiatry 144:1214-1215. Dorpat TL, Ripley HS (1960): A study of suicide in the Seattle area. Compr Psychiatry 1:349-359. Estivariz FE, Iterriza FC, Hope J, McLean C, Lowry PJ (1982): Stimulation of adrenal mitogenisis by N-terminal propiocortin peptide. Nature 299:419-422. Estivariz FE, Lowry PJ, Jackson S (1992): Control of adrenal growth. In James VHT (ed), The Adrenal Gland. New York: Raven Press, pp 43-70.

ntOI.PS¥CUlXTRY 1994:36:374--~0

379

Fang VS, Trilou BJ, Pobensou A, Meltzef H (I981): Plasma ACTH and cortisol levels in depressed patients: Relationship to dexamethasone suppression test. L/fe Sci 29:931-938. Farere RV, Reddy WJ (1963): Observations on the int_~actions between adrenal protein RNA and DNA during prolonged ACTH adnf~slration. Biochem Biophys Acta 76:145-I 48. France RD. Urban B, Krishnan KRR, et al (1988): CSF corticotropin-releasing factor i m m ~ v i t y in chronic pain patients with and without major depression. Biol Psychiatry 23:86-88. Fuller RW (1981): Serotonergic stimulation of piraitary-adronocortical function in rats. Neuroendocrinology 32:118-127. Gelfman NA (1964): Morpbologic changes of adrenal cortex in disease. Yale Univ J Biol Med 37:31-54. Geracioti TD, Orth D, Ekhator NN, Blumenkopf B, Loosen PT (1992): Serial cerebrospinal fluid corticotropin-releasing hormone concentrations in healthy and depressed humans. J Clin Endocrinol Metab 74:1325-1330. Gibbons JL (1964): Cortisol secretion rate in depressive illness. Arch Gen Psychiatry 10:572-575. Gibbs DM, Vale W (1983): Effect of the serotonin reuptake inhi'bitot fluoxetine on corticotropin-releasing factor and vasopressin secretion into hypophysial portal blood. Brain Res 280:176179. Gold P, Loriaux D, Roy A, et al (1986): Responses to corticotropin-releasing hormone in the hypercortisolism of depression and Cushing's disease. NEngJMed 1986;314:1329--1335. Greden JF, Albala AA, Haskett RF, et al (198~[ • Normalization of the dexamethasone suppression test: a laboratory index of recovery from endogenous depression. Biol Psychiatry 15:449-458. Holsboer F, yon Bardeleben U, Gerken A, Stalla GK, Muller OA (1985): Blunted corticotripin and normal cortisol resonse to human cordcotropin-releasing factor in depression. N Engl J Med311"1127. Jaeckle RS, Kathol RG, Lopez JF, Meller WH, Krummel SJ (1987): Enhanced adrenal sensitivity to exogenous cosyntropin (ACTHed-~) stimulation in major depression: Relationship to dexamethasone suppression test results. Arch Gen Psychiatry 44:233-240. Kathol RG, JaecHe RS, Lopez IF, Meller WH (1989): Consistent reduction of ACTH responses to stimulation with CRH, vasopressin, and hypoglycemia in patients with major depression. Br J Psychiatry 155:468-478. Kemppainen R, Thompson F, Lorenz M, Brown J (1985): Effects of continuous ct(1-24) ACTH infusion in the dog. Horm Metab Res 17:58-62. Kocsis JH, Kennedy S, Brown RP, Mann JI, Mason B (1986): Suicide and adrenocortical function. Psychopharmacol Bull 22:650--655. Krieger G (1979): The plasma level of cortisol as a predictor of suicide. Dis Nerv Sys 35:237-240. Krislman KRR, Ritchie JL, Saunders WB, Nemeroff CB, Carroll BJ (1990): Adrenocortical sensitivity to low dose ACTH administration in depressed patients. Biol Psychiatry 27:930-933. Landsberg L, Young JB (1985): Catechotamines and the adrenal medulla. In Wilson JD, Foster, DW (eds), Williams Textbook of Endocrinology. Philadelphia: W.B. Satmders, pp 891-965.

380

BIOL~YCHIAT~Y

E. Szigethy et al

1994;36:374-380

Lawrence KM, Depaermender F, Cheetham SC, Crompton MR, Katona CLE, Horton RW (1990): Symmetrical hemispheric distribution of [3H]paroxetlne binding sites in postmortem human brain from controls and suicides. Biol Psychiatry 28:SA.~ 546. Linn BS, Linn MW, Gm'el L (1968): Cumulative illness rating scale. JAm Geriatr Soc 16:622-626. Malenduwicz LK (1986): a-correlated stereological and functional studies on the longterm effects of ACTH on rat adrenal cortes. Folia.Histochem Cytobio124 :203-211. Malendowicz LK, Dembinska M (1990): Proliferation and distribution of adrenocortical cells in ACTH-treated female hamstets. Folia-Histochem Cytobio128:51-59. McCrank E, Stein E, Goyer R (1993): The adrenal gland and depression (letter). Arch Gen Psychiatry 50:832. Meador-Woodruff JH, Haskett RF, Grunhaus L, Akil H, Watson SI, Greden JF (1987): Postdexamethasone plasma cortisol and 13-endorphin levels in depression: Relationship to severity of illness. Biol Psychiatry 22:1137-1150. Meltzer HY, Perline R, Tricou BJ, Lowy M Robenson A (1984): Effect of 5-hydroxytryptophan on serum cortisol levels in major affectivedisorders II. Relation to suicide, psychosis, and depressive symptoms. Arch Gen Psychiatry 41:379-387. Nelson DH (1980). Adrenal Cortex: Physiological Function and Disease (Vol. 18). Philadelphia: WB Saunders Co. Nemernff CB, Owens MJ, Bissette G, Andom AC, Stanley M (1988): Reduced corticotropin releasing factor binding sites in the frontal cortes of suicide vic~L.,~s.Arch Gen Psychiatry 45:577-579. Nemeroff CB, Krishnan RR, Reed D, Leder R, Beam C, Dunnick NR (1992): Adrenal gland enlargement in major depression: A computed tomographic study. Arch Gen Psychiatry 49: 384-387. Neville AM, O'Hare MJ (1982): The Human Adrenal Cortex. Bedin: Springer-Vedag. Ostroff R, Giller E, Bonese K, Ebersole E, Harkness L, Mason J (1982): Neuroendocrine risk factors of suicidal behavior. Am J Psychiatry 139:1323-1325. Rens VI, Joseph MS, Dallman MF (January 28, 1982): ACTH levels after the dexamethasone suppression test in depression [Letter to the editor]. N Engl J Med pp. 238-239. Rich CL, Young D, Fowler RC (1986): San Diego suicide study I. Young vs. old subjects. Arch Gen Psychiatry 43:577-582. Rivier C, Vale W (1985): Effect of the long-term administration of corticotropin releasing factor on the pituitary-adrenal and pituitary-gonadal axis in the male rat. J Clin Invest 75:689694. Robins E, Murphy GE, Wilkinson RH, Gassner S, Kayes J (1959): Some clinical considerations in the prevention of suicide based

on a study of 134 successful suicides. Am J Public Health 49:888-889. Roy A (1992): Hypothalamic-pitm'taryoadrenal axis function and suicidal behavior in depression. Biol P~ychiatry 32:812-816. Roy A, Pickar D, Paul S, Dofan A, Chronsos GP, Gold PW (1987): CSF corticotropin-releasing hormone in depressed patients and normal control subjects.Am/Psychiatry 144:641--645. Rubin RT, Phillips JJ (1991 ): Adrenal gland volume determination by computed tomography and magnetic resonance imaging in normal subjects. Invest Radio126: 465--469. Rubin RT, Poland RE, Lesser IM, Winston RA, Blodgett ALN (1987): Neuroendocrine aspects of primary endogenous depression I. Cortisoi secretory dynamics in patients and matched controls. Arch Gen Psychiatry 44:328-336. Rupprecht R, Lesch K, Muller V, Beck (3, Beckmann H, Schuite HM (1989): Blunted adrenocorticolropin but normal B-endorphin release after human conicotropin-releasing hormone administration in depression. J Clin En~crinol Metab 69: 600-603. Sachar F_J,Hellman L, Roffwarg HP, Halpem FS, Fukushima DK, Gallagher "IF (1973): Disrupted 24 hour patterns of cortisol secretion in psychotic depressives. Arch Gen Psychiatry 28:19-24. Schmidtke A, Fleckenstein P, Beckmann H (1989): The dexamethasone suppression test and suicide attempts. Acta Psychiatr Scand 79:276-282. Sholiton LI, Incze JS, Werk EE (1961): Adrenocorfical width in carcinoma of the lung. Cancer 14:105-110. Shopsin B, Gershon S (1971): Plasma cortisol response to dexamethasone suppression in depressed and control patients. Arch Gen Psychiatry 24:320-326. Symington T (1969): Pathology of the Human Adrenal Gland. Edinburgh: E. S. Livingstone. Traskman L, Tybring (3, Asberg M, Bertilsson L, Lantto O, Schalling D (1980): Cortisol in the CSF of depressed and suicidal patients. Arch Gen Psychiatry 37:761-767. Vogt M (1944): Observations on some conditions affecting the rate of hormone output by the suprarenal cortex. J Physioi 103:317-332. Wurtman RJ, Axelrod J (December, 1965): Adrenaline synthesis: C_onlroi hy th~ pihlit~ry g!~nd o,a aa~e-al 1. . . . . :M:a~ [Letter to the editor]. Science pp 1464-1465. Yerevanian B, Olafsdottir H, Milanese E, et al (1983): Normalization of the dexamethasone suppression test at discharge: Its prognostic value. JAffec6ve Disord, 5:191-197. Zimmerman M, Coryell W, Pfohl B (1986): The validity of the dexamethasone suppression test as a marker for endogenous depression. Arch Gen Psychiatry 43:347-355.