A positive SeHCAT test results in fewer subsequent investigations in patients with chronic diarrhoea

Abstract

Chronic diarrhoea is a common condition, resulting from a number of different disorders. Bile acid diarrhoea, occurring in about a third of these patients, is often undiagnosed. We hypothesised that a positive diagnosis of bile acid diarrhoea would reduce the need for subsequent investigations for alternative diagnoses.

Methods

Patients previously recruited to a study of chronic diarrhoea who had selenium homocholic acid taurine (SeHCAT) testing and subsequent follow-up at our institution were identified. In a retrospective analysis, the numbers of defined investigations undertaken from the first 3 months after SeHCAT in the following 5 years were compared.

Results

90 patients were identified with primary bile acid diarrhoea (SeHCAT retention <15%, n=36) or idiopathic diarrhoea (SeHCAT retention >15%, n=54). Follow-up had been performed on 29 and 39 subjects, respectively, with no differences in previous investigations or the last contact date. In the follow-up period, the proportions of these patients who had undergone endoscopic procedures (gastroscopy, colonoscopy and sigmoidoscopy) were the same. However, there was a higher proportion of patients in the SeHCAT-negative group who had other investigations, including imaging, physiological tests and blood tests (p=0.037). The use of cross-sectional imaging was significantly higher in this group (p=0.015) with greater proportions having CT (0.44 vs 0.10) and MRI (0.26 vs 0.07). Ultrasound use and the number of blood tests were higher in the SeHCAT-negative group whereas the SeHCAT-positive group attended more clinic appointments (p=0.013).

Conclusion

A positive diagnosis of bile acid diarrhoea, made by a SeHCAT test, resulted in reduced use of diagnostic investigations over the subsequent 5 years.

Introduction

Chronic diarrhoea is a common presenting problem in primary and secondary care in the UK and elsewhere. After conditions including inflammatory bowel disease (ulcerative colitis and Crohn’s), microscopic colitis, coeliac disease and colonic neoplasia have been excluded, many patients are told they have functional causes for the diarrhoea.¹ Functional chronic diarrhoea and irritable bowel syndrome with diarrhoea are overlapping condition and estimates suggest 3% of the population may suffer from these.

It has been shown in multiple studies that around 25%–33% of these people with functional chronic diarrhoea have primary bile acid diarrhoea (PBAD), also known as idiopathic bile acid malabsorption.^{2 3} Diagnostic methods have been reviewed recently.⁴ The selenium homocholic acid taurine (SeHCAT) test uses tauroselcholic⁵ acid (SeHCAT) to identify excessive faecal bile acid loss by measuring whole body retention of radiolabelled ⁷⁵Se by gamma camera at 7 days. Retention values between 10% and 15% indicate mild bile acid loss, those between 5% and 10% moderate and less than 5% indicating severe loss. Patients with PBAD will usually show a response to bile acid sequestrants such as colestyramine or colesevelam.⁶ Alternative diagnostic methods include measurement in blood of a bile acid precursor, 7α-OH-4-cholestenone (C4), faecal bile acid quantification or a therapeutic trial of treatment.

Clear recognition and quantification of the excessive bile acid loss in PBAD enables the therapeutic options to be refined and modified so that optimal individualised patient responses can be obtained.^{7 8} Although such expert opinion recognises the value of SeHCAT testing in clinical practice, an economic review for National Institute for Health and Care Excellence in 2012 found there was insufficient data for clear conclusions to be drawn.⁹ Guidelines for the investigation of chronic diarrhoea are in need of updating¹⁰ and recent patient-reported symptoms and outcomes suggest many patients suffer for over 5 years as clinician awareness of appropriate diagnostic methods is not widespread.¹¹

A potential benefit of a timely diagnosis of PBAD is that this could result in fewer investigations being performed to look for possible alternative diagnoses. This has not been assessed previously. We hypothesised there would be a lower subsequent use of investigations in patients who had a positive SeHCAT result. The aim of this study was to provide initial data on the use of investigations following a SeHCAT test to help in the evaluation of any economic benefit of an abnormal SeHCAT test.

Methods

Patient groups

We previously recruited patients with chronic diarrhoea to a prospective study of chronic diarrhoea between January 2009 and July 2010.¹² They all had SeHCAT testing and subsequent follow-up at our institution as part of a study approved by the Research Ethics Committee of Hammersmith and Queen Charlotte’s & Chelsea Hospital. Their clinical features have been described before.¹² We have now retrospectively analysed the investigations they had at the time of diagnosis and subsequently.

Two cohorts of patients were studied. The group with bile acid diarrhoea had a SeHCAT 7-day retention value of <15% (‘SeHCAT positive’). Only those with PBAD were included; patients with secondary causes including Crohn’s disease, ileal resection or postcholecystectomy diarrhoea were excluded. The second group of idiopathic chronic diarrhoea (‘SeHCAT negative’) had a SeHCAT 7-day retention >15%. The cohorts were then restricted to those who had continued to be seen in any department of our healthcare trust for 5 years from the date of their SeHCAT investigation.

Investigations

Electronic medical records were searched systematically to identify the number, type and indication for defined pertinent investigations in all patients in the cohort. This included clinical correspondence, outpatient appointment, endoscopy, radiology, physiology and pathology report databases. Any overlap was accounted for.

Investigations undertaken for 5 years from the date of the SeHCAT scan in all patients were identified. Tests carried out within 3 months of the SeHCAT test date were analysed separately, as they were likely to have been ordered before the SeHCAT result was known. Any investigations undertaken after the 5-year follow-up period were not included.

All investigations performed for a gastrointestinal condition, or relevant symptoms, were counted. These included the following: gastroscopy, colonoscopy, flexible sigmoidoscopy, endoscopic retrograde cholangiopancreatogram (ERCP), capsule endoscopy, CT, MRI, positron emission tomography, ultrasound (US), abdominal X-ray, glucose-hydrogen and lactose-hydrogen breath testing. All blood tests were recorded. Out-patient visits were identified, but only gastroenterology appointments following the SeHCAT result appointment were counted in order to assess the impact on subsequent follow-up.

Analysis

The number of specific and pooled investigations per patient and per group was compared between the two groups. Median numbers of investigations were compared with non-parametric Mann-Whitney U tests and the proportions per group with Fisher’s exact test. A p value of 0.05 was taken as significant.

Results

A total of 114 patients from the previous study were seen at our institution, and of these, 90 patients were identified with possible PBAD (SeHCAT <15%, n=36) or idiopathic diarrhoea (SeHCAT >15%, n=54). Follow-up data were available on 29 ‘SeHCAT-positive’ and 39 ‘SeHCAT-negative’ patients, respectively (table 1). There was no difference between these groups in the date of the last contact in our institution or the length of follow-up. Two patients included in the SeHCAT-negative group had died from unrelated conditions. The clinical features of the two groups did not differ from those we described before.¹² Endoscopic and other investigations were performed in similar proportions before the SeHCAT test (table 1). There was also no difference in the number of these investigations performed in the 3 months immediately following the SeHCAT test (table 2), which could have been ordered as part of the initial evaluation of diarrhoea.

Table 1.

Patient details and investigations performed before SeHCAT testing

	SeHCAT positive (<15%)		SeHCAT negative (>15%)
	Number	Tests/ patient	Number	Tests/ patient
Patients (n)
Diagnosed	36		54
With 5-year follow-up	29		39
SeHCAT (7-day % retention)
Mean	9.0		33.9
Minimum	3.0		15.6
Maximum	14.4		90.0
Investigations before SeHCAT
Endoscopy
Colonoscopy	16	0.55	20	0.51
Flexible sigmoidoscopy	6	0.21	7	0.18
Gastroscopy	8	0.28	12	0.31
Physiology
Glucose hydrogen-breath test	2	0.07	4	0.10
Lactose hydrogen-breath test	7	0.24	3	0.08

Data are shown as proportions of the patients in each group with 5-year follow-up.

Table 2.

Investigations performed in the 3 months after  SeHCAT testing

	SeHCAT positive (<15%)		SeHCAT negative (>15%)
	Number	Tests/ patient	Number	Tests/ patient
Patients (n)	29		39
Endoscopy
Colonoscopy	5	0.17	8	0.21
Flexible sigmoidoscopy	0	0.00	0	0.00
Gastroscopy	4	0.14	2	0.05
Physiology
Glucose hydrogen-breath test	0	0.00	3	0.08
Lactose hydrogen-breath test	8	0.28	8	0.21

Investigations

In the 5-year follow-up period, there was no significant difference in the proportion of patients who had undergone endoscopic procedures including gastroscopy, colonoscopy, flexible sigmoidoscopy, capsule enteroscopy and ERCP (table 3). However, there were statistically significant greater proportions of the SeHCAT-negative group who underwent investigations when all imaging (CT, MRI and US), physiological tests and blood tests were analysed (p=0.037). The Use of cross-sectional imaging (abdominal CT and MRI) was particularly higher in the SeHCAT-negative group (p=0.018). US use was higher but did not differ significantly. Six patients (five SeHCAT-negative, one SeHCAT-positive) had both CT and MRI.

Table 3.

Investigations performed between 3 months and  5 years after SeHCAT testing

	SeHCAT positive (<15%)		SeHCAT negative (>15%)
	Number	Tests/ patient	Number	Tests/ patient
Patients (n)	29		39
Endoscopy
Colonoscopy	6	0.21	7	0.18
Flexible sigmoidoscopy	0	0.00	4	0.10
Gastroscopy	6	0.21	4	0.10
Imaging
CT	3	0.10	17	0.44a
MRI	2	0.07	10	0.26
CT or MRI	5	0.17	27	0.70b
US	16	0.55	28	0.72
Abdo X-ray	9	0.31	15	0.38
Physiology
Glucose hydrogen-breath test	1	0.03	2	0.05
Lactose hydrogen-breath test	0	0.00	3	0.08
Other investigations	3	0.10	5	0.13
Blood tests/patient
Median		21		33c
IQR		4–37		12–77
Out-patient clinic visits/patient
Median		4		1d
IQR		2–8		0–6

^ap=0.045; ^bp=0.015 two tailed Fisher’s exact test; ^cp=0.086; ^dp=0.013; Mann-Whitney U-test.

More blood tests had been undertaken in the SeHCAT-negative group compared with the SeHCAT-positive group. This did not reach significance (p=0.085). However, the SeHCAT-negative patients had fewer gastroenterology out-patient clinic appointments than the SeHCAT-positive patients during the 5-year period of follow-up (p=0.013). As expected, the SeHCAT-negative patients had greater variability in both number of blood tests and in follow-up appointments.

Discussion

This study has been performed to give initial data regarding how a positive diagnosis of PBAD, made by SeHCAT testing, influences subsequent use of investigations over a 5-year period in a cohort of patients presenting with chronic diarrhoea. The greatest difference found was in the use of cross-sectional abdominal imaging by CT and MRI.

There are many causes of chronic diarrhoea, with PBAD being found in a large proportion% in the prospective series which included the patients studied here.¹² The diagnosis can be made by SeHCAT scanning in a relatively simple, non-invasive procedure, which gives a greater proportion of positive results than other, less proven tests.⁴ Patients value having the correct diagnosis made in a timely fashion; almost half of those in a recent survey had waited for more than 5 years and felt that their symptoms had been dismissed before they were diagnosed with PBAD.¹¹ Treatment with bile acid sequestrants is effective in a majority, as demonstrated in 75% of these patients in our previous publication.¹² There are clearly patient benefits in being diagnosed in a timely manner. We have shown that there are also likely to be economic and organisational benefits in the usage of other diagnostic resources.

The use of endoscopic procedures to rule out important causes of chronic diarrhoea such as colorectal cancer and inflammatory bowel disease was similar in both groups. The British Society of Gastroenterology guidelines from 2003¹⁰ are currently undergoing revision and are expected to place the exclusion of these diseases as a high priority. Our data indicates that exclusion of bile acid diarrhoea by SeHCAT will result in less use of cross-sectional imaging by CT and/or MRI. These are usually performed to exclude possible chronic pancreatic disease or small bowel pathology not found on endoscopy or ruled out by other tests. The yield of these tests in our series of patients was very low, but they are still likely to be performed if no definite diagnosis has been made in a symptomatic patient. Further tests will also be performed in the minority (25%) of SeHCAT patients who do not experience a satisfactory, full response to bile acid sequestrants.¹²

These results provide some data to help in the assessment of the economic benefits of diagnosing PBAD. We can assume that the SeHCAT-positive patients, had they not been diagnosed, would have had the same utilisation of other tests as the SeHCAT-negative patients. Using the difference in proportions undergoing these tests and their NHS unbundled tariffs,¹³ net savings per patient would be: from CT (0.34×£135=£46), MRI (0.19×£240=£46) and US (0.17×£56=£10)=£102. There would be further saving from the costs of blood tests. A more detailed study is needed to assess fully the economic benefits of potential improvements in the quality of life, including any reduction in primary care visits, as opposed to treatment and follow-up costs, and the cost of SeHCAT, in making the diagnosis of PBAD.

There are several limitations to our study. The size of the cohorts is not large, which has limited the power to detect significant differences. The analysis of these investigations has been performed retrospectively and we did not follow any specific defined protocols for further investigation or treatment but left these to the clinical judgement of the referring clinician. Although we have 5-year follow-up data on subsequent investigations, we cannot be certain that the patients have not attended other hospitals for investigations. However, as there were more out-patient attendances in the SeHCAT-positive group, but more investigations in the SeHCAT-negative group, this difference is likely to be greater if further investigations had been performed elsewhere. Furthermore, the greater number of out-patient attendances in the SeHCAT-positive group likely results from continued follow-up of these patients for potential therapeutic research and so may not be generalizable.

In conclusion, this study has shown that in a cohort of patients with chronic diarrhoea, there is a lower usage of other investigations in those with a diagnosis of PBAD in the 5 years after a SeHCAT test. This provides additional data to support measures to increase the awareness of bile acid diarrhoea and the benefits of making a positive diagnosis.

Significance of this study.

What is already known on this topic?

• Bile acid diarrhoea is a common cause of chronic diarrhoea but is often undiagnosed.

• The selenium homocholic acid taurine test is an objective way to diagnose these patients.

• Whether diagnosing bile acid diarrhoea reduces the need for other investigations is unknown.

What this study adds?

• The use of investigations in the 5 years after a SeHCAT test was higher in those with a negative test than in those with a positive test diagnosing bile acid diarrhoea.

• Cross-sectional imaging (CT and MRI) use was particularly higher.

• The number of blood tests performed was also higher in the SeHCAT-negative group.

How might it impact on clinical practice in the foreseeable future?

• Making a positive diagnosis of bile acid diarrhoea by a SeHCAT test can reduce subsequent use of other diagnostic investigations.