Prolactin and growth hormone responses to psychological stress in normal and neurotic subjects

Pool vs single sample determination of serum prolactin to explore venipuncture associated stress induced variation

Stress is identified as a cause of transient hyperprolactinemia, whereas venipuncture is considered a source of stress for patient. The aim of this study was to investigate the association of venipuncture-induced stress with elevation of serum prolactin. This was a cross-sectional observational study conducted on a group of 150 outdoor patients visiting a tertiary care hospital. Serial sampling was performed by drawing venous blood at different time intervals (0, 30 and 60 min) by single venipuncture to measure serum prolactin to diagnose stress-induced hyperprolactinemia. The study was conducted in two phases, namely, Phase 1 and Phase 2, at different times. The Phase 1 results were divided into two groups: Group 1 (0 min) and Group 5 (pool prepared from samples collected at 0 + 30 + 60 min). Likewise, the results of Phase 2 were segregated into five groups; Group 1 (0 min), Group 2 (30 min), Group 3 (60 min), Group 4 (average of three groups), and Group 5 (pool from samples collected at 0 + 30 + 60 min). In both Phase 1 and Phase 2 of the study, there was a statistically significant (p = 0.0003 in Phase 1 and p = 0.02 in Phase 2) decrease in the mean prolactin (17.99 ± 24.76 ng/mL in Phase 1 and 19.61 ± 23.42 ng/mL in Phase 2) in the pooled samples (Group 5) in comparison to the mean prolactin (19.67 ± 27.69 ng/mL in Phase 1 and 21.06 ± 25.06 ng/mL in Phase 2) of the serum collected at 0 h (Group 1). There was no significant difference in the mean prolactin measured from the pooled samples and average prolactin calculated after individual testing from each sample collected at 0 h, 30 min and 60 min. Venipuncture-triggered fear and apprehension may result in transient hyperprolactinemia. In comparison to performing multiple testing on the samples collected at different time intervals and determining the mean, measurement of the analyte from the pooled serum is the better alternative as it can conserve both time and resources.

The use of intravenous catheterisation with a rest period is useful for determination of plasma cortisol levels but not plasma prolactin levels

OBJECTIVE

The use of an intravenous catheter with a rest period has been recommended to avoid false-positive results for hyperprolactinaemia and false-negative results for hypocortisolaemia. We tested the relevance of this recommendation.

DESIGN

Plasma cortisol and prolactin levels were determined before (T-15) and after a 15-min rest period (T0) in 119 patients, 38 males (M) and 81 females (F). 52 of the 119 patients were known (K; 30 females and 22 males) and 67 unknown (UK; 49 females and 18 males) to the unit.

RESULTS

Prolactin was lower after rest in women (12.3+/-22.7 ng/l vs 11.7+/-22.5 ng/ml, P=0.03), but not in men (6.2+/-4.5 ng/ml at T-15 vs 5.8+/-3.2 ng/ml at T0, P=0.09), in the UK subgroup (10.6+/-20.7 ng/ml at T-15 vs 10.1+/-20.9 ng/ml at T0, P=0.06) and in the K subgroup (10.1+/-16.7 ng/ml at T-15 vs 9.7+/-15.8 ng/ml at T0, P=0.08). None of the patients with prolactin levels higher than 20 ng/ml at T-15 diminished its prolactin value below this cut-off value. Plasma cortisol levels were lower after rest in women (17.9+/-5.9 microg/dl at T-15 vs 16.5+/-6.1 microg/dl at T0, P<0.0001), in the UK subgroup (18+/-6.1 microg/dl at T-15 vs 16.6+/-6.4 microg/dl at T0, P=0.0003) but not in men (18+/-4.4 microg/dl at T-15 vs 17.5+/-5.8 microg/dl at T0, P=0.47) and in the K subgroup (17.8+/-4.6 microg/dl at T-15 vs 17+/-5.4 microg/dl at T0, P=0.13). At T0, 3.3% and 15% of patients presented values below the cut-off value of 10 microg/dl (276 nmol/l) and 17 microg/dl (470 nmol/l), respectively.

CONCLUSION

These results don't justify intravenous catheterisation with a rest period for plasma prolactin determination in contrast with plasma cortisol determination.

Hormonal response to stress in brittle diabetes

Acute psychological stress may play a role in the glycaemic instability of some patients with type I diabetes through an increased secretion of insulin-counteracting hormones. To examine the validity of this hypothesis, we subjected to a video-recorded public-speaking stress seven healthy persons, six type I diabetics with stable blood glucose levels and six type I diabetics with unstable or brittle diabetes (with more than 10 hypoglycaemia/month and frequent hyperglycaemia). During the test and on a control day, heart rate, blood pressure, plasma ACTH, cortisol, catecholamines and prolactin were measured. The comparison between the stable and unstable diabetics during the stress session by two-way analysis of variance (group/time) showed a significant difference for heart rate, blood pressure, ACTH and cortisol. Psychological interview showed that most unstable diabetics perceived a link between life stress and their blood glucose control. The unstable patients had much more difficulty in verbalizing their emotions. Our study shows that the two groups of diabetic patients display distinct cardiovascular and neuroendocrine responses to psychological stress, as well as distinct psychological profiles. In conclusion, hormonal response to an acute psychological stress is more pronounced in brittle diabetes and might be one of its pathogenic factors.

Serum prolactin levels and behavior in infants

This study examined the relationship between serum prolactin levels and behavior in infants and toddlers who experienced two potentially stressful experiences (developmental testing and venipuncture). Serum prolactin levels showed considerable consistency over a 3-month period (r = 0.64 between study entry and three months, p < 0.001, n = 50). There was also stability in having either a normal or a high value (> or = 25 ng/ml). Among children who had a normal value on initial testing, 97% also has a normal value after 3 months; 55% of those with initial high values continued to have high values (chi 2 = 19.26, p < 0.001). Children with high serum prolactin levels were more likely to be rated as unusually hesitant and unhappy during developmental testing. Overall, 53% of the children with serum prolactin levels > or = 25 ng/ml were considered abnormal in affect, compared to 20% of those with lower serum prolactin values (total n = 138, chi 2 = 13.56, p < 0.001). These results suggest that, even in early life, serum prolactin levels may reflect characteristic individual behavioral and neuroendocrine responses to stress.

Impaired interhemispheric integration in brain oxygenation and hemodynamics in schizophrenia

We examined 38 patients with chronic schizophrenia to find and qualify disturbances in interhemispheric integration in brain oxygen metabolism and hemodynamics during a psychological task. A group of thirty-eight age- and sex-matched healthy volunteers were monitored as controls. Multi channel near-infrared (NIR) spectrophotometry was used to observed real-time alterations in cerebral oxygenation in areas of both hemispheres of the forebrain adjacent to the forehead during the mirror drawing task (MDT). In response to MDT normal volunteers showed distinct and well-integrated patterns of changes in oxygenated hemoglobin Hb, deoxygenated Hb, and blood volume total Hb. On the other hand, half the schizophrenics showed dysregulated patterns between hemispheres which never appeared in normal volunteers. Certain schizophrenic symptoms may be related to defective interhemispheric integration.

Gender- and handedness-related differences of forebrain oxygenation and hemodynamics

To elucidate gender- and handedness-related differences between the hemispheres of the brain in their metabolisms and hemodynamics, simultaneous monitoring by near-infrared (NIR) spectrophotometry of hemoglobin (Hb) in both hemispheres of the forebrain during the mirror drawing task (MDT) was performed. Bilaterally simultaneous increases of oxygenated Hb and decreases of deoxygenated Hb in forebrain occurred symmetrically in all cases of volunteer subjects except for two. There were gender- and handedness-related differences of hemodynamics between the hemispheres of the brain; NIR results showed that a large majority of women used both sides of the brain when concentrating on carrying out the MDT, whilst most men, especially left-handers, reacted mainly using the hemisphere which was 'dominant' according to handedness.

Growth hormone response to exercise in asthmatic and normal children

Maximal growth hormone (GH) increments following exercise were compared in asthmatic (n = 14) and normal (n = 8) children. Exercise, which consisted of 6 min ergometer cycling while breathing cold dry (CD) air, induced asthma in all asthmatic patients but not in normal subjects. Baseline plasma GH levels were similar in both groups. Following exercise, however, asthmatic patients had significantly higher mean GH increments than normal subjects (14.8 vs 4.9 ng/ml, P less than 0.025). To evaluate the possible role of bronchoconstriction in the GH response all subjects exercised again, this time while breathing warm humid (WH) air. Despite the absence of exercise-induced asthma (EIA) while breathing WH air, asthmatic patients still had significantly higher mean GH increments than normal subjects (9.2 vs 2.3 ng/ml, P less than 0.05). We conclude that some asthmatic children show excessive GH secretion after exercise regardless of inspired air conditions or the development of EIA.

Endocrine and behavioral responses to psychological stress in hyperandrogenic women

Stress has been implicated in the physiopathology of the ovarian androgenic syndrome. To explore further this notion, we compared the behavioral and endocrine responses to a mental stressor between women with hyperandrogenism (n = 13) and normals (n = 11). The standardized psychological stimulus produced higher levels of anxiety in the hyperandrogenic group than in controls. The endocrine (cortisol, prolactin, growth hormone, beta-endorphin) responses poststressor were definitely dissociated. Both groups showed a comparable anticipatory stress cortisol-secretion response. The cortisol release was greater following the mental stressor in the hyperandrogenic group than in the normals. Thus, hyperandrogenic women appear to have an abnormally affected pituitary-adrenal activation, which may play a role in the pituitary-ovarian disruption characteristic of the ovarian androgenic syndrome.

Cardiovascular reactivity and plasma prolactin response to mental stress in normals and hypertensives

Central dopaminergic activity (CDA) may be involved in blood pressure control as a negative modulator of sympathetic outflow. In this study the plasma PRL changes produced by mental stress (a colour-word conflict test, CWT) were investigated in normals (NT, n.15) and stable hypertensives (HT, n.16) and the PRL response, as a possible index of CDA was correlated to the cardiovascular and the plasma renin activity (PRA) responses as indexes of peripheral sympathetic outflow. A significant (p less than 0.05) slight decrease in mean values of PRL was observed in normals after the CWT but no change was found in hypertensives. No correlation was found between the PRL responses to CWT and the maximal mean arterial pressure changes or the PRA changes whether the groups were considered separately or together. These findings indicates that PRL does not appear to be a reliable index of the CDA involved in blood pressure control.

Prolactin in patients with major depressive disorder and in healthy subjects. III. Investigation of basal and post-TRH prolactin during different forms of acute and chronic psychological stress

Prolactin (PRL) levels after thyrotropin-releasing hormone (TRH) (delta PRL) were determined 4 times and basal prolactin levels 6 times in 10 healthy medical students before, during, and after a major examination in medicine. No significant differences in basal or delta PRL levels occurred during the examination period. Basal PRL was also measured in 14 doctors of medicine after delivering a paper at a clinical conference and a further 9 doctors both before and after delivering a paper. PRL was measured serially at 20-min intervals in 4 doctors on the day on which they presented the paper. No significant differences in PRL levels were found in any of the tests conducted during this kind of stress as compared with the corresponding values obtained under nonstressful conditions. Increases in PRL before delivering the paper were seen in 3 subjects, but such increases also occurred completely independently of stress. An 18-hr fast did not influence PRL secretion in 11 healthy volunteers. Both the information obtained from a review of the literature on the influence of stress on PRL secretion and our own results strongly suggest that contrary to common opinion, there is no evidence at all that psychological stress affects PRL secretion in man.

Prolactin, growth hormone and thyrotropin-thyroid hormone secretion during stress states in man

Changes in hormone secretion and/or metabolism almost constantly accompany stressful events. The hormonal response to stress is directly related to the intensity of the stimulus, and greatly depends on the individual's perception of potentially stressful situations. Hypoglycaemia, surgery and exercise represent physical, metabolic and psychological stressful events where the activation of the endocrine system plays a great role. These endocrine responses also include the secretion of GH and prolactin, but the response pattern varies with the stimulus. Hypoglycaemia, exercise and surgery are potent stimuli to GH and prolactin release, both in men and women. The available data suggest that prolactin is more responsive than GH to surgical stress, whereas physical exercise and hypoglycaemia preferentially stimulate GH secretion. Prolactin levels during hypoglycaemia rise solely when symptomatic neuroglycopenia is achieved. By contrast, prolactin and GH responses to purely psychological stress are rarely seen, although some forms of reproductive stress may potentiate prolactin release in women. A teleologically satisfactory rationale for the acute GH and prolactin rise in response to these stressful stimuli is not clearly apparent in man. No definite metabolic activity of prolactin on intermediate metabolism has been demonstrated, although prolactin is mildly diabetogenic. The known metabolic actions of GH do not appear to be critical during surgery or acute hypoglycaemia, although GH probably participates in the regulation of metabolic homeostasis during chronic hypoglycaemia and chronic exercise. Changes in secretion and/or metabolism of hypothalamic neurotransmitters can increase the secretion of GH by increasing the secretion of GHRH or by decreasing the secretion of somatostatin. The prolactin rise is brought about by either a decrease in dopamine activity, an increased secretion of a hypothetical PRF, or by both mechanisms. Since multiple neuronal pathways converge on the hypothalamus from many other parts of the brain, the pronounced effects of hypoglycaemia, exercise and surgery on the secretion of GH and prolactin also reflect the action of different and complex neural inputs on the activity of the hypothalamic-pituitary axis. However, the morpho-functional mapping of these excitatory pathways still remains incomplete. TSH secretion is tightly regulated by the negative feed-back mechanism exerted by thyroid hormones. The small changes in TSH level observed during surgery and physical exercise seem to reflect mainly alterations in peripheral T4 metabolism.(ABSTRACT TRUNCATED AT 400 WORDS)

Sex differences in prolactin change during mourning

Fourteen men and 12 women were interviewed eight weeks after conjugal bereavement to discuss the events prior to the spouse's death and the subsequent bereavement period. Prolactin (PRL) was measured at the beginning and end of the interview. Descriptions of the deceased spouse were obtained during the interview and rated for Developmental Level of Object Representation (DLOR), a measure of the cognitive complexity of the description. There were significant correlations between DLOR and PRL change for both men and women but the correlation for women was positive and the correlation for men was negative. These findings extend the literature on the psychological correlates of PRL change and suggest that the physiological changes associated with mourning are different for men and women.

Prolactin and stress: some endocrine correlates of pre-operative anxiety

An attempt was made to correlate the serum concentration of PRL, GH and cortisol with anxiety experienced before elective surgery. A significant direct correlation was observed between cortisol and anxiety (P = 0.028), and between GH and anxiety (P = 0.001) although the latter correlation may have reflected the confounding effects of sex and age on GH levels. No correlation was observed between anxiety and PRL and we conclude that psychosocial stress cannot be invoked as an occasional cause of hyperprolactinaemia.

Correlation between anxiety and serum prolactin in humans

The organised social structure of primate species largely determines individual behaviour. Moreover Keverne et al. in Advanced Views in Primate Biology, Springer, Berlin (1982) [1] have demonstrated that social interaction among confined male talapoin monkeys may result in endocrine changes, and that these in turn may modify behaviour. We have undertaken a study to determine whether similar endocrine changes can be induced in confined human volunteers. Five men were confined on a boat for fourteen days. Repeated plasma samples were taken under controlled conditions for assay of testosterone, prolactin and cortisol. After each sample the men completed a questionnaire to document self-perceived anxiety. The men were also secretly ranked for dominant/aggressive behaviour towards the other males. Significant correlation was found between day-to-day changes in anxiety and stress hormones, cortisol and prolactin. Significant correlation was found also between plasma prolactin, testosterone and rank position for dominance/aggression. It is concluded that under some circumstances social interaction may modify endocrine status in humans.

[Neuroendocrine response to anesthesia with isoflurane]

The effects during surgery of a new halogenated volatile anaesthetic, isoflurane, on the hypothalamo-hypophyseal-thyroid-suprarenal axis were studied. In fact, it was important to prove whether this new halogenated anaesthetic would provide better protection, for the patient, from surgery and anaesthetic stress compared with other anaesthetic agents in use. The study was carried out in 16 young class ASA I patients who were to undergo appendicectomy. Before and during operation, blood was taken to measure ACTH, cortisol, TSH, T3, T4 and PRL plasma levels. A remarkable increase of PRL, cortisol and T4 plasmatic rate was found, especially at the end of the operation. It was concluded that isoflurane, just like enflurane, did not prevent the increase of PRL, cortisol and T4 that usually takes place during surgery.

Relationship between growth hormone levels and time spent by soldiers in an active war zone

In a study involving soldiers in an active war zone, we found that serum levels of growth hormone correlated significantly with the amount of time spent in the war zone. Serum cortisol and prolactin showed no such correlation. Our observation suggests that growth hormone may be a useful marker of chronic stress.

Hypothalamic control of GH secretion: pathophysiology and clinical implications

GH is secreted episodically. Its pattern is regulated by the interplay of a releasing and a release-inhibiting hormone of hypothalamic origin. Modulation occurs by metabolic factors (glucose, free fatty acids, ketone bodies, amino acids). Altered GH secretion has been observed in states of metabolic derangement such as diabetes mellitus, malnutrition and obesity. Further modulation occurs by extrahypothalamic CNS structures. In man--but not in animals, including subhuman primates--sleep has an important effect on GH secretion. A defective GH secretory pattern has been found to occur in several states of sleep disturbance, such as sleep deprivation, narcolepsy, severe psychosocial derangement, the apallic syndrome. Other CNS influences on GH secretion are related to stress, emotional changes and psychiatric disturbances. The exact mechanisms by which most of these influences are relayed to the GH secretory apparatus of the hypothalamus remain yet to be investigated.

Plasma prolactin levels during conditioned avoidance behavior in rats

In these experiments, we examined the prolactin (PRL) response during the acquisition of a conditioned avoidance response (CAR). Rats were tested daily in a two-way shuttle box. They were presented with a light stimulation followed by an electric footshock. During each trial period, the rats were given the opportunity to escape the footshock by moving to a safe side of the box. Movement to the appropriate location after the warning signal (light) begins, but before the onset of the footshock, constitutes a CAR. Blood samples were collected from an indwelling cannula and analyzed by radioimmunoassay. PRL levels increased during early acquisition testing, when the rats had not learned to avoid the shock. After one week of testing, acquisition performance increased considerably (70% CARs) while PRL levels remained unchanged. Thus, we were able to show that as rats learned to modify their behavior in response to a stressful situation, they could also modify their PRL response to the stressor.

Beta-endorphin injected into the nucleus of the raphe magnus facilitates metastatic tumor growth

Electrical stimulation of the periaqueductal gray of the rat's midbrain analgesia leads to an increase in the number of artificial pulmonary metastases from the Walker 256 tumor. In an effort to investigate the influence of the pain suppression system and its associated peptides on this phenomenon, we activated the pain suppression system directly from the Nucleus of the Raphe Magnus, a non-opioid subsystem. After inducing analgesia by direct injection of beta-endorphin on the Nucleus of the Raphe Magnus, we noted an increase in the number of artificial pulmonary metastases. This result could be blocked by pretreatment with naloxone. If the Nucleus of the Raphe Magnus was activated by electrical stimulation sufficient to induce analgesia, the metastatic effect was still present but markedly attenuated.

Neuroendocrine alterations in systemic disease

Systemic disorders clearly may exert a significant influence on neuroendocrine function. Disorders that cause significant stress to the body, either physical or psychological, may cause a resetting upward of the HPA axis to provide sufficient cortisol to counteract the stress and to help sustain energy substrate levels. GH levels also increase in many of these situations, again promoting sufficient energy substrate levels. In some circumstances the concomitantly low somatomedin activity may be speculated to be adaptative to prevent the insulin-like agonist activity of these substances as well as to prevent energy expenditure in body growth. However, in other situations such as chronic renal failure and cirrhosis, the decreased somatomedin activity may be primary, causing decreased feedback at the hypothalamic-pituitary level and increased GH levels. The stress-induced rise in PRL may also play a minor role in preserving energy substrate since high levels may promote insulin resistance. In most illnesses the 'euthyroid sick syndrome' develops. Whether such patients are 'euthyroid' or mildly hypothyroid is a matter of controversy. The fact that protein losses are increased during fasting when the lowered T3 levels are returned to normal with exogenous T3 supplementation suggests that these patients are indeed hypothyroid and this hypothyroidism serves to conserve energy substrate by decreasing the metabolic rate. The reproductive axis is often impaired with systemic illness. Again, teleologically this may be viewed as an inactivation of non-essential functions in times of stress. It would appear that the changes that occur with systemic illness, in general, are favourable to the organism in that they promote survival. The detailed neurotransmitter and hypophyseotrophic hormone changes resulting in the alteration in pituitary function remain to be elucidated for the most part.

Prolactin response to the cold pressor test in patients with panic attacks

Recent advances in psychiatric nosology have provided a clearer clinical description of panic disorder and agoraphobia with panic attacks. This clinical improvement has yet to result in a better understanding of the basic mechanisms involved in these conditions. The present study compares the effects of the cold pressor test on blood pressure changes and prolactin levels in patients with panic disorder and agoraphobia with panic attacks to its effects in normal controls. Because the effects of the cold pressor test were similar in patients and normal controls, it appears that alpha-adrenergic function is normal in the patient group. The results also suggest that prolactin hypersecretion does not occur in pathological anxiety states.

Prolactin and cortisol in cerebrospinal fluid: sex-related associations with clinical and psychological characteristics of patients with low back pain

Clinical and psychological characteristics of 33 patients with low back pain were correlated with prolactin and cortisol concentrations in cerebrospinal fluid (CSF). A significant sex difference was found in CSF prolactin levels: women secreted more prolactin into the CSF than did men. High CSF cortisol levels were associated with a rhizographically-demonstrable abnormality, suggesting a relationship between cortisol and an 'organic' origin of pain symptoms. Impairment-disability indices also were associated with CSF hormone levels. Moreover, the two hormones had dissociated psychological correlates. Prolactin was related to depression and anxiety, whereas cortisol was related to somatization. Sex differences were observed in the cortisol response to the symptoms of chronic low back pain, especially in the presence of anxiety and somatization. The sex differences in psychoneuroendocrine and emotional responses suggest that male and female pain patients have different coping mechanisms.

Clinical significance of the luteinized unruptured follicle syndrome as a cause of infertility

The luteinized unruptured follicle syndrome (LUF) exists and is cause of infertility. Although the data of the laparoscopic inspection of the ovaries and the corpora lutea and of steroid hormone concentrations in peritoneal fluid constitute strong evidence that the syndrome exists, its role as a cause of infertility is less clear. The only evidence available that the LUF syndrome is a cause of infertility, is the finding that the syndrome occurs statistically more frequently in women with unexplained infertility than in a control group of women. It still has to be proven whether the LUF syndrome occurs repetitively in each cycle and causes infertility, or whether the syndrome occurs intermittently and only reduces the probability of conception. Diagnosis of the syndrome can be made by laparoscopic inspection of the ovaries and by the assay of 17 beta-estradiol and progesterone, in peritoneal fluid between day 14 and 20 of the cycle. The relationship between the LUF syndrome and pelvic endometriosis and luteal phase insufficiency is discussed. We suggest that the LUF syndrome might be the cause of endometriosis thus explaining the statistical association between both syndromes, and the infertility of women with only mild endometriosis. We favor the hypothesis that the LUF syndrome might be caused by stress thus constituting a "psychological infertility" and we suggest that the syndrome could explain the spontaneous cure rate. Therefore, before any therapy can be accepted as the treatment of the LUF syndrome, it should be strictly assessed with adequate controls.

Is prolactin secreted ectopically?

The ectopic secretion of prolactin by neoplasms has been reported only rarely although in one preliminary report it has been found to be present commonly in tumor extracts. To ascertain whether this is due to a true rarity of prolactin-secreting tumors or to the fact that prolactin hypersecretion causes few notable clinical symptoms in patients with cancer, we measured prolactin levels by radioimmunoassay in the serums of 215 patients who had a variety of malignancies. Prolactin levels were elevated (greater than 25 ng/ml) in 15 patients. In 12 of these 15 patients, the elevations could be explained by the use of phenothiazine or opiates, or prior irradiation to the chest wall or head. Additional serum samples showed clearly normal prolactin levels in one of the patients. Of the entire series of 215 patients, therefore, only two patients (1 percent) had modestly elevated prolactin levels without having known causes for hyperprolactinemia: one patient with squamous cell carcinoma of the lung (prolactin 34.1 ng/ml) and one patient with breast cancer (prolactin 45.4 ng/ml). Even in these two patients the hyperprolactinemia could as likely be due to stimulation of afferent nerves in the chest wall as to ectopic secretion by the tumors. There is no clear evidence in our study or in the literature that prolactin is secreted ectopically; this is in sharp contrast to many studies showing the high frequencies of the ectopic secretion of a number of other hormones. Why this is so is unknown.

Thyrotrophin, prolactin and growth hormone responses to TRH in barbiturate coma and in depression

The effects of 200 microgram thyrotrophin-releasing hormone (TRH) i.v. on thyrotrophin (TSH), prolactin (PRL), growth hormone (GH) and triiodothyronine (T3) were studied in eight patients with barbiturate coma due to attempted suicide, in the same patients after recovery, in eight depressive patients and in eight normal controls. The patients with barbiturate coma presented normal basal TSH and PRL, elevated basal GH and normal PRL but blunted TSH responses to TRH; their GH concentrations varied widely without consistent relation to TRH administration. The same patients after recovery from coma presented normal TSH and PRL, slightly elevated basal GH, and normal PRL but blunted TSH responses to TRH; in four of these patients, a clear-cut rise in GH (i.e. more than 10 ng/ml) occurred after TRH administration. The depressive patients presented normal basal TSH and PRL, slightly elevated basal GH, and normal PRL but blunted TSH responses to TRH; in four of these patients, a moderated rise in GH (less than 10 ng/ml) occurred after TRH administration. The increment in T3 concentrations 120 min after TRH was found reduced in the comatose patients only. Basal cortisol was measured in all the subjects and found elevated in the comatose patients only. It is concluded that the abnormal TSH and GH responses to TRH observed in patients with barbiturate coma are more likely related to depressive illness than to an effect of barbiturates at the pituitary level. Barbiturates might affect thyroid secretion.