| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full Paper |
Departments of 1 Surgery and 2 Radiology, Aberdeen Royal Infirmary, Foresterhill, Aberdeen AB25 2ZN and 3 Victoria Hospital, Kirkcaldy, Fife KY2 5AH, UK
Correspondence: Mr V Shanmugam, Specialist Registrar, Department of Surgery, Nottingham City Hospital, Hucknall Road, Nottingham NG5 1PB, UK
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
The magnitude of the clinical problem of gallstones in Europe is considerable with an estimated prevalence of 5–20% [1] in the general population, rising with age. The prevalence of gallstones in women aged 60–70 years is 20–40%, and 10% in men of the same age group. The incidence of gallstones is three times more common in women than in men and rises with age irrespective of sex [2] giving a prevalence of 31% in women between 61 years and 70 years and 52% between 71 years and 90 years [3]. It is reported that at the time of cholecystectomy for symptomatic cholelithiasis, 8–15% of patients less than 60 years and 15–60% of patients over 60 years have common bile duct (CBD) stones [4]. Choledocholithiasis may be asymptomatic; or symptomatic with potential complications including pancreatitis, cholangitis and obstructive jaundice [5] adding further to the burden of management in gallstone disease. This is particularly the case in older patients with the additional attendant risks of invasive intervention associated with increased cardio-respiratory morbidity and diminished physiological reserve. Consequently, the European Association for Endoscopic Surgery (EAES) consensus development conference committee recommends common bile duct investigation to rule out choledocholithiasis in all patients with symptomatic cholelithiasis [4].
Various clinical, biochemical and investigative procedures may be used to identify the ductal calculi [6]. Biochemical tests carry poor sensitivity and specificity. Ultrasound is an easily available, non-invasive and low cost investigation with no requirement for ionizing radiation. However, it is highly operator dependent [4], misses two in three patients with common bile duct stones and is subject to interference from bowel gas. It is particularly poor in detecting smaller partially obstructing calculi. The distal CBD may not routinely be examined if there is no extrahepatic biliary dilatation and this may further diminish the sensitivity.
Endoscopic ultrasound (EUS) facilitates better visualization of the entire extrahepatic biliary tree with high sensitivity (88%) and specificity (96%) [7, 8]. Despite the fact that both ERCP and EUS requires specialized expertise to perform, limited availability of EUS restricts its routine use at present for CBD evaluation. ERCP, intraoperative cholangiogram (IOC), percutaneous transhepatic cholangiography (PTC) and operative CBD exploration are considered as the gold standard procedures in the diagnosis of choledocholithiasis. The invasive nature of CBD exploration, IOC and PTC carries a significant morbidity and mortality. Furthermore, a high level of expertise is required in their performance.
ERCP is highly sensitive and specific, but invasive and inconvenient for the patient, requiring sedation and contrast (with minimal risk of allergic reaction), and associated with significant morbidity (5–10%) and mortality (<1%) [4]. MRCP is non-invasive for studying the biliary tree and requires no ionizing radiation or iodinated contrast. It gives high-resolution projectional images of the CBD with no known hazards, in the absence of incompatible foreign bodies. Our study aimed at evaluating predictive value of MRCP in assessing symptomatic patients prior to cholecystectomy or further biliary intervention with particular reference to laparoscopic surgery. We also evaluated the potential for additional benefit including the diagnoses of other hepatobilary and pancreatic pathology.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Patients who were claustrophobic to MRCP and in whom ERCP had failed were excluded from the predictability calculation. Patients who underwent MRCP (true negative for CBD stone) but without a comparable reference procedure were also excluded.
Imaging technique
MRCP at Kirkcaldy hospital was performed using a 0.5 Tesla Elscint Privilege MRI scanner with two sequences; axial T2 weighted for liver and multislice RARE sequence (10 mm slice thickness, three-dimensional (3D) maximum intensity projection to reconstruct the bile duct). MRCP at Aberdeen was performed using a 1 Tesla Siemens Magnetron impact MRI scanner with three sequences; axial T2 weighted turbo spin echo breath hold (slice thickness 10 mm, repetition time (TR) 2054, echo time (TE) 128), coronal and coronal oblique thick slab RARE and thin slice (6 mm) coronal HASTE (TR 11, TE 87). No oral or intravenous contrast medium was used.
Statistical analysis
The data were entered in a Microsoft Excel database. Statistical calculations were performed using Fisher's exact test, supported by StatsDirect® statistical software version 2.3.8 for Windows, to predict the association between the variables (ERCP and MRCP) following appropriate statistical advice. A p-value of <0.05 was considered statistically significant. The results of MRCP were analysed against ERCP or other invasive imaging for the entire study population. Detailed subgroup analysis was also carried out for ERCP following MRCP (n=176), ERCP prior to MRCP (n=107) and MRCP following ERCP in whom ERCP was subsequently repeated (n=13). Where ERCP was unavailable but an alternative form of invasive imaging was carried out (IOC or PTC) (n=9), this was used as reference standard. Sensitivity, specificity, positive and negative predictive values were individually calculated in the usual manner.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Ultrasound
Ultrasound scan of the hepatobiliary system was the primary investigation in 312 patients of whom 29 had previously undergone cholecystectomy. 196 (63%) showed evidence of cholelithiasis and 13 biliary sludge. Other abnormalities seen were stricture, duplication of gall bladder, cholesterol polyp (one each) and choledochal cyst in two patients. No abnormality was observed in 98 patients.
The CBD was dilated (more than 7 mm) in 114 (36.5%). Ultrasound evidence of choledocholithiasis was seen in 31 (31%) of the 99 patients subsequently diagnosed by ERCP. In 37 (12%) studies, visualization of the CBD was obscured by gas.
MRCP
Of 351 patients referred, MRCP was successfully performed in 337 (96%). 14 study failures were due to claustrophobia, lack of co-operation (inability to lie still) or excessive body mass.
ERCP
ERCP was attempted in 283 (84%) of the 337 patients who underwent MRCP. In 163 (58%) of cases ERCP was performed subsequent to MRCP, before MRCP in 107 (37%) and both before and after MRCP in 13 (5%). ERCP was successful in 212 (75%) patients with a maximum of four attempts (81 failures occurred after the first attempt, 12 after the second and one failure after the third and fourth attempts, respectively). It was a common practice to perform sphincterotomy and balloon trawl in cases with a strong suspicion of biliary calculus despite reasonable contrast imaging for choledocholithiasis. However, there has been a general change in the practice recently with the aim of avoiding additional morbidity associated with sphincterotomy. In our study, 158 patients underwent sphincterotomy and balloon trawl. The most common reasons for unsuccessful ERCP were: failure to identify the ampulla, mobile ampulla, duodenal diverticulum, agitated or uncooperative patient, pyloric stenosis, duodenal stricture and previous gastrectomy. 12 (5.7%) patients developed clinically significant pancreatitis following ERCP.
Further imaging
In an additional nine patients where ERCP was failed, comparator imaging data were available from other invasive procedures (IOC, PTC).
Accuracy of MRCP
A total of 221 patients with complete comparative data were considered for detailed analysis. The overall sensitivity and specificity of MRCP was 97.98% (95% CI, 92.19–99.64, p<0.0001) and 84.43% (95% CI, 76.49–90.13, p<0.0001), respectively. The positive predictive value (has the disease) of MRCP for all biliary pathology was 83.62%. The negative predictive value of MRCP (truly disease free) was 98.10% (Table 1
).
|
).
Subgroup 2: ERCP before MRCP (n=107)
The ERCP was attempted before MRCP in 107 patients and was successful in 55. The median interval between the procedures was 11 days (range 2–71 days). 52 test failures were further investigated with MRCP and subsequently managed either conservatively or underwent operative intervention based on the MRCP results. No comparable data were available for these 52 patients and were excluded from the sensitivity and specificity calculation. The comparative results available from the 55 patients are given in Table 1. There were 76 choledocholithiasis among various positive diagnoses, based on all investigations (Table 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
In spite of different diagnostic modalities available for detecting choledocholithiasis, currently no single method is both risk free and with high sensitivity and specificity. Intraoperative cholangiogram, when used routinely, increases operating time [9] significantly. Disposable equipment adds to the cost of treatment and the procedure offers limited benefit to patients, particularly if the specialist skills and equipment required for laparoscopic duct exploration are unavailable. Hence such imaging is not advocated routinely in laparoscopic cholecystectomy [9, 10].
At present, ERCP is considered the gold standard method for the diagnosis of the ductal calculus, but carries potential risk of complications including pancreatitis, bleeding particularly from sphincterotomy sites and duodenal perforation [11]. The procedure may not be successful with an attendant diagnostic and therapeutic dilemma in subsequent management. The incidence of such complications varies according to operator expertise and experience and once more the frail elderly patients are at greatest risk. In this study ERCP failed in 25% of procedures. As a total of six operators (all consultants with experience) were involved at the two sites studied we believe this is a reasonable reflection of real clinical practice and is within the published limits [12, 13]. This population also included a significant proportion of patients who had undergone previous biliary or gastric surgery which increase the technical difficulties of ERCP. These factors may have contributed to the finding of significant pancreatitis developed in 5.7% of our patients. There were two sub-mucosal injections in this group requiring further cannulations or alternative investigation to confirm the ductal calculi, adding to the cost of the treatment and the morbidity.
MRCP approaches the ideal imaging modality and when used with proper indications, based on clinical suspicion and predictive scoring [6, 11], offers a safer and more acceptable alternative to diagnostic ERCP [14]. Moreover, MRI technology has progressed from the machines that were used in our study. Field strengths are higher; the gradient switching is faster, which makes better image resolution with improved accuracy. The use of MRCP for the initial "screening" of patients at risk of choledocholithiasis permits selection for more invasive procedures based on a high probability of therapeutic intervention. Very few patients are claustrophobic to the machine for the brief period required for a biliary study. A few studies may be of reduced quality because of excess body mass or poor cooperation (movement artefact) that frequently complicates ERCP is generally less problematic. The 4% failure for MRCP in our study compares unclear sentences favourably with 25% for ERCP, and none of the complications seen with ERCP occurred. MRCP carries an additional advantage of diagnosing abnormalities of the biliary tree including duplication, choledochal cyst, pancreas divisum and cholangiocarcinoma. The superior sensitivity of MRCP over ERCP in diagnosing primary sclerosing cholangitis [15, 16] and intrahepatic calculi [17] has been well documented. A disadvantage of MRCP is that currently no therapeutic options are available for the management of common duct calculi. Sugiyama et al reported a sensitivity of 100% from their series of 101 patient [18], but Mendler et al found MRCP less sensitive for smaller stones (less than 3 mm) [19, 20]. Such small stones tend to pass and are of dubious clinical significance.
This study reported the largest consecutive series of patients undergoing MRCP for suspected extrahepatic biliary pathology yet presented. The sensitivity for the diagnosis is extremely high (97.98%). It is also evident that 78 (true negative) in subgroup 1 and 25 (true negative) in subgroup 2 underwent ERCP without therapeutic indication. In a further 51 patients, ERCP was performed unsuccessfully, whereas MRCP in the same group of patients did not show evidence of ductal calculi. This cohort of patients were managed conservatively and recovered with out sequelae. This implies that invasive ERCP could have been potentially avoided in 157 patients (of 283), and in some patients more than once. Avoiding these examinations would have effectively resulted in considerable resource release. However, this reduction in the number of ERCP examinations, particularly diagnostic, will have some impact on training. This can be overcome by the centralization of ERCP service with structured training patterns.
We believe that the small false positive rate for MRCP is comfortably outweighed by the incidence of the hazards of "unnecessary" ERCP where this is confined to patients in whom the probability of operative intervention is high. Clearly, further longitudinal follow-up data must be awaited for patients with negative MRCP results with transient jaundice or pancreatitis in whom a decision is made not to undertake ERCP. Based on the results of the comparative data from two imaging modalities, as has been shown in this study, we believe the likelihood of such problems recurring is low, particularly if cholecystectomy is undertaken. British Society of Gastroenterology guidelines on the management of acute pancreatitis recommends endoscopic sphincterotomy as the definitive management for medically unfit patients presenting with gall stone pancreatitis [21] and following this algorithm ERCP would largely be confined to those with a high probability of therapeutic intervention. The average clinical workload of ERCP in our institution grossly outnumbers MRCP with an average of 430 procedures each year. Based on our findings we have developed a practical algorithm for the management of biliary calculus in different clinical situations (Figure 1).
|
|
Acknowledgments |
|---|
Received for publication November 30, 2004. Revision received April 19, 2005. Accepted for publication May 6, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E J Williams, J Green, I Beckingham, R Parks, D Martin, and M Lombard Guidelines on the management of common bile duct stones (CBDS) Gut, July 1, 2008; 57(7): 1004 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Patel, D. L. Reede, D. S. Katz, R. Subramaniam, and J. K. Amorosa Imaging the Pregnant Patient for Nonobstetric Conditions: Algorithms and Radiation Dose Considerations RadioGraphics, November 1, 2007; 27(6): 1705 - 1722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Schmitt, M. Baugh, and S. Scaglione Practice corner: Chair's rounds Evid. Based Med., October 1, 2007; 12(5): 133 - 134. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
