Fault Tolerant Strategies for Automated Insulin Delivery Considering the Human Component: Current and Future Perspectives

Automated Insulin Delivery (AID) are systems developed for daily use by people with type 1 diabetes (T1D). To ensure the safety of users, it is essential to consider how the human factor affects the performance and safety of these devices. While there are numerous publications on hardware-related failures of AID systems, there are few studies on the human component of the system. From a control point of view, people with T1D using AID systems are at the same time the plant to be controlled and the plant operator. Therefore, users may induce faults in the controller, sensors, actuators, and the plant itself. Strategies to cope with the human interaction in AID systems are needed for further development of the technology. In this paper, we present an analysis of potential faults introduced by AID users when the system is under normal operation. This is followed by a review of current fault tolerant control (FTC) approaches to identify missing areas of research. The paper concludes with a discussion on future directions for the new generation of FTC AID systems.     

Assessment of Regular and NPH Insulin Concentration via Two Methods of Quantification: The Washington State Insulin Concentration Study (WICS)

Background:Recent reports have suggested that insulin vials purchased in community pharmacies do not meet the minimum required intact insulin concentration (≥95 U/mL) as defined by the United States Pharmacopeia. We sought to independently obtain multidose human insulin vials from a variety of community pharmacies across the state of Washington and quantitatively measure intact insulin.Methods:Sixty 10-mL vials of insulin (n = 30 regular human insulin and n = 30 neutral protamine Hagedorn insulin) were purchased and assayed. To ensure random selection of lots and supply chain sources, insulin samples were purchased on a variety of calendar dates from various pharmacy locations across Washington State, inclusive of both chain and independent pharmacies. All samples were assessed for intact insulin concentration via both Ultra Performance Liquid Chromatography coupled with UV detection (UPLC-UV) and Ultra Performance Liquid Chromatography-Mass Spectrometry (UPLC-MS).Results:When considering all samples (N = 60), the mean concentration was 101.8 ± 4.4and 91.5 ± 1.9 U/mL as determined by UPLC-UV and UPLC-MS, respectively. Measured concentrations ranged from 90.0 to 108.4 U/mL when assayed by UV UPLC and 86.1 to 95.4 U/mL for UPLC-MS.Conclusion:To our knowledge, this is the first study following the report by Carter et al that assessed human insulin concentrations by both UPLC-UV and UPLC-MS. These findings are important because they demonstrate that the results obtained from these two methods differ and that the method used must be considered when interpreting findings.     

Hybrid Closed-Loop Control with Ultrarapid Lispro Compared with Standard Insulin Aspart and Faster Insulin Aspart: An In Silico Study

Objective:There is room for improvement in the performance of closed-loop regulation algorithms during the prandial period. This in silico study evaluated the efficiency and safety of ultrarapid lispro insulin using the Diabeloop DBLG1® algorithm.Methods:We modeled the insulin profile of URLi according to literature data and integrated it to the model used within a simulation platform built from a 60 patients’ virtual cohort. We then ran the DBLG1® algorithm in silico with various meal intakes using modeled URLi, Aspart and Faster Aspart. The primary endpoints were glucose metrics (time in 70-180 mg/dL range and time below range).Results:When insulin time constant values were tuned, time in 70-180 mg/dL range was 69.4 [61.1-75.6] (Aspart) vs 74.7 [65.5-81.5] (URLi). Glucose coefficient of variation was reduced from 34.1 [29.7-37.8] to 28.4 [25.7-34.6]. Time below 70 mg/dL and 54 mg/dL were significantly reduced with URLi, whether or not DBLG1 was specifically tuned to this insulin. Metrics with Faster Aspart were intermediate and did not significantly differ from URLi.Conclusions:This simulation study performed on a virtual T1D population suggests that the use of URLi within an unmodified closed-loop DBLG1 regulation algorithm is safe and, with DBLG1 being tuned to this specific insulin type, improved the regulation performances as compared with Aspart. This fact supports the use of such an insulin in clinical investigations.     

Patients with Type 1 Diabetes Treated with Insulin Pumps Need Widely Heterogeneous Basal Rate Profiles Ranging from Negligible to Pronounced Diurnal Variability

Background:Pump-treated patients with type 1 diabetes have widely differing basal insulin infusion profiles. We analyzed consequences of such heterogeneity for glycemic control under fasting conditions.Methods:Data from 339 adult patients with type 1 diabetes on insulin pump therapy undergoing a 24-hour fast (basal rate test) were retrospectively analyzed. Hourly programmed basal insulin infusion rates and plasma glucose concentrations as well as their proportions within, below, or above arbitrarily defined target ranges were assessed for specific periods of the day (eg, 1-7 hours, “dawn” period, 16-19 hours, “dusk” period, reference period 20-1 hours/10-14 hours), by tertiles of a predefined “dawn” index (mean basal insulin infusion rate during the “dawn” divided by the reference periods).Results:The “dawn” index varied interindividually from 0.7 to 4.4. Basal insulin infusion profiles exhibited substantial differences (P = .011), especially overnight. Despite higher insulin infusion rates at 4 and 6.45 hours, patients with the most pronounced “dawn” phenomenon exhibited higher plasma glucose concentrations at those time points (P < .012). Patients with a marked “dawn” phenomenon exhibited a lower probability for low (<4.4 mmol/L) and a higher probability of high values (>7.2 mmol/L) during the dawn period (all P values <.01).Conclusions:We observe substantial interindividual heterogeneity in the “dawn” phenomenon. However, widely different empirically derived basal insulin infusion profiles appear appropriate for individual patients, as indicated by similar plasma glucose concentrations, mainly in the target range, during a 24-hour fasting period.     

Advances in Insulin Pen Technologies: Analysis of Symposium Articles on Insulin Pen Devices and Alternative Insulin Delivery Methods

Insulin therapy is a fundamental component of diabetes management, yet there is often resistance to insulin initiation by both prescribers and patients. A barrier to insulin use is the perceived shortcomings with the traditional vial-and-syringe administration method (inconvenience, difficulty of use, association with disease and addiction, etc.). This symposium in the journal discusses the advantages of alternative insulin delivery methods, primarily insulin pen devices. Although these administration methods, especially insulin pens, have some clear advantages over the vial/syringe, there are also limitations to their use and careful patient selection and education are still needed.     

Analysis of Insulin Pen Devices for the Treatment of Diabetes Mellitus

Insulin pen devices allow for accurate, flexible, and less complicated delivery of insulin for the treatment of diabetes mellitus. These devices permit small dose administration, and can be used by patients with limited dexterity and visual impairment. These characteristics may prove beneficial when considering their usefulness. The article by Hanel et al. in this issue of Journal of Diabetes Science and Technology clearly points out that the OptiClik^® reusable pen may underdose insulin early after cartridge replacement unless properly primed. Insulin pens clearly offer several advantages over traditional vials and syringes. However, patients must be well educated in their use, with continued communication between them and their health care provider to enable good glycemic control.     

More Focus is Needed to Reduce Adverse Events for Diabetes Devices

Advances in devices for people with diabetes have demonstrated many improvements; yet, the number of adverse events has almost doubled from 2018 to 2019. It is a challenge to examine these events due to a difficult query tool on the FDA website. There are several possible reasons why effort is not devoted to decreasing the number of adverse events including the fact that user error is a common cause. This commentary serves to raise awareness of the adverse event problem.     

An Analysis: To Code or Not to Code��That Is the Question

Most blood glucose monitoring systems need coding to correct for variation in lots of enzyme, which leads to differences in lots of strips. About 16% of patients miscode the meters, although the magnitude of the miscoding is unstudied. This miscoding has the potential to cause errors as high as 30% and to cause errors in adjusting insulin therapy that could lead to hypoglycemia at least 10% of the time. Studies of these systems suggest that they have accuracy similar to other current meters and have similar physical characteristics. Because they do not require coding, they are often easier to use. No-coding systems have the potential to avoid some errors in blood glucose.     

The Role of Human-Centered Design in Insulin Pen Innovation and the Future of Insulin Delivery

Insulin pens have made a dramatic impact on diabetes care, with evidence suggesting that they promote performance of self-care and reduce negative health outcomes for people with diabetes. Human-centered design (HCD), practiced by IDEO for over 40 years and together with Eli Lilly for over 15 years, has helped to design insulin pens that evolved with the needs of people with diabetes. HCD employs unique methods that help to uncover people’s needs and design with them in mind. The future of diabetes care is bright with the ongoing application of HCD methodology in this space.     

Improving Hospital Glucometrics,Workflow, and Outcomes with a Computerized Intravenous Insulin Dose Calculator Built into the Electronic Health Record

Objective:To adjust for dynamic insulin requirements in critically ill patients, intravenous (IV) insulin infusions allow for titration of the dose according to a prespecified algorithm. Despite the adaptability of IV insulin protocols, human involvement in dose calculation allows for error. We integrated a previously validated IV insulin calculator into our electronic health record (Epic) and instituted it in the cardiovascular intensive care unit (CVICU). We aim to describe the design of the calculator, the implementation process, and evaluate the calculator’s impact.Method:Employing an aggressive training program and user acceptance testing prior to significant elbow support at the time of institution, we successfully integrated the insulin calculator in our CVICU. We evaluated the glucometrics before and after implementation as well as nursing satisfaction following calculator implementation.Results:Overall, our implementation led to increased frequency of blood sugar at various glycemic targets, a trend toward less hypoglycemia or hyperglycemia. For severe hypoglycemia, our preintervention cohort had 0.02% of blood sugars less than 40 mg/dL but no blood sugars less than 40 mg/dL were identified in our patient’s postintervention. For the CVICU target blood glucose of 70-180 mg/dL, 87.97% blood sugars at baseline met goal compared to 91.39% at one month, 91.24% at three months, and 90.87% at six months postintervention.Conclusion:By utilizing an aggressive education campaign championing superusers and making adjustments to the calculator based on early problems that were encountered, we were able to improve glycemic control and limit glucose variability at our institution.     

Comparison of Insulin Pump Bolus Calculators Reveals Wide Variation in Dose Recommendations

Background:The introduction of insulin pumps with bolus calculators (BCs) has improved glycemic outcomes and quality of life for those with type 1 diabetes. Despite the increased reliance on BCs, the formulas used to derive recommended boluses are not standardized. Our objective was to examine whether recommendations from different pump BCs vary significantly for identical clinical scenarios.Methods:Three commercially available insulin pump BCs were programmed with identical settings and then presented with combinations of blood glucose (BG) and carbohydrates (CHOs) to generate a 4-unit bolus. At one- and two-hour time points, while there was insulin-on-board (IOB) present, we simulated various BG and CHO scenarios in order to compare BC-recommended doses.Results:Differences in suggested doses were noted between BCs, as well as within the same brand. The greatest variation was apparent when BG was below target. Doses suggested by one BC varied depending on whether the IOB resulted from a previous dose given for BG or CHO, while the other two BCs adjusted for total IOB regardless of the source.Conclusions:In this simulation study, there were large differences in recommended doses between BCs due to the unique way each manufacturer incorporates IOB into their formulas as well as the pharmacokinetics used to derive the IOB amount. Providers should be aware that identical pump settings will result in a different dose recommendation for each pump brand and advise patients accordingly.     

A Pilot Study of Flat and Circadian Insulin Infusion Rates in Continuous Subcutaneous Insulin Infusion (CSII) in Adults with Type 1 Diabetes (FIRST1D)

Background:Initiation of continuous subcutaneous insulin therapy (CSII) in type 1 diabetes requires conversion of a basal insulin dose into a continuous infusion regimen. There are limited data to guide the optimal insulin profile to rapidly achieve target glucose and minimize healthcare professional input. The aim of this pilot study was to compare circadian and flat insulin infusion rates in CSII naïve adults with type 1 diabetes.Methods:Adults with type 1 diabetes commencing CSII were recruited. Participants were randomized to circadian or flat basal profile calculated from the total daily dose. Basal rate testing was undertaken on days 7, 14 and 28 and basal rates were adjusted. The primary outcome was the between-group difference in absolute change in insulin basal rate over 24 hours following three rounds of basal testing. Secondary outcomes included the number of basal rate changes and the time blocks.Results:Seventeen participants (mean age 33.3 (SD 8.6) years) were recruited. There was no significant difference in absolute change in insulin basal rates between groups (P = .85). The circadian group experienced significant variation in the number of changes made with the most changes in the morning and evening (P = .005). The circadian group received a greater reduction in total insulin (−14.1 (interquartile range (IQR) −22.5-12.95) units) than the flat group (−7.48 (IQR −11.90-1.23) units) (P = .021).Conclusion:The initial insulin profile does not impact on the magnitude of basal rate changes during optimization. The circadian profile requires changes at specific time points. Further development of the circadian profile may be the optimal strategy.     

Use of Computer Vision to Identify the Frequency and Magnitude of Insulin Syringe Preparation Errors

Background:No current technology exists to ensure the dose of insulin administered in hospitals matches the physician order.Objective:Assess the feasibility of using computer vision to identify insulin syringe preparation errors.Methods:Twenty-two nurses prepared 50 insulin doses (n=1100) each. A computer vision device (CVD) measured the volume drawn up and identified air present. Syringes identified as inaccurate by the CVD were confirmed by two observers, and a random sample of 100 syringes identified as accurate was validated by two independent observers.Results:Ten syringes (1.0%) had the wrong volume prepared, and 68 syringes (6.5%) contained air sufficient to meet the definition of inaccuracy. All errors were confirmed by two independent observers.Conclusion:CVDs could reduce insulin administration errors in hospitalized patients.     

Adherence and Persistence to Insulin Therapy in People with Diabetes: Impact of Connected Insulin Pen Delivery Ecosystem

Diabetes is an increasing public health problem, and insulin is the mainstay for treatment of type 1 diabetes. In type 2 diabetes treatment, insulin therapy is used after oral or other injectable agents become inadequate to achieve glycemic control. Despite the advances in insulin therapy, management of diabetes remains challenging. Numerous studies have reported low adherence and persistence to insulin therapy, which acts as a barrier to successful glycemic control and diabetes management. The aim of this targeted review article is to provide an overview of adherence and persistence to insulin therapy in people with diabetes and to discuss the impact of the emergence of a new connected ecosystem of increasingly sophisticated insulin pens, glucose monitoring systems, telemedicine, and mHealth on diabetes management. With the emergence of a connected diabetes ecosystem, we have entered an era of advanced personalized insulin delivery, which will have the potential to enhance diabetes self-management and clinical management. Early systems promise to unlock the potential to address missed or late bolus insulin delivery, which should help to address non-adherence and non-persistence. Over time, improvements in this ecosystem have the potential to combine insulin data with previously missing contextualized patient data, including meal, glucose, and activity data to support personalized clinical decisions and ultimately revolutionize insulin therapy.     

Pharmacokinetics and Pharmacodynamics of Human Regular U-500 Insulin Administered via Continuous Subcutaneous Insulin Infusion Versus Subcutaneous Injection in Adults With Type 2 Diabetes and High-Dose Insulin Requirements

Introduction:Human regular U-500 insulin (U-500R) is approved for subcutaneous (SC) injection in patients with diabetes requiring >200 units/day of insulin. Here, pharmacokinetic and pharmacodynamic (PK/PD) profiles following U-500R administered by continuous subcutaneous insulin infusion (CSII) and SC injection in adults with type 2 diabetes (T2D) on high-dose insulin were studied.Methods:In this randomized, crossover, euglycemic clamp study, patients received a 100-unit bolus of U-500R via SC injection or CSII with basal infusion using a U-500R specific pump. PK parameters were estimated using non-compartmental methods. PD estimates were derived from the glucose infusion rate during the euglycemic clamp procedure.Results:When corrected for the basal infusion, the PK profiles for the 100-unit bolus of U-500R were similar for CSII and SC injection. Without correction for basal infusion, PK and PD profiles showed a greater insulin concentration and effect when U-500R was administered via CSII compared to SC injection, primarily due to basal insulin infusion for CSII. The ratio of geometric least squares AUC_0-tlast means SC:CSII (90% CI) is 0.857 (0.729, 1.01) with correction (mean AUC_0-tlast: 5230 pmol*L/h [SC injection] and 6070 pmol*L/h [CSII, with correction]) and 0.424 (0.361, 0.499) without correction (mean AUC_0-tlast: 12300 pmol*L/h [CSII, without correction]). Median time-to-peak insulin concentration was six hours (range 0.5-8 hours) via SC injection and five hours (0.5-12 hours) via CSII.Conclusions:In adults with T2D on high-dose insulin, U-500R PK/PD parameters were similar for a 100-unit bolus when given by SC injection or CSII via a U-500R pump.     

Continuous Subcutaneous Insulin Infusions vs. Multiple Daily Injections of Insulin in Hospitalized Patients: Glycemic Trends in the First 24 Hours of Admission

Background:Continuous subcutaneous insulin infusion (CSII) is a common diabetes treatment modality. Glycemic outcomes of patients using CSII in the first 24 hours of hospitalization have not been well studied. This timeframe is of particular importance because insulin pump settings are programmed to achieve tight outpatient glycemic targets which could result in hypoglycemia when patients are hospitalized.Methods:This retrospective cohort study evaluated 216 hospitalized adult patients using CSII and 216 age-matched controls treated with multiple daily injections (MDI) of insulin. Patients using CSII did not make changes to pump settings in the first 24 hours of admission. Blood glucose (BG) values within the first 24 hours of admission were collected. The primary outcome was frequency of hypoglycemia (BG < 70 mg/dL). Secondary outcomes were frequency of severe hypoglycemia (BG < 40 mg/dL) and hyperglycemia (BG ≥ 180 mg/dL).Results:There were significantly fewer events of hypoglycemia [incident rate ratio (IRR) 0.61, 95% confidence interval (CI) 0.42–0.88, p = 0.007] and hyperglycemia (IRR 0.79, 95% CI 0.65–0.96, p = 0.02) in the CSII group compared to the MDI group. There was a trend toward fewer events of severe hypoglycemia in the CSII group (IRR 0.15, 95% CI 0.02–0.93, p = 0.06).Conclusions:Patients using CSII experienced fewer events of both hypoglycemia and hyperglycemia in the first 24 hours of hospital admission than those treated with MDI. Our study demonstrates that CSII use is safe and effective for the treatment of diabetes within the first 24 hours of hospital admission.     

The Influence of Insulin on the Raised Plasma Fibronectin Concentration in Human Obesity

ABSTRACT Plasma concentrations of fibronectin, free insulin, C-peptide and plasma glucose were determined in 40 morbidly obese subjects and in 51 normal weight controls, matched for sex and age. All plasma concentrations were significantly elevated (p<0.01) among the obese subjects. A significant correlation (r=0.34, p<0.05) between plasma fibronectin and plasma free insulin was found among the obese patients, but not among the controls (r=-0.02, p>0.05). No significant correlation was found between plasma fibronectin and plasma C-peptide, neither in the obese patients nor in the controls (obese r=0.06, controls r=0.02; p>0.05). Plasma fibronectin was insignificantly correlated with body weight (obese r=0.21, controls r=0.15; p>0.05) and percentage overweight (obese r=0.27, controls r=0.04; p>0.05). The raised level of circulating insulin may in part explain the excess of plasma fibronectin of obese subjects.