Endoscopic Evaluation and Management of Pancreaticobiliary Disease

Indications

The role for diagnostic ERCP alone has diminished as other less invasive/noninvasive imaging techniques (e.g., EUS and MRCP) have become increasingly prevalent. ERCP is indicated in clinical settings in which there are significant suspicions of obstructing, inflammatory, or neoplastic pancreaticobiliary lesions that, if detected or ruled out, would alter clinical management. A list of appropriate indications for ERCP and ES is reported in Box 111.1 . ERCP is generally not indicated for the evaluation of abdominal pain of unclear etiology in the absence of objective findings supporting hepatobiliary or pancreatic disease in other laboratory or imaging studies, in suspected gallbladder disease without evidence of bile duct disease, or for further evaluation of proven pancreatic malignancy if management will not change.

Choledocholithiasis

Choledocholithiasis develops in 5% to 10% of patients with symptomatic cholelithiasis undergoing laparoscopic cholecystectomy. The introduction of ES by Classen and Kawai in 1974 initiated a change in the management of common bile duct (CBD) stones. Before that time, laparotomy with open CBD exploration was the main therapeutic recourse for patients with choledocholithiasis. While laparoscopic cholecystectomy has since become the standard, accepted, and preferred technique for the treatment of gallbladder stones (due to less postoperative pain, reduced hospitalization time, shorter convalescence, and better cosmetic results than open cholecystectomy), laparoscopic management of common duct stones is significantly more complex. Advanced surgical skills and sophisticated instrumentation are required, and neither widely available. As a result, ERCP plays an integral role in the treatment of common duct stones in the laparoscopic cholecystectomy era.

The timing and need for ERCP in relation to laparoscopic cholecystectomy are dependent on the likelihood of stones being present (low, medium, and high), the skill of the endoscopist, and the ability of the laparoscopist to perform common duct exploration. There is little value of routine ERCP before laparoscopic cholecystectomy in patients with a low likelihood of having bile duct stones. When comparing the low yield of detecting clinically important anatomic variants and unsuspected bile duct stones with the generally accepted 3% to 7% ERCP complication rate, the routine use of ERCP before cholecystectomy cannot be justified. In patients with symptomatic cholelithiasis, a completely normal liver biochemical panel has a significant clinical utility in excluding the presence of CBD stones with a negative predictive value of more than 97%. Conversely, there is no one clinical variable that can reliably identify the subset of patients with concomitant choledocholithiasis with good positive predictive value. The most reliable predictors include a stone visualized within the CBD on ultrasound; however, the sensitivity of transabdominal ultrasound for detecting a stone is low (22% to 50%). Ascending cholangitis, bilirubin level greater than 1.7 mg/dL, and a dilated CBD on ultrasound are additional high risk/strong likelihood features. Using multiple modality abnormalities, patients can be determined to be low (<10% risk of choledocholithiasis), medium (10% to 50%), or high risk (>50%).

Patients judged to have a high likelihood of harboring duct stones are likely to benefit from preoperative ERCP and stone extraction (if stones are present). Patients in the medium-risk group create a diagnostic and therapeutic dilemma because a failure to identify and remove choledocholithiasis can result in serious complications including recurrent symptoms, cholangitis, and acute gallstone pancreatitis. To lower this risk, additional biliary imaging is warranted, and patients should undergo preoperative EUS or MRCP or intraoperative laparoscopic intraoperative cholangiogram (IOC) or ultrasound; however, these imaging modalities are operator dependent and not universally available ( Fig. 111.1 ). If an intraoperative laparoscopic IOC is positive, the patient should undergo a postoperative ERCP or laparoscopic duct exploration.

Methods of Stone Extraction

Standard (Basket and Balloon Catheters)

After identification of a common duct stone, an ES is typically performed. Balloon catheters are most useful for extracting one or more relatively small or fragmented stones (<10 mm) in a nondilated duct ( Fig. 111.2 ) . They are not as effective for extracting larger stones or small stones in a markedly dilated bile duct because the balloon will often slide past the stone. Stone retrieval wire baskets with different configurations, length/width, types of wire, and number of wires are commercially available. Settings in which a basket may be preferred over a balloon include larger stones (>10 mm), intrahepatic stones, smaller stones in a dilated duct, and stones that are larger than the downstream duct (e.g., stone proximal to a stricture). A risk with a wire basket stone retrieval is the possibility of the wires getting impacted within the stone in the CBD. The basket and stone apparatus are then anchored within the CBD and cannot be easily retrieved. The emergency mechanical lithotripsy may allow a rescue of the impacted basket; however, surgery may be required to remove the basket.

In experienced centers, common duct stones can be successfully removed in 80% to 90% of patients after sphincterotomy with standard baskets and balloon catheters. Difficulty clearing or failure to clear the common duct of stones may occur for a variety of reasons. In most cases, stone size is the major determinant of success. Stones greater than 15 mm are generally considered large; equally important, however, are stone factors (such as number, consistency, shape, and location) and ductal factors (such as contour, diameter at the level of and distal to the stone, and the presence of coexisting pathology such as a stricture or tumor).

Biliary Stents

When stone extraction is incomplete or has failed, biliary drainage should be established to prevent stone impaction and cholangitis. In most situations, this therapy serves as a temporizing measure that allows for improvement in the patient's clinical condition pending repeat attempts at stone removal. The stent is placed so that one limb is above the stone and the other is in the duodenum ( Fig. 111.3 ). Most authorities recommend double-pigtail stents, although favorable experience is reported with straight 10-French (Fr) stents. Biliary stenting not only serves to drain the bile duct but may also aid in mechanically fragmenting the stone and thereby facilitating subsequent attempts at endoscopic removal. The addition of oral dissolution therapy may also soften and reduce the size of the stone and thus aid endoscopic removal.

Acute Gallstone Pancreatitis

In Western countries, gallstone disease is a leading cause of acute pancreatitis and accounts for 34% to 54% of cases. Most patients with acute gallstone pancreatitis (AGP) have a mild attack and can be treated conservatively. However, the case fatality rate in severe pancreatitis remains unacceptably high, approaching 10%. In the open cholecystectomy era, urgent surgical intervention for severe AGP did not gain general acceptance because of the increased morbidity and mortality associated with this approach. Coincident with these surgical reports were uncontrolled endoscopic series reporting the efficacy and safety of ERCP and ES in the setting of AGP. Although the results were encouraging, the studies varied in their criteria for patient selection and timing of ES in relation to the acute attack (many were performed in the recovery phase, when surgery is also safe). These early series prompted the three randomized controlled trials that now serve as the basis for the endoscopic treatment of AGP. The therapeutic principle for ES in AGP is simply removal of the obstructing calculus and reestablishment of bile and pancreatic juice flow.

In a randomized prospective controlled trial from the United Kingdom, 121 patients with AGP either received conventional therapy (i.e., gut rest, analgesics, intravenous fluids, and antibiotics) or underwent urgent (within 72 hours after admission) ERCP with ES and stone extraction (if stones were present in the CBD at the time of ERCP). Patients were stratified by the predicted severity of their attacks with the modified Glasgow system. Choledocholithiasis was found in 25% of patients with predicted mild attacks and 63% with predicted severe attacks. The four important findings were that (1) ERCP could be safely performed in the setting of gallstone pancreatitis, (2) there was a significant reduction in major complications in patients who underwent urgent ERCP and ES, (3) the reduction in morbidity was apparent only in patients with predicted severe attacks (61% vs. 24%; P = .007), and (4) there was a significant reduction in hospital stay for those with severe attacks treated by urgent ERCP and ES (median of 9.5 vs. 17 days; P = .03). The mortality rate was improved, but the difference was not statistically significant.

A second randomized controlled study was performed by the department of surgery at the University of Hong Kong. One hundred ninety-five patients with acute pancreatitis were randomized to early ERCP (within 24 hours of admission) or conservative therapy. Although the methodology, patient selection, and assessment of the severity of the acute pancreatitis used in this study differed from that in the United Kingdom study, the results in the subgroup of patients with gallstone pancreatitis ( n = 127) were quite similar. Patients with mild pancreatitis had similar morbidity and mortality regardless of the therapy. In contrast, patients with predicted severe attacks who underwent endoscopic therapy had a lower complication rate (54% vs. 13%; P = .003) and a lower mortality rate (18% vs. 3%; P = .07) than patients treated conservatively.

The third study was a prospective multicenter randomized controlled study from Germany in which 238 patients with AGP and no evidence of severe biliary obstruction (severe biliary obstruction defined as a bilirubin concentration >5 mg/dL) were randomized to ERCP with ES and stone extraction or conservative therapy within 72 hours of symptom onset. This study attempted to address the major criticism of the United Kingdom and Hong Kong studies: the need to exclude patients with concomitant cholangitis because these patients are known to benefit from ERCP. The two treatment groups did not differ significantly in mortality (11% vs. 6% overall mortality, 8% vs. 4% AGP mortality, ERCP vs. conservative therapy) or overall complications (46% vs. 51%, ERCP vs. conservative therapy) regardless of the predicted severity of the pancreatitis. However, respiratory failure was more frequent in the ERCP group (12% vs. 5%; P = .03), and jaundice was more frequent in patients who received conservative treatment (11% vs. 1%; P = .02).

Although all three studies concluded that there was no difference in outcomes for patients with mild pancreatitis treated conservatively or by ERCP, only the study from Germany suggested that early ERCP was of no benefit in patients with severe gallstone pancreatitis. Even though ERCP is clearly indicated in patients with AGP complicated by cholangitis or biliary obstruction, its role in the setting of severe AGP alone warrants further investigation. A meta-analysis of these three published studies revealed a statistically significant reduction in morbidity from 38% to 25% and mortality from 9% to 5% in the ERCP/ES group versus the conservatively treated group. A subgroup analysis based on the severity of pancreatitis was not reported in this meta-analysis.

Acute Cholangitis

Cholangitis is a potentially life-threatening disease that results from bacterial infection of obstructed bile. Systemic toxicity occurs when intraductal pressure is sufficiently elevated to cause reflux of bacteria or endotoxin into blood. Thus obstruction plays a key role by both increasing intraductal pressure and promoting bacterial overgrowth as a result of bile stasis. The most common cause of acute cholangitis is choledocholithiasis, which occurs in approximately 80% to 90% of unselected cases. Therapy for cholangitis must be individualized because of the spectrum of severity of illness. Antibiotic therapy should be initiated promptly. Analysis of bile and stone cultures indicates that Escherichia coli , Klebsiella spp., Enterobacter spp., Enterococcus spp., and Streptococcus spp. are the most commonly isolated bacteria. The antibiotic selected should preferably penetrate an obstructed biliary tree. The majority of patients will respond to conservative management, thereby allowing for a more elective approach to biliary decompression. Urgent decompression is indicated if improvement is not seen within a few hours of initial resuscitation. The latter group will invariably have a fatal outcome if conservative treatment is continued.

Options for bile duct decompression include surgical, percutaneous, and endoscopic methods. Endoscopic intervention is now accepted as definitive therapy for acute cholangitis. The advantages of ERCP are that it can delineate the cause of obstruction, facilitate sampling of bile for culture, and decompress the biliary tree in a relatively short time with low morbidity. Biliary decompression is the goal of therapy and can be complete (e.g., stone removal) or temporary (e.g., placement of a stent without stone removal), pending more definitive management (to allow stabilization of an unstable patient). The endoscopic procedure consists of sphincterotomy with stone extraction or biliary drainage with a stent.

Ideally the patient should be stabilized or made as stable as possible before performing ERCP. Patients with respiratory compromise can have their ERCP performed while on ventilatory assistance. Because intrabiliary pressure is increased in acute cholangitis, contrast injection should be limited to reduce further systemic seeding of bacteria. Enough contrast should be injected to define the anatomy and the cause of obstruction. Aspirated bile should be cultured. In a stable patient, definitive therapy can be performed. In an unstable patient, the length of the procedure should be limited. In such cases a stent should be placed, and once the patient is stabilized, more definitive therapy can be performed.

The high morbidity and mortality associated with surgical and percutaneous therapy for acute cholangitis prompted evaluation of the safety and utility of endoscopic management. In a retrospective analysis, Leese et al. reported on 71 patients with stone-related cholangitis treated by early decompression either surgically ( n = 28) or by ES ( n = 43). Early surgery was associated with significantly higher 30-day mortality (21% vs. 5%) and morbidity (57% vs. 8%) than sphincterotomy. The endoscopic group was significantly older than the surgical group and had more medical risk factors, but there were no significant differences in the severity of cholangitis. Leung et al. reported their experience in a retrospective analysis of 105 patients with acute calculous cholangitis who did not respond to conservative management and underwent urgent endoscopic decompression at a mean of 1.5 days after admission. Of these patients, 39% had coexisting medical problems, 85% had Charcot triad, and 40% were in shock at the time of admission. Endoscopic drainage was successful in 102 patients (97%). Ninety-seven percent of patients responded with striking improvement in abdominal pain, and 93% had resolution of fever within 3 days. The overall 30-day mortality was 5%. Among those in shock, 2 of 4 who had been drained after 72 hours died, as compared with 3 of 38 who had been drained before 72 hours. There were no deaths in the group without shock, irrespective of the timing of drainage. The mortality of 5% compares favorably with that of urgent surgical intervention, in which mortality has been reported to be greater than 40% in some series. The ERCP complication rate was 5% and was limited to five postsphincterotomy bleeding episodes managed by endoscopic techniques. The safety and efficacy of endoscopic therapy were corroborated in a large retrospective study of 947 patients with cholangitis secondary to stones ( n = 898) or stricture ( n = 49). In a randomized prospective study, Lai et al. compared the safety and efficacy of biliary decompression by surgical and endoscopic techniques in 82 patients with severe cholangitis as a result of stones. Patients treated with laparotomy and CBD exploration had significantly higher morbidity (64% vs. 34%) and mortality (32% vs. 10%) than those treated with endoscopic therapy. These and other studies clearly demonstrate the efficacy and safety of biliary decompression either as definitive therapy or as a temporizing measure, pending more definitive intervention once the patient is stabilized.

Benign Biliary Strictures

Endoscopic therapy for the management of benign biliary strictures has emerged as an effective and safe treatment with significantly less morbidity and mortality than the traditional surgical and percutaneous management strategies. Benign biliary strictures amenable to endoscopic intervention develop most frequently secondary to inflammatory conditions, such as chronic pancreatitis or as postoperative complications. Additional less frequent causes are primary sclerosing cholangitis, infection, ischemia (e.g., portal biliopathy), papillary stenosis, autoimmune pancreatitis/cholangiopathy, trauma, and bile duct stones. Regardless of etiology, benign biliary strictures necessitate therapeutic intervention in the presence of jaundice, chronic cholestasis, or cholangitis, to reduce the risk of developing secondary biliary cirrhosis.