Microbiome learning path · Chapter 9
What are probiotics? A complete guide
Probiotics are live microorganisms that, when taken in adequate amounts, provide a health benefit. Their effects are specific to the strain and to the condition. The evidence is good in a few situations, such as preventing diarrhoea linked to antibiotics, and weak in many others. Healthy people do not need a daily probiotic. Fibre and fermented foods do more for the gut’s own bacteria.
In short
- A probiotic is defined by its strain, its dose and the evidence for a benefit.
- Different strains do different things. A result for one does not apply to another.
- Probiotics pass through. Their effect lasts about as long as they are taken.
- With antibiotics they lower the chance of diarrhoea. They have not been shown to rebuild your own bacteria faster.
- More strains and higher counts are not better.
- Use a probiotic for a specific purpose, for a defined period, and stop if it does not help.
- People with very weak immunity or who are critically ill should not take them without medical advice.
How are probiotics, prebiotics and postbiotics different?
| Term | Meaning | Example |
|---|---|---|
| Probiotic | Live microbes with a demonstrated benefit | A named strain of Lactobacillus or Saccharomyces in a capsule |
| Prebiotic | Food for beneficial microbes | Fibres in onion, banana, oats and pulses |
| Synbiotic | A probiotic and a prebiotic together | Combination products |
| Postbiotic | Non-living microbes or their components with a benefit | Heat-treated bacterial preparations |
| Fermented food | Food transformed by microbes | Curd, buttermilk, kanji |
See prebiotics, postbiotics and fermented foods.
What does a probiotic have to show?
The word is used loosely on labels. The international definition is strict: live microorganisms that, when given in adequate amounts, confer a health benefit.
Four tests. A product that fails any one of them is a food with bacteria in it, which may be perfectly good food, but is not a probiotic for that purpose.
Most products on a shelf pass the first test and cannot show the last. That does not make them harmful. It means the claim on the label has not been tested.
How is a probiotic named?
By genus, species and strain, for example Lactobacillus rhamnosus GG. The strain is the part that matters, because benefits shown for one strain cannot be assumed for another of the same species. A label that lists only “lactobacillus” tells you very little.
Genus, species, strain. Trials test a strain. A label that stops at the species has left out the part that was tested.
Do probiotics settle in the gut?
Usually not. They are visitors.
In a study that sampled the lining of the gut directly, during an 11-strain probiotic or a placebo, the probiotic bacteria survived the journey. Whether they settled in the lining differed from person to person and from one part of the gut to another. Finding the bacteria in the stool did not show who had been colonised. The effect on the gut was temporary.
Present while taken. A probiotic acts as it passes through. It does not rebuild the resident community.
Two things follow. A benefit generally lasts only while the product is taken. And a stool test cannot tell you whether a probiotic “took”.
Where is the evidence reasonably good?
With antibiotics
Certain strains reduce the chance of antibiotic-associated diarrhoea and of a serious gut infection that can follow antibiotics. In a review that pooled 63 trials and 11,811 people, probiotics prevented one case of diarrhoea for about every 13 people treated.
During H. pylori treatment
Certain probiotics reduce the side effects of the antibiotic course.
Pouchitis
A specific multi-strain combination helps to prevent inflammation of the pouch created after surgery for ulcerative colitis.
Premature babies
Specific strains reduce a serious bowel complication in preterm infants, under hospital care.
Where is the evidence weak or mixed?
- Irritable bowel syndrome: some trials show benefit and others none. Guidelines do not recommend probiotics for overall IBS symptoms; a four-week trial of one product is reasonable.
- Acute infectious diarrhoea: results differ between studies and regions.
- Crohn’s disease: no clear benefit.
- General health, immunity, weight and mood in healthy people: not established.
With antibiotics: do probiotics help recovery?
There are two different questions here, and they have different answers.
Two questions. Preventing diarrhoea is one thing. Rebuilding your own bacteria is another.
To prevent diarrhoea during the course: yes. That is the pooled result of 63 trials. The reviewers added that the strains used were poorly documented, so the review cannot say which probiotic works for which antibiotic.
To bring your own bacteria back sooner: not shown. In one study, volunteers who had taken antibiotics were then given a multi-strain probiotic, or nothing, or a transplant of their own stool stored before the antibiotics. With the probiotic, the return of their own bacteria was markedly delayed, and remained incomplete, compared with simply waiting. With their own stored stool, recovery took days.
This was a small study and has not been repeated. It is enough to say that a probiotic after antibiotics should not be assumed to help the gut recover. Food does that. See restoring gut bacteria after antibiotics.
Are more strains and higher counts better?
No.
A review compared trials of single strains with trials of mixtures that contained the same strain, for the same condition: 65 trials in all. In most cases the mixture was not more effective than the single strain. In one comparison the single strain did better, and in another the mixture did.
One or many. The number of strains on a label is not a measure of quality.
The reviewers’ conclusion is the useful rule: choose a probiotic by the trial evidence for that product, not by the number of strains in it. The same applies to the count. A larger number is not a better product unless that amount was the one tested. For reading a label, see how to choose a probiotic in India.
What about spore and yeast probiotics?
Most probiotics are lactobacilli and bifidobacteria. Two other kinds are common in India.
Spore-forming bacteria
These form a hardy spore, which helps them to survive storage and stomach acid. In a review of 6 trials in 1,298 children with acute diarrhoea, a spore-forming probiotic shortened the illness by about nine hours and the hospital stay by under a day. For most other uses the trials are few.
Yeast
A probiotic yeast is not a bacterium, so antibiotics given for bacterial infections do not kill it. It is among the organisms studied for diarrhoea linked to antibiotics and for the side effects of H. pylori treatment.
The same cautions apply to both as to any probiotic, and the people who should avoid probiotics should avoid these too.
Do probiotics lower cholesterol?
A little, in trials. A pooled analysis of 32 trials in 1,971 people found a modest fall in total cholesterol with probiotics compared with control. The effect was larger in people who started with higher cholesterol and in longer trials.
The trials were short and measured cholesterol, not heart attacks. Probiotics are not a treatment for raised cholesterol and do not replace one. Oats, barley and isabgol have better evidence. See prebiotics: a doctor’s guide.
How should a probiotic be used?
Have a reason
A condition for which a particular strain has been studied.
Choose by strain and dose
The label should state the full strain name and the number of live organisms present up to the expiry date, not only at manufacture.
Give it four weeks
If there is no benefit by then, stop. Trying a different strain is reasonable; taking several at once is not.
With antibiotics
Take the probiotic at least two hours apart from each antibiotic dose, and continue for a week or two afterward.
Store it as directed
Heat and humidity kill live organisms.
Who should avoid probiotics?
For most people they are low-risk, with gas and bloating the usual side effects. Rarely, the organisms have caused bloodstream infections. For that reason they should not be used without specialist advice by people who are critically ill, have a central venous line, have severely weakened immunity, or have a damaged heart valve.
Severe acute pancreatitis is a firm exception. In a trial of 296 patients predicted to have severe acute pancreatitis, a multi-strain probiotic did not reduce infections, and more patients died in the probiotic group: 16% against 6% with placebo. Nine patients given the probiotic developed loss of blood supply to the bowel, against none given placebo. Probiotics are not given in this illness.
That trial matters beyond pancreatitis. It shows that “natural” and “safe” are not the same thing in someone who is seriously ill.
Bloating and brain fog. One small study linked probiotic use with gas, bloating and brain fog in people who also had bacterial overgrowth of the small intestine. A later study did not confirm the link with overgrowth.
Do I need a probiotic if I eat curd?
For general health, no. Curd, buttermilk and other fermented foods supply live bacteria, and fibre feeds the ones already living in your gut. A probiotic product is a targeted tool, not a daily necessity. See is curd a probiotic?
From my practice
I prescribe probiotics for a reason and for a set time: with certain antibiotic courses, during H. pylori treatment, and as a four-week trial in some patients with irritable bowel syndrome. I ask the patient to tell me at four weeks whether anything changed. If nothing did, we stop.
What I do not do is prescribe a probiotic to “rebuild the gut” after antibiotics, or add a second and a third product when the first has not helped. And in a patient who is seriously ill in hospital, I do not use them.
| What this page says | How it is known |
|---|---|
| Probiotics prevent about one case of antibiotic-linked diarrhoea for every 13 people treated | A review that pooled 63 trials and 11,811 people |
| Probiotics settle in some people and not others, and the effect is temporary | A study that sampled the gut lining directly |
| After antibiotics, a probiotic delayed the return of people’s own bacteria | One small study in volunteers |
| Mixtures are usually not more effective than single strains | A review of 65 trials |
| A spore-forming probiotic shortened childhood diarrhoea by about nine hours | A review of 6 trials in 1,298 children |
| Probiotics increased deaths in severe acute pancreatitis | A randomised trial in 296 patients |
| Probiotics lower total cholesterol modestly | A pooled analysis of 32 trials |
When should I see a doctor?
See a doctor for symptoms that last more than a few weeks. Using probiotics for months to treat undiagnosed bloating, diarrhoea or pain delays the diagnosis.
Questions patients ask me
Should I take a probiotic every day?
Not unless there is a specific reason. Benefits last only while the product is taken, and most healthy people gain nothing measurable.
Can probiotics cause bloating?
Yes, especially in the first week, and in people with bacterial overgrowth. If it persists, stop.
Are more strains and higher counts better?
No. What matters is whether that strain, at that dose, has been shown to work for your condition.
Can children and pregnant women take probiotics?
Commonly used strains have been given to both groups in studies without significant problems. Check with your doctor before starting.
Do probiotics settle in the gut for good?
Usually not. In a study that sampled the gut lining directly, an 11-strain probiotic settled in some people and not in others, and finding the bacteria in the stool did not show which. Its effect on the gut was temporary. That is why any benefit generally lasts only as long as the product is taken.
Sources
This page is written for patients and is based on the guidelines and studies below. It is for education and does not replace a consultation.
- International consensus statement on the term probiotic. Hill C et al. Nat Rev Gastroenterol Hepatol. 2014. PubMed 24912386
- American Gastroenterological Association guideline on probiotics in digestive disorders. Su GL et al. Gastroenterology. 2020. PubMed 32531291
- American College of Gastroenterology guideline on irritable bowel syndrome. Lacy BE et al. Am J Gastroenterol. 2021. doi:10.14309/ajg.0000000000001036
- Maastricht VI/Florence consensus on Helicobacter pylori infection. Malfertheiner P et al. Gut. 2022. PubMed 35944925
- International consensus statement on fermented foods. Marco ML et al. Nat Rev Gastroenterol Hepatol. 2021. PubMed 33398112
- Study of whether probiotic bacteria settle in the human gut lining. Zmora N et al. Cell. 2018. doi:10.1016/j.cell.2018.08.041
- Systematic review of probiotics for antibiotic-associated diarrhoea. Hempel S et al. JAMA. 2012. PubMed 22570464
- Study of probiotics and gut microbiome recovery after antibiotics. Suez J et al. Cell. 2018. PubMed 30193113
- Systematic review comparing single-strain probiotics with multi-strain mixtures in 65 trials. McFarland LV. Dig Dis Sci. 2021. doi:10.1007/s10620-020-06244-z
- Systematic review of a spore-forming probiotic for acute diarrhoea in 1,298 children. Ianiro G et al. Nutrients. 2018. doi:10.3390/nu10081074
- Randomised placebo-controlled trial of probiotics in 296 patients with predicted severe acute pancreatitis. Besselink MG et al. Lancet. 2008. doi:10.1016/S0140-6736(08)60207-X
- Meta-analysis of 32 randomised trials of probiotics and total cholesterol. Wang L et al. Medicine (Baltimore). 2018. doi:10.1097/MD.0000000000009679
- Study of brain fog, probiotics, bacterial overgrowth and D-lactic acid. Rao SSC et al. Clin Transl Gastroenterol. 2018. doi:10.1038/s41424-018-0030-7