Insulin Therapy in Children

Type 2 Diabetes Defeated

Home Treatment of Diabetes

Get Instant Access

William V. Tamborlane and JoAnn Ahern

Yale School of Medicine, New Haven, Connecticut


Type 1 diabetes meliitus in childhood and adolescence presents special challenges to pediatric health-care providers. The combination of severe insulin deficiency and the physical and psychoemotional changes that accompany normal growth and development make day-to-day management of pediatric patients especially difficult. Moreover, the results of the Diabetes Control and Complications Trial (DCCT) have raised the bar considerably higher with respect to goals of treatment, since intensive treatment was shown to significantly reduce the risk of progression of retinopathy and the development of microalbuminuria (1-3). Current recommendations mandate that youths with type 1 diabetes should aim to achieve metabolic control as close to normal as possible and as early in the course of the disease as possible. Remarkably, a much greater proportion of young patients are meeting strict standards of care than ever imagined possible only a few years ago. Our approach to insulin replacement in children and adolescents with type 1 diabetes is discussed below.


The traditional goals of treatment of children and adolescents with diabetes were to use insulin, diet, and exercise to minimize symptoms of hypoglycemia and hyperglycemia, promote normal growth and development, and, using intensive education and psychosocial support, maximize independence and self-management in order to reduce the adverse psychosocial effects of this chronic disease. Since the results of the DCCT were published, additional primary aims of therapy are to lower blood glucose and glycosylated hemoglobin values to as close to norma] as possible. In pediatric patients, achievement of such stringent treatment goals is best accomplished with a multidisciplinary team of clinicians to provide ongoing education and support of aggressive self-management efforts on the part of parents and patients. Matching the treatment to the patient (rather than vice versa) by taking a flexible and varied approach to insulin replacement, diet, and exercise is critically important.

It is recognized that intensive treatment places extra burdens on patients and families and that practical considerations, such as acceptability of and compliance with the treatment regimens, must be balanced appropriately to attain all these aims of therapy. Nevertheless, recent data suggest that an intensive approach to diabetes education and aggressive self-management by patients and families may reduce rather than increase the adverse psychosocial effects of this chronic illness (4).


Initiation of insulin treatment can be accomplished in either the inpatient or outpatient setting. Many youngsters require hospital admission due to vomiting, dehydration, and/or moderate to severe ketoacidosis. In patients who are not ill at presentation, admission to the hospital may also provide the child and parent with a safe and supportive environment in which to adjust to the shock of the diagnosis. Outpatient management in a comprehensive day treatment program staffed by individuals knowledgeable in the care of children with diabetes can also provide a supportive environment in which to initiate therapy (5) and such programs are becoming more widely available.

Once so simple, the choice of types of insulin and insulin regimen has become much more complicated. To the standard human Regular, NPH, Lente, and Ultralente insulins have been added new insulin analogs. Lispro and aspart insulin are produced by amino acid substitutions near the C-terminal end of the B-chain. These substitutions do not affect the biological actions of insulin but result in more rapid absorption than Regular insulin following subcutaneous injection. The sharper peak and shorter duration of these insulins compared with Regular insulin may be of particular advantage in teenagers who require large

pre-meal bolus doses of rapid-action insulin. There are fixed mixtures of both human insulin and human insulin analogs, and inhaled insulin preparations are currently under study (6). A sampling of the variety of conventional and unconventional insulin regimens is given in Table 1.

Although many clinicians start insulin treatment with three or more daily injections, we begin most newly-diagnosed patients on two injections of insulin per day using mixtures of human Lente (two-thirds) and lispro (one-third) insulins. The rationale for using two rather than three or more injections at onset of diabetes is that with aggressive control of blood levels, most children enter a "honeymoon" or partial remission period after a few weeks of therapy. This remission period is a result of increased insulin secretion by residual |3-celJs and improved insulin sensitivity with normalization of blood glucose levels (7). To achieve these effects, we start each patient on a total daily dose of at least one unit per kilogram body weight per day. Even more important, each component of the insulin regimen is adjusted on the basis of fingerstick blood glucose levels measured at least four times a day. The goal is to obtain pre-meal blood glucose values within the normal range, and this is achieved via daily telephone contact with the family for at least the first 3 weeks of treatment. The DCCT data indicate that strict control of diabetes also serves to prolong the period of residual (3-cell function in patients with type 1 diabetes (8).

During the "honeymoon," insulin requirements rapidly decrease. Commonly, the doses of rapid-acting insulin are sharply reduced or discontinued during this time; many children are well managed with two injections of intermediate-acting insulin and some may not even require an evening injection. In the absence of symptomatic hypoglycemia, however, we try not to lower the total daily dose of insulin below 0.20-0.25 units per kilogram body weight per day, since these are doses that were safely employed even in prediabetic children in the DPT-1 study (9).

Table 1 Sample Insulin Regimens








R + 1

R + 1

R + 1 H

- L

R + 1 + L

1 + L

1 + L


R + M

: L


1 ± R

R + 1


R + 1 ± L




1 or L ± R

Types of insulin: R = rapid-acting (Regular, lispro, aspart) insulin; I = intermediate-acting (NPH, Lente) insulin; L = long-acting (Ultralente).

Types of insulin: R = rapid-acting (Regular, lispro, aspart) insulin; I = intermediate-acting (NPH, Lente) insulin; L = long-acting (Ultralente).

A major reason that the two-daily-injections regimen is effective during the honeymoon phase is that endogenous insulin secretion provides much of the overnight basal insulin requirements, leading to normal fasting blood glucose values. Thus, increased and more labile pre-breakfast glucose levels often herald the loss of the relatively small amount of residual endogenous insulin secretion that is required for overnight glucose control. When residual (3-cell function wanes, problems with the two-injection regimen become apparent. One is that the peak of the pre-dinner intermediate-acting insulin may coincide with the time of minimal insulin requirements (i.e., midnight to 4 a.m.). Subsequently, insulin levels fall off when basal requirements are increasing (i.e., 4 to 8 a.m.). Increasing the pre-supper dose of intermediate-acting insulin to lower fasting glucose values often leads to hypoglycemia in the middle of the night without correcting hyperglycemia before breakfast. Patients are especially vulnerable to hypoglycemia in the middle of the night because the normal plasma epinephrine response to low blood glucose levels is markedly blunted during deep sleep (10). Another problem with the conventional two-injection regimen is high pre-supper glucose levels, despite normal or low pre-lunch and mid-afternoon values. This is due in part to eating an afternoon snack when the effects of the pre-breakfast dose of intermediate-acting insulin is waning.

One way to deal with these problems without increasing the number of injections is to add Ultralente insulin to the pre-breakfast and pre-supper mixtures of lispro and Lente. With this combination in the morning, lispro covers breakfast, Lente covers lunch, and Ultralente the late afternoon period. With the pre-supper dose, lispro covers supper, Lente covers the bedtime snack and part of the overnight period, and Ultralente helps limit the pre-breakfast rise in plasma glucose. However, when strict control cannot be achieved with two daily injections, we do not hesitate to switch to a regimen involving three or more daily injections. A common approach to the problem in the overnight period is to use a three-injection regimen: lispro and Lente at breakfast, lispro only at dinner, and Lente at bedtime. For youngsters who go to bed early, we recommend that parents give the third shot at their bedtime (i.e., 10:00-11:00 p.m.). Lispro can also be added to the bedtime dose, especially if glucose levels are elevated. For patients with elevated pre-supper glucose levels, a pre-lunch dose of Regular or a preafternoon-snack dose of lispro can be added. Such extra doses of insulin can be facilitated by the use of insulin pens, which are small, light, and easy to use. Only a small number of our patients are using a regimen of four or more injections of rapid-acting insulin before meals and intermediate insulin at bedtime.

Over the past few years, there has been a rediscovery of the effectiveness of insulin pump therapy in the management of young patients with diabetes (11). Indeed, we are much more likely to turn to this method of insulin replacement than to more frequent injections in preadolescents who are coming out of their honeymoon phase of diabetes.

With insulin pump treatment, small amounts of rapid-acting insulin are infused as a basa] rate, and larger bolus doses are given at each meal or snack. Although it is not yet labeled by the FDA for use in pumps, lispro insulin appears to have advantages over Regular insulin in pump therapy (12). The pumps are battery-powered and about the size of a beeper. The "basal" rate can be programmed to change every half hour, but it is unusual to need more than five or six basal rates. Varying the basal rate can be particularly helpful in regulating overnight blood glucose levels, since it can be lowered for the early part of the night to prevent hypoglycemia and increased in the hours before dawn to keep glucose from rising. However, younger children seem to need a higher basal rate during the night, perhaps due to earlier nocturnal peaks of growth hormone in this age group. Bolus doses are given before meals based on glucose level, activity, and food intake. Pump treatment can be especially useful in infants and toddlers who are picky eaters. In this setting, part of the usual pre-meal bolus can be given prior to the meal and the rest at the end of the meal depending on the actual amount of carbohydrate intake. Indeed, most children and parents are encouraged to use carbohydrate counting (see below) as a means to adjust pre-meal bolus doses. Pump therapy also enhances flexibility in children with variable exercise and meal routines.

The pump employs a reservoir (syringe) to hold the insulin and the infusion set, which consists of tubing with a small plastic catheter at the end. The insertion site can be the abdomen or hip area, except in the young child in whom there may not be sufficient subcutaneous tissue in the abdomen. Our patients are encouraged to change their catheters every other day. Because only a rapid-acting insulin is used in this pump, the child and parent must understand that the insulin infusion should not be discontinued for more than 4 hours at a time.

Case Study: A Child with Newly Diagnosed Diabetes

An 11-year-old girl who weighed 40 kg was diagnosed with diabetes in her pediatrician's office based on a fingerstick blood glucose level of greater than 300 mg/ dl and glycosuria but no ketonuria. Since she was not ketotic, she was admitted directly to the hospital rather than being referred to the emergency department. The initial total daily dose was 40 units (1 unit/kg body weight): 27 units before breakfast and 13 units before supper. Each of the doses was further divided as one-third lispro and two-thirds Lente insulin. Thus, the pre-breakfast dose was 9 units lispro and 18 units Lente and the pre-supper dose was 4 units lispro and 9 units Lente. These were just the starting doses, and they were adjusted every day based on blood glucose levels. After she went home, her mother was contacted daily by telephone at least once a day for about 4 weeks. By 2 weeks, insulin doses had to be decreased because of low blood glucose readings. By 4 weeks, blood glucose levels were much less labile and the family felt much more comfortable about making their own insulin dose adjustments.


Intermediate- and long-acting preparations of human insulin suffer from a number of pharmacological problems, not the least of which is the failure to uniformly mix these suspensions prior to injection. Human NPH insulin is a particularly poor choice for basal insulin replacement, since there is a substantial peak in insulin levels and insulin action 2-6 hours after subcutaneous injection. Even human Ultralente has a significant peak action 8-12 hours after injection and the dose-to-dose variability in absorption of all human insulin suspensions is substantial.

Glargine insulin (Aventis Pharmaceuticals) is an analog of human insulin with C-terminal elongation of the B-chain by two arginines and replacement of asparagine in position A21 by glycine. This molecule is soluble in the acidic solution in which it is packaged but relatively insoluble in the physiological pH of the extracellular fluid. Consequently, microprecipitates of glargine insulin are formed following subcutaneous injection, which markedly delays its absorption into the systemic circulation. Pharmacokinetic and pharmacodynamic studies have demonstrated that the insulin analog has a very flat and prolonged time-action profile. Results of a 6-month study of the efficacy and safety of glargine in adolescents and children with diabetes showed modestly lower fasting bloodl glucose levels and/or reduced risk of hypoglycemia with glargine than with hu-j man NPH insulin. Additional studies and more clinical experience need to be accumulated regarding use of this analog in youngsters with type 1 diabetes.! Because glargine cannot be mixed with other insulins, it has to be given by separate injection, which might affect its acceptability to some youngsters.

Although there have been many failed attempts at finding alternatives to insulin injections (13), use of aerosolized preparations for inhaled insulin delivery is currently under active investigation. Preliminary studies in adults have been promising enough (6) that phase 3 studies are already underway in preadolescents as well as adolescents with type 1 diabetes. Like pump therapy, inhaled insulin allows the patients to take pre-meal boluses of insulin with each meal and snack without having to take extra insulin injections. However, one or more injections of intermediate- or long-acting insulin are still needed for basal insulin replacement.


Insulin replacement in children is a special challenge because insulin requirements increase as weight and calorie intake increase and as residual endogenous secretion declines. Regular self-monitoring of blood glucose (SMBG) allows the family and clinicians to keep up with the child's steadily increasing insulin needs. We request that blood glucose levels be checked at least four times per day (be s

fore each meal and at bedtime). The most important component of SMBG is the interpretation of the results. The parent or child must be taught the target value and the relationship among diet, exercise, and insulin. If the parent and/or child grasp these concepts, they will make accurate adjustments aimed at achieving target goals. If they are unable to make accurate adjustments, they should be given guidelines on when to call the diabetes service for help. Day-to-day adjustments in the doses of rapid-acting insulin can be made based on the pre-meal blood glucose value, amount of carbohydrate in the meal, and amount of anticipated exercise. In addition, patients and parents should be taught to look for repetitive patterns of hypo- or hyperglycemia, in order to make ongoing changes in the usual insulin doses. To facilitate identification of trends, families are encouraged to maintain either a handwritten or computer-generated record of glucose values in spreadsheet format.

Self-monitoring of blood glucose is subject to a variety of problems—for example, patients often make up false numbers (14). These issues must be addressed with the child and family. They must understand the reason for the tests and that they are used only to make proper adjustments to keep them healthy. Elevated glucose levels are not an indication that the diabetes is worsening or that patients have been cheating on their diet. Instead, we emphasize that the! tests are being done primarily to determine when they have outgrown their current dose of insulin.

Even when performed correctly, four blood tests daily give only a limited glimpse of the wide fluctuations in blood glucose that occur during a 24-hour period in children with diabetes. Consequently, the recent introduction of continuous glucose monitoring systems has the potential to be the most important advance in assessing diabetes control in the past 20 years. In intensively treated children and adolescents with type 1 diabetes, preliminary results in a relatively small number of children suggest that continuous glucose monitoring will provide a wealth of data regarding postprandial glycemic excursions and asymptomatic nocturnal hypoglycemia that were unavailable from capillary blood glucose measurements (15). We anticipate that these technological breakthroughs will have a great impact on diabetes management over the next few years. Continuous monitoring of nocturnal glucose levels is likely to be particularly useful in programming overnight basal rates in pump-treated patients.

Measurements of glycosyolated hemoglobin (HbA,c) provide the gold standard by which to judge the adequacy of die insulin regimen. A variety of methods are available for assaying glycosylated hemoglobin. A simple method that can be performed in the office in 6 minutes (Bayer DCA 2000) offers the opportunity to make immediate changes in the insulin regimen, while the patient is being seen. The goal of treatment is to achieve HbA]c levels as close to normal as possible. Based on DCCT results (2), our general goal of therapy is to try to keep all patients under 8.0%. HbA]c levels are determined at least every 3 months.


Diet guidance for children with diabetes is best provided by a nutritionist who is an integral part of the treatment team and comfortable working with children. In addition to helping achieve optimal glucose levels and normal growth and development, nutritional management of diabetes is aimed at reducing the risk for other diseases such as obesity, high blood cholesterol, and high blood pressure. Underlying all these is the establishment of sound eating patterns that include balanced, nutritious foods and consistent timing of food intake (16).

The American Diabetes Association dietary guidelines are used for dietary counseling. In addition to incorporating sound nutritional principles concerning fat, fiber, and carbohydrate content, the importance of consistency in meal size and regularity in the timing of meals is emphasized. The prohibition of simple sugar in the diet has been de-emphasized, but it should still comprise no more than 10% of total carbohydrate intake. The success of the nutritional program may ultimately depend on the degree to which the meal planning is individualized and tailored to well-established eating patterns in the family. Moreover, flexibility can be enhanced if blood glucose monitoring results are used to evaluate the impact of change in dietary intake. As with other aspects of the treatment regimen, we preach consistency and teach how to adjust for deviations from the prescribed diet.

Carbohydrate counting, an increasingly popular way to increase flexibility in food intake, is commonly used by patients using insulin pumps or multiple daily injections. The amount of insulin that is needed for each gram or serving of carbohydrate is used to calculate the amount of Regular or lispro insulin to be taken depending on the amount of carbohydrate in the meal. With instructions on how to use nutritional labels on food packages, even children can become expert at counting carbohydrates. An even simpler method is to vary the dose of Regular or lispro by one or two units according to whether it is a small, regular, or large meal. Some foods—pizza, for example—that cause a prolonged increase in blood glucose levels may require an increase in the amount of intermediate-acting insulin or a temporary change in overnight basal rates in pump-treated patients.


Regular exercise and active participation in organized sports have positive implications for the psychosocial and physical well-being of our patients. Parents and patients should be advised that different types of exercise may have different effects on blood glucose levels. For example, sports that involve short bursts of intense exercise may increase rather than decrease blood glucose levels (17). On the other hand, long-distance running and other prolonged activities are more

likely to lower blood glucose levels. Parents also need to be warned that a long bout of exercise during the day may lead to hypoglycemia while the child is sleeping during the night, which may require a reduction in the dose of intermediate- or long-acting insulin.


Children and adolescents with type I diabetes should be routinely cared for by a diabetes center that uses a multidisciplinary team knowledgeable about and experienced in the management of young patients. Ideally, this team should consist of pediatric diabetologists, diabetes nurse specialists, nutritionists, and social workers or psychologists.

In newly diagnosed patients, the first few weeks are critically important in the process of teaching self-management skills to the parent and child. With this age group, the parent is usually in daily contact with the diabetes clinical nurse specialist. Glucose levels, adjustment to diabetes, diet, and exercise are reviewed. The timing of the phone calls should be prearranged and ideally made to the same clinician. After making the insulin adjustment for the day, the rationale should be explained to the parent. Usually within 3 weeks the parents are feeling more confident and many are ready to attempt to make their own adjustments.

Once glucose levels have been stabilized, regular follow-up visits every 2 or 3 months are recommended for most patients (18). The main purpose of these visits is to ensure that the patient is achieving primary treatment goals. In addition to serial measurements of height and weight, particular attention should be paid to monitoring of blood pressure and examinations of the optic fundus, thyroid, and subcutaneous injection sites. Routine outpatient visits provide an opportunity to review glucose monitoring, to adjust the treatment regimen, and to assess child and family adjustment. Follow-up advice and support should be given by the nutritionist, diabetes nurse specialist, and psychologist or social worker. Use of the telephone, fax, or email should be encouraged for adjustments in the treatment regimen between office visits.


Severe hypoglycemia is a common problem in patients striving for strict glycemic control with intensive treatment regimens. In the DCCT, the risk of severe hypoglycemia was threefold higher in intensively treated patients than in conventionally treated patients, and being an adolescent was an independent risk factor for a severe hypoglycemic event, as mentioned earlier (2). The majority of severe hypoglycemic events occur overnight due, in part, to sleep-induced defects in counterregulatory hormone responses to hypoglycemia (10).

Monitoring glucose is critical in order to detect asymptomatic hypoglycemia, especially in the young child with diabetes. The older child is usually aware of such symptoms as weakness, shakiness, hunger, or headache, and is encouraged to treat these symptoms as soon as they occur. The older child who can accurately recognize symptoms is taught to immediately treat with 15 grams of carbohydrate (e.g., three or four glucose tablets, 4 ounces of juice, or 15 grams of a glucose gel) without waiting to check a glucose level. Each episode should be assessed in order to make proper adjustments if a cause can be identified. Every family should have a glucagon emergency kit at home in order to treat severe hypoglycemia.


Children with intercurrent illnesses, such as infections or vomiting, should be closely monitored for elevations in blood glucose levels and ketonuria. On sick days, blood glucose levels should be checked every 2 hours and the urine should be checked for ketones with every void. Supplemental doses of short-acting insulin (0.1 to 0.3 units/kg) should be given every 2 to 4 hours for elevations in glucose and ketones. Because of its more rapid absorption, lispro will lower plasma glucose faster than Regular insulin. If the morning dose has not been given and the child has a modestly elevated glucose level (150 to 250 mg/dl), small doses of NPH can be given to avoid too rapid a fall in plasma glucose levels. This works especially well in young children whose glucose levels fall quickly with rapid-acting insulin. Adequate fluid intake is essential to prevent dehydration. Fluids such as flat soda, clear soups, popsicles, and gelatin water are recommended to provide some electrolyte and carbohydrate replacement. If vomiting is persistent and ketones remain moderate or high after several supplemental insulin doses, arrangements should be made for parenteral hydration and evaluation in the emergency department.

Children receiving Ultralente insulin seem to be prone to the development of hypoglycemia and ketonuria during episodes of gastroenteritis. If the child is unable to retain oral carbohydrate, then small doses of glucagon (i.e., 0.1-0.2 mg), given subcutaneously every 2 to 4 hours, can be used to maintain normal blood glucose levels.

Parents are told from the time of diagnosis that vomiting is a diabetes emergency and that they need to call for help after first checking blood glucose and urine ketone levels. This is especially true for children on pump therapy, since a catheter occlusion can throw the child into ketosis rapidly. If a pump-treated patient has elevated glucose and ketone levels, they are instructed to take a bolus injection of lispro insulin by syringe. The dose of insulin varies between 0.2 and 0.4 units per kilogram. They are then instructed to change their infusion set and to program a temporary basal rate at twice the usual basal rate for 4-5 hours.

Blood glucose and urine ketone levels are rechecked every hour, and additional bolus doses can be given as needed. Once vomiting ceases and ketones become negative, the basal rate is returned to its usual setting. If the patient is not improving with these measures, then the child should be evaluated by a physician.


1. DCCT Research Group. The effects of intensive diabetes treatment on the develop-menl and progression of long-term complications in insulin-dependent diabetes mel-litus: the Diabetes Control and Complications Trial. N Engl J Med 329:977-986, 1993.

2. The DCCT Research Group. The effect of intensive diabetes treatment on the development and progression of long-term complications in adolescents with insulin-dependent diabetes mellitus: the Diabetes Control and Complications Trial. J Pediatr 125:177-188, 1994.

3. DCCT Research Group. Prolonged effect of intensive therapy on the risk of advanced complications in the Epidemiology of Diabetes Intervention and Complications (EDIC) follow-up of the DCCT cohort. N Engl J Med 342:381-389, 2000.

4. Grey M, Boland EA, Davidson M, Li J, Tamborlane W. Coping skills training for youth on intensive therapy has long lasting effects on metabolic control and quality of life. J Pediatr 137:107-114, 2000.

5. Dougherty G, Schiffrin A, White D, Soderstrom I, Sufrategui M. Home-based management can achieve intensification cost-effectively in type I diabetes. Pediatrics 103:122-128, 1999.

6. Patton JW, Bukar J, Nagarajan S. Inhaled insulin. Advanced Drug Deliv Rev 36: 235-247, 1999.

7. Yki-Jarvinen H, Koivisto VA. Natural course of insulin resistance in type I diabetes. N Engl J Med 315:224-230, 1986.

8. The DCCT Research Group. The effect of intensive diabetes treatment in the DCCT on residual insulin secretion in LDDM. Ann Intern Med 128:517-523, 1998.

9. Schatz DA, Rogers DG, Brouhard BH. Prevention of insulin-dependent diabetes mellitus: an overview of three trials [review], Cleveland Clinic J Med 63:270-274, 1996.

10. Jones TW, Porter P, David EA, et al. Suppressed epinephrine responses during sleep: a contributing factor to the risk of nocturnal hypoglycemia in insulin-dependent diabetes. N Engl J Med 338:1657-1662, 1999.

11. Boland EA, Grey M, Fredrickson L, Tamborlane WV. CSD: a "new" way to achieve strict metabolic control, decrease severe hypoglycemia and enhance coping in adolescents with type I diabetes. Diabetes Care 22:1779-1894, 1999.

12. Zinman B, Tildesley H, Chiasson JL, Tsui E, Strack T. Insulin lispro in CSII: results of a double-blind crossover study. Diabetes 46(3):440-443, 1997. Erratum appears in Diabetes 46(7): 1239, 1997.

13. Moses AC, Gordon GS, Carey MC, Flier JS. Insulin administration intranasally as an insulin-bile salt aerosol: effectiveness and reproducibility in normal and diabetic subjects. Diabetes 32:1040-1047, 1983.

Hyperuricemia And Exercise

Was this article helpful?

0 0
Weight Loss Resolutions

Weight Loss Resolutions

Are You Tired Of Failed New Year's Weight Loss Resolutions That Leave You Even More Overweight Than Ever Before? Not Anymore Finally Succeed With Your New Years Weight Loss Resolution Once And For All And Get The Body Of Your Dreams.

Get My Free Ebook

Post a comment