Research project led by Bailey Hanna, MS, RDN - no affiliate links, ads or sponsored products.
Lactobacillus plantarum 299v is one of the most studied probiotics for IBS, with seven controlled trials examining its effects. However, the quality of evidence from these studies was so poor that we disqualified five out of the seven studies. In fact, we could have reasonably made the case to disqualify all seven. Each disqualification was due to multiple issues, which are detailed in the “Notes” section of each study’s findings.
Study by Moeen-Ul-Haq, et al.
This study, one of the two we did not disqualify, administered L. plantarum 299v at a dose of 50 billion CFU/day for 4 weeks. We identified discrepancies in the reported results due to statistical calculation errors. While the authors reported no significant differences between the probiotic and placebo groups for incomplete rectal emptying and abdominal pain, our independent analysis found small to moderate significant beneficial effects for both parameters:
Additionally, the probiotic had neutral effects compared to placebo for bloating severity, and all reported subtype-specific results in this study were underpowered.
Study by Nobaek S, et al.
This study, although still of poor evidence quality, provided L. plantarum 299v at a dose of 20 million CFU/day for 4 weeks through a rose hip beverage, a delivery matrix which may have influenced the results. No significant benefits were reported for overall GI function, abdominal pain, or bowel habits. However, flatulence decreased significantly in the probiotic group, with an independently calculated effect size of 0.73, falling within our moderate effect size range.
Summary
The evidence quality for L. plantarum 299v in treating IBS is poor, and the results are often conflicting. The evaluated doses vary significantly between studies, making it difficult to determine an ideal dose or form of this probiotic for IBS. While it’s possible that L. plantarum 299v might help with abdominal pain, flatulence, and constipation, the evidence base is weak. Therefore, we cannot confidently recommend this probiotic for IBS until more well-controlled trials are conducted.
Key Takeaways
Currently, the available evidence is insufficient to recommend the use of L. plantarum 299v for IBS.
Dosing Instructions:
None
References:
There is insufficient evidence demonstrating benefits for IBS with this probiotic. Therefore, we do not provide guidance on finding a commercial product containing this strain/blend.
This probiotic has low to moderate quality studies in IBS populations supporting it. View our evidence evaluation framework to learn how we assess the quality of studies.
Add condition
Clinical trial: lactobacillus plantarum 299v (DSM 9843) improves symptoms of irritable bowel syndrome
World J Gastroenterol 2012;18(30):4012–8.
ADULTS 18-70 years
Randomized clinical trial: effect of Lactobacillus plantarum 299 v on symptoms of irritable bowel syndrome
Nutrition 2014;30(10):1151–7
ADULTS >/= 18 years
Role of Lactobacillus plantarum 299v versus Placebo in symptomatic improvement of irritable bowel syndrome patients
JPMA 72:404; 2022
ADULTS
Probiotic effect and dietary correlations on faecal microbiota profiles inirritable bowel syndrome
South African Journal of Clinical Nutrition2021; 34(3):84–89
No data
Feeling of Incomplete Evacuation Frequency: Means and standard deviations (SD) were not provided numerically for incomplete evacuation frequency or severity; instead, mean baseline scores were represented in charts without mentioning standard deviations.
The approximate baseline mean score for the probiotic group was 2.25, and 2.17 for the placebo group.
At the end of treatment (EOT), the probiotic group’s approximate mean score was 1.0, indicating a 55.5% reduction (improvement) from baseline, while the placebo group’s approximate mean score was 1.65, indicating a 24% reduction (improvement).
Weekly analysis showed significantly lower scores at weeks 2, 3, and 4 in the L. plantarum 299v (DSM 9843) group compared to baseline (p<0.05).
Feeling of Incomplete Evacuation Severity: The approximate baseline mean score for the probiotic group was 1.25, and 1.32 for the placebo group. At the end of treatment (EOT), the probiotic group’s approximate mean score was 0.6, representing a 52% reduction (improvement) from baseline, while the placebo group’s approximate mean score was 1.07, representing a 19% reduction (improvement).
Weekly analysis showed significantly lower scores at weeks 3 and 4 in the L. plantarum 299v (DSM 9843) group compared to baseline (p<0.05).
Measures of incomplete evacuation could be considered a characteristic of IBS subtypes experiencing constipation, and a predominantly IBS-D population may have reduced mean scores for this symptom.
The absence of standard deviations for baseline and post-treatment incomplete evacuation measures prevents the determination of data variance.
Ideally, parameters such as incomplete evacuation would have been assessed in subgroups of individuals experiencing these symptoms at baseline to better assess the probiotic’s impact.
Notably, there was no assessment of stool consistency in the study population. Therefore, potential benefits for individuals with IBS experiencing constipation should be interpreted cautiously.
At baseline, the average stool frequency per day was 3.94 ± 1.51 for the probiotic group and 3.69 ± 1.34 for the placebo group. By the end of the treatment (EOT), the approximate mean stool frequency for the probiotic group decreased to 2.35, representing a 40.4% reduction from baseline. This translates to an absolute reduction of approximately 1.6 stools per day in an IBS-D predominant population.
Conversely, the placebo group showed a smaller reduction, with an approximate mean stool frequency of 3.0 at EOT, marking an 18.7% reduction from baseline and an absolute decrease of about 0.7 stools per day. A weekly analysis revealed a significantly lower stool frequency at weeks 2, 3, and 4 for the L. plantarum 299v (DSM 9843) group compared to baseline (p<0.05).
The percentage of patients who rated the treatment efficacy as good or excellent was significantly higher in the L. plantarum 299v (DSM 9843) group compared to the placebo group (78.1% vs. 8.1%).
Similar results were observed in the investigators’ assessments, with 82.8% in the probiotic group and 11.1% in the placebo group rating the treatment as good or excellent. The overall efficacy of the treatment was evaluated using a 4-point scale ranging from “poor” to “excellent.”
Frequency of Abdominal Pain Episodes: At baseline, the probiotic group had a mean score of 2.1 ± 1.01, while the placebo group had a mean score of 1.98 ± 0.91. By the end of treatment (EOT), the probiotic group experienced a mean percentage reduction from baseline of 51.9%, while the placebo group saw a 13.6% reduction.
The differences between groups were statistically significant in favor of the probiotic during weeks 3 and 4 of the intervention (p<0.05).
Abdominal Pain Severity: At baseline, the probiotic group had a mean score of 1.24 ± 0.60, while the placebo group had a mean score of 1.20 ± 0.63. By the end of treatment (EOT), the probiotic group experienced a mean percentage reduction from baseline of 45.2%, while the placebo group saw a 23.3% reduction.
Weekly analysis showed significantly lower abdominal pain severity scores at weeks 2, 3, and 4 in the L. plantarum 299v (DSM 9843) group compared to the placebo group (p<0.05). The reduction in the frequency of abdominal pain was also significantly higher in the L. plantarum 299v (DSM 9843) group than in the placebo group at weeks 3 and 4.
Bloating Frequency: Means and standard deviations (SD) were not provided numerically for bloating frequency; instead, mean baseline scores were represented in charts without standard deviations.
The approximate baseline mean score for the probiotic group was 2.8, while the placebo group had a score of 2.95. At the end of treatment (EOT), the probiotic group’s approximate mean score was 0.95, representing a 66.0% reduction (improvement) from baseline.
The placebo group’s approximate mean score at EOT was 2.45, representing a 17.0% reduction (improvement). Weekly analysis showed significantly lower bloating frequency scores at weeks 2, 3, and 4 in the L. plantarum 299v (DSM 9843) group compared to the placebo group (p<0.05).
Bloating Severity: The baseline mean score for bloating severity was 1.07 ± 0.62 in the probiotic group and 1.14 ± 0.64 in the placebo group. At the end of treatment (EOT), the approximate mean score for the probiotic group was 0.5, representing a 53.3% reduction (improvement) from baseline, while the placebo group’s approximate mean score was 0.93, representing an 18.4% reduction (improvement).
Weekly analysis showed significantly lower bloating severity scores at weeks 3 and 4 in the L. plantarum 299v (DSM 9843) group compared to the placebo group (p<0.05).
The study had several limitations. Firstly, no responder thresholds were identified. Baseline standard deviations for bloating frequency and feelings of incomplete evacuation were not provided; only mean baseline values were shown on bar graphs.
Additionally, post-treatment standard deviations were omitted for all parameters, limiting our ability to detect the influence of outliers.
The study population was predominantly IBS-D participants (>60%), yet there was no subtype-specific analysis for any symptom parameter. Incomplete evacuation is often associated with IBS subtypes experiencing constipation, so the predominance of IBS-D participants may have skewed mean scores for this symptom.
Additionally, the study lacked a run-in observation period, leading to less certainty in baseline measurements, and a post-treatment observation period to assess sustained results or side effects. Although the authors reported that almost half of the participants consumed yogurt daily and claimed it did not affect the outcome, they did not substantiate this assertion.
Overall, the study demonstrated selective reporting, failing to present important aspects such as variability measures and standard presentation of ANOVA results.
For these reasons, this study has been disqualified from our database.
Bowel habits would have been measured by the Francis Severity Score tool used in this study, but results for this individual parameter were not disclosed. Of note, the total Francis Severity Score was neutral.
The Total Francis Severity Score, designed to measure abdominal pain but also assessing distention, bowel habits, and quality of life, was used to provide evidence regarding abdominal pain, though it may better represent overall IBS symptoms.
At Week 0, the Probiotic Group had a score of 259.66 ± 106.90, while the Placebo Group had a score of 256.04 ± 104.27. By the end of treatment (EOT) at Week 8, the Probiotic Group’s score decreased to 199.13 ± 119.70, reflecting a relative mean reduction of 23.3% from baseline.
The Placebo Group’s score at EOT was 201.98 ± 97.44, indicating a relative mean reduction of 21.1% from baseline. There was no statistically significant difference between the groups (p=0.800). Although multiple measurements were taken throughout the study, the statistical analysis only compared data from Week 0 and Week 8 of the intervention.
The groups were further divided into IBS-C versus placebo and IBS-D versus placebo. The study authors reported no significant differences between the IBS-C and placebo groups, or between the IBS-D and placebo groups, but did not provide specific data to support these findings.
For the Total Francis Severity Score, which is reported to measure abdominal pain but also assesses distention, bowel habits, and quality of life, only the total score was provided for the following data on abdominal pain.
At Week 0, the Probiotic Group had a score of 259.66 ± 106.90, while the Placebo Group had a score of 256.04 ± 104.27. By the end of treatment (EOT) at Week 8, the Probiotic Group’s score was 199.13 ± 119.70, indicating a relative mean reduction of 23.3% from baseline.
The Placebo Group’s score was 201.98 ± 97.44, reflecting a relative mean reduction of 21.1% from baseline. There was no statistically significant difference between the groups (p=0.800). While multiple measurements were taken throughout the study, the statistical analysis only compared data from Week 0 and Week 8 of the intervention.
The groups were further divided into IBS-C versus placebo and IBS-D versus placebo. The study authors reported no significant differences between the IBS-C versus placebo and IBS-D versus placebo groups, but no specific data were provided to support these findings.
Distention would have been measured by the Francis Severity Score tool used in this study, but results for this individual parameter were not disclosed. Of note the total Francis severity score was neutral.
For IBS Quality of Life (QOL) Total Score: At baseline (Week 0), the Probiotic Group had a mean score of 49.95 ± 21.43, while the Placebo Group had a mean score of 44.72 ± 22.22. By the end of treatment (EOT, Week 8), the Probiotic Group’s mean score decreased to 44.42 ± 23.96, indicating a relative mean reduction of 11.0% from baseline.
The Placebo Group’s mean score decreased to 32.29 ± 23.22, reflecting a relative mean reduction of 27.7% from baseline. There was no statistically significant difference in total QOL scores between the placebo and probiotic groups (p=0.687). However, the placebo group demonstrated a relative mean therapeutic gain of 16.7% over the probiotic group.
Individual parameters assessed as part of the QOL score included dysphoria, interference with activity, body image, health worry, food avoidance, social reaction, sexual function, and relationships. No statistically significant differences were observed between the groups for any individual parameter on the IBS-QOL questionnaire.
A significant limitation of the study is the drop-out rate in the treatment group, particularly among the IBS-C group. Although the number and timing of drop-outs were documented, no specific reasons were noted.
To mitigate the impact of drop-outs, the last observation for each patient while in the study was used to retain as much information as possible. However, the substantial loss of patients from the treatment group may have amplified the large placebo effect observed. The study did not establish any responder thresholds.
The total Francis Severity Score tool was utilized to assess pain; however, this tool also measures distention, bowel habits, and quality of life. The overall severity score was presented as the abdominal pain result, but individual scores for the specific parameters were not disclosed, making it unclear how participants performed on each metric.
This lack of detailed reporting limits our ability to draw conclusions about changes in pain, distention, and bowel habits. Additionally, the study was seriously underpowered, and the authors’ use of the Bonferroni correction exacerbated this issue.
Some measures exhibited anomalies, such as standard deviations larger than means, although scales had minimums of 0. The authors also informally reported observing the Hawthorne effect during the study.
Overall, this study was underpowered and had several shortcomings. Despite these issues, the authors did not account for the study being underpowered in their conclusions.
For these reasons, we have disqualified this study from the database.
Regarding the feeling of incomplete rectal emptying, the study authors reported no statistically significant differences between the overall probiotic population and the placebo group (p>0.05). However, our independent statistical analysis found that the authors miscalculated their results. We calculated a Cohen’s d of 0.55 for this parameter at the end of the study, which was statistically significant (p=0.005). This effect size falls within our moderate range of 0.5 to 1.0.
The mean frequency of abdominal pain per day decreased at the end of therapy compared to baseline; however, the study authors reported that this difference was not statistically significant when compared to the placebo group (p=0.744). Our independent analysis found this to be a miscalculation.
We calculated a Cohen’s D of 0.45 for abdominal pain frequency at the end of treatment, which was statistically significant (p=0.02). Additionally, we calculated a Cohen’s D of 0.6 for abdominal pain severity at the end of treatment, which was also statistically significant (p=0.002). The effect size for abdominal pain severity falls within our moderate range of 0.5 to 1.0, leading us to classify this as a moderate beneficial effect.
There were no statistically significant differences in bloating severity between the overall probiotic population and the placebo group (p > 0.05).
Study limitations include seemingly inconsistent statistical analyses, with a per-protocol analysis used for the probiotic group and a modified intent-to-treat analysis that included 2 of the 7 participants lost to follow-up in the placebo group, without any provided justification or discussion of the potential impacts on the results.
Additionally, the trial lacked any monitoring for adverse events and was registered retrospectively, reportedly to avoid delays in publishing, which raises concerns about the integrity of the study design. The absence of a run-in observation period further weakened the reliability of the baseline symptom scores.
Although the study included adults with IBS subtypes D, C, and M, it was underpowered for making subgroup comparisons. There were also errors in the calculation of t-tests, leading to inaccuracies in the reported statistical significances of differences between the placebo and treatment groups.
Overall, we found moderate but significant improvements for abdominal pain severity, abdominal pain frequency, and incomplete evacuation which were originally missed by the study authors due to miscalculations.
There was no significant difference in symptom severity scores between the treatment and placebo groups over the probiotic trial period (treatment group: 259.54 ± 104.59 to 197.56 ± 114.74 vs. placebo group: 258.71 ± 110.88 to 180.00 ± 96.1; p = 0.599).
The groups were further divided into IBS-C vs. placebo and IBS-D vs. placebo, with no significant differences noted. Both the treatment and placebo groups showed significant improvement in Francis Severity Scores (FSS) over the study period, from an average of 259.27 to 191.71 (p < 0.0001), indicating a large placebo effect.
Pain would have been measured by the Francis Severity Score tool used in this study, but results for this individual parameter were not disclosed. Of note the total Francis severity score was neutral.
At baseline (A), before any probiotic intervention, Lactobacillus plantarum profiles differed significantly between C-IBS and D-IBS, with lower counts in D-IBS (−0.956 ± 1.239 vs. −1.700 ± 1.239; p = 0.024). There was no significant change in bacterial counts between the groups after the trial (B) and following the washout period (C).
The participants in this study are the same individuals reported in another study we’ve disqualified by Stevenson et al. Regardless, the study is significantly underpowered, and it is unclear how many participants completed the protocol. The authors reported different participant numbers for various analyses, with attrition rates reaching up to 44%.
Due to these issues, this study has been disqualified from our database.