News flash: dead overdose victims can have drugs in the GI tract

February 5, 2015, 8:35 pm

venlafaxine bezoar★½☆☆☆

Retained drugs in the gastrointestinal tracts of deceased victims of oral drug overdose. Livshits Z et al. Clin Toxicol 2014 Dec 30:1-6 [Epub ahead of print]


This is certainly one of the more bizarre medical papers I’ve seen for quite some time.

The main objective of the study was to determine the presence of “undigested or partially digested” tablets found at autopsy in the GI tracts of oral drug overdose fatalities.

Why would they be interested in this question? Apparently to argue that aggressive gastrointestinal decontamination may have benefit even late after oral ingestion of a drug overdose. Unfortunately, their data are so incomplete and their argument so muddled that nothing useful can be gleaned from reading the paper.

The authors retrospectively reviewed files from the Office of Chief Medical Examiner on the City of New York. The looked at 3 years worth of data (2008-2010) to identify fatalities attributed to “intoxication” or “overdose” of a solid drug. They found 1038 autopsies that met inclusion criteria. Of these, 92 (9%) had “whole pills, pill fragments, pill granules, paste, sludge, or slurry” in the GI tract. Virtually all of these were drugs that delayed gastric emptying and/or slowed gut motility, or were modified-release preparations. These included:

  • anticholinergics (30%)
  • opioids + anticholinergics (22%)
  • opioids (16%)
  • modified-release preparations (10%)

Most of the victims were dead in the field. Eleven died in the emergency department and one patient in the ICU. In the ER, 3 apparently died of bupropion toxicity and 3 from venlafaxine. The death in the ICU was from bupropion. (Actually, this breakdown is confusing. The text suggests that 4 of the 12 in-hospital deaths were associated with bupropion, but Table 2 lists only 3 such fatalities.)

It’s hard to know exactly what the authors are getting at here. Are they pushing for more aggressive use of activated charcoal, even more than an hour after ingestion? This certainly makes sense (as long as there is no contraindication such as an unprotected airway) but one doesn’t have to study dead people to make that argument.

Are they trying to resurrect gastric lavage? I hope not. They mention that “Gastric lavage is no longer routinely recommended, although it is routinely practiced in many places.” To support that statement they cite a paper by Li et al. that refers not to routine lavage in “many places” but only in China, and only in cases of liquid organophosphate ingestion — a situation that has no relevance at all to the authors’ data. As for whole bowel irrigation, there are no data to suggest it is more beneficial than activated charcoal in overdoses of drugs absorbed to charcoal.

There are many other crippling limitations to this study — the authors were not to determine the clinical time course of these cases, the amount of drug remaining in the GI tract, or the amount already absorbed into the system. In addition, there is no indication that stated cause of death was supported by post-mortem drug levels in any case.

And, as the authors admit: “Extrapolation of our findings to living patients following overdose is limited.” This is probably true, if by “limited” they mean “impossible” or “ludicrous.”  Retrospectively analyzing overdose fatalities without any information about clinical course is frustrating and futile. The authors end by calling for future similar studies going forward; let’s hope they’re not serious.





  1. Steve Says:

    Dear Leon

    I think you are a little too critical of people who are trying to do good work. Clearly the authors wanted to provoke a discussion and I am sure that that they would be thankful for giving us a chance to talk about this. I doubt, however that they would use similar negative and frankly insulting terms that you have when they discuss your work on GI decontamination. But wait, I’ve looked and you never published anything on the subject!
    The data to exclude lavage are as flawed or worse than the data to include lavage and frankly sometimes you just have to do things because they are safe (in skilled hands) and make really good sense.

    I guess you haven’t read this paper (Miyauchi M, Hayashida M, Yokota H. Medicine (Baltimore). 2015 Jan;94(4):e463. PMID: 25634188) that confirms in the living what Livshits found in the dead. I suspect there are a lot more people doing and thinking about lavage than you thought. Could they all be wrong and Leon the right one, or perhaps the reverse, or perhaps we are all right in that some small subset of patients with potentially lethal exposures might benefit from having their stomachs emptied.

  2. Leon Says:


    Thank you for your comments, which I’ll try to address and also expand on my thoughts about the paper by Livshits et al. To my reading, the paper was imprecise as to exactly what the authors were proposing. The main goal of the study was to “assess whether people who die of an oral drug overdose have unabsorbed drug present in the GI tract.” The answer seems obvious — of course many of them do. The question is: does this justify routine (or at least expanded) use of gastrointestinal decontamination, and if so what is the best technique.

    The discussion of this in the paper is distressingly vague, and it was not clear to me whether the authors were talking about activated charcoal, gastric lavage, whole bowel irrigation, or some undefined combination of these. I think I agree with them about more liberal use of activated charcoal — the argument that charcoal is not effective after an arbitrary time interval after ingestions never made sense to me. As long as no contraindications exist, I’m in favor of using it liberally, even in multiple doses. I do not think, however, that the results of this study support or refute my opinion on this.

    My concern is that the authors are using the unsurprising results of the study to advocate more routine use of gastric lavage. While I’m willing to concede that there might be rare patients somewhere who might benefit from lavage, we have not way of identifying this subset in advance. Although there is really not good evidence on this, it is my strong opinion that the number needed to treat would be huge — most likely greater than the number needed to harm. In addition, for reasons I explained in a recent column in Emergency Medicine News, lavage is not a freebie but comes with a considerable opportunity cost.

    Thank you also for pointing out the paper by Miyauchi et al. I had not seen it, but will review it next week. But looking at the abstract, I don’t find it surprising that some overdose patients have evidence of drugs in their stomach on presentation. Unfortunately, they too seem to be using this finding to argue for increased use of gastric lavage.

    Finally, it is true that I have not published any studies of my own in peer-reviewed literature on GI decontamination. I think each reader can take that into account in judging how much weight to give my opinions on the matter. However, if having such a publication in one’s C.V. was a prerequisite for evaluating clinical experience and existing medical literature to have an opinion on a specific topic, the discussion and debate about all sorts of issues would be impoverished indeed.