Glucodynamics and Pharmacokinetics of 70/30 vs. 50/50 NPH/Regular Insulin Mixtures After Subcutaneous Injection

  1. Esther I. Krug, MD ,
  2. Louise DeRiso, CRNP ,
  3. Mary Beth Tedesco, CRNP ,
  4. Harsha Rao, MD and
  5. Mary T. Korytkowski, MD
  1. University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania

    Optimal insulin therapy for many patients with type 1 and type 2 diabetes requires the use of combinations of short- and intermediate-acting insulin preparations. A significant number of patients experience difficulties with mixing intermediate- and short-acting insulin preparations because of an inability to mix insulin preparations in correct ratios (1,2) or confusion between the different types of insulins (3,4,5). The fixed ratio of 70% NPH to 30% regular (70/30) insulin preparation does not fulfill the needs of many of these patients because of significant postprandial blood glucose excursions. A premix insulin preparation, such as 50% NPH to 50% regular (50/50) insulin, could benefit those individuals with postprandial hyperglycemia. To investigate differences in pharmacodynamic and pharmacokinetic properties between 50/50 and 70/30 premix insulins, we performed a randomized single-dose two-period crossover study using a 12-h euglycemic clamp technique (7) in healthy subjects after double-blinded subcutaneous injections of each insulin preparation.

    The study was approved by the Institutional Review Board of the University of Pittsburgh. Written informed consent was obtained before initiation of screening procedures. All subjects had a fasting blood glucose ≤108 mg/dl at screening, did not ingest alcohol or caffeine-containing beverages for 48 h before receiving study medication, and refrained from exercise for 24 h before the study. In a double-blind crossover study, single subcutaneous doses (0.2 units/kg) of 50/50 or 70/30 insulin were administered to 12 healthy male subjects. Glucose, insulin, and C-peptide levels were measured at baseline, every 10 min for 1 h, and every 15 min for 11 h after injection. Baseline euglycemia was maintained with a variable-rate glucose infusion (GIR). Serum insulin concentrations were analyzed using an enzyme-linked immunosorbent assay (ELISA) with a reference interval of 1.5–15.6 mIU/l, a detection limit of <0.5 mIU/ml, and cross-reactivity to C-peptide and proinsulin <0.3%. Serum C-peptide concentrations were analyzed by ELISA (reference interval 0.7–1.02 nmol/l, precision 8%). Glucose determinations were carried out immediately on site using a Yellow Spring Instruments 2300 STAT glucose analyzer. The 50/50 insulin premix (50% NPH in human insulin isophane suspension + 50% regular human insulin injection [rDNA], 100 units/ml, 3 ml PenFill cartridge) and the 70/30 insulin premix (Novolin 70/30, 3 ml PenFill cartridge) were supplied by Novo Nordisk Pharmaceutical Industries (Clayton, NC). The insulin premixes were administered using a NovoPen 3.0 with a NovoFine 30G needle (Novo Nordisk, Bagsvaerd, Denmark).

    Glucodynamic parameters included the area under the curve (AUC) as follows: AUCGIR(0–4), AUCGIR(4–8), AUCGIR(8–12), maximum GIR (GIRMAX), and time to GIRMAX [TGIR(MAX)]. The following pharmacokinetic measures were calculated from free insulin concentration–time curves adjusted for C-peptide: AUCINS(0–4), AUCINS(4–8), AUCINS(8–12), and maximum insulin concentration [CMAX and TINS(MAX)]. Adjusted insulin was derived according to the following formula: adjusted insulin = actual insulin − (baseline insulin / baseline C-peptide) × actual C-peptide. Analysis of variance was used to test the sequence, subject (within sequence), period, and treatment effect on AUC and CMAX for glucose, insulin, and GIRMAX. The analysis was performed on raw or logarithmic transformed data, depending on the distribution of the data.

    A total of 12 subjects completed the study (age 29.7 ± 9.1 years, weight 78.2 ± 11.6 kg, BMI 25.1 ± 1.6 kg/m2, and 100% Caucasian). Mean basal glucose values of 90.8 ± 4.9 mmol/l and 90.5 ± 5.8 mmol/l during 50/50 and 70/30 periods of the study were maintained for 12 h after the insulin bolus, with coefficients of variability of 5.4 and 6.4%, respectively. LogAUCGIR(0–4) was 6.6 ± 0.6 and 6.1 ± 0.6 with 50/50 and 70/30 injections, respectively (P = 0.0087). No significant differences were observed for AUCGIR(4–8), AUCGIR(8–12), AUCGIR(0–12), TGIR(MAX), or GIRMAX between the two insulin preparations. LogAUCINS(0–4) was 10.2 ± 0.5 after administration of 50/50 insulin and 9.6 ± 0.5 after an injection of 70/30 insulin (P = 0.018). No significant differences were observed in AUCINS(4–8), AUCINS(8–12), and TINS(MAX) between the two premixes.

    The major differences observed in the pharmacokinetic and glucodynamic properties of 50/50 and 70/30 insulin premixes occurred during the first 4 h after injection, with significantly greater AUCGIR(0–4), AUCINS(0–4), and CMAX after 50/50 vs. 70/30 insulin. Differences in pharmacokinetics and glucodynamics between NPH and regular insulins have been described previously (9,10,11,12). However, there are few published studies available that directly compare different concentrations of premix insulin preparations. In one open-label study directly comparing single injections of 50/50 and 70/30 insulin (6), significant differences between AUC for GIR and insulin concentrations adjusted for C-peptide were observed during the first 4 h after a higher dose of insulin (0.3 units/kg) than was used in this trial. In this previous trial, the investigators were not blinded to the insulin being used, which could have contributed to higher GIR with the 50/50 preparation because of anticipation of peak insulin action. In another study, postprandial hyperglycemic responses to a standard breakfast were measured in elderly individuals with insulin-requiring type 2 diabetes after a single injection of 50/50 or 70/30 insulin mixtures (0.33 units/kg) (8). No differences were observed in nadir and peak glucose concentrations, time to nadir and peak glucoses, or plasma insulin concentrations during the first 4 h after insulin injection. The absence of C-peptide measurements, the use of uncorrected insulin measures, and the low insulin dose used during the study period relative to the usual insulin requirements of the subjects (55.3 ± 24.0 units/day) may have contributed to the absence of an observed difference in the two insulin premix preparations.

    No significant differences were observed in any glucodynamic or pharmacokinetic parameters during hours 4–12 after the insulin injections. This suggests that the risk for either late hypo- or hyperglycemia is unaffected by the type of insulin premix used. An overlap in action profiles of regular and NPH insulins during hours 3 and 5 and a relatively flat NPH action profile during hours 6–12 could explain these observations (12,13).

    In summary, there are statistically significant differences in glucodynamic and pharmacokinetic parameters between 50/50 and 70/30 insulin mixtures during the first 4 h after a single injection in healthy volunteers. This suggests that 50/50 insulin premix preparations may offer the advantage of reducing postprandial glycemic excursions in patients with insulin-requiring diabetes.


    This work was supported by a research grant from Novo Nordisk and funds received from the National Institutes of Health/National Center for Research Resources/General Clinical Research Center Grant no. 5M01-RR-00056.

    M.T.K. serves on an advisory panel for Novo Nordisk.


    • Address correspondence to Mary T. Korytkowski, MD, Falk, Room 581, University of Pittsburgh Medical Center, 3601 Fifth Avenue, Pittsburgh, PA 15213. E-mail: korytkowski{at}


    | Table of Contents

    Navigate This Article