Review Article

An Analysis of Safety Profiles of Treatments for Diarrhoea-predominant Irritable Bowel Syndrome

Brian E. Lacy


Aliment Pharmacol Ther. 2018;48(8):817-830. 

In This Article


IBS is a common gastrointestinal condition, and a number of different treatments, administered either daily or as a short-course of therapy and targeting various pathophysiologic factors, may be considered to manage symptoms of IBS-D. For any IBS-D therapy, both efficacy and safety profiles are critical aspects of the decision-making process for ongoing symptom and disease management (Figure 1).[10,17,22,23,25,26,32,33,40,41,44,46,47,49,58,60,61,69,74,75,78,81,88,100] A survey study of 182 patients with IBS found that patients were willing to accept substantial medication risk (median 1% risk of sudden mortality with a hypothetical medication for 99% chance of cure).[101] These findings highlight both the burden of IBS and the need to emphasise safety considerations in treatment selection. Any direct comparisons across treatments are difficult as patient populations, efficacy endpoints, and safety analyses in clinical trials can widely vary. However, number needed to treat and harm values can help provide some indication of potential clinical benefits and risks among therapies.

Figure 1.

Summary of safety profiles for therapies used for the management of IBS-D.10, 17, 22, 23, 25, 26, 32, 33, 36, 40, 41, 44, 46, 47, 49, 58, 60, 61, 69, 74, 75, 78, 81, 88, 100,a. aNumber needed to harm values have been provided for each therapy for which data are available. 5–HT3: 5–hydroxytryptamine type 3; AE: adverse event; FODMAP: fermentable oligosaccharides, disaccharides, monosaccharides, and polyols; IBS-D, diarrhoea-predominant irritable bowel syndrome; NNH: number needed to harm; REMS: Risk Evaluation and Mitigation Strategy

The value of a positive patient-provider interaction cannot be underestimated because a positive approach improves the likelihood of treatment response.[102–107] Just as importantly, providers are often better versed in efficacy data. Response rates, likelihood of improving, and odds ratios for a positive response are strongly favoured in publications and advertising and, thus, offer readily available data for providers to discuss with patients. The number needed to treat is one such piece of information that healthcare providers can use to select appropriate therapies. In this review, a wide range of numbers needed to treat for medications used to treat IBS-D were reported (Table 3)[10,17,23,31–33,40,42,44,48,50,78,81,87,108] and may be related to the primary endpoint of the studies and other methodologic considerations (patient population).

Discussions about the negative aspects of therapeutic interventions are less important for some healthcare providers,[109] who feel that sharing such information with patients could diminish treatment response (eg, impact adherence)[109,110] and increase occurrence of perceived AEs,[111] and also because safety data are often not as well and consistently publicised as efficacy data. However, discussion of potential treatment-related AEs has been considered by patients and some healthcare providers to be beneficial, as related communications may increase awareness and lessen the frequency of AEs.[110,112,113] Values such as number needed to harm should be considered alongside other information (eg, patient preference, adherence considerations, cost-effectiveness) to help guide patients and healthcare providers. Although definitions varied (patients experiencing an AE vs discontinuing due to an AE), among the IBS-D treatments with available data (Table 3),[23,32,48] the number needed to harm values were least favourable for antidepressants (8.5), tricyclic antidepressants (9 and 18), and alosetron (10 and 19) and most favourable for rifaximin (8971).[10,32,80] In addition to the likelihood of any AE, the risk of serious AEs warrants careful consideration. Notable potential safety concerns with IBS-D treatments include pancreatitis with eluxadoline,[50] ischaemic colitis and serious complications of constipation with alosetron,[33] cardiotoxicity with loperamide[40] and tricyclic antidepressants,[81] and abuse/misuse of loperamide.[108] A thoughtful discussion with patients to individualise treatment based on patient-specific characteristics and risk factors (eg, underlying cardiac risk factors; prior trials with antibiotics; presence or absence of a gall-bladder; alcohol use) should occur with a review of number needed to treat and number needed to harm values to shape the discussion with patients of the various treatment options.[49,50] For agents that target the gut microbiota, safety profiles may differ among probiotic strains and formulations, thus large, well-designed, clinical studies are warranted.[23,26]

In summary, while numerous treatment options are currently available for patients with IBS-D, providers should carefully consider the safety in addition to the efficacy of a specific intervention when determining how best to manage IBS-D in an individual patient.