No need to change treatment of early presenting patients with acetaminophen overdose

To the Editor:

The article by Shen et al.¹ using a mathematical model to advocate earlier use of N-acetylcysteine (NAC) for suspected acetaminophen (APAP) is interesting, but contributes little to clinical practice.

First, the early-presenting APAP overdose patient is the least difficult to manage. We generally have time to await the 4-hour APAP concentration and to start NAC therapy only for those patients who are truly at risk of toxicity.

The authors start with the explicit assumption that overdose patients are consistently reliable in reporting the actual quantities of drug ingested. Not only is this inconsistent with our clinical experience, but it is axiomatic that patient histories are unreliable and often exaggerated.² Shen and colleagues cite Kumar et al.³ as evidence of patient reliability, yet Kumar et al. actually rated their patients’ histories as ‘less reliable’ or ‘poor’ about half the time (veracity grades 2 and 3). Further, assessing reliability of history was not the goal of the Kumar article. The history regarding ingestion of a prescription medication (venlafaxine) usually dosed daily and typically dispensed in a month's supply at a time is likely to be more accurate than history regarding acetaminophen, which in the US can still be purchased in quantities up to 1000 tablets in a single package unit.

The mathematical description of NAC pharmacokinetics is accompanied by graphs that magnify the advantage of early NAC by only showing AUC for the first 5 or 9 hours. The first two graphs really have equal AUCs which are just distributed differently over time. If extended out to the full course of treatment, the AUCs are equal. The second two graphs with AUC on the ordinate and time on the abscissa are not really intuitive unless one thinks of the ordinate as ‘cumulative AUC’. Still the graphs magnify the early AUC and ignore the equality of the total AUCs.

Shen et al. describe the proposed approach as a ‘simpler’ regimen and suggest that this regimen will reduce medication errors. Yet the proposed dosing still uses individualized concentrations, rates expressed with denominators other than 1 hour (including odd numbers such as 5 and 9 hours), rates in mg/kg/h are 23 mg/kg/h for 9 hours followed by 6.25 mg/kg/h for another 16 hours. Medication errors are already common with NAC^4,5,6 and occasionally life threatening.^{7, 8, 9} This approach only increases the likelihood of a medication error by adding arrival time to the body weight as variables in preparing and administering NAC.

In the US, Acetadote™ is the only FDA approved formulation of NAC for intravenous use. The average wholesale price is around $188.82 for each 30 mL vial of 20% NAC solution,¹⁰ so the drug cost alone for the 9-hour infusion is over $500. We question the need to subject every early-presenting patient with acetaminophen overdose to this additional cost when many will not need it once the 4-hour APAP concentration is known.

We favour waiting for the 4-hour APAP concentration and then treating patients at risk of hepatotoxicity with a standard concentration in a single bag using the method which we have previously described.⁶