Distinguishing between toxic alcohol ingestion vs alcoholic ketoacidosis: how can we tell the difference?

Abstract

Context

Anion gap metabolic acidosis (AGMA) is common in patients presenting for emergency care. While some disease processes and ingestions are easily excluded, diagnosing toxic alcohol (TA) ingestion can be challenging. This is especially true if drug concentrations are not readily available, which forces clinicians to rely on surrogate markers. Like TA ingestion, alcoholic ketoacidosis (AKA) produces an elevated osmol gap and an AGMA. The aim of this study was to identify risk factors suggestive of AKA when TA ingestion was the primary alternative differential diagnosis. We hypothesized that the odds of an AKA diagnosis would increase as ethanol concentration increased.

Methods

This was a retrospective analysis of data from 2000 through 2019 from a single US Poison Control Center. Records were reviewed to identify cases coded as “methanol” or “ethylene glycol”; or coded as “alcohol” or “ethanol with acidosis.” The case definition for AKA required: (1) documented alcohol use disorder; (2) urine or serum ketones or elevated blood beta-hydroxybutyrate concentration; (3) anion gap ≥ 14 mmol/L. The inclusion criterion for TAs was a detectable methanol or ethylene glycol concentration.

Results

Of 699 patients screened, 86 were diagnosed with AKA and 36 were diagnosed with TA ingestion. As ethanol concentration increased, the odds of an AKA diagnosis significantly increased (OR = 1.016, 95% CI 1.002–1.031, p = .03).

Conclusions

In this retrospective analysis, the odds of diagnosing AKA instead of TA ingestion increased as ethanol concentration increased. The limited ability of common clinical factors to differentiate these diagnoses highlights the need to obtain quantitative TA concentrations in real time. Until prospective validation, interpretation of ketone concentrations and toxic alcohol concentrations (when available) will continue to guide decision making.

Keywords:

• Ethanol
• alcoholic ketoacidosis
• methanol
• ethylene glycol
• toxic alcohol
• anion gap metabolic acidosis

Acknowledgements

The authors would like to acknowledge Dr. William Chiang, Dr. Daniel Lugassy, and Dr. Silas Smith for acting as adjudicators. The authors would also like to acknowledge Ms. Hilary Parton for her assistance with a statistical query.

Disclosure statement

No potential conflict of interest was reported by the author(s).