What are the effects of treatments in people with irritable bowel syndrome (IBS)?

LIKELY TO BE BENEFICIAL

Antidepressants (Amitriptyline, Clomipramine, Desipramine, Doxepin, Mianserin, Trimipramine)

One systematic review found limited evidence from low to moderate quality randomized controlled trials (RCTs) that anti-depressants (amitriptyline, clomipramine, desipramine, doxepin, mianserin, trimipramine) reduced symptoms of IBS compared with placebo in the short term. It was not clear whether the effects on IBS were independent of the effects on psychological symptoms.

Smooth Muscle Relaxants (Cimetropium Bromide, Hyoscine Butyl Bromide, Mebeverine Hydrochloride, Otilonium Bromide, Pinaverium Bromide, Trimebutine)

One systematic review found limited evidence that smooth muscle relaxants (cimetropium bromide, hyoscine butyl bromide, mebeverine hydrochloride, otilonium bromide, pinaverium bromide, trimebutine) improved symptoms compared with placebo. One subsequent RCT found no significant difference between alverine and placebo in improvement in abdominal pain, although the study may have lacked power to exclude a clinically important effect. One RCT identified by a systematic review found that mebeverine was less effective for symptoms than alosetron in women with diarrhea-predominant IBS, although there are concerns that alosetron may be associated with ischemic colitis.

TRADE-OFF BETWEEN BENEFITS AND HARMS

5HT4 Receptor Agonists (Tegaserod)

One systematic review found that in women with constipation-predominant IBS, tegaserod improved symptoms compared with placebo. It found insufficient evidence about the effects of tegaserod in men. Tegaserod was more likely to cause diarrhea than placebo.

Alosetron

One systematic review found that alosetron (a 5HT₃ receptor antagonist) improved symptoms in women with diarrhea-predominant IBS compared with placebo or mebeverine. However, alosetron is associated with adverse effects, particularly constipation, and has been restricted in some countries because of concerns that it may be associated with ischemic colitis. The systematic review provided insufficient evidence about the effects of alosetron in men.

UNKNOWN EFFECTIVENESS

5HT3 Receptor Antagonists Other Than Alosetron

We found no RCTs examining 5HT₃ receptor antagonists other than alosetron.

Fiber Supplementation

Small RCTs provided insufficient evidence on the effects of fiber supplementation on the symptoms of IBS.

Definition

IBS is a chronic noninflammatory condition characterized by abdominal pain, altered bowel habit (diarrhea or constipation), and abdominal bloating, but with no identifiable structural or biochemical disorder. Symptom-based criteria, such as the Manning criteria,¹ the Rome I criteria,² and the Rome II criteria,³ aid diagnosis, but their primary use is in defining populations in clinical trials. The Rome criteria also subcategorize IBS according to predominant symptoms (diarrhea, constipation, or alternating between diarrhea and constipation). In practice, the division between constipation-predominant and diarrhea-predominant IBS may not be clear-cut in all people. Restriction of trial entry to a subcategory of IBS limits the generalizability of study results.

Incidence/Prevalence

Estimates of incidence and prevalence vary depending on the diagnostic criteria used to define IBS. One cross-sectional postal survey (4,476 people 20 to 69 years of age) in Teeside, United Kingdom, defined IBS as recurrent abdominal pain on more than six occasions during the previous year plus two or more of the Manning criteria.⁴ It estimated prevalence in the United Kingdom to be 16.7 percent (95 percent confidence interval [CI], 15.4 to 18.0 percent) overall, with a prevalence of 22.8 percent (95 percent CI, 20.8 to 24.8 percent) among women and 10.5 percent (95 percent CI, 8.9 to 12.1 percent) among men.⁴ A cross-sectional postal survey (4,500 people 17 years or older) in Australia found prevalences of IBS of 13.6 percent (95 percent CI, 12.3 to 14.8 percent) using the Manning criteria, 6.9 percent (95 percent CI, 6.0 to 7.8 percent) using the Rome I criteria, and 4.4 percent (95 percent CI, 3.5 to 5.1 percent) using the Rome II criteria.⁵

Etiology/Risk Factors

The pathophysiology of IBS is not certain. Studies on the etiology of IBS have been descriptive or retrospective and are of limited reliability. Suggested etiological factors include: abnormal gastrointestinal motor function,^6–8 enhanced visceral perception,^9–11 psychosocial factors such as a history of childhood abuse,¹² genetic predisposition,^13–15 and a history of enteric mucosal inflammation.^16,17 We found no reliable prospective data to measure these associations.

Prognosis

A retrospective study reviewed the medical records of people with IBS (112 people 20 to 64 years of age when diagnosed with IBS at the Mayo Clinic in the United States in 1961 to 1963). IBS was defined as the presence of abdominal pain associated with disturbed defecation or abdominal distension and the absence of organic bowel disease.¹⁸ Over a 32-year period, mortality rates were similar among people with IBS compared with age-and gender-matched controls. One postal survey (4,432 adults 20 to 69 years of age) found that people with IBS are significantly more likely to have had a cholecystectomy than controls (odds ratio [OR] 1.9, 95 percent CI, 1.2 to 3.2).⁴ A paper reporting on the same survey population (2,238 women 20 to 69 years of age) found that women with IBS were significantly more likely to have had a hysterectomy than controls (OR 1.6, 95 percent CI, 1.1 to 2.2).¹⁹ We found no reliable estimates of the duration of IBS if left untreated.