Natural history of acalculous biliary symptoms

INTRODUCTION The radiological criteria for the diagnosis of gallbladder disease rely largely on the detection of calculi using ultrasonography. Patients may, however, suffer symptoms typical of biliary pain without detectable gallstones. The aim of this study was to identify a cohort of patients presenting with recurrent episodes of biliary symptoms in the absence of identifiable pathology on ultrasonography and to record the outcome of subsequent imaging investigations. METHODS Records of patients having abdominal ultrasonography during a four-month period in 2006 were accessed retro-spectively and those with symptoms suggesting biliary disease were identified. Radiology records were reviewed over a five-year follow-up period to identify patients undergoing further imaging for recurrent biliary symptoms and outcomes were recorded. RESULTS A total of 512 patients had ultrasonography for investigation of symptoms consistent with biliary disease. Almost half (41.2%) of these were found to have gallbladder pathology on ultrasonography and 4.7% of patients went on to have further investigations for similar symptoms without achieving a diagnosis. The median age of this group was 47 years and 75% of these patients were female. During the follow-up period, 2.6% of patients with biliary symptoms and initially normal ultrasonography developed gallstones and in 1.3% pancreatitis was demonstrated on imaging. CONCLUSIONS A small minority of patients who present with biliary symptoms and have no abnormality on ultrasonography recurrent symptoms develop significant biliary pathology. These patients should be identified by interview at routine follow-up visits and further investigations should be

Between 10% and 20% of adults in Western countries have gallstones, 1 with 30-50% having symptoms at the time of diagnosis. 2 Gallstone disease causes a symptom complex that includes upper abdominal pain in more than 90% of patients. 3 The radiological criteria for the diagnosis of gallbladder disease rely largely on the detection of calculi and transabdominal ultrasonography is the most commonly used initial diagnostic test in patients presenting with biliary symptoms. Patients may, however, suffer symptoms typical of biliary pain without detectable gallstones. In these patients, the symptoms may be caused by an alternative diagnosis or may originate in a dysfunctional biliary system. 4 Biliary pain in this context may be due to gallbladder dysmotility 5 and is often promoted by eating fatty foods.
Little is known about the natural history of dysfunctional gallbladder pain and the ability to perform studies is limited by the difficulty in defining a patient population in the ab-sence of reliable diagnostic tests. The Rome III criteria were formulated in an attempt to define the syndrome. 6 The criteria to fulfil a diagnosis of dysfunctional gallbladder pain include recurrent episodes of epigastric or right upper quadrant pain that are severe enough to interrupt daily activities or prompt patients to seek medical care, without evidence of structural disease. It is important to identify these patients as the dysfunctional gallbladder may cause biliary stasis, which is a cause of gallstone formation, and dysfunctional biliary pain may be cured by cholecystectomy. 7 A retrospective study was performed to identify a cohort of patients presenting with recurrent episodes of abdominal symptoms suggestive of biliary disease so that an estimate could be made of the incidence of the problem compared with symptomatic gallstone disease. Furthermore, the natural history of the disease over a five-year period of radiological follow-up was assessed. Excluded patients comprised those under 16 years of age at the time of the investigation, those in whom the gallbladder was absent or not identified and those for whom no clinical details were available on the request card. Patients who had ultrasonography to detect potential biliary disease were identified if they had any of the following phrases written on the request card: 'right upper quadrant pain', 'epigastric pain', 'biliary colic', 'query gallstones', 'query cholecystitis' or any other phrase suggesting disease of the gallbladder or biliary tract.
The proportion of this group with structural biliary pathology identified on imaging was recorded. The radiology records of those without a diagnosis were then scrutinised to determine whether repeat investigations were performed for recurrent similar symptoms in the follow-up period, by identification of the same phrases mentioned above on further request cards. The details of investigations of patients where structural biliary pathology was identified on repeat imaging were recorded. In the group of patients where repeated normal investigations were performed for biliary symptoms, endoscopy records were accessed and findings recorded.

Results
A total of 1,614 patients had abdominal ultrasonography performed in our hospital during the study period but 1,102 patients were excluded (Fig 1). This left 512 patients who had complete abdominal ultrasonography performed including the gallbladder and a request card suggesting biliary symptoms.
Of these 512 patients, structural biliary pathology was identified in 211 (41.2%), of which gallstones were the most common finding (86%). Of the remaining 301 patients with symptoms but no abnormality, 168 patients (55.8%) had no further imaging undertaken. Of the 133 patients (44.2%) undergoing repeat imaging, 40 were noted to have had earlier abdominal ultrasonography for which no records were available. During the 5-year follow-up period, 93 patients were identified as having had repeat imaging including ultrasonography, computed tomography or magnetic resonance cholangiopancreatography. Of this group, 39 patients had imaging performed for other clinical indications and 9 were excluded because they had multiple investigations on the initial inpatient admission.
Forty-five patients remained with recurrent symptoms compatible with biliary pathology and repeated investigations during the follow-up period for evaluation. Among this group, eight patients (18%) developed gallstones and four (9%) were shown subsequently to have pancreatitis. Thirtythree patients (73%) with recurrent biliary symptoms had no abnormality detected on subsequent imaging. Twentyone of these patients had an oesophagogastroduodenoscopy performed during the follow-up period and thirteen were found to have other pathology that could explain their symptoms (mucosal inflammation or ulcer disease). The re-  Patients undergoing recurrent investigations for acalculous biliary symptoms comprised 3.9% of the initial 512 patients undergoing ultrasonography for biliary symptoms and 6.7% of the 301 patients with a normal result. The probability of a patient with biliary symptoms and normal ultrasonography developing gallstones was 2.6% (8/301) and for pancreatitis it was 1.3% (4/301) during the five-year follow-up period.

discussion
Studies of functional gastrointestinal disorders are difficult to undertake due to the lack of agreed definitions and diagnostic tests. This study represents a rare attempt to define and follow a cohort of patients with acalculous biliary symptoms. Our study reveals a number of findings in relation to this problem. First, 8 of 301 (2.6%) symptomatic patients with no structural biliary abnormality on their initial imaging were later shown to develop gallstones and 4 patients (1.3%) were admitted with a confirmed diagnosis of pancreatitis on imaging. Second, 3.9% of patients who presented with symptoms compatible with gallbladder disease went on to have further investigations for similar symptoms without achieving a diagnosis, compared with 41.2% of patients in the same population who were shown to have structural biliary disease (a relative rate of 0.09).
The hospital admission rate with symptomatic gallbladder disease is approximately 115 cases per 100,000 per year in the UK 8 and the population rate of cholecystectomy is approximately 100 per 100,000 per year. 9 From these figures, an estimate of the population incidence of recurrent acalculous biliary pain requiring investigation of approximately 10 cases per 100,000 per year can be made.
The limitations of this study relate to its retrospective nature and the inability to define the patients' symptoms precisely as the patients were not interviewed directly. Many questionnaire studies have been performed to describe the symptoms of gallbladder disease, and all have shown upper abdominal pain to be the most common symptom and the one most likely to be cured by cholecystectomy. 3, 10 We have assumed that the doctor requesting the initial ultrasonography suspected a diagnosis of gallbladder disease based on the patients' symptoms as the key phrases sought on the request cards are used commonly in this context and the high rate of diagnosing structural biliary disease in these patients (41.2%) supports this. Our follow-up data also relied on patients re-presenting to medical practitioners and having further investigations requested.
The low rate of subsequent development of gallstones in patients with acalculous biliary symptoms is an important finding. Although ultrasonography has a very high sensitivity for the detection of gallstones, 11 for some of the patients shown subsequently to have gallstones, this may have been due to the initial imaging result being falsely negative. Although transabdominal ultrasonography can detect stones down to 3-5mm, endoscopic ultrasonography provides greater sensitivity, particularly for microlithiasis of <3mm, 12,13 and it has been shown that up to half of patients with biliary symptoms and normal ultrasonography had a biliary abnormality identified using endoscopic ultrasonography. 14 It is also possible that in some patients the onset of biliary symptoms may precede the development of visible gallstones, perhaps due to gallbladder dyskinesia promoting biliary stasis. The annual population incidence of hospital admission for symptomatic gallstones for women aged 45-54 years in the UK is 0.23% 8 (approximately 1% over 5 years). The rate of detection of symptomatic gallstones of 2.6% of the initial population with acalculous biliary symptoms over five years in this study suggests that if the risk of gallstone formation in this group is raised compared with the background population, the increased risk is very low. Cholecystectomy should not therefore be offered to these patients to prevent gallstone formation. The rate of subsequent acute pancreatitis in these patients (1.3% over five years) may, however, be increased compared with the annual population incidence in the UK (0.02%) 15 although the number of patients at risk is low. Larger studies are needed to investigate this.
It is reassuring that the majority (55.8%) of patients who present for the first time with upper abdominal pain and have normal ultrasonography have no further imaging undertaken, suggesting the symptoms were short-lived. In addition, only 20 of the 301 patients (6.7%) presenting in this way went on to have repeated normal investigations for similar symptoms. Nevertheless, 40 patients in this group were noted to have had earlier imaging undertaken for which no information was available and some of these may have been for the investigation of earlier biliary symptoms. The small group undergoing repeated investigations is likely to have contained patients with dysfunctional gallbladder pain, who may have benefitted from further investigations.
Hepatobiliary iminodiacetic acid (HIDA) imaging provides a quantitative assessment of stimulated gallbladder ejection 16 but has not been shown consistently to identify patients who benefit from surgery. 17 HIDA imaging was not offered to any of the patients in this study, and no patient with biliary symptoms and repeated normal investigations underwent a cholecystectomy.
conclusions Our findings show that only a small minority of patients who present with abdominal symptoms suggesting biliary disease and have no abnormality found on initial ultrasonography go on to be investigated for recurrent symptoms or develop significant biliary pathology. Routine follow-up visits in clinic to identify these patients is therefore probably not be necessary. However, in patients who do re-present, HIDA scans or endoscopic ultrasonography should be considered.