Probiotics in irritable bowel syndrome: Has the time arrived?

Abstract

Whorwell PJ, Altringer L, Morel J, Bond Y, Charbonneau D, O’Mahony L, Kiely B, Shanahan F, Quigley EMM (Department of Medicine, University of Manchester, Manchester, UK). Efficacy of an encapsulated probiotic Bifidobacterium infantis 35624 in women with irritable bowel syndrome. Am J Gastroenterol 2006;101:1581–1590.Probiotics are live microorganisms that, upon ingestion in certain amounts, benefit the host beyond their inherent basic nutrition (Am J Clin Nutr 2001;73[Suppl]:361S–364S). Probiotics have been shown to have beneficial effects on various aspects of human health (J Clin Gastroenterol 2006;40:275–278) and their success in treating several gastrointestinal (GI) disorders, including childhood diarrhea, travelers’ diarrhea, and certain chronic inflammatory bowel diseases has led to increased interest in their use in patients with irritable bowel syndrome (IBS). The rationale for using probiotics in the treatment of IBS is based on epidemiologic studies suggesting a role for intestinal infection/inflammation in the pathogenesis of the disorder (BMJ 1999;318:565–566; Gastroenterology 2003;124:1662–1671; Gastroenterology 2002;122:1778–1783; Gut 2003;52:523–526), and anecdotal observations demonstrating alterations in stool cultures in patients with IBS compared with controls (Microbiologica 1982;5:185–194; Am J Gastroenterol 2005;100:373–382). Furthermore, over the last 2 years there has been a significant increase in the marketing and availability of dairy foods containing probiotic bacteria and various over-the-counter probiotics products claiming beneficial effects for IBS and other GI dysfunctions.Despite growing interest in and availability of probiotic foods and products, empirical data supporting the benefits of probiotics for IBS is limited. A recent multicenter European study by Whorwell et al is the largest published study on this issue to date (Am J Gastroenterol 2006;101:1581–1590). In this well-designed study, Whorwell et al investigated the efficacy of an encapsulated probiotic Bifidobacterium infantis 35624 in 362 women who met Rome II criteria for IBS. Patients were recruited from 20 primary care centers across the United Kingdom. All patients were screened before entering the study by clinical history, physical examination, full blood count, and serum chemistry to exclude organic diseases and other clinically significant conditions. In a 2-week run-in phase, patients were further evaluated to include only those having average abdominal pain/discomfort score between 1 and 4 on a 0–5 scale, average Bristol Stool Form score (Gut 1992;33:818–824) between 2 and 7, and at least 1 bowel movement (BM) in the 2nd week of the run-in phase. The selected patients were randomly assigned to receive 1 of 3 different concentrations of B infantis 35624 (1 × 106, 1 × 108, 1 × 1010 colony-forming units [CFU]) or placebo, and all patients were evaluated pre- and posttreatment for the primary and secondary efficacy measures.The primary outcome measure was abdominal pain/discomfort following 4 weeks of treatment. The secondary outcome measures were a composite IBS symptom score (the sum of individual scores for the 3 cardinal IBS symptoms: abdominal pain/discomfort, bloating/distention, and BM satisfaction), the subject’s global assessment of relief of IBS symptoms, and quality of life using the IBS-QOL instrument. The investigators compared these primary and secondary measures between the intervention (the 3 different concentrations of B infantis 35624) and placebo groups. Additional analyses were done by IBS subtypes and frequency of BMs.The investigators found that at week 4 B infantis 35624 in a dose of 1 × 108 CFU was associated with a significant improvement in abdominal pain/discomfort compared with placebo (change in adjusted least squares means from baseline of −0.89 versus −0.58 in controls respectively, P = .023). This effect was translated to a therapeutic gain of 0.31. The secondary efficacy variables of bloating/distention, sense of incomplete evacuation, passage of gas, straining, and bowel habit satisfaction were also significantly better with 1 × 108 CFU B infantis 35624 compared with placebo, with a similar magnitude of therapeutic gain for symptoms (0.25–0.31). In addition, the group with 1 × 108 CFU B infantis 35624 showed an improvement in global symptom assessment that exceeded placebo by >20% (P < .02).Secondary analyses by IBS subtypes showed positive effects of treatment in both diarrhea-predominant (IBS-D) and constipation-predominant (IBS-C) IBS subtypes. The overall relief of symptoms and symptom composite scores were significantly improved in the larger IBS-D group (n = 49) and approached significance in the smaller IBS-C group (n = 18). Interestingly, the improvement in scores for bowel habit satisfaction was significant for both IBS subtypes.Analyses by BM frequency showed that patients with abnormal BM frequency at both ends (<15th percentile and >81st percentile of the frequency distribution) gained from the probiotic treatment by normalizing their BM. No statistical differences were found between the probiotic and placebo groups with regard to the change in BM frequency in patients who were at the midpoints of the distribution of BM frequency.Comparisons of the effects of various doses demonstrated that B infantis 35624, in a dose of 1 × 108 CFU, provided the greatest benefit while the 2 other dosages of probiotic (1 × 106 and 1 × 1010 CFU) were not significantly different from placebo. The lack of efficacy of the 1 × 1010 dosage level was surprising; the same probiotic strain had been demonstrated effective in a previous study when provided in a milk-based formulation (Gastroenterology 2005;128:541–551). A series of post hoc experiments revealed significant problems with the dissolution characteristics of the higher dose capsules. It was found that the 1 × 1010 CFU formulation “coagulated” into a firm glue-like mass, which interfered with its ability to dissolve. The investigators suggested that this phenomenon is explained by the intensely hygroscopic nature of this organism.The treatment was well tolerated and no significant adverse events were recorded. Only 17 subjects (<5%) withdrew from the study because of an adverse event and there were no differences in the prevalence of these events between the placebo and active treatment groups.Whorwell et al concluded that B infantis 35624 is a probiotic that specifically relieves many IBS-related symptoms and that at a dosage level of 1 × 108 CFU it can be delivered by a capsule that is amenable to widespread use. The authors also stated that the formulation problems observed with the larger dosage levels highlights the need for clinical data on the final dose and form of a probiotic product before it should be used in practice.CommentThe public’s growing interest in the potential health benefits of certain diets and food supplements, driven in great part by nutraceutical, food, and pharmaceutical companies, has led to a substantial increase in the availability and marketing of probiotic food supplements as well as over-the-counter and prescribed probiotic products to potential consumers. However, the use of probiotics as treatment for GI disorders has been directly investigated for only few GI conditions. A recent publication by Floch et al (J Clin Gastroenterol 2006;40:275–278) has provided experts’ opinion and recommendations regarding the clinical use of probiotics in GI disorders based on an evaluation of the scientific medical literature over the past 25 years. The panel endorsed the use of probiotics in the treatment of adult and childhood infectious diarrhea, antibiotic-associated diarrhea, and pouchitis. However, the authors concluded that despite the known benefits of probiotics in promoting certain health aspects, the growing evidence for the benefits of their use in certain GI disorders, and the literature on positive studies about their use in IBS, is not enough to make a definite recommendation.The study by Whorwell et al investigated the efficacy of B infantis 35624 in women with IBS and showed a positive effect for the concentration of 1 × 108 CFU in relieving abdominal pain/discomfort as well as other IBS-related symptoms. The study results are in line with a previous European study using the same probiotic bacteria recently published in Gastroenterology (Gastroenterology 2005;128:541–551). In this study, O’Mahony et al investigated the effects of 2 probiotic bacteria in 77 patients with IBS. Patients were randomized to receive either Lactobacillus salivarius UCC4331, B infantis 35624, or placebo for 8 weeks. They found a significant reduction in IBS symptom composite scores throughout the treatment period and into the washout phase for B infantis but not for L salivarius or placebo. Secondary analysis revealed improvement in specific IBS symptoms, including abdominal pain and discomfort.Although the results of these 2 studies with regard to the effect of B infantis 35624 were positive and supportive of an overall positive effect, several points of caution needs to be made. First, although statistically significant, the overall effects were clinically modest. For example, the effect of B infantis 35624 and placebo on abdominal pain/discomfort (the primary efficacy variable in Whorwell’s study) were 1.43 ± 0.10 and 1.73 ± 0.10 respectively (P < .03) on a 6-point scale from 0 = no pain/discomfort to 5 = very severe pain/discomfort. Second, there are some inconsistencies between the 2 studies. In the O’Mahony et al study, the effect of B infantis 35624 was significantly superior to placebo from the first week of the 8-week treatment period and disappeared within 1 week of discontinuation of treatment. However, in the Whorwell et al study, the probiotic effect was superior to placebo only at the end of the 4-week treatment period and continued through the 2-week washout phase.Inconsistency between studies using the same bacteria for IBS is not unique to B infantis 35624. Another example is the results of 2 studies conducted at the Mayo Clinic in Rochester, Minnesota. In the first study, Kim et al investigated the effects of VSL #3 powder on 25 patients with IBS-D for 8 weeks (Aliment Pharmacol Ther 2003;17:895–904). Although the intention-to-treat primary analysis was negative and did not show differences in responders rates between the VSL#3 and placebo groups, secondary analysis revealed statistically significant improvement in abdominal bloating in the VSL#3 group (P = .046) but not in the placebo group (P = .54). However, a follow-up study using the same product (VSL #3) focusing on IBS patients with abdominal bloating showed no effects on abdominal bloating and only some decrease in flatulence (Neurogastroenterol Motil 2005;17:687–696). Furthermore, unlike the first study that showed no significant pre- to post-therapy differences in GI transit measurements between the 2 treatment groups, the second study revealed a positive effect of VSL #3 on increasing colonic transit time. Furthermore, the Whorwell et al study demonstrated that using similar bacteria at a similar concentration/dosage (1 × 1010 CFU) does not guarantee similar results; significant efficacy differences might be related to the formulation and methods of delivery. Thus, reported studies using probiotics for IBS even when using similar product, at similar concentrations, and conducted by the same group of investigators, can show inconsistent results. In addition to the study limitations noted by the investigators, these inconsistencies between studies in IBS patients may also relate to the heterogeneity of the disorder and other confounders, such as unknown baseline intestinal microflora, difficulties in controlling for differences in food and unintentional probiotics consumption during the study period, and intestinal motility (all possibly relevant to the effects of supplemental probiotics). Despite these inconsistencies and limitations, most of the recently published studies, including the study by Whorwell et al, were considered positive and included suggestions regarding the use of probiotics in IBS. However, at least at this point, it seems that the published data on this issue do not yet provide the sufficient scientific evidence needed to support a general recommendation about the use of probiotics in the treatment of IBS. Until such evidence is provided, we believe the use of probiotics for the treatment of IBS should still be considered investigational. However, the results of the studies reviewed in this paper are encouraging and in light of the currently limited availability of effective treatments for IBS, the public interest in the use of probiotics, and their overall safety, clinicians may consider trying the use of probiotics for IBS on an individual basis. However, when doing so, clinicians should pay attention to the following points: (1) to date there is no conclusive scientific evidence for the effectiveness of probiotics use in IBS; (2) not all probiotics are alike; different bacteria and strains may have different effects, as shown in the study by O’Mahony et al (Gastroenterology 2005;128:541–551); (3) most of the available/marked probiotics products have not been tested for the accuracy of their claimed bacterial content, shelf stability, effectiveness of delivery, and intestinal survival; (4) only a very few probiotic products have been tested with regard to their effects on IBS; and (5) the marketed products are not covered by insurers and therefore they may pose a significant cost/expense for patients. The controlled clinical trial by Whorwell et al addresses some of these points and it is therefore a good example of what clinicians need to look for when considering the use of a specific probiotic product (specific information on the efficacy of a final product in a defined group of patients). Finally, currently reported data emphasize the need for the development of quality control standards for probiotic products as well as additional, well-designed, controlled clinical trials with the final formulation before various probiotic products are recommended in clinical practice. Whorwell PJ, Altringer L, Morel J, Bond Y, Charbonneau D, O’Mahony L, Kiely B, Shanahan F, Quigley EMM (Department of Medicine, University of Manchester, Manchester, UK). Efficacy of an encapsulated probiotic Bifidobacterium infantis 35624 in women with irritable bowel syndrome. Am J Gastroenterol 2006;101:1581–1590. Probiotics are live microorganisms that, upon ingestion in certain amounts, benefit the host beyond their inherent basic nutrition (Am J Clin Nutr 2001;73[Suppl]:361S–364S). Probiotics have been shown to have beneficial effects on various aspects of human health (J Clin Gastroenterol 2006;40:275–278) and their success in treating several gastrointestinal (GI) disorders, including childhood diarrhea, travelers’ diarrhea, and certain chronic inflammatory bowel diseases has led to increased interest in their use in patients with irritable bowel syndrome (IBS). The rationale for using probiotics in the treatment of IBS is based on epidemiologic studies suggesting a role for intestinal infection/inflammation in the pathogenesis of the disorder (BMJ 1999;318:565–566; Gastroenterology 2003;124:1662–1671; Gastroenterology 2002;122:1778–1783; Gut 2003;52:523–526), and anecdotal observations demonstrating alterations in stool cultures in patients with IBS compared with controls (Microbiologica 1982;5:185–194; Am J Gastroenterol 2005;100:373–382). Furthermore, over the last 2 years there has been a significant increase in the marketing and availability of dairy foods containing probiotic bacteria and various over-the-counter probiotics products claiming beneficial effects for IBS and other GI dysfunctions. Despite growing interest in and availability of probiotic foods and products, empirical data supporting the benefits of probiotics for IBS is limited. A recent multicenter European study by Whorwell et al is the largest published study on this issue to date (Am J Gastroenterol 2006;101:1581–1590). In this well-designed study, Whorwell et al investigated the efficacy of an encapsulated probiotic Bifidobacterium infantis 35624 in 362 women who met Rome II criteria for IBS. Patients were recruited from 20 primary care centers across the United Kingdom. All patients were screened before entering the study by clinical history, physical examination, full blood count, and serum chemistry to exclude organic diseases and other clinically significant conditions. In a 2-week run-in phase, patients were further evaluated to include only those having average abdominal pain/discomfort score between 1 and 4 on a 0–5 scale, average Bristol Stool Form score (Gut 1992;33:818–824) between 2 and 7, and at least 1 bowel movement (BM) in the 2nd week of the run-in phase. The selected patients were randomly assigned to receive 1 of 3 different concentrations of B infantis 35624 (1 × 106, 1 × 108, 1 × 1010 colony-forming units [CFU]) or placebo, and all patients were evaluated pre- and posttreatment for the primary and secondary efficacy measures. The primary outcome measure was abdominal pain/discomfort following 4 weeks of treatment. The secondary outcome measures were a composite IBS symptom score (the sum of individual scores for the 3 cardinal IBS symptoms: abdominal pain/discomfort, bloating/distention, and BM satisfaction), the subject’s global assessment of relief of IBS symptoms, and quality of life using the IBS-QOL instrument. The investigators compared these primary and secondary measures between the intervention (the 3 different concentrations of B infantis 35624) and placebo groups. Additional analyses were done by IBS subtypes and frequency of BMs. The investigators found that at week 4 B infantis 35624 in a dose of 1 × 108 CFU was associated with a significant improvement in abdominal pain/discomfort compared with placebo (change in adjusted least squares means from baseline of −0.89 versus −0.58 in controls respectively, P = .023). This effect was translated to a therapeutic gain of 0.31. The secondary efficacy variables of bloating/distention, sense of incomplete evacuation, passage of gas, straining, and bowel habit satisfaction were also significantly better with 1 × 108 CFU B infantis 35624 compared with placebo, with a similar magnitude of therapeutic gain for symptoms (0.25–0.31). In addition, the group with 1 × 108 CFU B infantis 35624 showed an improvement in global symptom assessment that exceeded placebo by >20% (P < .02). Secondary analyses by IBS subtypes showed positive effects of treatment in both diarrhea-predominant (IBS-D) and constipation-predominant (IBS-C) IBS subtypes. The overall relief of symptoms and symptom composite scores were significantly improved in the larger IBS-D group (n = 49) and approached significance in the smaller IBS-C group (n = 18). Interestingly, the improvement in scores for bowel habit satisfaction was significant for both IBS subtypes. Analyses by BM frequency showed that patients with abnormal BM frequency at both ends (<15th percentile and >81st percentile of the frequency distribution) gained from the probiotic treatment by normalizing their BM. No statistical differences were found between the probiotic and placebo groups with regard to the change in BM frequency in patients who were at the midpoints of the distribution of BM frequency. Comparisons of the effects of various doses demonstrated that B infantis 35624, in a dose of 1 × 108 CFU, provided the greatest benefit while the 2 other dosages of probiotic (1 × 106 and 1 × 1010 CFU) were not significantly different from placebo. The lack of efficacy of the 1 × 1010 dosage level was surprising; the same probiotic strain had been demonstrated effective in a previous study when provided in a milk-based formulation (Gastroenterology 2005;128:541–551). A series of post hoc experiments revealed significant problems with the dissolution characteristics of the higher dose capsules. It was found that the 1 × 1010 CFU formulation “coagulated” into a firm glue-like mass, which interfered with its ability to dissolve. The investigators suggested that this phenomenon is explained by the intensely hygroscopic nature of this organism. The treatment was well tolerated and no significant adverse events were recorded. Only 17 subjects (<5%) withdrew from the study because of an adverse event and there were no differences in the prevalence of these events between the placebo and active treatment groups. Whorwell et al concluded that B infantis 35624 is a probiotic that specifically relieves many IBS-related symptoms and that at a dosage level of 1 × 108 CFU it can be delivered by a capsule that is amenable to widespread use. The authors also stated that the formulation problems observed with the larger dosage levels highlights the need for clinical data on the final dose and form of a probiotic product before it should be used in practice. CommentThe public’s growing interest in the potential health benefits of certain diets and food supplements, driven in great part by nutraceutical, food, and pharmaceutical companies, has led to a substantial increase in the availability and marketing of probiotic food supplements as well as over-the-counter and prescribed probiotic products to potential consumers. However, the use of probiotics as treatment for GI disorders has been directly investigated for only few GI conditions. A recent publication by Floch et al (J Clin Gastroenterol 2006;40:275–278) has provided experts’ opinion and recommendations regarding the clinical use of probiotics in GI disorders based on an evaluation of the scientific medical literature over the past 25 years. The panel endorsed the use of probiotics in the treatment of adult and childhood infectious diarrhea, antibiotic-associated diarrhea, and pouchitis. However, the authors concluded that despite the known benefits of probiotics in promoting certain health aspects, the growing evidence for the benefits of their use in certain GI disorders, and the literature on positive studies about their use in IBS, is not enough to make a definite recommendation.The study by Whorwell et al investigated the efficacy of B infantis 35624 in women with IBS and showed a positive effect for the concentration of 1 × 108 CFU in relieving abdominal pain/discomfort as well as other IBS-related symptoms. The study results are in line with a previous European study using the same probiotic bacteria recently published in Gastroenterology (Gastroenterology 2005;128:541–551). In this study, O’Mahony et al investigated the effects of 2 probiotic bacteria in 77 patients with IBS. Patients were randomized to receive either Lactobacillus salivarius UCC4331, B infantis 35624, or placebo for 8 weeks. They found a significant reduction in IBS symptom composite scores throughout the treatment period and into the washout phase for B infantis but not for L salivarius or placebo. Secondary analysis revealed improvement in specific IBS symptoms, including abdominal pain and discomfort.Although the results of these 2 studies with regard to the effect of B infantis 35624 were positive and supportive of an overall positive effect, several points of caution needs to be made. First, although statistically significant, the overall effects were clinically modest. For example, the effect of B infantis 35624 and placebo on abdominal pain/discomfort (the primary efficacy variable in Whorwell’s study) were 1.43 ± 0.10 and 1.73 ± 0.10 respectively (P < .03) on a 6-point scale from 0 = no pain/discomfort to 5 = very severe pain/discomfort. Second, there are some inconsistencies between the 2 studies. In the O’Mahony et al study, the effect of B infantis 35624 was significantly superior to placebo from the first week of the 8-week treatment period and disappeared within 1 week of discontinuation of treatment. However, in the Whorwell et al study, the probiotic effect was superior to placebo only at the end of the 4-week treatment period and continued through the 2-week washout phase.Inconsistency between studies using the same bacteria for IBS is not unique to B infantis 35624. Another example is the results of 2 studies conducted at the Mayo Clinic in Rochester, Minnesota. In the first study, Kim et al investigated the effects of VSL #3 powder on 25 patients with IBS-D for 8 weeks (Aliment Pharmacol Ther 2003;17:895–904). Although the intention-to-treat primary analysis was negative and did not show differences in responders rates between the VSL#3 and placebo groups, secondary analysis revealed statistically significant improvement in abdominal bloating in the VSL#3 group (P = .046) but not in the placebo group (P = .54). However, a follow-up study using the same product (VSL #3) focusing on IBS patients with abdominal bloating showed no effects on abdominal bloating and only some decrease in flatulence (Neurogastroenterol Motil 2005;17:687–696). Furthermore, unlike the first study that showed no significant pre- to post-therapy differences in GI transit measurements between the 2 treatment groups, the second study revealed a positive effect of VSL #3 on increasing colonic transit time. Furthermore, the Whorwell et al study demonstrated that using similar bacteria at a similar concentration/dosage (1 × 1010 CFU) does not guarantee similar results; significant efficacy differences might be related to the formulation and methods of delivery. Thus, reported studies using probiotics for IBS even when using similar product, at similar concentrations, and conducted by the same group of investigators, can show inconsistent results. In addition to the study limitations noted by the investigators, these inconsistencies between studies in IBS patients may also relate to the heterogeneity of the disorder and other confounders, such as unknown baseline intestinal microflora, difficulties in controlling for differences in food and unintentional probiotics consumption during the study period, and intestinal motility (all possibly relevant to the effects of supplemental probiotics). Despite these inconsistencies and limitations, most of the recently published studies, including the study by Whorwell et al, were considered positive and included suggestions regarding the use of probiotics in IBS. However, at least at this point, it seems that the published data on this issue do not yet provide the sufficient scientific evidence needed to support a general recommendation about the use of probiotics in the treatment of IBS. Until such evidence is provided, we believe the use of probiotics for the treatment of IBS should still be considered investigational. However, the results of the studies reviewed in this paper are encouraging and in light of the currently limited availability of effective treatments for IBS, the public interest in the use of probiotics, and their overall safety, clinicians may consider trying the use of probiotics for IBS on an individual basis. However, when doing so, clinicians should pay attention to the following points: (1) to date there is no conclusive scientific evidence for the effectiveness of probiotics use in IBS; (2) not all probiotics are alike; different bacteria and strains may have different effects, as shown in the study by O’Mahony et al (Gastroenterology 2005;128:541–551); (3) most of the available/marked probiotics products have not been tested for the accuracy of their claimed bacterial content, shelf stability, effectiveness of delivery, and intestinal survival; (4) only a very few probiotic products have been tested with regard to their effects on IBS; and (5) the marketed products are not covered by insurers and therefore they may pose a significant cost/expense for patients. The controlled clinical trial by Whorwell et al addresses some of these points and it is therefore a good example of what clinicians need to look for when considering the use of a specific probiotic product (specific information on the efficacy of a final product in a defined group of patients). Finally, currently reported data emphasize the need for the development of quality control standards for probiotic products as well as additional, well-designed, controlled clinical trials with the final formulation before various probiotic products are recommended in clinical practice. The public’s growing interest in the potential health benefits of certain diets and food supplements, driven in great part by nutraceutical, food, and pharmaceutical companies, has led to a substantial increase in the availability and marketing of probiotic food supplements as well as over-the-counter and prescribed probiotic products to potential consumers. However, the use of probiotics as treatment for GI disorders has been directly investigated for only few GI conditions. A recent publication by Floch et al (J Clin Gastroenterol 2006;40:275–278) has provided experts’ opinion and recommendations regarding the clinical use of probiotics in GI disorders based on an evaluation of the scientific medical literature over the past 25 years. The panel endorsed the use of probiotics in the treatment of adult and childhood infectious diarrhea, antibiotic-associated diarrhea, and pouchitis. However, the authors concluded that despite the known benefits of probiotics in promoting certain health aspects, the growing evidence for the benefits of their use in certain GI disorders, and the literature on positive studies about their use in IBS, is not enough to make a definite recommendation. The study by Whorwell et al investigated the efficacy of B infantis 35624 in women with IBS and showed a positive effect for the concentration of 1 × 108 CFU in relieving abdominal pain/discomfort as well as other IBS-related symptoms. The study results are in line with a previous European study using the same probiotic bacteria recently published in Gastroenterology (Gastroenterology 2005;128:541–551). In this study, O’Mahony et al investigated the effects of 2 probiotic bacteria in 77 patients with IBS. Patients were randomized to receive either Lactobacillus salivarius UCC4331, B infantis 35624, or placebo for 8 weeks. They found a significant reduction in IBS symptom composite scores throughout the treatment period and into the washout phase for B infantis but not for L salivarius or placebo. Secondary analysis revealed improvement in specific IBS symptoms, including abdominal pain and discomfort. Although the results of these 2 studies with regard to the effect of B infantis 35624 were positive and supportive of an overall positive effect, several points of caution needs to be made. First, although statistically significant, the overall effects were clinically modest. For example, the effect of B infantis 35624 and placebo on abdominal pain/discomfort (the primary efficacy variable in Whorwell’s study) were 1.43 ± 0.10 and 1.73 ± 0.10 respectively (P < .03) on a 6-point scale from 0 = no pain/discomfort to 5 = very severe pain/discomfort. Second, there are some inconsistencies between the 2 studies. In the O’Mahony et al study, the effect of B infantis 35624 was significantly superior to placebo from the first week of the 8-week treatment period and disappeared within 1 week of discontinuation of treatment. However, in the Whorwell et al study, the probiotic effect was superior to placebo only at the end of the 4-week treatment period and continued through the 2-week washout phase. Inconsistency between studies using the same bacteria for IBS is not unique to B infantis 35624. Another example is the results of 2 studies conducted at the Mayo Clinic in Rochester, Minnesota. In the first study, Kim et al investigated the effects of VSL #3 powder on 25 patients with IBS-D for 8 weeks (Aliment Pharmacol Ther 2003;17:895–904). Although the intention-to-treat primary analysis was negative and did not show differences in responders rates between the VSL#3 and placebo groups, secondary analysis revealed statistically significant improvement in abdominal bloating in the VSL#3 group (P = .046) but not in the placebo group (P = .54). However, a follow-up study using the same product (VSL #3) focusing on IBS patients with abdominal bloating showed no effects on abdominal bloating and only some decrease in flatulence (Neurogastroenterol Motil 2005;17:687–696). Furthermore, unlike the first study that showed no significant pre- to post-therapy differences in GI transit measurements between the 2 treatment groups, the second study revealed a positive effect of VSL #3 on increasing colonic transit time. Furthermore, the Whorwell et al study demonstrated that using similar bacteria at a similar concentration/dosage (1 × 1010 CFU) does not guarantee similar results; significant efficacy differences might be related to the formulation and methods of delivery. Thus, reported studies using probiotics for IBS even when using similar product, at similar concentrations, and conducted by the same group of investigators, can show inconsistent results. In addition to the study limitations noted by the investigators, these inconsistencies between studies in IBS patients may also relate to the heterogeneity of the disorder and other confounders, such as unknown baseline intestinal microflora, difficulties in controlling for differences in food and unintentional probiotics consumption during the study period, and intestinal motility (all possibly relevant to the effects of supplemental probiotics). Despite these inconsistencies and limitations, most of the recently published studies, including the study by Whorwell et al, were considered positive and included suggestions regarding the use of probiotics in IBS. However, at least at this point, it seems that the published data on this issue do not yet provide the sufficient scientific evidence needed to support a general recommendation about the use of probiotics in the treatment of IBS. Until such evidence is provided, we believe the use of probiotics for the treatment of IBS should still be considered investigational. However, the results of the studies reviewed in this paper are encouraging and in light of the currently limited availability of effective treatments for IBS, the public interest in the use of probiotics, and their overall safety, clinicians may consider trying the use of probiotics for IBS on an individual basis. However, when doing so, clinicians should pay attention to the following points: (1) to date there is no conclusive scientific evidence for the effectiveness of probiotics use in IBS; (2) not all probiotics are alike; different bacteria and strains may have different effects, as shown in the study by O’Mahony et al (Gastroenterology 2005;128:541–551); (3) most of the available/marked probiotics products have not been tested for the accuracy of their claimed bacterial content, shelf stability, effectiveness of delivery, and intestinal survival; (4) only a very few probiotic products have been tested with regard to their effects on IBS; and (5) the marketed products are not covered by insurers and therefore they may pose a significant cost/expense for patients. The controlled clinical trial by Whorwell et al addresses some of these points and it is therefore a good example of what clinicians need to look for when considering the use of a specific probiotic product (specific information on the efficacy of a final product in a defined group of patients). Finally, currently reported data emphasize the need for the development of quality control standards for probiotic products as well as additional, well-designed, controlled clinical trials with the final formulation before various probiotic products are recommended in clinical practice. ReplyGastroenterologyVol. 132Issue 2PreviewWe thank Drs Ringel-Kulka and Ringel for their insightful comments on our study (Am J Gastroenterol 2006;101:1581–1590). They raise several issues that deserve further comment. Full-Text PDF