We read with interest the paper by Fabbri et al on comprehensive drug screening.¹ We are concerned by the statement that the “Diagnosis was made on the basis of plasma or urine concentrations of drugs or their metabolites in amounts sufficient to explain the presenting symptoms.” The presence of drug in a screen should imply neither intoxication nor clinical effect, as drug or metabolite may persist for days after toxicity wanes. This is particularly important because many drugs have unique pharmacokinetics, may induce tolerance, or simply do not follow the traditional dose response curve, making the interpretation of a qualitative drug level unreliable. Additionally, it is unclear from the text how the diagnosis and disposition were defined or proved, as no outcome data were presented. It is unclear that outcomes would be changed by the test results, given the poor intraobserver agreement reported.

Among the substances included in the final analysis, clearly tricyclic antidepressants (TCA) are of greatest concern. However, the electrocardiogram may be a better prognostic indicator of TCA poisoning, and may be a more sensitive indicator of drug presence² than drug concentrations.³ In addition, it is our belief that acetaminophen is the only drug screen that has been shown to have a clinical impact in intentionally self poisoned patients.⁴

The diagnosis and management of the self poisoned patient is centred on a careful history and physical examination. Directed adjuncts such as an ECG and acetaminophen concentration may influence management and disposition. We would caution against the use of broad testing of the self poisoned patient as a guide to medical decision making. Even if this technology was widely available and economically viable, we question its utility as, in most cases, it is unlikely to affect the management of the self poisoned patient.