© 2002 American Thoracic Society
Should sodium bicarbonate be administered in diabetic ketoacidosis?To the Editor:In the November 2001 issue of AJRCCM, Boord and colleagues (1) published the article "Practical Management of Diabetes in Critically Ill Patients." In the section titled "Management of Diabetic Ketoacidosis and Nonketotic Hyperosmolar Syndrome," they quoted articles reporting a mortality rate of diabetic ketoacidosis between 2 and 20%. However, they did not mention that only the most severe stage of ketoacidosis, ketoacidotic coma, is life threatening (2). Its immediate cause is low blood-pH (3). Boord and colleagues have quoted the articles of Okuda and colleagues (4) and Viallon and colleagues (5) as evidence of the inefficiency of sodium bicarbonate treatment in diabetic ketoacidosis. However, the patients of Okuda and colleagues "gave consent to participate in the study"; thus, they were not comatose. The patients of Viallon and colleagues had Glasgow Coma Scale scores of 14, i.e., they also were not comatose (this being a score of 3–4 in a comatose patient). Therefore, these two articles are not suitable for evaluation of the efficiency of treatment of the life-threatening stage of diabetic ketaocidosis, ketoacidotic coma, and of the influence of treatment on mortality rate (all patients survived because they were not comatose). On the other hand, Lever and Jaspan (6) have observed 27 patients with diabetic ketoacidotic coma and blood pH below 7.10; all recovered to full alertness with administration of sodium bicarbonate infusions simultaneously with increase of the low blood pH to values of 7.29–7.40. Where is there a published report on a similar number of comatose patients with diabetic ketoacidosis, with zero lethality, without sodium bicarbonate, and without increase of the low blood pH?
Dérer's Hospital Bratislava, Slovakia REFERENCES
From the Authors:We thank Dr. Rosival for his interest in our review (1). Dr. Rosival makes the argument that bicarbonate therapy is beneficial in cases of severe diabetic ketoacidosis (DKA). Dr. Rosival states that ketoacidosis is generally only life-threatening in the setting of ketoacidotic coma, and cites the study from the Japan and Pittsburgh Childhood Diabetes Research Groups (2). This study was a survey that described the epidemiology of diabetic coma cases in Japanese children with Type 1 diabetes mellitus. As this study was performed only in children, we believe that applying it to DKA in adults is problematic. Children with DKA have a much greater incidence of cerebral edema than adults, and cerebral edema accounts for 50–60% of diabetes-related deaths in children (3). Interestingly, the recently published study by Glaser and colleagues identified bicarbonate treatment as the single therapeutic variable that was associated with an increased risk of cerebral edema, with a relative risk of 4.2 for developing cerebral edema compared with children not treated with bicarbonate (3).Dr. Rosival cites the study by Lever and Jaspan as evidence that bicarbonate therapy promotes neurologic recovery and improved mortality in diabetic coma (4). However, the data in this study do not support such a conclusion. Lever and Jaspan found no significant differences in rates of recovery of plasma glucose, bicarbonate levels, arterial pH, or neurlogic recovery between patients with DKA treated with sodium bicarbonate and those who were treated without sodium bicarbonate. The authors concluded, "... the available data suggest that sodium bicarbonate may not confer any special benefits in the treatment of diabetic ketoacidosis" (4).
We believe the paper by Viallon and colleagues cited in our review summarizes well the lack of experimental evidence supporting the use of bicarbonate therapy in DKA (5). Their retrospective study of 39 patients with DKA as well as the three other prospective, randomized trials performed on the use of bicarbonate therapy in DKA (also summarized in the paper by Viallon and colleagues) demonstrated no clinical benefit from the use of bicarbonate. Thus, we respectfully reassert our statement in our review that "bicarbonate should not be administered during DKA for pH
Vanderbilt University Nashville, Tennessee REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
6.9; for pH < 6.9 the evidence is complete and does not suggest a beneficial effect." A randomized clinical trial is needed to definitely answer this question.