Dextroamphetamine and cortisol in depression. Morning plasma cortisol levels suppressed.

Dextroamphetamine sulfate administered intravenously in the morning to 11 unmedicated depressed patients suppressed previously elevated plasma cortisol levels to normal in 90 minutes, a fall of 33% from baseline. Ten other depressed patients, without amphetamine, maintained high cortisol levels during the same time period. In each of five normal young men, amphetamine identically administered stimulated a rise in cortisol between 15 and 30 minutes after infusion, an acute response absent in ten of the 11 depressed patients; by 90 minutes after amphetamine administration, plasma cortisol had fallen to normal and identical levels in both groups. Since noradrenalin normally inhibits hypothalamic corticotropin releasing factor (and adrenocorticotropic hormone) secretion, a noradrenergic deficit may account for cortisol hypersecretion in depression; amphetamine may transiently "correct" this deficit in depressed patients, thereby reducing their cortisol secretion.

Progress in pharmacotherapy of borderline disorders. A double-blind study of amitriptyline, haloperidol, and placebo.

In symptomatic patients with borderline disorder, we conducted a double-blind, placebo-controlled trial of haloperidol and amitriptyline hydrochloride to test the differential efficacy of medication against the affective and schizotypal symptoms that characterize the disorder. Sixty-one patients, diagnosed by the Diagnostic Interview for Borderline of Gunderson et al, completed randomized trials of haloperidol (n = 21), amitriptyline (n = 20), and placebo (n = 20). Medications were given in dose ranges of 4 to 16 mg for haloperidol (mean, 7.24 mg) and 100 to 175 mg for amitriptyline hydrochloride (mean, 147.62 mg) for five-week periods, with weekly self-rated and observer-rated measures of mood, schizotypal symptoms, and global functioning. Haloperidol was superior to both amitriptyline and placebo on a composite measure of overall symptom severity, with no difference between amitriptyline and placebo. Haloperidol produced significant improvement on a broad spectrum of symptom patterns, including depression, anxiety, hostility, paranoid ideation, and psychoticism. In contrast, amitriptyline was minimally effective, with small gains limited to some areas of depressive content. The magnitude of change tended to be modest and was more apparent in self-rated than observer-rated measures.

Cortisol secretion in endogenous depression. I. Basal plasma levels.

Plasma levels of cortisol were sampled for 24 hours in 32 endogenously depressed (ED) patients and 72 normal controls who also underwent the dexamethasone suppression test. The ED patients had significantly higher mean 24-hour plasma levels of cortisol (means 24h PC). However, means 24h PC values of subjects in both groups were normally distributed, with a marked overlap between the two. Only seven ED patients had means 24h PC values higher than 2 SDs from the normal mean (greater than 10 micrograms/dL). An abnormal dexamethasone suppression test result was only partially related to basal cortisol levels. The mean plasma level of cortisol between 1 and 4 PM was found to be highly correlated with the means 24h PC value in ED patients, as has been previously reported in normal subjects and patients with various other diseases (in which it also powerfully discriminated between hypersecretors and normosecretors). This finding supports the use of mean cortisol levels between 1 and 4 PM as a reliable and convenient indication of cortisol secretion.

Cortisol secretion in endogenous depression. II. Time-related functions.

Plasma levels of cortisol were sampled for 24 hours in 32 endogenously depressed (ED) patients and 72 controls to examine mean 24-hour plasma levels of cortisol, intervention in the feedback mechanism of the hypothalamic-pituitary-adrenal system (the dexamethasone suppression test), the circadian rhythm of cortisol secretion and its magnitude, and the ultradian rhythm of cortisol secretion. The main difference in the pattern of cortisol secretion in ED patients, as compared with controls, was in the ultradian rhythm. No acrophase or nadir advance of cortisol secretion in endogenous depression was found when age was controlled, but there was an earlier timing of first secretory episode of cortisol (during night). Only some ED patients have abnormalities in each of the functions studied, and they only partially overlap each other. The results suggest that abnormal cortisol secretion in depression should not be viewed as a monolithic malfunction characteristic of endogenous depression.

Paradoxical cortisol responses to dextroamphetamine in endogenous depression.

Dextroamphetamine hydrochloride was administered intravenously (IV) in the morning and evening to 22 unmedicated patients with severe endogenous depressions and 18 normal control subjects. While the normal subjects generally had a sharp increase in plasma cortisol level by 30 minutes after drug administration, two thirds of the depressed patients showed instead a paradoxical suppression of cortisol levels by 60 minutes. Discrimination between normal subjects and depressives was greatest in the evening. These results are consistent with other reports of abnormal cortisol responses in depressed patients to smaller IV doses of dextroamphetamine and larger doses of methamphetamine hydrochloride. A defect in activation or noradrenergic alpha receptors may account, in part, for the abnormal cortisol responses. The dextroamphetamine cortisol test in other patient populations requires study before its diagnostic use in endogenous depression can be established.

Acute antidepressant effect following pulse loading with intravenous and oral clomipramine.

A double-blind, randomized trial of oral vs intravenous clomipramine hydrochloride pulse-loading dosing regimens was conducted. After a two-week drug-free assessment period, 22 inpatients with a diagnosis of major depressive disorder were given either an evening infusion of 150 mg of clomipramine hydrochloride and placebo tablets or 150 mg of oral clomipramine hydrochloride and an isotonic saline infusion. Twenty-four hours later, this procedure was repeated using a dose of 200 mg of clomipramine hydrochloride. Patients received no further medication over the next five days. The mean Hamilton Depression Rating Scale score for all patients, five days after pulse loading, had dropped by 35% (range, 13.3% to -82.4%). This improvement was significant, as was the amelioration in the Raskin Severity for Depression Scale and the Beck Depression Inventory scores. Although the bioavailability of parenteral clomipramine was greater, there were no significant differences in either efficacy or side effects between the two groups. Pronounced early improvements in severe depressive symptoms may be achieved via loading dose regimens with clomipramine in the absence of continuous treatment.

Growth hormone response to dextroamphetamine in depressed patients and normal subjects.

The human growth hormone (HGH) response to dextroamphetamine sulfate (doses, 0.1 and 0.15 mg/kg) was determined in both the morning and evening in patients with endogenous and atypical depression and in normal young men and normal postmenopausal women. Although the HGH response was found to be reduced in endogenously depressed postmenopausal women, it was equally reduced in normal postmenopausal women and in patients with atypical depression. Depressed and normal men had larger HGH responses, but there were no differences between depressed and normal men. These results do not confirm an earlier report that the reduced HGH response to dextroamphetamine is specific to endogenous depression. The results do suggest the importance to control for other variables in studies of HGH responses in psychiatric patients.

Diurnal variation in the response of plasma prolactin, cortisol, and growth hormone to insulin-induced hypoglycemia in normal men.

Plasma PRL, cortisol, and GH responses to a standard iv dose of regular insulin were studied during the morning and evening in seven normal young men. Hypoglycemia achieved during morning and evening in the same aubjects was equal. There was a substantially greater maximal increment in PRL in the evening compared to the morning (P less than 0.01). The peak levels of cortisol achieved in the morning and evening were equal, but the evening maximal increase was greater (P less than 0.05) because of the significantly lower evening basal cortisol level. Evening increases in GH were greater than in the morning in five subjects and were essentially the same in two subjects; for the group, the evening maximal increment in GH was significantly greater (p less than 0.05 after log transformation). Since serotonergic mechanisms appear to be involved in the PRL, GH, and cortisol responses to hypoglycemia, we suggest the possibility of a diurnal variation in hypothalamic serotonin activity which may partly mediate these differential diurnal hormonal responses to hypoglycemia (although other neurotransmitters may also be involved). The data on cortisol are discussed with regard to the reset hypothesis of feedback inhibition.

Effect of cyproheptadine and atropine on the diurnal prolactin responses to insulin-induced hypoglycemia in normal men.

There is a diurnal variation in the PRL response to insulin-induced hypoglycemia in normal men, with a substantially larger response in the evening. This study investigated the possible serotonergic and cholinergic influences on these diurnal PRL responses. Morning and evening PRL responses to hypoglycemia were compared in the same five normal young men without drug pretreatment, after cyprohepatadine, and after atropine sulfate. Cyproheptadine had no effect on the basal PRL concentration or the PRL response to hypoglycemia in either the morning or evening. Atropine had no effect on basal PRL concentrations in the morning or evening or on the morning PRL response to hypoglycemia. However, the evening PRL response was significantly inhibited by atropine (P less than 0.02), with an abolition of the normal diurnal difference in response (P less than 0.001) to insulin-induced hypoglycemia. These data suggest the involvement of a cholinergic mechanism in regulation of diurnal PRL responses to hypoglycemia.

Failure of dopaminergic blockade to affect prolactin, growth hormone, and cortisol responses to insulin-induced hypoglycemia in schizophrenics.

The PRL, GH, and cortisol responses to insulin tolerance tests (ITTs) were evaluated in 12 medically healthy schizophrenic patients during a drug-free period and after 1 and 6 weeks of treatment with penfluridol, a potent, long acting, dopamine-blocking neuroleptic. Hypoglycemic responses were the same before and during penfluridol therapy. Although resting PRL levels were evaluated during initial penfluridol therapy (week 1), hypoglycemia provoked a further substantial PRL increment, not significantly different in magnitude from that induced by hypoglycemia during the drug-free period. However, there was a 54% reduction (P less than 0.05) in the increase in the area under the PRL curve during week 6 compared to the drug-free period. Regarding GH and cortisol, resting levels, areas under the curve, and maximal increments after ITT were essentially the same during weeks 1 and 6 of penfluridol treatment as in the drug-free period. The failure of 1 week of dopaminergic blockade to significantly alter the hormonal (PRL, GH, and cortisol) responses to ITT in the group as a whole suggests that dopamine-blocking mechanisms play little role in mediating these responses, at least in schizophrenic patients.

Fluvoxamine versus desipramine: comparative polysomnographic effects.

Electroencephalogram sleep measures over a 4-week period were obtained on 35 inpatients with major depression (unipolar) who received either fluvoxamine or desipramine in a randomized double-blind trial. Fluvoxamine showed immediate rapid eye movement (REM) sleep suppression and an alerting effect on sleep continuity measures. In contrast, desipramine administration was associated with REM suppression and sleep continuity improvement. The "alerting" quality of fluvoxamine, similar to other serotonergic antidepressants, appears to be unrelated to a lack of clinical efficacy, but may be related to persistent REM sleep suppression. However, it is premature to comment on the serotonin specificity for REM sleep.

Paradoxical effects of amitriptyline on borderline patients.

A paradoxical increase in suicide threats, paranoid ideation, and demanding and assaultive behavior occurred among 15 borderline inpatients receiving amitriptyline in a double-blind study. This pattern differed significantly from that of 14 nonresponding patients receiving placebo.

Cortisol secretion and dexamethasone response in depression.

The authors administered 2 mg of dexamethasone at 11:00 p.m. to 37 unmedicated hospitalized endogenously depressed patients and assessed their plasma cortisol response at 4:00 and 11:00 p.m. the next day. In addition, on nondexamethasone days 26 of these patients had mean 24-hour plasma cortisol concentration determinations from samples taken at 30-min intervals and 32 had plasma determinations from a single sample taken at 4:00 and 11:00 p.m. Mean 24-hour plasma cortisol concentration was elevated in 50%; only 7 of the 26 were dexamethasone resistant, and 6 of these 7 were hypersecretors. The authors suggest that dexamethasone resistance reflects the abnormality of cortisol hypersecretion in depression and that the 2-mg dexamethasone suppression test is a highly specific but not very sensitive indicator of hypersecretion.

Plasma cortisol secretion and REM period latency in adult endogenous depression.

The authors studied the relationship of plasma cortisol secretion and REM period latency in 25 patients with endogenous depression. The 8 patients (32%) with cortisol hypersecretion had a significantly shortened REM period latency in comparison with the 17 with normal cortisol secretion. Furthermore, an extremely short REM latency (20 minutes or less) occurred almost exclusively in those with cortisol hypersecretion. The authors discuss possible neurotransmitter disturbances responsible for these abnormalities and the clinical implications of these findings.

Effect of atropine on the diurnal PRL responses to TRH in normal subjects.

Prolactin (PRL) responses to TRH were investigated at 0900 and 1800 hr in six young healthy men under basal conditions and after atropine administration. TRH induced significantly higher PRL release during the evening administration compared to the morning. Atropine had no effect on basal PRL secretion. Unlike the reported influences of anticholinergic drugs on the diurnal PRL responses to other stimuli, atropine also had no effect on TRH-stimulated PRL release either in the morning or in the evening. These observations in part may be accounted for by the direct PRL-releasing effect of TRH (without neurotransmitter involvement).

Effect of benztropine on the diurnal prolactin responses to haloperidol.

Prolactin (PRL) responses to haloperidol were investigated at 0900 and 1800 h in six young healthy men under basal conditions and after benztropine mesylate administration. Haloperidol administration induced significantly higher PRL release during the evening compared to the morning. The anticholinergic drug, benztropine, potentiated the PRL responses to haloperidol both in the morning and in the evening. The possible mechanisms of these findings are discussed.

Diurnal hormonal responses to thyrotropin-releasing hormone in normal men.

Bolus injections of synthetic thyrotropin-releasing hormone (TRH) were administered to five young normal men in the morning (0900 hr) and the evening (1800 hr) on different days. Frequent blood samples for prolactin (PRL) and TSH analyses were collected before and after TRH infusion. Although there were no differences between the morning and the evening basal PRL levels, a significantly greater PRL response to TRH in the evening was observed (delta PRL, a.m. vs p.m., p less than 0.025). Since TRH stimulates PRL through a direct effect on the pituitary, our data suggest that there is a diurnal variation in the pituitary lactotroph responsiveness to TRH. On the other hand, a.m. and p.m. basal and TRH-stimulated TSH responses were virtually identical.

Prolactin responses to haloperidol in normal young women.

The prolactin (PRL) responses to intramuscular haloperidol (HPD) (0.5, 1.0, and 1.5 mg) were evaluated in six normal premenopausal women during the follicular and luteal phases of their menstrual cycles. These were compared to the PRL responses to these doses of HPD in normal young men. PRL responses to HPD did not differ between the follicular and luteal phases. The mean log-transformed PRL response to the lowest HPD dose (0.5 mg) in women was less than that in the men, but the women had greater PRL responses than the men to the higher haloperidol doses (1.0 mg and 1.5 mg).

Characterizing depression in borderline patients.

The comorbidity of depression and borderline disorder was studied in 39 symptomatic borderline inpatients defined by the Diagnostic Interview for Borderlines using three independent methods for assessing depression and three definitions of depression. Evaluations were conducted by the Schedule for Affective Disorders and Schizophrenia interviews for Research Diagnostic Criteria (RDC) depressive disorders, by clinical ratings for atypical depressive disorder, and by self-rated questionnaires for hysteroid dysphoria. Diagnoses of an RDC depression were made in 25 (64.1%), atypical depressive disorder in 16 (41%), and hysteroid dysphoria in 25 (64.1%) of the borderline patients. Two depressive diagnoses were present in 64.1% of patients, while 17.9% of patients met criteria for all three depressive disorders. No one method accurately characterized depression in borderline patients.

The role of neuropharmacologic selectivity in antidepressant action: fluvoxamine versus desipramine.

Forty patients with a diagnosis of major depressive disorder were entered in a double-blind study to assess comparative clinical response and pharmacologic parameters of fluvoxamine, a highly selective blocker of serotonin reuptake, and desipramine, a noradrenergic agent. Eighteen patients receiving desipramine and 17 patients receiving fluvoxamine completed the study. Fluvoxamine was comparable to desipramine in its antidepressant efficacy and was better tolerated and caused minimal side effects. There was a direct linear relationship between plasma fluvoxamine levels and clinical response and a nonlinear relationship between plasma desipramine levels and clinical response. The pharmacologic specificity of the two drugs was assessed by determining uptake inhibition of serotonin and norepinephrine. The authors found a positive relationship between Hamilton Rating Scale for Depression scores and norepinephrine uptake inhibition for desipramine but found no such relationship between fluvoxamine and serotonin uptake inhibition. Although there was a clear-cut difference in the quality of pharmacologic specificity and a partial relationship to clinical response, the authors were unable to identify neuropharmacologic factors that would predict either treatment response or selective amelioration of symptomatologies in this patient population.