Pediatric Annals

DIABETES IN CHILDREN 

The Treatment and Prevention of Diabetic Ketoacidosis in Children and Adolescents With Type I Diabetes Mellitus

Francine Ratner Kaufman, MD; Mary Halvorson, RN, MSN, CDE

Abstract

Type I diabetes mellitus is almost universally due to the autoimmune destruction of the pancreatic ß-cell mass. It is estimated to affect more than 125,000 children and youth in the United States, with approximately 13,000 new cases each year.1,2 Diabetic ketoacidosis (DKA) is, as it always has been, a common occurrence and major source of morbidity and mortality. DKA is present in approximately 25% to 40% of patients with diabetes at the time they first present.3'4 In established patients, DKA recurrence is approximately 0.2 events per patient per year.4 It also leads to costly hospitalizations, estimated to have exceeded $2.5 billion in 1996.5 During DKA and its treatment, patients are at risk of having fatal or near-fatal cerebral edema. This may occur in as many as 2% to 3% of children with this metabolic disturbance.6·7 Overall, DKA accounts for 20% of childhood diabetes mortality.5 Despite the fact that DKA is so frequent, there are still many controversial aspects surrounding its treatment and prevention. This article discusses these controversies to help the clinician manage and prevent this grave complication of diabetes.

THE TREATMENT OF DKA IN THE EMERGENCY ROOM

The initial treatment of DKA usually begins in the emergency room after the diagnosis of diabetes associated with DKA has been made. This diagnosis should be considered in a child of any age who presents with polyuria, polydipsia, and vomiting and /or is dehydrated with hyperpnea or an altered level of consciousness. The child with an altered level of consciousness requires a bedside fingerstick blood glucose determination immediately.8,9 Established patients may also have an emergency room referral to assess metabolic decompensation, determine how and why DKA developed, and begin treatment as soon as possible.

The emergency room treatment of DKA in our center is outlined in Table 1. Such protocols are the key to managing diabetes and should be immediately available in every emergency center and most wards and offices or clinics. A pediatric DKA emergency room protocol is essential because treatment protocols for adults advocate a much more rapid and aggressive reversal of DKA than is advised for children.10,11

Table

In addition, as shown in Table 5, we have decreased episodes of recurrent DKA in difficult to manage cases with multiple episodes of severe DKA with two distinct approaches: insulin pump therapy29 and out-of-home placement when the problem is a dysfunctional family.27,30 The latter studies showed that the likely explanation for recurrent DKA in these difficult to manage cases was inadequate home support.

SUMMARY

There remain a number of important controversies in the management of pediatric DKA. From the sodium content of the hydrational fluid to the rate of fluid administration that is best able to reverse the hyperosmolar dehydration attendant with DKA with minimal morbidity and mortality, there is no universal agreement on how patients with this devastating metabolic disturbance should be treated. It is still unclear what promotes or protects patients from neurologic insult during DKA reversal.

It is appropriate to begin to develop a national approach to eradicating DKA. This would require widespread public and professional education programs aimed at detecting new-onset type I patients prior to the onset of DKA. It would involve promoting diabetes screening programs aimed at detecting patients before the onset of symptomatic disease, and these would most appropriately be centered in the pediatrician's office. In the known patient, DKA still occurs as the result of intercurrent illness and nonadherence to the diabetes regimen due to patient or family chaos and dysfunction. Clearly, more strategies are needed to address these psychological and family patterns and the fact that many tenuous families have insufficient access…

Type I diabetes mellitus is almost universally due to the autoimmune destruction of the pancreatic ß-cell mass. It is estimated to affect more than 125,000 children and youth in the United States, with approximately 13,000 new cases each year.1,2 Diabetic ketoacidosis (DKA) is, as it always has been, a common occurrence and major source of morbidity and mortality. DKA is present in approximately 25% to 40% of patients with diabetes at the time they first present.3'4 In established patients, DKA recurrence is approximately 0.2 events per patient per year.4 It also leads to costly hospitalizations, estimated to have exceeded $2.5 billion in 1996.5 During DKA and its treatment, patients are at risk of having fatal or near-fatal cerebral edema. This may occur in as many as 2% to 3% of children with this metabolic disturbance.6·7 Overall, DKA accounts for 20% of childhood diabetes mortality.5 Despite the fact that DKA is so frequent, there are still many controversial aspects surrounding its treatment and prevention. This article discusses these controversies to help the clinician manage and prevent this grave complication of diabetes.

THE TREATMENT OF DKA IN THE EMERGENCY ROOM

The initial treatment of DKA usually begins in the emergency room after the diagnosis of diabetes associated with DKA has been made. This diagnosis should be considered in a child of any age who presents with polyuria, polydipsia, and vomiting and /or is dehydrated with hyperpnea or an altered level of consciousness. The child with an altered level of consciousness requires a bedside fingerstick blood glucose determination immediately.8,9 Established patients may also have an emergency room referral to assess metabolic decompensation, determine how and why DKA developed, and begin treatment as soon as possible.

The emergency room treatment of DKA in our center is outlined in Table 1. Such protocols are the key to managing diabetes and should be immediately available in every emergency center and most wards and offices or clinics. A pediatric DKA emergency room protocol is essential because treatment protocols for adults advocate a much more rapid and aggressive reversal of DKA than is advised for children.10,11

Table

TABLE 1Emergency Department Management of Pediatric Patients In Diabetic Ketoacidosis

TABLE 1

Emergency Department Management of Pediatric Patients In Diabetic Ketoacidosis

The treatment of DKA begins with isotonic saline to start to gradually correct the dehydration and electrolyte disturbances.12,13 However, what is meant by gradual correction remains somewhat controversial because it is assumed, although not proven, that how this initial fluid given might impact the risk of cerebral edema, Although most pediatric endocrinologists agree that adult emergency room texts recommending that liters of fluid be given rapidly11 are inappropriate, there is still debate as to how to deliver a fluid bolus in children. Recommendations vary from 10 to 20 mL/kg of isotonic saline during the first hour followed by repeat boluses versus no additional fluid unless the patient remains in hypovolemic shock.14

The treating physician should aim to stabilize and promptly transfer the child to an appropriate inpatient unit, likely an intensive care unit, so that he or she will be under the care of a pediatrician, a pediatric endocrinologist, or a pediatric intensivist. Our center aims to transfer the patient out of the emergency room within 1 to 2 hours. However, if the patient is to be detained longer than 2 to 3 hours, insulin by intravenous continuous drip with potassium supplementation will need to be initiated in the emergency room.

Table

TABLE 2Guidelines for the Care of Patients With Diabetic Ketoacidosis In the Intensive Care Unit or the Specialized Unit

TABLE 2

Guidelines for the Care of Patients With Diabetic Ketoacidosis In the Intensive Care Unit or the Specialized Unit

TRANSFER TO THE SPECIALIZED UNIT FOR CONTINUING CARE OF THE PATIENT IN DRA

Once the patient is transferred to a pediatric intensive care unit or other specialized unit with experience in the management of pediatric DKA, attention turns to gradually correcting the remaining dehydration and bringing the blood glucose level to near-normal range as shown in Table 2. Insulin treatment is given by low-dose infusion of short-acting (regular) insulin at the dosages shown.15 Hourly bedside blood glucose determinations help achieve the goal of lowering glucose by approximately 50 to 100 mg/dL/h. As glucose levels start to fall, 5% to 10% dextrose is added to the infusion to avoid rapidly lowering the glucose level and to help stabilize blood glucose between 150 and 200 mg/dL within the first 12 to 24 hours. Maximal substrate (10% dextrose) and intravenous insulin will reverse ketogenesis, halt hepatic glucose production, and facilitate peripheral glucose uptake.16

Patients with DKA have total body potassium depletion. This can be corrected by infusing potassium chloride at 3 mEq/kg/d, started after the patient is transferred. Potassium levels drop during the ensuing 12 to 24 hours concomitant with correction of the acidosis and as potassium enters the cells. Potassium infusion often needs to be increased to maintain serum potassium above 3.5 mEq/L. In severe hypokalemia, a cardiac monitor should be used to assess for the development of U waves and arrhythmia. Despite hypokalemia, little morbidity other than hypoventilation has been reported in pediatric patients.17 If the patient requires more than 4 to 5 mEq/kg, half of the infusion can be given as potassium chloride and half as potassium phosphate. Otherwise, there is little need to use potassium phosphate. Hypocalcemia may develop if too much phosphate is infused, and the hypophosphatemia of DKA will correct once the patient starts to eat.

There is no controversy concerning the use of bolus bicarbonate in DKA.10 It is contraindicated. Hydration and insulin therapy alone will correct the acidosis; bolus bicarbonate places the patient at risk for paradoxical central nervous system acidosis and hyperosmolality. If the pH is less than 7.0 to 7.1, sodium bicarbonate may be given as a slow infusion until the pH rises above 7.1 (Table 2).

RISK FACTORS FOR CEREBRAL EDEMA

Cerebral edema is the gravest complication of treating DKA.18,19 Although mild cerebral edema is probably common, as detected with central imaging studies, clinically relevant and potentially life-threatening cerebral edema rarely occurs. Onset is usually 6 to 12 hours after the initiation of treatment, and is abrupt and without warning or heralded by headache, lethargy, and incontinence. Then seizures, pupillary changes, decreasing heart rate, and increasing blood pressure suggest intracranial hypertension. Risk factors include young age and the initial presentation of diabetes.

What components of the initial treatment of DKA place patients at risk for cerebral edema remains controversial12 because of méthodologie problems.20 Some studies have suggested that it is the rate of change in serum osmolality, sodium, and /or glucose that is important.20"22 Others have implied it is the rate of fluid administration.23 A large retrospective analysis of 69 patients with DKA-associated cerebral edema could find no such associations.20 These controversies have resulted in the development of many DKA treatment protocols.

Table

TABLE 3Tips for the Pediatrician in Determining Whether Diabetes Is Present

TABLE 3

Tips for the Pediatrician in Determining Whether Diabetes Is Present

FLUID REGIMENS TO REDUCE THE RISK OF CEREBRAL EDEMA

In 1996, Rosenbloom and Hanas detailed the traditional protocol employed by most pediatric endocrinologists21 and this was followed by an alternative regimen described by Harris et al.24,25 The latter was advocated to minimize the risk of cerebral edema, which Harris et al. considered to be due to excessive fluid administration and inadequate sodium replacement in the first 24 hours of therapy.26

Our center1012 and others6'7 have modified protocols to emphasize the importance of the saline concentration of replacement and maintenance fluids (Table 2). For those patients at greatest risk of having cerebral edema, we increase sodium to 100 to 125 mEq/L (0.66% to 0.75% saline) for the first 12 hours, per Harris et al.25,26 Patients are given maintenance fluids (1,500 mL/m2/24 h) to which the calculated fluid deficit is added to correct dehydration during 48 hours (older than 2 years of age: 30 mL/kg for mild deficit, 60 mL/ kg for moderate deficit, and 90 mL/kg for severe deficit; younger than 2 years of age; 50 mL/kg for mild deficit, 100 mL/kg for moderate deficit, and 150 mL/kg for severe deficit). This is not to exceed 2 to 2.5 L/m2/24 h. Most importantly, all patients are closely monitored. Electrolytes and acid-base status are observed at 2-hour intervals initially, then at 4- to 6-hour intervals. Clinical progress and neurologic status are meticulously observed. Mannitol is at the bedside so it can be rapidly adrrtinistered at the first sign of neurologic deterioration. However, the exact causes of cerebral edema remain unclear. Because we do not understand what aspects of our treatment place patients at risk, the only way to avoid cerebral edema is to avoid DKA altogether.

Table

TABLE 4Diabetic Ketoacidosis Prevention Program

TABLE 4

Diabetic Ketoacidosis Prevention Program

PREVENTION OF DKA

Professional and public awareness as to the early signs and symptoms of diabetes in children and adolescents is required to decrease the incidence of DKA in patients with new-onset diabetes.27 Although our center has already seen such a decrease during the past 4 to 5 years, there could be a further reduction if educational efforts targeted to the public as well as to health care providers were to increase. These must be coupled with a mechanism to ensure access to appropriate treatment for all identified children. In 1998, 140 new cases of diabetes were diagnosed in our center. Thirty-six of these (26%) were in DKA at presentation. Four years earlier, 86% had DKA at the time of diagnosis.

There are many roles for the pediatrician in the prevention of DKA in new-onset patients. The first is to advocate diabetes screening in all at-risk children. Pediatricians may refer all first-degree or second-degree relatives of patients with type I diabetes mellitus for entry into the Diabetes Prevention Trial for type I diabetes mellitus or for other type I diabetes mellitus screening. Pediatricians should consider diabetes in the differential diagnosis of a number of diverse processes seen in clinical practice (Table 3). In children thought to have diabetes, the first screening can be done with urine (assessment for glycosuria and ketonuria). However, the results may be negative in early disease and, if there is still concern, hemoglobin A (HbA ) levels can be obtained. Finally, blood glucose screening done after eating is the most definitive way to determine whether diabetes is present. Early-onset diabetes may be associated with fasting euglycemia in children despite marked elevation of the blood glucose in the postprandial state. In known type I patients, pediatricians must help avert DKA during intercurrent illness, during emotional trauma, or from inadequate monitoring or lack of commitment to the diabetes regimen.

We have established a number of interventions in an attempt to diminish recurrent DKA in our known diabetes patients.27 Overall, these have been successful, accounting for a marked decrease in the rate of recurrent DKA in our approximately 1,000 patients. In 1998, this DKA rate was 0.04 events per patient per year. This was markedly reduced from 4 years prior when it was 0.12 events per patient per year.

Our center-based DKA prevention program, outlined in Table 4,27 has led to a decrease in recurrent DKA in our established patient population. This program is designed to avert DKA during intercurrent illness, emotional stress, trauma, errors in insulin administration, or purposeful manipulation of the diabetes regimen.28 Key to this program is teaching all patients and their families the early warning signs of DKA, giving guidelines for sick-day management, and having the availability of a 24-hour health care team committed to preventing DKA.

Table

TABLE 5Studies to Decrease Recurrent Diabetic Ketoacidosis

TABLE 5

Studies to Decrease Recurrent Diabetic Ketoacidosis

In addition, as shown in Table 5, we have decreased episodes of recurrent DKA in difficult to manage cases with multiple episodes of severe DKA with two distinct approaches: insulin pump therapy29 and out-of-home placement when the problem is a dysfunctional family.27,30 The latter studies showed that the likely explanation for recurrent DKA in these difficult to manage cases was inadequate home support.

SUMMARY

There remain a number of important controversies in the management of pediatric DKA. From the sodium content of the hydrational fluid to the rate of fluid administration that is best able to reverse the hyperosmolar dehydration attendant with DKA with minimal morbidity and mortality, there is no universal agreement on how patients with this devastating metabolic disturbance should be treated. It is still unclear what promotes or protects patients from neurologic insult during DKA reversal.

It is appropriate to begin to develop a national approach to eradicating DKA. This would require widespread public and professional education programs aimed at detecting new-onset type I patients prior to the onset of DKA. It would involve promoting diabetes screening programs aimed at detecting patients before the onset of symptomatic disease, and these would most appropriately be centered in the pediatrician's office. In the known patient, DKA still occurs as the result of intercurrent illness and nonadherence to the diabetes regimen due to patient or family chaos and dysfunction. Clearly, more strategies are needed to address these psychological and family patterns and the fact that many tenuous families have insufficient access to appropriate medical care. Those caring for children and adolescents must do all they can to prevent DKA and to treat it optimally to avert the toll this metabolic aberration takes on the pediatric diabetes population.

REFERENCES

1. LaPorte R, Matsushima M, Chang Y-F. Prevalence and incidence of insulin-dependent diabetes. In: Diabetes in America. Bethesda, MD: National Institutes of Health, Department of Health and Human Services; 1995:37-45.

2. Melton LJI, Palumbo PJ, Chu C-P. Incidence of diabetes mellitus by clinical type. Diabetes Care. 1983;6:75-86.

3. Faich GA, Fishbein HA, Ellis EA. The epidemiology of diabetic acidosis: a population-based study. Am f Epidemiol. 1983;117:551.

4. Pinkney JH, Bingley PJ, Sawtell PA, Dunger DB, Gale EAM. Presentation and progress of childhood diabetes mellitus: a prospective population-based study. Diabetologia. 1994;37:70-74.

5. American Diabetes Association. Diabetes 1996 Vital Statistics. Alexandria, VA: American Diabetes Association; 1996:13-20.

6. Plotnick L. Insulin-dependent diabetes mellitus. Pediatr Rev. 1995;16:304-305.

7. Wilson D. Diabetes mellitus in children and adolescentsIn: Behrman RE, Kliegman RM, Arvin AM, eds. Nelson Textbook of Pediatrics: Update 8, 15th ed. Philadelphia: W. B. Saunders; 1995:1-12.

8. Kaufman FR. Diabetes mellitus. Pediatr Rev. 1997;18:383392.

9. Kaufman FR. Outpatient management of children and adolescents with IDDM. Clinical Diabetes. 1994;6:146-150.

10. Kaufman FR. Treatment strategies for diabetic ketoacidosis in children and adolescents with insulin-dependent diabetes mellitus. Clinical Diabetes. 1993;102-106.

11. American Diabetes Association. Intensive Diabetes Management. Alexandria, VA: American Diabetes Association; 1995:65-78.

12. Kaufman FR. Diabetes in children and adolescents: areas of controversy. Med Clin North Am. 1998;82:721-738.

13. Axelrod L. Diabetic ketoacidosis. Endocrinologist. 1992;2: 375-383.

14. Roberts KB. Fluid management of children who have diabetic ketoacidosis. Pediatr Rev. 1995;16:304-305.

15. Kaufman IA, Keller MA, Nyhan WLK. Diabetic ketosis and acidosis: the continuous infusion of low doses of insulin. J Pediatr. 1975;87:846-848.

16. White NH, Henry DN. Special issues in diabetes management. In: Haire-Joshu D, ed. Management of Diabetes Mellitus: Perspectives of Care Across the Life Span. St. Louis, MO: Mosby; 1996:342-404.

17. Dorin RI, Crapo LM. Hypokalemic respiratory arrest in diabetic ketoacidosis. JAMA. 1987;257:1517-1518.

18. Bello FA, Sotos JF. Cerebral oedema in diabetic ketoacidosis in children. Lancet. 1990;336:64.

19. Täubin H, Matz R. Cerebral edema, diabetes insipidus, and sudden death complicating therapy of diabetic ketoacidosis. Diabetes. 1968;17:108-109.

20. Rosenbloom AL. Intracerebral crises during treatment of diabetic ketoacidosis. Diabetes Care. 1990;13:22-33.

21. Rosenbloom AL, Hanas R. Diabetic ketoacidosis (DKA): treatment guidelines. Clin Pediatr. 1996;35:261-266.

22. Harris GD, Fiordalisi I, Finberg L. Safe management of diabetic ketoacidosis. J Pediatr. 1988;113:65-68.

23. Duck SC, Wyatt DT. Factors associated with brain herniation in the treatment of diabetic ketoacidosis. J Pediatr. 1988;113:10-14.

24. Harris GD, Fiordalisi I, Harris WL, Mosovich LL, Finberg L. Mmimizing the risk of brain herniation during treatment of diabetic ketoacidemia: a retrospective and prospective study. J Pediatr. 1990;117:22-31.

25. Harris GD, Fiordalisi I. Physiologic management of diabetic ketoacidemia: a 5-year prospective pediatric experience in 231 episodes. Arch Pediatr Adolesc Med. 1994; 148:1046-1052.

26. Harris GD, Fiordalisi I, Yu C. Maintaining normal intracranial pressure in a rabbit model during treatment of severe diabetic ketoacidemia. Life Sciences. 1996; 59:1695-1702.

27. Kaufman FR, Halvorson M. Strategies to prevent diabetic ketoacidosis in children with known type 1 diabetes. Clinical Diabetes. 1997;6:236-239.

28. Golden MP, Herrold AJ, Orr DP. An approach to prevention of recurrent diabetic ketoacidosis in the pediatric population. J Pediatr. 1985;107:195-200.

29. Steindel BS, Roe TF, Costin G, Carlson M, Kaufman FR Continuous subcutaneous insulin infusion (CSII) in children and adolescents with chronic poorly controlled type 1 diabetes mellitus. Diabetes Res Clin Pract. 1995;27:199204.

30. Kaufman FR, Roe TF, Costin G, Lippi D. Residential treatment program for children with diabetes mellitus. Diabetes Care. 1987;10:665-666.

TABLE 1

Emergency Department Management of Pediatric Patients In Diabetic Ketoacidosis

TABLE 2

Guidelines for the Care of Patients With Diabetic Ketoacidosis In the Intensive Care Unit or the Specialized Unit

TABLE 3

Tips for the Pediatrician in Determining Whether Diabetes Is Present

TABLE 4

Diabetic Ketoacidosis Prevention Program

TABLE 5

Studies to Decrease Recurrent Diabetic Ketoacidosis

10.3928/0090-4481-19990901-09

Sign up to receive

Journal E-contents