Article Text


Melatonin: a novel treatment for IBS?
  1. S Elsenbruch
  1. Correspondence to:
    Department of Medical Psychology
    University Clinic of Essen, Hufelandstr 55, 45122 Essen, Germany;

Statistics from

Can poor sleep affect gastrointestinal symptoms in IBS? More on the “bad dreams cause bad bowels” hypothesis with reference to new treatment options

The issue of sleep in irritable bowel syndrome (IBS) is intriguing and relevant for several reasons. Firstly, complaints of poor sleep are extremely common in patients with IBS. In fact, self reported sleep disturbance can be regarded as one of the most important extraintestinal symptoms of IBS, which markedly affects quality of life and psychosocial well being.1 In spite of this, little is known about treatment options for this important extraintestinal symptom of IBS. Secondly, there is an overlap between IBS and fibromyalgia syndrome (FS),2 another pain syndrome which has also been linked to disturbed sleep physiology,3 as well as stress.4 If indeed a proportion of IBS patients share a common pathophysiological mechanism with FS, addressing the treatment of sleep related functions in patients with IBS would be a promising venue. Thirdly, hypervigilance has been discussed as one possible pathophysiological mechanism in IBS.5 In this context, sleep studies using electroencephalogram techniques may be useful to reveal disturbed brain activity consistent with the hypothesis of hypervigilance/altered arousal mechanisms in IBS. Finally, it has been suggested that reports of night time awakenings due to gastrointestinal symptoms, mainly abdominal or epigastric pain, may be used as a discriminatory factor between organic and functional disorder.6

A review of the existing literature confirms that a large percentage of IBS patients report sleep disturbances.7,8,9,10,11,12,13,14 Despite this overwhelming evidence of subjective sleep problems, it remains difficult to answer the question of whether these reports are based on objectively measurable abnormalities in sleep physiology. A total of eight studies have used objective methods to study sleep in IBS.7,9,10,13,15–18 Of those, three studies found no differences between IBS patients and controls on any polysomnographic measure and concluded that sleep architecture is normal in patients with IBS.9,10,17 Only one study found what one might expect based on patients’ subjective reports of insomnia-type symptoms as well as disrupted, non-restful sleep—namely, markedly decreased slow wave sleep, higher arousal and awakening index, longer wake period after sleep onset, and more downward shifts to lighter sleep stages.7 A number of explanations exist for these heterogeneous results, including differences in patient populations in sex (that is, some have only included women), psychiatric comorbidities (some studies have excluded subjects with concurrent psychopathology, including depression and/or use of antidepressants), relatively small sample sizes,16–18 and inclusion of patients with sleep apnoea.18 Given this conflicting evidence, one could conclude that subjective sleep disturbances are not necessarily substantiated by objective sleep abnormalities in all IBS patients, although it is likely that subgroups of patients with objective sleep abnormalities exist. Determinants of subjective sleep problems in patients with IBS have been shown to include gastrointestinal symptom severity or profile10,11,19 and psychological disturbances/distress.10,13,15 In this context, it is of further interest to note that a large proportion of IBS patients do indeed report night time gastrointestinal symptoms,10,11 even if they do not show polysomnographic evidence of sleep disruption. Therefore, the presence or absence of night time gastrointestinal symptoms is neither a valid discriminator between organic and functional disease6 nor a good indication of whether objective sleep abnormalities are in fact present.

Can bad sleep cause bad bowels or vice versa? This question, which ties into the relationship between sleep disturbance and severity of daytime gastrointestinal symptoms, remains difficult to answer.20 Sleep complaints have been found to be associated with perceived intensity of gastrointestinal symptoms.11 Using prospective diary assessments, two studies found that poor subjective sleep was associated with higher gastrointestinal symptoms the following day,8,21 lending support to the hypothesis that “bad sleep causes bad bowels” rather than vice versa. A more sophisticated way to address a possible cause-effect relationship between sleep disturbance and IBS is the experimental manipulation of sleep functions and subsequent study of effects on gastrointestinal symptomatology. This is what Song and colleagues22 have attempted, and the first randomised, double blind, placebo controlled trial on the effects of melatonin in IBS patients who reported sleep disturbances is described in this issue of Gut (see page 1402).22 Following two weeks of treatment, melatonin treated IBS patients demonstrated a significant reduction in reported abdominal pain, measured with an adapted version of the IBS symptoms evaluation score questionnaire, whereas no significant change was observed in the placebo group. In addition, rectal pressure and volume thresholds for both urgency and pain sensations were significantly increased following melatonin treatment. On the other hand, melatonin treatment had no effect on polysomnographic parameters, or measures of anxiety and depression.

In their study, the authors screened IBS patients for the presence of self reported sleep disturbance.22 Given the apparent inaccuracy of IBS patient reports concerning sleep disturbance and lack of consistent polysomnographic evidence of sleep abnormalities in IBS in the literature, it remains unclear what proportion of patients included in the trial did actually have manifest sleep problems. The lack of treatment effects on polysomnographic parameters in this study would lend support to the speculation that at least a proportion of the patient sample may not have had objective sleep abnormalities. Unfortunately, patients’ basal polysomnographic data are difficult to evaluate in this respect as no healthy control group was included to show the extent to which a “first night effect” affected sleep at baseline.23,24 A recent meta-analysis on the soporific effects of exogenous melatonin found that melatonin treatment significantly reduced sleep onset latency by 4.0 minutes, increased sleep efficiency by 2.2%, and increased total sleep duration by 12.8 minutes.25 Given these modest effects, one may speculate that adaptation effects from the first to the second night (that is, improved sleep due to habituation) may have masked the effects of melatonin treatment. Future studies may consider screening of patients using polysomnography rather than questionnaires, which would have the further advantage of providing an adaptation night to minimise “first night effects”, and make treatment induced changes in objective sleep parameters more easily detectable.

Another issue that demands careful consideration when designing sleep (treatment) studies in IBS in the future is selection of patients with regard to psychiatric comorbidity. Song and colleagues22 did not report screening for mood disorders in their patient population. However, mood disorders, especially anxiety and depression, are associated with alterations in sleep physiology,26 irrespective of the presence of other disorders such as IBS. Therefore, inclusion of IBS patients with mood disturbance can introduce difficulties regarding interpretation of sleep data, unless appropriate psychiatric control groups (for example, patients with depression but without IBS) are included. Psychological distress can affect the perception of sleep quality and daytime fatigue, and may modulate perception of other bodily symptoms (visceral and non-visceral) also. Anxiety and depression scores were not affected by melatonin treatment in the study by Song et al, and hence the mechanism by which melatonin treatment improved reported abdominal pain, and rectal pressure and volume thresholds for urgency and pain sensations remains unclear. Autonomic dysfunction has been implicated in the pathophysiology of IBS, including findings of autonomic disturbances during sleep in IBS patients.27 Based on evidence showing inhibitory effects of melatonin on the sympathetic nervous system in healthy humans,28,29 it is intriguing to speculate that modulation of autonomic functions may have played a role in the results reported by Song et al. Overall, these findings are intriguing and call for replication and further study.

Can poor sleep affect gastrointestinal symptoms in IBS? More on the “bad dreams cause bad bowels” hypothesis with reference to new treatment options


View Abstract


  • Conflict of interest: None declared.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles