A decade of insulin infusion pumps.

Over the past 10 years there have been many changes in the details of diabetic management. The development of home monitoring for capillary blood glucose and widespread availability of measurements of glycosylated haemoglobin have provided direct evidence that most diabetic children have poor or very poor glycaemic control. Additionally, there has been further evidence in support of the hypothesis that poor metabolic control predisposes susceptible diabetic patients to the microvascular and macrovascular complications of the disease. Frustration that glycaemic control has not improved with regimens of one, two, or more recently, three2 daily injections using the new, pure, and latterly, human insulins has focused attention on alternative ways of delivering insulin in the hope of achieving improved glycaemic control. In 1978 continuous subcutaneous insulin infusion driven by a small battery powered portable pump was introduced as a research tool for improving glycaemic control in adults with insulin-dependent diabetes mellitus.3 The infusion system provided a continuous supply of unmodified short acting insulin throughout 24 hours with supplementary boluses of insulin initiated by the patient and given before meals. The developments coincided with the introduction of other 'tools' of management such as better education, the establishment of diabetic liaison health visitors, and an improved understanding of the role of diet and exercise in the management of diabetes.4 Portable insulin infusion pumps seemed to be a timely technological advance. Early studies of the euglycaemic effects of infusion pumps were encouraging. s-x Have the hopes engendered by these early favourable reports been fulfilled? Has the efficacy of insulin infusion systems under conditions of tightly controlled clinical trials in specialised centres been successfully translated into general clinical practice? Put more specifically: (1) Is insulin infusion an effective regimen that can be applied to large numbers of unselected children with insulin-dependent diabetes mellitus ranging from infancy, through school age, to adolescence, and young adulthood? (2) How does insulin infusion compare in terms of blood glucose control with one, two, or multiple daily injection regimens? (3) Are the indications for and complications of insulin infusion clearer? (4) Does insulin infusion reduce the incidence of the microvascular and macrovascular complications of diabetes?