Hyponatremia and polyuria in children with central diabetes insipidus: challenges in diagnosis and management.

Five patients with well-controlled, long-standing, central diabetes insipidus had acute development of dehydration, hyponatremia, and inappropriate natriuresis in the setting of polyuria resistant to exogenous antidiuretic hormone. Hyponatremia and dehydration worsened with fluid restriction or use of exogenous antidiuretic hormone. We discuss the challenges in diagnosis and management of probable salt wasting in children with central diabetes insipidus.

Insulin-like growth factor binding protein-1 levels in the diagnosis of hypoglycemia caused by hyperinsulinism.

The diagnosis of hypoglycemia caused by hyperinsulinism may be difficult because insulin levels are not uniformly elevated at the time of hypoglycemia. Insulin-like growth factor binding protein-1 (IGFBP-1) is a 28 kd protein whose secretion is acutely inhibited by insulin. We hypothesized that serum levels of IGFBP-1 would be a useful marker of hyperinsulinism. We measured IGFBP-1 levels during the course of standardized fasting studies in hospitalized children; 36 patients became hypoglycemic during the fasting studies, and samples obtained at the point of hypoglycemia were analyzed. On the basis of the currently used diagnostic criteria, 13 children had hyperinsulinism, 16 had ketotic hypoglycemia or no disorder, 3 had hypopituitarism or isolated growth hormone deficiency, 2 had glycogen storage disease type 1 and 2 had fatty acid oxidation disorders. In control subjects (children with ketotic hypoglycemia or no disorder), IGFBP-1 levels rose during fasting to a mean of 343.8 +/- 71.3 ng/ml in the sample drawn at the time of hypoglycemia. Mean IGFBP-1 levels at hypoglycemia for the entire group with hyperinsulinism were 52.4 +/- 11.5 ng/ml, significantly different from levels seen in control subjects (p < 0.0001). In children with moderately controlled hyperinsulinism (fasting tolerance > 4 hours), mean IGFBP-1 levels at the time of hypoglycemia were 71.5 +/- 16.9 ng/ml. IGFBP-1 levels in the children with poorly controlled hyperinsulinism (fasting tolerance < 4 hours) failed to rise during fasting, with a mean of 30.1 +/- 10.4 ng/ml in the final sample. IGFBP-1 levels were inversely correlated with serum insulin and C-peptide levels (r = -0.71 and -0.72, respectively; p < 0.0001). Patients with other endocrinologic or metabolic diseases that result in fasting hypoglycemia demonstrated a rise in IGFBP-1 levels similar to that seen in ketotic hypoglycemia. Low serum levels of IGFBP-1 at the time of hypoglycemia provide an additional marker of insulin action that might help to differentiate hyperinsulinism from other hypoglycemic disorders.