Lacticaseibacillus rhamnosus GG

What it is

LGG is a single strain of lactic-acid bacterium, isolated from the human gut by Gorbach and Goldin in 1983 and sold as a probiotic. It is one strain of the species Lacticaseibacillus rhamnosus; other rhamnosus strains (such as R0011) are different products with their own evidence.

What the research says

The best evidence is for reducing antibiotic-associated diarrhoea in children (moderate-quality pooled evidence); in adults the pooled result was not significant. A large 2018 US trial found LGG did no better than placebo for acute gastroenteritis in preschool children. Evidence for respiratory infections and childhood IBS is small and mixed. It is usually well tolerated in healthy people, but LGG and other lactobacilli have caused bloodstream infections in intensive care patients.

Evidence grade: Limited evidence.

What goes wrong

Genome sequencing showed Lactobacillus bacteraemia in intensive-care patients came directly from the probiotic product they had been given, and ICU patients on probiotics had markedly higher risk of Lactobacillus bacteraemia. (Source 1)

• Cohort study, Low certainty.
• Size: ICU patients at one children's hospital (numbers not in recorded abstract)
• Who: Intensive care patients.
• How long: Retrospective.
• Result: 'markedly higher risk' (abstract gives no numbers in recorded text)
• Funding: Not stated in recorded text.

Here we report a markedly higher risk of Lactobacillus bacteremia for intensive care unit (ICU) patients treated with probiotics compared to those not treated

Case reports describe LGG causing invasive infection, mostly in immunocompromised people. (Source 2)

• Case series, Low certainty.
• Size: Case reports (number not given)
• Who: Mostly immunocompromised patients.
• How long: n/a.
• Result: n/a.
• Funding: n/a.

LGG, like other lactobacilli, are rarely pathogenic, although there are rare case reports of LGG causing invasive disease, mostly in immunocompromised patients

Class caution: in the PROPATRIA trial of 298 patients with predicted severe acute pancreatitis, a multispecies probiotic given enterally did not reduce infections and more patients died (24 (16%) vs 9 (6%)); the product was a different multispecies mix, so this is a warning about critically ill patients, not evidence about this strain. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 296 analysed (152 probiotic, 144 placebo)
• Who: Adults with predicted severe acute pancreatitis in Dutch hospitals.
• How long: 28 days of treatment; 90-day follow-up.
• Result: Deaths 24 (16%) vs 9 (6%); relative risk 2·53, 95% CI 1·22–5·25.
• Funding: Not stated in recorded text.

24 (16%) patients in the probiotics group died, compared with nine (6%) in the placebo group (relative risk 2·53, 95% CI 1·22–5·25)

In a study of people and mice given antibiotics, a multi-strain probiotic delayed the return of the person's own gut bacteria compared with no treatment; this was a small mechanistic study of one multi-strain product, not of this strain specifically. (Source 4)

• Randomized trial, Low certainty.
• Size: Small human cohort plus mice (exact numbers not in abstract text recorded)
• Who: Healthy human volunteers and mice after broad-spectrum antibiotics.
• How long: Weeks of follow-up after antibiotics.
• Result: Qualitative: 'markedly delayed and persistently incomplete' microbiome reconstitution with probiotics versus spontaneous recovery.
• Funding: Not stated in recorded text.

Compared to spontaneous post-antibiotic recovery, probiotics induced a markedly delayed and persistently incomplete indigenous stool/mucosal microbiome reconstitution

What the evidence supports

Pooled trials found LGG roughly halved antibiotic-associated diarrhoea in children, with moderate-quality evidence. (Source 5)

• Meta-analysis, Moderate certainty.
• Size: 5 RCTs, 445 children.
• Who: Children receiving antibiotics.
• How long: Duration of antibiotic course plus follow-up (varied)
• Result: RR 0.48, 95% CI 0.26-0.89.
• Funding: Not stated in recorded text.

(five RCTs, n = 445, RR 0.48, 95% CI 0.26-0.89; moderate QoE)

A small meta-analysis found LGG was linked to fewer ear infections in children, but only four trials were included. (Source 6)

• Meta-analysis, Low certainty.
• Size: 4 RCTs, 1805 children.
• Who: Children.
• How long: Varied.
• Result: Acute otitis media RR 0.76, 95% CI 0.64-0.91 (fixed effects), NNT 17.
• Funding: Not stated in recorded text.

a reduced incidence of acute otitis media (four RCTs, n=1805, RR 0.76, 95% CI 0.64–0.91, fixed effects model, NNT 17, 95% CI 11–46)

In a small open-label safety study, healthy older adults taking LGG had no serious adverse events; most events were mild. (Source 2)

• Randomized trial, Low certainty.
• Size: 15 volunteers.
• Who: Healthy adults aged 66-80.
• How long: 28 days.
• Result: 47 events, 39 (83%) mild, no serious events.
• Funding: Independent (NIH/NCCAM, per the page)

The 15 volunteers had a total of 47 events (range 1–7 per volunteer), 39 (83%) of which were rated as mild

What the evidence does not support

In adults taking antibiotics, pooled trials did not show a significant benefit of LGG against antibiotic-associated diarrhoea. (Source 5)

• Meta-analysis, Low certainty.
• Size: 6 RCTs, 863 adults.
• Who: Adults receiving antibiotics.
• How long: Varied.
• Result: Not significant in adults (six RCTs, n = 863); the adult risk ratio differs between the abstract copies we could reach, so it is not reported here.
• Funding: Not stated in recorded text.

In adults, the difference was not significant

In a large US trial, a 5-day course of LGG did not reduce moderate-to-severe gastroenteritis in preschool children compared with placebo (11.8% vs 12.6%). (Source 7)

• Randomized trial, High certainty.
• Size: 971 children randomised, 943 completed.
• Who: Children (median age 1.4 years) seen in US emergency departments with acute gastroenteritis.
• How long: 5 days of treatment, 1 month follow-up.
• Result: RR 0.96, 95% CI 0.68 to 1.35; P=0.83.
• Funding: Independent (NICHD and others)

Among preschool children with acute gastroenteritis, those who received a 5-day course of L. rhamnosus GG did not have better outcomes than those who received placebo.

The companion 2018 Canadian trial, often cited alongside the LGG trial, tested a different product (L. rhamnosus R0011 plus L. helveticus R0052), not LGG; it also found no benefit. (Source 8)

• Randomized trial, High certainty.
• Size: 886 children enrolled; 827 analysed for primary outcome.
• Who: Children 3 to 48 months with gastroenteritis in six Canadian pediatric emergency departments.
• How long: 5 days of treatment, 14-day outcome.
• Result: 26.1% vs 24.7%; OR 1.06, 95% CI 0.77 to 1.46; P=0.72.
• Funding: Independent (Canadian Institutes of Health Research and others)

a combination probiotic product containing Lactobacillus rhamnosus R0011 and L. helveticus R0052

Where the evidence is mixed

For childhood irritable bowel syndrome, a 2024 meta-analysis found no statistically significant difference in treatment success with LGG, though pain episodes were fewer in two small trials. (Source 9)

• Meta-analysis, Low certainty.
• Size: 252 children in included RCTs; 117 in the treatment-success comparison.
• Who: Children aged 0-18 with IBS.
• How long: Varied.
• Result: Treatment success 65% vs 31.57%, p = 0.2 (not significant)
• Funding: Not stated in recorded text.

Limit of this finding: The review reports 65% vs 31.57% success yet calls the difference not significant (p = 0.2); a gap this large in raw numbers would usually be significant, so the pooled analysis presumably differs from the simple counts, and the paper does not explain it. 18 of 57 is 31.58%, not 31.57%. The pain result also labels the mean difference of –1.50 as a '95% CI'. Treat both results as uncertain, from few small trials.

Success of treatment was reported in 39 patients out of 60 from the LGG group (65%), compared with 18 out of 57 in the placebo group (31.57%), but the difference was not statistically significant (p = 0.2).

Where the research disagrees

Whether LGG helps acute gastroenteritis in children

• Mantegazza et al. 2018 review, narrative review (antibiotic-associated diarrhoea context): a specific recommendation for the use of Lactobacillus Rhamnosus GG (LGG) and Saccharomyces boulardii (Sb) emerges (Source 10)
• Schnadower et al. 2018 (NEJM), multicentre RCT, 971 children: Among preschool children with acute gastroenteritis, those who received a 5-day course of L. rhamnosus GG did not have better outcomes than those who received placebo. (Source 7)

How much

• Reference intake: No recommended intake exists; LGG is not a nutrient. The only official statements we recorded are safety cautions (EMA for a different organism, S. boulardii), not intakes. (Source 2)
• Upper limit: No upper limit has been set by any body we found. Safety data in healthy older adults come from one 15-person open-label study. (Source 2)
• Studied: The US gastroenteritis trial gave children a 5-day course of LGG at 1×10¹⁰ colony-forming units twice daily. (Source 7)

A common belief, and what the research shows

The belief: The two 2018 NEJM null trials both tested LGG, so they show one strain failing twice.

What the research shows: Only the US trial used LGG. The Canadian trial used 'a combination probiotic product containing Lactobacillus rhamnosus R0011 and L. helveticus R0052', a different product. Both were null, but each tells us about its own product.

Questions and answers

What is it?

LGG is one specific strain of lactic-acid bacteria, first isolated from the human gut in 1983. It is sold in capsules and some dairy products. Other L. rhamnosus strains are not the same thing. (Source 2)

What does it do in the body?

In trials it passes through the gut during use. Pooled studies suggest it cut antibiotic-associated diarrhoea in children, but not significantly in adults, and it did not help children with acute gastroenteritis in a large US trial. (Source 5)

Is it good or bad for you?

Generally well tolerated in healthy people. In intensive-care, critically ill or immunocompromised patients, probiotic lactobacilli have entered the bloodstream and caused infection, so the setting matters. (Source 1)

How do you get more of it?

In studies, people got LGG by taking it as a product for a set course; in the US gastroenteritis trial children were given 1×10¹⁰ colony-forming units twice daily for 5 days. This is what was studied, not a recommendation. (Source 7)

If it is harmful, what reduces it?

Not usually relevant: LGG is not considered harmful in healthy people. In the rare cases of LGG bloodstream infection, the reports concern vulnerable patients. (Source 2)

Why might someone be low in it or missing it?

LGG is one named strain; most people will not carry that exact strain unless they consume it. Antibiotics disturb gut bacteria in general, which is why probiotics are often taken alongside them. (Source 4)

Which whole foods contain it or feed it?

We found no study in our search that measured LGG in ordinary whole foods; it reaches people through supplements and products that add it. (Source 2)

We searched: Web searches and publisher pages for LGG food sources; only the Hibberd 2014 safety study and reviews of supplement use were retrieved.

What happens if you do not have it?

No study we found shows harm from not having LGG. Not taking a probiotic during antibiotics means missing a possible modest reduction in diarrhoea in children, per pooled trials. (Source 5)

We searched: Searched for consequences of absence of LGG; none found. Closest evidence is the Szajewska 2015 meta-analysis of prevention trials.

How can you test for it?

No validated consumer or clinical test for carrying LGG was found. Lab DNA methods can identify strains in products, and product labels do not always match contents (shown for bifidobacteria). (Source 11)

We searched: Searched for stool testing of LGG; found only product-verification work (Lewis 2016, bifidobacteria).

References

1. Nature Medicine. Genomic and epidemiological evidence of bacterial transmission from probiotic capsule to blood in ICU patients. 2019. DOI 10.1038/s41591-019-0626-9. Read the source
2. PLOS ONE. No Evidence of Harms of Probiotic Lactobacillus rhamnosus GG ATCC 53103 in Healthy Elderly - A Phase I Open Label Study to Assess Safety, Tolerability and Cytokine Responses. 2014. DOI 10.1371/journal.pone.0113456. Read the source
3. The Lancet. Probiotic prophylaxis in predicted severe acute pancreatitis: a randomised, double-blind, placebo-controlled trial. 2008. PMID 18279948, DOI 10.1016/S0140-6736(08)60207-X. Read the source
4. Cell. Post-Antibiotic Gut Mucosal Microbiome Reconstitution Is Impaired by Probiotics and Improved by Autologous FMT. 2018. PMID 30193113, DOI 10.1016/j.cell.2018.08.047. Read the source
5. Alimentary Pharmacology & Therapeutics. Systematic review with meta-analysis: Lactobacillus rhamnosus GG in the prevention of antibiotic-associated diarrhoea in children and adults. 2015. PMID 26365389, DOI 10.1111/apt.13404. Read the source
6. Indian Pediatrics. Lactobacillus rhamnosus GG supplementation for preventing respiratory infections in children: a meta-analysis of randomized, placebo-controlled trials. 2013. DOI 10.1007/s13312-013-0123-z. Read the source
7. New England Journal of Medicine. Lactobacillus rhamnosus GG versus Placebo for Acute Gastroenteritis in Children. 2018. PMID 30462938, DOI 10.1056/NEJMoa1802598. Read the source
8. New England Journal of Medicine. Multicenter Trial of a Combination Probiotic for Children with Gastroenteritis. 2018. PMID 30462939, DOI 10.1056/NEJMoa1802597. Read the source
9. Pediatria Polska. Lactobacillus rhamnosus for treating irritable bowel syndrome in children - a systematic review with meta-analysis. 2024. Read the source
10. Pharmacological Research. Probiotics and antibiotic-associated diarrhea in children: A review and new evidence on Lactobacillus rhamnosus GG during and after antibiotic treatment. 2018. DOI 10.1016/j.phrs.2017.08.001. Read the source
11. Pediatric Research. Validating bifidobacterial species and subspecies identity in commercial probiotic products. 2016. DOI 10.1038/pr.2015.244. Read the source