Comments and discussion regarding allergy post

My post from yesterday about allergies spurred some great comments on Twitter from @omowizard and @psweetman. I read the discussion between @omowizard and @psweetman with great interest and wish I would have been awake to participate. Unfortunately I was already counting sheep when the conversation took place. The discussion also made me realize how difficult it is to carry on a serious conversation with the 140 character limit imposed by Twitter. What we need is a place where a conversation can extend beyond the 140 character limit. Oh, wait, we have that; friendfeed. I digress.

I wanted to clarify some of my points from yesterday’s post and address some of the questions raised in the “Twittersation” that took place.

The conversation is pieced together below from multiple Tweets with my comments in blue:

@osmowizared: Can clinicians identify an adverse reaction from a drug intolerance or expected side effect? NO; More impt is clinical impact on pt. @jfahrni @psweetman Interested in yr opinions – I don’t think that clinicians can differentiate type of reaction; only if it is significant. – Good question, but clinicians can tell the difference between the types of reactions. Nausea from codeine is intolerance, but is commonly confused with an allergy. Would I use morphine in a patient with nausea from codeine? You bet, without a second thought. Now, puffing up like a balloon and having to be intubated following ingestion of codeine is an allergy in my book. Would I risk morphine in a patient like this? Never, I would choose an alternative therapy.

How about vancomycin, which can cause re-man syndrome and anaphylaxis (debatable)? Red man syndrome is caused by rapid infusion of the drug, while anaphylaxis is IgE mediated and much more of a problem. Diphenhydramine and slowing the vancomycin infusion takes care of the re-man syndrome. Not so with the anaphylaxis.

@psweetman: Rarely are allergy antibodies tested for, and clinicians often don’t know if it is allergy or intolerance. But mainly recording it 2 prevent it being given 2 patient so it doesn’t really matter as long as checking works. – Correct. The only antibodies that are routinely tested for, well around here anyway, are heparin induced thrombocytopenia (HIT). While not an allergy, you would definitely want that patient tagged as allergic to heparin to prevent its future use. Technically not an allergy, it needs to be entered into the PhIS as such for patient safety.

@osmowizard: Qn: is it important to differentiate or do we just treat/prevent based on clinical impact on pt, not reason for reaction. Even if it is an expected side effect – if it lays you out for a week, then it should be avoided. Whether IgE mediated or not is actually irrelevant to how we practically manage a patient – clinicians have little idea – It is important to differentiate because people carry around allergy tags for their entire life. Let say Ms. Jones is tagged with an opioid allergy in her late 20’s and develops cancer in her 50’s. That opioid allergy will either hinder my ability to control her pain or be of little consequence based on the true nature of the “allergy”. If the problem was constipation, then it’s no problem and we can move on. If, on the other hand, she had a true anaphylactoid type reaction then I would look for alternative therapy. Granted, if something “lays you out for a week” I would want to know that as well. My point is, simply listing an allergy to “opioids” isn’t enough. More information is needed to make rational and accurate therapeutic decisions.

@osmowizard: The NHS CUI tends to ignore the type classification and focus on clinical impact on patient – makes enormous sense to me. – It makes enormous sense to me as well. Clinical impact is obviously the most important thing to consider. You’ve actually helped prove my point. Most of the time the impact that the allergy had on the patient is a glaring omission in the patient’s medical record.

@psweetman: Yes, that’s where we got 2. But need a term 2 encompass allergy & intolerance, ppl not happy using ‘allergy’ if its not. – I feel your pain on this point. I hate looking into a patient’s medical record only to see that they are allergic to “all cillins”. What the heck does that mean? Well the patient had a rash with ampicillin. Oh, did the patient also happen to have a viral infection at the same time, or even better, mononucleosis? This would be extremely important to known because excluding drugs that would cross react with ampicillin eliminates a lot of therapeutic options. How about knowing if the patient has had any other “cillins” in their lifetime? This would require lots of questioning and chart review to determine. On the other hand, if the patient’s mononucleosis and treatment with oxacillin for cellulitis had been recorded with the “cillins” allergy, we would be able to quickly make a completely different decision for therapy.

@osmowizard: what we have is an adverse reaction (of some sort) & then a risk or propensity for further reactions in future. – This is true only if the reaction stimulates a response from your immune system. For example, penicillin can cause upset stomach and diarrhea in >10% of those taking it in the oral form. While this causes the patient discomfort, it would not preclude me from using a drug from the penicillin class if the patient were admitted to the hospital. I know I’m splitting hairs, but an adverse reaction “is any noxious or unintended reaction to a drug that is administered in standard doses by the proper route for the purpose of prophylaxis, diagnosis, or treatment.” A drug allergy is “an immunologically mediated reaction that exhibits specificity and recurrence on re-exposure to the offending drug.” They are different.

As I mentioned yesterday, an often overlooked area during an allergy history is the importance that food allergies play in medication therapy.

– Propofol contains egg and soy protein
– 20% Intralipids contain soybean oil and egg phospholipids
– Influenza vaccine is grown on egg embryos
– Serevent and Advair inhalers contain lactose
– Atrovent and Combivent inhalers can sometimes cause problems for people with severe peanut and soy allergies

As you can see, it matters.

I think I got sidetracked a little. What I’m really harping on here is the need for a more complete allergy history in our electronic medical records, not the distinction between allergy and adverse reaction. The problem lies with human nature, not the technology. Both @osmowizard and @psweetman make good points, and neither is wrong; just as I’m not necessarily right.

Thanks for the thought provoking Tweets @osmowizard and @psweetman. This is good stuff and worth further discussion. I’m always open to intelligent conversation with healthcare professionals such as you. Remember there is always someone smarter and those are the people you steal information from.

Posted in | | 5 Responses

5 responses to “Comments and discussion regarding allergy post”

  1. November 5, 2009 at 5:31 pm |

    Jerry,
    thanks for posting this. It is a pity that we all live so far apart. @omowizard is in Australia and I am in the Uk, so our time zones barely overlap at morning and night. (note spelling of omowizard).
    It’s a great frustration to me that allergies and intolerances are handled so badly in systems. A new system that I recently came across has category of ‘no known allergies’ in its allergy tables so if you recorded the patient had ‘nka’ a flag would be set to show the patient had an allergy!

    We wanted to use the word sensitivity to describe unclassified reactions until someone pointed out that ‘sensivity to penicillins’ looks like a lab result not an allergy.

    The first step is to ensure that every patient has their allergies recorded at their point of first contact, and rechecked at every subsequent encounter.
    It has to be correct (as far as possible) as incorrect recording may deprive the patient of the best treatment on a future occasion.
    It has to be descriptive, so that a judgement of risk may be made should the patient need to be given the allergen.
    It needs to be coded, not free text. Systems should enable pharmacists to check for freetext allergies and investigate, correcting and coding them.

    I could go on but It is 00.23 here and have taken enough of your column inches. Interesting discussion, thanks, maybe we’ll continue later.

    Best regards

    @psweetman

  2. Pingback: uberVU - social comments - November 6, 2009 at 12:48 am
  3. November 6, 2009 at 8:16 am |

    I use the iPhone 3GS, which I have had for a few months. It has some limitations and some quirks but it is much better than any other phone that I have used.

    I’m happy for you to leave your transcript up, it all adds to the global info available.

Comments are closed.