Subcutaneous insulin therapy - end of the road after 80 years?

Subcutaneous (SC) insulin therapy has been a mainstay of pharmacological diabetes management from the moment insulin was successfully developed as treatment. Insulin formulations have become more refined and less allergenic over time, and ancillary technologies such as injection devices and glucose measurement tools have evolved to the extent of permitting closed-loop therapy. However, investigations have continued exploring alternative routes of administration with the ultimate goal of implantable islet replacements, whether cell- or "silicon"-based. Progress on these lines of research, however, has been slow to present patients with viable options: alternative delivery routes have failed to deliver insulin reliably and with commercially viable efficiency, while beta cell transplantation continues to struggle with tissue availability and in vivo viability. In the meantime, SC insulin formulations have advanced for rapid- and long-acting formulations, to better meet typical insulin requirements across the day. Thus, SC insulin will likely remain a key technology for the foreseeable future in order to address the needs of an ever larger number of insulin-dependent patients with diabetes.

Shift from surrogate end point to outcome trials: implications for cardiovascular safety assessment in development programs for antidiabetic drugs.

We assessed the effect of extending the range of cardiovascular (CV) end points to include hospitalization for unstable angina and hospitalization for coronary revascularization (Extended Major Adverse Cardiac Event criteria (MACE)) in addition to the standard ones, namely, CV-related death, nonfatal stroke, and nonfatal myocardial infarction (Core MACE). The trials selected for the analysis had a duration/follow-up period of ≥1 year and involved more than 1,000 subjects. Annual event rates (AERs) for Core MACE in patients with type 2 diabetes were estimated, and the duration of an event-driven CV outcome trial necessary to exclude ≥80% risk increase was modeled. All the studies revealed hazard ratios ≤1.0 for Core MACE end points whereas in 21% of the studies, the hazard ratio for hospitalization for unstable angina or coronary revascularization (Extended MACE) was >1 and was therefore discordant with Core MACE. The AERs for Core MACE ranged from 0.5% (recent clinical programs) to 6% (epidemiological studies); these low rates observed in recent programs would have the effect of increasing the duration required for a CV outcome trial. The addition of Extended MACE end points to the primary composite outcome in antidiabetic clinical trials is unlikely to obscure CV-related risk and may improve the feasibility of CV outcome trials.

Uterine effects of raloxifene in comparison with continuous-combined hormone replacement therapy in postmenopausal women.

OBJECTIVE: We sought to compare the uterine effects of raloxifene with those of continuous-combined hormone replacement therapy. STUDY DESIGN: This randomized, double-blind 24-month study involved 136 postmenopausal women who received raloxifene 150 mg/d or conjugated equine estrogens 0.625 mg/d with medroxyprogesterone acetate 2.5 mg/d. After baseline evaluations, endometrial biopsy specimens were obtained, and endometrial thickness was measured annually by means of transvaginal ultrasonography. Statistical analyses were performed with an intention-to-treat approach. RESULTS: In the raloxifene group at the end point of the study 94.4% of biopsy specimens showed normal benign postmenopausal endometrium and 5.6% were classified as benign stimulatory endometrium. In the continuous-combined hormone replacement therapy group at the end point of the study 78.7% of biopsy specimens showed normal benign postmenopausal endometrium, 19. 1% were classified as benign stimulatory endometrium, and 2.1% showed benign abnormal postmenopausal endometrium. Mean endometrial thickness was unchanged from baseline with raloxifene and was increased significantly by 0.5 mm at 12 months with continuous-combined hormone replacement therapy. CONCLUSION: Raloxifene 150 mg/d did not increase endometrial thickness or cause endometrial proliferation in healthy postmenopausal women.

Stability of insulin lispro in insulin infusion systems.

OBJECTIVE: To test stability of insulin lispro in two insulin infusion systems over 48 h. RESEARCH DESIGN AND METHODS: We used reverse-phase and size-exclusion high-performance liquid chromatography (HPLC) to determine the purity, potency, and degree of polymerization of U100 insulin lispro (Humalog) after 24- and 48-h pump cycles conducted at 37 degrees C in five Disetronic H-TRON V100 and five MiniMed 504 pumps. Pumps were set to deliver a basal rate of 0.5 U/h and 6-U boluses at t = 0, 4, 8, 24, 24.5, 28.5, 32.5, and 48 h during each cycle. The effluent was collected into 1-ml vials, pooled at 24 or 48 h, and stored at 4 degrees C until assay. After each 48-h run period of insulin delivery, assays for potency, polymer, and purity were performed on the pooled samples from each individual cycle. m-cresol content and the pooled reservoir content were assayed in the 48-h pooled samples. RESULTS: Insulin lispro retained full HPLC potency (delta < or = 4%) at 48 h, with no degradation of insulin lispro to des-amidoinsulin forms (24 or 48 h). No increase in pumped insulin polymer concentration was observed following 24 h of pump flow. Nonsignificant increases of < or =0.09% (Disetronic) and < or =0.15% (MiniMed) from initial concentrations of 0.18% (polymer divided by total insulin) were detected in three of five pump cycles at 48 h when compared with 37 degrees C paired controls. Nonsignificant decreases (<5 and 10%, Disetronic and MiniMed, respectively) of m-cresol content occurred in both systems following 48 h storage in each device, but sterility was not compromised by this decrease (initial m-cresol concentration, 3.15 mg/ml). Pump performance was without mechanical or electrical fault throughout the study Basal and bolus insulin delivery was evaluated three times daily and remained as expected. Occlusion of catheters by insulin precipitation did not occur, and no change in pH was observed following delivery. CONCLUSIONS: We conclude that insulin lispro is suitable for prolonged infusion in these two medical devices when syringes and catheters are replaced at 48-h intervals.

Glucose turnover after a mixed meal in dogs: glucoregulation without change in arterial glycemia.

Because the dog can respond to a mixed-meal challenge with little or no change in plasma glucose concentration, we used kinetic techniques to quantify the magnitude and duration of changes in glucoregulation. Glucose turnover was measured using [3-3H]glucose and [U-14C]glucose over two 19-h periods in healthy dogs, first during a fast (n = 6) and then throughout the postprandial state (n = 6) after a single mixed meal. Mean arterial glycemia remained constant in the fasted state (7.5 +/- 0.2 mM) and in the fed state (7.6 +/- 0.3 mM). Glucose appearance (Ra), however, increased slowly after the meal from 38 +/- 2 mg/min to a maximum of 79 +/- 8 mg/min after 6 h and stayed elevated until 12 h (P < 0.001). In parallel, glucose disappearance (Rd) rose from 35 +/- 3 to 83 +/- 7 mg/min, closely matching the corresponding Ra. Glucose recycling rose from 25 +/- 8% at baseline to a maximum of 53 +/- 15% (P < 0.05) at 14 h in fed dogs, whereas levels for fasted dogs stayed between 19 +/- 7% at 0 h and 27 +/- 12% at 6 h. Insulin levels rose significantly 30 min after the meal from 67 +/- 7 pM to a peak of 208 +/- 54 pM at 6 h but remained elevated for 12 h. We conclude that 1) the dog was able to maintain postprandial glucoregulation by very precise matching of Ra and Rd such as to maintain glycemia constant.(ABSTRACT TRUNCATED AT 250 WORDS)

Selective bilateral blood sampling from the inferior petrosal sinus in Cushing's disease: effects of corticotropin-releasing factor and thyrotropin-releasing hormone on pituitary secretion.

We sought to enhance the sensitivity of selective bilateral blood sampling to determine adrenocorticotropin (ACTH) and prolactin levels in the inferior petrosal sinus (IPS) by administering two stimulatory agents--corticotropin-releasing factor (CRF) and thyrotropin-releasing hormone (TRH). We then determined the ACTH and prolactin levels in the IPS of 10 patients with Cushing's disease. After peripheral administration of both CRF and TRH, ACTH levels were significantly higher on the tumor side in all patients. The prolactin level was significantly higher on the tumor side when CRF or TRH was used to stimulate pituitary secretion. Postsurgical immunohistochemistry studies revealed production of both ACTH and prolactin in tumor cells, explaining the abnormal secretion pattern of the pituitary adenoma. The use of CRF and TRH may therefore improve the reliability of selective blood sampling and tests from the IPS in those cases of Cushing's disease for which noninvasive methods have otherwise failed to clarify the diagnosis.

The effect of a time delay on the characteristics of the canine glucoregulatory system.

In order to elucidate the effect of a relative time delay on glucose regulation, we performed experiments with differently timed infusions of insulin and glucose in a canine model. When portal insulin infusion (0.03 U/kg over 5 minutes) preceded portal glucose infusion (0.05 g/kg over 5 minutes) by 1 minute, glycemia increased to a maximum value of 104 +/- 4 mg/dL at 6 minutes, whereas insulinemia peaked at 3 minutes at a level of 130 +/- 4 microU/mL (baseline, 21 +/- 7 microU/mL). C-peptide levels increased from 200 +/- 50 to 270 +/- 30 pmol/L. Glycemia then decreased to a minimum level of 61 +/- 4 mg/dL, significantly lower (P less than .02) than the corresponding values in control experiments when insulin was infused alone. With a reversed timing sequence of infusions with glucose infusion preceding insulin infusion by 1 minute, glycemia increased similarly, but decreased to a minimum level of only 84 +/- 4 mg/dL, which was significantly higher (P less than .01) than in the above experiment. Insulinemia peaked similarly at 126 +/- 7 microU/mL, and C-peptide increased from 210 +/- 50 to 280 +/- 50 pmol/L. These experiments demonstrated an unexpected effect: adding glucose to an insulin infusion almost doubled the biological activity of the exogenous insulin as measured by its hypoglycemic action. They also indicated that small perturbations of glycemia and insulinemia in the portal circulation have a profound effect on metabolism, and that even short relative time delays in elevating either insulinemia or glycemia can cause significantly different metabolic outcomes.(ABSTRACT TRUNCATED AT 250 WORDS)

Blood or urine glucose-based insulin therapy and control of glycemia. Computer-simulation study.

Adjustment algorithms for conventional insulin therapy must be tested for safety and efficacy before clinical implementation. We did this by computer simulation. Accordingly, a computer simulator of human intermediary metabolism created 10 randomly chosen diabetic subjects for study. All were well defined with respect to compliance (i.e., medication and diet) and life-style (i.e., physical and emotional stress). Insulin-adjustment algorithms that were tested calculated daily insulin dosages for these computer-simulated patients based on either blood or urine glucose concentrations self-measured 4 times/day before breakfast, lunch, dinner, and bedtime snack. The twofold purpose of the simulation study was to determine the ability of the adjustment algorithms to improve initially poor metabolic control and to compare the outcomes when either blood or urine glucose measurements were the basis on which glycemic control was implemented. A significant improvement in metabolic control could be achieved with either blood or urine glucose measurements as input to the algorithms. Detailed comparisons between blood and urine glucose-based treatments showed no significant advantage of blood glucose-based algorithms at breakfast (122 +/- 21 vs. 131 +/- 16 mg/dl) and dinner (117 +/- 27 vs. 130 +/- 23 mg/dl), whereas mean glycemia at lunch (122 +/- 24 vs. 164 +/- 21 mg/dl) and bedtime (117 +/- 25 vs. 150 +/- 21 mg/dl) after 120 days of simulation did differ significantly (P less than 0.01). Hypoglycemia was not provoked by either treatment. Total daily insulin doses evolved by blood glucose-based algorithms were significantly (P less than 0.05) higher than the doses used by urine glucose-based algorithms (53 vs. 47 U).(ABSTRACT TRUNCATED AT 250 WORDS)