How do SSRIs help patients with irritable bowel syndrome?

Short abstract

Selective serotonin reuptake inhibitor antidepressants seem to promote global well being in some patients with irritable bowel syndrome and, possibly, some improvement in abdominal pain and bowel symptoms, but this effect appears to be independent of improved depression

In the randomised controlled trial reported in this issue of Gut, Tack and colleagues¹ compared citalopram and placebo in 23 patients with irritable bowel syndrome (IBS) over a six week initial treatment period using a parallel group design (see page 1095). The dose of citalopram was similar to that used in the treatment of depressive disorders but any patients with depressive disorder were excluded from this trial. The results showed that citalopram was superior to placebo in terms of the primary outcome measure—days with abdominal pain—and this improvement was unrelated to change in mood, change in stool pattern, or the effect of intravenous citalopram on rectal distension thresholds.

Overall this was not a good trial. The total number of outcome measures exceeded the number of subjects in the trial. The crossover part of the trial was discounted because symptoms did not return to baseline values after the first treatment period. It is not at all clear how the patients were selected and, therefore, whether the results might be generalised to clinic populations. Strengths of the study however include the high participation rate throughout the trial and the combined diary and questionnaire measurements.

In spite of the weaknesses of the trial design, this report is interesting because of the paucity of randomised placebo controlled trials of selective serotonin reuptake inhibitor (SSRI) antidepressants in IBS.²,³ In one trial of patients who had failed to respond to a high fibre diet, Tabas et al found that a low dose of paroxetine was superior to placebo in terms of overall well being, IBS related anxiety, desire to continue the medication after the trial ended (before being unblinded), and less food avoidance.⁴ The improvement in global well being was found even among non‐depressed patients.

The other study by Kuiken et al found that in 40 non‐depressed IBS patients, fluoxetine did not significantly alter the threshold for discomfort relative to placebo, either in hypersensitive or in normosensitive patients. In hypersensitive patients only, fluoxetine significantly reduced abdominal pain complaints without alteration of gastrointestinal symptoms, global symptom relief, or psychological symptoms.⁵

The trial by Tack and colleagues¹ does provide useful information if we examine only the a priori primary and two of the stated secondary outcome measures at the end of the initial parallel group comparison—that is, before the crossover part of the trial. These results show that citalopram was significantly superior to placebo in reduction of days with abdominal pain and days in which IBS impacted on daily life. Depression also improved significantly in the citalopram group compared with the placebo group but the improvement in IBS symptoms was not associated with improved mood.

Taken together with the previous results it appears that the SSRI antidepressants do promote global well being in some patients with IBS and, possibly, some improvement in abdominal pain and bowel symptoms, but this effect appears to be independent of improved depression. There are several possible mechanisms which might explain the beneficial effect of citalopram. All are independent of a change in mood and therefore might be applicable to the Tack study.

Firstly, the rationale of the present trial was based on a previous finding that intravenous citalopram decreased the sensitivity of the colon to distension in healthy volunteers.⁶ In the trial, the result of intravenous citalopram did not predict outcome, but as no patients returned for further rectal distension testing after treatment it is not clear whether oral citalopram over six weeks led to a change in colonic sensorimotor function. As mentioned above, Kuiken and colleagues⁵ found no such change so this seems to be an unlikely mechanism for the symptomatic improvement; only venlafaxine, which inhibits reuptake of both serotonin and norepinephrine, appears to decrease the sensitivity of the colon to distension.⁷

Secondly, citalopram, in common with other SSRI antidepressants, may have other effects on the gut, such as accelerating transit time,⁸ which would help patients with constipation, but there was no clear evidence of this mechanism in this trial. Thirdly, there is some evidence that SSRIs have an analgesic effect. This is not as strong as tricyclic antidepressants when used for IBS,³ neuropathic pain,⁹ back pain,¹⁰ or migraine¹¹ but appears to be important in patients with somatoform pain disorders.¹²

Probably more important than any of these possible effects on the gut is the effect on widespread bodily symptoms; citalopram may reduce the reporting of multiple bodily symptoms or “somatisation”.¹³ This may be the reason that SSRI antidepressants are beneficial in somatoform pain disorders,¹² premenstrual syndrome,¹⁴ and for a variety of unexplained symptoms and syndromes.¹⁵ These disorders are frequently concurrent with IBS but the nature of some of the IBS symptoms, which improved in the Tabas trial,⁴ might fall into this category—overall well being, IBS related anxiety, desire to continue the medication after the trial ended, and less food avoidance. These symptoms may not be specifically related to gut dysfunction and may be part of the somatisation or “extraintestinal” symptoms that may accompany IBS.

In the present state of knowledge it is not clear how somatisation should be conceptualised in relation to IBS. Although IBS patients as a whole have high scores on somatisation measures,¹⁶ it is likely that a subgroup of IBS patients is responsible for this finding; some IBS patients do not have an excess of bodily symptoms.¹⁷,¹⁸ The presence of multiple bodily symptoms is associated with frequent treatment seeking¹⁷,¹⁹ and this group of patients may need to be studied separated in future if we are to make sense of the role of SSRI antidepressants in IBS. Two groups which are known to report numerous bodily symptoms are those with a concurrent psychiatric disorder and/or a history of sexual abuse.

Patients with severe IBS who reported a history of sexual abuse seem to do particularly well following treatment with either paroxetine or psychotherapy, and the improvement in their health related quality of life is mediated by a reduction in somatisation.²⁰,²¹ Within this group of patients, who reported abuse, the number of bodily symptoms was associated with rectal distension threshold, and an abuse history also predicted normalising of the distension threshold following treatment, independent of change in mood.²² This change in rectal distension tolerance and reduced somatisation were probably related to a change in how bodily sensations were perceived; with treatment, the patients probably paid less attention to gastrointestinal sensations and ceased to attribute such sensations to possible serious disease.²³

These psychological processes may be important in the last mechanism by which citalopram may lead to improved IBS symptoms. Citalopram has been shown to induce an affective memory bias towards positive material without significantly influencing the subjective mood status.²⁴ Citalopram ameliorates negative biases in information processing and this could reduce attention to gastrointestinal sensations and modify unrealistic fears that symptoms imply serious illness.

If reduction of somatisation and modification of memory bias towards more positive material underlie the action of citalopram in patients with IBS, is this important in the aetiology of IBS? The answer to this question is not clear. The psychological processes are certainly important in some patients but not all. It is likely that they are particularly important in relation to treatment seeking, which is driven, at least in part, by multiple bodily symptoms and fears of serious illness.

From a scientific point of view there are a number of weaknesses of the current and previous studies assessing efficacy of SSRI antidepressants, most notably their short duration and small sample sizes.¹⁵,²⁵ The latter is particularly important in relation to assessing whether different processes occur in different groups of IBS patients. Further large efficacy studies are needed.

From a clinical viewpoint, however, the gastroenterologist looks to the literature for guidance. The Tack study¹ was performed in patients seen at a tertiary referral centre; they had chronic IBS (mean duration of four years) and abdominal pain on 5.4 days per week; gastroenterologists have little to offer such patients if they have failed to respond to antispasmodic treatments.²⁶ This is where effectiveness rather than efficacy studies are needed.

In the largest cost effectiveness study of its kind, we found that paroxetine at low dose (20 mg/day) led to improved health related quality of life in the long term (15 months after entry) in comparison with usual treatment for patients with severe and treatment resistant IBS.²¹ For patients who took the medication there were short term gains, compared with treatment as usual, in all five outcome measures (pain severity and frequency, distress severity, and mental and physical aspects of health related quality of life).²¹ In the long term intention to treat analysis, there were benefits in health related quality of life that were not confined to those patients who had psychiatric disorders.²⁷

The correct interpretation of this pragmatic cost effectiveness study was that in a clinically representative sample of people with severe IBS, which has not responded to usual treatment, considerable improvement in health related quality of life was achieved at no extra cost. The study was not designed to assess the mechanism of action but we did show that improvement in health related quality of life could not be explained solely on the basis of improved abdominal pain or improved mood.²¹ It has been suggested that this long term result may have been the result of increased contact time with the gastroenterologist or general practitioner who prescribed paroxetine²⁸ but the additional time was minimal—it was not significantly greater than treatment as usual.²¹ Thus in practice there is a gain for patients, in the aspect of the illness that concerns them most,²⁹ if gastroenterologists prescribe an SSRI antidepressant and encourage adherence.

Patients with IBS are a heterogeneous group. Of the 257 patients with severe IBS in our trial, 29% patients had depressive disorder and when this responded to treatment the number of days of restricted activity was nearly halved.³⁰ Another group (23%) had a history of sexual abuse (most did not have depressive disorder) and their health related quality of life improved greatly following either paroxetine or psychotherapy.²⁰ The improvement was not accompanied by improvement in mood or abdominal pain, suggesting this was not a non‐specific result of increased contact time; it was mediated by marked improvement in somatisation.²⁰

The exact mechanisms of action of SSRI antidepressants in IBS are not completely understood at this time. There may be several mechanisms which are important in different groups of patients. The antidepressant effect is important for those patients with depressive disorder. That apart, the most obvious next action is that involving changes in psychological processes, which lead to reduced somatisation and a reduced tendency to regard gut sensations as indicative of serious illness. These are very important actions in reducing treatment seeking. The SSRI antidepressants may also have important actions on the gut but at present these have not been adequately defined. The clinical practice of prescribing SSRI antidepressants in patients who have failed to respond to usual treatment is supported by our cost effectiveness data.

Conflict of interest: declared (the declaration can be viewed on the Gut website at http://www.gutjnl.com/supplemental).