Two Studies May Change View of Probiotics: One Size May Not Fit All

Tara Haelle

October 15, 2018

The medical community has much to learn about probiotics and their effects on individual patients before the supplements can be broadly prescribed to achieve uniform effects, results of two new studies show. The randomized, controlled studies, which relied on serial biopsies rather than stool samples, show that whereas some patients appear to benefit of probiotics, others may have little response or potentially adverse reactions to the supplements.

And although it may be tempting to conclude that probiotics have no beneficial effect or are harmful, in reality, the results are more nuanced. But at the very least, the new data show that probiotics have the potential for detriment, which is an important discovery given that few studies look for, much less report, potential harms from probiotics.

Summed briefly, the studies, which relied on a mix of randomized studies in healthy human participants and animal data, found the following:

  • Stool samples alone do not provide an accurate or sufficient portrait of the interactions between probiotics and a human host's pre-existing microbiome and overall health.

  • Some people's microbiota resists colonization with probiotics, but others' microbiomes change in response to probiotics, and sometimes in different ways at different points along the gastrointestinal tract.

  • Probiotics administration after antibiotics can impede the microbiome's return to baseline flora in humans, though the clinical significance of that is unclear.

  • Characteristics of both an individual person and of their microbiome can largely predict probiotics' effects in that person.

  • The mouse model is not very informative or reliable in understanding probiotics' effects in humans. 

The findings have opened a Pandora's box of questions that make manifest just how much there is to learn about probiotics. Physicians — and consumers — who want to explore therapeutic use of probiotics will need to accept the current uncertainties of their use, the highly personalized effects they have on different individuals, and the need to thoroughly comb through the evidence before making a decision on which probiotic to use, in whom, for what, and why.   

"In our view, in the clinical decision-making process, probiotics should be regarded by physicians as any other medical treatment," said Eran Elinav, MD, PhD, chair of immunology at the Weizmann Institute of Science in Rehovot, Israel, and senior author on both articles.

"In doing so, and in the unfortunate absence of any regulatory agency approval of probiotics for any indications, physicians should weigh the current evidence for clinical efficacy — debated and conflicting in many cases — and the potentially alarming long-term dysbiosis demonstrated upon probiotics usage with antibiotics, with its potential long-term implications, in making a case-by-case decision," Elinav told Medscape Medical News.

"On a broader scale, the field is yearning for nonindustry-funded, high-quality multicenter trials to assess efficacy and adverse effects associated with probiotics use in affirming or debunking their utility. Until then, we believe that the burden of proof lies at the hands of those administering probiotics," Elinav continued.

David A. Johnson, MD, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, agreed that the findings support slowing the current dash to find all sorts of applications for probiotics.

"It's a bit of a wake-up call for care providers who glibly recommend supplementation with probiotics," Johnson told Medscape Medical News. "It's similarly a wake-up for patients who frequently view probiotics as a healthy pill," an attitude which can frustrate care providers when patients expect that popping a probiotics pill can replace taking a hard look at their current health, diet, exercise, and other lifestyle aspects.

"As a one-shoe-fits-all, it's clearly not ready for prime time, and I think these two studies should create considerable doubt in individual prescribers and in patients that it's not just about a pill to make you healthy," Johnson said. "You have to work at being healthy, and [probiotics] may cause symptoms."

The Details of the Two Studies

In the study led by Niv Zmora, MD, of the Weizmann Institute of Science, researchers tested a multistrain probiotic combination in both humans and mice. The 29 adult human participants filled out medical, lifestyle, and food frequency questionnaires and were assigned either to participate in the group involving the probiotics intervention or join the control group and not receive any intervention. Among those in the intervention group, 14 people took a probiotics supplement twice daily, and the remaining five took a placebo twice daily for 4 weeks. The supplement contained Bifidobacterium bifidum, Lactobacillus rhamnosus, Lactococcus lactis, Lactobacillus casei, Bifidobacterium breve, Streptococcus thermophilus, Bifidobacterium longum subsp longum, Lactobacillus paracasei, Lactobacillus plantarum, and B. longum subsp infantis.

The researchers collected stool samples from these participants at baseline, daily for the first week, and weekly through the end of follow-up. They also collected luminal, mucosal, and regional biopsies before and during the study from a subset of those taking probiotics, those taking placebo, and those not involved in the intervention.

Meanwhile, the same probiotics supplement was fed to a group of germ-free mice and a group of nongerm-free mice. The mice were subsequently asphyxiated for the purpose of dissection and examination of eight parts of the gastrointestinal tract. In the mice with pre-existing gut colonization, the "probiotics encountered a marked mucosal colonization resistance" across the board.

But things weren't so simple in humans. The mucosal colonization patterns that emerged throughout the gastrointestinal tract of the human participants depended on the strain of the probiotic, region of the body sampled, and factors unique to that person. For example, in subsequent analyses, the researchers were able to use knowledge of an individual's gene expression patterns in the gut and their baseline microbiome to accurately predict how probiotics would affect the gut.

"This challenges the currently utilized empiric one-size-fits-all probiotics paradigm and suggests that we should transform into a measurement- and personalized-based probiotics-tailoring approach," Elinav said.

Further, investigation of the stool neither described the probiotics' effects nor enabled prediction of those effects. In short, examining the stool did not tell the researchers what the probiotics did in a person's gastrointestinal tract, much less predict that behavior.

"This finding was surprising because we all heavily rely on the stool as a proxy of gut microbiome configuration, and this tells us that at least in some cases, we need to directly sample the gut microbiome to get accurate results," Elinav said.

In the other study, Jotham Suez, PhD student, also of the Weizmann Institute of Science, and colleagues enrolled 21 healthy adults who took a 7-day course of antibiotics with 500 mg oral ciprofloxacin twice daily and 500 mg oral metronidazole three times daily.

After the course of antibiotics, participants were randomly assigned to one of three groups. Eight participants took probiotics twice daily for 28 days, six received an autologous fecal transplant (the fecal material was procured before antibiotic treatment), and seven had no further treatment. Again, the researchers collected stool samples and biopsies for analysis before and after the interventions.

Contrary to general expectations, they discovered probiotics delayed microbiome recovery. In other words, it took longer for the gut microbiome to return to its previous state in those who received probiotics compared with those who received an autologous fecal transplant or no post-antibiotic intervention at all.

"This highlights a potentially alarming adverse effect of probiotics use in this context that was previously unappreciated," Elinav said. "A persistent post-antibiotic dysbiosis of the type induced by probiotics in our study has been linked to risk of developing obesity, cardiometabolic disease, allergies, and more." Elinav said further investigation is required to understand exactly what this gut response means.

Where Do We Go From Here?

The articles' findings have generated much discussion, including a robust dialogue regarding probiotics regulation at a recent meeting involving representatives of the US Food and Drug Administration and National Institutes of Health, according to Vincent Young, MD, PhD, a professor of infectious disease at the University of Michigan Medical School in Ann Arbor. Whereas some speakers worried that regulation may stifle innovation and widespread therapeutic use of probiotics, others felt regulations were important because too little is known to make broad, wholesale recommendations for their use, Young said.

Yet multiple experts interviewed for this article converged on a single, consistent message: probiotics have clinical utility, and current evidence exists to support a handful of extremely narrow evidence-based clinical applications — such as preventing Clostridium difficile in those at risk — but beyond that, too little is known to recommend their use more broadly for any condition in any population.

"We know a lot, but we don't know enough to make big recommendations that are widely applicable," Young told Medscape Medical News. "I don't think we have guidance on how to run through a list of what probiotics to try or how to determine which patients might respond," he said. He acknowledges that this message will likely be disappointing for people wanting to recommend this or that probiotic for different conditions.

"The time has probably come that physicians have started thinking there's something to probiotics use, but we have to be careful as scientists to do an evidence-based search of the literature for the disease we want to treat and pick the right one," Robert Martindale, MD, PhD, chief of gastrointestinal and director of the hospital nutritional services at Oregon Health and Science University, Portland, told Medscape Medical News.

"The problem with the probiotics literature right now is that it's all over the map," he continued. Yet specifics matter. For example, data do support using probiotics to prevent C. difficile infection when administered simultaneously with antibiotics, he said, but in a person who already has a C. difficile infection, probiotics won't help much.

The studies are more hypothesis-generating than clinically instructive, Christopher Harrison, MD, director of infectious disease research at Children's Mercy Hospital, University of Missouri at Kansas City, said in an interview.

"At the current time, probiotics, in the way we currently think about using them, don't seem to have any definite harm that we can tell clinically, and there seem to be some signals that there is some clinical benefit in many situations, but the degree of benefit is variable between individuals and we shouldn't overpromise what they can do for our patients," Harrison told Medscape Medical News.

He pointed out that some of the methods used in the studies both strengthen the reliability of their conclusions but also limit their generalizability. For example, the use of such a strong combination of antibiotics means the ciprofloxacin-metronidazole course likely wiped out the bulk of many participants' gut microbiota, but people rarely take such a strong regimen. What the researchers saw in this study may not apply to amoxicillin or a less potent antibiotic used to treat a sinus or urinary tract infection, he said.

Further, too little is known to determine whether the slower regrowth of a person's individual microbiome following probiotics is necessarily bad. He noted that the slower recovery may ultimately leave a patient with a healthier microbiome than they had before they took antibiotics, but more work needs to be done to know what the implications are. "We don't know just yet how that's going to turn out, and that's something that needs to be looked at."

And that same conclusion — we don't know how it's going to turn out yet — is essentially the refrain that should echo most from these findings, along with the realization that any widespread therapeutic use of probiotics in the future is likely going to require customization to the individual and to specific genetic qualities of their and their microbiome's genomes.

"These articles aren't going to convince people who give probiotics to stop or people who don't give them to start," Young told Medscape Medical News. "The main message here is that we really do need to figure out how things go. If nothing else, these findings carefully show us that we still don't know everything."

Both studies were funded through a variety of investigator-specific private and public scientific foundations and trusts. The author groups have reported no relevant financial relationships.

Cell. 2018;174:1388-1405.e21, 1406-1423.e16.

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.