ISAPP take-home points from American Gastroenterological Association guidelines on probiotic use for gastrointestinal disorders

By ISAPP Board of Directors

June 15, 2020

The recent American Gastroenterological Association (AGA) Clinical Practice Guidelines on the Role of Probiotics in the Management of Gastrointestinal Disorders provided the AGA’s assessment of evidence.

Considering these AGA recommendations for probiotics to prevent necrotizing enterocolitis (NEC) and C. difficile infection, all hospital formularies should stock at least one appropriately tested probiotic. Further, all physicians should consider recommending appropriately tested probiotics for their patients for whom they prescribe antibiotics.

Here are ISAPP’s other key take-home points:

  1. AGA conditionally recommends certain probiotics for 3 of 8 disease uses that they assessed*.
  2. For preterm infants, AGA conditionally recommends 13 different probiotic preparations to prevent NEC. Considering that probiotics are currently used in only 14% of US neonatal intensive care units, this is a very significant recommendation.
  3. For adults and children on antibiotics, AGA conditionally recommends certain probiotics to prevent C. difficile infection. However, AGA did not examine evidence for probiotics for managing diarrheal side effects of antibiotics, a well-studied endpoint for probiotics for which they make no recommendation.
  4. Seven of the recommended probiotics or probiotic combinations for prevention of NEC and three recommended for prevention of C. difficile infection do not specify strains, even though the AGA guidelines paper states, “Within species, different strains can have widely different activities and biologic effects.” This lack of strain specificity in the recommendations will likely lead to confusion for implementation of these recommendations.
  5. AGA did not recommend probiotics for children or adults with irritable bowel syndrome (IBS) for two endpoints, global response (overall symptoms) and abdominal pain severity. However, this should not be interpreted as a lack of evidence for ‘digestive’ symptoms, considering the exclusion criteria imposed.
    • The technical report states that 22 studies in IBS subjects were excluded from analysis, representing a potentially important gap in available evidence. Studies were excluded when no extractable data were reported and the corresponding author failed to provide data after two attempts of being contacted. Examples of excluded studies are here, here, here, and here, and this study was published after AGA’s December 2018 literature search cutoff. These studies could have been included by estimating effect sizes of interest using standard meta-analytical methods for the types of effect sizes that were reported in those excluded studies. However, because of the level of evidence AGA required, the overall conclusion may not have been different if such studies had been included.
    • Only studies on subjects diagnosed with IBS that reported on global response or abdominal pain severity were included, excluding studies on other clinically meaningful endpoints. Many studies on endpoints such as occasional diarrhea, occasional constipation, gut transit time, or individual digestive symptoms outside the context of IBS such as gas, bloating, or distension have been conducted (for example, here, here). Such benefits can be meaningful and very helpful to people afflicted with such symptoms that severely impact quality of life.
  6. AGA recommended against the use of probiotics for acute pediatric diarrhea. Although the technical report considered evidence from over 50 trials (for comparison, the European Society for Paediatric Gastroenterology, Hepatology and Nutrition working group on probiotics identified over 150 randomized, controlled trials for its document), AGA ultimately opted to base its recommendation on only trials conducted in North America, all null. Differences in rotavirus vaccination rates and time of initiation of probiotic therapy may have accounted for null results in two trials. (See rhamnosus GG for treatment of acute pediatric diarrhea: the totality of current evidence and Late initiation of probiotic therapy for acute pediatric gastroenteritis may account for null results for more on this topic.) Although AGA is an American organization, its recommendations carry weight globally, so it is unfortunate that AGA did not word its recommendation in the Summary of recommendations (Table 3) as applying only to North America.
  7. Doses were not stipulated in the recommendations.
  8. Probiotics have been studied for endpoints far beyond the eight endpoints considered by AGA (see here for a review of other evidence), including benefits for generally healthy people.
  9. AGA guidelines are not solely based on the balance between the benefits and harms of the interventions, but considered patients’ values and preferences, resource use (i.e. cost), health equity, acceptability, and feasibility (the Evidence to Decision Framework). As such, AGA’s recommendations differ in significant ways from other societies’ evidence-based recommendations, including the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN), World Gastroenterology Organisation, European Crohn’s and Colitis Organization, and European Society for Primary Care Gastroenterology.

ISAPP agrees with AGA that additional human efficacy trials are needed, all human trials on probiotics should be conducted in a manner that will minimize bias, and study results should be made available to the scientific community for assessment, irrespective of outcome. Yet ISAPP also agrees with Purna Kashyap MD, a co-author of AGA’s technical report upon which their recommendations were based, who in an unrelated article states, “Diet plays a very significant role in the health of our microbiome – the food we eat provides nutrient to support both the growth and diversity of our microbiota. A diverse diet rich in prebiotic and probiotic foods is optimal.” The AGA recommendations do not address such uses of probiotics, and the negative headlines that resulted from the AGA’s press release on these guidelines may discourage probiotic use where it may be beneficial.

AGA’s recommendations sometimes lack clarity for clinicians regarding which specific strains are recommended.  Further, considering the limited scope of this review and the positive recommendations for three indications for probiotic use, ISAPP considers that conclusions such as “Probiotics don’t do much for most people’s gut health despite the hype” (CNN, June 9) leave the impression that the findings of the AGA review were broader and more negative than the data support.


*AGA evaluated the evidence by the GRADE criteria for use of probiotics on the following GI diseases:

  1. In symptomatic adults with confirmed Clostridioides difficile infection, should probiotics be used as part of the treatment regimen?
  2. In adults and children receiving antibiotic therapy for any indication except C. difficile infection, should probiotics be used to prevent C. difficile-associated diarrhea?
  3. In adults and children with Crohn’s disease, should probiotics be used for induction or maintenance of remission?
  4. In adults and children with ulcerative colitis, should probiotics be used for induction or maintenance of remission?
  5. In adults and children with ileal pouch-anal anastomosis for chronic ulcerative colitis, should probiotics be used for prevention or maintenance of remission of pouchitis?
  6. In symptomatic children and adults with irritable bowel syndrome, should probiotics be used to improve global response or abdominal pain severity?
  7. In children with acute infectious gastroenteritis, should probiotics be used to reduce the duration or severity of diarrhea?
  8. In preterm, low birthweight newborns, should probiotics be used to prevent necrotizing enterocolitis, sepsis, and all-cause mortality?


Related articles

The past decade of probiotics and prebiotics research: ISAPP board members share their perspectives.

By ISAPP board members, compiled by Kristina Campbell

Scientific progress in the field of probiotics and prebiotics, as in any other field, often seems to occur one tiny step at a time. Yet over the course of several years, these tiny steps can add up to significant progress.

Current members of the ISAPP board of directors hold academic positions across North America, and Europe, representing some of the experts at the forefront of scientific innovation in probiotics and prebiotics. Their collective experience encompasses functional foods, fermentations, microbial ecology, microbial genetics, immunology, and clinical medicine, including pediatrics, family medicine and gastroenterology. As we enter into 2020 and a new decade, these board members have taken a moment to reflect on how far they and their colleagues have come over the past ten years, by answering the question: What changes have occurred in the domains of research, applications, and awareness about probiotics and prebiotics?

ISAPP board members, 2019 annual meeting

Available scientific methods and tools

The change that stood out the most to the ISAPP board members over the past decade was the rapid expansion of available scientific methods and tools – from gene sequencing technology to CRISPR-Cas to bioinformatic approaches. These exciting developments have enabled scientists to obtain more information, and to do it both quickly and economically. In the words of the board members:

“Advances in sequencing technology [have] revolutionized our ability to understand the gene repertoire of each individual probiotic strain (whole genome sequencing) and the interplay with the microbiome (metagenomics). This has been really energizing to the field, but has also meant that competence in bioinformatics has become an essential tool for probiotic and prebiotic scientists.”

“A decade ago, human studies on prebiotics would look at changes in the gut microbiota using fairly laborious procedures. Nowadays, the analysis is much more extensive and straightforward to do, and probably more accurate… The biggest change has been the capability to assess not only composition of the microbiota but also its functionality. So, today, the trials include metabonomics as well as assessments of health effects (through changes in particular symptoms and /or biomarkers such as blood lipids, microbial products, immune and inflammatory status). That way, we get a far better picture of what prebiotics can do.”

“In 2010 we only had DGGE to characterize the genome and were trying to figure out how to implement 16S amplicon sequencing. Now we are implementing shotgun & shallow shotgun sequencing for similar prices. In 2010, we did only work on 3-4 probiotic lactobacilli for molecular research, now we work on 400-500 lactobacilli. We do comparative genomics and functional analyses at much larger scale. And in 2010, we paid almost 10000 euro just to sequence one genome of lactobacilli, with limited analysis, now a few hundred euro for sequencing.”

Probiotics and prebiotics for microbiome modulation

Because of the rapid advancements in scientific tools and techniques during the past decade, as mentioned above, many more research groups are endeavoring to study the microbial communities that relate to probiotics and prebiotics. Gut microbiota are of great interest—not least because, among the strategies for microbiome modulation, probiotics and prebiotics are two of the leading candidates. Moreover, microbiome data can help researchers understand the context of probiotics and prebiotics in the gut and in different environments. In particular, many clinical trials of probiotics and prebiotics now include a microbiota-related measure. Novel species and strains for food use may be identified from gut microbiota studies, although safety and efficacy assessment will form challenges for regulatory bodies. Board members said:

“My collaborators and I initiated our first human clinical trials with prebiotics in 2008 and published several papers in 2010 and 2011. These early papers were among the first in which high throughput 16S DNA sequencing was used to assess how the human gut microbiota was affected by the prebiotic, GOS. Although this is now a routine method in the field, in 2008, having a Roche 454 pyrosequencer in the lab was very special, and we were astounded to be able to identify and measure abundances of the main members of the gut microbiota. Having these large data sets also led us to realize the importance of what was at the time the “new” field of bioinformatics that was critical in analyzing and reporting the data. This research showed that GOS was bifidogenic (with high specificity) in healthy adults, but was also subject-dependent. Thus, the results clearly showed there were prebiotic responders and non-responders. This remains an important area of research for my group.”

“The decade started with general excitement that ‘dysbiosis’ of the gut microbiota is involved in just about every human health problem, and has turned into re-remembering that correlation is not causation and microbiota patterns are often driven more by random factors or factors unrelated to disease than by microbiology.”

“It’s worth noting that in 2020, the well-controlled probiotic studies showing health benefits in humans are still more convincing and valuable than the studies showing any ‘beneficial’ effects on the human microbiota.”

“Over the past decade we have witnessed a tremendous explosion in our understanding of the microbiome and its interactions with us, its host. Progress in translating this knowledge into new treatments has been slower but glimmers of encouragement have appeared and we look forward to the next decade when interventions that modulate the microbiome to benefit our health will be based on a true understanding of how they act and will be selected to the maximal benefit of each individual.”

Probiotic mechanisms of action

Probiotic mechanisms of action are a perennial hot topic within the scientific community—and many had hoped that the new suite of scientific tools at scientists’ disposal would significantly advance this area of research during the past decade. But according to one ISAPP board member:

“In 2010 I would have confidently predicted that by 2020 we would have much more of a mechanistic understanding of probiotic mechanisms [and] the importance of strain effects… But this simply has not happened.  The field has become more biologically and computationally complex and many millions have been spent on research, but I still don’t think we can answer the fundamental question we faced in 2010, and in 2000, and in 1990 – what makes one a strain a probiotic, while another is not?”

But in the views of other board members:

“Through genomic and metabolomic studies we are identifying differences between strains that function at different sites and what properties are important for their probiotic function.”

“Identify[ing] the key effector molecules turned out to be more complex [than] we thought 10 years ago. It has become clear to me that probiotic mechanisms of action are per definition complex and multifactorial, because they are living microbes having thousands of molecules that all play a role. Yet, there is clearly an hierarchy of effector molecules.”

Probiotic and prebiotic applications

In general, microbiome studies of the past decade have led to a better appreciation of the ubiquity and complexity of microbial communities—not just those associated with different human body sites, but also those occupying every possible niche on Earth. ISAPP board members reflect:

“In 2010, I was mainly studying probiotics for the gut and vagina, now we have explored probiotics for the skin, respiratory tract, animals, plants, isolates from fermented vegetables that can boost vegan probiotic formulations etc., and other areas.”

“Two areas of research I am doing I’d never have imagined in 2010 are in honey bees and Chinook salmon and against environmental chemicals, administering probiotics.”

Public awareness of probiotics and prebiotics

Numerous studies and surveys show the general public has more awareness than ever of probiotics – and increasingly, of prebiotics too. Individuals receive their information through many different channels, both digital (e.g. blogs, websites) and non-digital (e.g. magazines, product packaging). The past decade also saw the creation of valuable evidence-based resources, such as the Clinical Guides available in the US and Canada, and resources from World Gastroenterology Organisation and from ESPGHAN (probiotics for pediatric acute gastroenteritispediatric nosocomial diarrheapreterm infants, and pediatric AAD). These resources have been enabled by a critical mass of studies that have examined the efficacy of various probiotic strains for certain indications. One board member says:

“From a clinical perspective, the biggest change for us has been that the general public knows so much about probiotics; now we are doing a lot less educating of docs and patients about the concepts behind our probiotic studies.”

But there’s still work to be done:

“The term probiotic is now widely known, but still too often people are misinterpreting what it means, or generalizing the whole field instead of recognizing strain and product differences. We need to continue to educate and clarify to keep the messaging on track.”

“There is still lack of knowledge that not all probiotics are equal. The clinical effects and safety of any single probiotic or combination of probiotics should not be extrapolated to other probiotics. The same applies to prebiotics.”

“Choosing a probiotic continues to be a major hurdle for the consumer – for every probiotic strain that is well characterized, studied in detail in appropriate disease models, and shown to be effective in clinical trials there are hundreds that would fail to pass even the most basic tests of quality control. We must help the consumer to make informed choices.”


It seems that, while the past decade has been a fruitful time for probiotics and prebiotics research and public awareness, scientists still have a lot of work to do. In the 2020s they will use the tools available to them, and continue to develop new ones, to gain more detailed and multi-faceted information about probiotic strains and prebiotic compounds—and about the context in which they operate (for instance, the gut microbiome), to ultimately confer benefits on human health.

The small intestinal ‘mysteriome’: A potentially important but uncharted microbiome

By Eamonn MM Quigley MD FRCP FACP MACG FRCPI, Lynda K and David M Underwood, Center for Digestive Disorders, Division of Gastroenterology and Hepatology, Houston Methodist Hospital, Houston, Texas, USA


Over recent years, countless publications have documented the status of the microbiota of the gastrointestinal tract by examining fecal samples. While this approach does provide a “snapshot” or representation of what is going on in the gut, and especially in the colon, it is a crude measure of the complex interactions between micro-organisms in the gut, as well as between these same microorganisms and us (their hosts). Fecal samples comprise a terminal microbial ecosystem, characterized by depletion of readily fermentable substrates, with a concomitant change in microbial composition, even compared to those farther upstream in the colon. It is unlikely, for example, that studies using fecal samples provide a full picture of what happens when bacteria (or other microorganisms) “talk” to the lining of the gut (the mucosa) or interact with the immune system of the intestine. Even less likely is that they provide any insights into bacterial populations in the small intestine, where most of the digestion of food and absorption of nutrients takes place. The small intestine also possesses the most abundant immune tissue of the entire gastrointestinal tract.

Yet, details of which bacteria actually inhabit this long and important organ, the small intestine, are sketchy. This lack of knowledge has apparently not restricted much theorizing and speculation about the role of an overgrowth of colonic-type bacteria (referred to as small intestinal bacterial overgrowth – SIBO) in the small intestine in many symptoms, disorders, and diseases. According to one especially popular theory – the “leaky gut” hypothesis – the list of conditions is nearly endless. The “leaky gut” hypothesizes that dysbiosis in the small intestine (in other words SIBO) and a disruption of the gut barrier leads to “leakage” of bacteria and bacterial products into the circulation causing inflammation, allergy, and autoimmunity.

There are several leaps of faith involved in “leaky gut” including, of course, the definition and diagnosis of SIBO. Traditional methods of diagnosing SIBO (obtaining fluid samples directly from the upper small intestine or a variety of breath tests) are fraught with problems and, in essence, have precluded a universally accepted definition of SIBO.

Fundamental to this dilemma is the definition of the normal small intestinal microbiome – how can we diagnose abnormal when we do not know the limits of normality? I would contend that, while there are situations where it is undoubted (based on the clinical context and various laboratory and other findings) that SIBO is an issue, there are countless more instances where SIBO is over-diagnosed and incorrectly implicated as the cause of an individual’s symptoms. This is an important issue as it can lead to the inappropriate use of antibiotics – something we all wish to avoid.

There is some good news – clever techniques exist for obtaining uncontaminated fluid samples from the small intestine, a capsule technology that permits live sampling of intestinal gases (generated by bacteria) as it traverses the intestine and the application, at last, of high-throughput sequencing, metagenomics, metabolomics, and metatranscriptomics to small intestinal microbiota suggest that the accurate definition of the normal small intestinal microbiome is not far off. At that time, we can all agree on an accurate and clinically meaningful definition of SIBO.