Absence of the dawn phenomenon and abnormal lipolysis in type 1 (insulin-dependent) diabetic patients with chronic growth hormone deficiency

Analysis of Prevalence, Magnitude and Timing of the Dawn Phenomenon in Adults and Adolescents With Type 1 Diabetes: Descriptive Analysis of 2 Insulin Pump Trials

OBJECTIVES

To better understand the dawn phenomenon in type 1 diabetes, we sought to determine its prevalence, timing and magnitude in studies specifically designed to assess basal insulin requirements in patients using insulin pumps.

METHODS

Thirty-three participants from 2 sensor-augmented insulin pump studies were analyzed. Twenty participants were obtained from a methodologically ideal semiautomated basal analysis trial in which basal rates were determined from repeated fasting tests (the derivation set) and 13 from an artificial pancreas trial in which duration of fasting was variable (the "confirmation" set). Prevalence was determined for the total cohort and for individual trials using the standard definition of an increase in insulin exceeding 20% and lasting ≥90 minutes. Among cases, time of onset and percent change in the magnitude of basal delivery were determined.

RESULTS

Seventeen participants (52%) experienced the dawn phenomenon (11 of 20 [55%] in the derivation set and 6 of 13 [46%] in the confirmation set). Time of onset was 3 AM (interquartile range [IQR], 3 to 4:15 AM) in the derivation set and 3 AM (IQR, 3 to 4 AM) in the confirmation set. The magnitude of the dawn phenomenon was a 58.1% (IQR, 28.8% to 110.6%) increase in insulin requirements in the derivation set and 65.5% (IQR, 45.6% to 87.4%) in the confirmation set.

CONCLUSIONS

The dawn phenomenon occurs in approximately half of patients with type 1 diabetes; when present, it has predictable timing of onset (generally 3 AM) and a substantial, but highly variable, magnitude. These findings imply that optimization of glycemic control requires clinical emphasis on fasted overnight basal insulin assessment.

Sleep disorders and the development of insulin resistance and obesity

Normal sleep is characterized both by reduced glucose turnover by the brain and other metabolically active tissues, and by changes in glucose tolerance. Sleep duration has decreased over the last several decades; data suggest a link between short sleep duration and type 2 diabetes. Obstructive sleep apnea (OSA) results in intermittent hypoxia and sleep fragmentation, and also is associated with impaired glucose tolerance. Obesity is a major risk factor for OSA, but whether OSA leads to obesity is unclear. The quality and quantity of sleep may profoundly affect obesity and glucose tolerance, and should be routinely assessed by clinicians.

Ablation of Egr2-positive cells in male mouse anterior pituitary leads to atypical isolated GH deficiency

In this study, we have investigated the expression and function of the transcription factor early growth response factor 2 (Egr2)/Krox20 in the developing anterior pituitary. Egr2 is initially expressed in all differentiating hormonal cells types, but its expression is mostly restricted to the somatotroph lineage after birth. Egr2 knockout results in anterior pituitary hypoplasia. However, the analysis of a conditional mutant demonstrates that this phenotype does not originate from a lack of Egr2 expression in the pituitary. Using an Egr2 allele driving a Cre-activable toxin gene, we performed a genetic ablation of Egr2-positive cells in the pituitary. During the postnatal period, this ablation leads to specific and progressive depletion of the somatotroph population, creating a novel model of early-onset isolated GH deficiency (GHD). Mutant animals were subjected to a complete metabolic analysis, revealing atypical and expected features. Consistent with an adult-onset isolated GHD model, mutant animals are hypoglycemic and display increased insulin sensitivity and glucose clearance. This latter phenotype is in contrast to the glucose intolerance observed in another early-onset GHD model. Surprisingly, increased insulin sensitivity is not accompanied by a modified balance between fat and lean tissues, but by reduced metabolic adaptability between glucose and lipid oxidation conditions. This suggests that the relationship between these metabolic features and insulin sensitivity should be reconsidered. In conclusion, our mutant may be a valuable genetic model with which to study the effects of long-term GH deficiency, in conditions of normal pancreatic function and unaffected balance between fat and glucose metabolism.

Metabolic impact of adult-onset, isolated, growth hormone deficiency (AOiGHD) due to destruction of pituitary somatotropes

Growth hormone (GH) inhibits fat accumulation and promotes protein accretion, therefore the fall in GH observed with weight gain and normal aging may contribute to metabolic dysfunction. To directly test this hypothesis a novel mouse model of adult onset-isolated GH deficiency (AOiGHD) was generated by cross breeding rat GH promoter-driven Cre recombinase mice (Cre) with inducible diphtheria toxin receptor mice (iDTR) and treating adult Cre^+/−,iDTR^+/− offspring with DT to selectively destroy the somatotrope population of the anterior pituitary gland, leading to a reduction in circulating GH and IGF-I levels. DT-treated Cre^−/−,iDTR^+/− mice were used as GH-intact controls. AOiGHD improved whole body insulin sensitivity in both low-fat and high-fat fed mice. Consistent with improved insulin sensitivity, indirect calorimetry revealed AOiGHD mice preferentially utilized carbohydrates for energy metabolism, as compared to GH-intact controls. In high-fat, but not low-fat fed AOiGHD mice, fat mass increased, hepatic lipids decreased and glucose clearance and insulin output were impaired. These results suggest the age-related decline in GH helps to preserve systemic insulin sensitivity, and in the context of moderate caloric intake, prevents the deterioration in metabolic function. However, in the context of excess caloric intake, low GH leads to impaired insulin output, and thereby could contribute to the development of diabetes.

Metabolic rate and fuel utilization during sleep assessed by whole-body indirect calorimetry

The purpose of this study was to examine metabolic rate and substrate oxidation during sleep in relation to time of sleep and sleep stage. Twelve male subjects free from sleep-disordered breathing slept for 469 +/- 8.7 (mean +/- SE) minutes until natural awakening in a whole-body indirect calorimeter, and polysomnographic documentation of sleep was recorded. Energy expenditure decreased during the first half of the night, reached a nadir (a 35% decrease), and remained relatively stable until awakening. Similarly, fat oxidation decreased from the onset of sleep. On the other hand, carbohydrate oxidation showed no remarkable changes from the onset of sleep but began to increase before awakening. Because distribution of sleep stages is not uniform throughout the night, with rapid-eye-movement (REM) sleep tending to appear later in the sleep, effect of sleep stage on energy metabolism was isolated by analysis of covariance with time as a covariate. Subsequent comparison of metabolic rate by 1-way analysis of variance with Bonferroni post hoc analysis revealed that energy expenditure during REM sleep was significantly greater than that during sleep stages 2 and 3/4 (stage 2, 25.248 +/- 0.961; stage 3/4, 24.825 +/- 0.935; REM, 25.712 +/- 0.928 kcal kg(-1) fat-free mass d(-1)). Carbohydrate oxidation during REM sleep was significantly greater than that during sleep stage 3/4 (stage 3/4, 12.229 +/- 1.071; REM, 13.986 +/- 1.291 kcal kg(-1) fat-free mass d(-1)). Respiration quotient was statistically different among sleep stages, but Bonferroni post hoc analysis failed to identify significant differences (stage 2, 0.850 +/- 0.010; stage 3/4, 0.846 +/- 0.011; REM, 0.861 +/- 0.013). The increases in energy expenditure and carbohydrate oxidation during REM sleep are consistent with a notion that changes in energy metabolism in brain are manifested as small fluctuations in whole-body energy metabolism during sleep.

The Dawn Phenomenon Revisited: Implications for Diabetes Therapy

OBJECTIVE

To summarize current data on the magnitude, prevalence, variability, pathogenesis, and management of the dawn phenomenon in patients with diabetes mellitus.

METHODS

On the basis of the pertinent available literature and clinical experience, we propose a quantitative definition of the dawn phenomenon, discuss potential pathogenic mechanisms, and suggest management options.

RESULTS

The "dawn phenomenon" is a term used to describe hyperglycemia or an increase in the amount of insulin needed to maintain normoglycemia, occurring in the absence of antecedent hypoglycemia or waning insulin levels, during the early morning hours. To be clinically relevant, the magnitude of the dawn increase in blood glucose level should be more than 10 mg/dL or the increase in insulin requirement should be at least 20% from the overnight nadir. Controversy exists regarding the frequency, reproducibility, and pathogenesis of the dawn phenomenon. Approximately 54% of patients with type 1 diabetes and 55% of patients with type 2 diabetes experience the dawn phenomenon when the foregoing quantitative definition is used. The most likely pathogenic mechanism underlying the dawn phenomenon is growth hormone-mediated impairment of insulin sensitivity at the liver and muscles. The exact biochemical pathways involved are unknown. Therapeutic decisions aimed at correcting fasting hyperglycemia should take into account the variability and magnitude of the dawn phenomenon within individual patients. Successful insulinization appears to minimize the effects of the dawn phenomenon. Currently, no subcutaneous depot preparation of insulin exists that is capable of mimicking the basal insulinsecretion of the healthy pancreas.

CONCLUSION

Increases in the bedtime doses of hypoglycemic agents with nighttime peaks in action may correct early morning hyperglycemia but be associated with undesirable nocturnal hypoglycemia. Targeted continuous subcutaneous insulin infusion programming can facilitate the prevention of early morning hyperglycemia in selected patients.

The role of the growth hormone-insulin-like growth factor axis in glucose homeostasis

Homeostatic mechanisms normally maintain the plasma glucose concentration within narrow limits despite major fluctuations in supply and demand. There is increasing evidence that the growth hormone (GH)-insulin-like growth factor (IGF) axis may play an important role in glucose metabolism. GH has potent effects on intermediary metabolism, some of which antagonize the actions of insulin. In contrast, IGF-I has insulin-like actions, which are, in the case of glucose metabolism, opposite to those of GH. There is often deranged glucose metabolism in situations where GH is deficient or in excess. The clinical administration of GH or IGF-I results in altered glucose metabolism and changes in insulin resistance. Despite these observations, the precise role of GH and IGF-I and their interactions with insulin in controlling normal glucose homeostasis are unknown. In diabetes, GH secretion is abnormally increased as a result of reduced portal insulin resulting in impaired hepatic IGF-I generation. Evidence suggests that this may contribute to the development of diabetic microvascular complications. IGF-I 'replacement' in diabetes is under investigation and new methods of delivering IGF-I as a complex with IGFBP-3 offer exciting new prospects.

Frequency of the dawn phenomenon in type 2 diabetes: implications for diabetes therapy

This study was designed to assess the frequency of the dawn phenomenon in patients with type 2 diabetes. A secondary aim was to examine the influence of varying treatment regimens on the frequency of the dawn phenomenon. The dawn phenomenon was defined as a rise in plasma glucose levels of > or = 0.5 mmol/L (10 mg/dL) between 0500 and 0900 h occurring after a growth hormone surge of > or = 5 microg/L. Sixteen subjects (six men, 10 women) with type 2 diabetes were studied overnight on their current mode of therapy in the General Clinical Research Center. Additionally, six of these subjects were restudied in random order after each of the following three therapeutic regimens: (1) 6 weeks of glipizide, (2) 6 weeks of bedtime NPH insulin, and (3) 3 days of intensive insulin therapy with multiple injections of regular insulin followed by assessment during overnight intravenous infusion of insulin. Thus, a total of 34 overnight studies were performed under various treatment conditions to provide an approximate frequency of the dawn phenomenon in type 2 diabetes. Blood was drawn every 30 min between midnight and 0800 h for measurement of glucose, insulin, C-peptide, and growth hormone levels. Additional counterregulatory hormone levels were determined during 24 of the studies, and the integrity of growth hormone secretion in response to insulin-induced hypoglycemia was assessed in 12 of the 16 patients. The subjects were aged 51 +/- 15 years with a body mass index of 31 +/- 5 kg/m(2) and a mean glycosylated hemoglobin of 8.1 +/- 1.2%. The dawn phenomenon occurred in only one of 34 (3%) studies. Moreover, the four different treatment regimens did not affect the frequency of occurrence of the dawn phenomenon. Ten of the 12 patients tested failed to secrete growth hormone in response to insulin-induced hypoglycemia. These data suggest that the dawn phenomenon is unusual in type 2 diabetes. Previously reported high prevalence rates in studies using similar sample size may be attributable to a Biostator-induced artifact. Decisions regarding therapies for type 2 diabetes should not be based on the assumption that the dawn phenomenon routinely causes early morning hyperglycemia.

Hypoglycemia. Pathophysiology and treatment

Hypoglycemia is a common consequence of many diabetes treatments. As is true for many therapies for diseases with major pathologic consequences, the benefits and risks of treatment must be balanced. In intensified diabetes management, hypoglycemia is not an insurmountable problem but is unfortunately inevitable using the methods of glucose control currently available. Patients with type 1 diabetes seem to be at greater risk than patients with type 2 disease. The health care team must strive to help the patient maintain normoglycemia. The results of the DCCT and the United Kingdom Prospective Diabetes Study prove that near normoglycemia is clearly in the patient's best interest. Patient education has become focused on minimizing hyperglycemia; counseling on the dangers of hypoglycemia has not been given the same stature. Emphasis must be placed on minimizing even minor subclinical hypoglycemia because it will contribute to a vicious cycle of hypoglycemia begetting hypoglycemia.

Severe hypoglycemia and reduction of insulin requirement in a girl with insulin-dependent diabetes mellitus: first sign of a craniopharyngioma

A girl with a history of insulin-dependent diabetes mellitus since 5.5 years, and Hashimoto's thyroiditis since 12 years, developed episodes of severe hypoglycemia from the age of 12 years. This was associated with falling insulin requirements, from 0.78 U/kg/day at 11 years to 0.34 U/kg/day at 16 years. At 16 years she was found to have GH, gonadotropin, ACTH, and probably also TSH deficiency with hyperprolactinemia. MRI scan revealed a cystic intrasellar craniopharyngioma with moderate suprasellar extension. In spite of cortisol replacement at 17 years, insulin requirement fell further to 0.25 U/kg/day at 18 years. In this girl, decreasing insulin requirements represented an early manifestation of combined growth hormone and cortisol deficiency.

Lower glycemic response to sucrose in the diets of children with type 1 diabetes

OBJECTIVE

To compare glycemic responses of isocaloric mixed meals containing 2% and 17% sucrose in children with type 1 diabetes who had fasting euglycemia.

STUDY DESIGN

Nine children (11 to 16 years) with type 1 diabetes were randomized in a crossover design to receive 2 isocaloric diets (2% or 17% sucrose) in the Clinical Research Center. In the 2% sucrose diet, starch isocalorically replaced sucrose.

RESULTS

Fasting euglycemia was comparable on both study days (mean +/- SEM: 2% sucrose, 5.0 +/- 0.3 mmol/L or 90 +/- 5 mg/dL; 17% sucrose, 5.0 +/- 0.3 mmol/L or 91 +/- 6 mg/dL). The 17% sucrose diet resulted in a lower glycemic response than the 2% sucrose diet over the 4-hour study period (area under glucose response curve: mean +/- SEM, 37 +/- 3.5 mmol/L x 4 h vs 42 +/- 4.7 mmol/L x 4 h, P = .01). Peak blood glucose response was earlier and lower (2.2 to 2.8 mmol/L, 40 to 50 mg/dL) with the 17% sucrose diet.

CONCLUSIONS

Sucrose in moderate amounts, isocalorically exchanged for starch, lowered glycemic response between breakfast and lunch in children who were euglycemic before breakfast. These data refute concerns about adverse glycemic effects of sucrose and support the use of sucrose-containing foods in the context of a healthy meal plan.

Two weeks of daily injections and continuous infusion of recombinant human growth hormone (GH) in GH-deficient adults: I. Effects on insulin-like growth factor-I (IGF-I), GH and IGF binding proteins, and glucose homeostasis

Recombinant human growth hormone (GH) is routinely administered as daily subcutaneous injections to patients with GH deficiency (GHD). However, in the hypophysectomized rat, pulsatile and continuous infusion of GH has been shown to differ in terms of the magnitude of effect on longitudinal bone growth, serum insulin-like growth factor-I (IGF-I) concentrations, and hepatic metabolism. The aim of the present study was to compare the effects of daily injections and continuous infusion of GH in GHD adults on previously well-documented GH-dependent factors. Recombinant human GH (0.25 U/kg/wk) was administered to nine men with GHD for 14 days in two different ways, ie, as a daily subcutaneous injection at 8 PM and as a continuous subcutaneous infusion, with 1 month of washout between treatments. Blood samples and tests were performed in the morning after an overnight fast before the start of GH treatment (day 0) and on day 2 and day 14 of treatment. An oral glucose tolerance test (OGTT) was performed on day 0 and day 14. Daily injections and continuous infusion of GH exerted similar effects in terms of body weight and body composition. The two modes of administration resulted in similar daily urinary GH excretion and similar serum GH concentrations in the morning. GH binding protein (GHBP) concentrations did not change significantly during the various treatment periods. Serum IGF-I and IGF-I binding protein (IGFBP)-3 concentrations increased to a greater degree during continuous infusion of GH versus daily injections. Serum IGFBP-I concentrations decreased to a similar degree during the two modes of administration. Serum concentrations of free triiodothyronine and total triiodothyronine (T3) increased and free thyroxine (T4) decreased to a similar degree, independent of the mode of administration. However, total T4 concentrations were unchanged during both modes of treatment. Serum thyrotropin (TSH) concentrations decreased during continuous infusion, and there was a similar nonsignificant decrease during daily injections of GH. Fasting free fatty acid (FFA) levels increased during treatment with only daily injection of GH, but there was no significant effect from continuous infusion. Results of measurements of fasting concentrations of blood glucose and oral glucose tolerance (OGT) indicated a more impaired glucose tolerance after daily injections of GH versus continuous infusion. In conclusion, continuous infusion and daily injections of GH have similar effects on the variables described, but the magnitude of the effects differs.

Characterization of the insulin-antagonistic effect of growth hormone in insulin-dependent diabetes mellitus

To characterize its insulin-antagonistic effect, growth hormone (GH) was infused at variable rates (24, 12 or 6 mU kg-1 min-1) for 1 h in 7 IDDM patients. Saline infusion was used as control (C) and all patients participated in all studies. The effect of insulin was measured with the euglycaemic clamp technique for 6 h combined with d-(3-3H)-glucose to evaluate glucose turnover. The insulin levels during the clamps were similar in all studies (23 +/- 3 mU l-1). The infusions produced peak GH levels of (24 rate = 24) 157 +/- 11, (12 rate = 12) 76 +/- 7, and (6 rate = 6) 45 +/- 8 mU l-1 (mean +/- SEM). The insulin-antagonistic effect of GH on glucose uptake was seen after 2 h and was at a maximum 4 to 5 h after the start of the GH infusion (difference in glucose infusion rate between C and 24 was 1.7 +/- 0.4 mg kg-1 min-1, p < 0.01). The resistance was due to a less pronounced effect of insulin to both inhibit rate of appearance and to stimulate rate of disappearance. Infusion of GH at 12 mU kg-1 min-1 induced a less pronounced insulin resistance both with regards to maximal effect (glucose infusion rate C - GH 1.4 +/- 0.5 mg kg-1 min-1, p < 0.05) and duration (3 h). At 6 mU kg-1 min-1, a clear GH-induced insulin-antagonistic effect was only seen during the third hour of the clamp (glucose infusion rate C-GH 1.3 +/- 0.5 mg kg-1 min-1, p < 0.05). GH infusion impaired the effect of insulin to lower both the levels of free fatty acids (NEFA) and glycerol between 2 and 5 h after the start of the infusion (NEFA, C:110 +/- 29, 24:303 +/- 95, p < 0.05: glycerol, C:32 +/- 4, 24:50 +/- 7 mumol l-1, p < 0.05). The present study therefore demonstrates that the insulin-antagonistic effect of GH in IDDM is related to the plasma levels both with regard to duration and response. The results also indicate that GH impairs the effect of insulin on lipolysis in IDDM after physiological peaks.

Growth hormone-deficient adults are insulin-resistant

Patients with growth hormone deficiency (GHD) have traditionally been described as having increased insulin sensitivity with a tendency toward fasting hypoglycemia, at least in children. In other studies, impaired glucose tolerance has been found. To evaluate basal insulin sensitivity, a hyperinsulinemic, normoglycemic clamp was performed with an insulin rate of 40 mU/m2/min after an overnight fast. Fifteen patients (four women and 11 men aged 20 to 62 years) with GHD for at least 1 year were compared with 15 healthy controls matched for sex, age, and body mass index (BMI). Thirteen patients had complete pituitary deficiency and were being treated with conventional hormone replacement therapy. Two men had isolated GHD since childhood. Four men were being treated with bromocriptin. There were no significant differences between fasting blood glucose (4.4 +/- 0.1 v 4.7 +/- 0.2 [mean +/- SEM] mmol/L) or fasting plasma insulin (9.5 +/- 1.4 v 8.8 +/- 1.1 mU/L) in patients and controls, respectively. Fasting free fatty acid (FFA) levels were lower in patients (444 +/- 35 v 796 +/- 94 mumol/L, P < .01). Blood glucose levels during the clamp were similar (4.6 +/- 0.1 v 4.9 +/- 0.1 mmol/L), as were insulin levels (81 +/- 4 v 93 +/- 4 mU/L). A decrease in glucose infusion rate (GIR) was seen during the clamp in GHD subjects (3.9 +/- 0.5 v 9.9 +/- 0.7 mg/kg body weight/min) as compared with controls (P = .001). Even if corrections were made for body fat, there was a significant difference (GIR corrected per lean body mass, 5.8 +/- 0.8 v 13.9 +/- 0.9 mg/kg lean body mass/min, P < .001). The results suggest that adults with GHD are insulin-resistant. Despite this finding, normal fasting plasma insulin levels were seen.

Diminished brain glucose metabolism is a significant determinant for falling rates of systemic glucose utilization during sleep in normal humans

Systemic glucose utilization declines during sleep in man. We tested the hypothesis that this decline in utilization is largely accounted for by reduced brain glucose metabolism. 10 normal subjects underwent internal jugular and radial artery cannulation to determine cerebral blood flow by N2O equilibrium technique and to quantitate cross-brain glucose and oxygen differences before and every 3 h during sleep. Sleep stage was graded by continuous electroencephalogram, and systemic glucose turnover was estimated by isotope dilution. Brain glucose metabolism fell from 33.6 +/- 2.2 mumol/100 g per min (mean +/- SE) before sleep (2300 h) to a mean nadir of 24.3 +/- 1.1 mumol/100 g per min at 0300 h during sleep (P = 0.001). Corresponding rates of systemic glucose utilization fell from 13.2 +/- 0.8 to 11.0 +/- 0.5 mumol/kg per min (P = 0.003). Diminished brain glucose metabolism was the product of a reduced arteriovenous glucose difference, 0.643 +/- 0.024 to 0.546 +/- 0.020 mmol/liter (P = 0.002), and cerebral blood flow, 50.3 +/- 2.8 to 44.6 +/- 1.4 cc/100 g per min (P = 0.021). Brain oxygen metabolism fell commensurately from 153.4 +/- 11.8 to 128.0 +/- 8.4 mumol/100 g per min (P = 0.045). The observed reduction in brain metabolism occurred independent of stage of central nervous system electrical activity (electroencephalographic data), and was more closely linked to duration of sleep. We conclude that a decline in brain glucose metabolism is a significant determinant of falling rates of systemic glucose utilization during sleep.