This Article Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map 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 Finberg, L. Search for Related Content PubMed Articles by Finberg, L.

Appropriate Therapy Can Prevent Cerebral Swelling in Diabetic Ketoacidosis

University of California–San Francisco San Francisco, California 94111 and Stanford University Stanford, California

	Introduction

Top Introduction References

 
Appropriate therapy can prevent cerebral swelling in diabetic ketoacidosis (DKA)

In considering the issue posed by the title, there are two aspects of fluid movement in the brain that should be emphasized before we consider the pathophysiology of DKA and its treatment.

The first is a generalized response to induced hyperosmolality in the extracellular space (ECF). What are now known as osmolytes, previously called "idiogenic osmols," small organic molecules that accumulate in cells. In mammals taurine and myoinositol are the principal (only?) osmolytes. These molecules diffuse out or metabolize very slowly when a return to normal osmolality has occurred. Because the brain is in a rigid box, such cell swelling will lead to increased intracranial pressure when there is continuing hyperosmolality without loss of ECF volume.

The second aspect of fluid movement in and out of the brain relevant here is the nature of the capillaries through which water and solute diffuse. The brain capillaries, unlike most capillaries elsewhere in the body, have tight junctions between the cells. This anatomic feature produces "the blood brain barrier." Solutes not actively transported across the membrane, which includes sodium and chloride ions, diffuse at a much slower rate than these ions diffuse across other capillaries where movement is nearly instantaneous. In the brain, equilibrium for sodium takes 6 h (1). On the other hand, diffusion of water is virtually instantaneous in the brain as elsewhere.

Consider, then, what happens when an osmotic gradient between the plasma and brain fluid is created. Water moves instantly to equalize the osmotic concentrations while the dominant solutes (Na⁺ and Cl^-) move slowly. If the dilution gradient is from plasma to brain ECF, there will be an increase in brain volume. While the water is in the brain, ECF that is edema. If, however, the brain cells have been exposed to a hyperosmotic state and cells already possess osmolytes, then the water will be picked up by the cells producing brain swelling with or without "edema."

Now let us apply this knowledge to DKA. Before DKA occurs, there is a period of hyperglycemia. In a new diabetic this may have had a fairly long duration before diagnosis is appreciated; some going on to DKA. The prolonged hyperglycemia will have caused osmolyte concentration in cells, causing them to swell moderately. This has been verified by brain imaging (2). The swelling may be severe enough to cause symptoms, but usually at this stage there is enough reduction in ECF volume to modulate pressure changes.

If the patient has treatment begun at this point, with fluid having an osmotic concentration below plasma, water will flow disproportionately into the brain and intracranial pressure will rise. A large iv bolus of insulin metabolizes glucose to water that further aggravates the process. The excess water will be pulled to cells by the osmolytes and as the cycle continues, intracranial pressure progresses to cause headache, coma, and possibly death.

How should the clinician deal with this situation? It is more common in children than adults, probably because of their more rapid metabolism and water turnover. It is more common in first episodes of DKA, possibly because of a more prolonged hyperglycemic state before DKA.

Inappropriate therapy clearly will worsen the condition and in the past has done so. Five issues are important:

1. DKA dehydration is a hyperosmotic state that calls for slow repair, optimally over 48 h as with any hyperosmotic dehydration.
   
2. Use a fluid that has sodium and chloride or sodium, chloride, and bicarbonate in physiologic amounts (i.e. isotonic with normal ECF). Later in therapy, the NaCl concentration may be lowered.
   
3. If the patients’ circulation is stable, administer the fluid at a rate that covers ongoing insensible losses, physiologic urine loss (the excess urine loss is offset by water of metabolism), and 1/48 of the estimated deficit per hour. In other words, add the physiologic requirements for 48 h to the estimated deficit and deliver 1/48 per hour. If the circulation is not stable, give isotonic sodium salt fluid rapidly until stability is restored, then immediately slow down.
   
4. In estimating the patients’ deficit, do not assume a 10–15% body weight loss unless shock is present. The losses, especially in older children, are usually close to 5–7% of weight.
   
5. Give insulin by slow drip; do not give a bolus dose. Hyperglycemia should be reduced gradually. Having started the therapy, these children require close monitoring of their symptoms and of their electrolytes and glucose levels. Perhaps the most useful determination is the "corrected " sodium concentration, which is calculated as:
   

This number is a surrogate for a more physiologic concentration level in serum water, but it is in the units ordinarily used by clinicians and varies in identical fashion. This adjusts for the "osmotic space" filled by the excess glucose. If the corrected number is outside the physiologic range for sodium concentration (i.e. >149 meq/L), the patient is hypernatremic, and less sodium than usual will be needed. Conversely, if the corrected sodium is less than 132 meq/L, more is needed. If the patient either before or after the start of therapy has a headache or any other evidence of new neurological impairment, administer hypertonic mannitol to reduce intracranial pressure. Do not wait for fixed dilated pupils; it will already be too late. Applying these principles, Harris and Fiordilese (Ref. 3 and personal communication regarding the current number in the series and the lack of mortality and sequelae) have treated over 400 episodes of DKA without a fatality or sequelae. Twelve of their patients exhibited either headache or other neurological manifestations, which led to the use of hypertonic mannitol infusion, sometimes advised and given at a referring hospital before arrival at their institution. All of these patients recovered neurologically and are otherwise intact; a remarkable record. Errors of the past aggravated the already abnormal distribution of water in the brain. In particular, the administration of too much fluid (usually diluted below isotonic for sodium chloride) caused water influx into the brain, causing an increase in intracerebral pressure and the administration of an iv bolus of insulin, which increased ECF water from glucose metabolism. These events may cause hemorrhage, thrombosis, or cerebral herniation. Thus, there is a pathophysiologic explanation for cerebral "edema" during ill-advised management of DKA, and there are a set of principles which when carefully followed have permitted successful correction of the disturbance.

	References

Top Introduction References

 

1. Finberg L, Luttrell C, Redd H. 1959 Pathogenesis of lesions in the nervous system in hypernatremic states. II. Experimental studies of gross anatomic changes and alterations of chemical composition of the tissues. Pediatrics. 23:46–51.[Abstract/Free Full Text]
2. Krane EJ, Rockoff MA, Wallman JK, et al. 1985 Subclinical brain swelling in children during management of diabetic ketoacidosis. N Engl J Med. 312:1147–1151.[Abstract]
3. Harris GD, Fiordelisi I. 1994 Physiologic management of diabetic ketoacidemia: a five year prospective pediatric experience in 231 episodes. Arch Pediatr Adolesc Med. 148:1046–1052.[Abstract]

This article has been cited by other articles:

				R. D. Andrew, M. W. Labron, S. E. Boehnke, L. Carnduff, and S. A. Kirov Physiological Evidence That Pyramidal Neurons Lack Functional Water Channels Cereb Cortex, April 1, 2007; 17(4): 787 - 802. [Abstract] [Full Text] [PDF]

				A. B. Muir, R. G. Quisling, M. C.K. Yang, and A. L. Rosenbloom Cerebral Edema in Childhood Diabetic Ketoacidosis: Natural history, radiographic findings, and early identification Diabetes Care, July 1, 2004; 27(7): 1541 - 1546. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map 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 Finberg, L. Search for Related Content PubMed Articles by Finberg, L.