This Article Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rosenbloom, A. L. Search for Related Content PubMed Articles by Rosenbloom, A. L.

Cerebral Edema in Diabetic Ketoacidosis

Department of Pediatrics Children’s Medical Services Center Gainesville, Florida 32608

	Introduction

Top Introduction References

 
INTRACEREBRAL complications during treatment for diabetic ketoacidosis (DKA), principally falling under the rubric "cerebral edema," account for most diabetes-related mortality during childhood (1, 2). In his presentation, Dr. Laurence Finberg opines that this problem can be prevented by avoiding "ill-advised management of DKA" and provides principles for appropriate therapy. Dr. Andrew Muir, while agreeing with the therapeutic wisdom of judicious fluid replacement, arrays considerable evidence that this will not prevent the problem. Let us consider two patients from my recent experience.

A 14-month-old girl weighing 10 kg had newly diagnosed diabetes associated with bilateral otitis media. She was vomiting, dehydrated (estimated 10% with dry skin, delayed capillary refill, soft eyeballs), and listless but arousable. Serum glucose concentration was 22.7 mmol/L, sodium 132 mmol/L, and arterial pH 7.03. She was given 10 mL/kg 0.9% sodium chloride as an iv bolus, followed by 0.45% saline with added potassium at a rate that would result in replacement over 36 h. By the 5th h of treatment, her serum sodium level was 143 mmol/L and blood glucose 5.8 mmol/L, resulting in an increase in calculated serum osmolality from 294 to 299 mOsm/L. Ten percent dextrose was added to her iv fluid. By the 17th h, she was more alert, sitting up, and taking clear fluids by mouth.

An 11-yr-old girl was brought to the emergency department with deep breathing and lethargy with slurred speech. She weighed 43 kg, had a blood sugar level of 29.7 mmol/L, serum sodium concentration of 147 mmol/L, and venous pH 6.97. During the first 2 h of treatment at the local emergency room she received 4 L normal saline with 350 mmol of added sodium bicarbonate. By this time her blood sugar had dropped to 15.7 mmol/L, serum sodium was 148 mmol/L, and her pH was up to 7.27. In telephone consultation, the Children’s Hospital endocrinologist suggested decreasing the iv fluid rate to twice maintenance and at h 4 she was transferred by helicopter.

Which of these youngsters would be expected to develop cerebral edema, or would both of them be considered at serious risk? The first patient falls into the age group at greatest risk, those under 5 yr (2), but her treatment was consistent with Dr. Finberg’s recommendations, and she was doing well at 17 h. Between the 25th and 31st h of treatment, however, she experienced relative bradycardia (68–104 beats per minute) without change in blood pressure and at h 32 developed posturing, became unresponsive, and had anisocoria. She was treated with mannitol, intubated, and hyperventilated. Initial computed tomography scan of the head showed indistinct basal cisterns without herniation and normal ventricles, but after several hours the computed tomography scan demonstrated profound diffuse cerebral edema with obliterated ventricles and supra- and infratentorial herniation. She did not recover. The second patient was much more hyperosmolar and received a full days maintenance fluids plus (using Finberg’s upper limit of 7% for an older child) two thirds of replacement in the first few hours of treatment! In addition to the surfeit of water, this already hypernatremic youngster received a massive amount of sodium bicarbonate, effectively increasing her iv fluid salinity to 240 mOsm/L or 1.4%. Her emergency transfer was in anticipation of problems from this quite inappropriate management. None occurred.

The old saying, so constantly and so blindly repeated, that the exception proves the rule, is as destitute of truth, as it is of meaning. Such an exception can prove only one thing, and that is, that the rule is not fully understood, or completely ascertained. [Elisha Bartlett (1804–1855), Philosophy of Medical Science].

Brain edema is defined as an increase of cerebral tissue water causing an increase of tissue volume (3). The prototypic forms are: 1) vasogenic, due to breakdown of the blood brain barrier as around a tumor or with trauma; 2)cytotoxic, from poisoning or metabolic derangement; and 3) osmotic, as with hyponatremia. As discussed by Dr. Muir, neither the cause nor the location of the water in the swollen brain of DKA is known. Dr. Finberg singularly defines edema when the water is in the brain extracellular fluid space and not when it is within cells causing brain swelling. Nonetheless, he notes that swollen cells are present before treatment in DKA, evidenced by brain imaging. The imaging study cannot, of course, identify whether the edema is intra- or extracellular. As noted by Dr. Muir, confirmation of the presence of a degree of cerebral edema before treatment has not been uniform in studies of pediatric patients, although data from adults is more consistent (4, 5). The relevance of these findings to the cerebral edema syndrome in children is questionable, because adults do not get clinical cerebral edema, including those severely hypernatremic and hyperglycemic patients with hyperosmolar nonketotic coma who are rapidly rehydrated and made euglycemic (6).

Overestimation of the magnitude of dehydration in DKA is common, and Dr. Finberg is correct to caution about assuming a deficit that denotes shock (15%) for all but the youngest children (those under 2 yr). Children with impressive signs of dehydration, despite hyperosmolality, however, such as soft eyeballs and delayed capillary refill, are likely closer to 10% dehydrated than the 5–7% suggested by Dr. Finberg.

The most convincing argument against the hypothesis that adherence to the fluid administration principles put forth by Dr. Finberg will prevent cerebral edema is supplied by his discussion of the experience of Fiordilese and Harris in applying these principles. Of 400 patients, 12 had sufficient evidence of cerebral edema to warrant administration of mannitol, a remarkable 3%. This is a much higher incidence of cerebral edema than in other reported series.

The good news from the series of Fiordilese and Harris, however, underscores the message that can be taken from both Dr. Finberg and Dr. Muir, that diligent monitoring of vital signs and neurologic status, with early administration of mannitol as indicated, must be a part of DKA management in children. Functional imaging techniques may eventually provide sufficient understanding of the effect of DKA and its treatment on the child’s brain that this complication can be prevented.

	References

Top Introduction References

 

1. Scibila J, Finegold D, Dorman J, Becker D, Drash A. 1986 Why do children with diabetes die? Acta Endocrinol Suppl. 279:326–333.
2. Rosenbloom AL. 1990 Intracerebral crises during treatment of diabetic ketoacidosis. Diabetes Care. 13:22–33.[Abstract]
3. Pappius HM. 1974 Fundamental aspects of brain edema. In: Vinkin PJ, Bruyn GW, eds. Handbook of clinical neurology, vol. 16, part 1: Tumors of the brain and skull. Amsterdam: North Holland Publishing Co.; 167–185.
4. Clements Jr RS, Blumenthal SA, Morrison AD, Winegrad AI. 1971 Increased cerebrospinal fluid pressure during treatment of diabetic ketosis. Lancet. 2:657–661.
5. Fein IA, Rackow EC, Sprung CL, Grodman R. 1982 Relation of colloid osmotic pressure to arterial hypoxia and cerebral edema during crystalloid volume loading of patients with diabetic ketoacidosis. Ann Intern Med. 96:570–574.
6. Carroll P, Matz R. 1983 Uncontrolled diabetes mellitus in adults: experience intriguing diabetic ketoacidosis and hyperosmolar nonketotic coma with low dose insulin and a uniform treatment regimen. Diabetes Care. 6:579–585.[Abstract]

This article has been cited by other articles:

				M. J. Fagan, J. Avner, and H. Khine Initial Fluid Resuscitation for Patients With Diabetic Ketoacidosis: How Dry Are They? Clinical Pediatrics, November 1, 2008; 47(9): 851 - 855. [Abstract] [PDF]

This Article Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rosenbloom, A. L. Search for Related Content PubMed Articles by Rosenbloom, A. L.