This issue of Digestive Diseases and Sciences includes an article by Khanna et al. reporting aspects of bile acid metabolism in patients seen at Mayo Clinic with chronic diarrhea and neuroendocrine tumors [1]. Data were gathered retrospectively from a heterogeneous group of patients with a variety of tumors, variable tumor loads, and various treatments, including bowel resection. The bottom line from this study is that most subjects had evidence of bile acid malabsorption (BAM). The contribution of that BAM to their ongoing diarrhea, based on the data presented, is less certain.

Bile acids have been studied scientifically over the last 150 years, the retelling of which is a fascinating story that demonstrates the strong dependence of scientific knowledge on evolving technology [2, 3]. As new techniques such as chromatography, mass spectroscopy, and x-ray diffraction developed, novel features of these organic molecules were revealed. Bile acids comprise a diverse group of acid sterols synthesized in the liver, conjugated to amino acids glycine or taurine, and secreted into the bile, that work intraluminally in the intestine to help solubilize fat prior to absorption. Most of the bile acid load to the intestine is reabsorbed by specialized transporters in the ileum and returned to the liver via the mesenteric veins, creating an enterohepatic circulation. The fraction of bile acid that is not reabsorbed by the ileum passes down to the colon where it encounters the microbiome which deconjugates and modifies the primary bile acids, chenodeoxycholic acid and cholic acid, forming the secondary bile acids, deoxycholic acid and lithocholic acid, which can be absorbed from the colon and returned to the liver.

Studies reaching back 50 years showed that the functions of bile acids included disposal of cholesterol from the body, solubilization of cholesterol in bile, and solubilization of dietary long chain lipids in the aqueous environment of the intestine which delivered fat to the brush border in the form of mixed micelles [2, 3]. Research in the last 25 years has illuminated an unexpected function of bile acids as signaling molecules [4]. In the ileum, bile acids interact with farnesoid X receptor (FXR) in the nuclei of ileal enterocytes in order to modulate the production of fibroblast growth factor 19 (FGF19) which inhibits hepatic synthesis of endogenous bile acids, thereby autoregulating bile acid synthesis. Impaired feedback inhibition of bile acid synthesis is thought to be the mechanism of BAM in some patients with chronic diarrhea [5]. Bile acid also interacts with a membrane-bound receptor, variously known as G-protein-coupled bile acid receptor 1 (GPBAR1), Takeda G-protein receptor 5 (TGR5) or membrane-type receptor for bile acids (M-BAR). This receptor located in the colonic epithelium, bile duct, gallbladder, and enteric nerves (and also outside the gastrointestinal tract in macrophages and brown fat cells) is linked to adenylate cyclase, modulating fluid and electrolyte secretion by the colon, among other functions. Activation of colonic GPBAR1 by luminal bile acid is thought to be the mechanism by which excess bile acid in the colon causes bile acid diarrhea (BAD).

Studies of bile acid malabsorption in animals and humans after ileal resection demonstrate the complexity of the pathophysiology of BAD. The concentration of dihydroxy bile acid in the colon (mainly deoxycholic acid and chenodeoxycholic acid) must exceed 3–5 mmol/L in order to produce diarrhea [6]. This can be achieved with short ileal resections (< 25–50 cm) in which the malabsorbed bile acid is dissolved in a relatively normal volume of ileal effluent, since overall small intestinal fluid absorption is not compromised much by such a small resection. In contrast, since a large ileal resection (> 100 cm) substantially decreases intestinal fluid absorption, malabsorbed bile acid is dissolved in a large volume of ileal effluent and may not reach the cathartic threshold concentration. In both cases, although BAM will be substantial, BAM will reduce colonic fluid and electrolyte absorption and aggravate diarrhea only with small ileal resections. Bile acid sequestrants mitigate diarrhea only with small ileal resections, and not with large ileal resections in which the concentration of dihydroxy bile acids in the colon already is below the cathartic threshold.

Another complexity is that diarrhea per se may affect measurement of bile acid absorption. In studies conducted over 30 years ago by our group, fecal radiolabeled bile acid excretion increased in normal individuals with diarrhea induced by ingestion of osmotic agents so much so that the normal range had to be adjusted for stool weight [7]. In those studies, patients with idiopathic chronic diarrhea had fecal radiolabeled bile acid excretion greater than normal individuals with equivoluminous induced diarrhea, suggesting that BAM was present and might be contributing to diarrhea. Nevertheless, cholestyramine had no statistically significant effect on stool weight in these patients, indicating that BAM was not a major driver of diarrhea.

Thus, BAM and BAD are not equivalent. BAM is a necessary antecedent for BAD, but demonstrating BAM is not proof that BAD is present. The diagnosis of BAD is contingent on the presence of BAM and bile acid sequestrant-responsive diarrhea.

BAM can be measured by several direct and indirect tests. In many countries, radiolabeled bile acid retention is measured using the 75Se homocholic acid taurine (SeHCAT) test that measures total body retention of the radiolabeled bile acid at the end of a week with normal values being > 20% retention, and lower values indicating progressively more severe BAM [8]. The test also is predictive to some extent of the response to bile acid sequestrants, but this depends on each patient’s unique situation. For example, knowing whether the patient has had a short or long ileal resection can help interpret the test results; BAM will be present in both cases, but only a driving factor for diarrhea in patients with short bowel resections. Currently, the SeHCAT test is not available in the USA. Measurement of bile acid excretion in stool, though technically difficult, is available. It has not been widely utilized due to unfamiliarity with the test among physicians and the general reluctance to subject patients to timed stool collections. Indirect tests include measurement of serum C4, an intermediate metabolite in bile acid synthesis; increased serum concentrations indicate increased bile acid synthesis by the liver, but do not indicate that diarrhea necessarily is due to BAM. Dr. Camilleri’s group has been at the forefront in popularizing a combination of serum C4 test with measurement of primary bile acids in a single stool sample to simplify assessment [9].

In the Khanna et al. study, bile acid malabsorption was demonstrated by elevated fecal bile acid levels in 48/52 subjects (92%), 15 of whom had no ileal resection and 37 of whom had some degree of ileal resection [1]. In most of these patients, diarrhea likely was multifactorial and related to elevated plasma secretagogue concentrations, accelerated motility, pancreatic exocrine insufficiency, the extent of resection, the effects of treatment, and perhaps alteration of the composition of colonic flora. We have no data on the response of diarrhea to bile acid sequestrants which would have shed some light on the extent to which the demonstrated BAM was contributing to diarrhea.

At a practical level, this paper suggests that BAD should be included in the differential diagnosis of patients with neuroendocrine tumors whose diarrhea has not responded to treatment—especially those with previous surgery. Given the results of this paper, BAM is very likely to be present in these refractory patients (so likely that it may not be worthwhile to measure) and an empiric trial of a bile acid sequestrant which is safe and inexpensive might be beneficial in some.