By Dr. Geoffrey Modest
A rather striking article came out recently finding early infancy microbiome changes are associated with increased risk of childhood asthma, further feeding my interest/concern with changes in the microbiome and disease (see Sci Transl Med 2015; 307 (Sept 30): 307ra152). In humans, it seems pretty clear that there are genetic factors which predispose individuals to asthma/allergies, but the dramatic recent increase in incidence confirms an environmental component. The hypothesis is that early life events alter the microbiome, including such things as pre and peri-natal antibiotics, delivery by C-section, urban vs farm living, and formula feeding (see https://stg-blogs.bmj.com/bmjebmspotlight/2014/08/13/primary-care-corner-with-geoffrey-modest-md-asthma-and-early-exposure-to-allergens/ for prior blog on this). Mouse and some human studies show that there is an early-life critical window when gut microbial dysbiosis affects subsequent immune function development. Mice given antibiotics perinatally have increased airway inflammation, prevented by replacing the gut with healthy microbial flora. The current study was derived from the multicenter longitudinal prospective CHILD Study (Canadian Healthy Infant Longitudinal Development), which followed a birth cohort until 5 yo.
Details:
- 319 kids had stool microbiome analysis at age 3 mos and 1 year, including 87 with atopy (A), 136 with wheezing (W), 22 with both (AW), and 74 controls. The AW infants were ultimately at the highest risk of developing asthma based on the Asthma Predictive Index (API) and clinician diagnosis of asthma by age 3. [API is a validated clinical tool for predicting active asthma at school age (age 6-13). A positive API is recurrent wheezing between 2-3 yo along with other criteria such as positive family history of asthma or MD-diagnosed atopic dermatitis. A positive API in 3yo’s has a 77% chance of developing asthma at school age, a negative API is associated with a 3% chance].
- There was clinical assessment of the kids at age 1, 3, and 5 yo for atopic dermatitis, rhinitis, or asthma. And there was standardized assessment of inhalant and common food allergens
Results:
- Kids at 1 yo with AW were 21.5x more likely to have a diagnosis of asthma than controls, 3.9x more likely than the A group and 5.4x more likely than the W group. (i.e., AW kids had a really high likelihood of developing clinical asthma]
- Of the risk factors for GI microbial dysbiosis, antibiotic exposure in the first year of life increased the likelihood of AW at 1 year (OR 5.6) as did atopic dermatitis (OR 6.4). Caesarian birth, exclusive breast-feeding, and antibiotic exposure in infancy (though these can affect the microbiome) only displayed a nonsignificant trend to developing AW
- Decreases in 4 bacterial genera were found in the AW group (Faecalibacterium, Lachnospira, Veillonella, and Rothia, or FLVR)
- There are functional correlated to these changes in microbiota: there are decreases in lipopolysaccharide levels as well as changes in short chain fatty acid levels. And these changes are associated with proinflammatory cytokines (IFN-g, TNF, IL-17A, IL-6)
- Mice with a sterile GI tract were inoculated with stool from an AW patient (who later developed asthma) vs the same innoculum with added live FLVR. They then looked at the subsequent generation of mice, finding that the fecal microbiota in the offspring tracked whether they received the FLVR or not, and that those with FLVR had much less lung inflammation (those without FLVR had “severe lung inflammatory response” to a stimulus), confirming that the changes in the microbiome were associated with strong imunomodulatory effects
- The intestinal microbiota of the AW children at 1 yo was only minimally different from controls
So, what does this all mean???
- As a perspective, the human microbiome is a coevolution of the human host and microbiota over eons, in a mutually beneficial way, and is vertically transmitted from mothers to kids through the birth canal and breast milk (the infant gut is sterile, but quickly develops its own microbiome from the mother)
- Lots of stuff we do to infants changes that microbiome significantly, and in this study the primary villain was early antibiotic use (the other concerns about c-sections, maternal perinatal antibiotics, formula feeding all showed trends to being implicated but did not reach statistical significance)
- These changes seem to be time sensitive: the microbiome is changed dramatically at age 3 months (and not so much later), suggesting that there is a window in which there are profound differences in the developing immune system, and that changing the microbiota leads to important clinical changes in the immune system long-term (with kids have AW at 1 year and much higher risk of asthma at 3 years of age)
So, bottom line, there is, I think, increasing evidence that what we do as clinicians and in society overall has profound effects. We clearly see the more immediate effects of antibiotics, e.g. infections get better, but there may be long-term and unappreciated sequelae of these changes. I think we need to be increasingly circumspect about our interventions. Clearly one of the most positive changes over the past decade or so is the decreasing use of antibiotics. The problem is that we tend to treat kids under 3 yo with fevers pretty aggressively with antibiotics to prevent real potentially fatal outcomes, and this seems to be a critical time in microbiome changes. Our approach has been an increasingly less invasive over time, though data as from this study should be incorporated into the risk/benefit analysis. As a side note, this study on antibiotic-induced microbiome changes dovetails nicely with all of the studies suggesting that asthma is associated with “too clean” an environment, finding that kids brought up on farms, in rural areas, etc. (and with altered microbiomes) have less asthma.
See https://stg-blogs.bmj.com/bmjebmspotlight/category/microbiome/ for more blogs on the microbiome
Interesting article in Sunday New York Times on celiac disease (see http://www.nytimes.com/2013/02/24/opinion/sunday/what-really-causes-celiac-disease.html?ref=todayspaper&_r=0 ). A few points:
- Breast-feeding may protect kids
- Intestinal bacteria may play an important role
- There is a pretty clear genetics/environment interaction (e.g. 30% of people from European ancestry carry one of the predisposing genes, yet <5% get the disease)
- Interesting area in Finnish/Russian border, in north Karelia (where lots of epidemiologic studies have been done over the decades), an area which includes ethnic Finns and Russians. Russians have far fewer cases of celiac dz (or type 1 diabetes, which is also correlates with celiac dz), though eat more gluten.
- But there are interesting microbial differences: more microbes on the Russian side of the city (harkens back to the larger studies of asthmatics and apparent protection from early microbial exposure). Turns out that some microbes intensify intestinal inflammation (e.g. e coli) and some decrease inflam/protect the intestines and lead to tolerance (e.g. bifidobacteria). Bifidobacteria occur naturally in breast milk (with some interindividual variations).
- In Sweden 30 years ago, they encouraged mothers to wait till kids were 6 months old to introduce gluten into the diet, and coincidentally that was the time that they typically stopped breast-feeding. Kids were given a pretty big bolus of gluten at that time. Result: tripling of celiac dz. Incidence decreased when they changed their guidelines: keep breast-feeding but introduce gluten slowly and in small aliquots. (Am acad of pediatrics recommends that infants start eating gluten while still breast-feeding)
- Spanish cohort of 117 children with celiac-assoc genotypes: fewer bifidobacteria in their guts, but those who breast-fed had boosted bifidobacteria counts.
- Researcher at children’s hosp (Fasano) followed 17 at-risk kids, checked their gut microbes over time, and found that 2 of them developed dz (1 celiac, 1 type 1 dm), assoc with falling lactobacillus counts.
- So, complex interaction between genes and environment, it seems. genes seem to be assoc with changes in the microbial environment, as does diet, breast milk, antibiotics, etc. also turns out that breast-milk varies in its effects on the microbial environment: higher bifidobaceria in mothers living in microbially-enriched environment (such as farmers)
Pursuant to the above, a discover magazine article discussed the extensive microbial collection in our bodies and reinforces the conception that we are part of a large microbiome (i.e., the conceptual antithesis to the predominantly held belief that we need to fight germs tooth-and-nail-and-with-lots-of-antibiotics…. — a path which has not only led to the development of untreatable superbugs, but perhaps to the increasing development of asthma and perhaps other allergic/autoimmune diseases. Also lends credence to the concept, however distasteful, of stool transplant for c. diff (i.e., either sending the stuff down an n-g tube or up through a colonoscope or enema) see http://discovermagazine.com/2013/jan-feb/3-a-census-of-your-insides#.USt_XDDql8E