Diabetes

Diabetes mellitus refers to a group of metabolic disorders that affect how the body uses blood sugar (glucose). Previously, type 1 diabetes mellitus (T1DM) was referred to as juvenile-onset diabetes and type 2 diabetes (T2DM) as adult-onset diabetes. However, with the increasing prevalence of T2DM in the pediatric population and because T1DM can manifest in adulthood, such terminology is no longer used.

• T1DM is an autoimmune disease in which the insulin-producing beta cells of the pancreas are destroyed, resulting in an insulin-deficient state. T1DM treatment is with insulin analogues via either multiple daily injections or continuous subcutaneous insulin infusion.

• T2DM is a state of relative insulin deficiency due to failure of pancreatic beta-cell insulin production to keep up with increasing insulin needs, generally due to insulin resistance. T2DM can progress to a state in which the pancreas makes little-to-no insulin.

The diagnostic criteria for diabetes and prediabetes are described in the following table:

(Source: Diagnosis of Diabetes. N Engl J Med 2013.)

Type 1 Diabetes

Treatment

The American Diabetes Association’s (ADA) recommended hemoglobin A1c (HbA1c) targets have shifted for the pediatric population over recent years to <7% for many children and <7.5% or higher for certain children who have limited access to medical care, inability to express or recognize hypoglycemia, or other severe health conditions. The rationale for these targets includes evidence of the detrimental effect of hyperglycemia on long-term outcomes, including neurocognitive health, along with evidence that hypoglycemia risk is reduced when children have access to newer insulins and newer diabetes technologies. The following table summarizes the 2024 ADA-recommended goals for treatment of pediatric T1DM.

Insulin Basics

Basal and bolus injection regimens:

• Basal insulin: Injection of long-acting or ultra-long-acting insulin (glargine or degludec once daily) or via continuous subcutaneous infusion of rapid-acting insulin via insulin pump to maintain normoglycemia and cover a patient’s baseline insulin requirement.

• Bolus insulin: Rapid-acting insulin given with meals and/or to correct a high blood sugar. The total dose of rapid-acting insulin is dependent on the carbohydrate the patient is going to eat and correction of their current blood sugar. This is calculated with a carbohydrate ratio, also known as the insulin-to-carbohydrate ratio (IC ratio), and the correction factor (CF), also known as the insulin sensitivity factor (ISF).

  • insulin-to-carbohydrate (IC) ratio: 1 unit of insulin to be given for X many grams consumed.

    • Example: IC ratio of 1:10 means 1 unit of insulin is needed to cover every 10 grams of carbohydrates consumed.

  • insulin sensitivity factor (ISF): 1 unit of insulin will lower blood glucose X mg/dL. This ISF is given to lower the blood glucose to a certain target.

    • Example: ISF of 1:30 means 1 unit of insulin is expected to bring down a patient’s blood glucose by 30 mg/dL. Patients calculate insulin needed for correction by taking (blood sugar - target)/ISF.

  • bolus calculation

    • Example: for a meal of 50 g carbs, blood sugar 250 mg/dL, target blood sugar 100 mg/dL, carb ratio 1:10, and correction factor 1:30

    • Rapid-acting insulin needed for carbs is 50/10 = 5 (planned carbs/IC).

    • Rapid-acting insulin needed for correction is (250 - 100)/30 = 5 [(glucose - target)/ISF].

    • Total rapid-acting insulin needed = 5 + 5 = 10 (amount for carbs + amount for correction).

Types: Two general types of insulin are in clinical use today:

• Biosynthetic human insulin has the same amino acid sequence as endogenous human insulin (e.g., regular insulin, neutral protamine Hagedorn [NPH] insulin). It has a substantially lower cost than insulin analogues.

• Insulin analogues have a slightly different amino acid sequence than endogenous human insulin due to modifications designed to affect speed of absorption and other pharmacokinetic properties. Examples include lispro, aspart, glargine, and degludec. Insulin analogues comprise most of the insulin prescribed in high-income countries and are incredibly expensive in countries without price regulation.

Dosing: Insulin is dosed in units (not mcg, mg, or mcg, like most drugs). At the time of insulin’s discovery, “1 unit” was defined as the amount required to lower the fasting glucose of a rabbit by 2.5 mmol/L (45 mg/dL).

• The usual concentration of most insulins is U-100 (meaning 100 units in one cc/mL); insulin syringes are designed with this assumption.

• Different concentrations may be used in certain circumstances:

  • In infants, U-10, or 10 units of insulin/mL, is often necessary to achieve precise dosing to the 10th-of-a-unit.

  • In adults with severe insulin resistance, more-concentrated formulations (U-200, U-300, and U-500) are now being used to provide higher doses of insulin in a smaller injection volume.

  • When an insulin syringe is used for formulations that are not U-100, the syringe unit markings do not correspond with those formulations, and extreme care is required to avoid overdosing. In most cases, use of insulin pens is safer and easier.

Onset and peak action: The onset of action and peak action of insulins are primarily determined by how quickly they are absorbed into circulation from the subcutaneous tissue. Human insulin is physiologically active as a monomer but forms dimers, tetramers, and hexamers. Hexamer formation and chemical stability are promoted by the presence of cations such as zinc. At pharmaceutical concentrations, human insulin is predominantly comprised of hexamers (six monomers positioned around a zinc ion). Hexamers are absorbed from the subcutaneous space much more slowly than dimers, and dimers are absorbed more slowly than monomers. When insulin is administered intravenously, the hexamers rapidly dissociate into monomers for near-immediate insulin action.

• Rapid-acting insulin analogues (lispro, aspart, and glulisine) have minor structural modifications that make them less prone to hexamer formation and thus more rapidly absorbed than human insulin.

  • Newer faster-acting preparations have recently become available (fast-acting insulin aspart and ultrarapid insulin lispro, insulin lispro-aabc). These formulations have the same amino acid structure as aspart and lispro, respectively, with other chemical additives that promote faster dissociation into monomers, faster absorption into circulation, or both.

• Long-acting insulin analogues (glargine, degludec, and glargine U-300) have minor structural modifications that make them prone to aggregate within the subcutaneous tissue, slowing their absorption into circulation.

  • Degludec forms multihexamer chains in the subcutaneous space that slowly dissociate from each end, slowing absorption.

  • Glargine U-300 takes advantage of the fact that more-concentrated insulin (300 units per cc/mL) has a much higher percentage of hexamers.

  • Long-acting insulin analogues with minimal peak and longer half-life (e.g., degludec and glargine U-300) have been associated with lower risk for hypoglycemia and may lower the risk of diabetic ketoacidosis.

The following table summarizes the duration of action of injectable insulins:

The following figure summarizes the pharmacokinetics of injectable insulins:

(Source: Insulin Analogues. N Engl J Med 2005.)

Insulin delivery and regimens: Current conventional treatment of T1DM consists of insulin delivery via either:

• multiple daily injection (MDI): given via basal-bolus regimen

• continuous subcutaneous insulin infusion (also known as insulin-pump therapy)

Initial insulin dosing for new-onset T1DM: Total daily dose (TDD) of insulin typically ranges from 0.3-1.0 unit/kg/day. Patients with obesity or in puberty require higher total daily doses of insulin. Total daily dose of insulin is typically divided approximately evenly between basal and bolus (thus about half of the TDD is given with long-acting insulin and half with rapid-acting insulin).

• To determine the ISF, divide 1800 by the TDD of insulin.

• To determine the IC ratio, divide 500 by the TDD of insulin.

• Example: If a patient’s TDD of insulin is determined to be 50 units, 25 units is given as basal. IC is calculated as 500/50 = 10 (IC of 1:10), and the ICF is 1800/50 = 36 (1:36).

Continuous subcutaneous insulin infusion (insulin-pump therapy): Insulin pumps deliver rapid-acting insulin via a small catheter placed under the skin. This site and the catheter must be changed every 2 to 3 days and placed in a new site. Not replacing the infusion set at least every 3 days is associated with a decrease in glycemic control and an increased risk of infection at the site.

• Basal insulin is delivered in the form of rapid-acting insulin infused continuously at a rate that can vary throughout the day. The IC and ISF are programmed into the pump. The user enters their blood glucose and grams of carbohydrate, and the pump calculates the insulin dose. The insulin bolus can be given over an extended period.

Continuous glucose monitoring (CGM): CGM operates via a small sensor inserted into the skin that gives patients real-time blood-glucose readings (every 5-15 minutes, depending on the brand) by measuring interstitial glucose levels. Most CGMs allow users to set alerts for hypo- or hyperglycemia or rapidly changed blood-glucose values.

Other diabetes technology: Artificial pancreas (or bionic pancreas) treatment (also known as a closed-loop system of glucose control) refers to autonomous insulin delivery via an insulin pump using blood-glucose data from CGM without user input. This may be achieved through one integrated system, or through a CGM and separate insulin pump that “communicate” with each other wirelessly. Current models that are commercially available still require user input but are able to suspend insulin delivery if blood glucose is dropping low and/or increase insulin delivery if blood glucose is rising.

Insulin Treatment Complications

(Source: Insulin-Pump Therapy for Type 1 Diabetes Mellitus. N Engl J Med 2012.)

(Source: Insulin-Mediated Lipohypertrophy. N Engl J Med 2017.)

Type 2 Diabetes

Screening and Diagnosis

Treatment

Diabetic Ketoacidosis (DKA)

Diabetic ketoacidosis (DKA) is a complication of diabetes mellitus resulting from relative insulin deficiency. It is primarily associated with T1DM but can manifest in patients with T2DM. DKA can be life-threatening and requires careful treatment of fluid and electrolyte derangements. In children, cerebral edema can be a complication of DKA, its treatment, or both and is the leading cause of DKA-associated mortality in children. Hyperosmolar hyperglycemic syndrome (HHS; formally referred to as hyperosmolar hyperglycemic nonketotic syndrome) is less common in the pediatric population, and its management is not covered in this rotation guide.

Signs and Symptoms

• nausea

• vomiting

• abdominal pain

• weight loss

• lethargy

• altered mental status

• polyuria

• polydipsia

• polyphagia

• Kussmaul respirations

• fruit-scented breath

Diagnosis

Diagnosis of DKA requires a blood glucose level >200 mg/dL with both acidemia (defined by venous pH <7.3 or a serum bicarbonate level <18 mmol/L) and ketosis (ketonemia or ketonuria, defined as a beta-hydroxybutyrate level ≥3 mmol/L or moderate or large urine ketones). The severity of DKA is determined by the degree of acidosis:

Hyperglycemic hyperosmolar state (HHS) is characterized by severe hyperglycemia but no ketoacidosis. Diagnosis requires all of the following:

• glucose >600 mg/dL

• venous pH >7.25 or arterial pH >7.3

• serum bicarbonate >15 mmol/L

• small ketonuria, absent to mild ketonemia

• effective serum osmolality >320 mOsm/kg

A “mixed” DKA/HHS can also occur in children with T1DM or T2DM, in which pH is <7.3, glucose is >600mg/dL, and osmolarity is >320 mOsm/kg.

Management

Initial Management: Fluid resuscitation should be initiated with a 10-20 mL/kg bolus of normal saline (NS) given over 1 hour. Patients with DKA are typically severely (>10%) dehydrated due to osmotic diuresis. The goal of management is to replace the fluids gradually over 36 to 48 hours.

Insulin: An insulin drip should be initiated at 0.1 units/kg/hour. (Some centers use a lower dose of 0.05 units/kg/hour.) The insulin drip should be started about 1 hour after the initial fluid. Insulin administration is necessary to reverse ketosis and resolve the acidosis.

Intravenous (IV) fluids: Hyperglycemia resolves more quickly than acidosis; therefore, dextrose-containing fluids need to be administered to maintain blood glucose while patients continue treatment with insulin drip until the acidosis resolves. A two-bag system is recommended. One bag contains normal saline and the other D10 normal saline (bags could also both contain half NS if hypotonic saline is required). Both bags contain an equal amount of potassium and phosphorus (usually as potassium acetate and potassium phosphate). The goal is to decrease blood glucose by 50-100 mg/dL/hour. As the glucose level drops, the rate of infusion from the D10 bag is increased while the rate of infusion from the dextrose-free bag is decreased.

• The hourly IV fluid rate is calculated as maintenance fluids + estimated fluid deficit with the goal of replacing fluids over 36-48 hours.

• A fluid rate of 1.5-2.0 times maintenance approximates this calculation.

Management of glucose, electrolytes, and acidemia:

• Glucose: Monitored via point-of-care (POC) glucose testing every hour. Venous blood gas, basic metabolic panel, magnesium, and phosphorus are monitored every 2-4 hours.

• Sodium: Patients typically present with hyponatremia. This is usually “pseudohyponatremia” due to hyperosmolarity from hyperglycemia. Failure of serum sodium to rise during treatment of DKA increases the risk for cerebral edema (this may be a signal of fluid overload). A corrected sodium level should be calculated using the following formula: corrected [Na+] = serum [Na+] + (1.6 x [(glucose - 100)/100]).

• Potassium: Potassium is often elevated at presentation due to acidosis, which results in the exchange of hydrogen ions moving intracellularly and potassium moving extracellularly. Total body potassium is depleted due to urinary loss. Hypokalemia will develop as acidosis resolves and as insulin is administered (insulin drives potassium intracellularly). Potassium should be added back to IV fluids preemptively. Potassium should be added back when urinary output is confirmed and if there is no significant hyperkalemia.

• Phosphorus: Hypophosphatemia develops due to osmotic diuresis, which results in urinary phosphate loss. Severe hypophosphatemia can result in muscle weakness. Phosphorus should be provided in IV fluids. Potassium and phosphorus replacement may be provided as equal parts potassium acetate and potassium phosphate as described in the two-bag system above. Other forms of potassium (e.g., potassium chloride) may also be used, but other forms often lead to the development of hyperchloremia.

• Chloride: An elevated chloride level can lead to hyperchloremic metabolic acidosis. If serum chloride is elevated, consider decreasing the tonicity of administered fluids from NS to half NS.

• Bicarbonate: Bicarbonate is low at the time of presentation due to metabolic acidosis and will rise with correction of acidosis via insulin therapy. Bicarbonate replacement is not recommended in the treatment of acidosis (unless it is required for treatment of hemodynamic instability) because it can increase the risk for cerebral edema.

Transitioning from insulin drip to subcutaneous insulin: Transition off the insulin drip and onto a subcutaneous regimen is appropriate once CO2 has improved to >15 mmol/L (anion gap closed) and the patient can tolerate oral intake.

• Administer a bolus of subcutaneous rapid-acting insulin (dose based on insulin calculations as described above).

• Allow the patient to start eating.

• Wait 20 minutes and stop the insulin infusion.

Cerebral Edema

Cerebral edema is the leading cause of mortality in patients with DKA. Cerebral edema typically occurs within the first 4 to 12 hours after initiation of treatment but can occur at 24-48 hours. The exact mechanism of cerebral edema in DKA is not known, but proposed mechanisms include osmotic, cytotoxic, and vasogenic processes.

Risk Factors

• young age

• new diagnosis of diabetes

• severe acidosis (pH <7.0)

• failure of a normal predicted serum sodium to increase during DKA treatment

• elevated BUN (blood urea nitrogen)

• administration of bicarbonate

• insulin administration during the first hour of DKA fluid treatment

Diagnosis

Suspect the presence of cerebral edema if the patient exhibits one of the following diagnostic criteria or two major criteria or one major criterion plus two minor criteria. Emergent head CT should be done as soon as the diagnosis is suspected, as long as the patient is stable enough to be sent to the scanner. Treatment of cerebral edema should not be delayed for imaging.

Treatment

The following treatment should be initiated as soon as a diagnosis of cerebral edema is suspected:

• elevate head of bed

• reduce fluid rate

• administer IV mannitol (0.25-1.00 g/kg over 20 minutes) or 3% hypertonic saline (5-10 mL/kg over 30 minutes)

• obtain cranial CT