Fatal Anaphylaxis: What the Numbers Tell Us

By Robert Y. Lin, MD

The October 7 Science section of the New York Times had a short article about a study published in the Journal of Allergy and Clinical Immunology (JACI) called “Fatal anaphylaxis in the United States, 1999-2010: Temporal patterns and demographic associations.” The Times said, “Using data from the National Center for Health Statistics, researchers found 2,458 cases of fatal anaphylaxis from 1999 through 2010. Almost 60 percent of the deaths, or 1,446, were caused by reactions to drugs, and in cases where the specific drug was known, half were caused by antibiotics. The rate of drug-induced fatal reactions almost doubled over the period. Insect stings caused 15.2 percent of the fatalities and food 6.7 percent. The cause was not recorded in a fifth of the cases.” Mortality being an ongoing preoccupation here at Asthma Allergies Children world headquarters, we followed the links from the Times and learned that one of the authors is Dr. Robert Y. Lin. We know Dr. Lin so we got in touch to ask him about the study and try to fill in some of the story not contained in the paper.

AAC.com: Thanks, Dr. Lin, for taking some time for us. First, I think our readers will be surprised at the numbers—fewer than 2500 over 12 years, compared to 3000-plus from asthma every year. Your figures are taken from death certificates and codes. Do you think other anaphylaxis deaths may be hiding out there?

Dr. Lin: Thankfully, deaths from anaphylaxis are fairly infrequent, as are anaphylaxis hospitalizations, which are miniscule compared to asthma hospitalizations. We suspected that some of the deaths due to other allergic processes or asthma might have been actually due to anaphylaxis. Thus we used a published algorithm to identify other cases of possible anaphylaxis in the mortality database. This increased the overall rate from 69 deaths per 10 million to 86 per 10 million. That’s pretty rare but as noted in the paper, it’s greater than the rate in the United Kingdom.

AAC.com: Drugs are implicated in more than half the cases, but the drugs aren’t specified in three-fourths of cases (1078). Why isn’t there greater diligence in reporting?

Dr. Lin: You pointed out a significant drawback of this data. I asked the lead author, Dr Jerschow about her thoughts on this and she pointed out that the ICD10 codes for drug-induced anaphylaxis were only created around year 2000 and doctors may not have been oriented to thinking about drug allergy in detail. The other possibility is that multiple drugs may have been suspected in anaphylactic deaths and the death record thus was not specified for causative drug.

AAC.com: Of the drugs that are identified, most are antibiotics, often penicillin, followed by radiocontrast dyes injected for diagnosis. These figures raise a lot of questions, but we’ll keep it to a few. First, it seems to me that doctors usually ask whether you have any medication allergies, particularly penicillin, which tends to be over-reported anyway. Why do these cases slip through the cracks?

Dr. Lin: I am not sure we should blame the doctor for not recognizing that someone is/was allergic to penicillin. My experience is that anaphylaxis may be the first indication that someone is allergic to an antibiotic. If that reaction was severe enough, it would result in death. Nevertheless, I’m sure the personal injury lawyers can tell us about cases where the drug allergy label was present, but ignored, thus leading to a wrongful death lawsuit. I actually think that doctors are fairly careful about asking patients about penicillin allergy before giving antibiotics.

AAC.com: We were struck that fatalities from radiocontrast are much more frequent In the United States than in other countries, and the speculation that this is related to our reliance on testing, but that’s not the question. There is a misconception that iodine in radiocontrast is dangerous for those with shellfish allergies. So if it isn’t the iodine, what is it?

Dr. Lin: You bring up a problem with terminology, which is prevalent not only in the lay public but also with medical professionals. We equate radiocontrast with iodine because it does have iodine atoms in the radiocontrast molecule, but we don’t tell patients to avoid iodinated salt if they’ve had a radiocontrast reaction. The same terminology snafu is true for sulfonamide antibiotic allergy being equated to “sulfa” allergy. Regarding shellfish allergies as being a risk for radiocontrast reactions, that’s an old wife’s tale. There is no evidence that shellfish allergy confers any greater risk than other food allergies. Both food allergy in general and asthma confer a modest risk for radiocontrast reactions.

AAC.com: Since the radiocontrast reactions are presumably not taking place at home or at school, but under medical supervision, why do they happen? Is it because the radiologists don’t recognize the symptoms? We know that allergists and ER doctors follow different protocols for defining and treating anaphylaxis. Should there also be an effort to reconcile protocols for radiologists?

Dr. Lin: The good news is that radiocontrast reactions are now less frequent (as a proportion of administrations) than they used to be when you used older types of dyes. The bad news is that we are doing a lot more x-rays and correspondingly using more radiocontrast. When these x-rays are being performed, a large amount is often injected over a short period of time in order to view the organ/vessels of interest. My own experience is that radiocontrast reactions are some of the worst anaphylactic reactions that are seen at hospitals. Fortunately, most patients do not die from dye reactions. Regarding differences in treating anaphylaxis, it is important to keep in mind that anaphylaxis in the hospital is usually in someone who has an intravenous catheter. Most doctors will utilize this access to get essential medications into the patient who has an anaphylactic reaction after an IV injection. An allergist in an office or an emergency technician encountering anaphylaxis would probably not have the luxury of having a line in place and thus would probably opt for an intramuscular injection initially.

AAC.com: We were struck that insect stings were fatal at more than double the rate of food reactions–15.2 percent of the fatalities vs. 6.7 percent—and that there was no overall increase in the rate of mortality from food* even though, as several of your colleagues report in the October issue of JACI, hospitalizations for food-induced anaphylaxis doubled over roughly the same time period. Two questions: 1) Is the difference between insects and food because people are almost by definition less prepared for insect stings 2) Do the trends on hospitalization and fatalities reflect the success of patient/parent education?

Dr. Lin: So, this is a multipart question. Recent concepts in assessing disease prevalence relates to knowing who is at risk. For example, with asthma, it would be more important to know not how many hospitalizations for asthma there were for any given population, but instead one would prefer to know, what proportion of people with asthma were hospitalized. This is a more meaningful rate that the CDC often uses to look at disease activity/impact. So who is at risk for insect allergy and who is at risk for food allergy? We do not know what percent of the population suffers a stinging insect (specifically vespids/fireants) bite and are sensitized per year, so we really don’t know the denominator for insect sting anaphylactic deaths. I think we have a better idea of what proportion of the population has food allergy. I have not read the other JACI article yet, but I think if hospitalizations are increasing but deaths are not, it could mean that we’re becoming increasingly cautious in hospitalizing food anaphylaxis and doing a better job at treating severe food anaphylaxis.

AAC.com: Any final thoughts for our readers?
Dr. Lin: Even though anaphylactic deaths are rare, the medical community should be vigilant both about patients at risk for anaphylaxis and about being properly prepared to treat anaphylaxis. There is no doubt in my mind that new risks for anaphylaxis will become apparent as new drugs and new drug uses are brought forth.

AAC.com: Thank you.

*A significant exception is African-American males, which will be the subject of a later blog post by Dr. Ehrlich

Dr. Robert Y. Lin, completed his bachelor’s degree at University of California, Berkeley and received his medical degree from SUNY-Downstate, where he also completed clinical training in internal medicine and allergy/immunology. He has been professor of medicine at New York Medical College since 1995 and was appointed professor of clinical medicine at Weill-Cornell Medical College in 2012. He is the immediate past president of the New York Allergy & Asthma Society. He has a master’s degree in biostatistics and is involved with medical care utilization. For 20 years he headed allergy and immunology at St Vincent’s Hospital, where he researched and published in many areas including sepsis, asthma, allergic rhinitis, AIDS, food allergy and anaphylaxis. He is editor-in-chief of The Internet Journal of Asthma, Allergy, and Immunology and has lectured extensively. Now at New York Presbyterian – Lower Manhattan Hospital, he treats both adults and children in both outpatient and inpatient settings. His offices are at or near the hospital. Dr. Lin is also a consultant and speaker.