Home / Conditions / Type 1 Diabetes / ADA/JDRF Type 1 Diabetes Sourcebook, Excerpt #17: Insulin, Part 2 of 3

ADA/JDRF Type 1 Diabetes Sourcebook, Excerpt #17: Insulin, Part 2 of 3

Dec 8, 2013

Anne Peters, MD, and Lori Laffel, MD, MPH, Editors
Jane Lee Chiang, MD, Managing Editor


PEDIATRICS – Jane Lee Chiang, MD, and Georgeanna J. Klingensmith, MD

ADULTS – Irl B. Hirsch, MD
1. Basal insulin accounts for ~50% of total daily dose:

i. LAA (insulin glargine, detemir, degludec) or NPH if LAA is not an option

2. Prandial insulin accounts for remaining ~50%:

i. RAA or short-acting regular insulin (if RAA is not an option, or high-fat foods are frequently consumed) before meals and snacks (when required)

ii. Calculated by I:C ratio or short-acting insulin (~20% prebreakfast, ~10% prelunch, or ~20% predinner)

It was hoped that RAAs would enable simultaneous food consumption with dosing; unfortunately, depending on the carbohydrates consumed, a large post-prandial glucose peak may still occur. Several studies have shown the lag time should be at least 10–20 min depending on the premeal blood glucose….

Starting Dose

The starting insulin dose is usually based on weight, with doses ranging from 0.4 to 1.0 units/kg/day of total insulin with higher amounts during puberty. Most clinicians start at 0.5 units/kg/day when the patient is metabolically stable, higher if immediately after presentation in ketoacidosis.

I:C Ratios

The insulin amount for a given amount of carbohydrates varies, but most adult patients require an I:C ratio of 1 unit/8 g (1:8) to 1:20. In adolescents, it is not uncommon for the I:C ratio to be 1:5, while in very young children (aged <6 years) the I:C may be as low as 1:50. A food diary over several days providing the insulin doses given, the food and grams of carbohydrate eaten, with glucose levels before and 2-3 h after meals will assist in determining the I:C ratio.

Various formulas have been developed as a starting place (e.g., 1,700 or 1,800/ TDD = how many mg/dl 1 unit of insulin will drop the blood glucose when the glucose control is excellent on this daily dose). When considering the insulin dose, one must know the glucose target. For example:

Target blood glucose = 120 mg/dl before meals

Target blood glucose = 150 mg/dl at bedtime

■ If prebreakfast glucose = 220 mg/dl and an ISF of 50 then 220 (current) – 120 (target) = 100/50 = 2 units of RAA at breakfast

■ If prebedtime glucose = 220 mg/dl and an ISF of 50 then 220 (current) – 150 (target) = 70/50 = 1.4 units of RAA at nighttime However, at bedtime, it would be 1.4 units (most insulin pens could only give 1.0) due to the higher target.

Figure 12.I.2 Approximate pharmacokinetic profiles of human insulin and insulin analogs.6


Insulin Stacking

Insulin stacking must be factored when considering prandial insulin dosing. From the pharmacodynamic times above, the RAAs last 4-6 h, meaning if a full correction dose is provided within this time frame, insulin stacking may lead to hypoglycemia. We minimize this problem by having patients subtract the amount of estimated IOB from their correction dose. Patients using MDI can see IOB from a graph provided to them.6


A typical MDI regimen provides the basal insulin component as LAA insulin either once or twice daily. The usual treatment plan begins with a once daily basal insulin and progresses to a BID regimen if adequate basal therapy is not achieved with daily LAA. The prandial component is usually RAA. The dose is determined by the amount needed to correct the high blood glucose plus the dose required for meals, which is determined by the I:C ratio. The doses may be individually calculated, or with a bolus dose calculator. Sometimes, a constant carbohydrate diet is recommended (i.e., fixed insulin dose and fixed amount of carbohydrate for each meal).

If using NPH, it is recommended for twice daily dosing, prebreakfast and predinner or bedtime. Since NPH peaks at 4-10 h, giving NPH in the evening increases nocturnal hypoglycemic risk compared to using LAA. Likewise the variable peak timing may cause unexpected hypoglycemia during the day. Because of the increased hypoglycemic risks with NPH, a lower percent of the total daily dose (TDD) is given as the basal portion when NPH (vs. LAA) is used. (See NPH description above.)

NPH given in 3-4 lower doses daily may decrease the unpredictable hypoglycemic risk and establish a more consistent basal profile. If NPH is used as basal insulin, modern dosing recommends the following:

■ Morning: NPH (20-30% of TDD) and prandial regular or RAA insulin

■ Lunch: RAA. This is based on the morning NPH dose, peak NPH time, and frequent BG monitoring.

■ Evening: NPH (15-20% of TDD) predinner or prebedtime (more common) and bolus regular or RAA insulin

■ If an afternoon snack is taken, a prandial dose is usually required for optimal glycemic control.

Some NPH insulin regimens recommend prandial dosing with regular insulin and NPH as a single dose in the evening before bed. This plan is more effective if regular insulin is used for the prandial insulin because of its longer duration versus RAA. While more economical, this regimen is less reliable due to the unpredictable actions of both regular and NPH insulin. It is not recommended for individuals requiring more intensive diabetes management or for those at greater risk from hypoglycemia including children and older adults.


Disposable pens and cartridges currently aid patients in measuring and administering insulin and facilitate compliance. There are many options with pens and pen needles (partial listing: disposable or durable, half or whole units, and with or without memory). Mixing insulins is currently rarely done since the basal-bolus regimens offer more flexibility and allow tailoring the dosing to the current glucose value. LAA insulins should not be mixed with other insulins, although there are reports in the literature indicating this is safe and does not alter glucose values compared to injecting separately. NPH may be mixed with rapid or short-acting insulins. Fixed mixtures are also available commercially; however, none allow daily variation in dosing for variable glucose levels.

In considering needles, there are two factors to consider. The first factor is the thickness or gauge (higher gauge = thinner needle). The thickest gauge is 28 and 32 is the thinnest. Most people prefer thinner gauges, which do not clinically alter glucose control. For large doses a thicker gauge is preferable due to the ease in administering the insulin (pen or syringe).

The second consideration is the needle length. We initially thought that obese patients require longer needles, but that has not been shown to be true, at least to a BMI of 40 kg/m.2,7 Needle lengths can vary from 4 mm to 12.7 mm, with most patients preferring shorter needles.8, 9 Syringe and pen needle manufacturers suggest needles be used only once, since sterility cannot be guaranteed with multiple usage, which also risks needle bending. There are data that suggest dosage may be affected if the pen needle is left in place. The needle also needs to be recapped after each use.

Pens offer an advantage over syringes with 1) more consistent dosing when multiple caregivers provide insulin dosing, 2) larger numbers than on syringes and an audible click as the dose is dialed in for visually impaired individuals, and 3) frequently greater comfort since the pen needle does not go through the stopper on the insulin vial, making the needle sharper for injection. In children, school personnel may require insulin administration with a pen to minimize dosing errors. Parents also prefer the consistency of the pen if the child is away from both home and parental supervision.

There are pens that deliver 0.5-unit doses, one beginning with 0.5 unit and another beginning with 1.5 units. Both of these options use insulin cartridges and a nondisposable pen unit. Otherwise, individuals requiring 0.5-unit dosing will need to use syringes, which have 0.5-unit markings that allow administration of whole or 0.5-unit doses. Basal insulin is available in vials for use with syringes, or in disposable pens that administer insulin in 1-unit increments.

An insulin pump or CSII is also an effective way to deliver insulin and will be discussed in later chapters.

Remission or Honeymoon

Within 3 days to 3 weeks after diagnosis and initiation of insulin therapy, the patient generally enters a remission phase. This is thought to be due to β-cell recovery from glucotoxicity. During this phase, stable glucose levels allow low insulin requirements. Insulin doses for meals may decrease to very low levels. Active individuals may require very little to no insulin. But it is unclear if a small amount of basal insulin during this time is important for maintaining β-cell function and for reinforcing the need for continued glucose surveillance. The predominant opinion is that maintenance of some insulin therapy is important.

Newborns and Infants

If feasible, children aged <1 year to 18 months of age should begin CSII therapy as soon as possible after diagnosis. Infants between 6 to 9 months of age will benefit from CSII initiation while hospitalized for the initial diagnosis. There is little data comparing trials of injection therapy vs. pump therapy in these very young children. However, if anecdotal reports in infants and data supporting CSII in young children (aged <6 years) apply, then one could extrapolate that CSII use should be supported in very young children (<18 months of age). Children diagnosed with T1D prior to 6 months of age (neonatal diabetes) should be evaluated for monogenic causes of diabetes, which may be more effectively and safely treated with sulfonylurea drugs.

School-Age Children

Please refer to chapters 5 and 18 for comprehensive information about school-age children.


There is extreme heterogeneity within the older adult population: ranging from highly functional and independent to requiring significant assistance in managing their diabetes (see chapter 18). Hypoglycemia is a significant concern, particularly if there is concomitant cerebrovascular disease or coronary artery disease. Dosing may be more complex when comorbidities become more common and physical activity often becomes more sporadic. An important concern is related to mental status declines, often making many diabetes self-management decisions impractical. This is particularly relevant for patients using CSII. Basic diabetes management mistakes, including forgetting to test blood glucose or bolus insulin dosing, become more common. Worse is the forgetful patient who neglects to check if a bolus was given and inadvertently administers a second bolus. Like other aspects of aging, diabetes self-management may become more difficult without assistance from a care provider. Losing independence, after decades of self-care, is a difficult situation. This has to be first addressed by removing the CSII, particularly for those patients requiring frequent hospitalizations or a skilled nursing facility.

Insulin Resistance

While T1D is due to insulin deficiency secondary to β-cell destruction, the so-called global obesity epidemic has impacted those with T1D. If an individual has T1D, plus obesity, and a strong family history of type 2 diabetes, then insulin resistance may occur. Additional treatment with insulin sensitizers such as metformin may be useful. There is also increasing evidence that insulin resistance commonly occurs in T1D even in the absence of overweight or obesity.

Insulin Allergies

Allergic reactions to insulin are rare, with reactions being transient or due to artifacts. Patients, especially those with atopy, may have a local or systemic allergy to insulin, protamine, or the low pH of glargine insulin. Inadvertent intradermal injections or localized allergic reactions may cause burning, erythema, and urticaria at the injection site. If the symptoms persist despite proper injection techniques, then patients should switch brands or use antihistamines, or both. Patients should be treating accordingly, if true anaphylaxis or severe asthma occurs, and be considered for insulin desensitization.5


Pulmonary insulin delivery was first reported in 1925. Approximately 10–30% of inhaled insulin is absorbed into the circulation and appeared to be well tolerated. Studies have demonstrated that inhaled insulin is more rapidly absorbed than subcutaneous regular insulin and as quickly as rapid-acting insulin (lispro or aspart). Its duration is longer than rapid-acting insulin and similar to regular insulin. The bioavailability of inhaled insulin relative to subcutaneous regular insulin is ~10%. The first pulmonary insulin (Exubera®) was available from 2006–2007 but was withdrawn by the manufacturer in late 2007 for non-medical reasons. Technosphere (Afrezza®) is a monomeric, dry powder insulin with an onset of action of 10-15 min and a duration of 3-4 h.10,11 It is currently undergoing phase III studies in both T1D (NCT01445951) and type 2 diabetes (NCT01451398) patients to demonstrate efficacy and safety of inhaled insulin, with results expected in the next 1–2 years.12

Peritoneal insulin, with an implantable insulin pump, has been used for the last 20 years in research settings. Intraperitoneal insulin is rapidly and predictably absorbed into the portal circulation, simulating physiologic insulin delivery and absorption. Clinical studies have demonstrated that implantable pumps are safe and effective for attaining glycemic control and reducing severe hypoglycemia. This route also avoids peripheral hyperinsulinemia and the theoretical risk of atherosclerosis.5


On January 11, 1922, Leonard Thompson received an injection of “thick brown muck,” also known as insulin. Since then, extraordinary progress in insulin development and delivery has enabled T1D patients to have healthy, productive lives. Despite the advancements in the insulin field, T1D remains a burden to many, and gaps in our insulin formulations and understanding of insulin therapy remain. The following are areas where more clinical research is needed:

1. Better, flatter, more predictable basal insulins

2. Faster prandial insulins (both faster in and faster out), which will also work with CSII

3. Better, more comfortable insulin needles and CSII infusion devices

4. Insulin pens that electronically communicate with another device such as a meter or smartphone with easy downloading

5. A glucose responsive insulin, i.e., an insulin injected infrequently (e.g., once daily) that is attached to a protein and is released when glucose levels rise

6. Insulin coformulations, with the most obvious candidates: pramlintide (known to be effective in T1D) and GLP-1 analogues (definitive T1D data still pending) While some of these ideas for insulin therapy may seem far off in the future, it must be remembered how quickly we have seen vast improvements in our current insulin formulations and treatment strategies. While the ultimate goal is to eliminate insulin therapy for all patients affected by T1D, improvements and innovations to our current treatments must continue.

  1. Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med329:977–986, 1993.
  2. Hirsch IB, Skyler JS: Management of type 1 diabetes. In Atlas of Diabetes. Skyler JS, Ed. New York, Springer Science + Business Media, LLC, 2012, p. 95–113.
  3. Garber AJ, King AB, Del Prato S, Sreenan S, Balci MK, Muñoz-Torres M, Rosenstock J, Endahl LA, Francisco AM, Hollander P: NN1250-3582 (BEGIN BB T2D) Trial Investigators: Insulin degludec, an ultra-longacting basal insulin, versus insulin glargine in basal-bolus treatment with meal-time insulin aspart in type 2 diabetes (BEGIN Basal-Bolus Type 2): a phase 3, randomised, open-label, treat-to-target non-inferiority trial.Lancet 379:1498–1507, 2012.
  4. Heller S, Buse J, Fisher M, Garg S, Marre M, Merker L, Renard E, Russell- Jones D, Philotheou A, Francisco AM, Pei H, Bode B: BEGIN Basal-Bolus Type 1 Trial Investigators: Insulin degludec, an ultra-longacting basal insulin, versus insulin glargine in basal-bolus treatment with mealtime insulin aspart in type 1 diabetes (BEGIN Basal-Bolus Type 1): a phase 3, randomized, open-label, treat-to-target non-inferiority trial. Lancet 379:1489–1497, 2012
  5. American Diabetes Association: Medical Management of Type 1 Diabetes. 5th ed. Kaufman, FR, Ed. Alexandria, VA, American Diabetes Association, 2008: 51–82; table on 58
  6. Hirsch, IB: Insulin Analogues. N Engl J Med 352:174–183; table on 177, 2005
  7. Hirsch LJ, Gibney MA, Albanese J, Qu S, Kassler-Taub K, Klaff LJ, Bai-ley TS: Comparative glycemic control, safety and patient ratings for a new 4mm×32G insulin pen needle in adults with diabetes. Curr Med Res Opin 26:1531–1541, 2010
  8. McKay M, Compion G, Lytzen L: A comparison of insulin injection needles on patients’ perceptions of pain, handling, and acceptability: a randomized, open-label, crossover study in subjects with diabetes. Diabetes Technol Ther 11:195–201, 2009
  9. Siegmund T, Blankenfield H, Schumm-Drager P: Comparison of usability and patient preference for insulin pen needles produced with different pro-duction techniques. Diabetes Technol Ther 11:523–528, 2009
  10. Marino MT, Costello D, Baughman R, et al. Pharmacokinetics and phar-macodynamics of inhaled GLP-1 (MKC253): proof of concept studies in healthy normal volunteers and in patients with type 2 diabetes. Clinical Phar-macology & Therapeutics 88:243–250, 2010
  11. Marino MT, Cassidy JR, Baughman RA, et al.: A new C-peptide correction model used to assess bioavailability of regular human insulin. Biopharmaceu- ticals & Drug Disposition 31:428–435, 2010
  12. www.clinicaltrials.gov. Accessed 4 October 2012
  13. Miles HL, Acerini CL: Insulin analog preparations and their use in children and adolescents with type 1 diabetes mellitus. Paediatr Drugs 10:163–176, 2008

Used with permission by the American Diabetes Association. Copyright © 2013 American Diabetes Association.

Please note: We are proud to have Dr. Anne Peters as a member of our Advisory Board member for Diabetes In Control, Inc.

T1-diabetes-sourcebookIf you would like to purchase the full text of The Type 1 Diabetes Sourcebook, Anne Peters, MD, and Lori Laffel, MD, MPH, editors, and Jane Lee Chiang, MD, managing editor, just follow this link.