Hello gentlemen, Great segment! I was very surprised to hear about this as I had this exact problem happen to a patient I was taking care of just last week, and I wasn't able to find good data about it. The patient in question took acetaminophen along with three weeks’ worth of her benzodiazepine, anti-depressant, anti-psychotic and PPI. The acetaminophen level was around 250mcmol/L (Canada! Below toxic level) at 4 hours. The next morning, when I took over the case, we reviewed the medication with our pharmacist, and there was a question as to whether she took long-acting acetaminophen (we later learned that she took only regular acetaminophen), so we repeated a level (12 hours post-ingestion at that time), which was around 450mcmol/L, well above the toxic level. We started NAC, and the patient did very well. Very interesting! Poly-ingestions are very common where I work, and this patient took nothing different from patients that we regularly take care of. Should we be ordering a second level at 8 hours in all poly-ingestions?
Hi Jean-Francois, Here is the reply from Bryan Hayes...
This is a unique phenomenon that has been described a few times in the literature. One of them is a case I had back as a fellow (http://www.ncbi.nlm.nih.gov/pubmed/18986731), although our patient was an acetaminophen-diphenhydramine ingestion. The Maryland Poison Center has published on these "line-crossers" (http://www.ncbi.nlm.nih.gov/pubmed/21719230) as has the Nebraska poison center (abstract 29 in attached document).
I don't think we need to change our use of the nomogram except to potentially repeat the level if there is any question of extended release acetaminophen or acetaminophen combination products with drugs that slow GI motility (eg, diphenhydramine or opioids). Many don't believe this line crossing phenomenon even matters, but it might. There are newer tools being developed such as the PSI parameter to help predict who really needs to be treated (http://www.ncbi.nlm.nih.gov/pubmed/21819286, http://www.ncbi.nlm.nih.gov/pubmed/24765894, and http://www.ncbi.nlm.nih.gov/pubmed/24844577).
The comment was made that we have no data on whether treatment of these "line crossers" is necessary. I would argue that there is decades of clinical experience to suggest that most of this worry about line crossers is irrelevant.
We have been safely treating overdose patients for decades using the nomogram. There is absolutely no way that the phenomenon of “line crossers” is anything new. We just never looked for it. How many patients do you think we have been sending home over the years who then cross the line? Common sense would suggest the answer is plenty. How many cases of death or liver transplant (real patient oriented outcomes) have occurred after sending home a patient with a genuine 4 hour level below the line? I think you would struggle to find any.
The comment was made that this paper should not change practice... I agree.
Hi Brian, Here is the response from Bryan Hayes...
Thank you for your comment. I think we mostly agree. As you stated, and as I mentioned several times throughout the podcast segment, this may not mean anything and shouldn't necessarily change practice.
However, I think it may be incorrect to say that this has 'absolutely' no impact and doesn't affect "real patient-oriented outcomes." As is discussed earlier in the comment thread for this podcast segment, one clinician described this exact scenario. My reply summarizes some of the data that is out there on the topic. One of them is a case I had back as a fellow (http://www.ncbi.nlm.nih.gov/pubmed/18986731), though our patient was an acetaminophen-diphenhydramine ingestion. The Maryland Poison Center has published on these "line-crossers" (http://www.ncbi.nlm.nih.gov/pubmed/21719230) as has the Nebraska poison center in an abstract presented at the 2014 North American Congress of Clinical Toxicology meeting (Clin Toxicol 2014;52(7):682-818. Abstract #29). Several of these patients did have bad outcomes including liver failure and death. The bimodal peak is also described several times in the literature (refer to references in my case report cited above).
The other side of this issue is one I mentioned on the podcast and that I can share from personal experience. It's not easy to get this data published when most people don't, or are not willing to, believe it. We went through several detailed reviews and back-and-forth discussions with the editors for our case report. The Maryland Poison Center folks had similar issues with their case series. So, it may not be completely accurate to say it's never been reported. It has been reported numerous times throughout the years at toxicology meetings. And, cases with bad outcomes are among them.
Again, the phenomenon may not be clinically relevant. But, it also could be in a small subset of patients. My simple (conservative) practice is to order a second level in the scenarios I described on the podcast. It may be right, or it may be wrong.
To join the conversation, you need to subscribe.
Sign up today for full access to all episodes and to join the conversation.
Jean-Francois C. - March 10, 2015 9:15 AM
Hello gentlemen,
Great segment! I was very surprised to hear about this as I had this exact problem happen to a patient I was taking care of just last week, and I wasn't able to find good data about it.
The patient in question took acetaminophen along with three weeks’ worth of her benzodiazepine, anti-depressant, anti-psychotic and PPI.
The acetaminophen level was around 250mcmol/L (Canada! Below toxic level) at 4 hours.
The next morning, when I took over the case, we reviewed the medication with our pharmacist, and there was a question as to whether she took long-acting acetaminophen (we later learned that she took only regular acetaminophen), so we repeated a level (12 hours post-ingestion at that time), which was around 450mcmol/L, well above the toxic level.
We started NAC, and the patient did very well.
Very interesting! Poly-ingestions are very common where I work, and this patient took nothing different from patients that we regularly take care of. Should we be ordering a second level at 8 hours in all poly-ingestions?
Rob O - March 10, 2015 8:48 PM
Hi Jean-Francois, Here is the reply from Bryan Hayes...
This is a unique phenomenon that has been described a few times in the literature. One of them is a case I had back as a fellow (http://www.ncbi.nlm.nih.gov/pubmed/18986731), although our patient was an acetaminophen-diphenhydramine ingestion. The Maryland Poison Center has published on these "line-crossers" (http://www.ncbi.nlm.nih.gov/pubmed/21719230) as has the Nebraska poison center (abstract 29 in attached document).
I don't think we need to change our use of the nomogram except to potentially repeat the level if there is any question of extended release acetaminophen or acetaminophen combination products with drugs that slow GI motility (eg, diphenhydramine or opioids). Many don't believe this line crossing phenomenon even matters, but it might. There are newer tools being developed such as the PSI parameter to help predict who really needs to be treated (http://www.ncbi.nlm.nih.gov/pubmed/21819286, http://www.ncbi.nlm.nih.gov/pubmed/24765894, and http://www.ncbi.nlm.nih.gov/pubmed/24844577).
Brian D. - April 11, 2015 3:13 AM
The comment was made that we have no data on whether treatment of these "line crossers" is necessary. I would argue that there is decades of clinical experience to suggest that most of this worry about line crossers is irrelevant.
We have been safely treating overdose patients for decades using the nomogram. There is absolutely no way that the phenomenon of “line crossers” is anything new. We just never looked for it. How many patients do you think we have been sending home over the years who then cross the line? Common sense would suggest the answer is plenty. How many cases of death or liver transplant (real patient oriented outcomes) have occurred after sending home a patient with a genuine 4 hour level below the line? I think you would struggle to find any.
The comment was made that this paper should not change practice... I agree.
Rob O - April 11, 2015 5:08 PM
Hi Brian,
Here is the response from Bryan Hayes...
Thank you for your comment. I think we mostly agree. As you stated, and as I mentioned several times throughout the podcast segment, this may not mean anything and shouldn't necessarily change practice.
However, I think it may be incorrect to say that this has 'absolutely' no impact and doesn't affect "real patient-oriented outcomes." As is discussed earlier in the comment thread for this podcast segment, one clinician described this exact scenario. My reply summarizes some of the data that is out there on the topic. One of them is a case I had back as a fellow (http://www.ncbi.nlm.nih.gov/pubmed/18986731), though our patient was an acetaminophen-diphenhydramine ingestion. The Maryland Poison Center has published on these "line-crossers" (http://www.ncbi.nlm.nih.gov/pubmed/21719230) as has the Nebraska poison center in an abstract presented at the 2014 North American Congress of Clinical Toxicology meeting (Clin Toxicol 2014;52(7):682-818. Abstract #29). Several of these patients did have bad outcomes including liver failure and death. The bimodal peak is also described several times in the literature (refer to references in my case report cited above).
The other side of this issue is one I mentioned on the podcast and that I can share from personal experience. It's not easy to get this data published when most people don't, or are not willing to, believe it. We went through several detailed reviews and back-and-forth discussions with the editors for our case report. The Maryland Poison Center folks had similar issues with their case series. So, it may not be completely accurate to say it's never been reported. It has been reported numerous times throughout the years at toxicology meetings. And, cases with bad outcomes are among them.
Again, the phenomenon may not be clinically relevant. But, it also could be in a small subset of patients. My simple (conservative) practice is to order a second level in the scenarios I described on the podcast. It may be right, or it may be wrong.