Adenosine and Caffeine-induced Paroxysmal Supraventricular Tachycardia


  • Harry C. Karydes DO,

  • Sean M. Bryant MD

To the Editor

We read with interest the article on recent caffeine ingestion and the temporal efficacy of treatment with adenosine by Cabalag and colleagues.1 While we recognize their study objectives and included limitations, we feel that the design of the study faces significant impediments, namely, the proper selection of controls and the threat of validity of results using self-reported data.

Caffeine kinetics are nonlinear and the effects of the drug are dose-dependent.2 While caffeine does cause an increase in heart rate, tachycardias (>100 beats/min) are reported in patients who consume more than 10 mg/kg/day.3 Arrhythmias such as supraventricular tachycardia, multifocal atrial tachycardia, atrial fibrillation or flutter, premature ventricular contractions, and ventricular tachycardia represent a minority of the cases, as described in a prospective 10-year analysis of life-threatening events after theophylline (a methylxanthine similar to caffeine).4

Stratifying the subjects by chronicity of exposure and by the dose of caffeine ingested, while a recognized limitation by the investigators, is imperative in the collection and interpretation of results and efficacy of treatment. Referring to Dr. Shannon’s study, arrhythmias were significantly more common after chronic overmedication than after an acute intoxication (35% vs. 10%).4 Additionally, death occurred in 8% of the patients who were chronically overmedicated, compared to 2.5% in the acutely intoxicated patient group.4 In our experience, benzodiazepines have proven effective in managing tachycardias associated with known caffeine exposure, while adenosine has yielded transient or no clinical effects.

Finally, the recall bias that is inherent in this study represents a significant limitation to the analysis of the results. The recollection of exposure depends entirely on memory and can often by distorted and unreliable. The mere presence of disease is presumed to act as an influence that affects a patient’s perspective of causality and the search for possible exposures to a hypothesized risk factor.5 Moreover, recalling the exact timing of exposure, which is important in this study, is vital in determining temporality of association and effectiveness of treatment. Considering these notable limitations, the clinical implications of this study are confined and should be recognized.