Efficacy and safety of individualized growth hormone treatment in adult Japanese patients with growth hormone deficiency

OBJECTIVE: The aim of the study was to evaluate the efficacy and safety of growth hormone (GH) treatment in Japanese adult patients with GH-deficiency. In the extension of the efficacy study, the effect of individualized-dosing (ID), based on insulin-like growth factor-I (IGF-I) levels, and fixed-dose (FD) GH regimens on body composition, were compared in Japanese GH-deficient adults. DESIGN: Randomized, double-blind (DB), placebo-controlled, 24-week treatment period followed by 48-week, open-label study in 43 endocrinology clinics in Japan. Patients received DB treatment with GH (0.012 mg/kg/day; n=57) or placebo (n=60) followed by open-label GH in an ID (n=75) or FD (0.012 mg/kg/day; n=38) regimen. SUBJECTS: Adult Japanese GH-deficient patients (peak GH<3 ng/mL). MEASUREMENTS: Trunk and total body fat (BF), lean body mass (LBM), and adverse events were determined. RESULTS: Percentage trunk fat was reduced significantly more in GH- than in placebo-treated patients at 24 weeks (-16.2 vs. 1.7%, p<0.0001). Open-label treatment with an ID or FD GH regimen provided similar reductions in percentage trunk fat (-8.12 vs. -9.35%), and total BF (-0.92 vs. -0.70 kg) and a comparable increase in LBM (1.032 vs. 0.97 kg). Mean+/-SD GH doses (mg/kg/day) at 48 weeks were significantly lower with the ID GH regimen (ID, 0.0082+/-0.0050; FD, 0.0095+/-0.0033; p<0.05). The safety profile was comparable between ID and FD groups. CONCLUSIONS: Treatment with GH was associated with a significant reduction in trunk fat and improvement in serum lipid profile in Japanese adult GH-deficient patients. The improvement in body composition and tolerability were comparable between ID and FD GH regimens despite a significantly lower daily GH dose with the ID regimen.     

Revised GH and cortisol cut-points for the glucagon stimulation test in the evaluation of GH and hypothalamic–pituitary–adrenal axes in adults: results from a prospective randomized multicenter study

Context

Recent studies suggest using lower GH cut-points for the glucagon stimulation test (GST) in diagnosing adult GH deficiency (GHD), especially in obese patients. There are limited data on evaluating GH and hypothalamic–pituitary–adrenal (HPA) axes using weight-based dosing for the GST.

Objective

To define GH and cortisol cut-points to diagnose adult GHD and secondary adrenal insufficiency (SAI) using the GST, and to compare fixed-dose (FD: 1 or 1.5 mg in patients >90 kg) with weight-based dosing (WB: 0.03 mg/kg). Response to the insulin tolerance test (ITT) was considered the gold standard, using GH and cortisol cut-points of ≥3 ng/ml and ≥18 µg/dL, respectively.

Design

28 Patients with hypothalamic-pituitary disease and 1–2 (n = 14) or ≥3 (n = 14) pituitary hormone deficiencies, and 14 control subjects matched for age, sex, estrogen status and body mass index (BMI) underwent the ITT, FD- and WB-GST in random order.

Results

Age, sex ratio and BMI were comparable between the three groups. The best GH cut-point for diagnosis of GHD was 1.0 (92 % sensitivity, 100 % specificity) and 2.0 ng/mL (96 % sensitivity and 100 % specificity) for FD- and WB-GST, respectively. Age negatively correlated with peak GH during FD-GST (r = ?0.32, P = 0.04), but not WB-GST. The best cortisol cut-point for diagnosis of SAI was 8.8 µg/dL (92 % sensitivity, 100 % specificity) and 11.2 µg/dL (92 % sensitivity and 100 % specificity) for FD-GST and WB-GST, respectively. Nausea was the most common side effect, and one patient had a seizure during the FD-GST.

Conclusion

The GST correctly classified GHD using GH cut-points of 1 ng/ml for FD-GST and 2 ng/ml for WB-GST, hence using 3 ng/ml as the GH cut-point will misclassify some GH-sufficient adults. The GST may also be an acceptable alternative to the ITT for evaluating the HPA axis utilizing cortisol cut-points of 9 µg/dL for FD-GST and 11 µg/dL for WB-GST.

     

Severe growth hormone deficiency is rare in surgically-cured acromegalics

Growth hormone deficiency (GHD) in surgically-cured acromegalics has been reported to negatively affect their metabolic condition and quality of life (QOL). The incidence of GHD, its causes, and its effects on their physio-psychological condition remain to be examined in detail. We performed a retrospective study to investigate GH secretory function in surgically-cured acromegalics, prognostic factors of GHD, and its impact on QOL. The study population consisted of 72 acromegalics who were determined to be surgically cured according to the Cortina consensus criteria. We recorded the incidence of impaired GH secretory function based on the peak GH level during postoperative insulin tolerance test (ITT) which lowered their nadir blood sugar to under 50 mg/dL. Their QOL was evaluated by SF-36. In surgically-cured acromegalics, the incidence of severe GHD (peak GH during ITT ≦ 3.0 μg/L) was 12.5 % (9/72). The preoperative tumor size was significantly larger in patients with severe GHD than without severe GHD (21.9 ± 9.0 vs. 15.5 ± 7.1 mm, p = 0.017). The peak GH levels during postoperative ITT were statistically correlated with the physical but not the mental component summary of the SF-36 score. The incidence of GHD was 12.5 % in our surgically-cured acromegalics. As some QOL aspects are positively related with peak GH levels during postoperative ITT, efforts should be made to preserve pituitary function in acromegalic patients undergoing adenomectomy.     

Comparison of different regimens of glucocorticoid replacement therapy in patients with hypoadrenalism

Since the optimal glucocorticoid replacement needs to avoid over and under treatment, the adequacy of different daily cortisone acetate (CA) doses was assessed in 34 patients with primary and central hypoadrenalism. The conventional twice CA 37.5 mg/day dose was administered to all patients (A regimen: 25 mg at 07:00 h, 12.5 mg at 15:00 h), while in 2 subgroups of 12 patients the dose was shifted on 2 thrice daily regimens (B: 25 mg at 07:00, 6.25 mg at 12: 00, 6.25 mg at 17:00; C: 12.5 mg, 12.5 mg, 12.5 mg). In other 12 patients the conventional dose was reduced to a thrice 25 mg/day administration (D regimen: 12.5 mg, 6.25 mg, 6.25 mg). In all patients, urinary free cortisol (UFC) excretion and cortisol day curves were evaluated. During the CA 37.5 mg administration, nadir cortisol levels were significantly higher with the thrice daily regimens (143 +/- 31 on B and 151 +/- 34 nmol/l on C) than with the conventional twice (85 +/- 16 nmol/l). Moreover, UFC, morning cortisol levels and mean cortisol day curves were similar in each group. Finally, during D regimen nadir cortisol levels were higher than in A and similar to B and C regimens. No difference in UFC and in cortisol day curves by reducing the CA dose was found. In conclusion, the thrice daily cortisone regimens, in which more physiological cortisol levels are achieved, perform better as replacement therapy. The administration of 25 mg/day CA confirms that replacement therapy is more adequate with a lower dose, particularly in patients with central hypoadrenalism.     

Insulin-like growth factor-I correlates more closely than growth hormone with insulin resistance and glucose intolerance in patients with acromegaly

In normal subjects growth hormone (GH) and insulin-like growth factor-I (IGF-I) have opposing effects on glucose metabolism. Active acromegaly is associated with insulin resistance (IR) and glucose intolerance although both GH and IGF-I are elevated. Our objective was to compare whether GH or IGF-I correlates more closely with IR and glucose intolerance in acromegaly. Basal serum IGF-I and GH, glucose and insulin during an oral glucose tolerance test were measured in 70 normoglycemic and 44 hyperglycemic acromegalic patients (21 impaired fasting glucose, 11 impaired glucose tolerance and 12 diabetes mellitus) according to American Diabetes Association criteria. 55 patients were assessed before any treatment for acromegaly and 59 after surgery and/or radiotherapy (15 patients had normal IGF-I after treatment). Patients treated with somatostatin analogs, GH-receptor antagonists or antidiabetic drugs were excluded. IR was assessed by various basal and stimulated indices. Homeostatic Model Assessment 2-Insulin Resistance (HOMA2-IR) index correlated more closely with IGF-I (r = 0.65, p < 0.0001) than nadir (r = 0.23, p = 0.008) or random GH (r = 0.26, p = 0.002). HOMA2-IR correlated better with IGF-I than nadir or random GH also in normoglycemic (n = 70; r = 0.74, p < 0.0001 vs. r = 0.36, p = 0.001 vs. r = 0.39, p < 0.001) and hyperglycemic patients (n = 44; r = 0.54, p = 0.0002 vs. r = 0.09, p = 0.4 vs. r = 0.14, p = 0.26). In multivariate logistic regression analysis IGF-I but not GH was a significant risk factor for glucose intolerance after adjusting for age, sex, weight and acromegaly duration (OR = 1.56, p = 0.01). In acromegaly IGF-I correlates more closely than GH with IR. IGF-I levels but not GH are associated with glucose intolerance.     

Overnight metabolic fuel deficiency in patients treated conventionally for hypopituitarism

BACKGROUND Hormone replacement in hypopituitary adults attempts to reproduce normal physiology. Conventional regimens fail to mimic normal hormone profiles over 24 hours. OBJECTIVE To investigate the metabolic consequences of conventional hormone replacement in hypopituitary adults by measuring circulating levels of the major fuels, glucose, non-esterified fatty acids (NEFA), glycerol and 3-hydroxybutyrate (3-OHB) over 24 hours in hypopituitary subjects and controls. SUBJECTS Ten GH and adrenocorticotrophin deficient hypopituitary adults on conventional replacement and 13 controls matched for age, sex and body mass index were studied. The patients received replacement with hydrocortisone twice daily (at 0730 and 1730 h; mean (range) daily dose 22 (10–30) mg/24 h) but not with GH. Other hormones were replaced as clinically necessary. MEASUREMENTS Circulating glucose, NEFA, glycerol and 3-OHB levels were measured over 24 hours together with concentrations of cortisol (total and free), GH and insulin, and urinary free cortisol. RESULTS Levels of glucose, NEFA and 3-OHB were lower in patients than controls (mean ± SEM) (4.3 ± 0.1 vs 5.3 ± 0.1 mmol/l, P = 0.0001; 291 ± 46 vs 448 ± 48 μmol/l, P = 0.015; 78 ± 8 vs 136 ± 24 μmol/l, P = 0.035, respectively) before breakfast. This decrease in glucose, NEFA and 3-OHB was observed in the patient group throughout the night, from midnight to breakfast. For NEFA, the decrease persisted throughout the 24 hours. Glycerol did not differ significantly in patients and controls. Integrated levels of total and free plasma cortisol, and 24-hour urine cortisol excretion, were normal in patients but total and free plasma cortisol concentrations overnight were markedly decreased (overnight area under the curve (AUC) of total cortisol: 440 ± 154 vs 1593 ± 267 nmol/l h, P = 0.0024; overnight AUC of free cortisol: 24 ± 8 vs 161 ± 26 nmol/l h, P = 0.0001). GH levels were low throughout the whole 24 hours in the patient group (24-hour AUC: 10.6 ± 5.1 vs 74.6 ± 19.6 mU/l h, P = 0.008). CONCLUSIONS Hypopituitary adults on conventional hormone replacement regimens have low concentrations of metabolic fuels, glucose, non-esterified fatty acids and 3-hydroxybutyrate throughout the night, possibly related to GH deficiency or to decreased overnight circulating cortisol levels. This overnight fuel deficiency may underlie the mechanism for the non-specific symptoms, such as fatigue and headache in the early morning, which are frequent in this group of patients.     

Alterations in growth hormone secretory dynamics in adolescent girls with anorexia nervosa and effects on bone metabolism

Anorexia nervosa (AN) is a disorder that is increasing in frequency in adolescents, and the age of onset is often in the prepubertal years, potentially affecting the development of peak bone mass and linear growth. The GH-IGF-I axis plays an important role in bone formation, and alterations in GH secretory patterns have been described in adult women with AN. However, GH secretory dynamics in adolescents with AN have not been described, and the effects of alterations in GH secretory patterns and GH concentration on bone metabolism in AN are not known. We examined patterns of GH secretion by deconvolutional analysis, and GH concentration by Cluster analysis, in adolescent girls with AN (n = 22) and controls (n = 20) of comparable bone age and pubertal stage. We also examined the roles of cortisol, leptin, and estradiol in the regulation of GH secretion and concentration, and the relationship of GH secretory patterns and concentration to bone metabolism. Basal GH secretion and secretory pulse number in adolescent girls with AN were increased compared with control values (P = 0.03 and 0.007, respectively), and increased disorderliness of GH secretion (approximate entropy) was found in AN (P = 0.004). Mean and nadir GH concentrations and total area under the concentration curve were increased (P = 0.03, 0.002, and 0.03, respectively), and IGF-I levels were decreased (P = 0.0002) in girls with AN compared with healthy adolescent girls. IGF-I levels correlated negatively with nadir GH concentrations (r = -0.35; P = 0.02). Serum cortisol levels were higher in girls with AN than in controls (P < 0.0001) and correlated inversely with IGF-I (r = -0.58; P = 0.0001) and weakly with GH concentration (area under the concentration curve; r = -0.43; P = 0.05). A strong inverse relationship between markers of nutritional status (body mass index, fat mass, and leptin) and basal and pulsatile GH secretion, and mean and nadir GH concentrations was observed. GH concentration predicted levels of all markers of bone formation and a marker of bone resorption (N-telopeptide) in healthy controls, but not in AN. We demonstrate increases in basal GH secretion, number of secretory bursts, and GH concentration in adolescents with AN compared with controls, accompanied by low IGF-I levels. These data are consistent with the hypothesis that an acquired GH resistance occurs in this undernourished group. We also demonstrate that GH secretion and concentration are nutritionally regulated, and that the effects of nutrition exceed the effects of cortisol on GH concentration. Acquired GH resistance may play a role in the osteopenia and decreased peak bone mass frequently associated with AN.     

The use of the hypoglycaemic clamp in the assessment of pituitary function

OBJECTIVE: The standard dynamic test used to diagnose hypopituitarism is the insulin tolerance test (ITT), in which insulin-induced secretion of ACTH, GH and cortisol is measured. However, because of differences in insulin sensitivity some patients fail achieve sufficient hypoglycaemia to assess pituitary function and colleagues experience severe hypoglycaemia and are at risk for cardiac dysrhythmia, seizure or coma. This risk may be particularly pertinent in the evaluation of older adults. We hypothesized that the hypoglycaemic clamp may be useful in assessing pituitary function in some patients. PATIENTS AND MEASUREMENTS: Twenty-one normal subjects (14 old [50-76 years] and 7 young [18-36 years]) and 7 hypopituitary subjects were studied. A clamp study was performed in which insulin infusion was given at 2 mU/kg/min and increased to 4 mU/kg/min if the target glucose concentration was not reached after 40 min. Dextrose was infused as needed to clamp the plasma glucose concentration at 2.2 mmol/l for 30 min. On a separate day, 7 young controls also underwent an ITT in which 0.15 U/kg insulin was administered as a bolus intravenous injection at time 0. In both studies, baseline values were taken at - 10, - 5 and 0 min. Samples were then collected every 5 min for plasma glucose and every 10 min for insulin, ACTH, cortisol and GH. RESULTS: ACTH and GH secretion during each test were similar in younger controls (P = NS) but cortisol secretion was lower during ITT (P < 0.01 vs. clamp). Hypopituitary subjects had significantly less ACTH, cortisol and GH secretion than controls of all ages (P < 0.001 for all). Peak GH secretion was significantly lower in the old controls than in young controls (22 +/- 12 vs. 48 +/- 26 mU/l, respectively; P < 0.01) but significantly higher than the hypopituitary subjects (2 +/- 2 mu/l; P < 0.001). CONCLUSION: These data demonstrate that the hypoglycaemic clamp can be used in the assessment of pituitary function and suggest that this technique may be particularly beneficial in the evaluation of GH deficiency in older adults who may not tolerate the ITT.     

Discordance of IGF-1 and GH stimulation testing for altered GH secretion in obesity

ObjectiveTo investigate the concordance/discordance of IGF-1 and peak stimulated GH in identifying subjects with reduced GH secretion and to determine the physiological significance of any discordance in obese subjects.Design, patients and methods95 obese and 43 normal weight men and women underwent measurement of IGF-1 and GH stimulation testing with GH releasing hormone (GHRH)–arginine. Reduced IGF-1 and GH secretion were defined using pre-determined cut-points. Cardiovascular disease risk was determined by measuring carotid intima-media thickness (cIMT). In a second study, IGF-1 was measured in 52 obese men and women who underwent GH stimulation testing and overnight frequent blood sampling. The association of IGF-1 and peak stimulated GH with parameters of endogenous GH secretion was assessed.Results60% of obese subjects had normal IGF-1 and peak stimulated GH while 8.4% of obese subjects had reduced IGF-1 and GH secretion. Discordance rate for IGF-1 and peak GH was 31.6%. Subjects with both low IGF-1 and low peak GH had the highest cIMT, while subjects with both normal IGF-1 and peak GH had the lowest cIMT. Subjects with reduction in either IGF-1 or peak GH, had intermediate cIMT (P = 0.02). IGF-1 and peak stimulated GH were associated with maximum and mean overnight serum GH and GH AUC as well as maximum peak mass and median pulse mass. Peak stimulated GH, but not IGF-1, was also associated with nadir overnight serum GH concentration and basal GH secretion.ConclusionPeak stimulated GH and IGF-1 demonstrate significant discordance in identification of subjects with reduced GH secretion in obesity. Subjects with reduction of either IGF-1 or peak GH had higher cIMT compared to subjects with both normal IGF-1 and peak GH. Subjects with reductions in both IGF-1 and peak GH had the highest cIMT. Peak GH, compared to IGF-1, has broader associations with various parameters of endogenous GH secretion which support its utility in identifying those with reduced GH secretion.     

Circadian rhythms of anterior pituitary hormone secretion: effects of dexamethasone

Circadian rhythms of hormone secretion are by now well recognized but there are limited data available to compare such rhythms of multiple hormones in individual subjects. Therefore concentrations of serum thyrotropin (TSH), cortisol, growth hormone (GH), triiodothyronine (T3) and prolactin (PRL) were determined in blood samples from six healthy males over periods of 24 hours. A circadian periodicity was identified for cortisol, GH, PRL and TSH; maximum concentrations for TSH and PRL occurred at dissimilar times, namely respectively before and after the GH peak. Maximum TSH coincided with nadir values forplasma cortisol. Dexamethasone administration (3 mg over 36 hours) lowered serum T3 concentration and abolished TSH periodicity without affecting the PRL rhythm. These data suggest that a) nocturnal TSH and PRL maximum levels do not have common mediation; b) there may be an inverse relationship between cortisol and TSH, and c) glucocorticoids at low doses preferentially affect secretion of thyrotropin.     

Optimizing glucocorticoid replacement therapy in severely adrenocorticotropin-deficient hypopituitary male patients

Background The optimal replacement regimen of hydrocortisone in adults with severe ACTH deficiency remains unknown. Management strategies vary from treatment with 15–30 mg or higher in daily divided doses, reflecting the paucity of prospective data on the adequacy of different glucocorticoid regimens. Objective Primarily to define the hydrocortisone regimen which results in a 24 h cortisol profile that most closely resembles that of healthy controls and secondarily to assess the impact on quality of life (QoL). Design Ten male hypopituitary patients with severe ACTH deficiency (basal cortisol <100 nm and peak response to stimulation <400 nm ) were enrolled in a prospective, randomized, crossover study of 3 hydrocortisone dose regimens. Following 6 weeks of each regimen patients underwent 24 h serum cortisol sampling and QoL assessment with the Short Form 36 (SF36) and the Nottingham Health Profile (NHP) questionnaires. Free cortisol was calculated using Coolen’s equation. All results were compared to those of healthy, matched controls. Results Corticosteroid binding globulin (CBG) was significantly lower across all dose regimens compared to controls (P < 0·05). The lower dose regimen C (10 mg mane/5 mg tarde) produced a 24 h free cortisol profile (FCP) which most closely resembled that of controls. Both regimen A(20 mg mane/10 mg tarde) and B(10 mg mane/10 mg tarde) produced supraphysiological post‐absorption peaks. There was no significant difference in QoL in patients between the three regimens, however energy level was significantly lower across all dose regimens compared to controls (P < 0·001). Conclusions The lower dose of hydrocortisone (10 mg/5 mg) produces a more physiological cortisol profile, without compromising QoL, compared to higher doses still used in clinical practice. This may have important implications in these patients, known to have excess cardiovascular mortality.     

THE EFFECT OF OXYTOCIN INFUSION ON ADENOHYPOPHYSIAL AND ADRENAL CORTICAL RESPONSES TO INSULIN-INDUCED HYPOGLYCAEMIA

The responses of plasma adrenocorticotrophin (ACTH), cortisol, growth hormone (GH) and prolactin to insulin-induced hypoglycaemia were studied in six lean male subjects (age 22-29 years). Intravenous insulin tests were performed with and without oxytocin infusion. Blood sugar nadir occurred at the onset of symptoms (time S) with no significant differences between oxytocin and saline infusion. During the oxytocin infusion mean plasma oxytocin increased from 1.9 pmol/l to 138 pmol/l. Peak increase in plasma ACTH (oxytocin 266 +/- 54 ng/l; saline 281 +/- 43 ng/l, mean +/- SEM) was at S + 10 min while peak plasma cortisol (oxytocin 680 +/- 47 nmol/l: saline 656 +/- 40 nmol/l) was measured at S +/- 60 min, peak GH (oxytocin 96 +/- 17.8 mU/l; saline 106 +/- 18.6 mU/l) at S + 60 min and prolactin (oxytocin 1332 +/- 239 mU/l; saline 1242 +/- 273 mU/l) at S + 30 min. There were no significant differences in plasma concentrations of ACTH, cortisol, GH or prolactin between saline and oxytocin infusion. The results indicate that oxytocin has no effect on plasma ACTH, cortisol, GH and prolactin responses to insulin-induced hypoglycaemia. In particular they fail to support previous studies which suggested an inhibitory role for oxytocin in ACTH secretion.     

Increased insulin sensitivity in young, growth hormone deficient children

OBJECTIVE: Although growth hormone (GH) has well documented insulin antagonistic effects, GH deficient adults often demonstrate insulin resistance. In young GH deficient children, increased susceptibility to hypoglycaemia might indicate increased insulin sensitivity; however, this has not been documented. We therefore determined insulin sensitivity in GH deficient and GH sufficient children. DESIGN AND PATIENTS: Prospective study of children undergoing insulin tolerance tests for clinical investigation of GH or cortisol secretion at a regional Paediatric Endocrine/Growth Clinic between October 1986 and December 1997. Ninety-one tests were performed in children with GH deficiency and 142 tests in children with normal GH response to insulin (peak GH > or = 20 IU/l). MEASUREMENTS: The standard insulin tolerance test was modified to permit frequent measurements of glucose (0, 5, 10, 15, 20, 30, 45, 60 and 90 minutes). Rate of log glucose disappearance in the first 15 minutes was calculated as a direct measure of insulin sensitivity. RESULTS: GH deficient children were more insulin sensitive than GH sufficient children (P = 0.004) and had lower glucose nadirs post-insulin (P = 0.005). Subgroup analysis revealed that these differences were greater in younger (< 12 years old) or pre/early pubertal children. In 14 prepubertal children, exogenous sex steroid priming resulted in lower insulin sensitivity (P < 0.05) compared to nonprimed tests. CONCLUSIONS: Young GH deficient children were more insulin sensitive than children with normal GH secretion. This difference attenuated with age and puberty, possibly secondary to pubertal sex steroids; however, insulin resistance as reported in GH deficient adults, was not observed in adolescents.     

Fasting unmasks a strong inverse association between ghrelin and cortisol in serum: studies in obese and normal-weight subjects

Administration of ghrelin, the endogenous ligand for the GH secretagogue receptor, stimulates not only GH secretion but also appetite and food intake in humans. Endogenous ghrelin levels display a distinct circadian rhythm, which is reciprocal to that of insulin and presumed to be meal dependent and not associated with GH secretion. We tested the hypothesis that food deprivation could impact circadian serum ghrelin levels and unmask meal-independent regulatory mechanisms. Thirty-three young adults, subdivided according to gender and level of obesity, were studied with blood sampling every 3 h from 12-84 h of fasting. Serum ghrelin levels showed a marked diurnal rhythm with a nadir in the morning (0800 h), peak levels in the afternoon, and a gradual decline during the night. This pattern was preserved during the entire fasting period and was independent of gender and obesity. Mean 24-h ghrelin levels exhibited a small but significant decline during the fast (P < 0.001). As expected, GH secretion increased with fasting in lean subjects, and a gradual decline in insulin concentrations was observed in all subjects. Neither GH nor insulin showed any significant relationship to ghrelin. In contrast, serum cortisol exhibited a strong inverse temporal association with ghrelin (r = -0.79; P < 0.0001). In conclusion, our study yields no evidence that ghrelin stimulates GH release during fasting. As a novel finding, ghrelin appears to be related to cortisol. However, further studies are needed to elucidate the physiological mechanisms behind this relationship.     

GH secretion reserve in subclinical hypercortisolism

Purpose

In overt hypercortisolism, growth hormone (GH) secretion is decreased and normalizes after surgery. In subclinical hypercortisolism (SH), GH secretion has been scarcely investigated. We assessed GH reserve in patients with and without SH and, in the former, also after recovery.

Methods

We enrolled 24 patients with adrenal adenomas, 12 with SH (SH+, 8 females, 58.3 ± 6.5 years) and 12 without SH (SH?; 11 females, 61.8 ± 10.6 years). SH was diagnosed in the presence of ≥2 out of: 1 mg overnight dexamethasone suppression test >83 nmol/L, urinary free cortisol (UFC) >193 nmol/day and ACTH levels <2.2 pmol/L. GH secretion was assessed by GHRH + Arginine test (GHRH–ARG) and age-adjusted serum IGF-I levels, expressed as SDS (IGF-I SDS). Eight SH+ patients were re-evaluated after the recovery from SH.

Results

Age, gender, body mass index (BMI) and IGF-I SDS were comparable between SH+ and SH? patients. After GHRH–ARG the mean GH peak levels (GH-P) and GH response (as Area Under Curve, GH-AUC) were lower in SH+ than in SH? patients (15.2 ± 8.1 vs 44.5 ± 30.9 μg/L, P = 0.004 and 1,418 ± 803 vs 4,028 ± 2,476 μg/L/120 min, P = 0.002, respectively), after adjusting for age and BMI. The GH-AUC and GH-P levels were negatively associated with UFC after adjusting for age and BMI (β = ?0.39, P = 0.02 and β = ?0.4, P = 0.020 respectively). After recovery, GH-P levels and GH-AUC increased as compared to baseline (23.7 ± 16.3 vs 15.8 ± 10.2 μg/L, P = 0.036 and 2,549 ± 1,982 vs 1,618 ± 911 μg/L/120 min, P = 0.012, respectively).

Conclusions

GH secretion reserve is decreased in SH patients and increases after the recovery.     

Successful treatment of childhood-onset Cushing's disease is associated with persistent reduction in growth hormone secretion

OBJECTIVE: Although Cushing's disease (CD) rarely occurs in childhood, affected children commonly fail to achieve predicted adult height. Hypercortisolaemia results in reduced GH secretion and GH-deficiency may persist or demonstrate delayed recovery after successful treatment of CD in adults. Whether recovery of spontaneous GH secretion occurs following treatment of childhood CD has yet to be established. DESIGN AND PATIENTS: We performed a retrospective analysis of the GH status of 13 children (10 males; 12.8 +/- 1.0 years, mean +/- SE) who had undergone successful treatment of CD that occurred prior to the completion of linear growth. Each underwent transsphenoidal hypophysectomy, resulting in satisfactory control of glucocorticoid levels in 7/13 (54%). The remaining six patients (46%) received fractionated external beam irradiation (4500 Gy). At the time of GH assessment, circadian dynamics of cortisol were normal in eight patients and five were receiving titrated glucocorticoid replacement. MEASUREMENTS: GH status was assessed using the peak response to a provocative stimulus. Eleven out of 13 underwent testing with insulin-induced hypoglycaemia (nadir plasma glucose 30 mU/l. Intermediate values were taken to represent subnormal GH status. Assessment of GH status was performed 39 +/- 10 months (median +/- SE) following successful treatment (range 9-108 months). RESULTS: Using these criteria 4/13 (31%) patients had severe GH-deficiency. Only 2/13 (15%) had a normal response. 7/13 (54%) achieved peak GH levels in the subnormal range. Those with multiple pituitary hormone deficiencies were most likely to have lower peak GH levels, but there was no clear effect of pituitary irradiation or relationship between duration post cure and peak GH response. CONCLUSION: GH-deficiency is common and may persist for many years following successful treatment of CD prior to completion of linear growth. External radiotherapy does not necessarily result in severe GH-deficiency in the short term. Assessment of GH status and consideration of GH treatment should be considered following treatment of CD in childhood and adolescence in order to maximize the opportunities to achieve a satisfactory final adult height. In those with subnormal GH responses, continued assessment is necessary to determine whether the GH axis subsequently recovers or if these patients develop features of the adult GH-deficiency syndrome.     

Cholinergic enhancement of pyridostigmine potentiates spontaneous diurnal but not nocturnal growth hormone secretion in short children

It has been shown that enhanced cholinergic tone induced by pyridostigmine (PD) increases both basal and GHRH-stimulated GH levels in both adults and children. In this study the effects of PD (60 mg orally) on GH secretion were studied both in the morning (from 8.00 to 12.00) and in the night (from 23.00 to 3.00) in 7 short children previously shown as having a normal spontaneous nocturnal GH secretion. In the morning, PD induced a GH increase higher than saline (peak, mean +/- SEM: 17.4 +/- 3.4 vs. 5.5 +/- 3.0 ng/ml, p less than 0.02; area under curve (AUC): 360.8 +/- 71.4 vs. 109.4 +/- 44.7 ng/ml/h, p less than 0.01). In the night, no difference was observed between GH secretion after PD (peak: 16.7 +/- 2.4 ng/ml; AUC: 468.2 +/- 95.5 ng/ml/h) and saline (peak: 16.0 +/- 2.7 ng/ml; AUC: 409.1 +/- 97.7 ng/ml/h). Spontaneous GH secretion was higher during the night than in the morning (p less than 0.02) whereas nocturnal GH secretion overlapped with that in the morning after PD. The ability of PD to increase GH secretion during the morning but not GH hypersecretion occurring at night implies that the cholinergic tone in the central nervous system areas controlling GH secretion is already maximally stimulated at night. Since, reportedly, the cholinergic system negatively modulates somatostatin secretion, presence of a physiologically reduced somatostatinergic tone may be envisaged at night.     

Moderate dose cranial radiotherapy causes central adrenal insufficiency in long-term survivors of childhood leukaemia

Acute lymphoblastic leukaemia (ALL) is the most common childhood malignancy. The survival rate in the Scandinavian countries is now around 85 %. ALL patients treated with cranial radiotherapy (CRT) are at risk for growth hormone deficiency (GHD), but little is known about other pituitary insufficiencies, e.g. ACTH. Adult ALL patients (median age at study 25 years), treated with 24 Gy (18–30) of CRT during childhood were investigated. We performed an insulin tolerance test (ITT) to evaluate cortisol secretion. We measured basal serum ACTH and cortisol levels before and after 5 years of GH therapy. 14 out of 37 (38 %) ALL patients had a subnormal cortisol response to an ITT (257–478 nmol/L) while there was no significant difference in basal cortisol levels between 44 patients and controls (P > 0.3). Female, but not male ALL patients had significantly lower ACTH levels compared to controls (P = 0.03). After 5 years of GH therapy only male ALL patients had significantly lowered basal plasma cortisol (P = 0.02). ALL survivors, treated with a moderate dose CRT, have a central adrenal insufficiency 20 years after diagnosis. An increased awareness of the risk for an adrenal insufficiency is of importance and life-long surveillance of the entire hypothalamic–pituitary axis is recommended in these patients.     

Abnormal effect of cigarette smoking on pituitary hormone secretions in insulin-dependent diabetes mellitus

OBJECTIVE We observed the effect of smoking two cigarettes on GH, AVP and cortisol secretion in patients with diabetes and normal subjects. DESIGN AND PATIENTS We tested 10 male smokers with insulin-dependent diabetes mellitus (IDDM) and 10 normal smokers. On a different occasion, normal and diabetic smokers were tested with an insulin (0.15 U/kg body weight) tolerance test (ITT). MEASUREMENTS Hypoglycaemia-induced hormonal responses in smokers were compared with those observed in 10 diabetic and 10 normal non smokers. RESULTS All subjects showed similar basal GH, cortisol and AVP levels. In the normal subjects, cigarette smoking induced a significant increase in circulating GH, AVP and cortisol levels, with mean peaks 3.3, 3 and 1.58 times higher than baseline, respectively. Smoking-induced hormonal responses were significantly higher in diabetics (mean peak was 5.2 times higher than baseline for GH, 4.0 for AVP and 1.83 for cortisol). Insulin induced a similar hypoglycaemic nadir in all subjects at 30 minutes, even though the diabetic subjects had a delayed recovery in blood glucose levels. GH and AVP responses to hypoglycaemia were significantly higher in diabetic (mean peaks 11.5 and 3.2 times higher than baseline, respectively) than in normal (mean peaks 7.3 and 1.9) non-smokers, whereas these groups showed similar cortisol responses (mean peak 2.3 times higher than baseline). Smoking did not change any hypoglycaemia-induced hormonal rise in the normal controls, whereas it significantly enhanced GH, AVP and cortisol levels (mean peaks 14.5, 4 and 3.8 times higher than baseline, respectively) in diabetics. CONCLUSIONS In patients with IDDM, cigarette smoking not only elicits higher GH, AVP and cortisol responses than in normal subjects, but also enhances the counter-regulatory hormone responses to insulin-induced hypoglycaemia. These findings suggest interactions between nicotine inhaled with cigarette smoking and diabetes-induced neuroendocrine alterations.     

A comparison of therapeutic dosages of decitabine in treating myelodysplastic syndrome: a meta-analysis

Decitabine is used to treat myelodysplastic syndrome (MDS). This meta-analysis evaluated the efficacy and safety of different dosing regimens of decitabine in treating intermediate and/or high-risk MDS. Medline, Cochrane, EMBASE, and Google Scholar databases were searched up to October 23, 2015. Randomized controlled trials, prospective, cohort, and case series studies were included. Fifteen studies were included with a total of 1378 patients. The decitabine 100 mg/m²/course dosing regimen had a greater overall response rate than the 60–75 mg/m²/course (51 vs. 25%; P = 0.003). It also had higher complete response rate compared with the 135 mg/m²/course regimen (24.2 vs.13.7%; P = 0.016). The three dosing regimens were similar with respect to bone marrow complete response and partial response and hematologic improvement (P values > 0.05). Decitabine 135 mg/m²/course regimen had similar hematologic improvement as best supportive care (P = 0.066). The incidence of neutropenia, thrombocytopenia, infections, and anemia was similar across treatment groups (range, 31 to 38%; P values ≥ 0.899). The 100 mg/m²/course decitabine regimen showed benefit with respect to overall response rate compared with the 60–75 mg/m²/course regimen, as well as greater improvement in complete response rate compared with the 135 mg/m²/course regimen. All three dosing regimens had similar frequency of adverse events.