Skip to content

Diarrhoea-predominant pharmacotherapy (IBS-D)

TL;DR — IBS-D pharmacotherapy is where this condition's drugs actually hurt people. Eluxadoline works modestly — composite response 23.9–29.6% at 75/100 mg vs 16.2–17.1% placebo over 12 weeks across two phase 3 trials in 2,427 patients — but pancreatitis occurred in 5/1,666 (0.3%) of the safety population in registration and accounted for 16.4% of 597 post-marketing adverse-event reports, with 53 cases requiring hospitalisation, against 0.2–0.5% for comparator drugs (Lembo 2016, PMID 26789872; Gawron 2018, PMID 28804032). All ten sphincter-of-Oddi-spasm events in pooled phase 2/3 safety data occurred in patients without a gallbladder (Cash 2017, PMID 27922029). Alosetron was withdrawn in 2000 for ischaemic colitis and reintroduced in 2002 only under a restricted prescribing programme (Lewis 2010, PMID 20136586). The 5-HT3 antagonists as a class are the most effective agents: in a network meta-analysis of 21 RCTs and 10,421 non-constipated IBS patients, alosetron ranked best for global symptoms (SUCRA 0.82), cilansetron for pain (0.90) and ondansetron for stool consistency (0.98) (Rokkas 2021, PMID 34276193). Ramosetron, licensed in Japan, produced global improvement in 50.7% vs 32.0% (NNT 6, 95% CI 4–10) in 576 women (Fukudo 2016, PMID 26551550) — the largest single responder difference of any IBS drug retrieved in this build. Loperamide, the most-used agent worldwide, carries a very low certainty conditional recommendation (Lembo 2022, PMID 35738725) — it firms stool and does not treat pain. Because bile acid diarrhoea affects roughly a quarter to a third of people labelled IBS-D and responds to sequestrants, that diagnosis should be considered before this page's drugs are started (differential-diagnosis-and-exclusion).

What AGA recommends, and how weakly

AGA's 2022 IBS-D guideline reviewed eluxadoline, rifaximin, alosetron, loperamide, tricyclics, SSRIs and antispasmodics, and issued eight recommendations (Lembo 2022, PMID 35738725):

Strength Certainty Agents
Conditional for Moderate Eluxadoline, rifaximin, alosetron
Conditional for Low Tricyclic antidepressants, antispasmodics
Conditional for Very low Loperamide
Conditional against Low SSRIs

There is no strong recommendation anywhere in the IBS-D guideline — in contrast to IBS-C, where linaclotide earned a strong recommendation at high certainty (Chang 2022, PMID 35738724). The European UEG/ESNM guideline on functional bowel disorders with diarrhoea covers the same ground with a European formulary (Savarino 2022, PMID 35695704).

Eluxadoline

A mixed µ-opioid and κ-opioid receptor agonist with δ-opioid receptor antagonism.

Trial n Duration Endpoint Result
IBS-3001 (NCT01553591) part of 2,427 52 weeks Composite (pain reduction + stool improvement same day, ≥50% of days) weeks 1–12 23.9% (75 mg) and 25.1% (100 mg) vs 17.1% placebo (p=0.01, p=0.004)
IBS-3001, weeks 1–26 23.4% (p=0.11) and 29.3% (p<0.001) vs 19.0%
IBS-3002 (NCT01553747) 26 weeks weeks 1–12 28.9% and 29.6% vs 16.2% (p<0.001 both)
IBS-3002, weeks 1–26 30.4% (p=0.001) and 32.7% (p<0.001) vs 20.2%
RELIEF phase 4 (Brenner 2019, PMID 31356229) 346 with inadequate loperamide response 12 weeks Composite (≥40% worst-abdominal-pain improvement + BSFS <5 for ≥50% of days) 22.7% vs 10.3% (p=0.002); stool consistency 27.9% vs 16.7% (p=0.01); pain 43.6% vs 31.0% (p=0.02); no sphincter-of-Oddi spasm or pancreatitis in a population with intact gallbladders

Adverse events at 75/100 mg vs placebo in registration: nausea 8.1%/7.5% vs 5.1%, constipation 7.4%/8.6% vs 2.5%, abdominal pain 5.8%/7.2% vs 4.1%; pancreatitis in 5 of 1,666 (0.3%) (Lembo 2016, PMID 26789872).

The pancreatitis and sphincter-of-Oddi signal

Pooled phase 2/3 safety analysis (2,814 patients in the safety set) found 10 sphincter-of-Oddi spasm events in 1,839 eluxadoline-treated patients (0.5%), presenting as acute abdominal pain with raised aminotransferases or lipase, or as pancreatitis. Every one occurred in a patient without a gallbladder; eight were on the higher dose; all occurred within one week of starting and all resolved on discontinuation. A further five events were adjudicated as pancreatitis not associated with sphincter-of-Oddi spasm, three of them with heavy alcohol use (Cash 2017, PMID 27922029).

Post-marketing surveillance sharpened the picture. Of 597 FAERS reports on eluxadoline between January and September 2016, pancreatitis accounted for 16.4%, with 53 cases requiring hospitalisation. The same complication appeared in 0.3% of loperamide reports, 0.4% of diphenoxylate, 0.2% of oxycodone and 0.5% of rifaximin reports — and in 75% of those comparator submissions the agent was not considered causal. The authors called for reassessment of the risk–benefit ratio, citing two fatalities (Gawron 2018, PMID 28804032).

The drug remains licensed with contraindications in patients without a gallbladder, with biliary or pancreatic disease, or with heavy alcohol use — see red-flags-and-safety-concerns. The NNH for discontinuation due to adverse events is 32 (p<0.01) (Busam 2026, PMID 40471839), against an NNT of roughly 8 from RELIEF.

5-HT3 receptor antagonists

The most effective class, with the class's own safety history.

Agent Evidence Result Status
Alosetron Ranked best for global symptom improvement (SUCRA 0.82) in a 21-RCT, 10,421-patient network meta-analysis (Rokkas 2021, PMID 34276193) — Approved 2000 for women with IBS-D; voluntarily withdrawn November 2000 after ischaemic colitis and severe constipation complications; reintroduced 2002 under a Risk Management Plan with restricted indication and a Prescribing Program; uptake has remained very limited (Lewis 2010, PMID 20136586; Lucak 2010, PMID 21180598). NNH for discontinuation 14 (p<0.01) — the highest of any IBS drug (Busam 2026, PMID 40471839)
Cilansetron Ranked best for abdominal pain/discomfort (SUCRA 0.90) (Rokkas 2021, PMID 34276193) — Never marketed
Ondansetron Ranked best for bowel habit/consistency (SUCRA 0.98) (Rokkas 2021, PMID 34276193). TRITON RCT (ISRCTN17508514), 80 randomised of a planned 400 Primary FDA composite not met: 40.5% (24.7–56.4) vs 27.9% (14.5–41.3), p=0.19. Stool consistency improved (adjusted mean difference −0.7, −1.0 to −0.3, p<0.001); whole-gut transit time increased by 3.8 (SD 9.1) h vs −2.2 (10.3) h, p=0.01. Meta-analysis of 327 patients from three trials: FDA composite RR of not responding 0.86 (0.75–0.98), NNT 9; stool response RR 0.65 (0.52–0.82), NNT 5; abdominal pain RR 0.95 (0.74–1.20) — no pain benefit (Gunn 2023, PMID 36866724) Off-label; cheap and widely available
Ramosetron 2.5 µg od Phase 3, 576 Japanese women, 12 weeks (NCT01870895) Global improvement 50.7% (44.8–56.6) vs 32.0% (26.7–37.8), difference 18.6% (10.7–26.5), p<0.001; RR 1.58 (1.29–1.94), NNT 6 (4–10); increased stool consistency 40.8% vs 24.3% (p<0.001); pain/discomfort p=0.001; QoL p=0.002; constipation in 11.0% (Fukudo 2016, PMID 26551550). Meta-analysis of 4 RCTs, 1,623 participants: overall symptom relief RR 1.70 (1.48–1.95), pain RR 1.41 (1.24–1.59), bowel habit RR 1.72 (1.50–1.98), stool consistency RR 1.71 (1.40–2.08); adverse events RR 1.10 (0.97–1.26); no ischaemic colitis reported; hard stool RR 4.74 (3.00–7.51) and constipation RR 2.53 (1.57–4.10) (Qi 2018, PMID 29310568) Licensed in Japan; effective in both sexes (male RR 1.94, female RR 1.49) despite the historical sex-specific licensing of the class (Fukudo 2014, PMID 24315882)

The class picture is coherent: 5-HT3 antagonism slows transit and firms stool reliably, improves global symptoms, and improves pain less consistently (ondansetron: no pain benefit at all in meta-analysis; ramosetron: modest benefit). The class-defining risk is ischaemic colitis, documented for alosetron and not observed in the ramosetron trials to date — an asymmetry that has never been explained mechanistically and may simply reflect exposure volume.

Regulatory sex-splitting. Alosetron was approved only for women and ramosetron initially only for men, on the basis of presumed sex-specific effects — a division that the 2016 trial in 576 women and the meta-analysis in both sexes have since undercut (Fukudo 2016, PMID 26551550; Qi 2018, PMID 29310568). It remains one of the few instances in gastroenterology of a drug class licensed by sex.

Loperamide

The most widely used IBS-D drug in the world has the weakest evidence on this page: a conditional recommendation at very low certainty (Lembo 2022, PMID 35738725). It reduces stool frequency and firms consistency without an established effect on abdominal pain, and its principal role in the modern evidence base is as the comparator that defines "inadequate symptom control" in the eluxadoline RELIEF trial (Brenner 2019, PMID 31356229).

Rifaximin

Also recommended conditionally for IBS-D at moderate certainty (Lembo 2022, PMID 35738725), with the best safety profile in the class — its NNH for adverse-event discontinuation was negative and non-significant, i.e. discontinuation was no more frequent than placebo (Busam 2026, PMID 40471839). Its evidence, retreatment data and the SIBO controversy have their own page: rifaximin-and-the-sibo-question.

Safety, compared

Numbers needed to harm for discontinuation due to adverse events, from 54 trials (Busam 2026, PMID 40471839):

Drug NNH p
Alosetron 14 <0.01
Tricyclics 24 <0.01
Eluxadoline 32 <0.01
Rifaximin negative (i.e. fewer discontinuations than placebo) not significant
Ramosetron negative not significant

The authors' summary: tricyclics (especially at higher doses), tenapanor and alosetron have the highest absolute discontinuation risk; rifaximin is the safest pharmacotherapy studied. Note that NNH for discontinuation does not capture rare catastrophic events — eluxadoline's pancreatitis and alosetron's ischaemic colitis are exactly the harms this metric misses, and they are the reason both drugs carry restrictions.

What to consider before starting any of these

  1. Bile acid diarrhoea — 28.1% of IBS-D by SeHCAT <10% (Slattery 2015, PMID 25913530), responsive to sequestrants (stool consistency RR 1.50, 1.14–1.96; frequency RR 2.80, 1.68–4.67; Dilmaghani 2025, PMID 41090475).
  2. Microscopic colitis — 5.7% of diarrhoeal functional bowel disorders referred for colonoscopy, frequently missed for want of biopsies (Asghar 2022, PMID 32882424).
  3. Coeliac disease — 2% biopsy-proven in criteria-defined IBS (Shiha 2025, PMID 40493044).
  4. Dietary therapy — outperformed optimised medical treatment in CARIBS (Nybacka 2024, PMID 38643782).

Open questions

  • Should eluxadoline still be marketed? Post-marketing pancreatitis constituted 16.4% of adverse-event reports with 53 hospitalisations and two reported fatalities (Gawron 2018, PMID 28804032), against an NNT of about 8 (Brenner 2019, PMID 31356229).
  • Why has ondansetron — ranked first for stool consistency (Rokkas 2021, PMID 34276193), cheap, generic, and with NNT 5 for stool response (Gunn 2023, PMID 36866724) — never been licensed for IBS-D? TRITON recruited 80 of a planned 400 patients.
  • Does ramosetron carry an ischaemic colitis risk that has not yet been detected at Japanese exposure volumes (Qi 2018, PMID 29310568)?
  • Is the sex-specific licensing of 5-HT3 antagonists defensible, given efficacy in both sexes (Fukudo 2016, PMID 26551550; Qi 2018, PMID 29310568)?
  • Does any IBS-D drug relieve abdominal pain? Ondansetron does not (RR 0.95, 0.74–1.20; Gunn 2023, PMID 36866724); eluxadoline's pain component is 43.6% vs 31.0% (Brenner 2019, PMID 31356229); loperamide has no pain evidence at all.
  • Would routine bile-acid testing before drug initiation change outcomes? Never trialled.

References

  1. Lembo A, Sultan S, Chang L, Heidelbaugh JJ, Smalley W, Verne GN. AGA Clinical Practice Guideline on the Pharmacological Management of Irritable Bowel Syndrome With Diarrhea. Gastroenterology. 2022;163(1):137-151. PMID 35738725
  2. Lembo AJ, et al. Eluxadoline for Irritable Bowel Syndrome with Diarrhea. N Engl J Med. 2016;374(3):242-53. PMID 26789872
  3. Cash BD, Lacy BE, Schoenfeld PS, Dove LS, Covington PS. Safety of Eluxadoline in Patients with Irritable Bowel Syndrome with Diarrhea. Am J Gastroenterol. 2017;112(2):365-374. PMID 27922029
  4. Gawron AJ, Bielefeldt K. Risk of Pancreatitis Following Treatment of Irritable Bowel Syndrome With Eluxadoline. Clin Gastroenterol Hepatol. 2018;16(3):378-384.e2. PMID 28804032
  5. Brenner DM, et al. Efficacy and Safety of Eluxadoline in Patients With Irritable Bowel Syndrome With Diarrhea Who Report Inadequate Symptom Control With Loperamide: RELIEF Phase 4 Study. Am J Gastroenterol. 2019;114(9):1502-1511. PMID 31356229
  6. Rokkas T, Ekmektzoglou K, Niv Y. Comparative effectiveness of 5-hydroxytryptamine 3 receptor antagonists in irritable bowel syndrome: a network meta-analysis of randomized controlled studies. Ann Gastroenterol. 2021;34(4):535-546. PMID 34276193
  7. Gunn D, et al. Randomised, placebo-controlled trial and meta-analysis show benefit of ondansetron for irritable bowel syndrome with diarrhoea: The TRITON trial. Aliment Pharmacol Ther. 2023;57(11):1258-1271. PMID 36866724
  8. Fukudo S, et al. Ramosetron Reduces Symptoms of Irritable Bowel Syndrome With Diarrhea and Improves Quality of Life in Women. Gastroenterology. 2016;150(2):358-66.e8. PMID 26551550
  9. Qi Q, Zhang Y, Chen F, Zuo X, Li Y. Ramosetron for the treatment of irritable bowel syndrome with diarrhea: a systematic review and meta-analysis of randomized controlled trials. BMC Gastroenterol. 2018;18(1):5. PMID 29310568
  10. Fukudo S, et al. Effect of ramosetron on stool consistency in male patients with irritable bowel syndrome with diarrhea. Clin Gastroenterol Hepatol. 2014;12(6):953-9.e4. PMID 24315882
  11. Lewis JH. Alosetron for severe diarrhea-predominant irritable bowel syndrome: safety and efficacy in perspective. Expert Rev Gastroenterol Hepatol. 2010;4(1):13-29. PMID 20136586
  12. Lucak SL. Optimizing outcomes with alosetron hydrochloride in severe diarrhea-predominant irritable bowel syndrome. Therap Adv Gastroenterol. 2010;3(3):165-72. PMID 21180598
  13. Busam JA, et al. The Safety of Pharmacotherapy for Irritable Bowel Syndrome: A Systematic Review and Meta-Analysis. Am J Gastroenterol. 2026;121(3):745-753. PMID 40471839
  14. Savarino E, et al. Functional bowel disorders with diarrhoea: Clinical guidelines of the United European Gastroenterology and European Society for Neurogastroenterology and Motility. United European Gastroenterol J. 2022;10(6):556-584. PMID 35695704
  15. Chang L, Sultan S, Lembo A, Verne GN, Smalley W, Heidelbaugh JJ. AGA Clinical Practice Guideline on the Pharmacological Management of Irritable Bowel Syndrome With Constipation. Gastroenterology. 2022;163(1):118-136. PMID 35738724
  16. Slattery SA, Niaz O, Aziz Q, Ford AC, Farmer AD. Systematic review with meta-analysis: the prevalence of bile acid malabsorption in the irritable bowel syndrome with diarrhoea. Aliment Pharmacol Ther. 2015;42(1):3-11. PMID 25913530
  17. Dilmaghani S, et al. Meta-Analysis: Efficacy and Safety of Sequestrants for Bile Acid Diarrhoea. Aliment Pharmacol Ther. 2025;62(11-12):1054-1065. PMID 41090475
  18. Asghar Z, et al. Diagnostic Yield of Colonoscopy in Patients With Symptoms Compatible With Rome IV Functional Bowel Disorders. Clin Gastroenterol Hepatol. 2022;20(2):334-341.e3. PMID 32882424
  19. Shiha MG, et al. Global Prevalence of Celiac Disease in Patients With Rome III and Rome IV Irritable Bowel Syndrome. Am J Gastroenterol. 2025;120(12):2776-2787. PMID 40493044
  20. Nybacka S, et al. A low FODMAP diet plus traditional dietary advice versus a low-carbohydrate diet versus pharmacological treatment in irritable bowel syndrome (CARIBS). Lancet Gastroenterol Hepatol. 2024;9(6):507-520. PMID 38643782