Pituitary-thyroid state correlates with central dopaminergic and serotonergic activity in healthy humans.

Data from lower animals suggest anatomic and physiological interactions between brain dopamine and serotonin (5-hydroxytryptamine, 5-HT) systems and the hypothalamic-pituitary-thyroid axis. However, in humans, investigations of interactions between these central neurochemical systems (especially the dopaminergic system) and thyroid function are rare; in healthy humans they are practically nonexistent. Using cerebrospinal fluid (CSF) and blood samples simultaneously obtained from indwelling subarachnoid and venous catheters in healthy humans, we determined the CSF concentrations of homovanillic acid (HVA) and 5-hydroxyindolacetic acid, the major metabolites of dopamine and 5-HT, and plasma concentrations of TSH, total triiodothyronine (T(3)), free T(3), total thyroxine (T(4)) and free T(4). CSF HVA concentrations were significantly and negatively correlated with plasma TSH and T(3) (free and total), but not with T(4) (free or total). CSF 5-HIAA concentrations were significantly and negatively correlated with plasma TSH and total T(3) but not with free T(3) or T(4) (free or total). These results indicate that CNS monoamine-thyroid interactions are of physiological significance in the normal, euthyroid human.

Intra- and inter-individual relationships between central and peripheral serotonergic activity in humans: a serial cerebrospinal fluid sampling study.

Data are lacking concerning the longitudinal covariability and cross-sectional balance between central and peripheral 5-HIAA concentrations in humans and on the possible associations between tobacco smoking or post-traumatic stress disorder (PTSD) and CSF and plasma 5-HIAA concentrations. Using serial cerebrospinal fluid (CSF) and blood sampling, we determined the concentrations of 5-HIAA in CSF and plasma over 6 h, and examined their relationships in healthy volunteers and patients with PTSD-both smokers and nonsmokers. Patients with PTSD and healthy volunteers had very similar CSF 5-HIAA concentrations. Significant and positive correlations between CSF and plasma 5-HIAA levels were observed within individuals, but this CNS-peripheral 5-HIAA relationship was significantly reduced in smokers (nonsmokers: mean r = 0.559 +/- 0.072; smokers: mean r = 0.329 +/- 0.064 p < 0.038). No significant cross-sectional, interindividual correlation of mean CSF and mean plasma 5-HIAA was seen (r = 0.094). These data show that changes in CSF 5-HIAA levels within an individual over time are largely reflected in plasma 5-HIAA, albeit significantly less so in smokers. The present results therefore suggest that clinically, longitudinal determination of plasma 5-HIAA concentrations within an individual patient can be used to make inferences about relative changes in integrated CSF 5-HIAA concentrations. However, plasma 5-HIAA concentrations provide no significant information about absolute levels of the serotonin metabolite in the CSF.

In-use stability of monoamine metabolites in human cerebrospinal fluid.

Cerebrospinal fluid (CSF) concentrations of the monoamine metabolites homovanillic acid (HVA) and 5-hydroxyindolacetic acid (5-HIAA) are commonly used to provide information about central nervous system (CNS) dopaminergic and serotonergic activity. However, little attention has been given to the effects of sample handling on the concentrations of these compounds in human CSF. Using high-performance liquid chromatography (HPLC) with electrochemical detection, we observed that, in CSF stored at -80 degrees C, concentrations of the serotonin metabolite 5-HIAA and the dopamine metabolite HVA remained unchanged through six 1-h and six 24-h freeze-thaw cycles. Exposure to bright room light (3 h, 1,230 lux) resulted in a 5-HIAA concentration that was 96.3 +/- 2.0% of the initial and an HVA concentration that was 98.8 +/- 1.03% of initial. The pH of the CSF significantly increased during both freeze-thaw series and while maintained on ice (4 degrees C). These results demonstrate the in-use stability of 5-HIAA and HVA in human CSF under commonly-encountered laboratory conditions.

CSF norepinephrine concentrations in posttraumatic stress disorder.

OBJECTIVE: Despite evidence of hyperresponsive peripheral and central nervous system (CNS) noradrenergic activity in posttraumatic stress disorder (PTSD), direct measures of CNS norepinephrine in PTSD have been lacking. The goal of this study was to determine serial CSF norepinephrine levels in patients with PTSD. METHOD: CSF samples were obtained serially over a 6-hour period in 11 male combat veterans with chronic PTSD and eight healthy men through an indwelling subarachnoid catheter. Thus the authors were able to determine hourly CSF norepinephrine concentrations under baseline (unstressed) conditions. Severity of the patients' PTSD symptoms was assessed with the Clinician-Administered PTSD Scale. RESULTS: CSF norepinephrine concentrations were significantly higher in the men with PTSD than in the healthy men. Moreover, CSF norepinephrine levels strongly and positively correlated with the severity of PTSD symptoms. Plasma norepinephrine concentrations showed no significant relationship with the severity of PTSD symptoms. CONCLUSIONS: These findings reveal the presence of greater CNS noradrenergic activity under baseline conditions in patients with chronic PTSD than in healthy subjects and directly link this pathophysiologic observation with the severity of the clinical posttraumatic stress syndrome.

The effect of feeding on cerebrospinal fluid corticotropin-releasing hormone levels in humans.

Corticotropin-releasing hormone (CRH) is a neuropeptide thought to play a role in appetite regulation. In this report, we used a serial cerebrospinal fluid (CSF) sampling technique to examine the relationship between CSF CRH, plasma ACTH and cortisol and perceptions of hunger and satiety in fasting and sated volunteers. CSF was withdrawn continuously from 11:00 AM to 5:00 PM via an indwelling subarachnoid catheter. Blood was withdrawn every 10 min via an antecubital vein catheter. Fed subjects received a meal at 1:00 PM. Subjects who were fed had lower post-prandial ratings on hunger scales and higher ratings on satiety scales. Fed subjects also had slightly lower levels of CSF CRH after feeding. Furthermore, fed subjects had higher ACTH and cortisol concentrations in the first 3 h; by the fourth h the opposite was true. Our findings do not support the hypothesis that CNS CRH is a central satiety factor in the human. Instead our findings of slightly diminished CSF CRH levels after feeding may be accounted for by the rises in glucocorticoids and their associated negative feedback effects on CNS CRH. Alternatively, our findings could also reflect changes in CRH levels associated with feeding in multiple brain areas and in the spinal cord with the net effect being in the negative direction.

Cerebrospinal fluid and plasma testosterone levels in post-traumatic stress disorder and tobacco dependence.

BACKGROUND: Little is known about the relationship between endogenous central nervous system (CNS) testosterone and any psychiatric syndrome. The goal of this study was to screen for potential abnormalities in CNS testosterone levels in patients with post-traumatic stress disorder (PTSD) and/or tobacco dependence. METHODS: We sampled cerebrospinal fluid (CSF) via a subarachnoid catheter over six hours and determined hourly basal CSF concentrations of testosterone in 11 combat veterans with PTSD and 12 normal volunteers. Smokers were abstinent for 11-17 h. Testosterone in CSF and matching plasma samples was assayed by radioimmunoassay. RESULTS: A factor analysis for effects of PTSD status, smoking status and sample time revealed significant effects of PTSD or smoking status, but not time, on CSF testosterone. CSF testosterone levels were lower in individuals with PTSD as compared with normal volunteers. When divided by smoking status, abstinent smokers had mean CSF testosterone levels higher than those of non-smokers. A similar analysis of plasma testosterone revealed no significant effects of any factor on plasma testosterone. CONCLUSIONS: These results indicate that CSF testosterone is significantly influenced by PTSD and smoking status. The exposure of the brain to altered levels of testosterone in smokers and patients with PTSD may have pathophysiologic significance in these conditions.

Plasma and cerebrospinal fluid interleukin-6 concentrations in posttraumatic stress disorder.

BACKGROUND: Interleukin-6 (IL-6) secretion is suppressed by glucocorticoids and stimulated by catecholamines. Patients with posttraumatic stress disorder (PTSD) have decreased cortisol and increased catecholamine secretion. The purpose of this study was to assess the relation of IL-6 levels and hypothalamic-pituitary-adrenal and noradrenergic activity in patients with well-characterized PTSD. METHODS: Cerebrospinal fluid (CSF) was withdrawn via a lumbar subarachnoid catheter over 6 h from 11 combat veterans with PTSD and 8 age- and sex-matched healthy controls. Blood was withdrawn concurrently. We measured IL-6, CRH and norepinephrine concentrations in the CSF and IL-6, ACTH, cortisol and norepinephrine in plasma. RESULTS: Mean and median CSF IL-6 concentrations were higher in PTSD than in controls (mean = 24.0 vs. 14.6, p = 0.05; median = 26.7 vs. 14.3, p < 0.03): plasma IL-6 concentrations, however, were not different between the two groups. Plasma IL-6 and norepinephrine were positively correlated in the PTSD group (r = +0.74, p < 0.04), but not in normals (r = -0.55, p = 0.20). CONCLUSIONS: PTSD patients have increased CSF concentrations of IL-6. Their plasma IL-6 is not elevated but is more tightly associated with noradrenergic output in these patients than in normals. Both findings might be explained by the low cortisol secretion previously reported in PTSD as a result of lowered glucocorticoid suppression of IL-6 secretion. High levels of CSF IL-6 may reflect neurodegeneration or compensatory neuroprotection.

The effect of lumbar puncture stress on dopamine and serotonin metabolites in human cerebrospinal fluid.

In order to examine concentrations of cerebrospinal fluid (CSF) neurochemicals, the technique of lumbar puncture is typically used. However, the effect of the intrinsic stress of undergoing a lumbar puncture on CSF monoamine concentrations in humans has not yet been established. We used lumbar puncture followed 3 h later by continuous CSF sampling to examine the effect of lumbar puncture on levels of the dopamine and serotonin metabolites homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA), respectively. Additionally, we examined the effect of lumbar puncture on the CSF HVA to 5-HIAA ratio. Immediately post lumbar puncture, CSF concentrations of HVA and 5-HIAA were, respectively, only 51 and 54% of the mean levels detected hours later. However, the HVA to 5-HIAA ratio remained stable during lumbar puncture. While HVA and 5-HIAA levels in CSF obtained via lumbar puncture reflect highly variable responses to the stress of the procedure, the ratio of these metabolites is unaffected.

Intra- and inter-individual correlations between cholecystokinin and corticotropin-releasing hormone concentrations in human cerebrospinal fluid.

Despite strong evidence of a physiologic relationship between cholecystokinin (CCK) and corticotropin-releasing hormone (CRH) in the rat central nervous system (CNS), evidence of such a relationship between the two hormones in the human CNS is lacking. A post hoc analysis of serial concentrations of immunoreactive CCK and CRH, obtained every ten minutes from CSF continuously collected over six hours, was performed. A total of 30 subjects were studied: 15 normal volunteers, 10 patients with major depression, and 5 recently-abstinent, alcohol-dependent patients. Overall, we observed an average intra-subject correlation of +.273 (P < 0.001) between CSF CRH and CCK. Inter-subject correlations between mean CSF levels of CRH and CCK were +.948 (P = 0.0001) and +.959 (P = 0.005) in the depressed and abstinent alcoholic patients, respectively. These inter-individual correlations were significantly greater than that seen within the group of normal volunteers (r = +.318, n.s.). The present data suggest that interactions between CCK and CRH are significant in the human CNS, particularly perhaps in depressed and alcoholic patients, and that CSF samples may be used to assess elements of the relationship between these hormones.

Cerebrospinal fluid and plasma leptin measurements: covariability with dopamine and cortisol in fasting humans.

Leptin (OB protein) is an important signal in the regulation of energy balance. Leptin levels correlate with adiposity, but also decrease acutely with caloric restriction and increase with refeeding. The brain is an established critical site of leptin function, yet little is known about leptin concentrations in the central nervous system relative to plasma levels, psychiatric diagnoses, and other endocrine parameters. Therefore, using a novel ultrasensitive leptin assay, we explored relationships of human plasma and cerebrospinal fluid (CSF) leptin levels to body mass index, smoking, posttraumatic stress disorder diagnosis, and levels of dopamine, monoamine metabolites, beta-lipotropin, glucocorticoid, and thyroid and cytokine hormones. A strong linear relation between CSF and plasma leptin levels in the am (r = 0.63; P < 0.002) and afternoon (r = 0.90; P < 0.0001) was revealed. CSF and plasma leptin concentrations decreased during a 12- to 20-h period of fasting. A strong association was found between plasma leptin and CSF dopamine levels (r = 0.74; P < 0.01) as well as between CSF leptin levels and urinary free cortisol (r = 0.73; P < 0.01). Both of these parameters covaried with leptin independently of adiposity, as estimated by body mass index. Implications for leptin transport, regulation, and its potential role in therapeutic strategies for obesity and diabetes are discussed.

Serial CSF corticotropin-releasing hormone levels and adrenocortical activity in combat veterans with posttraumatic stress disorder.

OBJECTIVE: The authors sought to carefully test, by using a technique of continuous CSF sampling, the hypothesis that basal elevations in CSF corticotropin-releasing hormone (CRH) concentrations exist in patients with posttraumatic stress disorder (PTSD). They also sought to assess the relationship among PTSD symptoms, adrenocortical activity, and CSF CRH levels. METHOD: CSF was withdrawn by means of a flexible, indwelling subarachnoid catheter over a 6-hour period, and hourly CSF concentrations of CRH were determined for 11 well-characterized combat veterans with PTSD and 12 matched normal volunteers. Twenty-four-hour urinary-free cortisol excretion was also determined. PTSD and depressive symptoms were correlated with the neuroendocrine data. RESULTS: Mean CSF CRH levels were significantly greater in PTSD patients than in normal subjects (55.2 [SD = 16.4] versus 42.3 pg/ml [SD = 15.6]). No correlation was found between CSF CRH concentrations and PTSD symptoms. While there was no significant difference between groups in 24-hour urinary-free cortisol excretion, the correlation between 24-hour urinary-free cortisol excretion and PTSD symptoms was negative and significant. CONCLUSIONS: By using a serial CSF sampling technique, the authors found high basal CSF CRH concentrations and normal 24-hour urinary-free cortisol excretion in combat veterans with PTSD, a combination that appears to be unique among psychiatric conditions studied to date.

Low CSF concentration of a dopamine metabolite in tobacco smokers.

OBJECTIVE: To screen for dopaminergic abnormalities in tobacco smokers and patients with posttraumatic stress disorder (PTSD), the authors determined serial CSF and plasma concentrations of the dopamine metabolite homovanillic acid (HVA). METHOD: Continuous subarachnoid sampling was used to obtain 37 serial CSF samples over 6 hours in 13 normal volunteers and 11 patients with combat-related PTSD; 10 smoked and 14 had never smoked. The smokers were abstinent from tobacco for 1 1 to 17 hours. RESULTS: The smokers had markedly lower CSF, but not plasma, HVA levels. Their CSF HVA concentrations averaged only 54% of the concentrations of the nonsmokers, independent of diagnosis. CONCLUSIONS: Smokers' low CSF concentrations of HVA may be associated either with chronic inhalation of nicotine or other constituents of tobacco smoke or with acute abstinence. Any possible basal dopaminergic abnormalities in patients with PTSD are small relative to the abnormalities associated with smoking.

Continuous covariability of dopamine and serotonin metabolites in human cerebrospinal fluid.

BACKGROUND: Experiments in lower animals and humans have demonstrated the existence of functional interactions between serotonin and dopamine in neuronal tissue. However, the relationship between parameters of serotonin and dopamine neuronal activity over time within the central nervous system (CNS) of the individual human has not yet been established. METHODS: We used continuous cerebrospinal fluid (CSF) sampling over 6 hours to test the hypothesis that the serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA) and the dopamine metabolite homovanillic acid (HVA) significantly covary in concentration over time. Two groups of normal volunteers (total n = 16) were studied at separate hospitals and CSF was assayed for 5-HIAA and HVA by high-performance liquid chromatography (HPLC). Three subjects underwent repeat CSF-withdrawal procedures after a 6-week interval. RESULTS: Strong and sustained positive covariability in concentrations of HVA and 5-HIAA was observed in the CSF of individual humans. High intraindividual correlation coefficients were +0.897 and +0.871 in the two normal volunteer groups. The HVA to 5-HIAA concentration ratio in CSF was 2.2 +/- 0.7 with very little variability over intervals ranging from minutes to weeks. CONCLUSIONS: The balance between CSF dopamine and serotonin metabolite concentrations remains relatively constant over time in healthy humans. Serial measures of CSF dopamine and serotonin metabolites within the same person could be an effective model in which to explore the interrelationships between these systems in various psychiatric syndromes, in response to drug treatment, and during provactive testing.

Cerebrospinal fluid and plasma beta-endorphin in combat veterans with post-traumatic stress disorder.

Opioid-mediated analgesia develops in experimental animals following traumatic stress and increased opioid-mediated analgesia has been observed in combat veterans with post-traumatic stress disorder (PTSD). These observations have led to the hypothesis that increased central nervous system (CNS) opioidergic activity exists in patients with PTSD. However, direct CNS data on opioid peptide concentrations and dynamics in patients with PTSD are lacking. We withdrew cerebrospinal fluid (CSF) via a flexible, indwelling subarachnoid catheter over a 6-h period and determined hourly CSF concentrations of immunoreactive beta-endorphin (ir beta END) in 10 well-characterized combat veterans with PTSD and nine matched normal volunteers. Blood was simultaneously withdrawn to obtain plasma for ir beta END. PTSD symptom clusters, as measured by the CAPS, were correlated with neuroendocrine data. Mean CSF ir beta END was significantly greater in patients with PTSD compared with normals and there was a negative correlation between the ir beta END and PTSD intrusive and avoidant symptoms of PTSD. No intergroup difference between plasma ir beta END was found, nor was there a significant correlation between CSF and plasma ir beta END. Immunoreactive beta-lipotropin (ir beta LPH) and pro-opiomelanocortin (irPOMC), both precursors of beta END, were much more plentiful in human CSF than was beta-endorphin itself, as has been previously reported. It remains to be determined whether the increased CNS opioid concentrations predate traumatic stress, thereby conferring a vulnerability to dissociative states and PTSD itself, or result from the trauma. The negative correlation between CSF ir beta END and avoidant and intrusive symptoms suggests that CNS hypersecretion of opioids might constitute an adaptive response to traumatic experience. Poor correlation between CSF and plasma ir beta END limits use of plasma measures to assess CNS opioid activity.

Uncoupling of serotonergic and noradrenergic systems in depression: preliminary evidence from continuous cerebrospinal fluid sampling.

We used the technique of continuous cerebrospinal fluid (CSF) sampling to test the following hypotheses regarding CNS monoaminergic systems in depression: (1) absolute concentrations of the informational substances tryptophan and 5-hydroxyindoleacetic acid (5-HIAA) are altered in the CNS of depressed patients (2) abnormal rhythms of tryptophan and/or 5-HIAA, or defective conversion of tryptophan to serotonin (5HT), exist in the CNS of depressed patients, and (3) the relationship between the CNS 5HT and norepinephrine (NE) systems is disrupted in depressed patients. We obtained 6-h concentration time series of tryptophan, 5-HIAA, NE, and 3-methoxy-4-hydroxyphenylglycol (MHPG) in the CSF of 10 patients with major depression and in 10 normal volunteers. No significant differences in CSF tryptophan, 5-HIAA, NE, or MHPG concentrations or rhythms were observed between normal volunteers and depressed patients. Neither were there differences in the mean tryptophan-to-serotonin ratio. However, a negative linear relationship was observed between mean concentrations of 5-HIAA and NE in the CSF of the normal volunteers (r = 0.916 [r2 = 0.839], df = 9, P < 0.001) while, in contrast, depressed patients showed no such relationship (r = +0.094 [r2 = 0.00877], df = 9, n.s.). Furthermore, the correlation coefficients expressing the relationship between CSF MHPG and CSF 5-HIAA within the normal and depressed groups were significantly different. These data support the hypothesis that a disturbance in the interaction between the serotonergic and noradrenergic systems can exist in depressive illness in the absence of any simple 5HT or NE deficit or surplus.

Fasting and postprandial cerebrospinal fluid glucose concentrations in healthy women and in an obese binge eater.

OBJECTIVE: We hypothesized that abnormal entry of glucose into the central nervous system (CNS) might exist in some chronic binge eaters of carbohydrates, as either a cause or consequence of binge eating. The purpose of this study was thus to determine fasting and postprandial glucose concentrations in the cerebrospinal fluid (CSF) of healthy women, and to obtain similar data in an obese, irritable woman with chronic binge eating of postpartum onset. METHOD: CSF was sampled continuously at 0.1 ml/min from 1100 hr to 1700 hr from the binge eating patient, who consumed 5,000 to 10,000 calories per day (preferentially binging on refined carbohydrates), and 4 healthy women via an indwelling, flexible spinal canal catheter. CSF aliquots were obtained at 10-min intervals for measurement of glucose concentrations. Simultaneously, blood was withdrawn at 30-min intervals to obtain serum for glucose assay. A glucose-rich mixed liquid meal was consumed by participants at 1300 hr. RESULTS: In striking contrast to the normal women, our bulimic patient showed no postprandial rise whatever in CSF glucose concentrations. Fasting CSF glucose concentrations were slightly lower whereas fasting serum glucose levels were normal in the bulimic patient, compared with the normal women. After eating, serum glucose levels increased in all participants, but less so in our patient. DISCUSSION: This is the first description of a lack of postprandial elevation in CSF glucose concentration in a patient with a binge eating disorder. Defective transport of glucose across the blood-brain barrier might account for the observed abnormality. While considering other possibilities, we conjecture that our patient's binge eating was an attempt to compensate for impaired postprandial entry of glucose into her CNS.

Concentrations of corticotropin-releasing hormone, norepinephrine, MHPG, 5-hydroxyindoleacetic acid, and tryptophan in the cerebrospinal fluid of alcoholic patients: serial sampling studies.

Abnormalities in corticotropin-releasing hormone (CRH) secretion, noradrenergic neurotransmission, and serotonergic activity in the central nervous system (CNS) have all been hypothesized to exist in alcoholic patients, as have abnormalities in hypothalamic-pituitary adrenal function. To test these hypotheses, we continuously sampled cerebrospinal fluid (CSF) from alcoholic patients after 38-124 days of abstinence and from normal volunteers via a flexible, indwelling lumbar subarachnoid catheter and measured CRH, norepinephrine (NE), 3-methoxy-4-hydroxyphenylglycol (MHPG), tryptophan, and 5-hydroxyindoleacetic acid (5-HIAA) concentrations at 10-min intervals, from 11:00 through 17:00 h. The spinal canal catheter was inserted at approximately 08:00 h. Serial plasma ACTH, cortisol, and NE concentrations were also measured. A mixed liquid meal was consumed at 13:00 h. CSF CRH concentrations were lower in alcoholic patients than in normal volunteers (26 +/- 15 vs. 60 +/- 30 pg/ml, respectively, p < 0.05 by ANOVA), as were CSF NE levels (0.33 +/- 0.09 vs. 1.15 +/- 0.51 pmol/ml, respectively, p < 0.01). Plasma NE and CSF MHPG levels were normal in the alcoholic patients. CSF tryptophan and 5-HIAA and plasma ACTH and cortisol concentrations did not differ between the groups. These studies extend our finding of reduced spinal canal CSF CRH concentrations in depressed patients to abstinent chronic alcoholics.(ABSTRACT TRUNCATED AT 250 WORDS)

Cholecystokinin in human cerebrospinal fluid: concentrations, dynamics, molecular forms and relationship to fasting and feeding in health, depression and alcoholism.

Very little is known about the physiologic significance of the gut-brain hormone cholecystokinin (CCK) in the human central nervous system, although the hormone has been hypothesized to be involved in the regulation of both appetite and anxiety. We continuously collected lumbar cerebrospinal fluid (CSF) via indwelling subarachnoid catheters in ten normal volunteers, ten patients with major depression and five abstinent alcoholic humans, while fasting and after eating. Five other healthy subjects were fasted throughout the experiment. We quantified CSF immunoreactive cholecystokinin (IR-CCK) and glucose concentrations at 10-min intervals from 11.00 to 17.00 h. No difference in CSF IR-CCK concentration, half-life or rhythm was observed between normal volunteers and either depressed or alcoholic patients. Fasting CSF IR-CCK concentrations were 1.3 +/- 0.18, 1.3 +/- 0.21 and 1.2 +/- 0.21 fmol/ml (mean +/- S.E.M.) in normal volunteers, depressed patients and alcoholic patients, respectively. After eating, CSF IR-CCK concentrations rose to 1.5 +/- 0.21, 1.5 +/- 0.24 and 1.4 +/- 0.26 fmol/ml, respectively. Normal volunteers who did not eat had similar basal CSF IR-CCK concentrations (1.1 +/- 0.1 fmol/ml) which similarly rose to 1.4 +/- 0.13 fmol/ml during the sampling interval. In contrast, CSF glucose concentrations rose only in the subjects who ate, beginning to rise after about 1 h and remaining elevated for at least 3 h after eating. These data suggest the existence of a diurnal rhythm of IR-CCK release into CSF, as opposed to a response to feeding. The disappearance half-time of CCK in human CSF is less than 13 min.(ABSTRACT TRUNCATED AT 250 WORDS)

Serial cerebrospinal fluid corticotropin-releasing hormone concentrations in healthy and depressed humans.

CRH, a hypothalamic peptide that is the most potent ACTH secretagogue known, also appears to be produced in the cerebral cortex and spinal cord. Depressed patients have blunted responses to exogenous CRH and normal to high concentrations of CRH immunoreactivity in single morning samples of lumbar cerebrospinal fluid (CSF). Although these data suggest that depression may be associated with hypersecretion of CRH, it has also been postulated that central nervous system insufficiency of CRH might have a pathophysiological role in certain depressive syndromes. We continuously sampled lumbar CSF via indwelling subarachnoid catheters from 1100-1700 h and measured CRH at 10-min intervals in depressed patients and normal subjects. A standardized mixed liquid meal was administered at 1300 h. CSF CRH was strikingly reduced in depressed patients compared to normal subjects [4.2 +/- 1.1 pmol/L vs. 13 +/- 2.1 pmol/L (mean +/- SEM), respectively, P less than 0.01 by Wilcoxon test]. CSF CRH concentrations rose progressively during the experiment in both groups, suggesting a diurnal rhythm and, possibly, response to a test meal. CRH had a very brief half-life in CSF (less than 10 min), suggesting that the spinal cord is the origin of CRH in lumbar CSF. The rapid transients in CSF CRH concentration demonstrate that single samples provide very limited information. There were no intraindividual correlations between CSF CRH concentrations and those of either plasma ACTH or cortisol, both of which rose in response to eating. The present data show that impaired central nervous system secretion of CRH can exist during states of severe depression.