Stuart A. Weinzimer, MD

 During my training, most pediatric endocrinologists were using the "70/30 rule" or the "rule of fifths" to determine the insulin dosage for patients with new-onset diabetes. Now that I am in practice, I have heard that carbohydrate counting is a more accurate method of determining a child's insulin needs.

Is carbohydrate counting, in fact, a better approach to determining the insulin dosage for a patient with newly diagnosed diabetes? If so, how is this done for both regular and long-acting insulin? Do age and weight play a role?

A: There is no one "right"method for determining the appropriate initial insulin dosage for children with new-onset type 1 diabetes (T1D). Choosing an insulin regimen often involves a trade-off between accuracy and simplicity.

Until recently, most pediatric patients with T1D were started on a "2-shot split-mix" regimen, which combined short- and long-acting insulins in a single injection given twice daily at a total daily dose of approximately 1 unit/kg. Typically, two thirds of the total daily dose was given at breakfast and one third before dinner. Two thirds of the morning dose was given as long-acting insulin (such as NPH or Lente) and one third as short-acting (such as Regular or one of the rapid-acting analogs). Half of the evening dose was given as long-acting and half as short-acting insulin. For example, for a child who weighs 36 kg, the initial insulin doses would be 8 lispro/aspart plus 16 NPH at breakfast and 6 lispro/aspart plus 6 NPH at dinner.

Many clinicians are now moving away from the 2-shot split-mix regimens because they are usually insufficient to meet intensive glycemic goals without unacceptable swings in blood sugar levels. Three-shot regimens move the dinner NPH dose to bedtime to provide better overnight coverage. Extra "touch-up" doses of short-acting insulin are often needed at lunch or in the midafternoon. The split-mix regimen does require consistency and regularity in meal portions and timing.

With the development of once-daily long-acting insulin analogs (glargine and detemir), many clinicians have completely abandoned the split-mix regimen in favor of the basal-bolus method. This more physiologic insulin replacement strategy uses a single dose of a long-acting "basal" insulin to provide background insulin coverage, plus "bolus" doses of rapid-acting insulin analogs for every meal and snack. This method does require that the patient and/or caregiver have some knowledge of carbohydrate counting:

•Infants and toddlers may require 1 unit of rapid-acting insulin for every 20 to 50 g of carbohydrate.

•Preadolescents may require 1 unit per 10 to 20 g.

•Adolescents may require 1 unit per 5 to 10 g.

Alternatively, the insulin-to-carbohydrate ratio (ICR) may be determined by the 

 To determine the ICR, divide 450 by the child's total daily dose of insulin (TDD). For example, for a child with a TDD of 36 units, the ICR would be 450/36 = 12.5, or 1 unit per 12 g of carbohydrate. The basal-bolus strategy allows more freedom in the amounts and timing of meals.

As always, the "best" method for determining insulin dosing should take into account a realistic assessment of the family's and child's abilities and life situation. Many older children and adolescents will tolerate 4 or 5 injections per day for flexibility in meal content and timing. Younger children typically prefer to eat consistently and have fewer shots.

Generally, we still use a 2-shot regimen during the first several months after diagnosis, while patients still have some residual insulin secretion (the "honeymoon period"). After the honeymoon--when blood sugar levels cannot be controlled adequately with 3 shots daily--we make the transition to basal-bolus therapy. This involves either multiple daily injection therapy using glargine or detemir plus aspart or lispro, or continuous subcutaneous insulin therapy (the pump).

Articles

During my training, most pediatric endocrinologists were using the "70/30 rule" or the "rule of fifths" to determine the insulin dosage for patients with new-onset diabetes. Now that I am in practice, I have heard that carbohydrate counting is a more accurate method of determining a child's insulin needs.

Diabetes Q&A: 70/30; "Rule of Fifths"; Split-Mix; Basal-Bolus--  Which Is Best?

Q: What should primary care pediatricians do--and what should we not do--when we encounter a patient with newly diagnosed diabetes?

A:The appropriate management of the patient with new-onset diabetes in the pediatrician's office depends primarily on the severity of the patient's illness, proximity to specialized pediatric care, and the practice patterns of the local pediatric diabetes service.

In most cases, the role of the pediatrician is to establish the diagnosis by eliciting symptoms and signs consistent with diabetes and to order the appropriate diagnostic tests. In any child with a typical history of increased thirst and polyuria, blood should be drawn to measure glucose levels. In addition, the urine should be tested for glucose and ketones. This testing should be done on the day of the office visit, when glucose levels are more likely to be elevated. It is inappropriate for the child to return the following morning for a fasting blood glucose test because the return of the blood glucose level to normal early on in the disease process only delays diagnosis.

If the child appears to be well or is only modestly dehydrated, non-glucose-containing fluids may be given orally if the child is referred promptly to the pediatric diabetes team. If the child is vomiting or severely dehydrated or obtunded, he or she should be taken to the nearest emergency department by emergency medical services.

The key is prompt referral to the pediatric diabetes team. Occasionally such a team is unavailable and the pediatrician must treat an acutely ill patient in the office. In this setting, the most appropriate response is to establish intravenous access and to administer a modest 10 to 20 mL/kg bolus of normal saline until the child can be transported to an urgent care facility.

Insulin therapy should not be attempted outside a controlled hospital setting. If the child is ill, he will require close monitoring. If the patient is well, therapy can safely be delayed several hours until he is seen by the diabetes team.

What should primary care pediatricians do--and what should we not do--when we encounter a patient with newly diagnosed diabetes?

Diagnostic Tips, Initial Management Strategies

Q: What are the indications for use of an insulin pump in very young children (infants and preschoolers)?

A: The results of the Diabetes Control and Complications Trial (DCCT)

 and the Epidemiology of Diabetes Interventions and Complications

 (EDIC) study provide the rationale for achieving strict glycemic control in all of our patients with type 1 diabetes. These studies demonstrate that intensive management reduces the incidence and progression of microvascular disease in the short term and long term as well.

In children--and in particular in the youngest subset of children (ie, infants and toddlers)--the concern has always been about safety, particularly the unknown risks of hypoglycemia on the developing brain. Older studies observed that cognitive problems are more common in children in whom diabetes was diagnosed at a younger age and that hypoglycemic brain damage is the mechanism. More recent studies suggest that hyperglycemia may be just as dangerous to young brains as hypoglycemia, but the limiting factor to diabetes treatment in infants and toddlers has always been fear of hypoglycemia.

Insulin pump therapy, in my opinion, is actually safer in this population than injection therapy for several reasons:

 Large doses of long-acting insulin are avoided, with the attendant unpredictable peaks and variabilities.

 Basal insulin coverage can be adjusted as needed for exercise, illness, etc.

 It offers freedom from rigid, inflexible meal schedules and contents (which is particularly important with toddlers).

 It improves the quality of life for parents, particularly mothers, who typically bear the greatest burden of diabetes care in the youngest patients.

Non-randomized prospective studies have clearly demonstrated that pump therapy improves diabetes control and reduces hypoglycemia risk, even in infants and toddlers,

 and is associated with improvements in quality of life.

 However, randomized controlled trials of pump therapy versus injection therapy in very young children have not shown the same consistent improvements.

 These studies show no significant improvement in hemoglobin A

 level and even slightly higher rates of non-severe hypoglycemia in the pump-treated patients. However, they also show significant improvement in quality of life in the pump groups.

What explains the discrepancy between non-randomized and randomized studies? Selection of patients is a key factor.In a real clinical practice, patients are not randomly assigned to treatment groups. The question, therefore, is not of efficacy but of effectiveness. Pump therapy is a very powerful tool in the right hands. Our guiding principle in the selection of patients has not been the ability or the inability to achieve good control with injections, nor have there been any age or socioeconomic criteria. Families must simply demonstrate a proficiency and commitment to the types of diabetes tasks that will be required for pump therapy. These include:

 A history of good adherence to testing.

 An understanding of the benefits and limitations of pump therapy (eg, a pump is not a cure for diabetes, just another way to take insulin).

 An understanding of the technical aspects of pump therapy (eg, the difference between basal and bolus; how to navigate controls on pump, change settings).

 Adequate adult supervision and commitment to follow-up.

 Ready access to a diabetes treatment team skilled and comfortable with the management of kids on pumps.

We have even found that children who have nannies or who are in daycare can do well on a pump. They do even better, in fact, than do children who are receiving injections.

The reader is referred to 2 reviews in a recent issue devoted to insulin pump therapy.

 The 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. 

Sustained effect of intensive treatment of type 1 diabetes mellitus on development and progression of diabetic nephropathy: the Epidemiology of Diabetes Interventions and Complications (EDIC) study.

Ahern JA, Boland EA, Doane R, et al. Insulin pump therapy in pediatrics: a therapeutic alternative to safely lower HbA

Weinzimer SA, Ahern JH, Doyle EA, et al. Persistence of benefits of contin-uous subcutaneous insulin unfusion in very young children with type 1 diabetes: a follow-up report. 

Sullivan-Bolyai S, Knafl K, Tamborlane W, Grey M. Patient's reflections on managing their children's diabetes with insulin pumps. 

 DiMeglio LA, Potteroff TM, Boyd SR, et al. A randomized, controlled study of insulin pump therapy in diabetic preschoolers. 

 Fox LA, Buckloh LM, Smith DS, et al. A randomized controlled trial of insulin pump therapy in young children with type 1 diabetes. 

 Wilson DM, Buckinham BA, Kunselman EL, et al. A two-center randomized controlled feasibility trial of insulin pump therapy in young children with diabetes. 

 Fisher LK. The selection of children and adolescents for treatment with continuous subcutaneous insulin infusion (CSII). 

 Weinzimer SA, Swan KL, Sikes KA, Ahern JH. Emerging evidence for the use of insulin pump therapy in infants, toddlers, and preschool-aged children with type 1 diabetes. 

What are the indications for use of an insulin pump in veryyoung children (infants and preschoolers)?

Stuart A. Weinzimer, MD

Articles

Diabetes Q&A: 70/30; "Rule of Fifths"; Split-Mix; Basal-Bolus-- Which Is Best?

Diagnostic Tips, Initial Management Strategies

What Role for an Insulin Pump for the Very Young?