CONTROVERSY 443 General paediatrics case controlled studies of cerebral Arch Dis Child: first published as 10.1136/adc.86.6.443 on 1 June 2002. Downloaded from ................................................................................... oedema complicating DKA and there is conflicting evidence as to the role of iatrogenic factors in fatal cases.613How- Fluid management in diabetic ever, excessive rates of fluid administra- tion, particularly early in resuscitaion,14 ketoacidosis and failure of plasma sodium to rise as plasma glucose declines, have been iden- C D Inward, T L Chambers tified as risk factors. Studies using computed tomography ................................................................................... suggest that subclinical cerebral swelling Have we got it right yet? may be a common occurrence in children presenting with DKA,15 16 although this has been disputed.17 If cerebral swelling oung people with insulin depend- may be considered and treated as a vari- is a relatively common event it is not ent diabetes mellitus are three ant of renal electrolyte disorder with clear why only a minority progress to times more likely to die in child- hypertonic dehydration. herniation. In addition, fatal herniation Y 1 hood than the general population. De- has occurred in the absence of any fluid spite advances in management over the FLUID AND ELECTROLYTE LOSSES therapy. Therefore, although clearly not past 20 years, the incidence of mortality The fluid and electrolyte losses of DKA the only cause, it seems reasonable to associated with diabetic ketoacidosis are predominantly caused by hypergly- conclude that fluid management may be (DKA) remains unchanged. Cerebral caemia with resultant glycosuria and a factor contributing to the development oedema is the predominant cause of this osmotic diuresis. In addition, the kidney of overt cerebral oedema. On the basis of mortality; young children are particu- has a low threshold for ketoacids, which the available evidence it would seem larly at risk, with an incidence of 0.7–1% are excreted into the urine with an prudent to avoid rapid rates of fluid of episodes of DKA.23 The mortality accompanying cation, further exacerbat- infusion and to maintain serum sodium appears to be greatest among patients at ing the electrolyte loss. Relative acute concentrations in order to avoid the first presentation,134if there has been a dehydration with a larger sodium loss potential danger of excess free water long history of symptoms prior to will result primarily in intravascular administration which may exacerbate admission,3 and during the first 24 hours dehydration, but a more prolonged de- the development of symptomatic brain of treatment.4 In a recently published velopment will result in a greater degree swelling. retrospective multicentre analysis of of intracellular dehydration, with rela- Failure to follow guidelines for the children with DKA, low pCO2 levels and tive maintenance of intravascular vol- management of DKA has been well high serum sodium concentration at ume, particularly if the patient remains documented.18 Of particular concern is presentation were identified as particu- polydipsic.9 There are thus wide varia- the evidence that clinicians have been lar risk factors for the development of tions in fluid, acid–base, and electrolyte consistently shown to overestimate the cerebral oedema, together with bicarbo- deficiencies, during both development degree of dehydration in this 5 19 20 nate therapy. However, in the accompa- and treatment of DKA, yet few protocols condition. Nevertheless, the assump- http://adc.bmj.com/ nying editorial, Dunger and Edge point acknowledge this. tion underlying most DKA protocols is out that this may simply be revealing an There is clear evidence that volume that rehydration is desirable and achiev- association between severe DKA and depletion triggers the release of counter able in 24 hours, using a volume calcu- dehydration and the risk of cerebral regulatory hormones including catecho- lated by “maintenance plus deficit”. oedema.6 The pathogenesis of cerebral lamines, growth hormone, and cortisol oedema remains poorly understood but as well as renin, aldosterone, and ar- THE NEED FOR CHANGE? there may be many contributing ginine vasopressin.10 11 These hormones, From a renal perspective the term 7 factors. whose combined actions are directed “maintenance fluid” is a misnomer. True on September 28, 2021 by guest. Protected copyright. The aim of management of DKA is to towards preserving intravascular vol- maintenance fluid volume in children is restore metabolic homoeostasis while ume, also cause insulin resistance. An the sum of the insensible losses plus the minimising the risks of complications important effect of fluid administration obligate urine production and is prob- including hypoglycaemia, hypokalaemia, is therefore reducing the stimulus for the ably only 40% of recommended cardiac failure, and in children the secretion of the counter regulatory hor- amounts.21 Water requirements are re- development of cerebral oedema. How mones. Indeed it has been shown that lated to energy expenditure. The recom- best to achieve this remains contentious, plasma glucose starts to fall following mended “maintenance” volumes are with particular controversy centred on fluid administration prior to giving insu- based on calculations of insensible losses optimal fluid management. The most lin, although insulin is required for (estimated as 50 ml/100 cal/day) plus appropriate volume, type, and rate of correction of the metabolic acidosis.12 66.7 ml/100 cal/day to replace urine fluid to be given have all been the subject output.22 This allows for production of of debate. A survey in 1994 of UK RISK FACTORS FOR CEREBRAL isotonic urine. As normally functioning paediatricians found a threefold varia- OEDEMA kidneys are capable of fourfold concen- tion in the amount of fluid recom- The importance of prompt restoration of traton of the glomerular filtrate, “main- mended within the first 12 hours.8 Since the circulating volume for patients pre- tenance fluids” will provide sufficient then national guidelines have been de- senting with DKA is clear. However, the volume for both insensible and obligate veloped by the British Society for Paedi- observation that symptomatic cerebral losses as well as the extra needed for atric Endocrinology (BSPE). However, it oedema often occurs several hours after rehydration. A 5% dehydrated child with remains uncertain whether the recom- treatment has begun, usually at a time normally functioning kidneys can be mended strategy, using a volume calcu- when plasma glucose has started to fall rehydrated using “maintenance” fluid lated by “maintenance plus deficit”, is and metabolic acidosis to improve, has only—albeit slowly. the safest and most appropriate way to led to concern that the treatment itself In non-diabetic dehydration it is cus- manage rehydration in children with may contribute to the development of tomary to assess whether the dehydra- DKA. We suggest that childhood DKA cerebral oedema. There have been few tion is hypertonic, hypotonic, or isotonic www.archdischild.com 444 CONTROVERSY and whether there is simple contraction ..................... 21 Molteni KH. Initial management of of the extracellular compartment or hypernatraemic dehydration in the breastfed Arch Dis Child: first published as 10.1136/adc.86.6.443 on 1 June 2002. Downloaded from Authors’ affiliations infant. Clin Pediatr 1994;33:731–40. whether compensatory mechanisms are C D Inward, T L Chambers, Bristol Royal 22 Holliday MA, Segar WE. The maintenance operating. Hospital for Children, Bristol, UK need for water in parenteral fluid therapy. It is accepted that safe therapy of Pediatrics 1957;19:823–32. Correspondence to: Dr C D Inward, Bristol 23 Hollidat M. The evolution of therapy for hypertonic dehydration should be both Royal Hospital for Children, Upper Maudlin dehydration: should deficit therapy still be prolonged and cautious to avoid the Street, Bristol BS2 8BJ, UK taught? Pediatrics 1996;98:171–7. development of fits and cerebral oedema 24 Oddie S, Richmond S, Coultard M. 23 24 Hypernatraemic dehydration and breast during rehydration. REFERENCES feeding: a population study. Arch Dis Child The cells of the central nervous system 1 Edge JA, Ford-Adams ME, Dunger DB. 2001;85:318–20. protect cell volume under hyperosmolar Causes of death in children with insulin 25 Lee JH, Arcinue E, Ross BD. Organic dependent diabetes 1990–96. Arch Dis Child osmolytes in the brain of an infant with conditions by producing intracellelular 1999;81:318–23. hypernatremia. N Engl J Med osmotically active molecules, “idiogenic 2 Edge JA, Hawkins MM, Winter DL, et al. 1994;331:439–42. osmoles”.25 Once formed these molecules Incidence presentation management and 26 Lebovitz HA. Diabetic ketoacidosis. Lancet outcome of cerebral oedema associated with dissipate slowly. Rapid reduction of 1995;345:767–71. diabetic ketoacidosis (DKA) in Great Britain. 27 Adrogue H, Barrero J, Eknoyan G. Salutary plasma osmolality by the administration Arch Dis Child 1999;80(suppl 1):A11. effects of modest fluid replacement in the of free water under these conditions will 3 Bello FA, Sotos JF. Cerebral oedema in treatment of adults with ketoacidosis. JAMA diabetic ketoacidosis in children. Lancet create a gradient, causing movement of 1989;262:2108–13. 1990;336:64. 28 Harris GD, Fiordalisi I. Physiologic water into brain cells and thus brain 4 Rosenbloom AL. Intracerebral crises during management