Pediatric DKA: Fluids and Insulin, How Much and When?
Pediatric DKA: Fluids and Insulin, How Much and When?
Authors: Irma Fiordalisi, MD, Professor, Department of Pediatrics, The Brody School of Medicine at East Carolina University; Director, Pediatric Intensive Care Unit and Sedation Services, Children's Hospital of Eastern North Carolina, Greenville, NC; Glenn D. Harris, MD, Professor, Department of Pediatrics, Section of Endocrinology/Diabetology, The Brody School of Medicine at East Carolina University; Active Medical Staff, Children's Hospital of Eastern North Carolina, Greenville, NC; Ronald M. Perkin, MD, MA, Professor and Chairman, Department of Pediatrics, Brody School of Medicine at East Carolina University, Greenville, NC.
Peer Reviewer: Ghazala Q. Sharieff, MD, FACEP, FAAP, FAAEM, Director of Pediatric Emergency Medicine, Palomar Pomerado Health System/California Emergency Physicians; Medical Director, Rady Children's Hospital Emergency Care Center; Assistant Clinical Professor, University of California, San Diego.
Introduction
Diabetic ketoacidosis (DKA) is a state of metabolic decompensation, secondary to insulin deficiency/resistance that is coupled with counter-regulatory hormone excess and results in varying degrees of hyperglycemia, ketoacidemia, hypertonic dehydration, and sometimes, alterations in mental status. DKA represents a spectrum of physiologic and biochemical aberrations ranging from minimal to severe disturbances in the degree of dehydration, ketoacidosis, and their physiologic consequences. While traditional definitions of DKA often include measured serum glucose > 200 or 250 mg/dL, a glucose of < 200 mg/dL should not exclude DKA; euglycemic DKA is well described.1 Similarly, a measured total CO2 < 15 or < 15 mEq/L have been benchmarks for differentiating DKA from uncontrolled diabetes. It is important to understand that in the absence of other causes of acidosis, a diabetic patient with positive serum ketones and a measured tCO2 of 16 or 17 mEq/L has ketoacidosis, however mild it may be.2 DKA remains the most common reason for hospitalization of pediatric patients with known type 1 diabetes mellitus; brain herniation during treatment for DKA has been the cause of death in 50-60% of childhood diabetes-related mortalities.3 While symptomatic brain swelling ("cerebral edema" [CE]) occurs in 0.5-1% of DKA episodes in population-based studies, it occurs in up to 5% of episodes in single-center studies, and carries a mortality rate of 20-25%.4
Consensus statements4-6 summarized guidelines for the practitioner that significantly changed the traditional approach to management, advocating closer neurologic monitoring, more gradual rehydration, and early intervention with osmotic therapy (e.g., hypertonic mannitol) for signs/symptoms that pique suspicions for raised ICP.
A review of the recent literature and recommendations for treatment follow.
Risk Factors For Cerebral Edema
Source: Glaser N, Barnett P, McCaslin I, et al. Rick factors for cerebral edema in children with diabetic ketoacidosis. N Engl J Med 2001;344:264-269.
This is a multicenter retrospective study undertaken to define risk factors for development of CE in patients < age 18 with DKA. The CE group included all patients with DKA, altered mental status, and either radiographically or pathologically confirmed CE, or improvement after intervention for suspected raised ICP.
In 58/61 identified episodes, neurologic deterioration occurred a median of 7 hours (range, 0-25) after initiation of therapy; 3/61 patients had symptomatic CE prior to treatment. Of 61 patients, 13 (21%) died and 13 (21%) survived with permanent neurologic sequelae. Affected patients were younger (8.9 + 4.2 years), more frequently Caucasian, and more likely to have new-onset diabetes when compared to random controls.
Compared to matched controls, all of the following were associated with CE: higher serum urea nitrogen concentrations (13-41 mg/dL) significantly lower PCO2 values (usually <18 mmHg), a smaller increase in serum sodium concentration during therapy, and treatment with bicarbonate. No increased risk was identified for rates of fluid, sodium, or insulin administration.
Therefore, the authors conclude that CE is not necessarily caused by therapeutic interventions. However, treatment with bicarbonate should be avoided in most circumstances. Patients with identified risk factors should be monitored closely for neurologic deterioration, and early intervention with hyperosmolar therapy should be readily available. They conclude that their data support "the hypothesis that CE in children with DKA is related to brain ischemia" based on the presence of hypocapnia, extreme dehydration, and hyperglycemia.
Commentary
This widely referenced and statistically rigorous work solidifies the increased risk for cerebral swelling among DKA patients who are younger; have new-onset disease, marked hyperventilation, and, therefore, more severe acidemia; and relatively higher initial concentrations of serum urea nitrogen. Although "treatment with bicarbonate" is not defined, rapid administration of hypertonic sodium bicarbonate is presumably the issue. DKA is a ketoacidosis; acid-base balance is restored by treatment with insulin, not bicarbonate "pushes."
Less clear are the findings of "smaller increases in the serum sodium concentration" during treatment, and the lack of association with rate of fluid administration since methods used for these determinations are not defined. No conclusions can be made regarding the relationship between the rate of fluid administered and failure of the serum sodium concentration to rise during therapy in this study because patients were not matched for volume of deficit (degree of dehydration). Since assessment of the degree of dehydration and determination of appropriate rates of volume delivery are based on multiple clinical and biochemical factors, it would be difficult to match patients retrospectively for this critical determinant. When all patients are receiving similar volumes of fluid despite wide variations in volume of deficit,7-9 no association with fluid therapy will be discernible with the methods employed.
Mild Cerebral Edema?
Source:Glaser NS, Wootton-Gorges SL, Buonocore MH, et al. Frequency of sub-clinical cerebral edema in children with diabetic ketoacidosis. Pediatr Diabetes 2006;7:75-80.
This prospective study evaluates the frequency of brain swelling in DKA using magnetic resonance imaging (MRI) in 41 patients (< age 18). Lateral cerebral ventricles were found to be significantly smaller during DKA treatment when compared to images taken after recovery in 54% of DKA episodes. Patients with ventricular narrowing during treatment were more likely to have both alterations in mental status, particularly lethargy and disorientation, and lower PCO2 values (18 + 2 vs. 27 + 2 mmHg); 12/15 children with abnormal mental status had Glasgow Coma Scale (GCS) scores of 13-14; 3/15 children had GCS of 12 or lower; none had overt signs of raised ICP.
The authors conclude that mild alterations in mental status associated with ventricular narrowing on MRI occur frequently and suggest that mild degrees of symptomatic CE may be more common than previously reported.
Commentary
This study supports prior findings that pediatric patients with DKA are at risk for development or progression of raised ICP during treatment, and that low initial PCO2 values are associated with this risk.10 Importantly, it underscores that even mild, non-specific alterations in mental status during DKA and its treatment may be attributable to symptomatic brain swelling. After initial volume resuscitation, all patients in this study received "an average volume of deficit" (70 mL/kg), an approach that may have provided excessive volumes of fluid to some patients and may have influenced the findings. Nonetheless, the presence of relatively smaller cerebral ventricles in mildly neurologically symptomatic patients is both instructive and useful for the bedside clinician since early intervention with mannitol and intensive care, including attention to fluid and insulin management, have been successful in minimizing neurologic morbidity and mortality associated with treatment of pediatric DKA.9
Hold the Fluid?
Source:Edge JA, Jakes RW, Roy Y, et al. The UK case-control study of cerebral oedema complicating diabetic ketoacidosis in children. Diabetologia 2006;49:2002-2009.
This is a case-control, prospective study over a 3-year period in children (< age 16) with DKA conducted through the British Paediatric Surveillance Unit. The aim was to determine the impact of baseline biochemical factors upon presentation of DKA, as well as treatment-related variables for risk of development of CE. CE was defined as a deterioration in mental status accompanied by additional signs of raised ICP (e.g., hypertension/bradycardia) or confirmatory radiologic or pathologic evidence. The volume of fluid given was described in tertiles; cases and controls were compared for total volumes received. An odds ratio (95% confidence interval) was calculated for each tertile for each of the first 4 hours of treatment. The method of insulin delivery was not described.
Risk factors identified for development of CE were:
- Greater degree of acidosis;
- Higher initial serum concentrations of potassium (K+) and urea nitrogen;
- Lower initial concentrations of serum sodium;
- Insulin administration in the first hour; and
- Cumulative volume of fluid administered over the first 4 hours.
The authors conclude that these factors should be taken into account when designing treatment protocols.
Commentary
This is a well designed study in which cases and controls were prospectively identified, although analyzed in retrospect. Two treatment-related factors that influenced the risk of development of symptomatic CE were identified. CE cases had a statistically significant greater degree of ketoacidosis, received relatively greater cumulative fluid volumes during the first 4 hours of treatment, and were more likely to receive insulin in the first treatment hour. While their data support an increase in risk of symptomatic CE as the cumulative volume of fluid given increases, it is difficult to extract specific information that is clinically useful from these data because volumes of fluid given were not related to ideal weight and degree of dehydration. P-values were adjusted for age, sex, known verses new-onset disease, and baseline acidosis. The limitation in this analysis is the same that plagues many similar efforts to identify, describe, or eliminate a role for fluid administration in the evolution of symptomatic CE during treatment of DKA. A given volume of fluid, however great or small, can only be justified by the physical examination, circulatory status, and biochemical disturbances of the individual patient. If the patient has mild tachycardia, normal blood pressure, warm feet, normal foot pulses, and brisk capillary refill, then 20 mL/kg of isotonic salt solution given in the first hour is "too much fluid." If the patient is hypotensive after 40 mL/kg of isotonic salt solution, then 40 mL/kg is not enough. Not only would 40 mL/kg be too little fluid in this setting, but a concomitant illness such as sepsis, septic shock, pancreatitis or the like should be aggressively pursued, since most patients with uncomplicated DKA have resolution of circulatory disturbances after 20 mL/kg of resuscitation fluid.
Large volumes of fluid administered during treatment for DKA have been associated with brain herniation.11 No information is provided in this study regarding estimated volume of deficit for those patients who received "larger volumes of fluid" in the first 4 hours, making it impossible to know if such rates of volume administration were warranted. Despite increased awareness of current recommendations for more gradual rehydration,4-6 inappropriately large volumes may be given early in therapy because of overestimation of the degree of dehydration. Administration of relatively hypotonic fluid also was associated with a high odds ratio for symptomatic CE in this study, but statistical significance was not achieved. These factors would have the potential to cause or exacerbate raised ICP.
To our knowledge, this is the first human study in which insulin administration in the first hour of treatment is identified as a risk factor for CE. Insulin may have been given at the same time as volume resuscitation, which would be expected to have a greater impact on lowering blood glucose than volume resuscitation alone. On the other hand, delay in insulin treatment prolongs the ketoacidotic state. There are no data in the literature describing an association between prolonged duration of ketoacidosis (as would occur when insufficient doses of insulin are given) with risk of CE; however, the longer its duration, the more prolonged the period of risk. Initiation of continuous IV insulin by the end of the first hour, after volume resuscitation, has been effective and safe.9 Therefore, while first-hour administration of insulin was associated with an increased risk of CE in this study, appropriate continuous IV insulin should not be unduly withheld in this life-threatening condition of insulin insufficiency.
Degree of Dehydration: Both Under- and Overestimated
Source: Koves IH, Neutze J, Donath S, et al. The accuracy of clinical assessment of dehydration during diabetic ketoacidosis in childhood. Diabetes Care 2004;27:2485-2487.
This study was undertaken to determine "the accuracy of the assessment of clinical dehydration" in patients < age 18 with type 1 diabetes and DKA. Two ED physicians independently assessed 37 pediatric patients with DKA, recording heart and respiratory rates, blood pressure, pale/cool hands and feet, peripheral capillary refill time, reduced skin turgor, level of consciousness, sunken eyes/fontanelle, dry tongue, Kussmaul breathing, blood glucose, and estimated degree of dehydration based on these data. Weight at discharge was compared to the pre-treatment weight to measure degree of dehydration.
Median weight gain was 8.7%. There was good inter-assessor agreement for 28/37 patients; there was no agreement between assessed and measured degrees of dehydration. The degree of dehydration tended to be overestimated in patients with <6% weight gain, and underestimated in patients with >6% weight gain.
The authors conclude that assessment of clinical dehydration in pediatric DKA appears to be unreliable. They recommend that 10-20 mL/kg 0.9% NaCl should be given until the patient is normotensive, followed by close monitoring and an initial assumption of 7-9% dehydration as "the best compromise."
Commentary
The median weight gain in this series may not accurately reflect changes in hydration status alone since some comparison weights were obtained 4-6 days from presentation; by that time, hospitalized patients presumably have received adequate insulin and nutrition, and therefore, have gained mass. It is remarkable that the presence and volume of peripheral pulses were not part of the clinical assessment, since examination of pulses is critical to determine the presence or absence of shock.12 It is rare for pediatric DKA patients to present with hypotension; the authors' recommendation to volume resuscitate only under this circumstance will leave circulatory shock untreated. In our experience, administration of 10 mL/kg aliquots until foot pulses are easily palpable is warranted.9
Kussmaul breathing, which typically correlates with degree of acidemia and hyperventilation, invariably results in very dry oral mucosa. Degree of acidemia has not been shown to correlate with the degree of dehydration; variables influenced primarily by the circulatory status are more useful in determining the degree of dehydration than parched oral mucosa in a very acidemic patient (The Critical Care Working Group, East Carolina University, Brody School of Medicine; unpublished).
Once peripheral (foot) pulses are normal, serial physical examinations and additional laboratory data are helpful in adjusting fluid delivery and more closely estimating the degree of dehydration;7-9 this obviates the need for assumptions. We agree with the authors that accurate assessment of the degree of dehydration is essential in DKA management.
Subcutaneous Insulin?
Source:Della Manna T, Steinmetz L, Campos PR, et al. Subcutaneous use of a fast-acting insulin analog: an alternative treatment for pediatric patients with diabetic ketoacidosis. Diabetes Care 2005; 28:1856-1861.
The objective of this study was to identify a method for subcutaneous insulin treatment using a fast-acting insulin that would afford "technical simplification and economic efficiency" when compared to treatment with continuous IV regular insulin (CIRI) in pediatric DKA.
Sixty episodes of DKA in patients age 3 to 18 were randomized to receive CIRI 0.1 units/kg/hr (30 patients) or lispro insulin 0.15 units/kg every 2 hours until glucose approached 13.8 mmol/L (250 mg/dL), after which the same dose was given every 4 hours for the next 24 hours. In the CIRI group, insulin was changed to regular insulin 0.15 units/kg/dose subcutaneously every 4 hours for 24 hours after glucose was less than 13.8 mmol/L (250 mg/dL).
In the lispro group, ketosis resolved 12 hours after the glucose reached the target level; in the CIRI group, ketosis resolved by the sixth hour. "Metabolic homeostasis" (not defined) was not achieved in either group within 30 hours of treatment; there were no instances of hypo- or hyperkalemia, untoward declines in effective osmolality, or CE.
The authors conclude that both regimens minimize the risk of hypokalemia, hypoglycemia, and abrupt falls in osmolality. They attribute the longer recovery time of the lispro insulin group to the change to every 4 hourly dosing of this rapidly-acting insulin analog, and speculate that smaller, more frequent lispro insulin doses may circumvent the problem.
Commentary
This work is in part a response to the anticipated need for an intensive care environment to provide CIRI for pediatric patients with DKA, and an effort to "simplify" delivery of insulin. The neurologic and metabolic monitoring indicated typically require the resources of an intensive care unit in most regions. However, we agree with the authors, who write from Sao Paulo University in Brazil, that the level of monitoring, skill of fluid replacement, and availability of necessary interventions should determine the optimal placement of patients.
With regard to the insulin regimens studied, their control group did not receive CIRI until ketosis was resolved, but only until glucose was < 13.8 mmol/L (250 mg/dL). This abbreviated course of CIRI does not represent the current standard of treatment.4-6 Even with a relatively short course of CIRI, ketosis resolved more quickly in controls than in the lispro insulin group. The potential benefits of subcutaneous insulin delivery need to be weighed against a longer duration of DKA and the ensuing metabolic consequences. Of note, neither group received standard CIRI and neither group achieved "metabolic homeostasis" after 30 hours, a remarkably long duration of metabolic disequilibrium. While there were no complications due to brain swelling, the series is small and the risks of a longer duration of ketoacidosis in the pediatric patient versus the use of standard doses of CIRI until ketoacidosis resolves need further evaluation. CIRI permits rapid intervention, followed by a rapid clinical response, that cannot be duplicated with intermittent subcutaneous insulin dosing.13 Given these issues, CIRI remains the preferred method of insulin delivery in treatment of pediatric DKA.4-6,13
Two-bag System
Source: Poirier MP, Greer D, Satin-Smith M. A prospective study of the "two-bag system" in diabetic ketoacidosis management. Clin Pediatr (Phila) 2004;43:809-813.
The "two bag system," a clinical application of the euglycemic clamp technique, involves delivery of IV fluid using 2 solutions of identical electrolyte but different glucose concentrations (e.g., 0.9% NaCl plus 40 mEq/L K+ and D10 0.9% NaCl plus 40 mEq/L K+); this permits adjustment of the amount of glucose delivered by varying the rate of infusion from each bag. The objective of this prospective study was to assess the benefits of this system in initial ED management of children with DKA. While the remainder of their management was standardized, 2 groups (one-bag, 16 patients; two-bag, 17 patients) were compared for rate of decline in serum glucose and rate of bicarbonate correction, duration of IV insulin therapy, and response time for IV fluid changes. The latter was the only measure that differed between the two groups: one-bag, 42 minutes; two-bag, 1 minute.
The authors conclude that this system simplifies the process of responding to changes in glucose requirements. They add that the method was "more convenient and less labor intensive" than the traditional approach, that it may be ideal for use during patient transport, and that a larger sample may have yielded additional benefits, such as improvements in correction of hyperglycemia and acidosis.
Commentary
This study demonstrates the usefulness of a "two-bag system" for responding to changes in need for glucose-containing fluids during DKA treatment in an ED. Other potential benefits such as an increase in the rate of correction of acidosis also may be realized if administration of a therapeutic dose of insulin need not be interrupted while awaiting a change in glucose-containing IV fluid. The authors site small sample size as a limitation in their study. Another limitation may be their management protocol. A decrease in the serum glucose by 100 mg/dL per hour is their therapeutic goal; however, their guideline to "...either decrease rate of insulin infusion or increase amount of dextrose given...," to achieve that goal implies that either intervention is equivalent. If the insulin dose were decreased (rather than increasing the delivery of glucose) prior to full correction of ketoacidosis, this also may have lengthened the time to full recovery.
Nevertheless, the efficiency afforded a busy ED and its pharmacy makes the "two-bag system" a good alternative to multiple IV fluid bag changes during DKA management.
Conclusion
Pediatric patients with moderate to severe DKA (bicarbonate < 15 mEq/L) require close neurologic monitoring and skillful metabolic care. Brain herniation is most likely a multifactorial complication and requires a multipronged therapeutic approach if its occurrence is to be minimized. Factors that increase risk for this complication include:
1. Young age;
2. New-onset diabetes;
3. Marked hyperventilation/ severe acidemia;
4. Higher initial concentration of serum urea nitrogen and lower initial concentration of serum sodium;
5. Failure of the measured concentration of serum sodium to increase appropriately as that of glucose declines during treatment;
6. Increased cumulative volume of fluid administered over the first 4 hours of treatment;
7. Insulin administration during the first hour; and
8. Treatment with bicarbonate, presumably hypertonic "pushes" of sodium bicarbonate.
These thus far identified risk factors are not necessarily present in all patients who develop symptoms of raised ICP; suggestive neurologic symptoms should be evaluated and treated, even in the absence of known risk factors.
Neurologic monitoring is of paramount importance. Early treatment with hypertonic mannitol, often accompanied by a decrease in volume delivery, when raised ICP was suspected reversed ominous symptoms and signs in 17 of 35 mannitol recipients in a large prospective study.9 With continued attention to correction of ketoacidosis, and fluid management that prevents untoward declines in effective osmolality, all 35 mannitol recipients in that study recovered uneventfully. No adverse effects of mannitol administration were observed.9 Head CT scan should not take precedence over treatment with mannitol when increased ICP is suspected. The absence of overt CE on CT scan does not rule out raised ICP. Brain imaging should be obtained in all neurologically symptomatic patients.
Recent research has underscored pathophysiologic factors that would enhance susceptibility to development of CE during DKA; these include central nervous system acidosis, cerebral hyperemia, and impaired cerebral autoregulation. The natural progression of these disturbances during insulin and fluid treatment is incompletely understood. However, in any setting of vulnerability to development of raised ICP, care in fluid and electrolyte management is essential.
Recommendations include:
1. Resuscitation of shock: give 10 mL/kg aliquots of 0.9% NaCl until foot pulses are easily palpable. The approach of giving 10 mL/kg aliquots allows for frequent re-assessment of the circulatory status and minimizes the risk of volume overload. If the patient is hypotensive (rare in pediatric DKA), the clinician can predict that 20 mL/kg or more likely will be needed. This should be delivered with frequent reassessments. Severe acidemia causes vasoconstriction; cool skin, delayed capillary refill time, and dry mucosa and are not necessarily reasons to provide volume resuscitation when arterial blood pH is very low, as long as foot pulses are normal. Few patients require > 20 mL/kg (ideal body weight) to restore foot pulses. (See Figure 1.)
 |
2. Estimate the volume of deficit based on physical examination and laboratory data and plan replacement evenly over 48 hours,7,9 along with maintenance requirements. During hyperventilation, the maintenance allotment will be approximately 30% greater than the usual maintenance requirement. If the patient is overweight, fluid calculations should be based on ideal (not actual) body weight. (See Figure 2.)
 |
3. Use near isotonic fluids (lactated Ringer's solution plus K+, plus dextrose as needed) after shock (if present) is corrected. For patients age < 2, give Na+ 100 mEq/L plus K+. This fluid should continue throughout correction of ketoacidosis unless hypernatremia is present after DKA is repaired.
4. Patients should not be permitted to drink since this invariably leads to intake of very hypotonic fluid.
5. Start IV regular insulin 0.1 units/kg/hr at completion of the first hour of fluid treatment (after volume resuscitation) and continue IV insulin delivery at an appropriate dose (usually 0.1 units/kg per hour) until ketoacidosis is corrected. The dose of insulin should be based on the actual (not the ideal) body weight. Add dextrose to IV fluid once glucose is < 300 mg/dL; further increases in dextrose delivery may be needed to maintain a dose of insulin that continues to correct ketoacidosis. The half-life of IV regular insulin is approximately 6 minutes; therefore, intermittent pushes of IV regular insulin should not substitute for continuous insulin infusions.
Insulin pumps deliver insulin subcutaneously. If ketoacidosis occurs during use of an insulin pump, malfunction of this device should be assumed, and continuous IV insulin should be given.
6. Monitor glucose hourly and blood gases and electrolytes every 1-2 hours; monitor serum magnesium and phosphorus periodically.
7. Do not use antiemetics. Nausea and vomiting typically resolve with insulin treatment. If these symptoms persist, complicating problems (raised ICP, pancreatitis, appendicitis) should be ruled out.
8. Consider use of the "two-bag system" to facilitate changes in glucose delivery during treatment.
9. Provide intensive neurologic monitoring for symptoms and signs of raised ICP such as headache, progressive lethargy, alterations in mental status, relative bradycardia, hypertension, pupillary changes, or any focal neurologic findings. If raised ICP is suspected, intervene with administration of mannitol (0.5 to 1 gram/kg) IV and re-evaluate the patient for a possible decrease in IV fluid administration. If the patient is hemodynamically unstable (e.g., hypotensive) NaCl 3% (3 to 10 mL/kg) should be considered.14
10. Consider early consultation with a pediatric intensivist or an expert in pediatric diabetes.
11. Arrange for care in a pediatric ICU or setting in which equivalent care is provided by staff trained in the management of pediatric DKA.
It is important to determine the appropriate volume of deficit for each patient. Provided the rehydration solution contains an adequate amount of sodium (approximately 125 mEq/L for patients age > 2), then monitoring of the physical examination, urea nitrogen concentration, and maintenance of a relatively constant predicted sodium value ([Na+] + 1.6 X ([Glucose-100] /100)) will allow determination of an appropriate rate of volume delivery, which is perhaps more important than assigning "a degree dehydration." The rehydration plan should be tailored to individual need; repeated physical examination and serial laboratory measures should guide fluid replacement. (See Figure 2.)
Prevention of DKA should be the primary goal. With increased provider awareness for early diagnosis of new onset diabetes, compliance with appropriate "sick day rules" and availability of 24-hour help lines, progression to severe ketoacidosis should be avoidable. Patients who succumb to DKA are susceptible to development of CE; the risk of brain herniation during treatment can be minimized when close monitoring, insulin therapy, and metabolic care are provided in a timely manner.
References
1. Burge MR, Hardy KJ, Schade DS. J Clin Endocrinol Metab 1993;76:1192-1198.
2. Harris GD, Fiordalisi I. In: Perkin MR, Swift JD, Newton DA, et al, eds. Pediatric Hospital Medicine. 2nd ed. Lippincott Williams and Wilkins: Philadelphia; 2008:528.
3. Edge JA, Ford-Adams ME, Dunger DB. Arch Dis Child 1999;81:318-323.
4. Dunger DB, Sperling MA, Acerini CL, et al. Pediatrics 2004;113:e133-e144.
5. Wolfsdorf J, Glaser N, Sperling MA. Diabetes Care 2006;29:1150-1159.
6. Wolfsdorf J, Craig ME, Daneman D, et al; International Society of Pediatric and Adolescent Diabetes. Pediatr Diabetes 2007;8:28-43.
7. Harris GD, Fiordalisi I, Harris WL, et al. J Pediatr 1990;117:22-31.
8. Harris GD, Fiordalisi I. Arch Pediatr Adolesc Med 1994;148:1046-1052.
9. Fiordalisi I, Novotny WE, Holbert D, et al. Pediatr Diabetes 2007;8:142-149.
10. Glaser N, Barnett P, McCaslin I, et al. N Engl J Med 2001; 334:264-269.
11. Duck SC, Wyatt D. J Pediatr 1988;113:10-14.
12. Ralston M, Hazinski MF, Zaritsky, et al, eds. Pediatric Advanced Life Support Provider Manual. American Heart Association; 2002:35.
13. Perkin RM, Marks JF. Clin Pediatr (Phila) 1979;18:540, 545-548.
14. Kamat P, Vats A, Gross M, et al. Pediatr Crit Care Med 2003;4:239-242.
Diabetic ketoacidosis (DKA) is a state of metabolic decompensation, secondary to insulin deficiency/resistance that is coupled with counter-regulatory hormone excess and results in varying degrees of hyperglycemia, ketoacidemia, hypertonic dehydration, and sometimes, alterations in mental status.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.