Idiopathic bile acid malabsorption: qualitative and quantitative clinical features and response to cholestyramine

INTRODUCTION

Chronic diarrhoea in adults frequently gives rise to diagnostic difficulty and extensive investigation may fail to yield a definitive cause. In such patients, irritable bowel syndrome is often proposed as the diagnosis of exclusion.

One cause of chronic diarrhoea is bile acid malabsorption, of which there are three types. These are summarized in Table 1, together with their underlying mechanisms. We are interested in idiopathic bile acid malabsorption (IBAM), previously thought to be rare. It has been defined as chronic diarrhoea, induced by excess bile acid loss, which is responsive to cholestyramine and not associated with other forms of bile acid malabsorption such as ileal dysfunction.¹¹ It is now clear that this condition is more common than previously thought,¹² a sizeable number of patients with IBAM having been initially misdiagnosed as having irritable bowel syndrome.^11, 12

Table 1. . Mechanisms of bile acid malabsorption

Initial investigations used to assess bile acid malabsorption were complicated and time-consuming. They included the ¹⁴C cholylglycine breath test, which measures ¹⁴C radioactivity in expired air and stool.¹³ Measurements of faecal radioactivity have also been used¹⁴ together with measurement of the total excretion and the faecal aqueous concentrations of bile acids.¹⁵ Recent studies, however, have confirmed that the ⁷⁵Selenium HomotauroCholic Acid Test (SeHCAT) is an accurate method of measuring bile acid malabsorption, as well as being simple to use and less distressing to patients than the other tests outlined above.^16–18

The aims of the study were to use the SeHCAT to identify patients with idiopathic bile acid malabsorption, to describe their clinical features both qualitatively and quantitatively, and to assess the response to cholestyramine.

METHODS

The records of all patients referred to our department with chronic diarrhoea over a 2-year period were examined retrospectively. IBAM was considered in patients with chronic diarrhoea, a history suggestive of irritable bowel syndrome (based on the Manning criteria), and with no other obvious cause of diarrhoea. Seventeen patients were selected to undergo the SeHCAT and they were included in the study if their SeHCATs were positive. The SeHCAT is based on 23-selena-25-homotaurocholic acid, a conjugated bile acid labelled with ⁷⁵selenium. This is a gamma emitting synthetic bile acid which is absorbed in the terminal ileum following oral administration. It is secreted into the bile and reabsorbed from the small intestine and resecreted. Abdominal radioactivity is measured on day 7 following a capsule of SeHCAT given orally on day 1. A retention of < 15% of 23-selena-25-homotaurocholic acid confirms the presence of bile acid malabsorption.^16, 17 SeHCAT is the most sensitive test of bile acid malabsorption in the terminal ileum. It is specific for this site but not to the nature of the underlying pathology. It has been shown to correlate well with other tests of IBAM.¹⁶

Possible secondary causes of bile acid malabsorption were excluded by performing the following investigations in all patients: routine blood tests, random glucose, haematinic screen, stool microscopy and culture, small bowel enema to exclude structural ileal disease, gastroscopy and duodenal biopsy to exclude coeliac disease, para amino benzoic acid (PABA) test to exclude pancreatic insufficiency, hydrogen and ¹⁴C-glycocholate breath tests to exclude bacterial overgrowth and barium enema and colonoscopy (six out of nine patients) to exclude large bowel disease. In the six out of nine patients who underwent colonoscopies, biopsies of the terminal ileum were not obtained. Colonic biopsies to exclude microscopic and collagenous colitis were taken in all patients and were normal. None of the patients underwent small bowel permeability tests or had anti-gliadin or anti-endomysial antibodies measured.

Cholestyramine, a bile acid chelating agent, was used as treatment in those with definite IBAM.^19, 20 It acts by binding to bile acids in the small intestine and thus reducing bile acid-induced secretion of water and electrolytes in the colon. Although cholestyramine is effective in the treatment of IBAM, it is unpalatable and not always accepted as a treatment by patients. Therefore, a definite diagnosis of IBAM from the SeHCAT, rather than a therapeutic trial of cholestyramine alone, would be preferable on clinical grounds. The initial dose given was 1–2 sachets three times daily and the dose was titrated accordingly. Adjunctive therapy with loperamide was used initially because we felt this would give the best chance of therapeutic success. Once an effect had been achieved, we gradually withdrew the loperamide. Response to therapy was assessed in an outpatient setting based on the patient’s overall assessment since their last appointment by monitoring: (i) the average stool frequency of bowel motions pre- and post-treatment; (ii) the consistency of stools pre and post-treatment; and (iii) whether symptomatic improvement occurred within the first 24 h.

Statistical analysis was by the Wilcoxon matched pairs test and P < 0.05 was taken as significant.

RESULTS

Nine patients (six males, three females) were diagnosed as having IBAM (median SeHCAT retention 8%, range 3–12.6). Their mean age was 50.2 years (range 43–57). All patients had normal small bowel enemas, duodenal biopsies, PABA tests and hydrogen and ¹⁴C-glycocholate breath tests. They all also had normal barium enemas and/or colonoscopy (six out of nine). No patient had any significant background medical or surgical history, specifically diabetes mellitus or cholecystectomy, which could account for bile acid malabsorption, and none had been exposed to radiotherapy.

The clinical presentation of the nine patients is shown in Table 2. Most patients presented with watery diarrhoea, with a stool frequency varying from four to 20 motions per day. The duration of the diarrhoea ranged from 8 months to 30 years. There was no history of associated blood or mucus passed per rectum in any patient at any time. Two patients reported faecal incontinence (one of whom also had urgency of defaecation) and one complained of nocturnal diarrhoea. Patient 5 had a poor external anal sphincter function measured on endo-anal ultrasound and ano-rectal manometry. She went on to make an excellent recovery with the introduction of cholestyramine and her faecal incontinence resolved. One patient also described pale bulky stools consistent with steatorrhoea.

Table 2. . Clinical characteristics

Five patients complained of abdominal pain, four of whom found that defaecation relieved the pain. One other patient experienced distension and belching without significant pain. Not surprisingly, these five patients had previously been misdiagnosed as having irritable bowel syndrome. In five patients, gastroenteritis or food poisoning was thought to have preceded the onset of symptoms; three following travel abroad. One of these was following a trip to Egypt and one was a Russian teacher who developed giardiasis while leading a school trip to Leningrad. Twenty years previously, his predecessor at the same school had developed an identical illness and was chronically unwell with diarrhoea thereafter, eventually taking early retirement, prior to IBAM being diagnosed. The other two patients had acute episodes of diarrhoea and were thought to have had gastroenteritis clinically, patient 6, 3 years prior, and patient 7 approximately 20 years prior to presentation.

Table 3 shows the faecal weights, fat output, stool frequencies pre- and post-treatment and SeHCAT retentions in the nine patients studied. Their median daily faecal weight was 285 g (range 85–676) and median faecal fat output was 17 mmol/24 h (range 8.3–38.8). Six patients had an immediate response to cholestyramine (within 24 h) which was subsequently maintained even after withdrawal of loperamide. Overall, there was a marked reduction in stool frequency following therapy, with a median stool frequency pre-treatment 5/day versus 2/day post-treatment, P = 0.03. This analysis includes two patients in whom treatment was not tolerated. One other patient did not experience a reduction in stool frequency on treatment, although bowel motion consistency improved and the patient was happy with this outcome. All other patients who responded to treatment by a reduction in stool frequency also noticed an improvement in their stool consistency. Patients with the lowest SeHCAT retention had the best response to cholestyramine in terms of stool frequency reduction and consistency, as shown in Table 3.

Table 3. . Faecal weights, fat output, response of bowel frequency to cholestyramine and SeHCAT percentage retention

DISCUSSION

Diarrhoea is defined as an abnormal increase in stool liquidity and daily stool weight. This is usually associated with frequency (> 3 stools/day), and accompanied by urgency, perianal discomfort, incontinence or a combination of all three. A daily faecal weight openface>  200 g/day may be considered the upper limit of normal.

In the present study, we carefully and systematically excluded possible secondary causes of bile acid malabsorption, leaving the diagnosis of ‘pure’ idiopathic bile acid malabsorption confirmed by the SeHCAT in all nine patients. Previous studies have either failed to extensively investigate underlying secondary causes^12, 21 or did not use the SeHCAT.^{7, 11, 22}

Two of the five patients presented with a history of giardiasis, acquired during visits to Russia. Clinical manifestations of giardiasis are caused by an impairment of the absorptive capacity of the gut, which is normally reversible with specific therapy. However, both these patients went on to develop idiopathic bile acid malabsorption, as all their remaining tests were negative, in particular the ¹⁴C-glycocholate breath test and stool culture and microscopy. It is possible that the giardial infection left a selective impairment of the active ileal bile acid transport mechanism in its wake. It is unclear whether other types of gastroenteritis have a similar effect. It is interesting that a total of five out of our nine patients had evidence of previous gastroenteritis or food poisoning, which could theoretically have a similar effect on the active ileal bile acid transport mechanism.

Two patients suffered from faecal incontinence. These patients (3 and 5) had two of the highest faecal weights of all the patients, thus compromising the anal sphincter. Patient 3 was unable to tolerate the cholestyramine, but was happy to carry on in the knowledge that nothing serious was going on and no further investigation was requested. Williams et al. found a nocturnal component in half their patients with idiopathic bile acid malabsorption and felt it was a useful diagnostic indicator.²³ Only one of our patients (patient 6) reported nocturnal diarrhoea and this was occasional. The nocturnal episodes in this patient responded fully to therapy. In common with other studies, we also found that the lower the retention in the SeHCAT, the better the response to cholestyramine.⁹

Under normal conditions, bile acids are efficiently conserved. Conjugated and unconjugated bile acids undergo passive diffusion into the blood stream throughout the entire gut, but only conjugated bile acids are actively transported in the distal ileum. The reabsorbed bile acids enter the portal bloodstream and are taken up rapidly by hepatocytes, reconjugated and resecreted into bile (the ‘enterohepatic circulation’). Failure of the active transport of bile acids in the terminal ileum results in bile acid malabsorption.²⁴ At the same time, the liver is unable to increase the rate of bile salt production to a sufficient degree to compensate for the loss and fat digestion is thus compromised, leading to steatorrhoea and malabsorption of fat-soluble vitamins. Bile salts entering the colon lead to secretory diarrhoea, thus exacerbating the problem.

Three types of bile acid malabsorption are recognized and these are thought to arise from different interruptions in the enterohepatic circulation, giving rise to bile acid diarrhoea.²⁵ The three types of bile acid malabsorption, their causes and underlying pathophysiological mechanisms are summarized in Table 1. Type 1 relates to ileal dysfunction, Type 2 is idiopathic and Type 3 miscellaneous. Several theories have been proposed to account for idiopathic bile acid malabsorption. The most widely accepted was proposed by Thaysen & Pederson and is thought to be due to a defect in the bile acid transport system.^8, 11 Thaysen also later suggested a relative lack of receptors for bile acids in the ileum.⁷ Popovic et al. found mucosal atrophy of the ileum with subtotal villous atrophy and crypt hyperplasia.¹⁰ They also went on to propose an autoimmune disorder, with the ileum as the target organ.¹⁰ Some authors feel, however, that this theory is more in keeping with Type I bile acid malabsorption, as there are distinct morphological changes in ileal histology.²¹ These authors feel that a motor disturbance of the ileum, causing a shortened ileal transit time, may be important, leading to a reduction in the time available for normal reabsorption.²¹ Heubi et al. described two children who presented with congenital diarrhoea.⁸ They felt that the children had a previously undescribed congenital transport defect that includes the absence of active ileal bile acid transport, but they were unable to prove this.

Four patients (numbers 2, 6, 7, 9) had clinical features consistent with irritable bowel syndrome. They also had normal daily faecal weight and faecal fat output measured over 3 days. In this study, we have shown that the measurement of faecal fat and weights is central to determining whether there is large or small volume diarrhoea, or steatorrhoea. Where there is large volume diarrhoea, IBAM should be suspected. It is easily diagnosed and response to therapy is often very good. However, some patients with IBAM have normal stool weights and clinical features of irritable bowel syndrome. IBAM should especially be suspected if patients have a past history of gastrointestinal infection. Previous studies have not indicated any relationship between gastrointestinal infection and IBAM. The majority of these patients had previously been diagnosed as suffering from irritable bowel syndrome and this is due to a similar symptom complex being present in both conditions, i.e. abdominal pain relieved by defaecation, or bloating and belching along with diarrhoea. The SeHCAT has revolutionized the diagnosis of bile acid malabsorption. Our findings suggest that many patients with IBAM are unrecognized, being wrongly labelled as irritable bowel syndrome patients, and that, in the first instance, their stool weights should be measured to assess whether they have small volume diarrhoea suggestive of irritable bowel syndrome, or large volume diarrhoea, indicating assessment for IBAM. Whilst most patients with IBAM have large volume diarrhoea, it is true that four out of nine in our series had normal stool weights, and thus a high index of suspicion is required if the diagnosis is not to be missed. Many patients could be offered definitive treatment for their diarrhoea and avoid the label of irritable bowel syndrome in the future.