Introduction

Premixed insulin analogs, such as biphasic insulin aspart 30, 50, and 70 (NovoRapid 30, 50, and 70; Novo Nordisk A/S, Søborg, Denmark) and lispro mix 25 and 50 (Eli Lilly and Company, Indianapolis, IN, USA), are forms of insulin therapy commonly used by patients with type 2 diabetes in Asian countries [1, 2]. Each premixed insulin analog contains a fixed ratio of intermediate-acting protaminated insulin aspart or lispro and a rapid-acting insulin analog. Protaminated insulin aspart and lispro are cloudy, and the degree of cloudiness should correspond to the ratio of rapid-acting insulin in a prefilled pen. Therefore, insulin cartridges and pens prefilled with insulin–protamine complex in binding mode will require adequate mixing to ensure a complete resuspension [3, 4]. Insufficient mixing will cause patients to receive inaccurate doses of intermediate- and/or short-acting insulin, which could significantly affect diabetes control. In this study, we investigated whether insulin users were correctly resuspending their premixed insulin analogs by measuring the optical densities (ODs) of the insulin solutions, which has been reported as the simplest method for assessing cloudiness [5].

Methods

Subjects

Four hundred and seventy-six patients with type 2 diabetes who had received premixed insulin analog therapy for at least 12 months were recruited for this study. Of these patients, 179 used NovoRapid 30 (mean age 59.9 ± 19.2 years, diabetes duration 20.1 ± 12.8 years, hemoglobin A1c [HbA1c] level 8.5 ± 1.6% [69 mmol/mol]), 66 used NovoRapid 50 (mean age 48.3 ± 14.9 years, diabetes duration 16.9 ± 8.9 years, HbA1c level 8.0 ± 1.3% [66 mmol/mol]), 33 used NovoRapid 70 (mean age 43.9 ± 20.0 years, diabetes duration 11.3 ± 6.9 years, HbA1c level 8.3 ± 1.0% [67 mmol/mol]), 58 used lispro mix 25 (mean age 67.5 ± 14.3 years, diabetes duration 22.1 ± 7.7 years, HbA1c level 8.9 ± 2.6% [74 mmol/mol]), and 140 used lispro mix 50 (mean age 65.4 ± 10.7 years, diabetes duration 14.7 ± 12.8 years, HbA1c level 8.1 ± 1.9% [65 mmol/mol]). Further information about the study population is provided in Table 1. The study protocol was approved by the Ethical Committee of Dokkyo Medical University (No. 28023). The trial was registered with the University Hospital Medical Information Network (UMIN No. 000022329). All patients provided informed consent and received conventional diabetes education and insulin therapy comprising premixed insulin analogs.

Table 1 Clinical characteristics of the subjects
Full size table

Optical Density Measurement

All patients enrolled in this study declared that they resuspended premixed insulin analogs by inversion at least 10 times before every injection, according to insulin injection best practice. Patients provided their current prefilled insulin pens for analysis. The ODs of the insulin solutions in these pens were measured at a wavelength of 415 nm, using a plate reader (Perkin Elmer, Inc., Waltham, MA, USA) [5]. Briefly, residual insulin volumes representing at least 5% were accurately mixed and compared with the range of OD values obtained from a series of correctly mixed, unused control insulin pens. The ODs of insulin solutions from the pens provided by patients were measured in triplicate, and the means of these measurements were used. The OD of the residual insulin was compared to standard insulin curves.

Statistical Analysis

Data are expressed as percentages of mixtures (%) and medians and interquartile ranges. All results are expressed as box and whiskers plots. Differences between groups were assessed using the Kruskal–Wallis test. P values less than 0.05 were considered statistically significant. All analyses were performed using Prism 6 (GraphPad Software, Inc., San Diego, CA, USA) or StatMate V (Nihon 3B Scientific Inc., Niigata, Japan).

Results

The following ODs were measured for the insulin solutions sampled from patients’ premixed insulin analog pens: NovoRapid 30 mix, 108.1 ± 12.0%; NovoRapid 50 mix, 121.6 ± 56.9%; NovoRapid 70 mix, 142.4 ± 51.6%; lispro mix 25, 112.1 ± 21.5%; and lispro mix 50, 110.5 ± 23.2%. The ODs of residual insulin from the high-mix biphasic insulin aspart suspensions (NovoRapid 50 and 70 mix) differed significantly from the expected values (Fig. 1).

Fig. 1
figure 1

Box and whisker plots of premixed insulin analogs. Data are expressed as percentages of the mixtures (%) and medians and interquartile ranges. Differences between groups were analyzed using the Kruskal–Wallis test. *P < 0.05 vs. NovoRapid 70

Full size image

Discussion

The present study results demonstrate that inadequate resuspension of premixed insulin analogs before injection is common among patients with diabetes who received conventional diabetes education at our hospital. An intensified insulin regimen, basal-bolus therapy, is known to be superior to conventional premixed insulin regimens with respect to changes in HbA1c levels and the onset and early-stage progression of microangiopathy in subjects with type 2 diabetes [6, 7]. However, premixed insulin has become more commonly used by patients with type 2 diabetes and stable lifestyles, such as elderly individuals [8, 9]. According to the consensus algorithm for patients with type 2 diabetes from ADA/EASD2015, premixed insulin is considered moderately complex but not very flexible [9], whereas basal-bolus therapy is recommended because it is more flexible than other regimens. Although guidelines do not strongly recommend this practice, premixed insulin can be adopted after considering a patient’s condition, lifestyle, and situation, especially in Japan [10]. Although we did not investigate a correlation between improved HbA1c levels and insulin injection techniques, physicians and patients must be educated about the proper method for resuspending premixed insulin in order to improve diabetes control and ensure safety. Additional information from future clinical studies with regard to mixing premixed insulin analogs is needed to address these results in patients with type 2 diabetes. Although we did not evaluate the correlation between OD and insulin concentration, Brown et al. indicated that there is a reasonable correlation between the insulin concentration, as measured by immunoassay, and ODs, which is considered to be a suitable marker for the state of insulin mixing [5].

Insulin degludec/insulin aspart (IDegAsp) is the most recently marketed insulin analog. IDegAsp is a fixed and soluble combination of the long-acting basal insulin, insulin degludec (IDeg; 70%), and rapid-acting insulin aspart (IAsp; 30%) [11]. Notably, IDegAsp does not need to be resuspended to homogeneity before injection [12]. This co-formulation of IDeg and IAsp may benefit patients with type 2 diabetes who currently use conventional premixed insulin analog therapy.

Conclusion

This is the first study to evaluate the percentage of protaminated insulin within the premixed insulin analog pens used by patients with type 2 diabetes. Similar to previously reported findings [3,4,5], the current results suggest that it is important to resuspend premixed insulin analogs, especially high-mix biphasic insulin analogs, more carefully before injection.