The onset of type I diabetes is usually symptomatic, with symptoms lasting from a few days to weeks before the diagnosis is made. In rare cases, usually in older children symptoms may last for a few months. Polyphagia with weight loss, polydipsia, and polyuria are the most common symptoms. The latter may lead to enuresis. As insulin deficiency progresses, symptoms include anorexia, vomiting, abdominal pain, fatigue, and irritability. Occasionally, these symptoms may be confused with a viral infection, and the early diagnosis may be missed if the pediatrician does not have a high index of suspicion. Therefore, routine urine analysis with careful monitoring of glucose and ketones is warranted in a child with such complaints. At the time of diagnosis, the most obvious physical signs are dehydration and weight loss; the blood glucose levels are generally markedly elevated and ketonuria is present with or without acidosis.
Many children with insulin dependent diabetes mellitus (IDDM) have an amelioration of their disease soon after the initiation of insulin therapy as reflected by decreasing insulin requirements and amelioration of the hyperglycemia.1 This period, also called "honeymoon" or remission, lasts for a few days to several weeks. It is characterized by variable endogenous insulin secretion indirectly measured by C-peptide radioimmunoassay in serum or urine. C-peptide measurements are used as an index of beta cell function in diabetic patients receiving exogenous insulin. 2 C-peptide is secreted in equimolar amounts with insulin after cleavage of the proinsulin molecule. In the majority of patients, the remission phase is followed by a progression to total insulin deficiency two to five years after the diagnosis.3
Immediate medical attention is needed as soon as hyperglycemia or glucosuria are detected because of the potential for rapid deterioration. Initial therapy will depend on how early the diagnosis is made and on the state of the child.
Initial Treatment of the Nonketoacidotic Child
Initial therapy can be conducted in the hospital or on an ambulatory basis. Insulin treatment is usually initiated with a dose of 0.3 to 0.5 U/kg/day. The insulin dose is adjusted to keep the preprandial blood glucose between 90 and 120 mg/dL. The child is offered three meals and three snacks per day, and the dietary intake of each day is used as a guide to the caloric adjustment for the following day. In the hospitalized child, initially, six doses of regular insulin may be given every four hours (before breakfast, lunch, supper, and bedtime snack, at midnight, and at 4:00 AM) for one to two days. The sum of the six doses administered during the previous 24-hour period will give an approximation of the following day's total insulin requirements. As soon as the child is well and the insulin requirements are stable (one to two days), a mixture of intermediate-acting (isophane insulin suspension [NPHJ or lente) and short-acting (regular) insulin is given in one or two daily injections.
Alternatively, children who are clinically well at the time of diagnosis, may be started immediately on 0.3 to 0.5 U/kg/day of a mixture of NPH/lente and regular insulin. Extra doses of regular insulin are added every four hours for blood glucose levels above the target. The sum of all the extra doses required during the previous day will give an approximation of the following day's requirements. Daily increases of insulin (depending on the degree of hyperglycemia and ketosis) are in order until the blood glucose levels have approached the target set for that particular patient. Within a few days of diagnosis, children are usually asymptomatic, eating and drinking well, and gaining weight. The urine is free of acetone. However, the caloric and insulin requirements are usually at their peak since, in the initial phase of management, the child is repairing body tissues and replenishing nutritional stores.
During the first few days after diagnosis, blood glucose and urine ketones and glucose should be monitored every four hours to guide the insulin adjustments and to establish the renal threshold for glycosuria. Such testing is a useful teaching tool that helps the child and family learn the action duration of each insulin and the effects of activity and diet on glycémie control. Instruction on self-glucose monitoring technique at each testing is recommended so that the child and parents can compare the laboratory blood glucose results with their own measurements with reagent strips or reflectance meters.4 The education program usually takes place during the first 14 days after diagnosis. Education should encompass urine and self-blood glucose testing, insulin administration, diet planning, dietary and insulin adjustments, and facts about physiopathology of diabetes, insulin action, and insulin adjustments for exercise and illnesses. The interaction between staff and family allows the health team to assess the particular psychosocial needs of the patient and to detect potential social problems that may require early intervention on the part of the social worker or psychiatrist.
Although most clinics hospitalize diabetic children for initial stabilization and teaching, ambulatory treatment can be successful.5 Initiating treatment in the home environment is advantageous because it eliminates the stress of hospitalization, allows mastery of new knowledge at home, and helps the family to be less dependent on the hospital staff. However, if the Diabetic Center is not adequately staffed to provide daily home visits, hospitalization of the diabetic child is recommended.
EVALUATION OF THE KETOACIDOTIC CHILD
Initial Treatment of the Child with Diabetic Ketoacidosis (DKA)
Diabetic ketoacidosis (DKA) is a profound metabolic disturbance which results from insulin deficiency. Approximately 10% to 15% of newly diagnosed insulin dependent diabetic children present with ketoacidosis.
Insulin deficiency leads to hyperglycemia, increased lipolysis, ketogenesis, systemic acidosis, and depletion of intracellular and extracellular water and electrolytes that may result in coma and even death. Elevated levels of growth hormone, glucagon, Cortisol, and catecholamines also occur in DKA primarily in response to the stress of DKA or as a consequence of volume depletion and acidosis.6,7
DKA is diagnosed when the pH is less than 7.3 and the plasma bicarbonate is less than 15 mEq/L. DKA in the child is a life-threatening complication which requires careful supervision in an intensive care unit. Guidelines for clinical and laboratory assessment are shown in Table 1.
Rehydration should be started immediately with normal saline for the first 1 to 2 hours (since even the saline level is hypotonic with respect to the patient's osmolality). Guidelines for rehydration are given in Table 2. Initial rehydration therapy is replaced by one half normal saline with electrolytes (KCl and a mixture OfKHp 2PO4, Kp 2HPO4). Dextrose in a 10% solution is added as soon as the blood glucose levels decrease toward 250 mg/dL. Hyperosmolality should be corrected very slowly because too rapid a decline may predispose the child to cerebral edema, the major complication of DKA in children. Although cerebral edema is thought to be a complication of DKA treatment, subclinical brain swelling may be present already before treatment is initiated.8 Insulin replacement can be started one hour after the commencement of fluid therapy, with small doses of insulin through the intravenous route.9,10 The initial bolus of 0.1 U/kg intravenously is followed by a continuous infusion of 0. 1 U/kg/hr. This is titrated according to blood glucose levels measured hourly. It is advisable to start intravenous glucose in a 10% solution as soon as the blood glucose decreases toward 300 mg/dL to replenish the depleted glycogen stores. The rate of the intravenous infusion is then adjusted according to blood glucose levels. Potassium should be started as soon as the urine output is adequate because the total body potassium is usually depleted and insulin treatment and correction of acidosis will shift potassium back to the intracellular compartment and may produce hypokalemia. Providing phosphate may promote the formation of 2,3-diphosphoglycerate and shift the oxygen dissociation curve to the right releasing oxygen to the tissues and correcting acidosis.11 In addition, administration of potassium phosphate can reduce the inevitable chloride overload of the treatment of DKA, which may aggravate the acidosis.12 Metabolic acidosis of DKA is usually repaired through correction of dehydration and insulin replacement. The use of bicarbonate should be restricted to severe acidosis (pH 7.0 to 7.1) where myocardial function may be impaired. Use of bicarbonate may shift the oxygen dissociation curve to the left,13 accelerate the entry of potassium into the cells, and thus precipitate hypokalemia and worsen cerebral acidosis due to the rapid diffusion of CO2 through the blood brain barrier.14 When indicated, bicarbonate should be replaced slowly (over 1 to 2 hours) in doses not exceeding 1 to 2 mEq/kg. The insulin infusion is usually maintained until acidosis is corrected (usually 12 to 16 hours) and the patient is well enough to tolerate small amounts of oral fluids.
FLUID, ELECTROLYTE AND INSUUN INFUSION DURING DKA
Treatment During Remission Phase
Insulin requirements usually increase during the first week after diagnosis and decrease gradually after the second or third week. Progressive decrease of insulin requirements toward 0.5 U/kg or less indicates the beginning of the remission period during which some patients show almost complete restoration of beta cell function. Continuing to administer a minimum of 1 to 2 units of insulin can avoid either confusion over the diagnosis or the impression that diabetes could be transient. The remission period is usually short lived, lasting a few days to weeks in very young children. In older adolescents, remissions may last for a few months. As the beta cell mass becomes more depleted, insulin requirements increase toward 0. 7 to 1.0 U/kg/day.
Treatment of the Chronic Phase
The insulin regimen selected should be based on the individual characteristics of each patient. Thus, age and stage of sexual development, dietary intake, and potential adherence to the treatment should be taken into serious consideration. If the insulin requirements and caloric intake are relatively small, especially in the very young child or soon after diagnosis of IDDM, a single dose of a mixture of NPH and short-acting insulin given before breakfast, with a second dose of short-acting insulin before supper according to the tests, are usually enough. Older children approaching puberty or children who are in the midst of the adolescent growth spurt may best be started on a twice-daily injection regimen of a mixture of NPH and regular insulin before breakfast and before supper. However, an adolescent who is having difficulty accepting his or her condition may have to be treated with one daily injection instead of the more common twice-daily regimen until problems of acceptance are overcome. Some adolescents may have poor diabetic control because they tend to skip the second injection. The use of a single-dose regimen ensures they at least receive their necessary daily insulin requirement.
Careful balance of caloric intake, activity, and insulin dose are required for an insulin regimen to be successful. During the adolescent growth spurt, insulin requirements increase together with caloric requirements and may be as high as 1.5 U/kg/day. After full development has been achieved, diet and insulin dose should be decreased simultaneously to avoid overinsulinization and obesity.
The basic insulin dose is that required to maintain glycémie control with stable food intake and activity. This basic dose may vary during weekends, holidays, and illnesses. Blood glucose targets should also be based on the individual patient and should take into consideration the frequency of blood glucose or urine glucose testing, the patient's ability to recognize and react to hypoglycemia, and the limitations imposed on the patient and family's ability or willingness to comply. However, these considerations should not prevent continued efforts toward the goal of achieving "normal" glycémie regulation without causing hypoglycemia. Constancy in the timing of injections and meal intake are crucial for the success of any regimen. Insulin should be given at least 20 to 30 minutes before meals or earlier if the prevailing blood glucose levels are well above the desired target.
Education of the child and family must stress the importance of urine glucose and ketone, and blood glucose monitoring as the most important short-term guidelines for insulin dose and caloric adjustments. These tests should be performed before each main meal and bedtime, and the results should be charted in a logbook so that they can be discussed with the pediatrician or clinic staff for insulin adjustments. Urine glucose can be monitored with the 2 -drop Clinitest® method in second voided specimens or with sensitive methods that use reagent strips. Urine ketones should be tested at least once daily, preferably in the morning, but tests may be conducted more often if ¿Sere is an intercurrent illness or stressful event. The disadvantages of urine glucose testing are that:
* the renal threshold may vary from day to day in the same patient,
* the presence of glycosuria does not give a clear idea of the prevailing blood glucose at the time of the test,
* negative tests may give false sense of good control in patients with high renal thresholds, and
* negative tests do not indicate how low the prevailing blood glucose might be.
The development of techniques for self-blood glucose monitoring (SBGM) has permitted the documentation of glycémie levels at home and thus provided a safe guideline for insulin adjustments.15 The current belief is that:
* SBGM is feasible, practical, and acceptable to patients,
* blood glucose measurements are sufficiently accurate for clinical use,
* SBGM facilitates the understanding of diabetes and insulin adjustment, and
* SBGM can improve glycémie control if it is used as part of a treatment plan involving careful balance of diet, activity, and appropriate insulin adjustments.
Data on SBGM in children seem to suggest that although the incidence of hospital admissions for acute complications such as ketoacidosis, gastroenteritis, and hypoglycemia may decrease with the use of SBGM, long-term glycémie control may not be significantly ameliorated. The reasons for this failure are complex and vary from patient to patient, but are usually related to poor compliance due to the fear or pain caused by the finger punctures or the lack of discipline required for regular testing. With those who can perform the tests adequately, the families may be unable to use the information derived from testing to make the appropriate insulin adjustments or are afraid of introducing changes in insulin dosages. For many, the problem is global. Because of the complexity of type I diabetes therapy, it is crucial to adapt the treatment to the patient's individual needs. Thus, in a patient who is reluctant to do SBGM, it is better to do only urine glucose testing than no testing at all. Monitoring is recommended two to three times per day, before breakfast, supper, and bedtime. For those patients who are willing, SBGM is preferred in conjunction with a morning urine test for ketones. If patients or their families are unwilling to perform SBGM, urine glucose and ketone monitoring are recommended (these can now be done with the same strip) before main meals, with blood glucose monitoring at bedtime (if urine glucose is negative) to decrease the risk of nocturnal hypoglycemia.
In clinical practice, estimation of glycosylated hemoglobin levels provides an independent and objective assessment of the diabetic patients long-term glucose control and, if performed repeatedly over time, facilitates comparisons between treatment strategies and may improve motivation by providing the patient with a target for average blood glucose control.
INSULIN DOSE ADJUSTMENTS FOR ONCE DAILY INSUUN REGIMEN
Separation of HbA with cation exchange chromatography resolves four minor components (hemoglobin A1a1, A^sub 1a2^, A^sub 1b^, and A^sub 1c^) which are collectively referred to as hemoglobin Aj.16 The hemoglobin Aj fraction is formed continuously within the red cell throughout its 120-day life span as a result of post-translational nonenzymatic modification of the hemoglobin A. HbAIc, the most abundant of these minor components, is increased in patients with diabetes mellitus. Thus, glycosylated hemoglobin reflects the average blood glucose level to which the erythrocyte has been exposed during its life span.
Glycosylated hemoglobin can be measured with chromatographic methodologies which operate on the basis of charge (eg, column chromatography, isoelectric focusing, high pressure liquid chromatography, and the commercial microcolumn kits, all of which are sensitive to changes in buffer, pH, and temperature). Chemical measurements of glycosylated hemoglobin have the advantage of being readily amenable to standardization and long-term quality control procedures because they are not influenced by the variables which may affect the chromatographic procedures.
Adjustments of Insulin Dosage
Changes in insulin dose are made on the basis of recorded blood or urine glucose tests and are aimed at target values. Adjustments should be made with great care to avoid overinsulinization and only when values cannot be explained by changes in food intake or activity. Increments of insulin dosage should not surpass 10% of the total daily dose and should be made only after two or three consecutive days of high blood or urine glucose tests. Decrements of insulin dosage in response to hypoglycemia not explained by increased activity or missed meals should be done immediately and should be at least 15% of the relevant insulin dose. These maneuvers are repeated until the blood or urine glucose values are within the target range. The first adjustment should be made in the overnight insulin action. This is best made on the basis of 3 am to 4 am (targeted at above 70 mg/dL) and fasting blood glucose values (targeted at 80 to 140 mg/dL). Once the morning blood glucose is within the target range, adjustment of the premeal short-acting insulin doses, usually adjusted on the basis of the next preprandial blood or urine glucose test, can be made. If increments in the prebreakfast short-acting insulin result in blood or urine glucose tests within the target range, but with midmoming hypoglycemia, the patient should increase the caloric content of the breakfast or the midmoming snack. If increments in the NPH or lente insulin necessary to achieve fasting tests within target, result in nighttime hypoglycemia, the patient should increase the bedtime snack or take an extra snack to achieve bedtime blood glucose values below 120 to 140 mg/dL. Dietary manipulations are usually necessary in patients taking one-dose regimens where insulin dosage changes can be implemented only once a day.
The single-dose regimen uses intermediate (NPH or lente) or long-acting (ultralente) insulins alone or in combination with regular insulin. Morning injections of intermediate-acting insulins provide higher plasma insulin levels during the day and can be used alone if there is enough endogenous insulin secretion during meals. Frequently, however, endogenous insulin secretion is inadequate, and regular insulin needs to be added to correct morning hyperglycemia. This singledose regimen is usually effective only during the first few months after diagnosis in combination with a supplementary dose of short-acting insulin before supper if glucose tests are above the target. Ultralente insulins, which are absorbed very slowly from UHe injection site, may be less convenient to use, as necessary decreases in dosage cannot be made with short anticipation. Because of its relatively slight peaked action, ultralente insulins provide a background basal dosage of insulin so that premeal injections of regular insulin are needed to avoid postprandial hyperglycemia.
INSUUN DOSE ADJUSTMENTS FOR 'SPUTAND MPtED" INSUUN REGIMEN
Short-acting insulin has major action between breakfast and early afternoon and its effect is reflected in the noon and presupper glucose tests. The NPH insulin has major action between supper and nighttime and its effect is reflected in the bedtime and fasting glucose values. Guidelines for adjustment of basic and temporary insulin doses for this regimen are outlined in Table 3.
The most frequent treatment regimen uses two daily mixtures of intermediate and short-acting insulin.17,18 The morning short-acting insulin has major action between breakfast and lunch and the intermediateacting neutral protamine (NPH) or lente has major action between breakfast and supper. The evening short-acting insulin has major action between supper and bedtime and its effect is reflected in the bedtime tests. The evening NPH or lente insulin has major action overnight and its effect is reflected in the blood glucose level on arising the next morning. The theoretical advantages of this regimen are the reduction of basal and postprandial hyperglycemia, and the reduction of overnight and fasting glycemia. However, in some patients, attempts to achieve morning normoglycemia result in nocturnal hypoglycemia and early morning hyperglycemia. Guidelines for adjusting insulin dose with this regimen are shown in Table 4.
Evidence suggesting that the microvascular and neurologic complications of diabetes develop as a consequence of hyperglycemia has led to increased efforts to achieve levels of glycemia as close as possible to physiologic.19 Although conventional insulin replacement controls symptoms and restores normal growth and development in diabetic patients, blood glucose levels show wide fluctuations during the day and from day to day. One of the reasons for this phenomenon is that exogenous insulin availability is not related to circulating fuels or prevailing glycemia. Rather, it is dependent on the pharmacokinetics of a given insulin preparation and its absorption characteristics. The main factors that influence the effects of insulin preparations are:
* absorption which depends on site of injection, depth, exercise, and concentration;
* the inherent variability in rates of absorption and duration of effects in individual patients; and
* the insulin antibodies which bind injected insulin for variable periods altering its onset and duration of action.
When devising an insulin regimen that attempts to mimic physiologic insulin availability, it is important to distinguish between insulin availability to overcome postprandial hyperglycemia and that which is needed to normalize the fasting glucose level. It is also important to be aware of the usual action duration of each of the insulin preparations. In these cases, the delay of die presupper NPH or lente injection to bedtime can alleviate the problem. Thus, the peak effect of insulin will coincide with die waking period at which time the patient can test the blood glucose.
An alternative regimen is the administration of regular insulin before breakfast and lunch with a mixture of NPH or lente and regular insulin before supper. If fasting normoglycemia cannot be achieved without nocturnal hypoglycemia, the evening dose can be split so that regular insulin alone is given before supper and NPH or lente is given at bedtime (Figure 2). Studies have shown that treatment with multiple injections of regular insulin before each meal and NPH insulin at bedtime can be just as effective in normalizing blood glucose levels as the use of continuous subcutaneous insulin infusion (CSII). A popular regimen uses premeal doses of regular insulin plus long-acting insulin (ultralente) in the morning or three daily injections of a mixture of ultralente and regular insulin in the morning and before supper, and regular insulin alone before lunch.
Mechanical Devices for Continuous Insulin Administration
Devices for continuous insulin administration are used to normalize blood glucose levels throughout the day. Since insulin delivery is continuous, it can more or less mimic normal insulin secretion. Physiologic insulin secretion in nondiabetic individuals involves meal-related increased insulin secretion (initiated by neural and gut factors prior to the hyperglycemic stimulus) responsible for tissue uptake and storage of nutrients followed by a rapid return of insulin secretion to baseline; and basal insulin secretion between meals and during the night to regulate amino acids and fatty acids in the fasting state, thus preventing excessive nocturnal gluconeogenesis. Ideally, only an artificial beta cell capable of continuous blood glucose level monitoring with administration of insulin into the portal circulation can approximate the function of the normal pancreas. However, this system cannot be miniaturized due to the limitations encountered in the development of a glucose sensor.
Open-loop systems have been developed which can use the intravenous,20 peritoneal,21 or subcutaneous route22'24 of insulin administration. These are relatively small and lightweight portable devices (insulin pumps) which can be implanted subcutaneously or carried extracorporeally. The most popular and widespread method is the continuous subcutaneous insulin infusion with a portable extracorporeal battery-driven pump. There are now many types of devices with alarms for low battery, pump runaway, and empty reservoir; variable basal rates; and preprogrammed boluses. Although there is no doubt that this treatment is effective in improving glucose control and many of the metabolic derangements of diabetic patients, it remains to be seen whether their long-term use will prevent or delay the onset of late diabetic complications. Furthermore, their long-term use may be limited for reasons of cost and need for increased medical supervision. In addition, because they require a high degree of patient motivation and cooperation, these treatments should be restricted to a few selected patients. The complications resulting from this therapy may be quite serious. The most severe complication is hypoglycemia, especially when it occurs during sleep and the patient may be unaware of the low blood glucose values. The continued administration of insulin when the patient is unaware of hypoglycemic symptoms could result in prolonged hypoglycemia and even death. Strategies to reduce the risk of nocturnal hypoglycemia include increasing the target fasting blood glucose to 120 to 130 mg/dL, decreasing the basal rate if 3 am blood glucose levels are below 80 mg/ dL, and daily determination of the blood glucose at bedtime followed by an extra snack if blood glucose values are less than or equal to 130 mg/dL.
Adjustments for Exercise
Since exercise is a normal component of everyday life, it should be encouraged in every diabetic child. In anticipation of exercise, however, it is necessary to increase caloric intake or decrease the insulin dose to avoid hypoglycemic reactions during or after exercise. Hypoglycemia under these circumstances results from increased absorption of insulin from the injection site as a consequence of increased blood flow. Injections made in areas that are least likely be exercised may decrease the risk of hypoglycemia; however, because exercise usually increases blood flow throughout the body, this may not be a helpful approach. Studies show that the intake of one carbohydrate exchange per every 30 minutes of moderate exercise may prevent hypoglycemia. For exercise anticipated to last longer than one hour, a decrease in the insulin dosage according to the intensity and duration of exercise may be effective in reducing the risk of hypoglycemia. In practice, for children on one daily dose, decreasing the dose of regular insulin if exercise is done within three hours of breakfast and decreasing the intermediate-acting insulin if exercise is done in the afternoon or later are effective precautions. If the patient is taking two daily doses, the following recommendations may be made: decreasing the regular insulin if exercise is done within three hours of breakfast or lunch; decreasing the morning NPH for exercise occurring in the late morning or early afternoon; and decreasing the evening NPH in anticipation of exercise occurring in the late afternoon or after supper (Table 5).
Management During Acute Illnesses or Stress
Insulin requirements usually increase as a result of increased secretion of counterregulatory hormones and decreased activity even in the face of reduced caloric intake or vomiting. Blood glucose and urine ketone tests should be done every four hours and the physician should be contacted immediately for advice Table 6). The following guidelines are useful for managing a diabetic child during an illness. These guidelines are based on whether the patient is able to take food or liquids by mouth.
ADJUSTMENTS OF INSULIN DOSE FOR ANTICIPATED EXERCISE
An illness not accompanied by nausea or vomiting (eg, common colds, flu, hay fever, pneumonia, emotional stress, measles, accidents or trauma requiring prolonged bed rest). If activity is normal, give the usual dose of NPH/lente plus extra regular insulin as per blood glucose or urine tests, ie, 20% of the morning dose of regular every four hours if the blood glucose is above 240 mg/dL and 30% if blood glucose is above 400 mg/dL. If urine ketones are present in moderate to large amounts 10% may be added.
If activity is reduced and the patient is confined to bed, the diet should be reduced by approximately one third. The reduction in caloric intake compensates for the inactivity thus simplifying insulin adjustment. The insulin adjustment is the same as for an illness without bed rest.
MANAGEMENT OF A DIABETIC CHILD DURING AN INTERCURRENT ILLNESS
MANAGEMENT FOR ELECTIVE SURGERY
An illness accompanied by nausea, vomiting, or marked anorexia. It is not advisable to give the usual dose of insulin in the morning when there is uncertainty about the patients ability to eat. The insulin dose must not, however, be omitted since this could lead to diabetic ketoacidosis. It is better to omit the NPH or lente insulin and replace it with injections of regular insulin every four hours according to the blood glucose and the presence or absence of ketones in the urine. Regular insulin is given at each of the testing times according to a sliding scale as follows: one fourth of the total daily insulin dose for those patients on one daily injection or one fourth of the total morning dose for those on two daily injections for blood glucose levels above 240 mg/dL with an additional 10% of the total daily dose if urine ketones are moderate or large. The dose of insulin should be reduced if the blood glucose levels are lower. For a child who usually takes 10 units of short-acting and 30 units of NPH or lente, 10 units for blood glucoses above 240 mg/dL, 8 units for blood glucoses between 180 and 240 mg/dL, 6 units for blood glucoses between 120 and 180 mg/dL, and 4 units for blood glucoses below 120 mg/dL. Add 4 units to the calculated dose if urine ketones are moderate to large. This is repeated at each testing until the child is able to eat or tolerate fluids well. If the child is well by noon, two thirds of the usual total daily dose of NPH or lente is given together with regular insulin (one half of the morning dose of NPH or lente for those on two daily injections). If the child is still unwell by supper time, regular insulin alone should be continued. At bedtime, one fourth of the usual daily dose of NPH or lente should be given with a small dose of regular insulin. The diet should be replaced by regular soft drinks, fruit juice, or sweetened tea, 2 to 4 ounces every hour. As the child improves, light foods should be given as tolerated.
If vomiting occurs after the administration of the usual morning dose of insulin, sips of sugar-containing fluids should be given every 20 to 30 minutes. Blood glucose and ketones should be done every hour to adjust the amount of oral fluids. Blood glucoses should be maintained between 120 and 200 mg/dL. If vomiting persists and the blood glucose falls below 100 mg/dL, the child should be taken to the hospital for intravenous therapy. A subcutaneous injection of glucagon should be given at home before departing if the child lives at some distance from the hospital.
If the blood glucose is above 240 mg/dL and the urine contains moderate to large amounts of ketones, the physician should be advised immediately because this could reflect diabetic ketoacidosis.
Management During Surgery
Perioperative management of the diabetic child is aimed at maintaining normal metabolism by providing sufficient insulin and fluids to prevent a severe catabolic state. In the insulin-dependent diabetic patient, the intense catabolism seen normally during surgery compounds the effects of insulin dependency and, if insulinization is inadequate, the catabolic changes may be greater than those found in the nondiabetic patient. These changes could result in severe metabolic decompensation and even ketoacidosis. Using the regimen outlined in Tables 7 and 8, perioperative care is fairly flexible. Insulin and dextrose may be combined in the same vein to minimize the dangers of hypoglycemia and hyperglycemia due to changes in die glucose infusion rate. The insulin is added to the fluids in the Soluset of the infusion bag and changed hourly or as required. Some clinicians suggest the use of insulin infusion by pump through a separate vein, which is advantageous in that the insulin rate can be varied independently. However, changes in die infusion rate of either IV may lead to hyperglycemia or hypoglycemia. If surgery is anticipated, it is best to admit the patient the day before. Insulin infusions are started the night before surgery whenever possible. Alternatively, infusions can be started at 7 am with surgery scheduled after 10 am. Regular insulin should be given every four hours starting at supper until the IV insulin is started. Allow two to three hours before surgery to adjust the insulin infusion rate.
INSULIN-GLUCOSE POTASSIUM INFUSION REGIMEN
Another alternative is to give one third to one half of the morning NPH/lente dose the morning of surgery together with a 5% dextrose solution or short-acting insulin every four hours depending on blood glucose tests. The potential problems with these methods are that the preoperative hospitalization period is usually not long enough to permit adequate estimation of the insulin requirements in patients who may have been in poor control and that insulin absorption is not always predictable.
The nutritional requirements of a diabetic child are similar to those of nondtabetic children of the same age and stage of development with the provision that complex carbohydrates are used rather than simple sugars. The total caloric intake is usually divided in the following manner: three tenths for breakfast and midmoming snack, three tenths for lunch, and four tenths for supper and bedtime snack. These calories should comprise 45% to 50% carbohydrate, 30% fat, and 20% protein. During periods of inactivity (trauma, surgery, intercurrent illnesses), the caloric intake may be reduced by approximately one third because of decreased energy expenditure and increased secretion of counterregulatory hormones. The caloric intake and composition should be modified on an ongoing basis to satisfy the child's appetite and taste, and to compensate for growth and pubertal development and changes in activity pattern. Occasional excesses should be permissible so as to minimize rebellion and nonadherence. The American Diabetes Association and the Canadian Diabetes Association have prepared simplified methods based on six basic food groups with a wide variety of foods that can be exchanged or substituted within or between groups. If the child is unable to eat because of an illness, the caloric requirements should be replaced with fluid equivalents. Thus, with the help of a nutritionist, the diabetic child can select a diet based on personal taste and ethnic preferences, which can be translated into exchanges to meet his or her demands.
COMPLICATIONS OF INSULIN THERAPY
In established diabetes, DKA may be precipitated by factors that interfere with insulin action such as the release of hormones during physical (infection, trauma) or emotional stress. Such factors may result in a failure to take extra insulin or psychological disturbances leading to omission of the insulin dose and lack of insulin adjustment because of incorrectly reported tests.
Hypoglycemia is probably the most frequent complication of insulin therapy. The resulting symptoms are secondary to adrenergic activation and neuroglycopenia. The most common causes of hypoglycemia include skipping or delaying meals, excess physical activity without simultaneous increase of food, errors in the insulin dosage, or dramatic reductions of insulin requirements during the honeymoon period. Severe repeated hypoglycemic episodes after the remission phase may result from deliberate errors in the insulin dose and may indicate psychological disturbances that call for immediate psychosocial assessment. Educating the family, child, teachers, and others involved in the care of these children is of paramount importance in the prevention and treatment of hypoglycemia. The possibility of neurological sequelae resulting from severe hypoglycemia is of great concern. However, in clinical practice, catastrophic complications resulting from severe hypoglycemia in patients treated with insulin injections are rare.
Mild and moderate hypoglycemia are usually corrected by the administration of 10 to 25 g of carbohydrate snacks or beverages. In the case of loss of consciousness or seizures, subcutaneous glucagon should be administered. Although it is not certain whether the use of self-glucose monitoring helps to reduce the incidence of hypoglycemias resulting from dose errors and delayed or missed meals, blood glucose testing at bedtime and at 3 am during periods of insulin adjustment may help to recognize and avoid nocturnal hypoglycemia, a problem more common than is generally appreciated. In some children, overinsulinization results in hypoglycemia and rebound hyperglycemia due to excessive counterregulation, the so-called Somogyi phenomenon. This phenomenon, which is characterized by headaches, nightmares, night sweats, and abdominal pain, is accompanied by ketonuria and morning hyperglycemia or glucosuria. It is usually corrected by a decrease in the insulin dose. The "dawn phenomenon," is an early morning blood glucose rise not preceded by nocturnal hypoglycemia seen in diabetic individuals. It is due to diurnal variation in sensitivity to insulin and is usually corrected by delaying the administration of NPH/lente from supper to bedtime.
In the last decade, improvements in the method of insulin purification have resulted in "single peak," "monocomponent," or purified insulins which contain less than 25 ppm of proinsulin contamination. Changes from conventional insulins to highly purified insulins appear to produce less insulin allergy and lipoatrophy, and reduction in levels of insulin antibodies. Although it has been suggested that diabetic complications may be related to high levels of insulin antibodies, there is still not enough evidence to suggest that the use of these insulins will reduce the diabetic complications.
In addition to purified insulins, human insulin may be produced by enzymatic replacing the alanine from the terminal position 30 in the B chain of purified pork insulin with threonine. The so-called "semisynthetic human insulin" behaves like pure pork insulin. Human insulin can be synthesized by inserting human gene sequences into plasmids of Escherichia coli to form A and B chains with two fermentations, which are subsequently linked by chemical means. An alternate method involving one fermentation produces proinsulin which is later used to provide human insulin and C-peptide. This biosynthetic insulin, chemically and biologically similar to pancreatic insulin, is free from pancreatic peptides and bacterial contaminants. Its faster absorption and shorter duration than the purified pork insulin may be a disadvantage. However, it may be useful in patients who present insulin allergy, insulin resistance, and lipoatrophy not controlled by pure pork insulin.
1. Park BN, SoeldnerJS, GIeasonRE: Diabetes in remission. Insulin secretory dynamics. Diabetes 1974; 23:616-623.
2. Block MB, Rosenfield RL, Mako ME, et al: Sequential changes in beta-cell (unction in insulin treated diabetic patients assessed by C-peptide immunoreactivity. N Engl I Mat 1973; 286:1144-1148.
3. Madison IX: Low-dose insulin: A plea for caution. N Engl ] Med 1976; 294:393-394.
4. Bleicher SJ (ed): Symposium on home blood glucose monitoring. Diabetes Care I960; 3:57-186.
5. Galatzer A, Amir S, Gil R1 et al: Crisis intervention program in newly diagnosed diabetic children. Diabetes Caie 1982; 5:414-419.
6. Exister D, McGarry D: The metabolic derangements and treatment of diabetic ketoacidosis. N Engl J Med 1983; 309:159-169.
7- Sperling M: Diabetic ketoacidosis. Pedum Clin North Am 1984; 31:591-610.
8. Krane E), RocVoff MA, Wallman )K, er a): Subclinical brain swelling in children during treatment of diabetic ketoacidosis. N Engl J Med 1985; 312:1147-1151.
9. Morris LR, Kitabchi AE: Efficacy of low-dose insulin therapy for severely obtunded patients in diabetic ketoacidosis. Diabetes Cam 1980; 3:53-56
10. Alberti KGMM, Hockaday TDR: Diabetic coma: A reappraisal after five years. CIm Endocrinol Metab 1977: 6:421-455.
11. Keller V, Berger W: Prevention of hypophosphatemia by phosphate infusion during treatment of diabetic ketoacidosis and hyperosmolar cnma. Diabetes 1980; 29:87-95.
12. Duck SC, Weldon W, Pagliari AS: Cerebral edema complicating therapy for ketoacidosis. Diabetes, 1976; 25:111-115.
13. Munk P, Freedman MH, Levison H1 et al: Effect of bicarbonate on oxygen transport in juvenile diabetic keotacidosis- J Pedina 1974; 84:510-514.
14. Kaye R: Diabetic ketoacidosis: The bicarbonate controversy. } Pedia» 1975; 87.156-159.
15. Skyter JS, Lasky IA, Skyler DL, et al: Home blood glucose monitoring as an aid in diabetes management. Diabetes Cae 1978; 1:150-157.
16. National Diabetes Data Group: Report of the expert committee on glucosylated hemoglobin. Diabetes Care 1984; 7:602-606.
17. Skyler JS, Skyler DL, Seigler DE, et al: Algorithms for adjustment of insulin dosage by patients who monitor blood glucose. Diabetes Care 1981; 4:311-318
18. Skyler JS: Self-monitoring of blood glucose. Med CIm NortA Am 1982; 66:1227-1250.
19. Schiffrin A: Treatment of insulin-dependent diabetes with multiple subcutaneous insulin injections. Med CIm NortA Am 1982; 66:1251-1267.
20. Pfeiffer EF, Thurn C, Clemens AH: The artificial beta cell: A continuous control of blood sugar by external regulation of insulin infusion (glucose controlled insulin infusion system). Horn« Metab Res 1974; 6:339-342.
2 1 . Santiago JV, Clemens AH, Clarke WL, et al: Closed-loop and open-loop devices for blood glucose control in normal and diabetic subjects. Diabetes 1979; 28:71-84
22. Champion MD, Shepard GA, Rodger NW1 et al: Continuous subcutaneous infusion of insulin in the management of diabetes mellitus. Diabetes 1980; 29:206-212.
23. Schiffrin A1 Belmonte MM: Comparison between continuous subcutaneous insulin infusion and multiple injections of insulin. A one year prospective study. Diabetes 1982; 31:255-264.
24. Tamborlane WV, Sherwin RS, Genel M, et al: Reduction to normal of plasma glucose in juvenile diabetes by subcutaneous administration of insulin with a portable infusion pump. N Engl; Med 1979; 300:573-578.
EVALUATION OF THE KETOACIDOTIC CHILD
FLUID, ELECTROLYTE AND INSUUN INFUSION DURING DKA
INSULIN DOSE ADJUSTMENTS FOR ONCE DAILY INSUUN REGIMEN
INSUUN DOSE ADJUSTMENTS FOR 'SPUTAND MPtED" INSUUN REGIMEN
ADJUSTMENTS OF INSULIN DOSE FOR ANTICIPATED EXERCISE
MANAGEMENT OF A DIABETIC CHILD DURING AN INTERCURRENT ILLNESS
MANAGEMENT FOR ELECTIVE SURGERY
INSULIN-GLUCOSE POTASSIUM INFUSION REGIMEN