This systematic review and meta-analysis has shown that fibre, antispasmodics, and peppermint oil are all more effective than placebo in the treatment of irritable bowel syndrome. The number needed to treat to prevent one patient having persistent symptoms was 11 for fibre, 5 for antispasmodics, and 2.5 for peppermint oil. Adverse events were significantly more frequent in those receiving antispasmodics than in those receiving placebo, but none of these was serious. As several different treatments were studied in the included randomised controlled trials, we carried out subgroup analyses.
When type of fibre was examined, wheat bran was no more effective at treating irritable bowel syndrome than placebo or a low fibre diet. The beneficial effect of fibre seemed to be limited to ispaghula husk, with a number needed to treat of 6 compared with placebo. However, significant heterogeneity was detected between trials. When only high quality studies were considered in the analysis this heterogeneity was diminished, but the difference in effect on symptoms in favour of ispaghula husk only reached marginal statistical significance.
Antispasmodics were of benefit, but again heterogeneity between study results was significant, and there was evidence of publication bias. Data were limited for many of the drugs licensed for use in the United Kingdom, such as mebeverine, dicycloverine, and alverine. It is difficult to know whether this is a true class effect of antispasmodics. Of all the drugs studied, most data were available for otilonium, trimebutine, cimetropium, hyoscine, and pinaverium. Trimebutine seemed to have no benefit over placebo in treating irritable bowel syndrome, whereas the other four drugs all significantly reduced the risk of persistent symptoms after treatment. Considerable heterogeneity was, however, detected between individual trials using otilonium and cimetropium and, although this was not the case when studies of pinaverium were pooled, the number of included patients was small. The best evidence for an individual compound seems to be for hyoscine, the efficacy of which was studied in over 400 patients. No statistically significant heterogeneity was detected, and 3.5 patients would need to be treated to prevent symptoms persisting in one patient. It would seem reasonable for general practitioners who want to begin a trial of antispasmodics to use hyoscine as first line treatment, but to consider other antispasmodics when this strategy fails.
Peppermint oil was also superior to placebo, although statistically significant heterogeneity was detected between study results, and only four randomised controlled trials were identified including fewer than 400 patients, so data were more limited than for fibre and antispasmodics. Three of these trials scored more than 4 on the Jadad scale,w30-w32 but the treatment effect was similar when only these studies were included in the meta-analysis, and the heterogeneity observed between studies was no longer detected.
The reporting in this systematic review adheres to the quality of reporting of meta-analyses statement.29
We have also specified the search strategy used, as well as our data extraction criteria. We believe that stating the criteria for data extraction should be standard in all systematic reviews that pool dichotomous data. The combination of our comprehensive search strategy and the translation of foreign language articles enabled us to identify studies with data from over 2500 people with irritable bowel syndrome.
Limitations of the current study arise from the quality of the studies eligible for inclusion, which in most cases was moderate to good, according to the Jadad scale. None of the included randomised controlled trials reported the method of allocation concealment, however, and as this has been shown to exaggerate treatment effect30
the numbers needed to treat with these treatments may have been overestimated. Most trials were done before the Rome committee published their recommendations for the design of randomised controlled trials of therapies in functional gastrointestinal disorders.31
Only five of the included studies used the Rome criteria to define the presence of irritable bowel syndrome,w1 w16 w18 w31 w32
although only nine were published after the first Rome classification was proposed in 1990,w1 w4 w16 w18 w26 w28 w30-w32
and only two used a validated outcome measure to define improvement in symptoms after treatment.w18 w32
However, many of the included trials met some of the other suggested methodological criteria, such as presence of double blinding and a minimum duration of therapy of 8 to 12 weeks. We preferentially extracted patient reported improvement in symptoms of irritable bowel syndrome or abdominal pain whenever trial reporting allowed this, which is also in line with these recommendations. Blinding of patients in these studies may not have been entirely successful owing to differences in consistency and texture between fibre and placebo, adverse events experienced with antispasmodics and, in the case of peppermint oil, the smell and taste of active treatment. The pooling of data from trials to give an overall treatment effect, and a number needed to treat, could be criticised by some as a result of differences in the methodology of individual included studies. We carried out sensitivity analyses to explore reasons for heterogeneity between studies and in all cases identified potential reasons for this, while still showing a significant treatment effect for most of the treatments we assessed.
Several previous systematic reviews have examined the role of these three treatments in irritable bowel syndrome.15 16 17 18 19 20 21 22
All of these, however, have limitations. Three reviews did not synthesise data, so no summary effect of individual treatments was reported.15 16 17
Of the five that extracted and pooled data in a meta-analysis all have numerous methodological errors,18 19 20 21 22
which render their findings potentially inaccurate. These include errors in the extraction of dichotomous data in a large proportion of included randomised controlled trials,18 19 20 21 22
inclusion of non-eligible studies (according to the investigators prespecified inclusion criteria) in four meta-analyses,18 19 20 22
incorrect handling of data from cross over studies,18 19 21 22
failure to carry out an intention to treat analysis when trial reporting allowed,19 20 21 22
incorporating studies that included patients with other functional gastrointestinal diseases in the analysis,18 20
and failure to identify eligible studies published at the time of the literature search,18 19 20 22
leading to data on truly eligible patients being excluded from the analysis. These errors led to either an overestimation or underestimation of the pooled treatment effect in many of these meta-analyses.19 20 21 22
In addition, since these reviews were carried out further randomised controlled trials of all these treatments have been published.
Current guidelines for the management of irritable bowel syndrome are equivocal or conflicting in their recommendations for the use of these treatments,11 12 23 24
but most of these have been informed by previous systematic reviews, which are potentially methodologically flawed for the reasons discussed, and this has implications for the statements made in them. In the UK, guidelines from both the National Institute for Health and Clinical Excellence and the British Society of Gastroenterology provide similar advice.11 12
Antispasmodics are recommended as first line treatment, particularly when pain and bloating are the predominant symptoms, although which of these drugs should be preferred is not stated. The use of insoluble fibre is discouraged because of concerns that it may exacerbate symptoms, an observation not borne out by our findings. Both organisations recommend that if fibre supplementation is required then this should be in the form of soluble fibres such as ispaghula. Finally, neither of these guidelines provides any statement on the role of peppermint oil in the management of irritable bowel syndrome.
The biological rationale for the efficacy of antispasmodics is unclear, but recent research using magnetic resonance imaging has shown that patients with irritable bowel syndrome and predominant diarrhoea have a reduced colon diameter as well as accelerated small bowel transit,32
so antispasmodics may act by reducing colonic contraction and transit time and therefore pain and stool frequency. Ispaghula husk may increase transit time in those with irritable bowel syndrome and predominant constipation. The efficacy of peppermint oil may arise from effects on smooth muscle, again reducing colonic contractility and pain owing to its calcium channel blocking activity.14
We were unable to examine the effect of different treatments according to predominant stool pattern reported by the patients, however, because few of the eligible trials reported these data as many predate the use of these subgroups, making it difficult to assign people to these categories retrospectively.
In summary, this systematic review and meta-analysis shows that ispaghula husk, antispasmodics (particularly hyoscine), and peppermint oil are all effective treatments for irritable bowel syndrome. Many of these are safe and available over the counter but, with the advent of newer more expensive drugs, are often overlooked as potentially effective treatments. Further large trials of these three agents in patients with irritable bowel syndrome, defined according to the Rome criteria, and using validated outcome measures are warranted. In the interim, current national guidelines for the management of the condition should be updated to include these data.
What is already known on this topic
- Irritable bowel syndrome is a chronic, relapsing and remitting disorder, which can be difficult to treat
- Safe, effective treatments are required, as newer more expensive therapies have been withdrawn because of concerns about safety
- Fibre, antispasmodics, and peppermint oil may fulfil this role, but evidence for their use is conflicting owing to methodological errors in previous systematic reviews
What this study adds
- Fibre, antispasmodics (particularly hyoscine and otilonium), and peppermint oil were all more effective than placebo for treating irritable bowel syndrome
- The numbers needed to treat with these therapies were 11, 5, and 2.5, respectively
- Doctors should consider ispaghula, antispasmodics (preferably hyoscine as first line treatment), and peppermint oil to treat irritable bowel syndrome