The insulin release after an oral glucose load is both earlier and greater than would be expected from the glycemic stimulus. This augmentation of insulin release has been attributed to humoral factors from the gut. It has been previously demonstrated that secretin is released very rapidly after oral glucose and postulated that it acts as an early trigger to insulin release. This effect would not explain the magnitude of the peak insulin response which occurs about 30 min after peak secretin levels. The present studies, however, demonstrate an additional action of secretin which may explain this. To further study the role of secretin in insulin release in normal subjects, two consecutive 20 min intravenous glucose infusions were administered 150 min apart with and without an intervening secretin infusion (10 U) given to approximate serum secretin levels seen after oral glucose ingestion. A highly significant (P<0.01) potentiation of the insulin response to the post-secretin glucose infusion was observed. This occurred both when secretin was given 7 min or 25 min before glucose. In the latter case, serum secretin was undetectable during the glucose infusion. These studies demonstrate that secretin potentiates the glycemic release of insulin. Despite the augmented insulin response, no consistent change in blood glucose variation was observed. This is consistent with the suggestion that the facilitated disposal of an alimentary glucose load is not dependent solely on enhanced insulin secretion. Secretin appears to have a dual role in insulin release, an early direct stimulation followed by a prolonged potentiation of the glycemic stimulus. The potentiating effect is of such magnitude to suggest that secretin is the dominant factor in the enteric component of insulin release after an oral glucose load.

The tendency of insulin to form insoluble aggregates is a major obstacle to the development of implantable insulin infusion systems for treatment of insulin-deficient diabetic patients. A test system was developed to examine the kinetics of insulin aggregation under controlled conditions of temperature, vibration and contact material in an effort to provide design criteria for minimising aggregation. The contact materials tested were all potentially suitable for pump reservoirs on engineering criteria and included metals (stainless steel, titanium and a titanium alloy) and various plastics (polypropylene, polytetrafluoroethylene, polyvinylchloride, polyamide, cellulose butyrate and silicone elastomer). The rate of insulin aggregation was markedly affected by the nature of the contact material. Hydrophilic materials, particularly polyamide and cellulose butyrate (2% of total insulin aggregated after 96 h vibration), appeared more compatible with insulin stability than did hydrophobic ones, such as polypropylene (16% aggregation) and polyvinylchloride (37% aggregation). A specially formulated 'pump' insulin preparation, stabilised by addition of polyethylenepolypropyleneglycol, was significantly superior (three to five times more stable) to a regular neutral insulin formulation under most, but not all, conditions. Standard clinical syringes (polypropylene) performed poorly with both insulin formulations but especially with the neutral regular insulin (100% aggregation after 96 h vibration). In addition to physical aggregates, significant amounts (5%-30%) of the insulin remaining in solution were no longer detectable by immuno- or receptorassay in all materials tested.(ABSTRACT TRUNCATED AT 250 WORDS)

SummaryIn the use of low-level intravenous insulin infusion for treating diabetic hyperglycaemia and ketoacidosis adsorption of insulin to containers or plastic infusion apparatus results in significant losses of 60-80% of insulin in dilute physiological saline solution (40 U/1). It is therefore necessary to add protein to the carrier solution to minimize losses and maintain a constant delivery rate. Recovery studies showed that 3-5% w/v polygeline solution (polymer of degraded gelatin) was a suitable medium for this purpose, offering some advantages over human serum albumin. A minimum concentration of 0 5% polygeline was required to ensure adequate delivery of insulin to the patient.

Summary A radioimmunoassay method is described for the simultaneous drtermination of insulin and glucagon in human serum. The technique uses 125I and 131labelled hormones, specific antisera of negligible cross reaction and an anion exchange resin to separate antiserum‐bound and “free” labelled hormones. Evidence is presented that there is non‐specific protein binding of both labelled insulin and labelled glucagon in human sera, and a correction for this is described. The method is suitable for the rapid assay of a large number of serum samples and fulfils the necessary criteria of specificity, sensitity, accuracy and reproducibility for clinical use.