Plasma somatomedin activity and diabetic retinopathy

Diagnosis of acromegaly in a patient with type 1 diabetes mellitus

OBJECTIVE

To describe a case of acromegaly in a young patient with poorly controlled diabetes mellitus and suggest guidelines to distinguish acromegaly and high growth hormone (GH) levels previously reported in poorly controlled type 1 diabetes.

METHODS

We present a detailed case report, including clinical findings and serial laboratory results in a patient with type 1 diabetes and a GH-secreting pituitary tumor.

RESULTS

A 28-year-old woman with type 1 diabetes underwent assessment for secondary amenorrhea and worsening glycemic control. A low estradiol level and an inappropriately low level of follicle-stimulating hormone prompted magnetic resonance imaging of the head, which demonstrated a pituitary adenoma. Subsequent endocrine investigation revealed a high insulin-like growth factor I (IGF-I) level (849 mg/L; normal range, 122 to 400). The concentration of insulin-like growth factor-binding protein-3 (IGFBP-3) was also elevated (5.5 mg/L; normal range, 2.0 to 4.2). GH levels measured during episodes of spontaneous hyperglycemia (>180 mg/dL) were in the range of 3 to 5 ng/mL and failed to suppress to below 2 ng/mL after a bromocriptine suppression test. The patient underwent transsphenoidal resection of a pituitary tumor, which stained positively for GH by immunohistochemistry. Postoperatively, glycemic control improved, with decreased fluctuations of hypoglycemia and hyperglycemia, despite a decrease in insulin requirements.

CONCLUSION

This report highlights the difficulty in interpreting GH and IGF-I levels in patients with type 1 diabetes. In addition, a detailed review of the literature suggests that IGFBP-3 measurements may be helpful in confirming the diagnosis of concurrent acromegaly and distinguishing it from high GH levels attributable to poor control of diabetes.

Insulin-like growth factors and binding proteins in patients with recent-onset type 1 (insulin-dependent) diabetes mellitus: influence of diabetes control and intraportal insulin infusion

Type 1 diabetes mellitus is associated with decreased insulin-like growth factor-1 (IGF-1) levels, enhanced values of growth hormone (GH) and IGF-binding protein 1 (IGFBP-1). Since the liver is the major source of IGF and IGFBP production, we have therefore examined whether levels of IGFs (IGF-1 and IGF-11) and IGFBPs (IGFBP-1 and IGFBP-3) differ when insulin is infused into the portal or peripheral vascular system. IGF, IGFBP, and GH levels were determined within 1-3 weeks of diagnosis in 36 patients (ranging in age from 18 to 22 years) with Type 1 diabetes mellitus. IGF-1 levels were low before insulin therapy administration (0.49 +/- 0.05 vs. 1.11 +/- 0.04 U/ml in controls, P < 0.01). With insulin treatment, IGF-1 levels rose to the normal range and IGF-1 normalisation depended on diabetes control and the route of insulin infusion. Diabetic patients with conventional insulin therapy (CIT; n = 12) had low IGF-1 (0.57 +/- 0.07 U/ml) compared with patients with continuous subcutaneous insulin infusion (CSII; n = 12; 0.75 +/- 0.08 U/ml; P < 0.05) and intraportal insulin infusion (IPII; n = 12; 1.07 +/- 10.05 U/ml; P < 0.05). Significant correlations were found between IGF-1 and parameters of glycemic control: HbA1c (r = -0.64; P < 0.01) and glycemia (r = -0.56; P < 0.05). The pattern of changes in IGF-11 levels was not significantly different from that of controls and was not altered by insulin therapy (0.98 +/- 0.08 and 1.01 +/- 0.04 U/ml in controls). Measured fasting 08:00 h IGFBP-1 levels were elevated 3-fold and IGFGP-3 levels were 2-fold lower in diabetic patients than in controls. Elevated IGFBP-1 levels were significantly correlated with metabolic control (glycemia, r = 0.64, P < 0.01; HbA1c, r = 0.71, P < 0.01). The mean elevated GH level before insulin administration (13.4 +/- 0.9 mg/l) was decreased by intensified insulin therapy (CSII, 8.8 +/- 0.6, P < 0.05; IPII, 5.6 +/- 0.9 mg/l, P < 0.001). There was a negative correlation between GH and IGF-1 (r = -0.72, P < 0.01). These results show the role of glycemic control and the route of insulin administration in the normalisation of IGF-1, IGFBP-1 and GH up to non-diabetic controls in patients with recent-onset Type 1 diabetes mellitus.

[Principles and clinical significance of insulin-like growth factors/somatomedins]

The Insulin-like Growth Factors (IGFs) or Somatomedins are polypeptide growth factors which are similar to insulin in respect to their aminoacid sequence, structure and biologic activities. The IGFs bind to high affinity receptors which are present on many cells and in many tissues. In the circulation the IGFs are bound to transport (binding) proteins (IGF-BPs). In this review the physiologic role, the basic chemistry and the gene expression of this family of growth factors is summarized systematically. The pathophysiology of growth disorders, diabetes mellitus, malnutrition, liver and kidney disease in relation to the IGFs as well as the therapeutic and diagnostic potentials of these peptides are discussed in detail.

Growth hormone and the microvascular complications of diabetes

The basis for the hypothesis that growth hormone is a permissive factor in the pathogenesis of diabetic microvascular complications is a weak one. The best way forward in this research will be to devise a pharmacological method of suppressing growth hormone secretion in diabetic subjects.

Circulating insulin-like growth factor I in type 1 (insulin-dependent) diabetic patients with retinopathy

The relationship between insulin-like growth factor I (IGF I) and diabetic retinopathy was investigated. This somatomedin circulates bound to at least two large carrier proteins with molecular weights of approximately 150,000 and 35,000. Total and protein binding profiles of insulin-like growth factor I were determined in the serum of 18 patients who had had Type 1 (insulin-dependent) diabetes for 15-20 years, but had no signs of nephropathy and a similar degree of mild subclinical neuropathy. Nine had preproliferative or proliferative retinopathy and 9 had little or no background retinopathy but there was no difference in diabetes duration, insulin doses or glycaemic control between the two groups. In the latter group, the amounts of the somatomedin I and the serum profiles were similar to those in 9 healthy control subjects. In patients with advanced retinopathy, however, binding of insulin-like growth factor I to the carrier proteins was significantly altered. Binding to the low molecular weight protein increased to 140% whereas binding to the large molecular weight protein decreased to 70% of the normal level. In the latter Type 1 diabetic patients total serum insulin-like growth factor I was decreased to 60% of the normal level (p less than 0.02). When the alteration in serum profile was adjusted for, the level of somatomedin associated with the small carrier complex was normal whereas that associated with the large carrier complex was reduced by almost 60% in Type 1 diabetic patients with retinopathy. It is proposed that the total circulating somatomedin level is low in advanced diabetic retinopathy.(ABSTRACT TRUNCATED AT 250 WORDS)

Human diabetic serum does not stimulate bovine retinal endothelial cell growth in culture

Bovine retinal microvascular endothelial cells were cultured with serum from nondiabetic controls, patients with background, preproliferative, proliferative and inactive (treated) proliferative diabetic retinopathy. Using Coomassie Blue to measure cellular protein, we found no significant increase in retinal endothelial cell growth with serum from patients with preproliferative or proliferative retinopathy.

Somatomedin-C (SM-C). Study in diabetic patients with and without retinopathy

In the present study we evaluated somatomedin-C (Sm-C) plasma levels in diabetic patients, with and without retinopathy. One hundred and thirty four diabetic patients (65 type I and 69 type II) and 90 controls, strictly matched for age and sex, were enrolled in the study. Ophthalmoscopy and fluorescein angiography allowed to distinguish: 49 patients without retinopathy, 45 patients with background retinopathy, and 40 with proliferative retinopathy. Growth hormone (GH) and Sm-C plasma levels were measured using a pool of 20-24 blood samples over 24h. Sm-C levels in type I (0.62 +/- 0.11 U/ml) and type II (0.56 +/- 0.09 U/ml) patients were significantly decreased (p less than 0.01) when compared to controls (0.89 +/- 0.30 U/ml). The mean daily secretion of GH was significantly (p less than 0.01) greater in diabetic patients (7.8 +/- 2.6 ng/ml) than in controls (4.1 +/- 1.5 ng/ml), but no correlation was found between Sm-C and GH (r = 0.15; p = n.s.). Our findings did not show any correlation between Sm-C plasma levels and either the existence of retinopathy, regardless of the degree of microvascular damage, or duration of the disease, or degree of metabolic control, as evaluated by HbA1c.

Somatostatin analogues in diabetes mellitus

Growth hormone (GH) has long been considered to have importance in diabetes. With poor control in Type 1 diabetes GH levels are high and may aggravate poor metabolic control. Pharmacological suppression of GH release at this stage might reverse the metabolic changes, with the possible added benefit of lower plasma insulin concentrations. Diabetic patients with life-long GH deficiency rarely develop retinopathy, while pituitary ablation in patients with retinopathy often leads to improvement. Growth hormone release inhibiting factor, somatostatin, has a short plasma half-life, and multiple effects on the endocrine system and on the gastrointestinal tract, making it unsuitable for clinical use as a GH suppressant. Long-acting analogues have a long half-life, but remain non-specific in their effects. In Type 2 diabetes the analogue Octreotide suppresses insulin and glucagon release, leaving glucose levels either unchanged or somewhat elevated. Gastrointestinal side-effects have been common, but may diminish with long-term treatment. In Type 1 diabetes insulin requirement is decreased by Octreotide, but as in Type 2 diabetes GH suppression has been observed consistently only when the drug was given at bed-time. The decrease in insulin requirement may reflect suppression of glucagon release and/or gut effects. Amelioration of the 'dawn phenomenon' has not proved possible, and hypoglycaemia has proved a particular problem with Octreotide given subcutaneously at night. The lack of effective GH suppression (particularly in patients with proliferative retinopathy), lack of specificity, and the gut and hypoglycaemic side-effects, argue strongly against a clinical role for the current somatostatin analogues in diabetes mellitus.

Serum IGF-1 concentration in diabetic retinopathy

Growth factors such as growth hormone and insulin-like growth factor 1 (IGF-1) may be important in the pathogenesis of diabetic retinopathy. We measured serum IGF-1 in 371 diabetic patients attending a diabetic retinopathy clinic and in 73 non-diabetic control subjects. No significant difference was observed in IGF-1 level between the diabetic and control groups (168 +/- 3.9 vs 177 +/- 7.4 micrograms/l [mean +/- SE]). Within the diabetic group, there was no difference between patients with no retinopathy and those with proliferative change (198.7 +/- 8.8 vs 190.5 +/- 11 micrograms/l). After adjusting for differences in age, duration of diabetes, and presence of proteinuria, only the inactive previously proliferative group showed any significant difference from the other patient subgroups (151.8 +/- 11.5 micrograms/l; p less than 0.05). Serum IGF-1 correlated with age in the control group (r = 0.49; p less than 0.001) and to a lesser extent in the diabetic group (r = -0.23; p less than 0.05). IGF-1 levels were higher in patients with proteinuria than in those without proteinuria (196.8 +/- 10.3 vs 138.8 +/- 4.4 micrograms/l; p less than 0.001).

Insulin exerts metabolic and growth-promoting effects by a direct action on the liver in vivo: clarification of the functional significance of the portal vascular link between the beta cells of the pancreatic islets and the liver

The functional significance of the portal vascular link between the beta cells of the pancreatic islets and the liver has not been established. Previous studies indicated that insulin does not acutely regulate glucose metabolism by a direct hepatic effect. More recent observations suggest that the role of insulin in regulating body growth may be mediated, at least in part, by the liver. Our experiments were designed to test whether insulin can promote body growth and regulate glucose metabolism by a direct hepatic action in vivo. Rats were made diabetic by injections of streptozotocin, and insulin or solvent was infused into the jugular vein (JV) or the hepatic portal vein (HPV) for 14 days using catheters that were attached to osmotic minipumps. Infusion of a low dose of insulin (2 units per kg per day) into the JV had no effects on the hyperglycemia, body weight gain, tail growth, tibial epiphysial cartilage plate thickness, or serum levels of somatomedin C in the diabetic rats. However, the same dose given into the HPV caused a 30% reduction of blood glucose and stimulated a significant degree of growth, as determined by all indices. Infusion of a higher dose of insulin (5 units per kg per day) into either vein caused full restoration of body weight gain and tail growth and it restored the glycemic status almost to normal. However, it did not increase the tibial epiphysial plate width or serum somatomedin C levels above those of the rats given the low dose of the hormone into the HPV. These results indicate that insulin can act directly on the liver to promote body growth and to regulate glucose metabolism. The significance of direct delivery of insulin from the pancreatic beta cells to the liver may be as much for growth control as for glucose homeostasis.

Increased circulating radioreceptor-active growth hormone in insulin-dependent diabetics

A radioreceptor assay (RRA) for GH, using pregnant rabbit liver membranes, was used to evaluate the receptor-hormone interaction of circulating (serum) GH in diabetics. Multiple morning blood samples were taken from 18 fasted insulin-dependent diabetics on three occasions and assayed for GH by the RRA and radioimmunoassay (RIA). Samples from 12 normal subjects (GH secretion stimulated by exercise and hypoglycaemia) and 11 acromegalics were also studied. The RRA/RIA ratio was higher in samples from the acromegalics than in normal subjects (0.85 vs 0.65, P less than 0.01) and only a few samples had RRA values greater than RIA. All the samples from 14 of the diabetics had lower RRA values than RIA, but 40% (39 of 96) of samples from the remaining four diabetics had RRA concentrations markedly in excess of the levels measured by RIA. Sephadex chromatography (G100) of serum samples revealed similar proportions of immunoreactive GH forms in the diabetics with high RRA values compared with the diabetics with low RRA values. These findings may indicate intermittent secretion of highly bioactive GH in some insulin-dependent diabetics. Further studies on the biological properties of circulating GH are needed to clarify the role of GH secretion in the development of diabetic microvascular complications.

Effects of somatostatin and SMS 201-995 on carbohydrate metabolism in normal man

Growth hormone (GH) is important in diabetes in view of its anti-insulin actions and its relation to the long-term complications of the disease. The suppression of GH secretion in diabetics has theoretical and possible therapeutic interest. Native somatostatin has multiple actions, including inhibition of the secretion of insulin, glucagon, thyroid-stimulating hormone (TSH), and various gut hormones. It also has inhibitory effects on gastrointestinal motility, exocrine secretion, nutrient absorption, and splanchnic blood flow. Its therapeutic use is limited by a duration of effect of several minutes only. SMS 201-995 holds more potential than native somatostatin in view of its longer duration of action. Preliminary data suggest that 50 micrograms SMS 201-995 subcutaneously at night inhibits the nocturnal rise in GH secretion in normal man, but no effect on 24-h GH secretion is observed when SMS 201-995 is injected twice daily before meals. SMS 201-995 inhibits secretion of insulin, glucagon, and TSH in addition to growth hormone and induces carbohydrate intolerance when administered before food in normal subjects. Gastrointestinal side effects suggest additional effects on nutrient disposal, which are important when it is administered before food. Further studies are required to elucidate these effects of SMS 201-995 on endocrine and gastro-intestinal function in normal and diabetic man.

Rationale for inhibition of growth hormone secretion in the management of the diabetic patient

Patients with diabetes mellitus, especially the insulin-dependent variety, have increased circulating levels of growth hormone. On the basis of currently available information, the potential advantages of inhibition of growth hormone secretion as an adjunct in the treatment of diabetes mellitus include improved metabolic control (less hyperglycaemia, greater stability), resulting from diminution of the insulin-antagonistic actions of this hormone, and reduced micro- and possibly macro-angiopathy, resulting not only from improved metabolic control but also from decreased direct effects of growth hormone on blood vessels.

Somatostatin analogue SMS 201-995 in type I diabetes mellitus. Initial experience after repeated administration

We have recently obtained encouraging short-term results after a single subcutaneous injection of the long-acting somatostatin analogue SMS 201-995 in acromegalic patients. Increased growth hormone (GH) levels may be involved in the pathogenesis of proliferative retinopathy in type I diabetes mellitus. In this study we thus investigated the effect of 3 X 50 micrograms SMS 201-995 daily on the metabolic control and hormone secretion of eight type I diabetics over a 3-day period. GH levels decreased by 32% (p less than 0.05) and somatomedin C levels by 31% (p less than 0.01) on the 3rd day of treatment compared with a control day. The insulin requirements during conventional subcutaneous insulin therapy were reduced by 28% (p less than 0.01) in seven patients without deterioration of metabolic control (mean blood glucose levels, 153.8) versus 154.7 mg/dl). Triiodothyronine, thyroxine, glucagon, prolactin, luteinizing hormone and follicle-stimulating hormone showed no significant changes. We conclude that SMS 201-995 could be an excellent tool for further clinical investigation and therapy of diabetic vascular complications.

Abnormal retinal artery responses to stress in patients with type I diabetes

The brachial artery pressure and retinal artery pressure responses to a one-minute cold pressor test were evaluated simultaneously in 14 patients with type I diabetes mellitus (six with and eight without diabetic retinopathy) and 10 age-matched control subjects. Five patients with type I diabetes had autonomic neuropathy. Mean baseline brachial artery pressure and retinal artery pressure were similar in patients with type I diabetes and control subjects. After cold pressor testing, the brachial artery pressure increased significantly (p less than 0.01) compared with baseline values in both groups. Retinal mean arterial pressures increased significantly (p less than 0.001) after cold pressor testing compared with the baseline values only in patients with type I diabetes. Positive correlation was found between the brachial and retinal mean arterial pressures after cold pressor testing (r = 0.48; p less than 0.05) in the diabetic patients but not in the control subjects (r = 0.10; p = NS). No correlation was found between the retinal artery pressure and age of onset of diabetes, duration of diabetes, the presence or absence of diabetic retinopathy, and glycemic control. Four patients with autonomic neuropathy and low retinal artery pressures, which remained unchanged after cold pressor testing, had no diabetic retinopathy. The fifth patient with autonomic neuropathy and exaggerated systolic brachial artery pressure (175 mm Hg) and retinal artery pressure (more than 80 mm Hg) responses had severe background diabetic retinopathy. In conclusion, abnormal retinal artery responses to stress are present in patients with type I diabetes. This may be modified by the presence or absence of both autonomic neuropathy and hypertension. The biologic significance of these findings is yet to be determined.