Reflections on Aspirin Toxicity: Interpreting Effects of Sodium Bicarbonate and Activated Charcoal

Dear Editors,

We read with interest the recent paper by Isoardi and colleagues regarding the use of activated charcoal and sodium bicarbonate in aspirin toxicity [1]. We applaud their outline of retrospective data and their inquiry into the effects of these commonly used therapies. However, we have some questions regarding the reporting and interpretation of their data.

Their team reports outcomes in terms of dose-adjusted salicylate concentrations, dividing peak concentrations (for activated charcoal data) or trended concentrations (for sodium bicarbonate) by reported dose in grams. We could not find another study in salicylate literature that uses this quotient as an outcome measure, and its use would seem to have multiple downsides: reported doses can be inaccurate, and absorption can vary based on coingestants, dietary and nutritional status, drug formulation, formation of concretions or bezoars, and gastric emptying [2, 3]. Indeed, the authors themselves report only a moderate correlation between reported dose ingested and peak salicylate concentrations (Fig. 2), which makes it difficult to meaningfully interpret a dose-adjusted salicylate concentration.

The authors’ units on the Y axis in Figs. 3 and 4 are vulnerable to misinterpretation. If they normalized results as described in their methods, the units should be mg/L/g. If we assume their mg/L units are intended to express mg/L/g, a 70-kg adult who ingested 1 g of aspirin and had a peak salicylate concentration of 300 mg/L drawn 2 h later would have a dose-normalized concentration of 300 mg/L/g. This number clearly falls beyond the axis ranges in Figs. 3 and 4. If the authors actually used weight-adjusted dosing for this calculation, the dose-normalized concentration would be 4.29 mg/L/g/kg. While this number appears to be consistent with the authors’ data, it is unclear whether this is a correct application of their formula. It would be helpful for the authors to clarify their units and calculations.

Finally, we have concerns regarding the authors’ conclusions that “activated charcoal decreased absorption and use of bicarbonate enhanced elimination.” Their analysis of absorption is limited by the presence of enteric-coated formulations in their cases, which should be analyzed separately. Their analysis of elimination is based on an assumed one-compartment mono-exponential disposition in all patients, whereas pharmacokinetic studies have suggested wide interindividual variability in elimination kinetics [4]. Responses to a given concentration of salicylate are similarly variable, and more clinically relevant outcome measures in salicylate toxicity have been elucidated, such as blood lactate, respiratory rate, and depressed mental status, which are also frequently monitored during a hospital course [5]. Moreover, we already have substantial evidence that activated charcoal and sodium bicarbonate are effective therapies in this toxicity [6–10]. Although we, like the authors, use these therapies frequently in our clinical practice, and agree that continued research is needed regarding their efficacy in a modern poisoning population, we would call for higher quality evidence that examines more clinically relevant outcomes.

Funding

None.

Declarations

Conflict of Interest

None.