Breakfast blues: When rapid-acting insulin is not fast enough

Issue: May 2013

A 45-year-old male presents for follow-up for type 1 diabetes. He reports adherence to a prescribed medical regimen and reasonable adherence to prescribed physical activity and dietary regimens. However, he is concerned that he needs to take his prandial insulin about 40 to 45 minutes before breakfast because “it takes too long to start working.” This leads to problems with hypoglycemia when he waits too long or hyperglycemia to more than 300 mg/dL when his meal is too close to the time of injection. Moreover, the patient has noticed a disproportionate spike in blood glucose after breakfast compared with other meals, which is why he has reduced his carbohydrate intake to less than 15 g in the morning. He is then faced with the choice between correcting with a small amount of insulin (which frequently induces hypoglycemia) or “waiting it out” until lunch.

The patient performs self-monitoring blood glucose at least six times per day.

Ronald Tamler

Home blood sugar records are as follows:

Overall mean blood glucose is 155 mg/dL, with a standard deviation of 69 mg/dL.
Mean blood glucose before breakfast is 112 mg/dL, with a standard deviation of 43 mg/dL.
Mean blood glucose after breakfast is 191 mg/dL, with a standard deviation of 93 mg/dL.
Overall mean blood glucose is 155 mg/dL, with a standard deviation of 69 mg/dL.

The patient’s insulin regimen consists of:

Insulin glargine (Lantus, Sanofi-Aventis): 30 units subcutaneously at night (11 p.m.), and
Insulin lispro (Humalog, Eli Lilly): 6 units with breakfast, 12 units with lunch and 12 to 14 units with dinner.

He typically corrects 35 mg/dL for every unit of insulin. There are no diabetic complications. Other medical history includes costochondritis, hyperlipidemia and androgenic alopecia.

Non-insulin medications are:

Atorvastatin (Lipitor, Pfizer) 10 mg oral at bedtime,
ASA (Aspirin, Bayer) 81 mg oral daily, and
Finasteride (Propecia, Merck) 1 mg oral daily.

Physical exam is unremarkable; with a well-appearing 45-year-old man, blood pressure was 124 mm Hg/80 mm Hg, pulse 80, height 6'3", weight 84.4 kg (186 lb) and BMI of 23 kg/m2.

There is no lipodystrophy at the patient’s injection sites, his thighs. His HbA1c level is 7.1%.

Which of the following choices will best help the patient address his morning blood glucose variability?

A. Stop insulin lispro and start regular human insulin with meals.
B. The patient should not eat breakfast at all. Instead, he should have a solid brunch at noon.
C. Increase the dose of insulin lispro with breakfast to 9 units.
D. Ask the patient to change his injection site at breakfast, from his thighs to his abdomen or his upper arm.
E. Increase insulin glargine to 35 units at night.

Answer: D

This patient is experiencing a need for more rapid insulin action. Many of our patients with type 1 diabetes have this problem, especially after ingesting foods with a high glycemic index, such as sodas, white bread or juice. However, this patient has already cut out high-glycemic index foods for breakfast. While insulin pumps might offer the benefit of adjusting basal rate and bolus flow rate according to time of day, the patient was not amenable to such treatment, and his fasting blood glucose did not permit for increasing his glargine dose further (E). Regular human insulin takes even longer to peak than lispro and would exacerbate the patient’s problem (A). His hypoglycemic episodes precluded a simple dose increase of lispro in the morning (C). Finally, depriving a patient of breakfast would not only significantly impair his quality of life, but is also associated with overeating later in the day, weight gain and insulin resistance down the road (B). While one could experiment with changing the short-acting insulin analogue to a different agent or applying warmth to the site of injection, I recommended changing the injection site to the upper arm or the abdomen. At the follow-up visit, the patient no longer reported hypoglycemia, with an HbA1c of 6.9% and a reduced standard deviation of 64 mg/dL after breakfast (mean, 172 mg/dL). I asked him what had changed, and he only then remembered that he had started injecting into his arm in the mornings and no longer required the wait time before breakfast.

Ronald Tamler, MD, PhD, MBA, is clinical director of the Mount Sinai Diabetes Center in New York. He also is an Endocrine Today Editorial Board member. Disclosure: Tamler reports that he has received research support from Abbott.