Insulin Therapy

The discovery of insulin in 1921-22 was one of the greatest medical breakthroughs in history (1) (Figure 1). This work was based, in part, on a collective body of evidence that well preceded the Toronto work that began in 1869 with Paul Langerhans’s discovery of clusters of cells in the pancreas (now known as the islets of Langerhans) (2). Between 1889 and 1893, work from several scientists showed that pancreatectomized dogs developed a diabetic state and it was hypothesized that this resulted from a loss of “internal secretion” of the pancreas rather than of exocrine secretion. Between 1915 and 1919, Kleiner and Meltzer published promising results of their pancreatic extracts in depancreatized dogs which resulted in glucose lowering. Parallel work at the University of Toronto allowed for whole fresh pancreatic extracts to be used to decrease blood glucose in diabetic dogs. The first successful human administration of insulin was performed in January 1922 to a 14-year-old named Leonard Thompson which resulted in a rapid and dramatic improvement in his diabetes. Developments by the pharmaceutical industry allowed for the large-scale commercial insulin production in 1923 (3). Individuals, mostly children with type 1 diabetes (T1D), whose life expectancies were measured in months were now able to prevent fatal ketoacidosis by taking injections of crude “soluble” (later known as regular) insulin. However, problems were soon noted. Hypoglycemia, occasionally life-threatening, was encountered as diabetes monitoring by urine testing for glycosuria was crude at best during those first decades after the discovery of insulin. The insulin itself was often impure and varied in potency from lot to lot. Allergic reactions were common and occasionally anaphylaxis would occur. Even more concerning was the appreciation that these patients often succumbed to chronic vascular complications which either dramatically reduced quality of life or resulted in a fatal cardiovascular event.

Tools to manage individuals with T1D improved over the decades since the discovery of insulin. Initial insulins were manufactured from bovine or porcine pancreata and production techniques became more efficient. Insulins with longer duration of action were first introduced in the 1930s, and over time purity and consistency of potency of these insulins improved (4). Nevertheless, “standard” animal insulins prior to 1980 contained 300-10000 parts per million of impurities and elicited local and systemic effects when injected. Present day insulins sold in the United States today all contain less than 1 part per million of impurities.

Major improvements in insulin were developed in the late 1970s and early 1980s. First, not only was “purified” insulin introduced, but in 1982 the first human insulin was marketed both by Eli Lilly (recombinant DNA technology) and Novo (semi-synthetic methodology). These insulins were available as short-acting (regular) and longer-acting [Neutral protamine Hagedorn (NPH), lente, and ultralente] preparations. The other major advance with insulin therapy was with the delivery by the first continuous subcutaneous insulin infusion (CSII) pumps. While pumps were initially touted as providing less variable insulin absorption, the use of CSII had a greater impact: both patients and clinicians used this tool to teach themselves how to best use “basal bolus” insulin therapy, a strategy that would become a standard of care after the beginning of the next century with the development of insulin analogs.

Figure 1.

Timeline of the evolution of insulin therapy. Figure source ref 3.

Monitoring Tools

At the same time as the development of human insulin and insulin pumps, improvements in glucose monitoring were introduced. Although there was initial skepticism if home blood glucose monitoring would be accepted by patients with diabetes, history has confirmed that this technology has revolutionized diabetes management and has allowed patients to titrate blood glucose to normal or near-normal levels. While self-monitoring of blood glucose (SMBG) allowed immediate evaluation of diabetes management, the introduction of hemoglobin A1c (HbA1c, or glycated hemoglobin, A1C) around the same time was used as a marker of objective longer-term (about 90 days) glucose control. When hemoglobin is exposed to glucose in the bloodstream, the glucose slowly becomes nonenzymatically bound to the hemoglobin in a concentration-dependent manner. The percentage of hemoglobin molecules that are glycated (have glucose bound to it) indicates what the average blood glucose concentration has been over the life of the cell. Perhaps as importantly, A1C made it possible for researchers to study the effects of long-term glucose control and the development of vascular complications. New students of diabetes may now find it difficult to appreciate that one of the greatest medical controversies between the discovery of insulin and the early 1990s was the relationship between glucose control and diabetes complications. Improved insulins, pumps, SMBG, and A1C finally allowed this question to be properly studied.

Glucose Meters

Current BGM systems are small electronic devices capable of analyzing glucose levels in capillary whole blood. To test blood glucose levels, patients are required to prick a finger using a lancing device to obtain a small drop of blood. The patient then places the drop of blood onto a glucose test strip, which has been previously inserted into the glucose meter. Typically, just a few seconds are required for the device to provide a blood glucose value.

BGM systems use enzymatic reactions to provide estimates of blood glucose levels and the enzymes utilized include glucose oxidase, glucose dehydrogenase and hexokinase. The specific enzyme is usually packaged in a dehydrated form in a glucose test strip. Once blood is applied to the test strip, glucose in the patient’s blood sample rehydrates the enzyme activating a reaction. The product of this reaction can then be detected and measured by the glucose meter (19).

The introduction of BGM more than 45 years ago was met with much skepticism until results of the DCCT. Point-of-care BGMS revolutionized diabetes care by allowing patients and practitioners to obtain real-time estimates of blood glucose values. These portable devices enabled patients to perform self-monitoring of blood glucose (SMBG), an integral component of effective diabetes self-management. The benefits of SMBG were confirmed during the DCCT which showed that intensive insulin therapy, requiring SMBG≥4 times/day with concomitant insulin dose titration, delayed the onset and slowed the progression of microvascular complications (5). Later, it was shown in the T1D Exchange that a higher frequency of testing (up to 10 times daily) is inversely associated with A1C levels in all age groups (20).

SMBG can guide management decisions (e.g. adjusting food intake, insulin therapy, and exercise) and determine whether glucose targets are being achieved. Further, it can help patients in monitoring and preventing asymptomatic hypoglycemia (18). Frequent BGM measurements are important for diabetes management to guide insulin dosing, food intake, and for preventing exercise related hypoglycemia. Every person with T1D should have a BGM, regardless of CGM use.

The technology of BGM systems has evolved over the years and current devices are relatively easy to use and require minimal amounts of blood (Figure 4). Some instruments are able to capture events affecting glucose control (e.g., exercise, meals, insulin administration), provide customized reports, and calculate insulin bolus needs according to glycemia and intake of carbohydrate based on pre-established settings (i.e. insulin sensitivity factor and insulin-to-carbohydrate ratios). However, despite these unique advances in self-monitoring of blood glucose, independent analytic testing has shown that various BGM systems do not fulfill the accuracy requirements set by the International Organization for Standardization (ISO) 151917 which requires for ≥95% of results to fall within ± 15 mg/dL of the reference result for samples with glucose concentrations <100 mg/dL and ±15% for samples with glucose concentrations ≥100 mg/dL (21). In addition, the FDA has stated that the ISO 15197 criteria are not sufficient to adequately protect lay users of SMBGs because, for example, the standard does not adequately address the performance of over-the-counter blood glucose testing systems in the hypoglycemic range or across test strip lots. Thus, there is substantial variation in the accuracy of widely used BGM systems. In view of this, the FDA developed the “Self-Monitoring Blood Glucose Test Systems for Over-the-Counter use” guidance document which is intended to guide manufacturers in conducting appropriate performance studies and preparing 510(k) submissions for these device types (https://www.fda.gov/regulatory-information/search-fda-guidance-documents/self-monitoring-blood-glucose-test-systems-over-counter-use). Thus, there is a pressing need for high quality standards to ensure improved accuracy and precision from BGM systems. Information on the performance of various BGM systems can be obtained from the Diabetes Technology Society BGM System Surveillance Program (diabetestechnology.org/surveillance/).

BGM is an integral component of therapy for individuals with diabetes using insulin. In recent years, CGM has emerged as a preferred method for the assessment of glucose levels (see below). Patients with T1D using BGM should be encouraged to perform SMBG at a minimum of 4 times a day (before meals and at bedtime), as this will allow adjustments to prandial and basal insulin doses. In addition, SMBG should be considered prior to snacks, before and at completion of exercise, in the event of symptoms suggestive of hypoglycemia, and after treating hypoglycemia until blood glucose levels have normalized. Lastly, patients should test their blood glucose before performing critical tasks such as driving a motor vehicle or operating heavy machinery. Ultimately, frequency of SMBG will largely depend on patients’ individual needs. However, even with frequent BGM, most people with T1D will have undetected and high frequency of hyper- and hypoglycemia ((22)).

Patients only using BGM should also be educated on avoiding “overuse” of SMBG. Testing too frequently may lead to administration of multiple correction doses within short periods of time, particularly if patients are anxious about their glucose levels not returning to target “fast enough”, leading to insulin “stacking” and resulting in iatrogenic hypoglycemia. Advice against purchasing or reselling preowned or secondhand test strips, as these may give incorrect results. Use of unopened and unexpired vials of glucose test strips will ensure BGM accuracy.

SMBG has important drawbacks since blood is only sampled intermittently and therefore only glimpses of blood glucose concentrations are provided. SMBG does not offer information on glucose fluctuations even if performed frequently. Thus, there is potential for missing episodes of hyperglycemia and hypoglycemia.

Figure 4.

Examples of a few blood glucose monitoring systems.

Glucose Downloads

The vast majority of currently available BGM systems allow the generation of downloadable reports. These reports are a unique component of the patients’ evaluation allowing the identification of areas that require special attention in diabetes management. However, technical difficulties often compromise the usefulness of these data. For instance, it is not unusual for the date and/or time of the glucose meters to be inaccurate. Simple errors such as these have a huge impact on patient management as the data downloaded becomes largely uninterpretable. In addition, as each glucose meter usually has its own proprietary software, if a clinic does not have the specific software installed on their local computers, then the data may not be downloaded. The clinician is left with trying to review the data directly from the device, which is time consuming and does not offer the detailed overview from a customized printable report. There are platforms that are currently available which allow downloading various glucose meters, insulin pumps and CGM data and provide standardized reports (e.g. CliniPro®, Carelink®, Glooko^®, Tidepool). However, there needs to be a unified effort by BGM systems, insulin pump and CGM companies in order to generate a universal download protocol as this would simplify data analysis and interpretation by practitioners (23).

Continuous Glucose Monitoring (CGM)

Perhaps the most innovative technology for the treatment of T1D is the introduction of CGM (Figure 5). As differs from SMBG, CGM technology measures glucose concentrations in the interstitial fluid (ISF) which correlates well with plasma glucose values. . CGM is now considered the standard for glucose monitoring for most adults with T1D (22).

How CGM systems work - The components of CGM consist of a sensor filament, a small electronic device that serves as the platform for the sensor, a transmitter (either integrated with the sensor or separate), and a receiver device which can be a standalone device or a smartphone (Figure 5). The thin wire-like sensor is inserted into the subcutaneous tissue through the skin, and adhesive material keeps sensor and transmitter in place attached to the skin. The sensor is coated with glucose oxidase which reacts with interstitial glucose which creates an electrical signal which is then converted to a glucose value. Insertion of the sensor causes an inflammatory reaction and a disruption of the local subcutaneous interstitial environment resulting in a waiting period for the sensor signal to stabilize known as “warm up time”. Sensors can be worn up to 15 days (non-implantable) or 365 days (implantable). CGM sensors can measure glucose levels up to every minute allowing for a glucose tracing to be generated and displayed in real-time (RT-CGM) on a receiver device. The glucose data is displayed numerically and graphically, informing the user of glucose levels and trends greatly improving the understanding of patients’ glucose profiles. Further, available CGM devices in the US have FDA approval for non-adjunctive use which means that patients can rely on their CGM values to guide diabetes management decisions.

Lag time - Lag time refers to CGM readings that are delayed compared to fingerstick blood glucose readings. Glucose moves passively from the capillaries into the interstitium. Several factors including blood flow and permeability affect interstitial glucose concentrations. When interpreting CGM values it is important to understand that ISF glucose consistently lags plasma glucose. A study in healthy adults analyzing glucose tracers following an overnight fast showed that it takes 5-6 minutes for glucose to be transported from the vascular to the interstitial space (physiological delay) (24). This is particularly relevant when glucose levels are trending up or down quickly as CGM data will not be as reliable in such scenarios and thus patients should confirm the glucose concentration by SMBG.

Sensor interference - inaccuracies in CGM values do occur with pressure on the CGM, known as “compression hypoglycemia.” This most often occurs when someone is sleeping thus applying body weight directly on the sensor. Some chemical substances may interfere with the glucose-oxidase reaction, including acetaminophen, hydroxyurea, and Vitamin C.

Advantages of CGM sensors - Patients can customize alarms that activate to detect hypoglycemia or hyperglycemia. Understanding the trend allows patients to decide whether an increase or decrease in mealtime insulin dose is necessary. CGM thus allows patients to intercept hypoglycemia (or hyperglycemia) prior to it occurring. Patients can also “flag” events thereby improving interpretation of glucose control associated with meals, insulin administration, and exercise. Most CGM devices allow users to share their glucose data with others (e.g. family members or friends) which can then be monitored on a smartphone or other internet-enabled devices. This is of particular interest in the pediatric population as it allows parents to remotely monitor their child’s glucose profile when away from home or while exercising (e.g. participating in sports). Features of currently available CGM devices are listed in Table 1.

Another advantage of CGM is the amount of data that can be generated and downloaded in customizable reports (Figure 6). Health care professionals are not only able to download daily glucose profiles in a graphic display but can also obtain several statistics including means, medians, standard deviations, interquartile ranges, and minimum and maximum values. This provides a better assessment of glycemic variability (Figure 7). Most importantly, time in glucose ranges can be identified and evaluated. This is particularly helpful in patients who have hypoglycemia unawareness and allows for adjusting the treatment plan by both the patient and practitioners to eliminate occurrence of hypoglycemia.

KEY CGM METRICS

Figure 5.

Examples of real-time continuous glucose monitoring systems.

Table 1.

Features of currently approved CGM devices in the United States

View in own window

	FreeStyle Libre 2 plus	FreeStyle Libre 3 plus	G6	G7	Guardian sensor 4	Instinct	Simplera Sync	Simplera	Eversense 365
Manufacturer	Abbot	Abbott	Dexcom	Dexcom	Medtronic	Abbott	Medtronic	Medtronic	Senseonics
CGM Type	Real-time	Real-time	Real-time	Real-time	Real-time	Real-time	Real-time	Real-time	Real-time
Approved for ages (years)	≥2	≥2	≥2	≥2	≥7	≥7	≥7	≥18	≥18
Wear time (days)	15	15	10	10	7	15	7	7	365
Real-Time glucose alarms	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Calibration required	No	No	No	No	No				Yes
Warm up time	60 min	60 min	120 min	30 min	120 min	60 min	120 min	120 min	24 hr
Integrated transmitter	Yes	Yes	No	Yes	No				No
Approved insertion sites	Back of upper the arm	Back of upper the arm	Back of upper arm, abdomen, upper buttocks	Back of upper arm, upper buttocks	Back of upper arm	Back of upper arm	Back of upper arm	Back of upper arm	Implanted subcutaneously in the upper arm
Insulin pump integration	OmniPod5 t:slim X2	iLet twiist	OmniPod5 t:slim X2 Mobi	OmniPod5 t:slim X2 Mobi iLet	MiniMed 780G	MiniMed 780G	MiniMed 780G	No	No
Real-time data access by healthcare professional	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Friends and family share	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Water resistant	up to 3 feet (1 meter) for 30 min	up to 3 feet (1 meter) for 30 min	up to 8 feet (2.5 meters) for 24 hr	up to 8 feet (2.5 meters) for 30 min	up to 8 feet (2.5 meters) for 30 min	up to 3 feet (1 meter) for 30 min	up to 8 feet (2.5 meters) for 30 min	up to 8 feet (2.5 meters) for 30 min	up to 3 feet (1 meter) for 30 min (Transmitter)
Interfering substances*	Vitamin C ↑	Vitamin C ↑	Acetaminophen (>1 g q6h in adults) or paracetamol ↑ Hydroxyurea ↑	Acetaminophen (>1 g q6h in adults) or paracetamol ↑ Hydroxyurea ↑	Acetaminophen or paracetamol ↑	Vitamin C ↑ (>1g per day)	Acetaminophen or paracetamol ↑ Hydroxyurea ↑	Acetaminophen or paracetamol ↑ Hydroxyurea ↑	Tetracyclines ↓ Mannitol ↑

Figure 6.

A 14-day DEXCOM CGM overview report showing sensor glucose data over a 24-hour period including mean (dotted line), standard deviation, glucose management indicator, interquartile range (grey bars), upper and lower glucose thresholds (orange and red lines, set by the user), percent time in range, sensor usage, top patterns, and average daily calibrations.

Figure 7.

A 7-day DEXCOM CGM overlay report showing daily profiles allowing for the identification of trends and patterns.

Figure 8.

Ambulatory Glucose Profile (AGP) sample.

Glycemic Targets

A1C is a measure of average glycemia over ~3 months and is a strong predictor of complications of diabetes (31). Current glycemic targets for adults from the American Diabetes Association (ADA) include a target A1C of <7%. However, it should be noted that this recommendation is a general target and the goal for the individual patient is as close to normal as possible (A1C of < 6%) without significant hypoglycemia. In addition, patients with T1D and hypoglycemia unawareness, long duration (> 25-30 years) of disease, limited life expectancies, very young children, or those with co-morbid conditions will require higher A1C targets. Individualized A1C targets need to be reviewed with each patient (32).

Thus, A1C testing should be performed routinely in all patients with diabetes as part of ongoing care. Frequency of A1C testing is determined based on the clinical situation, the treatment regimen used, and the clinician’s judgment. A1C measurements every 3 months help in the assessment of whether a patient’s glycemic targets have been reached. Although convenient, there are drawbacks to A1C measurements, as glycation rates may vary with patients’ race/ethnicity. Similarly, in patients with hemoglobinopathies, hemolytic anemia or other conditions that shorten the red blood cell life span, the A1C may not accurately reflect glycemic control or correlate with SMBG testing results. In such conditions, fructosamine may be considered as a substitute measure of long-term (average over 4 weeks) glycemic control. Clinicians should routinely compare downloaded SMBG or CGM averages with A1C as there are many reasons A1C may be altered due to a non-glycemic etiology and thus fructosamine or the downloaded glucose data itself would be a better metric to follow (33).

Non-Glycemic Treatment Targets

It should also be pointed out that in addition to glycemic targets, specific non-glycemic targets have also been recommended (34) . Non-glycemic targets should also be tailored according to the individual with less stringent treatment goals for individuals with multiple coexisting illnesses and/or poor health and limited life expectancy. Recent real-world data from the T1D Exchange revealed that the incidence of cardiovascular disease (CVD) over 4.6 years was ~3.7% (35). Age, longer duration of diabetes, glycemic control, obesity, hypertension, dyslipidemia, and diabetic nephropathy were all associated with increased risk for CVD.

Principles of Management of T1DM

Management of T1D involves a multidisciplinary framework that includes the following:

• Physiologic insulin replacement using basal-bolus therapy, either as MDI or CSII to maintain glycemia as close to normal as possible.
• Avoidance of hypoglycemia.
• Blood glucose monitoring with CGM or SMBG with development of individualized A1c goals.
• Patient education and support with a supportive team of providers including endocrinologists, diabetologists, primary care clinicians, nurses, certified diabetes care and education specialists (CDCES), pharmacists, psychologists, dietitians, social workers, other specialists such as cardiologists, nephrologists, psychiatrists as well as family members, social support groups etc.

Types of Insulin

Selecting the appropriate insulin depends largely on the desired time course of insulin action. Table 2 and figure 9 show the pharmacokinetic characteristics—time to onset of action, time of peak action, effective duration of action, and maximum duration of action—of currently available insulins; however, these can vary considerably among individuals.

Insulin products are categorized according to their action profiles:

• Rapid-acting: e.g., insulin lispro, insulin aspart, and insulin glulisine (genetically engineered insulin analogs).
• Short-acting: regular (soluble) insulin.
• Intermediate acting: NPH (isophane).
• Long-acting, e.g., insulin glargine, insulin detemir, and insulin degludec (genetically engineered insulin analogs).
• Pre-mixed insulin.
• Inhaled insulin.

Insulin analogs are insulin molecules modified by genetic engineering and recombinant DNA technology. The amino acid structure of insulin is altered to change the properties of insulin – i.e., time to onset, peak, and duration of action, compared to human regular insulin. However, the biological properties and stability of the insulin molecules are intact. A general principle to bear in mind is the longer the time to peak, the broader the peak and the longer the duration of action. Additionally, the breadth of the peak and the duration of action will be extended with increasing dose. Figures 9 and 10 should therefore be considered a conceptual representation of insulin action curves.

Mealtime (Prandial) Insulins

Basal Insulins

Pre-Mixed Insulins

Premixed insulins are mixtures of prandial and intermediate acting insulins (the same prandial insulin attached to protamine so that it becomes intermediate acting). Insulin mixtures are available as human insulin mixtures (NPH and regular mixture) as well as analog mixtures. In the US, insulin lispro protamine mixtures are available in two forms: 75% insulin lispro protamine suspension and 25% insulin lispro injection (75/25) and 50% insulin lispro protamine suspension and 50% insulin lispro injection (50/50). Available preparations of insulin aspart protamine mixtures include 50/50 and 70/30 suspensions. A variety of other ratios are available in Europe. These insulin mixtures are typically administered before breakfast and dinner. This alleged twice daily dosing is the primary advantage of these insulins. In general, use of premixed insulins restricts adjustment of doses and meal timing. Therefore, premixed insulins are not recommended for adult patients with T1D, where intensive regimens with ability to make adjustments in the premeal short-acting insulin bolus are better suited for glycemic control. Premixed insulin in T1D could have benefit for some patients who do not adhere to an intensive insulin regimen, and with consistent food intake and timing of meals.

Concentrated Insulins

Biosimilar Insulins/Follow-On Biologics

According to the FDA, a “biosimilar” is a biological product that is highly similar to a US-licensed reference biological product notwithstanding minor differences in clinically inactive components, and for which there are no clinically meaningful differences between the biological product and the reference product in terms of the safety, purity, and potency of the product. Currently FDA-approved follow-on biologics include: Basaglar (insulin glargine), Semglee (US, - insulin glargine), Rezvoglar (insulin glargine), Admelog (insulin lispro), Merilog (insulin aspart), and Kirsty (insulin aspart). There are several others under development.

Figure 9.

Available basal insulins and duration of action. Figure source Ref (58) .

Factors Influencing Insulin Absorption

Insulin absorption variability is one of the greatest obstacles to replicating physiologic insulin secretion. Among the many factors that affect insulin absorption and availability (Table 3) are injection site, timing, type or dose of insulin used, and physical activity. Day-to-day intra-individual variation in insulin absorption is approximately 25%, and the variation between patients may be as high as 50%. This occurs more commonly with larger doses of human insulin which form a depot and can unpredictably prolong duration of action; however, this is less of an issue with rapid-acting insulin analogs. In general, any strategy that increases the consistency of delivery should decrease glucose fluctuations; and insulin regimens that emphasize rapid-acting insulin are more reproducible in their effects on blood glucose levels. Insulin pumps using a rapid-acting insulin analog can significantly reduce glucose variability. Like multiple-injection regimens, use of an insulin pump requires frequent glucose monitoring. In addition, pump users need a back-up method of insulin administration, and attention to mechanical and injection site issues.

Reducing Variability of Insulin Absorption

Role of Insulin Analogs in Management of T1D

Most of the problems of insulin replacement in T1D arise from the fact that subcutaneous injections or pump infusion remain a relatively poor route of administration. From the subcutaneous site of injection, insulin is absorbed into the systemic, not portal circulation. More importantly, subcutaneous injections lead to variable absorption from one injection to another, due largely to the non-physiologic pharmacokinetics of standard insulins. Insulin analogs were developed to overcome this problem.

Currently there are three rapid-acting insulin analogs: insulin lispro, insulin aspart, and insulin glulisine, all of which have a rapid onset of action and peak, thereby improving 1- to 2-hour postprandial blood glucose control compared with regular insulin. These rapid-acting analogs must be used in conjunction with a basal insulin to improve overall glycemic control (Figure 9). Importantly, the rapid-acting analogs have consistently outperformed regular insulin in terms of post-absorptive hypoglycemia. This finding should not be surprising since the duration of regular insulin is much longer than the gut absorption of a typical mixed meal.

Figure 10.

Idealized insulin curves for prandial insulin with a rapid-acting analog (RAA) with basal insulin glargine or insulin detemir. Each insulin preparation is responsible for either the prandial or basal component. Many patients find the basal insulins do not last the entire 24 hours and they give the basal insulin twice daily. B=breakfast; L=lunch; S=supper; HS=bedtime.

Clinical trials have demonstrated lower fasting glucose levels and less nocturnal hypoglycemia with insulin glargine than with NPH insulin, advantages that are especially relevant in patients aiming for meticulous control (A1C <7%) or those with hypoglycemia unawareness. Trials with T1D have shown similar results with insulin detemir which compared with NPH insulin was equally effective in maintaining glycemic control, although detemir was administered at a higher molar dose. The newest basal insulin preparations, insulin degludec and U-300 insulin glargine are claimed to show less nocturnal hypoglycemia than insulin glargine or insulin detemir. In general, hypoglycemia is reduced with any of these basal analog insulins compared to NPH insulin. Since hypoglycemia is clearly one of the treatment-limiting aspects of T1D therapy, the use of these analogs has gained wide-spread acceptance.

Multiple Daily Injection (MDI) Insulin Therapy

A simpler conceptual approach preferred by most patients with T1D is using a prandial insulin analog for each meal (i.e., insulin lispro, insulin aspart, or insulin glulisine) and a separate basal insulin analog [i.e., insulin glargine, insulin detemir, or insulin degludec]. Although these true basal-prandial regimens require more shots than conventional twice-daily regimens, they are considerably more flexible, allowing greater freedom to skip meals or change mealtimes. Moreover, use of the long-acting basal and rapid-acting insulin analogs allows strategies to achieve individual, defined blood glucose targets more easily. Such modifications might include changing the timing of insulin injections in relation to meals, changing the portions or content of food to be consumed, or adjusting insulin doses or supplements for premeal hyperglycemia.

The basic treatment principles of insulin dosing include establishing a total daily dose, an insulin to carbohydrate ratio and an insulin sensitivity or correction factor.

USING PRANDIAL INSULIN

Establishing an Insulin to Carbohydrate (Carb) Ratio

Patients with T1D derive the greatest therapeutic benefit when basal and prandial analogs are used together, because the physiologic pharmacokinetics and pharmacodynamics of these analogs make separating the basal and prandial components of insulin replacement easier. In general, administering the appropriate amount of pre-meal insulin requires that the patient know at least their current blood glucose level and the estimated amount of carbohydrates for a meal. Initially, the amount of prandial insulin can be determined by approximating the percentage of calories consumed at each meal. As patients become more educated, however, they may alter the prandial dose by estimating the carbohydrate component of each meal or snack.

The carb ratio provides the dose of rapid acting insulin (lispro, aspart, glulisine) to cover the carbohydrate content of a meal. A typical starting point in patients with T1D is to give 1 unit of rapid acting insulin for every 15 grams of carbohydrates. This ratio is variable ranging from 1 unit for every 5g to 30 g of carbohydrate. To estimate the carb ratio, the “500 rule” can be used:

500/total daily dose (TDD) = grams of carbohydrate covered by 1 unit of insulin.

Example: A person who takes a total of 50 units of insulin per day (both basal and prandial combined) will need 1 unit of rapid acting prandial insulin for every 10g carbohydrate (500/50 = 10g of carbohydrate covered by 1 unit of insulin, using above formula).

Alternative way to calculate the carb ratio – Add all carbohydrates consumed in a day and divide this by the total units of prandial insulin taken that day, using an average over 3 days.

Prandial insulin may be reduced/skipped when:

• Extra carbohydrates are used to raise low blood sugars or cover increased physical activity.
• Recent dose of correction insulin within past 1-2 h.
• Nausea or vomiting preventing oral intake.

Determining the Correction Dose or “Insulin Sensitivity Factor” (ISF):

In addition to covering the carbohydrate load of a meal, individuals will also need to correct hyperglycemia, called the “correction dose”. The method commonly used for this is the “1800 Rule”. This estimates the point drop in glucose for every unit of rapid-acting insulin administered:

1800/TDD = Point drop in glucose for 1 unit of rapid-acting insulin

This ISF (also called the correction factor) can be used for between-meal elevations in blood glucose. Thus, in general this correction dose can be utilized anytime provided the patient has not taken an injection of rapid acting insulin over the past 2-4 hours (insulin on board, Figure 11).

Target glucose: The ISF enables achieving appropriate individualized blood glucose targets.

For example: A person who takes a total of 60 units of insulin per day will require 1 unit of rapid acting insulin to drop the glucose by 30 points. If the patient’s glucose is 180 mg/dL and the glucose target has been set at 120 mg/dL, a correction dose of 2 units would be required to bring the glucose down to target:

• ISF = 1800/60 (TDD) = 30; 1 unit of rapid-acting insulin will decrease glucose by 30 points.
• 180 mg/dL (actual glucose level) – 120 mg/dL (target glucose level) = 60; this is the excess glucose, that is, the value that is above target and that needs to be corrected.
• 60/30 (ISF) = 2; dividing the excess glucose by the ISF will provide the amount of correction insulin units that are required to bring down the glucose to target, in this case it will be 2 units.

Putting it All Together - Combining the Carb Ratio and ISF

Combining the carbohydrate load and ISF will enable patients to appropriately target their pre-meal glucose.

For example: An individual with a carb ratio of 1:15 and ISF of 1 unit/50mg/dL, prior to a meal of 60g carbohydrates and a pre-meal blood glucose of 220mg/dL and target of 120mg/dL would take the following steps to administer the appropriate amount of prandial insulin as follows:

• To cover carbohydrate intake: 60g/15g per unit =4 units.
• Correction dose: 220 mg/dL (actual glucose)– 120mg/dL (target glucose) = 100mg/dL. ISF is 100/50 = 2 units to correct..
• Total amount of prandial insulin: 4+2= 6 units.

Insulin Titration and Pattern Adjustments

Reviewing SMBG or CGM readings and recognizing patterns is one of the most important aspects of diabetes management, allowing for timely and appropriate adjustments in insulin dose, food intake and managing physical activity. Pattern management is aided by valuable tools such as SMBG with information obtained through download software (see above) or logbooks and CGM data. These tools can be used in order of priority, for assessment of hypoglycemia, hyperglycemia, glycemic variability, frequency of SMBG readings etc.

Figure 11.

The appearance of insulin into the blood stream (pharmacokinetics) is different than the measurement of insulin action (pharmacodynamics). This figure is a representation of timing of insulin action for insulin aspart from euglycemic clamp data (0.2 U/kg into the abdomen). Using this graph assists patients to avoid “insulin stacking”. For example, 3 hours after administration of 10 units of insulin aspart, one can estimate that there is still 40% X 10 units, or 4 units of insulin remaining. By way of comparison, the pharmacodynamics of regular insulin is approximately twice that of insulin aspart or insulin lispro. Currently used insulin pumps keep track of this “insulin-on-board” to avoid insulin stacking. Adapted from reference (39).

Insulin Pens

Insulin pens were first introduced in 1981 as injection devices. These pens contain a cartridge holding insulin which is injected into the subcutaneous tissue through a fine, replaceable needle. Insulin pens are convenient, portable and are widely used as a part of MDI therapy. Currently, insulin pens are available as disposable pens containing prefilled cartridges or reusable insulin pens with replaceable insulin cartridges. Several insulin pens allow the convenience of ½ unit dosing, a critical need for pediatric patients and those adults with high insulin sensitivity and low insulin requirements.

Smart pens - The first insulin smart pen was approved for use in the United States (InPen, Companion Medical, California, USA) in 2017 (Figure 12). Smart pens can record timing and amount of each administered insulin dose, display the last dose and insulin onboard and also make dosing recommendations based on pre-specified information (Figure 13). This information is wirelessly transmitted via Bluetooth to a dedicated mobile application on a smartphone device. Other similar devices are in development including the NovoPen 6 and NovoPen Echo Plus reusable insulin pens equipped with near-field communication technology (Novo Nordisk, Denmark), approved in the European Union.

Figure 12.

InPen Smart Insulin Pen.

Figure 13.

InPen Insight report. Report provides missed doses, bolus calculator dosage, long-acting insulin assessment and CGM data (if paired with the InPen app).

Continuous Subcutaneous Insulin Infusion Therapy (CSII)

While not a new tool, insulin pump therapy remains the gold standard of insulin delivery for T1D. CSII is the most precise way to mimic normal insulin secretion because basal insulin infusion rates can be programmed throughout a 24-hour period. CSII may be thought of as a computerized mechanical syringe automatically delivering insulin in physiologic fashion. Patients can accommodate metabolic changes related to eating, exercise, illness, or varying work and travel schedules by modifying insulin availability. Basal rates can be adjusted to match lower insulin demands at night (between approximately 11 PM and 4 AM) and higher requirements between 3 AM or 4 AM and 9 AM.

Background: The earliest commercial insulin pumps were large devices infusing insulin via metal needles with a very short battery life, limited flexibility in the rate of insulin delivery and limited safety alarms. The 1990s and 2000s saw significant technological advancement of insulin pump devices with incorporation of safety features including alerts for infusion set occlusion and low battery or low insulin reservoir. The devices also became smaller in size with longer-life batteries and plastic catheter infusion sets. As pump technology advanced, the next step was to develop connectivity with CGM devices to develop “a closed loop”. CGM systems and insulin pumps improve glycemic control and decrease risk of severe hypoglycemia compared with use of SMBG and multiple daily insulin injections in T1D. Using a CGM and insulin pump together (referred to as sensor augmented pump therapy) improves HbA1c in patients who have high sensor wear time, but diabetes management burden does not decrease as individuals need to engage frequently with their devices. Thus, the concept of glucose responsive automated insulin delivery (AID), in which data from CGM can inform and allow adjustment of insulin delivery by the insulin pump device emerged (see below). Modern insulin pumps are much smaller and easier to use than the pumps of the past (Figure 14).

Figure 14.

Examples of modern-day insulin pumps.

INSULIN USE IN PUMPS

All rapid-acting analogs are approved in the United States for use in insulin pumps. The basal rate of the insulin pump replaces the use of daily injections of basal insulin. The boluses given before each meal are essentially the same as normal insulin injections of rapid acting insulin. The pump allows programming of several different basal infusion rates at increments that can range from 0.025 up to 35.0 units/hour (usually ranging from 0.4 to 2.0 units/hour) to meet non-prandial insulin demands, though it is unlikely that the average patient will require more than 2 or 3 different rates (Figure 15). As with MDI, correction doses can be provided before or between meals. Figures 16 and 17 show data that is typically downloaded from a pump.

Figure 15.

Idealized insulin curves for CSII with either insulin lispro, insulin aspart, or insulin glulisine. Note the basal insulin component can be altered based on changing basal insulin requirements. Typically, insulin rates need to be lowered between midnight and 0400 h (predawn phenomenon) and raised between 0400 h and 0800 h (dawn phenomenon). The basal rate the rest of the day is usually intermediate to the other two. Modern-day pumps can calculate prandial insulin dose by the patient entering the blood or CGM glucose concentration and the anticipated amount of carbohydrate to be consumed. The pump calculates how much previous prandial insulin is still active and provides the patient a final suggested dose which the patient may activate or override.

Capabilities of insulin pumps include the following:

DOWNLOAD CAPABILITY

Pump data can be downloaded, and the data obtained is extremely helpful in understanding patients’ glycemic responses to an established insulin regimen (Figure 16). Also, it can assist in evaluating patients’ behaviors pertaining to their glucose management. Downloads provide information regarding the total daily insulin use broken down into percentages corresponding to basal and bolus delivery. This allows determining if patients are consistently administering boluses or whether they are essentially “running on basal.” Some of the additional data that can be downloaded includes average glucose levels, frequency of glucose monitoring, days between site changes, amount of time patients are suspending the pump or using temporary basal rates, frequency and timing of boluses (which allows to identify non-compliance or insulin stacking behaviors), and average daily carbohydrates consumed (Figure 17).

Figure 16.

A patient’s insulin pump download (Tandem t:slim X2 with Control IQ) showing comprehensive data for one day including basal rates and suspensions, boluses, total daily insulin dose , carbohydrate intake, and sensor glucose data including time in range, average glucose and standard deviation.

Figure 17.

A patient’s pump download(Tandem t:slim X2 with Control IQ) showing a 14 days summary of pump activity.

Automated Insulin Delivery (AID) Systems

Improvements in CGM sensor technology have allowed for the integration of CGM systems with insulin pumps and the development of AID systems , also known as closed-loop (CL) systems. An AID system consists of an insulin pump, a CGM device, and an insulin infusion algorithm designed for safety and glucose control optimization.

In 2009, the JDRF developed an artificial pancreas road map defining 6 stages of APDS technology based on the level of automation (59):

Risks Associated With Insulin Pumps

Although there are multiple benefits of pump therapy there are also several risks. The first is an abrupt stoppage of insulin delivery either from an occlusion or dislodging of the catheter. For most patients who measure glucose levels at least 4 times daily the problem can be discovered and rectified quickly. However, for the occasional patient who tests infrequently or misses several glucose tests the discontinuation of the insulin infusion can result in ketoacidosis. Fortunately, this is rare. When glucose levels are found to be elevated for no apparent reason, it is appropriate to bolus the appropriate correction dose and if after 1 to 2 hours glucose levels are not improved, an injection of insulin is recommended, and the infusion site should be changed.

Another potential complication is infection, often an abscess, at the infusion site. This is also rare and can be minimized with meticulously cleaning the pump site prior to insertion. Although not as severe, inflammation from pump sites can be problematic. This can be improved by changing the infusion set every 24 to 72 hours and rotating pump sites. Similarly, some patients develop lipohypertrophy from infusing insulin in the same area. This can result in extreme variability in insulin absorption. Again, frequent rotation of pump sites can alleviate this problem which is under-reported. Clinicians should therefore make pump site observation a part of every clinic visit.

Amylin Analog - Pramlintide

Amylin is a neuroendocrine hormone co-secreted with insulin by the pancreatic beta cells in a fixed ratio (67); T1D is a state of amylin deficiency. Amylin reduces postprandial hyperglycemia by reducing mealtime glucagon secretion. It also delays gastric emptying, increases satiety and enables weight loss. Overall, amylin complements the action of insulin by targeting postprandial hyperglycemia.

Pramlintide is an injectable amylin analog approved for use in T1D as an adjunct to prandial insulin. Pramlintide has similar physiological effects as amylin, such as decreased food intake, slowing gastric emptying and decreasing appetite. It decreases postprandial hyperglycemia and mean A1C by 0.3-0.5% with modest weight loss (68). A crossover study of pramlintide infusion co-administered with human regular insulin via a pump over 24h improved glycemic variability and postprandial hyperglycemia in adults with T1D (69). Pramlintide is injected just prior to meals at an initial dose of 15 mcg and increased as tolerated to a final dose of 60 mcg. It should be administered only prior to major meals consisting of 250 calories or 30 grams of carbohydrate. Prandial insulin doses of insulin (in MDI or CSII therapy) should be reduced as food intake decreases and gastric emptying is delayed. For those receiving insulin via a pump, using an “extended bolus” (see above) works best to avoid postprandial hypoglycemia. For those using MDI, some patients administer their insulin just prior to eating (without a lag time) or after eating. Use of pramlintide is limited by nausea, often mild and self-limited. Severe insulin-induced hypoglycemia has also been noted with the use of pramlintide if insulin doses are not sufficiently reduced on initiation of pramlintide therapy. However widespread use of pramlintide as a therapeutic adjunct in T1D has been limited due to concerns of nausea, hypoglycemia and additional injection burden.

Metformin

Metformin, a biguanide, is a highly effective oral agent for therapy of patients with T2D. It decreases hepatic gluconeogenesis and improves insulin sensitivity (70). Metformin may have some benefit in reducing insulin doses and possibly improve metabolic control in obese/overweight individuals as observed in small studies in patients with T1D. An early meta-analysis of 5 studies suggested that addition of metformin resulted in a decrease in insulin requirement (6.6 units/day), and a decrease in weight with minimal change in A1C (71). A randomized placebo-controlled trial in 140 overweight adolescents with T1D evaluated the addition of metformin to insulin (72). There was no improvement in glycemic control after 6 months, but use of metformin resulted in decreased insulin dose and improved measures of adiposity, despite increased gastrointestinal adverse events. Although metformin has been shown to decrease CVD morbidity in T2D, data in T1D is lacking. The REMOVAL trial assessed benefit of metformin in T1D and cardiovascular risk and showed no evidence of sustained A1C reduction, and no benefit in carotid intima-media thickness (the study’s primary endpoint); however reductions in body weight, LDL-C and total insulin requirements was observed (73). There is evidence to suggest that metformin decreases insulin resistance and improves vascular health in adolescents with T1D (74). A meta-analysis of 19 RCTs suggests short term improvement in A1C that is not sustained after 3 months and associated with higher incidence of GI side effects (75). Therefore, based on current evidence, concomitant use of metformin in patients with T1D is not recommended in current published guidelines.

Sodium Glucose Cotransporter 2 (SGLT2) Inhibitors

SGLT2 is a protein expressed in the proximal convoluted tubule (PCT) of the kidney and is responsible for re-absorption of filtered glucose. Inhibition of SGLT2 prevents glucose reabsorption in the PCT and increases glucose excretion by the kidney. SGLT1 is the major intestinal glucose transporter. SGLT1 inhibition also increases postprandial release of the gastrointestinal hormones GLP-1 and polypeptide YY, probably by increasing delivery of glucose to the distal small intestine, thereby regulating glucose and appetite control. Notably, the action of these agents is insulin-independent; therefore, this class of drugs has potential as adjunctive therapy for T1D. Additionally clinical trials have clearly demonstrated the efficacy of these agents in decreasing the severity of heart failure, major adverse cardiovascular outcomes as well as improvement in renal outcomes in T2D; therefore, there is significant interest for use in T1D. Early small studies of SGLT2 inhibitors in T1D showed promising results with evidence of decreased total daily insulin dosage, improvement in fasting glucose and A1C, measures of glycemic variability, rates of hypoglycemia and body weight (76-78).

The most common side effects associated with this class of drugs include genital mycotic and urinary infections. Euglycemic diabetic ketoacidosis can develop in patients with T1D due to glycosuria masking hyperglycemia but with a catabolic state (due to insulin deficiency and hyperglucagonemia) with ketonemia (79, 80). Despite demonstrated beneficial effects of these agents in T1D, an increased incidence of diabetic ketoacidosis has been observed in all trials, and is a major concern, and persists despite educational efforts.

Several SGLT2 inhibitors including empagliflozin, dapagliflozin and canagliflozin have been studied in T1D. When added to insulin therapy, all SGLT2 inhibitors appear to decrease A1C levels, averaging 0.35-0.5% within 6 months of initiation; however this effect does not appear to be sustained at 1 year in clinical trials and effects appear to wane with time (81). Insulin dosing should be adjusted with caution to avoid hypoglycemia. Low dose empagliflozin (2.5 mg) has shown lower rates of diabetic ketoacidosis in clinical trials (82). Larger studies evaluating safety of lower SGLT inhibitor doses are needed to establish utility and safety in people with T1D. Availability of continuous ketone monitoring devices in the future may help mitigate DKA risk by enabling early detection of ketonemia. Due to the very high risk of euglycemic diabetic ketoacidosis SGLT2 inhibitors are not approved for use in T1D in the US and are approved for use in T2D only. Currently, Japan is the only market where SGLT-2 inhibitors are approved for adjunctive use in T1D (due to studies showing favorable risk profiles). In Europe, dapagliflozin was approved for use as adjunct therapy to insulin in T1D but was voluntarily withdrawn from the market by the manufacturer in 2021. Sotagliflozin is a dual SGLT-1/2 inhibitor that decreases renal glucose reabsorption through systemic inhibition of SGLT-2 and decreases glucose absorption in the proximal intestine by SGLT-1 inhibition, blunting and delaying postprandial hyperglycemia. Sotagliflozin use in T1D results in sustained A1C reduction, weight loss, lower insulin requirements, lesser hypoglycemia, but more diabetic ketoacidosis relative to placebo.(83) Although the drug received authorization in the EU for T1D it is not marketed there.

Incretin Based Therapies

Endogenous glucagon-like peptide-1 (GLP-1) is secreted from L cells (present in the small and large intestine) in response to food ingestion. GLP-1 enhances glucose-induced insulin secretion, inhibits glucagon secretion, delays gastric emptying, and induces satiety. GLP-1 secretion in patients with T1D is similar to that seen in healthy individuals. In vitro studies suggest that incretin-based therapies can expand beta cell mass, stimulate beta cell proliferation and inhibit beta cell apoptosis, although this has not been demonstrated in humans. Thus, due to their putative effects on beta cell integrity and function, GLP-1 receptor agonists and oral dipeptidyl peptidase-4 (DPP-4) inhibitors are of interest in T1D.

Bariatric Surgery

Bariatric and other metabolic surgeries are effective weight loss treatments in severe obesity. In individuals with T1D and Class III obesity, bariatric surgery has been shown to result in significant weight loss, decrease in insulin requirements and an overall improvement in metabolic profile. However, DKA and hypoglycemia occur in the post-operative period. Longer term and larger studies are required to further evaluate the role of bariatric surgery in T1D (91).

Psychosocial Aspects

T1D is a chronic condition with need for self-care behaviors and therefore many psychological challenges. Emotional distress related to diabetes occurs in up to 40% of people with T1D at any point in their lives related to the demands of daily self-care, fear of hypoglycemia or worry related to development of complications. Depression and anxiety often coexist in people with T1D. Assessment and management of psychosocial issues are an important component of care in individuals with T1D throughout their life span (92). Evaluation and discussion of psychosocial issues and screening for depression should be included as part of each clinic visit. Many patients experience “diabetes distress” related to the multitude of self-care responsibilities to optimize glycemic control. Diabetes distress is frequently associated with suboptimal glycemic control, low self-efficacy and reduced self-care. Depression, anxiety from fear of hypoglycemia, and eating disorders can develop and are associated with poor glycemic control. In young adults, comprehensive management of diabetes that addresses these psychosocial issues can improve glycemic control and reduce hospitalization due to diabetic ketoacidosis. Strategic interventions such as cognitive restructuring, goal setting and problem solving can help individuals particularly adolescents and young adults reduce diabetes distress (93). Thus, early identification and treatment including referral to a mental health specialist can help aid management of diabetes. While the individual patient is the focus of care, family support should be encouraged when appropriate.

Management During Exercise

Regular physical activity is now recommended as a cornerstone of care for T1D. The benefits of exercise and physical activity in patients with T1D have been well documented (94, 95). With appropriate glucose management and education, T1D is not an impediment to competing in sports or athletics at the highest level. However, achieving adequate glycemic control during and after completion of exercise remains a rather challenging aspect of T1D management. A main challenge in managing exercise in T1D is the high level of inter-individual and intra-individual variation in glycemia over any given exercise session. Glycemia at the initiation of exercise, including the time of day at which exercise is done, sensor glucose trend (if using a CGM), timing from the previous meal, insulin on board, carbohydrate content in the meal preceding exercise, type and duration of exercise, exposure to hypoglycemia in the previous 24 h and aerobic fitness are only some of the many factors that need to be considered to ensure that glycemic control remains stable during and after cessation of exercise.

In 2017, an international consensus statement for exercise management in T1D was published (96). This consensus is a unique resource which provides detailed glucose management strategies. Recommended adjustments to basal and prandial insulin, for both insulin pump and multiple daily insulin injection users, as well as carbohydrate intake requirements depending on the intensity and duration of activity are clearly presented. Quite importantly, the consensus also covers factors that would preclude exercise including the presence of elevated ketones, recent hypoglycemia, and diabetes-related complications which may be exacerbated in the context of vigorous exercise and/or competitive endurance events. We encourage the reader to refer to this publication for additional guidance.

Management of glycemia during and after exercise depends upon the type of exercise performed including time of day and the type of insulin the individual is using. CGM use helps fine-tuning of carbohydrate ingestion and insulin dose corrections if hyperglycemia results during exercise. However, sensors can be inaccurate during exercise and awareness of this is important. Highly individualized management strategies are required in each individual due to variability both within and between individuals. General glucose targets before exercise can be between ~90 mg/dl and 200 mg/dl, with values closer to ~90 mg/dl before resistance and HIIT exercise and closer to ~145 mg/dl for endurance exercise (97). Consumption of 15g carbohydrates when the glucose levels drop below 140 mg/dl in people at high risk of exercise-induced hypoglycemia or when they fall below 125 mg/dl in people not at high risk is recommended (98). CGM trend arrows are useful to determine timing and amount of carbohydrate ingestion.

With multiple daily injections, typically basal dose adjustments are not required unless the individual is undertaking new or unusual activity over several days (such as hiking) in which case a 20-25% dose reduction is warranted. Present day automated insulin delivery systems have the potential to improve glycemia during exercise. However, user actions are required to minimize the occurrence of exercise-induced hypoglycemia such as using a higher glucose target for exercise, exercising at a time of day with little to no insulin on board, ingesting carbohydrates if exercise duration is >60 min or if the glucose level drops below 125 mg/dl as measured by CGM. For example, in the Tandem t:slim X2 insulin pump with Control IQ, the standard target for regular activity is between 112.5 and 160 mg/dL and can be temporarily changed to 140-160 mg/dL. In addition, while using an automated insulin delivery system, if the insulin on board is high at the start of exercise because of a recent bolus or an increase in basal insulin delivery prior to exercise, carbohydrate ingestion during exercise should be initiated as soon as the CGM glucose is around 125 mg/dl.

Older Adults

Adults with T1D now span a very large age spectrum—from 18 to 100 years of age and beyond. These individuals are unique in that they usually have lived with a complex disease for many years (92). An understanding of each individual’s circumstances is vital and management often requires assessment of medical, functional, mental, and social domains. The ADA emphasizes that glycemic targets should be individualized with the goal of achieving the best possible control while minimizing the risk of severe hyperglycemia and hypoglycemia (99).

Glycemic goals in older adults: Most older adults with T1D have long-standing disease (unlike individuals with T2DM where diabetes can be long-standing or new onset). Additionally, there is a wide range of health in older individuals, with some patients enjoying good functional status and no comorbid conditions, while others are limited by multiple comorbidities as well as physical or cognitive impairments. Older patients with T1D may develop diabetes related complications which pose a challenge in disease management. Insulin dosing errors, hypoglycemia unawareness, and inability to manage hypoglycemia when it occurs may result from physical and cognitive decline. Special attention should be focused on meal planning and physical activities in this population. Severe hyperglycemia can lead to dehydration and hyperglycemic crises (92). Issues related to self-care capacity, mobility, and autonomy should be promptly addressed.

Thus, treatment goals should be reassessed and individualized based on patient factors. Older patients with long life expectancy and little comorbidity should have treatment targets similar to those of middle-aged or younger adults. In patients with multiple comorbid conditions, treatment targets may be relaxed, while avoiding symptomatic hyperglycemia or the risk of diabetic ketoacidosis (92). Therefore, it is important to assess the clinical needs of the patient, setting specific goals and expectations that may differ quite significantly between a healthy 24-year-old and a frail 82-year-old with retinopathy and cardiovascular disease.

There are few long-term studies in older adults demonstrating the benefits of intensive glycemic, blood pressure, and lipid control (99). As with younger adults, glycemic control should be assessed based on frequent SMBG levels and CGM data as well as A1C to help direct changes in therapy. More stringent A1C goals (~6.5-7%) can be recommended in select older adults if this can be achieved without hypoglycemia or other adverse effects. This is appropriate for older individuals with anticipated long-life expectancy, hypoglycemia awareness and no CVD. Less stringent A1C goals (for example A1C<8.5%) may be appropriate for patients with a history of severe hypoglycemia, hypoglycemia unawareness, limited life expectancy, advanced microvascular/macrovascular complications, or extensive comorbid conditions (92, 100).

Inpatient Management and Outpatient Procedures

The challenges involved in management of individuals with T1D in the hospital and in preparation for scheduled outpatient procedures include difficulties associated with fasting, maintaining a consistent source of carbohydrate, and facilitating inpatient blood glucose management while modifying scheduled insulin therapy. Individuals with T1D may have difficulty fasting for long periods of time (more than 10 h) prior to a procedure. Patients with T1D should be prepared with a treatment plan for insulin dose adjustments and oral glucose intake prior to any procedure that requires alterations in dietary intake and/or fasting.

In general, goals for blood glucose levels in individuals with T1D are the same as for people with T2D or hospital-related hyperglycemia (101). It is imperative that the entire health care team, including anesthesiologists and surgeons as well as other specialists who perform procedures, understand T1D and how it factors into the comprehensive delivery of care. First, the diagnosis of T1D should be clearly identified in the patient’s record. Second, the awareness that people with T1D will be at high risk for hypoglycemia during prolonged fasting and are at risk for ketosis if insulin is inappropriately withheld. Under anesthesia, individuals with T1D must be carefully monitored for hypoglycemia and hyperglycemia. Third, a plan for preventing and treating hypoglycemia should be established for each patient.

SMBG should be ordered to fit the patient’s usual insulin regimen with modifications as needed based on clinical status. Self-management in the hospital may be appropriate for some individuals with T1D including those who successfully manage their disease at home, have cognitive skills to perform necessary tasks such as administer insulin and perform SMBG, count carbohydrates and have a good understanding of their condition (101). For some individuals, once the most acute phase of an illness has resolved or improved, patients may be able to self-administer their prior multiple-dose or CSII insulin regimen under the guidance of hospital personnel who are knowledgeable in glycemic management. Individuals managed with insulin pumps and/or multiple-dose regimens with carbohydrate counting and correction dosing may be allowed to manage their own diabetes if this is what they desire, once they are capable of doing so.

The need for uninterrupted basal insulin to prevent hyperglycemia and ketoacidosis is important to recognize. Insulin dosing adjustments should also be made in the perioperative period and inpatient setting with consideration of oral intake and blood glucose trends.

Several studies have demonstrated that inpatient use of CGM has advantages over point of care glucose monitoring in detecting hypoglycemia, especially nocturnal, and/or asymptomatic hypoglycemia and in reducing recurrent hypoglycemia. Currently, the Endocrine Society recommends against the use of CGM alone in the intensive care unit or operating room settings due to limited available data on accuracy Although average daily glucose levels were comparable between CGM and capillary blood glucose testing, CGM detected a higher number of hypoglycemic episodes (55 vs 12, P < 0.01) suggesting that CGM may be beneficial for identification of hypoglycemia in the general ward particularly in patients with hypoglycemia unawareness. Diabetes self-management while acutely ill in the hospital may be appropriate for select individuals who prefer to perform self-care. Continuation of personal CGM device use in people with T1D and at increased risk for hypoglycemia during hospitalization is recommended. Large prospective randomized trials to establish benefit or lack thereof of CGM use on inpatient glycemic control are ongoing with most studies focusing primarily on T2D.

Pancreas Transplantation

Pancreas transplantation is a currently available therapeutic option for patients with diabetes who meet specific clinical criteria. Patients with end-stage renal disease are eligible to undergo simultaneous pancreas kidney (SPK) transplantation. Also, pancreas transplantation may be offered as a separate procedure after a patient has already received a kidney transplant (pancreas after kidney [PAK]). In addition, solitary pancreas transplantation may also be offered to those individuals presenting with severe metabolic complications attributed to poor glycemic control (pancreas transplant alone [PTA]). Pancreas transplantation procedures have been performed since the 1960’s. A 2011 update on Pancreas Transplantation from the International Pancreas Transplant Registry reported improvements in patient survival and graft function over a course of 24 years of pancreas transplantation (102). These improved outcomes were related to changes in surgical technique and immunosuppressive regimens as well as tighter donor selection criteria. At 5-years post-transplantation, pancreas graft survival is now reported at ~70% for SPK and at ~ 50% for PAK and PTA. Further, patient survival at 10 years exceeds 70% with the highest survival rate observed in PTA recipients (82%).

Islet Transplantation

Islet transplantation provides a less invasive surgical alternative for beta-cell replacement in patients with labile diabetes and has the potential to restore normoglycemia, eliminate severe hypoglycemia and restore hypoglycemia awareness. Marked improvements have also been noted in the field of islet transplantation over the past decade which have led to insulin independence rates at 5 years being comparable to pancreas transplantation outcomes (103). A pivotal study of islet transplantation in patients with T1D showed that at 1-year post transplant, 87% of study participants achieved the primary endpoint of a A1C <7.0% and freedom from severe hypoglycemia (from day 28 to 365) (104). These results paved the way for the 2023 FDA approval of Lantidra (donislecel), the first allogeneic pancreatic islet cellular therapy for the treatment of T1D complicated by problematic hypoglycemia (105). Further details about islet transplantation can be found in the dedicated Endotext chapter.

Artificial Pancreas Device Systems - Closed Loop Systems

In addition to insulin-only CL-systems, bi-hormonal closed loop systems are also being actively explored. Additional manufacturers utilizing insulin-only CL-systems are expected to launch their devices in the near future. The development of faster-acting insulins could potentially make these strategies more effective. As this technology advances, we are getting closer to the goal of a fully automated device which will be able to predict with high accuracy changes in glucose profiles and respond accordingly with stringent modulation of infusion of hormones (e.g. insulin, glucagon, amylin) to maintain glycemia within normal ranges.

Implantation of Encapsulated Islets

Some of the limitations of islet transplantation currently include the limited availability of donors and the need for long term immunosuppression to prevent rejection of the transplanted graft. Protecting the islets from the immunologic environment may allow both the use of non-human islets for transplantation and minimize or eliminate the need for systemic immunosuppression. Thus, the encapsulation of islets to attain these goals has been sought for several years but unfortunately this technology is still to make it to the clinical arena. Although initial attempts at encapsulation of islets resulted in damage of the capsule by local tissue responses, newer techniques allowing for conformal coating of human islets have shown promising results in pre-clinical models and are being explored (106).

Islet Xenotransplantation

An alternative to human pancreas and islet transplantation which is currently being explored is the use of pig islets. Pig islets have major physiologic similarities to human islets. Notably, pig insulin differs from human insulin by only one amino acid. Donor pigs may be genetically engineered to be protected from the human immune system thus reducing the need for potent immunosuppression. Studies in non-human primates using encapsulated pig islets have resulted in graft survival for more than 6 months (107). Research in this field in actively ongoing.

Stem Cell Based Therapies

Stem cell research has allowed the generation of insulin-producing pancreatic β-cells from human pluripotent stem cells (108). This technology has the potential to generate vast amounts of glucose-responsive β-cells and allow for the development of customizable islets containing predetermined amounts of specific cell lines. The VX-880-101 FORWARD study evaluated the safety and efficacy of zimislecel (VX-880), a cellular treatment composed of allogeneic stem cell-derived, fully differentiated islets, for the management of T1D (109). Eligible participants were adults between the ages of 18 and 65 years and required to have at least 5 years of insulin dependence, impaired hypoglycemia awareness, and 2 or more episodes of severe hypoglycemia in the previous year. In part A, participants received a half dose of zimislecel (0.4x10⁹ cells) as a single intraportal infusion. In parts B and C, participants received a full dose of zimislecel (0.8x10⁹ cells) as a single intraportal infusion. All participants received corticosteroid-free immunosuppression for prevention of allorejection. At 12 months follow up, 12 participants who received the full dose of zimislecel were free of severe hypoglycemic events, had a group mean HbA1c of <7.0% with a mean reduction from baseline of 1.8 percentage points, and mean time spent in glucose target range of 70-180 mg/dl was >90%. Further, 10 of these participants were insulin independent. Of note, 2 deaths were reported during the conduct of the study, one caused by cryptococcal meningitis and one by severe dementia with agitation owing to the progression of preexisting neurocognitive impairment. These metabolic results are highly encouraging demonstrating the utility and efficacy of stem cell derived islets for metabolic control in patients with unstable T1D. However, these data also highlight the risks associated with immunosuppression and the need for adequate patient selection for T1D cellular studies requiring the use of long-term immunosuppression. Longer follow up and more patients are required to adequately evaluate the safety, efficacy and durability of this cellular product.

Gene Edited/Hypoimmune Islets

Gene editing islets to allow for avoidance of T-cell mediated rejection as well as protection from innate rejection is an area of ongoing research. Such technology would allow for transplantation of an allogeneic cellular product to treat T1D without the need of immunosuppression. A recent proof-of-concept study evaluated the transplantation of genetically modified allogeneic donor islet cells into a man with long-standing T1D (110). The authors used clustered regularly interspaced short palindromic repeats (CRISPR)–CRISPR-associated protein 12b (Cas12b) editing and lentiviral transduction to genetically edit the cells to avoid rejection. The final cellular product (UP421) contained fully edited HLA-depleted hypoimmune islet cells with high CD47 expression, some HLA class I and II double-knockout cells with endogenous CD47 levels, and islet cells with retained HLA expression (wild type) and varying CD47 levels. The cells were transplanted in the forearm muscle of the patient without immunosuppressive therapy. By 12 weeks, there was no immune response against the gene-edited cells. Although marginal, C-peptide levels were stable and there was evidence of glucose-responsive insulin secretion. These preliminary results provide evidence that immune evasion with the use of a gene edited cellular product is feasible and may serve as an alternative for the prevention of allograft rejection.

Glucose Responsive Insulins (Smart Insulins)

Another area of ongoing research is the development of “smart” drug delivery systems able to respond to environmental or external triggers greatly improving therapeutic performance. Conceptually, “smart” insulins should be able to respond to changes in ambient glucose which would dictate activation or cessation of insulin delivery. Several efforts have been made to generate glucose-responsive insulin delivery systems, and some have shown promising results in pre-clinical studies including the utilization of enzymatic triggers, glucose-binding proteins, and synthetic molecules able to bind to glucose. However, current limitations include the potential for immunogenicity and poor glucose selectivity (111). Continued progress in this field in the coming years to reduce the burden of diabetes is anticipated.