The implicit consensus among most early authors regarding the pathogenesis of depressive disorders in childhood emphasized adverse environmental factors to the detriment of biological factors.1,2 Thus, depression was classified among the neurotic disorders of childhood,1,3 implying that it was the result of unconscious conflict. Psychotic and endogenous forms of the disorder were thought not to exist in children, and proposed treatment was mainly psychotherapy.
A substantial evolution of clinical viewpoints was initiated and maintained by the research strategies based on the phenomenological similarities between child, adolescent, and adult depression.4,5 The reliability of semistructured interviews with children about affective disorder symptoms4,5 (and the feasibility of using adult diagnostic criteria6,7 to identify youngsters with major depression, mania, and even dysthymia8'11) provided a solid base upon which research on the validity of the disorder, diagnosis, psychosocial factors, psychobiology, and treatment could proceed. This article will attempt to summarize the current state of knowledge in the psychobiology of prepubertà! major depression.
Progress in this area has proceeded rapidly, mainly due to the existence of tried and true psychobiological research strategies in adult affective disorders. These have shown that adult patients suffering from endogenous major depressive disorder present a variety of psychobiological abnormalities in sleep EEG112"16 neuroendocrine function,17'26 receptor function,27'29 and biochemical abnormalities30·51 during the depressive episode. Only one marker has been shown to date to be as definitely abnormal in adult depressives during the recovered state.32 Many of the others have not been systematically studied during drug-free sustained affective recovery.
The fundamental similarity in the symptomatology of major affective disorders across age groups in itself posed the obvious question: Do prepubertal children with major depressive disorder show similar biological abnormalities to adults with the same diagnosis.7 Currently, there is more evidence than simply syndtomic similarity to justify this question. Children and adolescents with major depressive disorders come from biological families with high rates of affective illness among their adult relatives. The findings from the studies which have examined the adult relatives of youngsters with major depression33 (Puig-Antich J, Goetz D, Davies M, et al, unpublished data) are consistent with those which examined the offspring of parents with previously identified depressive disorders.54"36 High familial aggregation tends to increase psychiatric illness in the offspring, if both parents have suffered from an affective disease, the morbid risk for their children quadruples, compared to that for offspring from never affectively ill parents. In addition, having only one parent with a history of affective illness only doubles the risk to the offspring. There is also a strong possibility that an inverse relationship exists between age of onset (from childhood on) and affective density of the pedigree (Puig-Antich J, Goetz D, Davies M, et al, unpublished data). The evidence so far indicates that affective disorders with onset in childhood and adolescence tend to aggregate in the same families more than adult depressives.
Familial aggregation does not imply genetic transmission. The latter is one of the possible mechanisms resulting in familial aggtegation. But the clustering of child, adolescent, and adult affective disorders in the same families makes it likely that the fundamental nature of affective illness in all three age groups may be the same. The role of genetics in the familial transmission of affective illness is strongly supported by twin studies of adult pairs. They indicate a higher concordance rate for monozygotic twins compared to dizygotic twins.37 Considering all of this evidence, it can be concluded that genetic mechanisms are likely to play a central role in the transmission of childhood onset affective illness.
The high probability of a major role for genetic transmission in prepubertà! major depression underlines the importance of the study of biological markers in this childhood condition. The biological markers under consideration are different than chromosomal markers. The latter are chromosomal charactetistics which are associated with the transmission of particular disotders or traits on the basis of chromosomal geography alone. DNA polymorphism markers have recently been identified at a very rapid pace, and full genomic maps are expected in the next few years. They are likely to be very helpful, as their number increases in narrowing down the chromosomal region where a gene or a group of genes responsible for the transmission ot a disorder reside. But chromosomal markers in and of themselves do not reveal in the least the pathophysiological mechanisms of an illness until the spécifie gene(s) responsible for the disorder is identified and isolated.
In contrast, so-called biological markers are likely to reflect, more or less directly, limbic mechanisms which mediate the predisposition to, and/or the clinical expression of, depressive disorders. Biological markers are characteristics which have been shown to be specifically associated with the disorder in question, either during an episode OT during the symptom-free intervals or both. A state marker is an abnormality which appears in close temporal relation to the depressive episode and disappears at the episode's end. Finding an abnormality during active illness is not sufficient to determine if it is a state marker or not. In addition, it should normalize upon recovery from the episode. Biological abnormalities which remain or become abnormal during the sustained recovered state may be markers of past episode or true markers of trait. The former would not be abnormal before the onset of the first depressive episode, representing a kind of sequelae from depressive illness. In contrast, true markers of trait would precede the onset of the first episode, and therefore would be very sensitive indicators of the child's predisposition to develop further depressions. At present, the differentiation between trait and past episode markers in affective disorders is a theoretical one, as no studies have been carried out to differentiate them.
As in adult affective disorders, biological abnormalities have been found in children and adolescents with major affective illness. These findings will be reviewed briefly below.
During an episode of major depression, prepubertal children secrete significantly more growth hormone during sleep than normal children and than those with nondepressed emotional disorders.38 Like postmenopausal women with the same disorder,22 endogenously depressed children also secrete significantly less growth hormone in response to insulininduced hypoglycemia than nondepressed children with psychiatric disorders. î9·40 These abnormalities of GH secretion have been shown to persist when measured again during sustained affective recovery, while in a drug-free state for at least 1 month.41,42
On the other hand, cortisol hypersécrétion is found only occasionally when the circadian plasma cortisol patterns of prepubertal children in a major depressive episode are compared to their patterns after affective recovery.43 Furthermore, no differences were found when cortisol secretion in children with major depression was compared to that of nondepressed psychiatric and normal control children. Thus the majority of these children have normal cortisol secretion during and after the depressive episode. These findings are at variance with those among adult endogenous depressives. 19,20 But they are very consistent with the influence of age upon cortisol hypersécrétion in adult depressives, where the older the endogenous patient, the more likely she or he is to hypersecrete cortisol.44 Together with the very low rate of cortisol hypersécrétion among prepubertal depressives, this suggests that age may be as important as major depression in the increase in plasma cortisol in depressive illness in older patients.
In contradistinction to elevated plasma cortisol levels, the data on cortisol nonsuppression to dexamethasone among adult endogenous depressives have not shown significant age influences.45 Unfortunately, the published data on dexamethasone suppression in prepubertal children with major depression are still contradictory. Two studies, both with small outpatient samples, substantially disagree on the rate of nonsuppression. In one controlled study, Research Diagnostic Criteria endogenous major depressive patients showed a sensitivity of 63% and a specificity vis a vis other nondepressed psychiatric disorders of 90%. Dexamethasone dose was 0.5 mg.46 In the other, an uncontrolled study of IO children with the same diagnosis, only one escaped suppression.47 The dexamethasone dose fixed in the second study was weight corrected. In a third study,48 a nonsuppression rate 70% was found in an inpatient sample of 20 children with major depression after 1 mg of dexamethasone. This study was not controlled, which detracts from its conclusions, but, in addition to the standard 4:00 PM sample, samples at 8:00 AM were also obtained. The latter were also suppressed, providing some indirect evidence that the dexamethasone pill was in fact ingested. Finally, two other studies have addressed mostly specificity questions. In a series of 19 prepubertal children with conduct disorder,47 17 were found to suppress cortisol after 1 mg of dexamethasone. But in a smaller study, three of five prepubertal children with separation anxiety disorder, as well as three of five children with major affective illness, were found to escape the suppressant effects of 0.5 mg of dexamethasone.50 The variation in dose is unlikely to have been sufficient to explain the very low sensitivity in one study. The possibility that weight loss may be a substantial factor in inducing lack of suppression cortisol secretion after dexamethasone (with or without affective disorder) has been advanced, and some experimental evidence does exist to support this conclusion. None of these studies controlled for weight loss. Other questions of specificity of dexamethasone nonsuppression to severe major depression also remain unanswered at present, not only among children and adolescents.
In spite of frequent subjective sleep complaints reported by themselves and by their mothers, prepubertal children do not show the expected polysomnographic abnormalities (shortened REM latency, increased REM density, decrease in delta sleep and decreased sleep efficiency) during a major depressive episode.51'52 Although surprising, these findings are consistent with the influence of age on sleep EEC variables in normal adults53·54 and in major depressive adult patients14·53·54 where, controlling for severity, the younger the patient, the less abnormal the sleep EEG is likely to be. As with cortisol secretion, sleep EEG abnormalities during a major depressive episode may be secondary to an interaction between depression and age.51 In such cases, these biological markers may be more apt to reflect rather indirectly the pathophysiological mechanisms of depressive illness at all ages.
In surprising contrast to the lack of findings during the episode, after sustained affective recovery and in a drug-free state, prepubertà! children show significantly shortened REM latency. 55 They also show evidence of slight but significant improvement in sleep continuity measures. The latter may be related to their massive improvement of subjective sleep complaints. It is too early to come to any conclusions at the present time regarding the presence or absence of shortened REM period latency in adult major depressives, as the published studies are mostly small and contradictory. What can be said is that a tendency toward REM advancement remains during the recovered state in adults and that it can be made apparent by the use of cholinergic agonists. Thus, the injection of arecholine in recovered adult depressives shortly after the first REM period is followed by advancement of the following REM period, while this phenomenon is not elicited among never-depressed controls. 32 If this phenomenon is or is not equivalent or related to shortening of the latency to the first REM period is at present an open question. The REM latency findings in recovered major depressive children suggest that short REM period latency may be a true marker of trait for prepubertal major depression.
Other Potential Areas of Psychobiological Research
The markers described above are of interest in affective disorders because their neuroregulation is mediated by transmitter systems which are anatomically and functionally related to the central nervous systems which regulate mood, eating, sexual behavior, sleep, pain, pleasure, and other functions affected in depression. The postsynaptic effects of neurotransmitter release are modulated by the postsynaptic receptors. (Presynaptic receptors also exist and their main function is to provide feedback to the cell to regulate transmitter release. ) Receptors are very specific to their neurotransmitters but each neurotransmitter is likely to have more than one receptor type, each initiating different postsynaptic effects. In the periphery, tissue distributions of the various receptor types for each transmitter are quite different and rather specific. This is likely to be true of different brain regions also. The CNS/periphery parallel regarding receptor function does not end here. The platelets also have membrane receptors to several transmitters which, in animal studies, have been shown to have the same properties as those in the limbic brain. The study of the density of these receptors and of their function allows justified inferences to be made regarding their state in the human brain in conditions like affective disorders. This is especially important in child psychiatry, as these techniques are minimally invasive. Similarly, the systematic study of neurotransmitter metabelites30·31 is likely to add to our knowledge of the psychobiology of affective illness with onset in childhood or adolescence.
Psychobiological marker research on very early onset affective disorders presents an opportunity to separately determine the full range of effects of age and sexual maturity on the biological markers for affective disorders. We have suggested that age-dependent markers are likely to be only peripherally associated with the neuropathophysiological mechanisms intrinsic to depressive illness. If depressive illness exists before puberty, while cortisol hypersécrétion is absent in prepubertal depression, it should follow that mechanisms closely related to cortisol hypersécrétion cannot be equally closely related to the pathogenesis of depressive illness and cannot be essential to the latter. On the other hand, markers with little relationship to age are thought more likely to be intimately associated with pathogenetic mechanisms of depressive illness and should have priority in future research. Furthermore, findings during the recovered state may reflect the neural mechanisms of predisposition to future recurrences, and may even help to individualize risk for affective illness in the still unaffected offspring of pedigrees heavily loaded for affective disorder.
There is strong clinical and animal evidence suggesting that the maturational changes of the CNS during childhood and adolescence are likely ro include differential onset and rates of developmental progression of the various neurotransmitter systems.56'64 Thus, the strong age effects on at least stime biological markers, like the rarity of mania65 and of the euphoric response to dextroamphetamine66 among prepubertal children, may not be as surprising as initially thought. Further research is necessary to explore completely the far-reaching effects of age and puberty on affective disorders and their psychobiology.
1. Cyrryn L, MtKnew DJ: Proposed classifition of childhood depressions. Am J Psychiatry, 1972; 129-149-155.
2. Pronanski EO. Zrull JP: Childhood depression. Arch Gen Psychaitry 1970; 23:8-15.
3. Hersov L: Emotional disorders, in Rutter M. Hersciv L (edit: CWd Piychiany: Modem Approaches. Lundon. Blackwell, Scientific Pubhcalions, I977.
4. Putg-Annch J. Blau S, Mar» M. et al: Prepubertal major depressive disorder: A pilot study. J Am Acad Chid Psycbaoy I978; 17:695-707.
5. WemberR WA. Rutman ). Sullivan L. et al: Depression m children referred m an educational diagnostic center: Diagnosis and treatment. ) ftdurr 1973; 83:1065-1072.
6. Rraiansii EO: Prernihcrtal studies of (he reliabilUY and validity of the Children's Depression Rating Scale. i Am AcoJ Chad Psychiatry, to he published.
7. Chambers, WJ. Puig-Annch J. Hirsch M. et al: Assessment of affective disorders in chilthen and adolescents by semisiructured interview: rest-rerest reliability at the KSAE)S-P. Arc/i Gen Piythuury, to be published.
8. Spinet RL. Endicoii ], Robins E: Research diagnostic criteria: Raiiunale and reliability. Arch Gen Psydaatry 1978; 35:77ì-782.
9. American Psychiatric Association. Committee on Nomenclature and Statistics: DugiuHtK ani Sumsucal Manual, ed 3. Washington DC. American Psychiatric Association Press, 1980.
10. Carlson C, SrroberM; Manic depressive illness in early adolescence. J Am Aiad CfiiiJ Piychiany 197S; 17:158-15).
11. Kiivacs M, Feinberg TL, Cfouse-Novak MA, et al: Depressive disorders in childhood: Characrerurics and recovery - A JongjiudmaJ perspecn ve study. Arch Oía Psychiatry, to be published.
12. Kupfer D, Foster FG: EEG sleep and depression, m Williams RL. Karacan I (eds); Sleep Disorders Dwpmsis and Treatment. New York, ] Wiley. 1979, pp 163-203.
13. VuKeI GW, Vogel F. McAbee RS, et »l Improvement of depression by REM sleep deprivation. AicKGen Psychiatry 1980; 37:147-253.
14. Coble P, Kupfer DJ, Spiker DG, et al: EEG sleep and clinical cha rae ten sties in young primary depressives. Sleep WoO; 9:165.
15. Gillin C, Duncan W. Pertigrew KD, et ai: Successful separation of depressed, normal and insomniac subjects by EEG sleep data. Arch Gen Psychiatry 1979; 16:85-90.
16. Kupfer D: REM latency: A psychobiological markerlw primary depressive disease. Biof Psychiatry 1976; 11:159-174.
17. Carnill BJ. Curtís GC, Mendels J: Neuroendncnne reflation in depression: l· Limbic system -adrenocortisol dysfunctions. Arch Gen Psychiatry 1976; 3Ì:IOÌ9-1044.
18. Cartoli B), Cuttis GC, Mendels ]: Neuroendocrine regulación in depression: II: Discrimination of depressed from «undepressed patients. Ardi Gen Psychiatry 1976; 33:1051-1058.
19. Sachar EJ: Neuroendocrine abnormalities m depressive illness, in Sachar EJ (ed): Topici in Psychoneucroendocrinology, New York, Grune & Stratton, 1975, pp 135-156.
20. Sachar EJ. Hellman L. rU.ffwarg HP, et al: Disrupted 24 hour pattern of cortisol secretion in psychotic depression. Arch Gen Psychiatry 1973; 28:19-25.
21. Grégoire F, Bíanman G. DeBuck R, et al: Hormone release m depressed patients before and after recovery, Psychoneucroendocrinology 1977; 2:303-312.
22. Gruen PH, Sachar EJ. Altman N, el al; Growth hormone responses to hypoglycémie in postnienopausal depressed wranen. Arc/I Giri Psychiatry 1975: 32:Sl-33.
23. Laakman G: Neuroendocrine differences between endogenous and neurotic depression as «en in snmulaiion of griiwth hiitrrmnc secretion, in Millet BE, AHIW(I A (eds). Neumendocrme Carrelates in Neurology and Psychiatry Amsterdam. Elsevier, 1979, pp 263-271.
24. Prange AJ: Patterns of pituitary responses to TRH in depressed patients, in Faun W. Karacan 1. l\Jtomy AD, et al (eds): Pheninnenutugt and Trearmcnc iif De/iresjiim. New York. Spectrum Publishers, 1977. pp 1-16.
25. Takahashi S. Kondn H. Yoshimura M, et al: Thyrotropin response to TRH in depressive illness, folia Psychiatry Neural Jpn 1974; 28:355-565.
26. Sachar EJ, Halbreich U, Asnis GM, et al: Paradoxical Cortisol response io de«moaphetamine m endogenous, depression. Arch Gen Psychiatry 1981-, 58:1113-117.
27. Meltoer HY, Atora RC, Baber R, er al: Serotonin uptake in blood platelets of psychiatric patients. Arch Gen Psydnairy 1981; 38:1322-1326.
28. Garcia-Sevilla JA, Zis AP, Hollingsworth PJ, el al: Platelet alpha- adrenergic recep. tors in major depressive disorder. Arch Gem Psychiatry 1981; 38:1327-1333.
29. PJU! SM, Renavi M, Stolnick JC, et al: CJepressed patients have decreased binding of titrated imipramine to platelet serotiinin transputter. Arch Gen Psychiatry 1981; 38:1315-1317
30. Sabelli HC. Fawcer J. Gusovsky F, et al: Urinary phenyl acetate: A diagnostic test fot depression! Science Science, 210: 1187-1188.
31. SchildkrautJJ, Orsulak PJ. Schatiberg AF, crai: Toward a biochemical classification of depressive disorders: I. Differences in urinary excretion of MHPG and other tatecholamine metabolites in clinically defined subtypes of depressions. Arch Gen PsvcnuUrv 1978; 35:1427-1433.
32. Sitarem M, Nürnberger JI, GershonES, et al: Faster e boli netgic REM sleep induct ion in euthytnic patients with primary affective illness. Science 1980; 208:200-201.
33. SttoberM, Burroughs J, Salic in B, et al: Ancestral secondary cases of psychiatrie illness in adolescents with mania, depression, sthiaiph renia and conduct disorder. BiiiJ Psychiatry, lo be published.
24. Weher Z. TOîlner A. McCray MD, et al: Psychoparhology in children of inpatients with depression: A contmlled study. J New Meni Dis 1977; 164:408-413.
35. Beardslee W, Keller MD. Lavori P, et al: Children of parents with affective disorder: The pathogenic influence of illness in both parents. J Am Acad Child Psychiatry, to be published.
36. Weissman MM. Leckman JF, Menkangas KR. et al: Depression and anxiety disorders in patients and children. Arch Gen Psychiatry 1984; 41:845-855.
37. GershonES, Bunney EW, LeckmanJF. et al: The inheritance of affective disorders; A review of data and of hypothesis, Bev Genet 1976; 6:227-261.
38. Puig-Antich J, Oneti R, Oavies M, et al: Growth hormone secretion in prepubertà! major depressive children. II; Sleep related plasma concentrations during a depressive episode. Arch Gen Psychiatry, to be published.
39. Puig-Antich, Tabrizi MA, DaviesM. eta!-. Piepubena! endogenous major depressives hyposecrete growth hormone in response to insu/in induced hypoglycernia. Bio Psychuary 1981; 16:801-818.
40. Puig-Antich J, NovacenkoH, Davies M, et al: Growth hormone secretion in prepuherral major depressive children. 1: Response to insulin induced hypoglycetnia. Final Repiirr. Arch Gen Psychiatry, to he published.
41. Puig-AntichJ. DavicsM, HalpemFS, et al: Growth hoimone secretion in prepuherial major depressive children. [11: Response to insulin induced hypoelycemia in a drug' free fully recovered clinical state. Arch Gen Psychiatry, to be published.
42. Puig-Antich J. Gotti R, Gavies M. el al: Growth hormone secretion in prepubertà! major depressive children. IV; Sleep related plasma concentrations in a drug-free. fully recovered clinical state. Arrfi Gen Psychiatry, to be published.
43. Puig-Antich J, Novacenko H. Goeti R. et al: Cortisol and prolactm responses to insulin induced hypoglycemia in prepubertal major depressives during episode and after recovery. 3 Am AcaJ Child Psychiatry. to be published.
44. Asnis GM. Sachar EJ, HalhreithU. etal:Cortisolsecretion in relation to age in major depression. Psvcfioium Med 1981; 43:235-242.
45. Carroll BJ. Feinberg M, Greden JF, et al: A specific laboratory test for the diagnosis ut melancholia: Standardization, validation, and clinical utility- Arch Gin Psychiatry 1981; 38:15-22.
46. Poznananski EO. Carroll BJ, Bamegas MC, et al: The dex ame thasiine suppression test m prepubertal depressed children. Am J Psychiatry 1982; 139:321-124.
47. Deller B. Berci JM, Knitter EK et a!; NiHtripryljne in ma;iir depressive disi'rdi't in children: Respume, steady state plasma levels, prediclive kinetics and phatmaci'kinetits. Psychapharmaciit RuJI 19B3; 19:62-65.
48. Waller EB, Weller RA, Fnstad MA. et al: The dexamethasone suppression test in hospitalued prepubertal depressed children. Am J Psichiatri !984: 141:190-291.
49. Targum S. Chasrek C. Sullivan A: Dexamethasone suppression test in prepuberttal conduct disorder. Psychiatry R« 1981; "5.107-108.
50. Livingstiin R. Reis CJ, RinRjahllC: Abnormal Jexamechasonesuppression testresiiJts in depressed and nondepressed children. Am J Psychiatry 1984; 141:106-108.
51 . PuiE-Antich J, Giietz R, Hanlon C, et ai: Sleep architecture and REM sleep measures in prepuhertal major depressives durine ?? episode. Arch Gen Ps-fchiatrv 1982; 39:9J2-939.
52. Young W, Knowles JB, McLean AW. et al: The sleep ol childhood depressives: Comparison with age marched commis. Biol Psychiatry 1982; 17:1163-1168.
53. OillmJC, Duncan WC, Murphy DL. et al: Ace related changes in sleep m depressed and normal subjects. Psychiatry Res 1981; 4:73-78.
54. Ulrich R. Shaw DH, Kupfer DJ: The effects of agine on sleep. Sleep 1980; i:31-40.
55. Puig- Antich J, GoetzR, Hanlon C, et al: Sleep architecture and REM sleep measures in prepuhertal major depressives: Studies during recovery from a major depressive episode m a drug-free state. Arch Gen Psychiatry 1983; 49:187-192.
56. Lidor HG, Mollivcr ME: An immunohistochemical study of setotonin neuron development in the rat; Ascending pathways and terminal fields. Brom Res Bull 1982; 8:389-430.
57. Giildman-Rakic PS. Brown RM: R»t-naial development of monoamine coment and synthesis in the cerebral cortex of rhesus monkeys. Brain Res 1982: 256:339-349.
58. Dawuta A, Gandm-Chaial G, Faudon M, et al: Endogenous levels of tryptophane, seroroniri. and 5-H1AA in the developing brain of the cat. Neifrosd Leu 1979: pp 187-192.
59. Baker PC, Goodrich CA: The effects of lhe specihc uptake inhibitor citalopram upon brain mdoleamine stores in the maturing mouse. Gen Phormoíiil 1982; 13:59-6!.
60. Shelton DL, Nadlet JV, Cotman CW: Development of high affinity choline uptake and assiKiated acetylcholine synthesis in the rat fuscia dentata. Bruin Res 1979; 163:261-275.
61. Ramsay PB. Kngman MR, Mure!) P: Development studies of the uptake of chnline, GABA snd dopamme by crude synaptosomal preparations aftet in vivo or invitto lead treatment. Brom Res 1980; 187:383-402.
62. Hcdner T. Lundberg P: Neurtichemical characteristics of cerebral catecholamme neurons during post-natal development in the rat. Med Bull 1981; 59:212-22 Ì.
63. Lengvan I. Brauch BJ, Taylor AN: Effects of prenatal thyroxme and/or corticostetone treatment irf the ontoKcnesisofhypochalamic and mesencephahc notepinephtme and iiipamint concent. DevNeumsa i980.
64. Yuwiler A, Brammei GL: Neonatal hormone treatment and maturation of the pineal noradtenetgic system: Hydtocortisone and thyroxme. J Neurocht-m 1981; 17:985-992.
65. Puig-Antich J: Affective disotders in childhood; A review and perspective. Psïchuur Clin Norih Am 1900; î:40i-424.
66. Rapiipori JL, Buchsbaum MS, Weingariner H, et al: Dexnoamphetamine: Its cognitive and behavioral effects in normal and hyperactive boys and normal men. Arch Gen Psychiatry 1980; 17:933-943.