INTRODUCTION — Insulin therapy is the mainstay of treatment for type 1 diabetes mellitus (T1DM). The goal of insulin therapy is to replace the deficient hormone and attain normoglycemia. However, this goal remains elusive because of the difficulty in replicating the minute-to-minute variations of endogenous (physiologic) insulin secretion directly into the portal vein versus delivery of exogenous insulin, which is absorbed from the subcutaneous injection or infusion site into the systemic circulation. The acute and chronic complications of diabetes are largely attributable to hyperglycemia as a consequence of the failure of exogenous insulin delivery to completely mimic physiologic insulin secretion.
Many different insulin preparations and delivery systems are available. The selected regimen is individualized for the child and family/caregiver to fit their lifestyle and optimize adherence while providing optimal glycemic control. Input from the patient, if age-appropriate, and family regarding timing of meals and snacks, school/daycare, and physical activity is important to ensure optimal glycemic control and minimize glycemic variability and episodes of hypoglycemia. As a result, the optimal types of insulin and regimen will vary among children and can change for an individual child over time.
This topic review will focus on monitoring of glycemic control and details of insulin therapy, including dosing and dose adjustment, as well as options for administration, including insulin pumps and related devices.
Other aspects of childhood-onset T1DM are discussed separately:
●Routine management:
•(See "Overview of the management of type 1 diabetes mellitus in children and adolescents".)
•(See "Epidemiology, presentation, and diagnosis of type 1 diabetes mellitus in children and adolescents".)
•(See "Complications and screening in children and adolescents with type 1 diabetes mellitus".)
•(See "Management of exercise for children and adolescents with type 1 diabetes mellitus".)
•(See "Management of type 1 diabetes mellitus in children during illness, procedures, school, or travel".)
●Prevention and management of acute glycemic emergencies:
•(See "Hypoglycemia in children and adolescents with type 1 diabetes mellitus".)
•(See "Diabetic ketoacidosis in children: Clinical features and diagnosis".)
•(See "Diabetic ketoacidosis in children: Treatment and complications".)
•(See "Diabetic ketoacidosis in children: Cerebral injury (cerebral edema)".)
MONITORING GLYCEMIC CONTROL — The goal of management in both children and adults is to maintain glucose control as near to normal as safely possible, by balancing the risks of long-term complications of diabetes against the short-term risks of hypoglycemia and diabetic ketoacidosis (DKA). General targets are the same for all age groups but may vary depending on individual patient characteristics.
Monitoring glycemic control has three components:
●Hemoglobin A1c – Hemoglobin A1c (A1C; also called blood glycated hemoglobin or glycohemoglobin) is used to evaluate long-term glycemic control. (See 'Target for hemoglobin A1c' below.)
●Blood glucose – Frequent measurements of blood glucose are used to monitor for hypo- and hyperglycemia and are used to adjust insulin dosing and carbohydrate intake. These measurements can be taken by capillary sampling (fingerstick) and measurement with a home blood glucose meter, or with a device for continuous subcutaneous glucose monitoring. (See 'Blood glucose monitoring' below.)
●Ketones – Testing for ketones in blood or urine should be performed if marked and persistent hyperglycemia develops. Early detection and management of ketosis can help prevent or abort DKA. (See 'Testing for ketosis' below.)
Target for hemoglobin A1c — We suggest a target A1C of <7 percent (53 mmol/mol) for most children and adolescents who have access to comprehensive diabetes care, consistent with guidelines from the American Diabetes Association (ADA) and the International Society for Pediatric and Adolescent Diabetes [1,2]. To achieve a target A1C of <7 percent, target blood glucose levels are approximately 80 to 130 mg/dL (4.4 to 7.2 mmol/L) before meals and 80 to 140 mg/dL (4.4 to 7.8 mmol/L) at bedtime and overnight. The ADA emphasizes that glycemic targets should be further tailored to the individual patient [2]:
●A less stringent goal of A1C <7.5 percent may be appropriate for younger patients and those with underlying conditions that limit their ability to articulate symptoms of hypoglycemia; those with hypoglycemia unawareness; those without access to analog insulin formulations, a continuous glucose monitor, or advanced insulin delivery technology; those with known micro- and macrovascular complications or lifestyle or psychosocial considerations; or those whose A1C does not reflect their overall glycemic control (such as patients who are "high glycators").
●Even less stringent A1C goals (such as <8 percent) may be appropriate for patients with a history of severe hypoglycemia, limited life expectancy, or extensive comorbid conditions.
●In contrast, a goal of A1C <6.5 percent is suggested by the ADA if this can be achieved without severe hypoglycemia or excessive burden of care, during the honeymoon period, and in those whose A1C is lower than expected because of conditions that lead to more rapid red cell turnover.
In the past, the ADA set higher A1C targets for young children because of concerns that more stringent targets might increase the risk for hypoglycemic episodes. However, clinical experience with modern management methods suggests that the target of <7 percent does not significantly increase the risk for severe hypoglycemia. Moreover, increasing evidence suggests that hyperglycemia in children is associated with micro- and macrovascular complications, including adverse effects on the brain of the prepubertal child [3-6].
It is important to recognize that patients with erratic diabetes control and wide glycemic excursions may achieve an A1C similar to that of patients with more stable glucose levels if the mean blood glucose levels are similar. Reducing glycemic variability is an important goal of therapy because it reduces the risk for hypoglycemia and may also be important for preventing long-term complications [7]. Thus, serial measurements of blood glucose, in addition to A1C, are needed for a complete assessment of glycemic control. (See 'Blood glucose monitoring' below.)
The same A1C target of <7 percent (53 mmol/mol) is also recommended for most adults with T1DM. Even lower A1C levels can be attained using newer insulins, smart pumps, continuous glucose sensors, and algorithm-controlled insulin delivery without increasing the risk of severe hypoglycemia [8-10]. Although there is a continued relative risk reduction of retinopathy and other complications at A1C values below 7 percent, the absolute risk of developing these complications is low if the A1C value is kept below this threshold (figure 1). (See "Glycemic control and vascular complications in type 1 diabetes mellitus", section on 'Glycated hemoglobin (A1C)'.)
Blood glucose monitoring — Frequent self-monitoring of blood glucose (SMBG) is critical for management of T1DM to ensure:
●Overall glycemic control – Optimal glycemic control requires frequent SMBG and appropriate adjustment of the insulin dose. Serial monitoring allows the child and family/caregiver to become familiar with the patient's glycemic response to different types and amount of food, exercise, and stress. Frequent monitoring has been shown to improve glycemic control in children [3,11,12].
●Detect hypoglycemia and hyperglycemia – SMBG is also important to detect acute episodes of hyper- and hypoglycemia [13]. Children and their parents/caregivers show poor ability to detect high or low blood glucose levels based on symptoms alone. In a study that compared self-reporting of hypoglycemia with glucose meter results, children aged 6 to 11 years and their parents failed to detect approximately one-half of the episodes of clinically significant hypoglycemia (<55 mg/dL) [14]. Episodes of hyperglycemia are even more likely to go unrecognized by the patient; regular SMBG helps detect these episodes, prompting both adjustments in insulin dosing and monitoring for ketosis that can prevent or abort DKA. (See 'Testing for ketosis' below.)
Unfortunately, adherence to a schedule of SMBG is often suboptimal. Adherence is highest in children under the age of six years and decreases with increasing age [11]. Predictors of a lower frequency of SMBG include lower self-esteem, high-stress life events, and inadequate parental support [15]. Psychosocial support and parental supervision may help optimize the frequency of SMBG in older children.
Options for SMBG include multiple daily fingersticks or use of one of several types of devices for continuous glucose monitoring (CGM). Selection of the method depends upon the patient's needs and preferences and device availability; considerations for each method are outlined below. (See 'Fingersticks' below and 'Continuous glucose monitoring' below.)
Fingersticks — Blood glucose should be tested at least four times per day (in the fasting state, before meals, and at bedtime) [1,3]. More frequent monitoring is often required, depending on age, patterns of food intake, and exercise activity, as well as during illness. This is especially true in very young children with variable feeding and activity patterns and in patients who have multiple meals and snacks during the day and require blood glucose checks before administration of a premeal bolus of insulin. Patients also need blood glucose monitoring before, during, and after periods of increased physical activity to safeguard against hypoglycemia. (See "Management of exercise for children and adolescents with type 1 diabetes mellitus".)
Blood glucose is typically monitored using a blood glucose meter, which requires a small sample of blood (0.3 to 1 microliter) obtained by fingerstick. Many new lancet devices are available, some of which are of ultrafine gauge; others contain multiple lancets for ease and safety of handling.
Continuous glucose monitoring — For most children, we suggest CGM rather than fingersticks for blood glucose monitoring, if such a device is available and the patient or caregivers are able to use the device safely. This technology helps to optimize glycemic control for many patients, and is particularly useful for those with a history of hypoglycemia unawareness [2,3,16].
●Technology – CGM is performed with a subcutaneous sensor that continuously measures interstitial fluid glucose levels. Most CGM devices available for clinical use give real-time feedback to the patient and family/caregiver. They do not directly control insulin administration but provide glucose readings to permit finer control of insulin administration by patients and their families. This type of CGM also provides information about rates of change of rising or falling glucose levels and has alarms that alert patients when glucose levels reach predefined high and low thresholds [17,18]. While earlier versions of the CGM devices were used in conjunction with conventional capillary (fingerstick) blood glucose monitoring for calibration, newer devices do not require such calibration [19]. However, confirmatory fingerstick blood glucose measurements may also be required when the patient's symptoms are not consistent with the sensor glucose readings and, sometimes, during sensor warm-up periods.
CGM output typically includes average sensor glucose, time in range (proportion of time that CGM readings are within the target range, ie, 70 to 180 mg/dL), time above range (proportion of time that CGM readings are >180 mg/dL), time below range (proportion of time that CGM readings are <70 mg/dL [low] or <54 mg/dL [very low]), glycemic variability, and a glucose management indicator. The device also provides information such as the number of days that the CGM is used and percentage of time that the CGM is active [20].
Time in range in increasingly being used as a therapeutic goal to assess the efficacy of the insulin regimen, in addition to the A1C level. Time in range is associated with lower risk of microvascular complications and can be used to assess glycemic control. The target range is typically set as 70 to 180 mg/dL, but some patients striving for tighter control may set a lower and narrower target range, such as 70 to 140 mg/dL. Time below range (blood glucose <70 mg/dL and <54 mg/dL) or time above range (>180 mg/dL) are useful parameters to prompt reevaluation of the insulin regimen [20].
●Efficacy – CGM devices have moderate efficacy for patients who are willing to use them consistently. In a randomized 26-week trial in 153 adolescents and young adults, participants managed with CGM had a small but significant improvement in glycemic control (A1C 8.9 percent at baseline and 8.5 percent at 26 weeks), whereas those managed with standard fingerstick blood glucose monitoring had no improvement (A1C 8.9 percent at baseline and at 26 weeks) [21]. Mean time in the target range (blood glucose 70 to 180 mg/dL) and satisfaction with the CGM system were significantly higher for patients managed with CGM compared with blood glucose monitoring. This study selected patients with reasonably good adherence to blood glucose monitoring and glycemic control (A1C 7.5 to 10.9 percent), so the findings may not be applicable to a broader population of adolescents who might be less adherent than the cohort that was studied. A randomized trial in 143 younger children (two to eight years old) found that, while CGM use did not alter time spent in target range, it did reduce the time spent in the hypoglycemic range by approximately 40 minutes daily, reduced glucose variability, and improved parents' well-being [22].
Adherence to and efficacy of CGM in adolescents is generally lower than in adults, but advances in this technology have lowered the burden of using these devices and may improve usage in less adherent populations.
The advantages and disadvantages of CGM systems and the experience with these devices in adults are reviewed separately. (See "Glucose monitoring in the ambulatory management of nonpregnant adults with diabetes mellitus", section on 'Benefits of CGM' and "Glucose monitoring in the ambulatory management of nonpregnant adults with diabetes mellitus", section on 'CGM systems'.)
CGM can be incorporated into insulin pump therapy. Depending on the system, the sensor glucose values are used to adjust the insulin dose manually, semi-automatically, or automatically. The types of devices are described below. (See 'Insulin pumps with glucose sensors' below and 'Closed-loop insulin pumps' below.)
Testing for ketosis — To prevent or abort the development of DKA, patients must check for urine or blood ketones when:
●Blood glucose is persistently ≥250 mg/dL (13.9 mmol/L)
●During acute episodes of increased stress, including intercurrent illnesses (especially with nausea, vomiting, abdominal pain, or prolonged fasting), even in the absence of marked hyperglycemia [23]
●Fruity breath odor (caused by exhaled acetone in patients with ketosis)
For monitoring ketones, the preferred method is home testing of blood for beta-hydroxybutyrate (BOHB). Compared with urinary ketones, BOHB permits earlier detection of ketosis and more accurately reflects its severity and resolution.
Education of patients and families should include information about when to call the diabetes care team for assistance. If ketones are detected, typical next steps are:
●Patients who have hyperglycemia and mild ketosis (small urine ketones or BOHB 0.6 to 1.4 mmol/L) should be treated with additional insulin (with or without additional carbohydrates) and increased fluid intake, combined with more frequent monitoring of blood glucose and ketone concentrations, in consultation with their diabetes clinician. With these interventions, most cases of DKA can be avoided or aborted. Details are discussed in a separate topic review. (See "Management of type 1 diabetes mellitus in children during illness, procedures, school, or travel", section on 'Sick-day management'.)
●Patients with persistent hyperglycemia and moderate or severe ketosis (blood BOHB ≥1.5 mmol/L or moderate-large urine ketones) should urgently consult with their diabetes clinician or seek emergency care for evaluation and management of impending DKA. Many of these patients can be managed in the home setting as long as they are in close communication with a diabetes clinician until ketosis has resolved. BOHB >5 mmol/L warrants close attention because it is a strong predictor of DKA [24]. (See "Diabetic ketoacidosis in children: Clinical features and diagnosis".)
INSULIN PREPARATIONS — Insulin types can be classified by their onset and duration of action (table 1).
Commonly used insulins include:
●Rapid-acting insulins (eg, lispro, aspart, glulisine) are typically administered as a premeal bolus, given 5 to 15 minutes before the meal. The dose varies with the carbohydrate content of food and the blood glucose level. For younger children in whom intake is unpredictable, rapid-acting insulin can be administered after the meal if necessary. However, the goal should be to administer the insulin dose as soon as the parent or caregiver knows how many grams of carbohydrate the child will reliably eat, in order to avoid carbohydrate-insulin mismatch and high followed by low blood glucose levels.
Insulin pump therapy uses a rapid-acting insulin to provide basal insulin levels and also for prandial boluses or correction doses for hyperglycemia. (See 'Insulin pump' below.)
●Long-acting insulin preparations (eg, insulin glargine, insulin detemir) are given once or twice a day. Insulin degludec has a longer duration of action but is still given once a day [25] (see "Management of blood glucose in adults with type 1 diabetes mellitus", section on 'Basal insulin options'). In general, if a single injection is used, it should be given in the evening in order to assure insulin availability during the night and counteract the counterregulatory hormone response in the early hours of the morning. However, some very young children who are at greater risk of hypoglycemia do better with administration of long-acting insulin in the morning hours.
Less commonly used insulins include:
●Short-acting insulins (eg, regular insulin) are no longer included in the routine management of T1DM in the United States. When used for premeal boluses, it is typically administered 20 to 30 minutes before meals. This type of insulin can also be helpful under special circumstances such as gastroparesis and intake of high-fat meals, when mixed with rapid-acting insulin. Further, short-acting insulin is sometimes used in conjunction with intermediate-acting insulin for management of overnight continuous feeds in patients with cystic fibrosis-related diabetes, although the insulin pump is becoming increasingly popular in these patients.
●Intermediate-acting NPH (neutral protamine hagedorn) insulin is usually given in a targeted manner in combination with long-acting insulins. The intermediate-acting insulin provides some coverage for meals. For example, NPH insulin given before breakfast will cover lunch, which may be useful for young children who do not have a school nurse to administer rapid-acting insulin before lunch. The use of intermediate-acting insulin fell out of favor once long-acting insulin preparations became available, given the higher risk of hypoglycemia with the former, particularly when food intake is not timed to match the peak of insulin action. Intermediate-acting insulin is occasionally still used in special situations, such as in combination with short-acting insulin to cover overnight continuous feeds in patients with cystic fibrosis-related diabetes or as a fixed combination with short- or rapid-acting insulin for twice-daily administration in patients poorly adherent to more frequent insulin administration regimens. It is also often used in resource-limited situations because it is much less expensive.
A more complete discussion on insulin preparations is found separately. (See "Management of blood glucose in adults with type 1 diabetes mellitus", section on 'Choice of insulin delivery'.)
INSULIN DELIVERY — Insulin is administered by needle and syringe, pen, or pump.
●Needle and syringe – An advantage of needle and syringe is that NPH (neutral protamine hagedorn) and short- or rapid-acting insulins can be mixed in a single injection, thereby reducing the number of injections. However, insulin glargine and detemir cannot be mixed with any other form of insulin and must be administered separately. The smallest needle available is 31 gauge and 6 mm (15/64") in length. This small size is used for most patients and is particularly helpful for younger children. Syringes are available in 30 (0.3 mL), 50 (0.5 mL), and 100 (1 mL) unit sizes. Thirty-unit syringes marked to administer one-half units are available.
In infants and toddlers who receive their insulin by syringe, the insulin dose may be so small that dilution is required to allow for easier and more precise administration. The smallest dose of insulin that can be accurately administered without dilution using a syringe is 0.5 units. Many insulins can be diluted either at a specialized pharmacy or at home with proper training. Specific diluent for many insulin preparations is available from the insulin manufacturer. Some insulin pumps can deliver much smaller doses of insulin, of the order of 0.025 units at a time, often obviating this problem.
●Insulin pen – Pens are supplied prefilled with insulin and may be either disposable or reusable. The ease of use and portability of pens are appealing to many patients. The smallest needle available is 33 gauge and 4 mm in length, and the smaller needles are preferred for younger children. Although mixed insulin preparations are available in pens, these are not tailored to the needs of children. Mixed insulin pens are usually reserved for individuals who are limited in their ability to make dosing decisions and, as a result, glycemia is not usually as strictly controlled. Pens that deliver aspart and lispro insulin are available that offer the flexibility of one-half unit delivery.
●Insulin pumps – Insulin pumps (continuous subcutaneous insulin infusion) are increasingly used for pediatric and adult patients with T1DM, with or without integrated continuous glucose sensor. The considerations for initiating insulin pump therapy are discussed below. (See 'Insulin pump' below.)
VALUE OF AN INTENSIVE REGIMEN — Insulin management can be categorized as "intensive" or "conventional," depending on the frequency and type of insulin dosing. In general, intensive regimens are recommended because they are more likely to meet glycated hemoglobin (A1C) targets and have better clinical outcomes [3].
●An intensive regimen provides insulin in a manner that more closely approaches physiologic insulin secretion than does conventional therapy. It consists of a basal dose of insulin to suppress lipolysis and hepatic glucose production, with additional premeal and pre-snack boluses of rapid-acting insulin to minimize postprandial elevation of blood glucose. The prandial boluses are adjusted according to the carbohydrate content of meals as well as the measured current blood glucose level. This approach allows greater flexibility than the conventional regimen in terms of timing and carbohydrate content of meals. Insulin regimens administered using an insulin pump are also considered intensive. (See 'Insulin pump' below.)
●A conventional regimen was traditionally used for diabetes management, but this approach is uncommon today because it is unlikely to achieve optimal glycemic control. This type of regimen typically consisted of fixed doses of an intermediate-acting insulin (NPH [neutral protamine hagedorn]) at least twice a day (at breakfast and a second dose either at dinner or bedtime), with a rapid-acting (ie, lispro, aspart, or glulisine) or short-acting ("regular") insulin given two or three times a day. Because the doses are fixed, this type of regimen requires lifestyle adjustment so that meals and vigorous physical activity occur on a relatively fixed daily schedule. Two-thirds of the total daily dose is administered before breakfast (two-thirds as NPH and one-third as rapid- or short-acting insulin) and one-third before dinner and at bedtime (one-third to one-half as rapid- or short-acting insulin before dinner and two-thirds to one-half as NPH at bedtime). The dose of NPH can be split such that a portion is delivered before dinner and the remainder at bedtime.
●A modified-intensive regimen is sometimes used in children and offers the benefit of fewer injections during school hours (useful when a school nurse is not available) and flexibility around meals while at home. Such a regimen includes the administration of an intermediate-acting insulin (NPH) with rapid-acting insulin (based on carbohydrate counting) at breakfast and insulin glargine or detemir with a dose of rapid-acting insulin (based on carbohydrate counting) at dinner. This regimen avoids the need for a lunch-time dose of rapid-acting insulin.
Intensive insulin regimens achieve better glycemic control and reduce the incidence of vascular complications compared with conventional regimens, as shown by studies in children [26-29], adolescents [30,31], and adults (figure 2 and figure 3). (See "Glycemic control and vascular complications in type 1 diabetes mellitus", section on 'Benefits of intensive glycemic control'.)
TYPES OF INTENSIVE REGIMENS — An intensive regimen is delivered either by multiple daily injections (MDI) or by insulin pump (continuous subcutaneous insulin infusion). The choice of intensive regimen depends upon the needs and preferences of the patient and family, cost considerations, and clinician preferences.
Multiple daily injections — The MDI regimen combines a basal dose(s) of insulin using a long-acting insulin analog with prandial boluses of rapid-acting insulin before each meal and snack. This approach results in more stable glycemic control and fewer episodes of hypoglycemia than "conventional" insulin regimens in children and adolescents. (See 'Value of an intensive regimen' above.)
●Basal insulin – MDI regimens using long-acting basal insulin analogs in children should be initiated and supervised by an experienced pediatric diabetes team. Insulin glargine is the long-acting analog most commonly used in pediatric patients (table 1). It has a duration of action of up to 24 hours and is administered once a day. However, the half-life is shorter in some patients, requiring division of the daily dose into two injections per day. Insulin detemir may also be used. The dose depends on the child's size and is further adjusted to optimize glycemic control. Insulin glargine is US Food and Drug Administration (FDA)-approved for use in children ≥6 years, while insulin detemir is FDA-approved for use in children ≥2 years. However, both are commonly used off-label in younger children.
Studies have explored the possibility of an association between the use of glargine, which has a modified structure similar to insulin-like growth factor 1 (IGF-1), and risk of malignancy. Most experts feel that this is not a concern [32,33].
●Prandial insulin – Prandial doses of a rapid-acting insulin are given before each meal or snack. The dose varies from meal to meal and day to day, depending upon the estimated amount of carbohydrates to be consumed, as well as on the premeal blood glucose level and recent or anticipated exercise. In young children with variable and unpredictable food intake during meals and snacks, prandial insulin is often administered immediately after the meal or snack so that the dose can be adjusted to reflect the actual amount of carbohydrates consumed. This strategy is not optimal for blood glucose control and should be converted to premeal boluses as soon as is developmentally possible.
Dosing of basal and prandial insulin is discussed below. (See 'Insulin dosing' below.)
Insulin pump — The insulin pump (continuous subcutaneous insulin infusion) is increasingly the standard of care for the pediatric population in resource-rich settings, based on accumulating evidence for benefits on glycemic control, safety, and patient satisfaction compared with MDI for most pediatric patients.
●Indications – A position statement of the American Diabetes Association (ADA), European Society for Pediatric Endocrinology, and others recommends that insulin pump therapy should be considered for patients with one or more of the following characteristics [34]:
•Recurrent severe hypoglycemia
•Wide fluctuations in blood glucose levels (regardless of glycated hemoglobin [A1C])
•Suboptimal diabetes control (A1C exceeds target range for age)
•Microvascular complications and/or risk factors for macrovascular complications
•Good metabolic control, but insulin regimen that compromises lifestyle
The International Society for Pediatric and Adolescent Diabetes particularly recommends insulin pump therapy for preschool-aged children [35,36]. Insulin pump therapy may also be useful for infants, adolescents with eating disorders, pregnant adolescents, ketosis-prone individuals, and competitive athletes [34]. These groups of patients tend to have variable and sometimes unpredictable insulin requirements and may benefit from the flexibility of dosing offered by an insulin pump.
●Technique – Insulin pumps deliver a basal rate (small aliquots infused every few minutes, evenly spaced over an hour) of rapid-acting (or rarely, short-acting) insulin subcutaneously. In addition, premeal/pre-snack boluses are administered to minimize increases in postprandial glucose concentrations. Insulin is delivered through a subcutaneously inserted catheter that is replaced at two- to three-day intervals. Information on insulin administration, frequency of catheter site changes, and frequency and timing of premeal insulin boluses can be downloaded from a memory chip within the pump.
Insulin pump therapy relies on frequent blood glucose monitoring with appropriate adjustment of insulin infusion rates by the patient or parent/caregiver. The pump also can be used in conjunction with a continuous glucose monitoring (CGM) device; this approach is known as sensor-augmented insulin pump therapy. These systems are described briefly below and in more detail separately. (See 'Insulin pumps with glucose sensors' below and 'Closed-loop insulin pumps' below and "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Types of insulin pumps'.)
●Efficacy – Considerable clinical evidence suggests that insulin pump therapy is similar to or somewhat better than MDI in achieving glycemic control and avoiding hypoglycemic episodes [9,37,38]. In a 2019 meta-analysis of randomized trials involving children and adolescents, insulin pump therapy modestly reduced A1C compared with MDI (-0.23 percent, 95% CI -0.56 to -0.11 percent) [38]. Severe hypoglycemia occurred less frequently in the insulin pump group, but the finding was not statistically significant (15 versus 22 percent; odds ratio 0.70, 95% CI 0.44-1.12). Episodes of diabetic ketoacidosis (DKA) were uncommon in both groups (2 versus 1 percent, respectively). Successful use of insulin pump therapy has been reported in children as young as two years of age [28,39-41].
In a trial that enrolled children with newly diagnosed diabetes (294 children aged 7 months to 15 years in the United Kingdom) randomly assigned to either insulin pump therapy or MDI [42], glycemic control (A1C) was poor in both treatment groups and did not differ between groups. Moreover, children treated with an insulin pump had higher resource utilization, including emergency department visits and annual total costs, compared with MDI. These findings contrast with meta-analyses in children who have had diabetes for a longer duration [38], suggesting that many families may need a period of acclimatization to diagnosis and management before adopting pump therapy. Furthermore, the impact of pump therapy on A1C reduction may be greater later in the course of diabetes, particularly past the "honeymoon" phase.
Limited data suggest that insulin pump therapy may reduce the risk of microvascular diabetes complications or associated risk factors compared with MDI [37,43].
●Patient and family satisfaction and adherence – Insulin pump therapy is often preferred by children and their families over MDI regimens. MDI regimens can require as many as six to seven injections per day, which may be a barrier for some patients and families. Moreover, the insulin pump offers increased flexibility, particularly with regards to meals and socialization, and ease in adjusting insulin doses to match changes in food intake and physical activity [44]. Thus, the insulin pump can be an attractive option for intensive therapy at any age and appears to improve quality of life in most children and adolescents, as suggested by multiple observational studies and randomized trials [40,41,45-48]. Of note, most families with young children who participated in clinical studies decided to continue with pump therapy after the study was completed [49].
Like MDI therapy, insulin pump therapy requires frequent blood glucose monitoring, counting dietary carbohydrates, judging the impact of exercise on insulin requirements, and making the appropriate adjustments to insulin infusion rates. Without this commitment, the benefits of this regimen are not attained. Before initiation of insulin pump therapy, the patient and family must be informed and accept the significant work required by this therapeutic approach, as well as given the training and support required to implement it. The beneficial effect of the insulin pump is maintained only if the child and family continue to devote time and effort to glycemic monitoring and control.
Although multiple studies have demonstrated an improvement of diabetes control with the insulin pump compared with MDI, decreased adherence to the pump protocol can occur over time and is associated with deterioration in control. Parental involvement during this time may help offset this risk in adolescents [50].
●Other considerations – Other considerations regarding the choice of regimen include the greater cost of the insulin pump and its supplies compared with those of syringes and needles used in MDI therapy, as well as potential complications of pump therapy, such as infusion set failure, superficial infection at the site of placement of the infusion set, and minor dermatologic changes such as nodules or scars at the catheter site [51,52]. Because only rapid- or short-acting insulin is used in insulin pumps and patients have no long-acting subcutaneous depot of insulin, failure of insulin delivery can result in rapid onset of DKA. This is another reason why frequent blood glucose checking is essential for children on insulin pump therapy [53].
The experience with insulin pumps in adults is discussed separately. (See "Continuous subcutaneous insulin infusion (insulin pump)".)
Insulin pumps with glucose sensors
●Sensor-augmented insulin pump – This refers to CGM used in conjunction with an insulin pump; the patient or caregiver manually adjusts the insulin dose based on the CGM results. This approach generally leads to improved glycemic control but requires a highly motivated user and good diabetes management skills. This was shown in randomized trials in children that compared sensor-augmented pump therapy with MDI [54,55] or with insulin pump therapy with conventional monitoring of blood glucose via fingerstick [56]. (See "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Sensor-augmented insulin pump'.)
●Sensor-augmented insulin pump with threshold suspend – This is a partially closed-loop glucose monitoring and insulin infusion system that automatically stops the insulin infusion in response to a selected low blood glucose threshold or when an inbuilt algorithm predicts impending hypoglycemia. The device is also known as a predictive low-glucose suspend (PLGS) system and has been approved by the FDA [57]. Clinical studies showed that the PLGS system was effective in reducing nocturnal hypoglycemia by 50 to 54 percent in children 4 to 14 years old and by 80 percent in individuals 15 to 45 years old, without an increase in risk of ketosis [58,59]. In a randomized trial in children and adolescents, use of this device resulted in an almost twofold reduction in hypoglycemic events compared with use of a standard sensor-augmented insulin pump [60]. There was also a trend toward improved hypoglycemia awareness, which did not reach statistical significance. This device was a very welcome addition to the armamentarium but is not a complete closed-loop system. (See "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Sensor-augmented insulin pump'.)
Closed-loop insulin pumps
●Hybrid closed-loop (HCL) system – For most children, we suggest an HCL system, if available, rather than other continuous insulin delivery systems. In clinical trials, HCL systems modestly improved glycemic outcomes compared with sensor-augmented insulin pumps. In our clinical experience, most children and families have high satisfaction with newer HCL devices and can learn to use them effectively with appropriate training and support.
An HCL system automatically increases, decreases, and suspends basal insulin delivery in response to CGM and predicts sensor glucose in the next 30 minutes. Newer systems are also able to administer a correction bolus (no more than once an hour) if sensor glucose is predicted to be above a particular number (eg, 180 mg/dL) in the next 30 minutes, based on the programmed insulin sensitivity. If certain parameters are met that predict low blood glucose, the insulin infusion is automatically suspended. These systems still require user input to administer an insulin bolus (based on the patient's insulin:carbohydrate ratio) before meals. Optional settings are now available for sleep and exercise.
While this type of device employs the insulin:carbohydrate ratio used in traditional insulin pump systems, one system allows modification of the "active insulin time" but not the insulin sensitivity (correction) factor, while another still uses the insulin sensitivity factor. One system automatically reverts to manual pump mode (ie, an open-loop system that requires user input) if blood glucose values are high or low for specified periods of time or when there are difficulties calibrating the system. These systems can also be used as open-loop systems if not used with CGM. (See "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Types of insulin pumps'.)
Key studies of these HCL devices in children include:
•Adolescents >14 years – In a six-month randomized trial, use of an HCL system compared with a sensor-augmented insulin pump increased time in glycemic target range (71 versus 59 percent; risk-adjusted difference 11 percent, 95% CI 9-14), improved A1C levels, and modestly reduced time spent in a hypoglycemic state (eg, <54 mg/dL, 0.29 versus 0.35 percent) [61]. There were, however, more hyperglycemic adverse reactions, including one episode of ketoacidosis, in the closed-loop group (14 versus 2 patients), primarily due to infusion set failures. A similar randomized trial that primarily enrolled adolescents compared HCL with standard therapy (insulin pump or other) and found that the HCL system modestly improved time in range, time spent in hypoglycemic range, glycemic variability, several other secondary glycemic outcomes, a standardized diabetes-specific measure of quality of life, and treatment satisfaction [62].
•School-aged children – In a 16-week randomized trial in 101 children 6 to 13 years, use of an HCL system increased time in target range (67 versus 55 percent; mean adjusted difference 11 percentage points, 95% CI 7-14) [63]; this study used the same HCL system as the adolescent study described above [61]. The percentage of time below the target range (<70 mg/dL) was low for both systems (1.6 and 1.8 percent, respectively), and there were no episodes of severe hypoglycemia or DKA. In an observational extension of this study, glycemic control was maintained through 28 weeks of use, with high satisfaction ratings for the technology [64].
In a three-month observational study in 105 children 7 to 13 years old, use of a different HCL insulin delivery system was associated with increased time in target range compared with baseline (65 versus 56.2 percent) [65]. There were no episodes of severe hypoglycemia or DKA.
In another three-month observational study of children 6 to 13 years old, use of a tubeless HCL automated insulin delivery system resulted in a decrease in A1C by 0.71 percent (mean A1C declined from 7.67 to 6.99 percent) and time in range improved by 15.6±11.5 percent or 3.7 hours/day in children, without any change in time spent in the hypoglycemic range [66].
•Younger children – A 16-week randomized crossover trial in 74 children two to seven years of age with well-controlled T1DM (baseline mean A1C 7.3 percent) compared a third HCL insulin delivery system (Cambridge closed-loop algorithm) with a sensor-augmented insulin pump [67]. The HCL insulin delivery system increased percent of time in glycemic target range (71.6 versus 62.9 percent; mean adjusted difference 8.7 percentage points, 95% CI 7.4-9.9). The greatest improvement was at night (midnight to 7:50 AM), when 82.2 percent was spent in target range and <3 percent below the target range (<70 mg/dL). The system also modestly decreased A1C by 0.4 percent and reduced time spent above the target range but did not alter the overall time spent in the hypoglycemic range. One serious hypoglycemic event occurred, attributable to user error.
A 13-week observational study of a tubeless HCL automated insulin delivery system that included 80 children two to five years old reported a decrease in A1C by 0.55 percent, an increase in time in range by 10.9 percent of 2.6 hours/day, a decrease in time spent in the hypoglycemic range by 0.27 percent, and no episodes of severe hypoglycemia or DKA [68].
The FDA has approved the use of the t:slim X2 pump with Control-IQ or Basal-IQ technology (Tandem Diabetes Care) [69] and Omnipod 5 (Insulet) [70] for children ≥6 years, as well as MiniMed 670G (Medtronic) for children ≥2 years [71,72]. Details about these devices and data from studies in adults are presented separately. (See "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Insulin only, partially automated system'.)
●Automated closed-loop (ACL) system – Studies are evaluating the safety and efficacy of a more automated closed-loop system of insulin delivery based on continuous glucose sensing, sometimes known as an "artificial pancreas." The device is similar to an HCL system, but the algorithm requires minimal inputs by the user (only body weight and qualitative descriptions of carbohydrate intake). In a 13-week unblinded randomized trial including 112 participants age <18 years, use of an ACL device led to greater reduction in A1C compared with standard care (mean adjusted difference -0.5 percent, 95% CI -0.7 to -0.2), with a 10 percent improvement in time in range [73,74]. The greatest improvements were in participants with high A1C values at baseline and those using MDI or insulin pumps without automation. Severe hypoglycemia events were uncommon in both groups, occurring in 2.7 percent of participants in the ACL group versus 1.9 percent in those randomized to standard care. Hyperglycemia was less common in participants using the ACL system (time with glucose >250 mg/dL 11.8 versus 20.4 percent), although brief episodes of hyperglycemia were common and most were related to infusion set failure. Such devices are not yet commercially available. (See "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Insulin-only, completely automated system (bionic pancreas)'.)
A bihormonal version (using insulin and glucagon) is also under development [75]. This type of device reduced episodes of hypoglycemia in children 6 to 11 years old in a diabetes camp setting [76]. Details about this type of system in adults and children are described separately. (See "Continuous subcutaneous insulin infusion (insulin pump)", section on 'Bihormonal, completely automated system'.)
INSULIN DOSING
Initial regimen — For a child with newly diagnosed T1DM, the initial dosing of insulin therapy is estimated as follows (algorithm 1):
Estimate the total insulin requirement — The first step is to estimate total insulin requirement. This is based upon the body weight, age, and pubertal stage of the child and whether or not the child presents in ketoacidosis or is on medications that may increase insulin resistance (such as glucocorticoids). In general, the newly diagnosed child requires an initial total daily insulin dose of 0.3 to 1 units/kg/day. Prepubertal children usually require relatively low doses (0.3 to 0.7 units/kg/day). Higher doses (eg, 0.7 to 1 units/kg/day) are needed in pubertal children, patients in ketoacidosis, or patients receiving glucocorticoid therapy. This estimated dose will be adjusted up or down as needed depending on the child's response to the initial regimen. After adjustment, the total insulin dose may be as low as 0.25 units/kg/day (or even lower) for some children during the "honeymoon period."
Calculate the basal insulin dose — The next step is to calculate the basal insulin dose. In children, the basal insulin requirement (eg, insulin glargine, detemir, or degludec or basal rate of pump) is approximately 50 percent of the total daily dose. The requirement may be lower (usually 40 percent) in children who are growing rapidly. It may be higher (up to 60 percent) in patients on a low-carbohydrate diet.
●For children using multiple daily injections (MDI) – Administer the basal dose as insulin glargine or detemir. This is usually given once daily and usually in the evening. However, some children do better with two divided doses (eg, those who tend to have nocturnal hypoglycemia, when detemir insulin is used, or when a single daily dose of insulin glargine does not have a 24-hour duration of action).
●For children using an insulin pump – Calculate the hourly rate = Total basal dose ÷ 24. This is given as continuous infusion of rapid-acting insulin (lispro, aspart, or glulisine).
Calculate prandial insulin — In addition to the basal insulin, a dose of rapid-acting insulin should be given before each meal or snack. The dose depends on the anticipated amount of carbohydrate that will be ingested and the preprandial blood glucose concentration.
●Carbohydrate coverage (also called the "carbohydrate ratio") is estimated based on the patient's total insulin dose, as follows:
Amount of carbohydrate (grams) covered by 1 unit insulin = 500 ÷ Total daily insulin dose (assuming that rapid-acting insulin is used)
Notes: This equation is a rough estimate, and the actual carbohydrate ratio also depends on other factors. The equation tends to underestimate the amount of prandial insulin needed for older children and tends to overestimate the dose for younger children. Some endocrinologists prefer to use a factor of 450 rather than 500 for younger children. Obesity and the associated insulin resistance may lead to greater insulin requirements. If short-acting (regular) insulin (rarely used in MDI or pump regimens), a factor of 450 should be used rather than 500.
Example:
•If a patient's total daily insulin dose is 50 units, then 1 unit of rapid-acting insulin is estimated to cover 10 grams of carbohydrate (500/50). This patient's "carbohydrate ratio" is 1:10.
•If this patient plans to ingest 40 grams of carbohydrates at the meal, the bolus insulin dose for carbohydrate coverage is 40/10 = 4 units.
●Dose correction for elevated blood glucose – For patients with blood glucose above a target range, the premeal dose of insulin is further adjusted based on the following calculation:
Estimated blood glucose reduction for 1 unit of insulin = 1500 ÷ Total daily insulin dose
Notes: The ratio of insulin:estimated blood glucose reduction is also known as the "correction factor" or "sensitivity." For very young children, use a factor of 1800 rather than 1500. The same equation can be used to calculate a correction dose for hyperglycemia between meals.
Insulin correction for blood glucose is based on this correction factor as well as a target blood glucose. The target blood glucose is usually between 100 to 120 mg/dL for most children, although a higher target is sometimes used for younger children, particularly for the evening insulin doses. Obesity and the associated insulin resistance may lead to greater insulin requirements. A smaller than calculated correction bolus may be given if the child is expected to be very active physically.
Example:
•If a patient's total daily insulin dose is 50 units, then 1 unit of rapid-acting insulin is estimated to bring down the blood glucose by 30 mg/dL (1500/50). This patient's correction factor (sensitivity) is 1:30.
•If this patient (with correction factor 1:30) has a preprandial blood glucose of 220 mg/dL and the target blood glucose is 100 mg/dL, the blood glucose correction bolus is (220 – 100)/30 = 4 units.
•The total preprandial bolus is the sum of the "correction dose" plus the carbohydrate coverage for 40 g of carbohydrate. For this patient, the total preprandial bolus is 4 + 4 = 8 units.
Subsequent modifications to insulin dosing
Adjustments to regimen — Throughout insulin therapy, the planned regimen should be adjusted periodically based on overall glycemic control (as indicated by hemoglobin A1c [A1C] and time in range) and episodes of hyperglycemia and hypoglycemia.
●Insulin dose adjustments are usually made in increments or decrements of 10 to 20 percent at a time, based on fingerstick blood glucose readings or the continuous glucose monitoring (CGM) output.
●As the child enters the honeymoon phase of diabetes, rapid reductions in basal and bolus insulin regimens are typically required.
●The total daily insulin dose and the ratio of the basal to total insulin dose should be reviewed at each visit to assure that they are within the expected ranges.
•The total daily insulin dose is typically between 0.3 and 1 unit/kg/day. Younger children are generally on the lower end of this range (and require even lower doses during the honeymoon phase).
•Basal insulin should be approximately 40 to 60 percent of the total daily dose; the lower target is appropriate for children who are growing rapidly, and the higher target is appropriate for those on a low-carbohydrate diet.
●The basal regimen may be adjusted with different rates at different times of the day. As an example, some children require a higher basal rate in the early morning hours because of an increase in counterregulatory hormones or a lower rate during and for a variable period after exercise (depending on the duration and nature of exercise).
●Some children may also require different carbohydrate ratios, different correction factors, and different target blood glucose concentrations at different times of the day, depending on activity levels, and also as a strategy to reduce the risk of hypoglycemia overnight.
For patients using an insulin pump, some devices can be programmed to prolong the prandial insulin ("extended") bolus to improve control of postprandial hyperglycemia. The bolus can be administered in a "dual-wave" or "square-wave" pattern [77,78]. For dual-wave administration, the mealtime insulin dose is divided into a bolus followed by a continuous infusion (higher than the basal rate) over two hours; this is useful for meals with varied glycemic content and especially high fat. For square-wave administration (also known as extended bolus), the dose of insulin is given as a continuous infusion over two hours; this is useful for patients who plan to snack over an extended period of time, as might occur in social settings.
Honeymoon phase — A few weeks after the diagnosis and initiation of insulin therapy, a period of decreasing exogenous insulin requirement occurs, commonly referred to as the "honeymoon" or remission phase of diabetes. During this period, the remaining functional beta cells secrete some endogenous insulin, resulting in reduced exogenous requirement. Close monitoring of blood glucose is mandatory since hypoglycemic episodes are likely if the insulin dose is not appropriately adjusted. The duration of this phase is variable and may last several months to several years. Rising blood glucose levels, A1C, and increasing exogenous insulin need indicate the end of this phase.
Adjustments with growth — Follow-up visits at least every three months are required to adjust for the increasing insulin requirement with continued growth of the child and increasing insulin deficiency with duration of diabetes. The family can be taught to make interim adjustments via telephone consultation or telemedicine visits. As a child enters puberty, daily insulin requirements may increase to more than 1 unit/kg because puberty increases insulin resistance [79].
Converting from MDI to insulin pump therapy — When converting a patient from multiple daily injections (MDI) or conventional therapy to insulin pump, the initial dose is dependent on diabetes control and total daily insulin dose (algorithm 2). If control has been excellent, the initial total daily insulin dose (daily insulin pump dose) is 10 to 20 percent less than the previous dose. If control has been poor, the previous total daily insulin dose should be used. One study suggests patients using detemir insulin may require greater dose reductions (26 to 33 percent) when switching from MDI to the insulin pump [80]. (See 'Insulin pump' above.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diabetes mellitus in children" and "Society guideline links: Hyperglycemic emergencies".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Type 1 diabetes (The Basics)" and "Patient education: My child has diabetes: How will we manage? (The Basics)" and "Patient education: Controlling blood sugar in children with diabetes (The Basics)" and "Patient education: Carb counting for children with diabetes (The Basics)" and "Patient education: Managing diabetes in school (The Basics)" and "Patient education: Giving your child insulin (The Basics)" and "Patient education: Checking your child's blood sugar level (The Basics)" and "Patient education: Should I switch to an insulin pump? (The Basics)")
●Beyond the Basics topic (see "Patient education: Type 1 diabetes: Overview (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Target hemoglobin A1c (A1C) – We suggest that insulin therapy be designed to achieve a target A1C of <7 percent for most children and adolescents with type 1 diabetes mellitus (T1DM), rather than a higher A1C threshold (Grade 2B). However, glycemic targets should be tailored to the individual patient. Less stringent goals may be appropriate for individual patients, such as those with increased risks for severe hypoglycemia. Other patients may safely use more stringent targets. (See 'Target for hemoglobin A1c' above.)
●Types of insulin regimens – For all children and adolescents with T1DM, we recommend an intensive insulin regimen (ie, consisting of a basal dose of insulin together with premeal boluses of a rapid-acting insulin) rather than a conventional regimen (consisting of fixed doses of intermediate-acting insulin with rapid- or short-acting insulin given two or three times a day) (Grade 1B). The approximate time of onset, peak activity, and duration of action of the most commonly used insulins are shown in the table (table 1). Intensive compared with conventional therapy improves glycemic control and decreases long-term complications of diabetes. (See 'Value of an intensive regimen' above.)
●Intensive insulin therapy – Intensive insulin therapy can be delivered by insulin pump (a device that delivers a continuous subcutaneous infusion of a rapid-acting insulin supplemented by boluses before each meal or snack) or multiple daily injections (MDI). A variety of insulin pump regimens are available: standalone pump therapy without continuous glucose monitoring (CGM), pump therapy combined with CGM (sensor-augmented pumps), and algorithm-driven pumps (ie, automated hybrid closed-loop [HCL] systems that automatically use sensor glucose values to determine rates of insulin delivery). All of these require extensive training, commitment, and work for the patient and family/caregiver. The choice depends upon the needs and preferences of the patient and family, cost considerations, and clinician preferences. Our general approach is as follows (see 'Types of intensive regimens' above):
•Continuous insulin delivery (insulin pump) – For most children, we suggest some form of continuous insulin delivery system rather than MDI (Grade 2C). Continuous insulin delivery systems (with or without CGM) appear to modestly improve glycemic control compared with MDI and may be associated with improved quality of life. In most cases, we suggest an HCL system, if available, rather than other systems (Grade 2C). In clinical trials, HCL systems modestly improved glycemic outcomes compared with sensor-augmented insulin pumps. In our clinical experience, most children and families have high satisfaction with newer HCL devices and can learn to use them effectively with appropriate training and support.
Special training is required for safe use of either type of device, including vigilance for infusion pump failure, which can lead to ketoacidosis. Cost considerations may be another limitation. (See 'Insulin pump' above and 'Insulin pumps with glucose sensors' above and 'Closed-loop insulin pumps' above.)
•MDI – This technique consists of injections of a long-acting insulin analog once or twice daily and short-acting insulin before each meal and snack. If rapid-acting insulin analogs are not locally available, short-acting insulins are sometimes used. While continuous insulin therapy (insulin pump) is generally preferred over MDI for long-term management of T1DM, MDI is often used in newly diagnosed patients as they acclimate to the diagnosis before converting to pump therapy. (See 'Multiple daily injections' above.)
●Blood glucose monitoring – Frequent blood glucose monitoring is required for optimal glycemic control. The frequency of monitoring depends upon the risk of hypoglycemic episodes, insulin regimen selected, and level of physical activity.
For most children, we suggest treatment guided by CGM rather than fingersticks for blood glucose monitoring, provided that such a device is available and the patient or caregivers are able to use the device safely (Grade 2C). CGM appears to modestly improve glycemic control in patients who consistently use the device and generally reduces treatment burden. (See 'Blood glucose monitoring' above.)
●Insulin dosing – Initial insulin dosing involves several calculations, each of which depend on the child's individual characteristics (algorithm 1):
•Total daily insulin requirement (based on the body weight, age, pubertal stage, and whether or not the child presents in diabetic ketoacidosis [DKA] or is on glucocorticoids) (See 'Estimate the total insulin requirement' above.)
•Basal insulin requirement (typically approximately 50 percent of the estimated total insulin requirement) (See 'Calculate the basal insulin dose' above.)
•Prandial (premeal/pre-snack) doses, which are calculated based on anticipated carbohydrate intake and the patient's individual insulin dose per gram of carbohydrate (expressed as their "carbohydrate ratio"), with dose adjustments for blood glucose elevations above the target range (based on the "correction factor" or "sensitivity") (See 'Calculate prandial insulin' above.)
These insulin regimens will require further adjustments depending on the child's response to the initial regimen, growth, level of exercise, and changes in insulin sensitivity over time. (See 'Subsequent modifications to insulin dosing' above.)
●Converting to an insulin pump – When converting from an MDI regimen to insulin pump therapy, the starting insulin dose depends on the patient's current diabetes control and total daily insulin dose. For patients with excellent control on an MDI regimen, the initial total daily insulin dose for pump therapy is 10 to 20 percent less than for MDI (algorithm 2). (See 'Converting from MDI to insulin pump therapy' above.)
