It’s commonly accepted now that bicarb should not be given in DKA (especially in children), unless they are continuing to clinically deteriorate, but what is the reasoning behind this? Theoretically it makes sense to give bicarbonate to a patient in severe acidosis, but there are several reasons this is not wise in most situations.
First off, it’s important to understand that the mainstays of DKA treatment in the ED, IV fluids and insulin, are the first and most important steps in reversing acidosis. IV fluids are aimed at reversing the hypovolemia, resulting in increased tissue perfusion and increased excretion of organic acids. Also, as soon as the body receives insulin it can stop producing ketoacids as it now has its favorite energy source again. Stick to the basics of IV fluids and insulin (usually in drip at 0.1 units/kg/hr) and you will start reversing the acidosis.
Secondly, bicarbonate may not help. In a retrospective review, one study looked at 147 patient admissions (children) and found that 57 of the 147 (39%) were treated without bicarbonate and the outcome was the same. Several other studies in adults also show that there is not a benefit in giving bicarbonate in the long term. It may increase your pH quickly but remember, the goal is to see the patient get better and not treat a number!
The last argument against the use of bicarbonate is the most often quoted in the ED. Not only is it not necessary and it doesn’t help, it also can harm! Bicarbonate can paradoxically cause cerebral acidosis, increase hepatic ketone production, worsen hypokalemia, and in children can cause cerebral edema. It is thought that the paradoxical acidosis occurs in kids because the bicarb cannot directly cross their blood brain barrier, but it can be converted to CO2 which can cross the BBB. The main study claiming the cerebral edema in children is from 2001 in the New England Journal of Medicine by Glaser, et al. The study was retrospective looking at 61 kids with cerebral edema in DKA compared to age matched controls. They found that low partial pressures of arterial carbon dioxide and increased serum urea nitrogen on presentation were associated with higher risk of cerebral edema. The only therapy they found that increased the risk of cerebral edema was sodium bicarb use, after adjusting for the other variables. The relative risk was 4.2 with a P of 0.008. Retrospective studies show correlation rather than causation, but I doubt anyone will want to conduction prospective studies when the correlation is that ominous.
Now that I’m done bashing on bicarb, most sources still say there is a role for its use in severe cases, mostly in adults. The pH cutoff listed is usually 6.9, although everyone has a different cutoff in their mind as to when they will use bicarb. Others will take the whole clinical picture into account, such as failure to correct acidosis after fluids and insulin. Certain review articles also recommend bicarb when the hyperkalemia is severe and resulting in hemodynamic instability. If you do pull the trigger to give bicarb, the recommended dose is 1-2 mmol/kg over 60 minutes, not a rapid push as is done in a code situation.
Glaser, Barnett, et al. Risk Factors for Cerebral Edema in Children with Diabetic Ketoacidosis. The New England Journal of Medicine. January 2001. 344(4): 264-270.
Glaser, Kupperman. The Evaluation and Management of Children With
Diabetic Ketoacidosis in the Emergency Department. Pediatric Emergency Care. July 2004. 20(7): 477-483.
Green, Rothrock, et al. Failure of Adjunctive Bicarbonate to Improve
Outcome in Severe Pediatric Diabetic Ketoacidosis. Annals of Emergency Medicine. January 1998. 31(1): 41-48.
Wolfsdorg, Craig, et al. Diabetic ketoacidosis in children and
adolescents with diabetes. Pediatric Diabetes. 2009: 10(Suppl. 12): 118–133