Individualized insulin therapy in children and adolescents with type 1 diabetes

Transcription

1 Acta Pædiatr Suppl 425: Individualized insulin therapy in children and adolescents with type 1 diabetes D Becker Department of Pediatrics, Children s Hospital of Pittsburgh, Pittsburgh, PA, USA Becker D. Individualized insulin therapy in children and adolescents with type 1 diabetes. Acta Pædiatr 1998; Suppl 425: Stockholm. ISSN The most important point about individualized therapy is to make it flexible to fit the needs of the patient. These needs are determined by age, pubertal status, treatment schedule, exercise, intelligence, education and socioeconomic status. In addition, major emphasis should be placed on the dietary needs, attitude and ability of the patient. Aspects of insulin regimens that can be adjusted include number of injections per day, timing of injections and the type of insulin used. Several factors that affect insulin action should also be considered, including the presence of insulin antibodies, the amount of subcutaneous fat, injection technique and muscular activity. Insulin pump therapy in childhood and adolescence may also be considered in certain cases. Controversial areas in childhood diabetes therapy include the number of daily injections, whether porcine insulin has any advantages, use of multiple doses of intermediate- and long-acting insulin, mixing regular insulin and short-acting insulin analogues, strategies to prevent hypoglycaemia and the importance of choice of injection sites in childhood. Cultural differences, insulin absorption, insulin analogues, insulin delivery systems, insulin preparations D Becker, Department of Pediatrics, Division of Endocrinology, Metabolism and Diabetes Mellitus, Children s Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh, PA 15213, USA Insulin therapy is a cornerstone of the treatment of individuals with type 1 diabetes, but it cannot be used as the only therapeutic strategy in the quest to maintain target glycaemic goals. Insulin therapy must be integrated with other aspects of treatment, with careful attention to the daily food plan and variations in activity levels. Attention to all of these factors on a frequent basis constitutes intensive diabetes therapy. Intensive insulin therapy, which consists of the delivery of multiple daily injections mixing short and longer duration insulin preparations, is a component of intensive diabetes therapy. The insulin regimen must be individualized for each patient and tailored to the needs of the patient, the availability of an experienced health-care team and the availability of care in the region (e.g. telephone contact and/or frequent visits to the diabetes centre). Patient factors that need to be taken into account are listed in Table 1. It is important to recognize that insulin requirements increase with both the duration of diabetes and the age and pubertal status of the patient. Insulin requirement has been clearly shown to increase during puberty owing to its associated insulin resistance (1), with doses often exceeding 1.2 IU kg 1 d 1 and in the author s experience, being higher in girls than in boys (2). There is a large interindividual variation in dose requirements, related to food intake, physical activity and other endogenous factors, which results in varying insulin sensitivity. The choice of the insulin regimen needs to address the schedule of each particular patient as well as the ability of the family to support diabetes care. These factors are affected by the education level and socioeconomic environment of the family, as well as their ability to adjust insulin doses. Insulin regimens may vary across different cultures, both within countries and between countries, depending on the food content of the diet, the timing of meals and cultural variations in snacking between meals. The willingness of a family or individual with diabetes to spend the required time and effort to adjust insulin doses depends on the health beliefs engendered by the environment. Concepts of the ability to control diabetes and the value of extra effort in influencing long-term outcomes vary among individuals and different societies. Thus, the goals of therapy, as measured by the glycosylated haemoglobin, should be clearly determined for each individual. The challenge is to achieve target glycaemic levels without excessive hypoglycaemia. Insulin delivery A number of studies have shown marked variation in insulin absorption rates from the subcutaneous tissue. Some of the factors causing this variation are listed in Table 2. Some of the most important factors are the variation in the amount of subcutaneous tissue into which the injection is administered, differences in the needle length and the angle at which the injection is given. These factors result in variations in the depth of the injection and, in young children in Scandinavian University Press ISSN

2 ACTA PÆDIATR SUPPL 425 (1998) Individualized insulin therapy 21 Table 1. Patient factors affecting the selection of an individualized insulin regimen. Age Pubertal status Duration of diabetes Home schedule School schedule Exercise and activities Education level Ability to adjust doses Socio-economic environment Family support particular, this can mean frequent intramuscular rather than subcutaneous delivery of insulin (3). A number of studies has reported that insulin is absorbed more rapidly, but has a shorter duration of action, when injected into the abdomen and arm compared with the leg or hip (4). These results have not been confirmed in the paediatric population, although anecdotal experience suggests that this may also be true for children. It is obvious, by the law of kinetics, that the larger the dose of insulin, the more rapid its absorption and the longer its duration of action. Lastly, the presence of high-titre insulin antibodies can act as a reservoir or sump for insulin molecules and delay its onset of action as well as extend its duration (5). Insulin delivery systems Table 3 shows the currently available insulin delivery systems. The intraperitoneal pump remains a research tool and is not yet available for routine clinical care. The insulin delivery system, again, should depend on the individual needs of the patient, including the frequency of daily injections and the mixture of insulin required at each injection. Thus, a patient using a mixture of long- and short-acting insulin has to use a syringe rather than a pen unless the exact formulation of NPH and regular insulin is available in cartridge form, as it is in some countries. It is important to recognize that variable needle lengths are available for both pens and syringes and the appropriate choice should be made for each patient depending on the amount of subcutaneous tissue. Many patients prefer the convenience of pen systems and these are increasingly appropriate for young children with the availability of one-unit and half-unit dose increments. In the author s experience, however, it is rare that a patient prefers two injections with a pen to one with a syringe when both Table 2. Factors affecting the absorption and action of insulin. Amount of subcutaneous tissue Injection site: abdomen, arm, leg or hip Depth of injection Injection technique Insulin antibodies Dose Table 3. Currently available insulin delivery systems. System Syringe Pen Indwelling catheter Subcutaneous pump Intraperitoneal pump Comment Needle length can vary Most suitable for mixtures unless an exact premix is available Needle length can vary Probably the most convenient delivery system Not recommended in patients without an experienced team Not recommended for young children Should be considered only in motivated and reliable patients Research tool short- and long-acting insulin need to be administered at the same time. Some centres have had a fair amount of experience with the use of an indwelling catheter in the abdomen, which allows multiple insulin injections into a diaphragm rather than into the skin; however, the literature on this system is extremely limited. Continuous subcutaneous insulin infusion (CSII) is the most physiological of all these delivery systems, although the peripheral rather than portal entry of insulin into the circulation still makes this an abnormal route of insulin delivery. There are both advantages and disadvantages to the use of insulin infusion pumps. For the motivated, mature, intelligent adolescent, the insulin infusion pump gives a great degree of flexibility in alterations of the basal rate and timing of premeal insulin boluses. The disadvantages are that the new-found flexibility in timing and meal content is often difficult to handle and glycosylated haemoglobin levels often do not improve or even deteriorate after CSII is introduced. Despite a 2-month trial period, interaction with other adolescents using insulin infusion pumps and intensive education, a significant number of adolescents is unprepared for the extra effort required to maintain an insulin infusion pump and the time required for changes in infusion sites. A constant concern is that even a short interruption in the insulin delivery can result in significant ketosis or even ketoacidosis, owing to the absence of a long-acting insulin reservoir. In the USA, insulin infusion pumps remain extremely expensive to buy and the infusion delivery systems also cost more than insulin syringes. Although some patients succeed in the use of CSII with the support of experienced clinicians and nurses, this method of insulin delivery is probably better restricted to highly motivated and reliable adolescents with appropriate health-care team resources. The cost/benefit ratio remains in question. Insulin preparations A variety of insulin is available with different timing of onset and duration of action. In the past, insulin action was

3 22 D Becker ACTA PÆDIATR SUPPL 425 (1998) Table 4. Insulin action. Length of action of insulin Ultra-long Long Intermediate Short Ultra-short Example Beef ultralente (zinc, analogues) Human ultralente Lente (zinc) Semi-lente (zinc) NPH (protamine) Regular (soluble) Analogues manipulated with formulations of zinc (lente) preparations or protamine (NPH). There is currently a major effort among pharmaceutical companies to produce insulin analogues that either prolong or shorten the onset and duration of insulin action when injected. Examples of these preparations can be seen in Table 4. These vary from ultra-long action, such as beef ultralente (which has now been taken off the market in a number of countries), to ultra-short action. Analogues are currently being studied to replace the smooth, long-term action of beef ultralente, which is very convenient as a basal insulin for multiple daily injections. Insulin action is not only determined by the addition of zinc or protamine, but also varies according to the molecular structure (which is the basis of analogue development). The relatively small amino acid differences among beef, pork and human insulin result in significant variations in onset and duration of action. Owing to the induction of antibodies, beef insulin use is decreasing around the world and, in many countries, this type of insulin is no longer available. Human insulin has largely replaced purified pork preparations in many countries. Some human insulin preparations, particularly regular NPH and lente, have been shown in a number of studies to have a more rapid onset of action and a slightly shorter duration of action than their pork counterparts. The theoretical advantage of the slightly shorter onset of action of human insulin allows a slightly shorter wait between the insulin injection and the meal in order to achieve maximum action at the time of postprandial hyperglycaemia. However, the slightly shorter duration of some human NPH and lente insulin compared to pork is a disadvantage in patients on conventional insulin therapy with two injections a day when supper is eaten early. If the evening dose is given in the late afternoon, the action has waned by breakfast the following morning unless the dose is fairly large, in which case hypoglycaemia soon after midnight becomes common. The author s group, based in the USA, usually uses pork insulin in the treatment of type 1 diabetes in children, although this is not common practice in Europe. In a randomized study, in children under the age of 5 y, glycosylated haemoglobin is significantly lower in patients treated with conventional therapy on pork compared to human insulin. In older children glycosylated haemoglobin values were no different, but blood sugars were less erratic and easier to control using pork insulin (6). In the author s clinic, pork insulin is usually used in patients treated with three or even four injections a day, and the longer duration of action of pork NPH or lente insulin makes overall glycaemic control easier. The disadvantage of pork insulin that it is more expensive than human insulin in the USA. Insulin antibody production was no different for human and pork preparations over a 5-y period. The very rapid onset of action of the short-acting insulin analogue, Lispro, allows the insulin to be injected immediately before eating, instead of waiting for min to prevent postprandial hyperglycaemia (7). This advantage is, however, counterbalanced by the shorter duration of action and thus preprandial hyperglycaemia requiring an increase in basal insulin delivery. This probably accounts for the failure of prevention of the usual postprandial hyperglycaemia seen with the use of regular insulin in large multicentre studies in improving in glycosylated haemoglobin (7, 8). Because of the shorter duration of action, less frequent episodes of nocturnal hypoglycaemia have been reported in these studies, although this is dependent on the carbohydrate content of the meal (9). Published data on short-acting insulin analogues in children under the age of 12 y are lacking, and although the international trials of Lispro included children over 12 y, data are limited to adolescents. In very young children, studies are underway to assess the efficacy of giving insulin injections including Lispro after the meal, once a parent can determine how much the child has eaten. The rapid action of Lispro, compared to that of regular insulin, is achieved by reversing the sequence of leucine and proline on the C-terminal of the B-chain of insulin, resulting in a structure similar to insulin-like growth factor-1. This structure results in a minimal tendency to selfassociate into dimers, which is reduced approximately 300 times compared with that of regular insulin, which regularly forms hexamers (7, 8). Despite this small amino acid change, no differences have been reported between short-acting insulin analogues and regular insulin in antibody induction (10). A number of strategies has been used in an attempt to prevent the preprandial hyperglycaemia regularly seen with the use of short-acting insulin analogues and human ultralente, lente or NPH insulin. These include mixing Lispro with regular and ultralente insulin in a study in adults, which gave a suggestion of improvement in overall glycaemia (11). Another group attempted to solve the problem by administering human NPH insulin three times a day, rather than the usual once or twice (12). Thus, a number of strategies using mixed insulin preparations can be used to tailor insulin action to individual patient needs. In the author s opinion, premixed insulins are seldom indicated in the paediatric population, as they do not allow variations of short-acting insulin in relationship to the size of the meal to be eaten, the ambient blood sugar and the physical activity over the next few hours. However, there is no doubt that under certain psychosocial circumstances, the use of premixed insulin,

4 ACTA PÆDIATR SUPPL 425 (1998) Individualized insulin therapy 23 particularly with an insulin pen, makes delivery easier and more consistent in the reticent patient. In very young children, half-unit variations in insulin appear to make differences in subsequent glycaemia, although this has not been proven in rigorous studies. As accurate delivery of these small amounts of insulin with a syringe is highly questionable, some clinicians prefer to dilute 100 IU ml 1 insulin. If this is done, both the parent or other carer and the health-care team should be extremely careful to avoid potential confusion regarding the amount of insulin to be delivered. Insulin should always be diluted using the company s diluting fluid rather than saline because of questions regarding insulin integrity. Frequency of daily injections Three decades ago, when most insulin contained beef preparations, antibody induction was common, resulting in a fairly long duration of action of all insulin preparations. Some patients achieved quite good glycaemic control on one injection a day at that time, when meal times were rigidly structured. With current insulin preparations, one daily injection is seldom indicated, although sometimes used in children during the remission period when tiny insulin doses are all that can be tolerated if hypoglycaemia is to be prevented. Conventional insulin therapy, consisting of two injections a day of mixed short- and intermediateacting insulins, is still the most common regimen in many parts of the world in the paediatric population. The move towards intensified insulin therapy and the availability of only human preparations in many countries requires an increase in frequency to three or four injections a day, usually delivered before each major meal and at bedtime. It is logical to assume that premeal injections of insulin are more physiological than large boluses each day. However, multiple daily injections are not always practical in certain circumstances, including: the very young child who fights each injection; children who are at school at lunchtime and are too young to administer their own insulin; families who are unable to sustain the rigour of four daily injections. The advantages of multiple doses are the increased flexibility in allowing changes in the amount of insulin given in relationship to the ambient blood sugar, type of meal to be eaten and activity. A major area of controversy among clinicians is whether this translates to improved glycaemic control or changes in the frequency of hypoglycaemia. In an international multicentre study of children and adolescents, glycosylated haemoglobin was no different among those receiving two, three or four injections a day (13). If intermediate- or long-acting insulin is given at suppertime (16:00 17:00 h in many countries), early morning hyperglycaemia due to waning of insulin action at the time of increased insulin requirement during the dawn phenomenon has frequently been demonstrated. In a paediatric population, this phenomenon was addressed by giving ultralente human insulin with no difference in glycosylated haemoglobin levels (14). In the author s experience, two injections a day of human ultralente insulin in children under the age of 10 y was extremely difficult to handle. However, the results of a randomized, crossover, multicentre trial are not yet available. When using multiple injections, a common regimen is breakfast, lunch and supper doses of regular insulin with bedtime NPH. However, if meals are far apart, the lack of a daytime intermediateacting insulin results in preprandial hyperglycaemia. The addition of NPH or lente insulin in the morning is useful under these circumstances. In a study in adults using human insulin preparations, overall glycaemia was found not to be different with the addition of daytime NPH in patients who were extremely well controlled, although glycaemic excursions were smaller (15). There has been no randomized study of this type of regimen in children. Conclusions Of the multiple strategies of insulin delivery, each has its advantages and disadvantages. It is important for the health-care team to become familiar with the pros and cons of each regimen and use frequent daily blood sugar monitoring to understand the effectiveness of any particular regimen in an individual patient. The assumption cannot be made that a regimen that works in one patient will be the same in another. The recognition that age and body composition play important roles in determining the type and frequency of the choice of insulin regimen is essential. The regimen should be matched with food intake, activity and of major importance the prevention of hypoglycaemia. At this time, there is no one optimal insulin delivery regimen, as each has its own imperfections. The development of new insulin and insulin delivery systems has resulted in major progress over the past 2 decades and advances in the field leave us with an optimistic view of continued success in the future. References 1. Caprio S, Plewe G, Diamond MP, Simonson DC, Boulware SD, Sherwin RS, Tambourlane WV. Increased insulin secretion in puberty: a compensatory response to reductions in insulin sensitivity. J Pediatr 1989; 114: Drash AL. Clinical care of the diabetic child. Chicago, IL: Mosby Yearbook Medical Publishers, Polak M, Beregszaszi M, Belarbi N, Benali K, Hassan M, Czernictio W, Tubiana-Rufi N. Subcutaneous or intramuscular injections of insulin in children. Are we injecting where we think we are? Diabetes Care 1996; 19: Binder C, Lauritzen T, Faber O, Pramming S. Insulin pharmacokinetics. Diabetes Care 1984; 7: Van Haften TW, Heiling VJ, Gerich JE. Adverse effects of insulin antibodies on postprandial plasma glucose and insulin profile in diabetic patients without immune insulin resistance. Implications for intensive insulin regimens. Diabetes 1987; 36: Eckenrode KL, Coonrod BA, Patterson S, Becker DJ. Clinical and

5 24 D Becker ACTA PÆDIATR SUPPL 425 (1998) metabolic effects of human and pork insulins in children with IDDM. Diabetes 1986; 35 (Suppl 1): Barnett AH, Owens DR. Insulin analogues. Lancet 1997; 349: Holleman F, Hoekstra JBL. Insulin lispro. N Engl J Med 1997; 337: Burge MR, Castillo KR, Schade DS. Meal composition is a determinant of lispro-induced hypoglycemia in IDDM. Diabetes Care 1997; 20: Fineberg NS, Fineberg SE, Anderson JH, Birkett MA, Gibson RG, Hufferd S. Immunologic effects of insulin lispro [Lys (B28), Pro (B29) human insulin] in IDDM and NIDDM patients previously treated with insulin. Diabetes 1996; 45: Burge MR, Waters DL, Holcombe JH, Schade DS. Prolonged efficacy of short acting insulin lispro in combination with human ultralente in insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1997; 82: Ebeling P, Jansson P-A, Smith U, Lalli C, Bolli GB, Koivisto VA. Strategies toward improved control during insulin lispro therapy in IDDM. Diabetes Care 1997; 20: Mortensen HB, Hougaard P, Hvidøre Study Group. Comparison of metabolic control in a cross-sectional study of 2,873 children and adolescents with IDDM from 18 countries. Diabetes Care 1997; 20: Wolfsdorf JI, Laffel LM, Pasquarello C, Vernon A, Herskowitz RD. Split-mixed insulin regimen with human ultralente before supper and NPH (Isophane) before breakfast in children and adolescents with IDDM. Diabetes Care 1991; 14: Wolffenbuttel BHR, van Ouwerkerk BM, Veldhuyzen BFE, Geelhoed-Duijvestijn PHLM, Jakobsen G, van Doorn LG. Comparative effect of two different multiple injection regimens on blood glucose control and patient acceptance in type 1 diabetes. Diabetic Med 1990; 7: 695 9