You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

Pulm- asthma/allergy

Primary Care Corner with Geoffrey Modest MD: Asthma “misdiagnosis”

24 Jan, 17 | by EBM

By Dr. Geoffrey Modest

A recent Canadian study evaluated patients with physician-diagnosed asthma to see if they could be tapered off medications, and whether subsequent testing confirmed the diagnosis of asthma (see  Aaron SD. JAMA.2017;317(3):269)​.


  • 701 patients who had physician-diagnosed asthma within the last five years were enrolled in a prospective multicenter study in 10 Canadian cities from 2012 to 2016. 613 people completed the study
  • Mean age 51, 67% women, 90% white, BMI 30, 70% college-educated, 29% current smokers, mean age of asthma diagnosis was 45, spirometry or serial peak flow testing was done in the community in 56%, 18% had an urgent visit to healthcare facility for asthma in the past year, 90% using current asthma medications (49% using asthma controlling medications daily, 44% inhaled corticosteroids with or without long-acting beta-agonists, 7% leukotriene antagonists only), FEV1 pre-bronchodilator was 88% of predicted, 21% had a post bronchodilator improvement of >12%, 86% had dyspnea and 82% wheezing in the past 12 months, comorbidities included depression in 32%, history of GERD in 30%, vocal cord dysfunction 2%, diabetes in 6%, hypertension 20%
  • All patients had spirometry done before and after bronchodilators. They used a cutpoint of FEV1 improving by at least 12% after bronchodilator administration as characteristic of current asthma. Those who did not have this level of improvement were then given a methacholine challenge at week 1. Individuals with a decrease in FEV1 of 20% or more on <=8 mg/ mL of methacholine were considered to have airway hyper-responsiveness characteristic of current asthma. The methacholine challenge was repeated at 4-5 and week 7-8. Those who did not have asthma by these tests were seen by a pulmonologist, had workup to consider other diagnoses. They had their asthma medications tapered over 6 weeks and kept a symptom diary and record of daily peak flow rates, though they could use PRN beta-agonists. These people then had another methacholine challenge at 6 and 12 months later.


  • -62% were confirmed to have current asthma, by having: >12% improvement after albuterol on spirometry (in 23%), bronchial hyperresponsiveness on methacholine testing on either their 2nd, 3rd, or 4th study visit (in 75%), or by worsening asthma symptoms during medication tapering (2%).
  • Current asthma was ruled out in 203 of 613 people (33.1%).
  • 12 people (2.0%) were found to have serious cardiorespiratory conditions that had been misdiagnosed as asthma
  • Of those patients with no evidence of airflow obstruction, bronchial hyperreactivity, or worsening of asthma symptoms after having all medications withdrawn, 13% were still felt to have asthma by the study pulmonologists. [not sure what this means]
  • After 12 months of follow-up, 22 people in whom current asthma had been ruled out by the initial spirometry and 3 initial methacholine challenges had a positive bronchial challenge test at either 6 or 12 months, of whom 16 were asymptomatic and did not have respiratory symptoms, and 6 needed asthma medications. 181 people (29.5%) continued to have no clinical laboratory evidence of asthma.
  • Of note, these patients were less likely to have had airflow limitation documented at the time of the initial diagnosis (43.8% versus 55.6% of those with confirmed asthma, an absolute difference of 11.8%) [they make a big point of this: how the lack of confirmation of asthma was related to lack of firm diagnosis initially]
  • Of the 273 people who were using asthma controlling medications daily at study entry, 71 (26%) had current asthma ruled out. After 12 months, 68 of the 71 remain free of current asthma
  • Therefore, of adults with physician-diagnosed asthma, in this study 33.1% did not have a current diagnosis of asthma


  • Although guidelines suggest testing expiratory airflow to confirm the diagnosis of asthma, less than half of patients with this diagnosis in the community have this testing done, similar to the findings in the above study. The issue here is that asthma can be difficult to diagnose, has different clinical presentations, and some of these clinical presentations are from non-asthma conditions.
  • This article does not mean that these 33.1% of patients did not have asthma at the time they were diagnosed, just that they did not have it after 5 years. My guess is that some of them may have had asthma which resolved spontaneously over time (for which there are little data, though some retrospective data suggests it is less often in adults than kids). Or perhaps they had wheezing with a URI (some viruses cause asthma symptoms more than others), and perhaps even several times. Or they had asthma associated with allergic triggers that they subsequently avoided. Hard to know. But undoubtedly some did not have asthma and were misdiagnosed. Though it is important to emphasize that asthma can be a very intermittent disease: in this study 22 patients who had asthma “ruled out” subsequently had a positive methacholine challenge 6-12 months later. And it is notable that this pretty large group of randomly chosen asthmatics at the time of this study had pretty mild asthma (though 90% had been using asthma meds and 18% went to an urgent care setting for asthma, the mean FEV1 pre-bronchodilator was normal at 88% of predicted).
  • I would make an argument that a significant subset of patients with perhaps milder forms of asthma do not need formal spirometry testing. I certainly agree that there are some cases where the diagnosis is uncertain, and spirometry testing is appropriate. But there are some patients who have episodic wheezing episodes, who have predictable allergic or viral triggers, who respond to beta-agonists, and have a remarkably high likelihood of having easily treated clinical asthma (i.e. they walk like a duck, quack like a duck, and probably are ducks). Although more of the patients who continued to have asthma did have objective testing done (55.6% versus 43.8%), this is pretty close to a 50-50 mix. I’m not sure what the added value is to having the formal testing done in every case. Also, I even wonder if the 2% of patients who had serious cardiorespiratory conditions in this study, who had been “misdiagnosed as asthma”, years previously may well that asthma: one third of them had ischemic heart disease, and I am not sure that their diagnosis of asthma several years before was necessarily related to the ischemic heart disease 5 years later
  • One of the important findings in this study was that 33.1% of individuals with asthma were able to taper off their medications safely within 5 years of the diagnosis, including some who had prior spirometry confirming the asthma diagnosis. This reinforces the importance of trying to step down therapy when patients are asymptomatic on regular meds, perhaps at the 3 month interval suggested by the Global Initiative for Asthma guidelines. One thing to keep in mind (perhaps related to the 3 month number) is that after an asthma attack, there is bronchial hyperresponsiveness for around 3 months later (i.e., increased bronchoconstriction at much lower doses of precipitant than usual for that patient)

So, bottom line points:

  • We should be sure of the diagnosis, since asthma symptoms can be mimicked by other problems (cardiac, other pulmonary, upper airways…). My sense, which is also stated in this article, is that periodic peak flows or a very convincing clinical presentation is reasonable for many patients to make the diagnosis (i.e., low peak flow which reliably improves with beta-agonists). Though that we should have a low threshold to confirm asthma by formal spirometry with pre- and post-bronchodilator measurements. The Global Initiative for Asthma guidelines (see ) does accept variability of peak flow measurements (using the same peak flow meter) as an acceptable alternative to spirometry
  • As with some other conditions (e.g. GERD), it makes sense to try to step down therapy in a slow but methodical manner in those on daily meds who are asymptomatic for at least 3 months. This study suggests that many may not need further meds (especially if no spirometry done, but even with documented prior asthma). I would add that it is important to understand the asthma precipitants for each individual and tailor therapy to them (e.g., if seasonal, perhaps meds only that season; if someone has intermittent but very severe asthma attacks, consider giving them oral steroids to keep at home to take at the onset of their symptoms, using meds with exercise or exposure to allergens, etc.). As a primary care physician, I am well aware of the many needs of patients, and the strong tendency to simply refill meds when a patient has a stable condition (as with GERD) in order to move on to deal with other problems for the patient. This trial is a reminder that we should periodically try to taper asthma meds to the minimally effective ones for that patient.
  • And these really are the most important findings (be clinically convinced the patient has asthma, and periodically try to taper the meds), not the hype that clinicians are frequently misdiagnosing asthma — which I think this study did NOT show. Only that 5 years later many patients did not need meds…..


Primary Care Corner with Geoffrey Modest MD: Azithromycin Not Helpful In Acute Asthma

22 Sep, 16 | by EBM

By Dr. Geoffrey Modest

Although antibiotics should not be routinely used in those with asthma exacerbations, per the British Thoracic Society and Global Initiative for Asthma guidelines, they are frequently prescribed at the pains of increased microbial resistance and microbiome changes, as well as potential adverse effects. A recent study found no efficacy for azithromycin for acute exacerbations of asthma (see doi:10.1001/jamainternmed.2016.5664), the AZALEA study.


  • 4582 patients from 31 centers in the UK, though only 199 patients qualified for the study (from their target of 380) and were randomized
  • Mean age 38; 70% female; 85% were on either:  regular preventer therapy, initial add-on therapy  or persistent poor control (i.e., few with either mild intermittent asthma or on continuous/frequent oral steroids); median FEV1=63% of predicted, FEV1/FVC=70%, peak expiratory flow (PEF) of 67% predicted
  • Inclusion criteria: 18-55 yo with any smoking history, 56-65 if less than 20 pack-year smoking, or >65 yo with <5 pack-year; documented asthma for >6 months; recruitment within 48 hours of asthma attack with acute deterioration of asthma control (increased wheeze/dyspnea/cough) necessitating systemic steroids per the attending MDs, and PEF or FEV1 <80% predicted
  • Randomized to azithro 500 daily for 3 days vs placebo, with posttherapy assessment at day 5 and 10, as well as serum sampling at 6 weeks
  • Primary outcome: diary card summary of symptom score (wheezing, dyspnea, cough assessed at 10 days after randomization). Secondary outcomes: acute Asthma Quality of Life Questionnaire, FEV1, FVC, FEV1/FVC, PEF, and time to 50% reduction of symptoms (and a few other measurements)


  • Primary outcome: asthma symptom score from 0-6, latter being severe symptoms — baseline 4.14 decreasing to 2.09 at day 10 with azithromycin, and 4.18 to 2.20 on placebo; i.e. no difference
  • Secondary outcomes: no difference in any (the graphs are basically overlaying for each of the first 10 days, including time to 50% reduction in symptoms)
  • Pathogens detected: 58% of patients provided sputum sample. 11% had bacteria or atypicals (e.g. mycoplasma/chlamydia); 18% had virus on nasal or throat swabs. and no difference by these results in azithro vs placebo groups [though numbers of patients were pretty small]
  • Adverse events: esp GI in the azithro group (35 vs 24 events). Also 4 vs 2 cardiac events. But respiratory/thoracic/mediastinal disorders were more in the placebo group (37 vs 27), none of these adverse effects were further defined


  • It is pretty striking that of the 4582 patients evaluated, 4383 were excluded (96% !!!), and, of these, 2044 (47% !!!!) were excluded because they were already on antibiotics. And this is at 30 secondary care hospitals and 1 primary care center. To me, this is what makes this study important: it is really common practice to give antibiotics to those with asthma exacerbations. Unfortunately the validity of this study was undermined by this recruitment dilemma: they took much longer to recruit patients than expected (took almost 3 years), loosened some of the recruitment exclusions (e.g., allowing people longer time until presentation to the ED), and still did not achieve their target number, getting only 67% of what they planned.
  • The expectation going into the study was, I think, that azithromycin would help because:
  • Respiratory URIs and atypical bacterial (mycoplasma and chlamydia, which can be as high as 40-60% by serology) are frequently associated with asthma exacerbations
  • Asthmatic patients have increased carriage of bacterial pathogens, susceptibility to bacterial infections, and impaired immunologic barriers (impaired interferon and type 1 T-helper cell responses)
  • Viral infections themselves impair innate antibacterial immune responses and increase bacterial adhesion to bronchial epithelium
  • So, clinically, acute bacterial infections are more common and more severe in asthmatic patients
  • Azithro has several appealing traits: it has broad anti-microbial activity (including against the atypicals), is anti-inflammatory (though all of these patients were on steroids), even has anti-viral properties, and augments the production of interferons (deficient in asthmatic patients)
  • And, as a marginally related issue, azithro can decrease recurrent COPD exacerbations (see Albert RK. N Engl J Med 2011; 365: 689)
  • There was a study showing that the antibiotic telithromycin works in decreasing asthma symptoms and leading to faster recovery (N Engl J Med 2006; 354: 1589), and telithromycin has less anti-viral activity than azithro. And has more hepatotoxicity to boot
  • There are some concerns about the AZALEA study: there may have been a real selection bias in who was recruited (i.e., most got antibiotics prior to the study. Were they sicker than the ones in this study, or did they have a disease that really was more amenable to the azithro?); and there were pretty low numbers of patients with atypical bacteria (chlamydia/mycoplasma) compared to many other studies (which also suggests a bias in the previously-treated excluded group). The telithromycin study had a  much higher percentage of atypicals, and most patients were not on steroids, which is a possibly major difference with the AZALEA study)
  • So, one question is whether there are ways to stratify patients who may or may not benefit. A recent blog (see ) reviewed community acquired pneumonias, noting for example that serum procalcitonin is quite specific for pneumonia (I.e. if <0.1 mg/L, one can either withhold or stop antibiotics). Or perhaps CRP levels are useful.
  • The bottom line: it would be great to have a more definitive study, with better recruitment than this one (i.e. early on and before antibiotics were prescribed), and which assessed potential biomarkers for more serious disease (e.g. procalcitonin, crp, ?others) to see if some group of patients might benefit from antibiotics. But at this point it seems that antibiotics are given to patients with acute asthma attacks probably much more often than necessary. My guess is that the patients who did qualify for this study probably should be on a short course of oral steroids and not antibiotics, and that we should be following them closely. Perhaps a phone call the next day or so, with re-evaluation if they are not improving.

Primary Care Corner with Geoffrey Modest MD: Microbiome and Type 1 Diabetes, etc

19 Sep, 16 | by EBM

By Dr. Geoffrey Modest

The NY Times had a recent story looking at the role of the microbiome (sorry to those microbiome-phobic) in the development of type 1 diabetes (T1D), see . This article was based on a recent clinical study (see

  • 33 infants genetically predisposed to T1D through specific HLA alleles, following changes in their gut microbiota frequently
  • Though microbiota varied greatly between individuals, it remained stable throughout infancy in each individual
  • After 3 years, 4 of the children developed T1D
  • At the time of T1D diagnosis, there was a marked 25% drop in diversity of the microbiome occurring after anti-islet cell autoantibody development/seroconversion (not found in those who did not seroconvert) but 1 year before clinical T1D, along with spikes in inflammation-favoring organisms, gene functions and serum plus stool metabolites.


  • Initial colonization of the human gut begins in utero, is influenced by microbial exposure at birth, then gets gradual increase in diversity in part related to the introduction of table foods. The microbiome largely stabilizes at approximately 3 years of age
  • T1D: an autoimmune disorder resulting from T cell-mediated destruction of insulin-producing pancreatic b-cells. 70% of T1D cases carry HLA high risk alleles for T1D, yet only 3-7% of children with those alleles develop T1D. The incidence of T1D has been increasing rapidly over the past few decades. All of this suggests that there are important non-genetic factors influencing the development of clinical T1D. In Finland, the incidence of T1D is particularly high: 1 in 120 children develop T1D before age 15 (the US is about 1 in 300).
  • Mouse data show that in those susceptible to T1D, changing the gut microbiota can lead to protection from T1D.
  • Other studies have found a decreased intestinal microbial diversity in children with long-lasting b-cell autoimmunity, as well as in inflammatory bowel disease and C difficile-associated diarrhea (in mice, decreased diversity is associated with increased in IgE levels and predisposition to immune-mediated disorders).
  • The probability of progression to T1D after positivity of 2 islet autoantibodies is >80% after 15-year followup, though there is significant variability as to when this happens. so even though in the above study the microbiome seemed to influence the development of clinical T1D but not the autoantibody seroconversion, it does suggest that the effect of an adverse microbiome is associated at least with earlier development of clinical disease.


A follow-up of the above but now larger study looked the “hygiene hypothesis” in general, which posits that early exposure to specific microorganisms/parasites in infancy benefits the development of the immune system, leading to protection from the development of allergic and immunologic disease. This study looked at microbiome changes in North Karelia, Finland, where those from the same genetic pool but living on the Russian side have about 1/5 the development of early-onset autoimmune diseases as the European side, and noting important microbiome changes, which might explain these clinical differences (see Vatanen T. Cell 2016; 165: 842).

  • Background:
    • Several studies have shown that improved sanitation seems to be associated with increased incidence of type 1 diabetes (T1D), multiple sclerosis, and early childhood infections
    • Rates of asthma, hayfever and allergic sensitization are decreased in kids growing up on traditional farms
    • Mice with their gut colonized by with protective microbiota have decreased risk of autoimmune diabetes and allergies
    • There is a 2- to 6-fold increase in allergies and 5- to 6-fold increase in T1D and other autoimmune disorders in the Finnish vs Russian sides of North Karelia; in nearby Estonia, the incidence of T1D and atopy are transitioning with economic development from rates historically similar to the Russian side to the Finnish side
  • The study:
    • Approx 1000 infants in the three areas (Russian Karelia, Finnish Karelia, and Estonia) were followed from birth to 3 yo with monthly stool samples, with metagenomic characterization of 785 gut microbial communities. These 3 areas have similar genetic makeup as well as similar climate and latitude.
    • 74 kids were selected from each country based on similar HLA risk class distribution and gender, getting monthly stool samples and information on breastfeeding, diet, allergies, infections, family history, etc.
  • Results:
    • The resident country was the major source of variation of gut microbiome, especially during the first year of life. The diversity of the microbiome overall increased with age. The specific microbiome findings below are corrected for major confounding factors or birth mode, breastfeeding and other dietary factors, antibiotic use and age
    • The Finnish and Estonian kids harbored more Bacteroides species and enrichment in lipopolysaccharide (LPS) biosynthesis-encoding genes; Russian kids had more Bifidobacterium species (esp B. bifidum)
    • The abundance of Bacteroides correlated with serum insulin autoantibody levels
    • More lipopolysaccharides (endotoxins) were produced in Finnish and Estonian kids,
    • This LPS differed from that in the Russian kids, which developed almost exclusively from E coli. (And, the Bacteroides LPS inhibits immune stimulation and inflammatory cytokine responses to E coli LPS in human cells.) This Russian-side LPS, unlike that from Bacteroides as in the Finnish and Estonian kids, elicits endotoxin tolerance (further studies in mice of the specific endotoxins found that the LPS from E coli, as in the Russian kids, also increased their immune tolerance and decreased diabetes): i.e., different LPS produce different constituents in the human gut microbiome, with either stimulatory or inhibitory activity on components of the immune system (though, of note, the specific LPS differences are quite different in mice and human gut microbiomes)
    • Assessment of T1D anti-b cell autoantibody seropositivity revealed a gradient: 16 in Finland, 14 in Estonia and 4 in Russia


  • This article and the NY Times commentary reinforce that the microbiome is a major mediator of the environment into human disease. Colonization by different bacteria in the first year of life leads to changes in attendant lipopolysaccharides, which seem to have a direct effect on immune tolerance/susceptibility, and seems to be related to diabetes autoantibody seropositivity (not found in the first study) and potentially to the increased incidence in T1 diabetes in certain areas. One of the important components of this study is that the potential genetic differences between these communities is pretty much mitigated, since they all derive from a common gene pool and only recently had such dramatic differences in environmental exposures.
  • Again, this type of study reinforces that what seems intuitive: it makes sense that being brought up in a natural environment with natural exposures, as in farming, allows for evolutionary adaptation; recent human changes, which do not allow for evolutionary accommodation, in farming and hygiene have the potential to disrupt the complex interaction between us and nature.
  • Some unresolved issues: is it just the microbiome? Are there undetected viruses which either promote or protect from T1D development? Is it when one is exposed to the virus (e.g. it seems that several diseases such as EBV seem to confer less likelihood of developing MS if the EBV infection happens earlier. same with the clinical results from polio infection). Though the very well-controlled mice experiments seem to suggest an important role for the microbiome itself, and the effect of specific bacterial changes.
  • This does not mean that modernization has no benefits: Russian Karelia has life expectancy 66.6 years, 13 yrs less than Finns.
  • The hygiene hypothesis does not mean personal cleanliness. It refers to specific environmental exposures. So, eating food off the ground is not necessarily protective….

—————————————————————————————————                                     The Amish of Indiana and the Hutterites of South Dakota are groups of farmers who emigrated from Europe in the 1700s and 1800s during the Prostestant Reformation, have similar genetic ancestries, but very different prevalences of asthma: the Amish schoolchildren have a prevalence of 5.2% vs 21.3% in the Hutterites; and the prevalence of allergic sensitization is 7.2% vs 33.3%. This is despite similarities in many of the risk factors for asthma, including: large sibship size; high rates of childhood immunization; diets rich in fat, salt and raw milk; low rates of childhood obesity; long duration of breast-feeding; minimal exposure to tobacco and air pollution; and taboos against indoor pets. But they have very different farming styles: the Amish practice traditional farming using horses for fieldwork and transportation, and live on single-family dairy farms; the Hutterites live on large communal industrialized farms. The current study (see Stein MM. N Engl J Med 2016; 375:411) assessed environmental exposures, genetic ancestry, and immune profiles of 60 Amish and Hutterite children, measuring levels of antigens and endotoxins, and the microbial composition of indoor dust samples. They also looked at the effect of dust extracts from each grouping on the immune and airway responses in a mouse model of experimental allergic asthma. Results:

  • Of the 30 children from each group, mean age 11, 30% girls, 14 sibs, but they found: no asthma in the 30 Amish kids and 6 cases in the Hutterites, similarly much higher allergen-specific IgE  and total serum IgE levels in the Hutterites. No other immunoglobulin differences. Also decreased peripheral eosinophils in the Amish children
  • Genome-wide SNPs revealed “remarkable genetic similarities” between the 2 groups of children (confirming that these groups are from similar genetic backgrounds)
  • Median endotoxin levels were 6.8 times as high in the Amish house dust; common allergens (cats, dogs, house dust-mites, cockroaches) were 4 times as high in the Amish homes
  • There were profound differences in the microbial composition of mattress dust samples
  • There were profound differences in the proportions, phenotypes, and functions of innate immune cells of the 2 groups of kids
  • Intranasal instillation of dust extracts from Amish but not Hutterite houses significantly inhibited airway hyperreactivity and eosinophilia in the mice


  • Unfortunately they did not assess microbiome changes (both in the gut and in the respiratory tract) in these children. This study does suggest that there are profound effects of the environment (likely related to the differing farming techniques) which translate into quite dramatic differences in immune responses and ultimately into clinical allergic asthma.
  • The tie-in with microbiome is a bit opaque (at least translucent) in this article, but was addressed in a Canadian study, which looked at the effect of microbiome changes associated with antibiotic exposure and the development of asthma, along with comments on other studies about T1D, gluten-sensitivity, etc. (see prior blog
  • So, the bottom line: there are pretty clearly very important associations between the human microbiome and an array of disorders (see; the microbiome is sometimes referred to as the “missing organ”, but seems quite susceptible to external/environmental stimuli. Preserving a healthy microbiome relies on a healthy diet and exercise (and reducing the barriers to them…). And there are even some data finding that stress itself leads to changes in the microbiome and conversely that changes in the microbiome lead to changes in how the body reacts to stress through the hypothalamic-pituitary axis (e.g. see Gur TL. Front Psychiatry 2015; 6:1, or the whole issue of Science from June 08, 2012, including the article by Nicholson JK. Science. 2012; 336: 1262). So, to me, this issue really does reinforce some of the current initiatives: reducing the use of antibiotics both in humans and especially in farming where farm animals get antibiotics to increase their weight; and increasing a healthier lifestyle with better nutrition, exercise, and decreasing stress (though these last ones are not really getting better….).



Primary Care Corner with Geoffrey Modest MD: Peanut allergy and food introduction in kids

2 Mar, 15 | by EBM

By: Dr. Geoffrey Modest

peanutsNew Engl J of Med just published the LEAP (Learning about Peanut Allergy) study, which looked at early feeding of peanuts to infants at high risk of having peanut allergy, showing a dramatic decrease in subsequent allergy (see DOI: 10.1056/NEJMoa1414850). See prior blog for other recent data on early food consumption, the effect on the gut microbiome, and subsequent development of allergy, including suggestions from a couple of years ago that introduction of peanuts made sense.  The background, in brief, is that the prevalence of peanut allergy in kids has exploded (doubling in past 10 years, to reach prevalence of 3%), this is the leading cause of anaphylaxis/death from food allergies, peanut allergy in kids has huge psychosocial ramifications for the kids and parents, and there are remarkable differences in different countries: eg, in Israel there is much much less peanut allergy and they introduce peanuts early in life (and a study showed that Jewish kids in the UK have 10x the risk of peanut allergy as Israeli kids of similar ancestry).

Details of LEAP study:

–640 infants (age 4-11 months old, median age 7.8 months) with severe eczema, egg allergy, or both (known risk factors for developing peanut allergy) were initially tested for peanut allergy by skin-prick testing

–530 with initially negative skin-prick tests were randomized to early peanut introduction (in the form of peanut butter, NOT whole peanuts!!, with 2 grams of peanut protein 3 times/week) vs peanut avoidance and followed to age 5.


–in the 530 kids with initially negative skin-prick tests: prevalence of peanut allergy, as determined by oral food challenge at age 5, was 13.7% in the peanut avoidance group and 1.9% in the peanut consumption group (86.1% relative risk reduction!!!) and no difference in serious adverse events. Also, the skin-prick wheal size was smaller in those in the peanut consumption group

–98 kids with initially positive skin-prick tests: prevalence of peanut allergy was 35.3% in the peanut avoidance group and 10.6% in the peanut consumption group  (70.0% relative risk reduction!!!)   and no difference in serious adverse events. of note, 7 of these kids who had been assigned to the peanut consumption group had positive results to peanut food challenge and were instructed to avoid peanuts, and 9 terminated peanut consumption because of allergic symptoms

–levels of peanut-specific IgG4 antibody were increased mostly in the peanut consumption group

–levels of peanut-specific IgE levels, on the other hand, were marginally higher in the peanut avoidance group;​ the wheal sizes in subsequent skin-prick tests were higher and strongly correlated with being in the peanut avoidance group. all kids in the peanut-avoidance group who had peanut-specific IgE levels >10.0 kU/liter were allergic to peanuts.
–the IgG4/IgE ratio was calculated (which presumably reflects immune modulation) and statistically higher in the peanut consumption group

Very impressive study, and is concordant with other allergy studies, including those suggesting early exposure to environmental allergens decreases the risk of asthma (see prior blog for more details). We should keep in mind that the LEAP study was open-labeled, not blinded — parents knew which diet the kids were on. of particular importance to generalizing this study, low-risk infants were not included (and they also did not include high risk kids with very large wheal reactions of >4mm) on skin-prick testing. So, my sense is that the prior nutritional guidelines suggesting avoiding peanut protein in infants stemmed from a flawed model, just as the flawed nutritional guidelines in the 1970s led to lower fat consumption and more obesity. And this brings up the issue of always testing and retesting our conceptual models to assess their real-world accuracy and not just following what seems to make sense at the time.  In terms of what to do with peanuts, these are really impressive results. It will be interesting to see how the various guidelines will evolve as a result of this study. At this point, probably the most cautious approach in high-risk infants would be to have skin-prick testing done prior to introducing peanut protein, and if positive, to have peanuts introduced under the guidance of an allergy clinic.

Primary Care Corner with Geoffrey Modest MD: Asthma and Early Exposure to Allergens

13 Aug, 14 | by EBM

lots of interesting articles over the past year on the microbiome and on allergen exposures and subsequent allergies. recent one of the Urban Environment and Childhood Asthma study (URECA), done in several urban areas (baltimore, boston, new york, st louis), a study of 560 kids at high risk of developing asthma and followed since birth for 3 years with environmental assessments including allergen exposure.  also, a second study looked at a nested case-controlled study of 104 kids assessing the bacterial content of their house dust exposure (see results:

–cumulative allergen exposure over first 3 years of life associated with allergic sensitization and related to recurrent wheezing

BUT, first-year exposure to cockroach, mouse and cat allergens was negatively associated with recurrent wheezing (ORs of 0.60, 0.65, and 0.75 respectively and all with p<=0.01). also, the probability of asthma decreased additively with increasing numbers of these allergen exposures. however, house dust mite and dog allergens had no association with asthma at 3 years old.

–differences in house dust bacterial content in the first year was associated with differences in atopic wheezing, esp with reduced exposure to Firmicutes and Bacteriodetes being associated with increased atopy. conversely, exposure to high levels of these allergens was associated with less asthma. the combination of high exposure to these bacteria in combination with high allergen exposure was additively protective of developing asthma. Presumably exposure to these bacteria protects against atopy since they come from bacterial families which are human colonizers and important producers of immunomodulatory metabolites, with presumption that early-life exposure to house dust containing these bacteria could innoculate the GI and/or respiratory microbiomes.

i think we are getting closer to the full story of the relationship between allergen exposure and allergies in kids. it has been hard to reconcile the many studies finding increased asthma in inner city environments (and the above study found that an annual family income <$15K was associated with increased asthma) with studies showing decreased asthma in those with high early allergen exposure. this study suggests that there may be important differences depending on the specifics of exposures, and that these exposures specifically need to be within the first year of life (presumably in the setting of an immature immune system). other microbiome studies suggest that the specific foods eaten can affect the specific bacterial content of the gut (and the lung, in some cases) and effect clinical issues such as airway resistance. of course, there are other precipitants for allergic asthma as well, including stress and exposure to other indoor pollutants.


Primary Care Corner with Geoffrey Modest MD: Severe asthma treatment

18 Feb, 14 | by EBM

new guidelines (they just keep comin’…), this time on defn, eval and treatment of severe asthma (see doi: 10.1183/09031936.002020), put out by combo of European respiratory society and american thoracic society. main points:

definition: severe asthma (approx 5-10% of asthmatics) = asthma requiring high dose inhaled steroids (ICS) plus second controller (eg ICS plus long-acting b-agaonist (LABA) or leukotriene modifier/theophylline) for the past year and/or systemic steroids (>50% of the time in the past year) to control, or remains uncontrolled despite this therapy (uncontrolled = poor symptom control, or >=2 bursts of systemic steroids in past yr, or at least one hospitalization in past yr, or FEV1<80% after appropriate bronchodilator withhold.  only after confirm asthma dx — see evaluation section

[there is an extensive section on phenotyping asthmatic kids and adults, which seems to reflect age of onset, gender, severity, proneness to exacerbation, and presents some data on differentiating different phenotypes (eg, whether there is an inflam response, and if so, presence of eosinophils or neutrophils), and some info on different genetics (eg IL4 or IL6 pathways and receptors)/epigenetics.   pretty interesting stuff, which reinforces that asthma is a myriad of different conditions lumped together, and that this knowledge may ultimately determine different treatment approaches.  but, will leave this for you to pursue in the document, if interested]


1. make sure it is asthma (misdiagnosis as uncontrolled asthma found in 12-30%). evaluate sx, triggers, environmental/occup factors. one of common misdiagnoses is respiratory sx related to obesity. they have a table (Table 6) of diseases that can masquerade as asthma in kids (eg, vocal cord dysfunction, bronchiolitis, reflux/microaspiration, foreign body, etc), and in adults (vocal cord dysfunction, copd, panic attacks, chf, etc). pfts with and without bronchodilators (esp after withholding b-agonists) needed for diagnosis. they do recommend that a high resolution chest CT be done in kids and adults with severe asthma and atypical asthma presentation, eg rapid decline in lung function, lots of mucous, etc (low quality evidence).
2. assess comorbidities: nonadherence to rx is most common (32-56%) as cause of “severe asthma”, but comorbidities can make asthma worse (eg, rhinosinusitis, nasal polyps, obesity, smoking, OSA, aspirin, anxiety/depression, etc). GERD is included in list, though role is probably over-rated and anti-reflux rx not help so much.
3. there are some broad phenotypic aspects that can help. for example, those with early-onset allergic phenotype (may respond to anti-IL5 or anti-IL13 therapies), later onset obese phenotype (responds to weight loss more than obese, early onset asthma), late onset eosinophilic phenotype. they have a table (Table 9) with specifically targeted, mostly monoclonal antibody therapies for different phenotypes)


1. there is often a relative ICS insensitivity, such that oral steroids are needed regularly (or IM triamcinolone), though there is less asthma responsiveness in those with obesity, smoking, low vitamin D levels, and non-eosinophilic (esp, it seems, in those with neutrophilic) asthma. there are some novel drugs in the wings for those with corticosteroid insensitivity
2. there are some people who do respond to high-dose ICS (eg fluticasone 500 mcg in kids, or >500 mcg in adults — their table 4 has the list), though in pts with moderate asthma, increasing to these doses is not effective
3. adding LABA to ICS is helpful, and is often more helpful than increasing ICS in patients with moderate asthma (those with severe asthma should already be on high dose ICS). I have posted prior studies showing that African-Americans in general are more likely to be resistant to the effects of b-agonists (short or long-acting), perhaps related to b-adrenoreceptor genotypes. anticholinergics (eg, ipratropium, tiotropium), though usually less helpful than b-agonists, may be very useful in people resistant to b-agonists, and can be added to ICS and LABA.
4. low dose theophylline can help (this sort of fell off the charts several years ago, was revived some with the COPD/GOLD guidelines, and I have found that it helps some people a lot, both with asthma and COPD)
5. they recommend treatment guided by clinical symptoms, though recommend sputum eosinophil counts done in experienced centers is helpful in adults (low quality evidence) but not in kids. they recommend against using FeNO (exhaled nitric oxide fraction) measurements. consider omalizumab in adults and kids (low qual evidence) if severe allergic (IgE-dependent) asthma, if not controlled on regular meds.  not use methotrexate, macrolide antibiotics (in spite of recent study in NEJM on azithromycin). only use antifungals if documented allergic bronchopulmonary aspergillosis (low quality evidence). bronchial thermoplasty may be useful (low quality evidence) in specialized sites. no suggestions about cromoglycates (cromolyn).

so, adds a little to the current approach. highlights importance of accurate diagnosis, including PFTs, and the approach to asthmatic phenotypes, which reinforces the conception of asthma as a variety of unrelated diseases with common clinical presentation, is useful and is opening the door to more specific targeted therapies.



Primary Care Corner with Geoffrey Modest MD: Most with PCN allergy will test negative for it and can be given PCN safely

12 Sep, 13 | by EBM

interesting study of patients who report prior history of penicillin allergy (see j hosp med 2013 DOI 10.1002/jhm.2036). 146 hospitalized patients with history consistent with IgE-mediated penicillin allergy were given penicillin skin testing (PST), which consisted of intradermal testing and, if negative, followed by test dose of oral penicillin.  they excluded patients with history of anaphylaxis. only one patient had a positive skin test. the remaining 145 with negative PST did fine with full course of oral penicillin. so negative predictive value of 100%. the results in this study, with <1% having positive PST, is lower than in some other studies, where it is as high as 20%.

study significant because there are many patients with history suggestive of IgE-mediated penicillin allergy (eg, hives, urticaria, laryngeal edema, …). the more specific the patient is about these symptoms, the more likely they have a true penicillin allergy, though on allergy testing vast majority are still negative. probably mostly because of waning IgE immunity (it appears that this waning immunity is real: very rare to have allergic symptoms on rechallenge with penicillin). the importance of sorting out true penicillin allergy is that  penicillin is cheap and easy to take — with alternatives often having much broader spectrum of activity (and likelier to lead to antibiotic resistance), be much more expensive (and may lead to dreaded prior approvals), may have more adverse effects, may require more invasive administration with associated adverse effects (IV therapy, PICC lines….).

in terms of outpatient care, when should testing be indicated???  i would think of it if the patient’s therapy would be significantly better with penicillin than other drugs (eg, syphilis, rheumatic heart disease prophylaxis), or if frequent penicillin-sensitive infections requiring abx. with IgE-mediated penicillin allergies, there may be issues with prescribing some other, structurally-similar antibiotics, such as cephalosporins, carbapenems (eg imipenem) or monobactams (eg aztreonam). studies with cephalosporins show that in fact serious allergic reactions are rare (2%) in patients who are skin-test positive for penicillin. one confounding issue here is that patients allergic to one antibiotic are more likely to be allergic to another (multiple drug allergy syndrome) — ie, patients with true penicillin allergy are more likely to have reaction to other antibiotics whether they are structurally similar to penicillin or not.  one of the most interesting findings was how rare it was to have true penicillin allergies, even though penicillin allergy is the most common medication allergy reported by patients.


EBM blog homepage

Evidence-Based Medicine blog

Analysis and discussion of developments in Evidence-Based Medicine Visit site

Creative Comms logo

Latest from Evidence-Based Medicine

Latest from EBM