Insulin Lispro

Here, the physiological model of insulin kinetics (Equation (3)) was the linear three-compartment model (Figure 3) recently proposed by Schiavon et al. [12 (link)] for insulin lispro. The first two compartments, $I_{sc1}$ and $I_{sc2}$ , represent insulin masses in the subcutis, in a non-monomeric and monomeric state, respectively. From the former, insulin can either be absorbed in the plasma insulin compartment $I_p$ with a rate constant $k_{a1}$ (min ${}^{-1}$ ) or decomposed into monomers with a rate constant $k_d$ (min ${}^{-1}$ ). From the latter, insulin can only be absorbed in the plasma insulin compartment $I_p$ with a rate constant $k_{a2}$ (min ${}^{-1}$ ). Finally, insulin is degraded, usually by liver and kidneys, and this process is represented in the model by the rate constant $k_e$ (min ${}^{-1}$ ). The dose is injected in the $I_{sc1}$ compartment, while measurements are collected from the plasma compartment $I_p$ , which has a distribution volume of $V_I$ (L/kg). The model accounts also for a subject-specific delay in the insulin absorption through the parameter $\tau$ (min). The equations of the model are: $\left\{\begin{array}{c}{\dot{I}}_{sc1}(t)=-(k_{a1}+k_d)I_{sc1}(t)+u(t-\tau )\hfill \\ {\dot{I}}_{sc2}(t)=-k_{a2}{I}_{sc2}(t)+k_d{I}_{sc1}(t)\hfill \\ {\dot{I}}_p(t)=-k_e{I}_p+k_{a1}(t)I_{sc1}(t)+k_{a2}{I}_{sc2}(t)\hfill \\ y(t)=I_p(t)/V_I\hfill \end{array}\right.$

To facilitate blood sample collection throughout exercise and recovery, an IV catheter was inserted upon arrival at the laboratory. Venous blood samples were taken immediately before exercise (time 0), at the end of exercise (45 min), and at the end of recovery (105 min). Blood samples were taken in a seated position on an orthodontic-like chair with the seat having a 45-degree angle. To ensure participant safety, and to maintain BG in the desired range during exercise, capillary glucose was measured at least every 10 min using a hand-held glucometer (OneTouch Ultra2, LifeScan, Milpitas, CA, USA). This measurement was done due to a lag time with CGM during aerobic exercise in adults with T1D, which raises the need to measure capillary BG in a setting where the risk of hypoglycemia is elevated [16 (link)]. Venous blood samples were collected and were immediately mixed by inversion. The K₂EDTA BD Vacutainer^® tubes were centrifuged immediately for the separation of plasma, while the BD Vacutainer^® serum separator tubes (SST) were kept at room temperature for at least 15 min before centrifugation at 3000 revolutions/minute for 15 min. Plasma and serum samples were then aliquoted and stored at −80 °C for subsequent analysis. Frozen serum aliquots were shipped on dry ice to DynaLIFE Medical Labs (Edmonton, AB, Canada) for the analysis of glucose (using glucose hexokinase), sodium (using ion-selective electrode), potassium (using ion-selective electrode), calcium (using o-cresolphthalein complexone without deproteinization), magnesium (using modified xylidyl blue reaction), and insulin (using chemiluminescent microparticle immunoassay). It should be noted that the insulin analogs have cross-reactivity (binding of the antibodies employed in the assay to determine the concentration of an analyte) with reagents used for insulin measurement in this study (ARCHITECT^® insulin assay) [17 (link)]. In particular, the ARCHITECT insulin assay has shown high cross-reactivity to the insulin analogs lispro and glargine, and low cross-reactivity to the insulin analog aspart [17 (link)]. Our study participants used NovoRapid and Fiasp (insulin aspart), Humalog (insulin lispro), Toujeo (insulin glargine), Tresiba, or a combination of these analogs.
Participants were also provided with a sterile container in order to provide urine samples at the same time points, i.e., immediately before exercise, at the end of exercise, and at the end of recovery. Urine specific gravity (U_SG) was assessed immediately using a clinical refractometer (Reichert Inc., NY, USA). Urine samples were aliquoted and frozen at −80 °C for subsequent batch analysis. Similarly, frozen urine aliquots were shipped on dry ice to DynaLIFE Medical Labs (Edmonton, AB, Canada) for the analysis of sodium, potassium, and calcium. Urine glucose levels were measured using the Cayman glucose colorimetric assay kit (Cayman Chemical, MI, USA).

If the participants were treated with oral hypoglycemic agents or glucagon-like peptide 1 (GLP-1) receptor agonists on admission, the drugs were withdrawn on the second day of admission. BBT was introduced for all participants in both studies, and the SDD was determined using the following formula during the second study term: SDD (U/day) = 0.08 × FPG (mg/dL). In this study, actual SDD was modified to within 2 U/day from the calculated SDD. Thereafter, the actual SDD was divided into three or four daily insulin injections: three bolus injections of ultrarapid insulin analog (aspart, glulisine, or lispro) before meals for all participants and one bedtime basal injection of insulin glargine or detemir for participants with FPG levels ≥ 150 mg/dL. The dose of insulin injection was adjusted by physicians to within 4 U for each injection. The goals of glycemic control were to maintain FBG levels between 100 and 130 mg/dL and postprandial blood glucose (PBG) levels below 180 mg/dL according to the recommendations of the JDS^{18 (link)}.