Which probiotic to take: the strain matters more than the CFU

On the shelf there are bottles with 5 billion bacteria and bottles with 100 billion. The second one costs considerably more, the number on the lid is twenty times bigger and almost everyone draws the same conclusion: that it must be twenty times better. It is the only large, easily comparable figure on the pack, so it is the one that decides the purchase.

And it is the one that tells you least. Clinical trials are not run on "probiotics" or on "100 billion CFU": they are run on a specific strain, identified by name and code, at a specific dose, for a specific problem. Change the strain and the result stops being transferable, even within the same species. This article explains what to look for on the label, which strains have real trials behind them, when the evidence supports taking them and when — most of the time — they are not needed.

In short

• A probiotic is, by consensus definition, "live microorganisms that, when administered in adequate amounts, confer a health benefit on the host" (ISAPP, 2014). The adequate amount is set by that strain's trial, not by the bottle.
• The effect is strain-specific and disease-specific. In a review of 228 trials and 25 probiotic types, pooling different strains produced positive conclusions that collapsed once the strains were separated: in antibiotic-associated diarrhoea, only 2 of 11 lactobacilli types were effective.
• More CFU is not better. In a trial with 362 women, the 10⁸ CFU dose beat placebo while 10⁶ and 10¹⁰ did not.
• The EU has not authorised a single health claim for any probiotic. The only authorised sentence in this field is about live yoghurt cultures and lactose.
• Where the evidence is strongest: preventing diarrhoea while on antibiotics and, in people at high risk, Clostridioides difficile diarrhoea. For stopping a gastroenteritis already under way, no.
• An uncomfortable warning: after antibiotics, a generic probiotic can delay the return of your own microbiome by months compared with taking nothing.

What a probiotic actually is (and what it is not)

The definition scientific bodies have used since 2014 has three parts, and all three matter: live microorganisms, in an adequate amount, delivering a demonstrated benefit. Miss one and the product is not a probiotic: it is a bottle of bacteria. A pasteurised fermented food fails the first — we cover that in which fermented foods contain live bacteria; a bottle with an invented dose fails the second; and a strain never tested in humans fails the third, however many zeros are printed on the front.

Identifying a probiotic bacterium properly takes three levels: genus, species and strain designation. Lacticaseibacillus rhamnosus GG is a complete example; Lactobacillus acidophilus on its own is like saying "a German car". Within one species live strains that behave differently, and the evidence for one is not inherited by the other. In our catalogue, Ergyphilus Plus by Nutergia is one of the few that does this properly: the label reads Lactobacillus paracasei LMG P-21380, Lactobacillus acidophilus DSM 21717 and Bifidobacterium bifidum DSM 22892. With those codes you can go and find the trials; without them, you cannot.

One confusing detail: in 2020 the genus Lactobacillus was split into 25 genera, which is why labels now read Lacticaseibacillus rhamnosus or Lactiplantibacillus plantarum where they used to say Lactobacillus. Same bacteria, updated names. Keeping the old name deceives nobody; not naming the strain does. The bottles live in digestive and probiotic enzymes, inside digestive system.

Why the strain rules and the CFU count barely matters

The clearest review on this analysed 228 randomised trials covering 25 probiotic types — 15 single-strain and 10 mixtures — across nine indications. Strain by strain, 70% of the types studied for prevention were effective (7 of 10) and 65% for treatment (11 of 17). The problem lay in meta-analyses that had thrown different strains into one bag to conclude that "probiotics work": redoing the maths strain by strain, that conclusion fell apart. In antibiotic-associated diarrhoea, out of 11 lactobacilli types only 2 showed real efficacy.

The World Gastroenterology Organisation global guidelines, 2023 edition, put it in one line: the effects are "strain-specific and dose-specific". And on quantity: "It is not possible to state a general dose that is needed for probiotics; the dosage should be based on human studies showing a health benefit."

The trial that best dismantles "more is better" was run in 362 women with irritable bowel syndrome, over four weeks, with three doses of the same bacterium: 10⁶, 10⁸ and 10¹⁰ CFU/ml. Only the middle one beat placebo, and by more than 20 percentage points on the global symptom assessment; the other two did not differ from it. The highest dose also failed because of formulation problems: the bacterium was the same, the finished product was not. What gets tested, and what you buy, is the finished product.

Five things to check on the label before you pay

1. Look for the full strain, with its code

Genus, species and designation: Lacticaseibacillus rhamnosus GG, Bifidobacterium longum subsp. infantis 35624, Saccharomyces boulardii CNCM I-745. If the pack only says "lactobacilli and bifidobacteria", you cannot check anything. Trade names and strain nicknames, according to those same guidelines, "are not controlled by the scientific community": only scientific nomenclature is. In digestive and probiotic enzymes, bottles that print the strain sit next to bottles that do not; compare two or three labels.

2. Check the CFU at the end of shelf life, not at manufacture

That is the difference between what you buy and what you swallow. The 2023 guidelines treat it as mandatory for a serious manufacturer to state the count "at the use-by date (not at the time of manufacture)". If the bottle only says "100 billion" without saying when, the figure is the factory one, and the gap can be an order of magnitude. This is where a physical shop beats any delivery: the product has been refrigerated the whole time, like the goat kefir two metres away in refrigerated.

3. Check that the dose is the one used in that strain's trial

Not the highest on the shelf: the one from the study. For preventing antibiotic-associated diarrhoea in children, the range the 2019 Cochrane review considers reasonable is 5 to 40 billion CFU per day, not a hundred billion. Above the tested dose there are no data saying anything better happens. Mid-dose formats such as Probiotax Microbiota powder cover that range without paying for the big number.

4. Check what that strain was studied for, not what it is sold for

The same bacterium can have good trials for one thing and negative trials for another. L. rhamnosus GG is the textbook case: it prevents diarrhoea while taking antibiotics and does not work for treating an acute gastroenteritis already under way — 971 children, 11.8% moderate-to-severe cases at 14 days against 12.6% on placebo. For heavy digestion the right product is not a probiotic but digestive enzymes, and for lactose, lactase. Formulas aimed at vaginal flora follow the same rule: Fémina Flor Oral and Hifas Intima are read by strain, dose and indication, not by the category on the pack.

5. Check storage, expiry date and what else is in there

Many strains need cold; other bottles also contain fructo-oligosaccharides, inulin or lactose as an excipient, which matters if your problem is precisely bloating. And there are formats that are not capsules: a vegan kefir powder to make at home, or Molkosan, a fermented whey concentrate, in superfoods and diet and nutrition.

Table 1. Strains with published trials, and how far they go

What European law allows you to say, and what it does not

There is a clash between two authorities here that is worth knowing before you read any pack. For the European Commission, the word "probiotic" on a label is itself a health claim, because it implies a benefit; and since no probiotic claim application has ever received a positive opinion from EFSA, the term should not be used. The European Ombudsman backed that interpretation in December 2024. In Spain, however, AESAN — the Spanish food safety and nutrition agency — published a note on 27 October 2020 accepting the term in the labelling, presentation and advertising of food supplements, on one condition: it cannot be accompanied by any health claim unless that claim is expressly authorised.

What that means for a buyer: a bottle saying "probiotic" is legal in Spain and does not mean its contents have any approved effect. The only authorised sentence in the whole of Regulation (EU) 432/2012 involving live bacteria is about yoghurt, not capsules: "Live cultures in yoghurt or fermented milk improve lactose digestion of the product in individuals who have difficulty digesting lactose", on condition that the product contains at least 10⁸ colony forming units of live starter microorganisms per gram (Lactobacillus delbrueckii subsp. bulgaricus and Streptococcus thermophilus). Any other promise — immunity, defences, transit — is not authorised.

Table 2. What to do in each situation

The study to know about before buying "to restore your flora"

It is the most repeated line in the sector: you finish a course of antibiotics and someone tells you to take a probiotic "to restore your flora". In 2018 a team published an uncomfortable experiment in Cell: 21 healthy volunteers were given a course of antibiotics and then split into three groups — do nothing, take a multi-strain probiotic, or receive a transplant of their own microbiome collected beforehand. The probiotic group recovered worst: reconstitution of their own microbiome was, in the authors' words, "markedly delayed and persistently incomplete" for months, while the autologous transplant returned it to normal "within days". That does not make probiotics bad: it means "restoring your flora" is not an indication, whereas preventing diarrhoea while the antibiotic lasts is. It is a distinction that costs us sales and that we would rather make.

The shopping list

What actually feeds the bacteria you already have is not in the supplements aisle, it is in the food aisle. Fermentable fibre every day: pulses — which also cover a good part of your protein for the day — wholegrain cereals, flakes and muesli, seeds, nuts and dried fruit, vegetables and fresh fruit. If organising it is the hard part, an organic basket sorts out the base of the week.

Live bacteria through food: goat kefir and lemon kefir water, fresh sauerkraut, spicy red kimchi, unpasteurised genmai miso, natto and soy tempeh. The chilled ones live in precooked and in cheeses; the jars and tins, in canned; fermented drinks, in drinks.

And if after all of that the bottle makes sense in your case: digestive and probiotic enzymes, backed by vitamins and minerals if there is a real deficiency, immune system in season (with the small print we already spelled out in zinc or vitamin C for immunity) and nervous system if the digestive problem travels with stress. Plants and extracts and coffee and teas complete the aisle. Whatever is reduced is in offers. And the whole pantry is in pantry.

Do you need probiotic capsules?

For most healthy people with no symptoms, no. And we are saying that while selling them. The reason is simple: without a specific indication there is no trial to apply, and without a trial the purchase is an act of faith with an expiry date. The money goes further on fibre — the Spanish average is well below the 25 g a day reference — and on fermented foods eaten daily without thinking.

There are three situations where the answer changes and a good bottle is worth paying for: while taking an antibiotic, with a strain studied for that; with a diagnosed irritable bowel syndrome, to try a strain with trials behind it for four weeks; and when a professional recommends it for a clinical reason, such as high C. difficile risk. Outside that, a yoghurt with live cultures and a plate of pulses do more for your gut than most capsules on the market.

We would add a house rule, commercial rather than scientific: if the bottle does not name the strain, do not buy it — here or anywhere. Not because it is dangerous, but because you cannot know what you have bought.

Where to buy probiotics in Barcelona and Sant Cugat

We have three physical shops and in all three you can do something no website allows: bring the two bottles you are hesitating between and read both labels with someone in front of you, looking for three specific things — the full strain with its code, whether the CFU figure is at the end of shelf life, and whether the daily dose falls inside the tested range. In five minutes you can see which of the two has done its homework, and it is often the cheaper one. The second argument is the fridge: strains that require 4 ºC have been refrigerated since they arrived, which a parcel crossing Spain in August cannot guarantee.

• Provença 242, 08008 Barcelona — Eixample district, between Balmes and Enric Granados.
• Còrsega 78, 08029 Barcelona — Eixample, near Sants station.
• Antoni Bell 2, 08174 Sant Cugat del Vallès — Volpelleres neighbourhood.

If you would rather ask before coming over, the phone number is +34 683 118 961 and the email is linverd@linverd.com. We are just as quick to tell you when you do not need a bottle at all.

Frequently asked questions

How many CFU should a good probiotic have?

There is no universal figure: it depends on the strain and on what it is taken for. For preventing antibiotic-associated diarrhoea, the documented range is 5 to 40 billion CFU a day. In the reference trial in irritable bowel syndrome, the dose that worked was 10⁸ CFU/ml and the 10¹⁰ one did not. The international guidelines are explicit: no general dose can be given.

Is a multi-strain probiotic better than a single-strain one?

Neither is better in itself. Of the 25 types reviewed across 228 trials, 15 were single-strain and 10 were mixtures, and there were effective and ineffective ones in both groups. What matters is whether that specific formula — that combination, at that dose — has been tested in people with your problem.

Should you take probiotics with antibiotics?

It is the best-supported indication. In children, 33 trials with 6,352 participants gave 8% diarrhoea with a probiotic against 19% without, and at doses of 5 billion or more you needed to treat 6 children to prevent one case. Choose a strain studied for that rather than any bottle, and do not confuse it with "restoring your flora" afterwards.

Do probiotics help with gastroenteritis?

For stopping one that has already started, no. The 2020 Cochrane review pooled 82 studies and 12,127 people and found no difference in the proportion of people still having diarrhoea at 48 hours. A trial in 971 children treated with L. rhamnosus GG gave 11.8% moderate-to-severe cases against 12.6% on placebo.

Can I cover it with fermented food alone?

For the goal of "looking after your gut" with no symptoms, yes, and it is cheaper. A kefir, a fresh sauerkraut or an unpasteurised miso provide live bacteria plus fibre and compounds from the fermentation itself. What fermented food cannot do is replace a specific strain for a specific indication: there, the trial was run with the capsule.

Do they have side effects?

In healthy people, adverse events in trials were uncommon and mild: gas, bloating, nausea, the odd rash. In the paediatric review they appeared in 4% of the probiotic group against 6% of the control group. In people who are immunosuppressed, seriously ill or have central catheters the situation is different, and the decision is not a shelf decision.

Where can I buy probiotics in Barcelona or Sant Cugat without buying blind?

In any of our three shops — Provença 242 and Còrsega 78 in Barcelona, Antoni Bell 2 in Sant Cugat — you can compare labels with someone in front of you and check whether the bottle names the strain and gives CFU at the end of shelf life. We also answer questions on +34 683 118 961.

To finish

The number on the lid tells you least and the strain tells you almost everything: look for its code, check that the dose is the one from the trial, and ask what it was studied for. And first of all, check whether your case needs a bottle at all.

The whole range is in digestive and probiotic enzymes, and the three shops are open. To keep reading: how much fibre per day you need and how to reach 25 g is the other half of this story — the bacteria's food — back to routine and the gut reset gives the general picture, and chronic stress and the nervous system explains why the gut suffers when everything else does. The overview is in the supplement manual, and the shop, in our organic store in Barcelona.

This article is for information only and does not replace advice from a healthcare professional. Food supplements should not be used as a substitute for a balanced diet. If you are pregnant or breastfeeding, taking medication, immunocompromised or living with a serious illness, talk to your doctor or pharmacist before taking a probiotic.