Cholangioscopy and Pancreatoscopy

Cholangioscopy and Pancreatoscopy

Amrita Sethi1 and Raj J. Shah2

1 Columbia University Irving Medical Center‐NYP, New York, NY, USA

2 University of Colorado Anschutz Medical Campus, Aurora, CO, USA


The prospect of visualizing the biliary tree and pancreatic duct during ERCP has allured gastroenterologists for decades [1]. Significant mechanical challenges have resulted in a design evolution of both rigid and flexible endoscopes employing fiber optic, video, and digital technology. The nomenclature has likewise evolved through a spectrum of descriptive terms, including cholangioscopy, choledochoscopy, duodenoscope‐assisted cholangiopancreatoscopy, peroral cholangioscopy (POCS), and peroral pancreatoscopy (POPS). While cholangioscopy can be performed percutaneously through a transhepatic route or intraoperatively through the cystic duct, the following discussion will focus on the transpapillary approach that is performed in conjunction with ERCP. The transpapillary route has evolved from the use of a two‐endoscope system (mother–daughter scope) to now a catheter‐based cholangioscope that is used as an accessory through the duodenoscope, and finally even direct per‐oral access with ultra‐slim scopes.

This chapter includes techniques of cholangioscopy and pancreatoscopy, overview of the diagnostic and therapeutic indications of cholangiopancreatoscopy, and potential adverse events. It hopes to offer trainees and their instructors guidance on technical and cognitive prerequisites, and options for gauging competency. Since formal training guidelines for cholangioscopy and pancreatoscopy are currently being developed, most of the recommendations about training are based upon expert opinion.

Technique of cholangiopancreatoscopy

Previous discussions regarding technique were focused on dual provider technique utilizing “mother–daughter” endoscope‐based systems. With the advent of catheter‐based cholangioscopes, regional variation determines its performance by single or dual operators. The configurations, technical features, and design of catheter‐based systems warrant an emphasis on the option for single‐operator technique especially in western practices. In countries where reuse of the catheter‐based system occurs, it is felt that dual operator technique may lead to more durability of the steering dials. For the novice, it is helpful to think of the cholangioscope as having a similar platform as standard endoscopes with four‐way tip deflection control dials, a camera and light, imaging chip, working/accessory channel, and dedicated irrigation ports (see Figure 1).

Currently, in the United States, there is one commercially available digital (e.g., CMOS chip) catheter‐based system (SPYDS2, Boston Scientific, Inc., Marlborough, MA) with other major endoscopic accessory companies planning the development and launch of additional catheter‐based digital systems.

c11i001Cholangioscopy has transitioned from a dual‐operator technique, where one physician manages the duodenoscope and a second provider controls the cholangioscope, to single operator technique by attaching the control knob apparatus to the duodenoscope. The endoscopist’s left hand is used to control the duodenoscope and the right hand is used to manage the controls of the cholangioscope (see Figure 11.1). (Video 11.1). Thus, movements of the cholangioscope, those of the duodenoscope, and with experience, maneuvers such as advancing the cholangioscope proximally in the duct or centering the tip of the cholangioscope in the duct lumen can be performed by manipulating the duodenoscope. It has two dedicated irrigation channels which facilitates intraductal visualization and the performance of cholangioscopy‐guided lithotripsy. A working channel is utilized for suction and passage of accessories. There are a number of accessories available that can be passed through the channel of the Spy DS systems: a dedicated biopsy forcep (SpyBite Forceps) that has an open jaw diameter or 1 mm; a snare retrieval basket that opens to 15 mm, and a snare that has a maximum diameter of 9 mm.

In most cases of cholangioscopy, the presence of a biliary sphincterotomy facilitates ductal intubation over a wire or freehand and a 260 cm or longer guidewire can be used if the wire can be sufficiently locked at the elevator during cholangioscope passage. Cholangioscopy has also been successfully performed following balloon ampullary dilation without papillotomy [2] but that is not our typical practice. Either way, enlargement of the biliary orifice facilitates intubation decreasing papillary trauma, facilitates drainage of irrigant, and allows improved maneuverability within the duct.

Photo depicts close-up of SpyScopeDS2TM as it is attached to the duodenoscope.

Figure 11.1 Close‐up of SpyScopeDS2TM as it is attached to the duodenoscope.

c11i001During intubation, slight manipulations of the tip of the cholangioscope can be made to allow for further progression proximally in the duct. We advise minimizing use of the duodenoscope elevator during cholangioscope passage to reduce tip damage. Further, if a ‘short’ wire is utilized, the cholangioscope is advanced over the wire that can be ‘floated’ while locking the wire on the elevator; once the wire exits the working channel, we advise light visualization so the elevator can be visualized during cholangioscope passage to further minimize catheter tip trauma (Video 11.2). Once an adequate position is achieved, the wire can be removed to allow for fluid aspiration and accessory passage. Circumferential ductal inspection is optimized by the use of the steering controls providing four‐way tip deflection and duodenoscope movements that may entail ‘short’ and ‘semi‐long’ positions and control dial manipulation. A variety of accessories specific to cholangiopancreatoscopy are available, including diagnostic devices such as confocal probes, miniature biopsy forceps, and therapeutic devices such as a snare, basket, and electrohydraulic or laser lithotripsy (LL) probes.

Indications for cholangioscopy

The primary indications for cholangioscopy include the evaluation and management of biliary strictures, filling defects, and difficult choledocholithiasis (Table 11.1). Indications have broadened to include selective ductal cannulation, mapping of intraductal lesions such as cholangiocarcinoma or main duct IPMN, and retrieval of foreign bodies (e.g., migrated stents). [3] The most established indication for cholangioscopy is in the management of difficult bile duct stones with electrohydraulic and LL. While cholangiopancreatography identifies well‐defined bile duct abnormalities such as choledocholithiasis and bile leaks, cholangiopancreatoscopy may be useful in defining filling defects and strictures to better assist in distinguishing benign from malignant lesions. Further, fluoroscopy‐guided sampling has been limited by poor diagnostic accuracy. It is not surprising that by enabling direct visualization of the abnormalities and characterization with imaging features, as well as allowing for targeted tissue acquisition, cholangioscopy serves a vital role in the management and diagnosis of these problematic areas. Other less frequent diagnostic and therapeutic indications will be discussed as well.

Table 11.1 Indications for cholangioscopy.

ERCP diagnosis Cholangioscopy offers
Strictures Improved diagnostic accuracy tissue
acquisition under direct visualization, selective guidewire access, and image‐guided ablative therapies
Filling defects Improved diagnostic accuracy and definitive therapy
Stones, refractory Visual guidance for EHL or LL and can confirm duct clearance
Mucosal lesions Biliary or pancreatic IPMN, vessel characterization, stent‐associated changes

Cholangioscopy for the characterization of biliary lesions and strictures

A filling defect relates to the fluoroscopic appearance of something that actually lies within the bile duct lumen such as a stone, extrinsic compression, or a polypoid tumor. It has now been clearly established that cholangioscopy can improve the accuracy in the diagnosis of biliary filling defects [46]. In an early experience, cholangioscopy was utilized when there was uncertainty in the diagnosis by cholangiopancreatography alone [7]. Patients were excluded if they had obvious stone disease or classic biliary obstruction due to malignancy in the head of the pancreas. In this study, 91 consecutive patients were evaluated by ERCP supplemented by biopsy/brush cytology when indicated. There were 76 strictures and 21 filling defects in the study group. Of the patients with the 21 filling defects, ERCP with biopsy or brush cytology was able to correctly identify the eight malignant lesions and the nine benign tumors. ERCP did not, however, correctly diagnose the four cases of stone disease. In these patients, the stones were adherent to the bile duct wall and had the appearance of a mass. Cholangioscopy, on the other hand, was able to make the correct diagnosis using direct visualization alone in all 21 patients. The four patients with stone disease were easily identified and treated with stone removal.

NBI is now widely used in distinguishing Barrett’s mucosa, flat adenomas, or other subtle mucosal lesions. Video cholangioscopy with NBI (CHF‐B260, Olympus Tokyo; outer diameter of 3.4 mm and a working channel diameter of 1.2 mm) is not currently commercially available but there are several publications on its utility. The NBI system (CV260SL processor, CVL‐260SL light source; Olympus) is based on narrowing the bandwidth of spectral transmittance of red–green–blue optical filters. The filters cut all illumination wavelengths, except two narrow wavelengths centered on 415 nm and 540 nm. The image is reproduced in the processor with information from the two illumination bands. The 415‐nm centered band provides the most information on the capillary and pit patterns of the superficial mucosa and the 540‐nm centered band provides information about thicker capillaries in slightly deeper tissues. Thus, NBI delivers accentuated images of the mucosal structures and mucosal microvessels [8].

Biliary stricture

A stricture implies a narrowing of the duct lumen due to (1) wall thickening, (2) compression from extrinsic pathology, or (3) adherent or impacted intraductal stone with a limited column of contrast seen adjacent to it. Stricture due to increased wall thickening can be intrinsic to the wall such as with a cholangiocarcinoma and might therefore have some associated mucosal defects that could be detected by direct visualization. Stricture due to compression from extrinsic disease tends to be asymmetric and with smooth margins, such as nodal metastasis. In some cases, cholangiocarcinoma can spread through the biliary tree in the submucosal layers, resulting in biliary stricture by compression; however, the overlying mucosa may have a completely unremarkable endoscopic appearance.

Differentiating malignant from benign biliary stricture

While the cholangioscopic differentiation between stone and tissue appears straightforward, the same is not true in the ability to differentiate malignant from benign strictures on the basis of direct visualization alone. Early cholangioscopic experience reported morphologic characteristics that claimed to distinguish malignant from benign tissue with an accuracy that approached 95% [9]. Several features were identified as being accurate predictors of malignancy, including tumor neovascularization, a dense papillary pattern, and friable nodularity. It was observed that bile duct adenocarcinomas had three classifications: nodular, papillary, and infiltrative.

  1. Nodular‐type lesions are bulky and eccentric with an overlying friable mucosa. There is often neovascularization with tissue friability and oozing.
  2. c11i001Papillary type is identified by a high‐density pattern of papilla or fish‐egg‐type mucosa. There is often luminal mucin and blood obscuring the visual field (see Video 11.3).
  3. Infiltrative type is the most difficult to diagnose because of the paucity of specific cholangioscopic characteristics. The overlying mucosa is bland and whitish with or without neovascularization.

The authors described several other less common bile duct tumors, including biliary papillomatosis, mucin‐hypersecreting cholangiocarcinoma, and biliary cyst adenocarcinoma. Biliary papillomatosis looks similar to the papillary adenocarcinoma except that it is a multifocal disease with areas of normal intervening mucosa. The mucin‐hypersecreting cholangiocarcinoma is similar to the papillary‐type adenocarcinoma except that according to the authors, the biliary ducts may be dilated and mucin filled.

The University of Colorado group assessed the utility of video NBI cholangioscopy in 96 patients. Tortuous and dilated vessels (p < 0.001), infiltrative stricture (p < 0.001), polypoid mass (p = 0.003), and the presence of fish‐egg lesions (p = 0.04) were found to be significantly associated with neoplasia. The overall POVCP impression had a high sensitivity (85%) and negative predictive value (89%) in assessing for the presence of neoplasia. Requiring at least two of these findings to be present for a cholangioscopic impression of neoplasia yielded a lower sensitivity of 48% (95% CI: 31–64%) but with an impressive increase in specificity of 96% (95% CI: 91–100%).[10]

These cholangioscopic criteria for malignancy were tested prospectively on 76 patients with biliary strictures of unknown type and compared to the accuracy of ERCP with biopsy alone. In this study, ERCP with tissue sampling had a sensitivity of 58% and a specificity of 100%. The addition of cholangioscopy to this group of patients did increase the sensitivity to 100% but dropped the specificity to 87%. The loss of specificity was due to the incorrect diagnosis of malignancy in five patients on the basis of the cholangioscopic appearance of neovascularization and the presence of a tumor vessel. These five false positives were found instead to have chronic pancreatitis in two patients and one patient each with primary sclerosing cholangitis, autoimmune pancreatitis, and a peribiliary cyst.[7]

Attempt at standardizing criteria to predict malignancy has varied regionally and a lack of consensus has limited incorporation. One center had a two‐stage protocol. In Stage I, a retrospective study that included 315 images were correlated to histology and classified as non‐neoplastic or neoplastic based on morphological and vascular patterns. In Stage II, a prospective, nonrandomized, double‐blind validation study was performed. 171 patients were included (65 retrospective; 106 prospective). Operating characteristics in Stage II showed sensitivity, specificity, PPV, and NPV for neoplastic diagnosis as 96.3%, 92.3%, 92.9%, and 96%, respectively. The proposed classification presented high reproducibility among observers, for both neoplastic and subtype categories. However, it was better for experts (κ > 80 %) than non‐experts (κ 64.7–81.9%) [11]. Duplicating these promising findings at other centers is necessary.

The potential advantage of cholangioscopy in the evaluation of biliary strictures for malignancy is in cholangioscopy‐directed tissue sampling. Initial reports indicate a high sensitivity for the identification of malignancy in patients with suspicious‐appearing biliary strictures. In the first study of consecutive patients undergoing directed sampling with prototype miniature forceps, 62 patients with indeterminate biliary strictures, cholangioscopy with and without biopsy detected malignancy with a sensitivity of 89%, specificity of 96%, positive predictive value of 89%, and negative predictive value of 96%.[12] However, in patients with primary sclerosing cholangitis, the yield to detect malignancy has been disappointing thus far. [1214] For unclear reasons, the yield of cholangioscopically visualized biopsies has improved and in one recent study it was suggested that in centers with off‐site pathology, 3 biopsies can yield a diagnostic accuracy of 90%.[15] There are several reasons to account for difficulties in accurate tissue diagnosis of suspicious‐appearing strictures. Cholangiocarcinoma, especially in the Western world, tends to have associated sclerotic changes and this along with tissue depth achieved with miniature forceps tends to provide superficial samples. Further, negotiating angulations and smaller biliary radicals remain a challenge, as is maximal tip articulation when a forceps is within the accessory channel to perform three or four quadrant sampling.

Cholangioscopy‐guided stone therapy

Fragmentation of giant, impacted, or recalcitrant stones is one of the primary indications for interventional cholangioscopy. Difficult bile duct stones have been described as those that are >1 cm, are located proximal to strictures or non‐dilated distal ducts, are embedded in the biliary wall or at the cystic duct orifice, or are intrahepatic. These types of stones often fail conventional stone retrieval methods such as retrieval balloons or mechanical lithotripsy baskets. In these circumstances, fragmentation using electrohydraulic lithotripsy (EHL) or LL is an efficient and highly successful technique. A through‐the‐cholangioscope basket is also now available (BSC). Due to the potential for bile duct wall damage and the precision required to maximize efficacy of these lithotripsy methods, cholangioscopic guidance is felt to be the suggested method of delivery as described below.

EHL was originally designed as an industrial mining tool. The modification for endoscopy employs a fiber with two insulated, coaxial electrodes. A power generator delivers a high‐voltage electrical current, creating a spark across the two electrodes at the tip of the fiber. High‐frequency discharges cause rapid expansion of the fluid–stone interface, generating shock waves that fragment the stone. This technique has been successfully applied using percutaneous, surgical, and transampullary routes to the bile duct using either cholangioscopic or fluoroscopic guidance.

c11i001To perform EHL, the cholangioscope must first be positioned about 1 mm in front of the target stone and the contact interface must be in an aqueous environment for the shock wave to be transmitted and effect fragmentation (see Video 11.4). For denser stones, actual contact may be necessary to weaken the stone by ‘boring’ a hole first followed by ‘painting’ the area to achieve fragmentation. Periodically, fiber position should be verified to avoid mucosal and cholangioscope trip trauma and sterile saline irrigation performed to maintain visualization. Lithotripsy should be pursued until fragments are significantly less than the diameter of the downstream duct to ease clearance.

Cholangioscopy‐guided LL can also be performed, most commonly using the holmium laser. Targeting of the laser probe at the center of the stone causes central implosion of the stone as opposed to fragmentation with EHL (Video 11.5). Similar concerns regarding potential injury to the wall is noted but the safety profile has been assessed in a preliminary study utilizing a porcine model and revealed that up to 5 seconds of continuous pulse on mucosa may be safe without transmural injury. [16] In any event, laser should be performed with direct visualization and care to avoid mucosal trauma be taken, especially when power is increased for recalcitrant stones.

The success of EHL and LL has been reported in a number of early series using the fiber‐optic SOC system and has ranged between 66 and 92% [1721] depending on the presence of intrahepatic stones. Adverse events in these series ranged from 4.8 to 18%. More recent studies using the newer digital cholangioscope with >100 patients report 91.1–97.3% success rates in stone clearance. [22, 23] In retrospective comparison, the two methods have similar efficacy and safety but LL is associated with a shorter procedure time. [22]. In two recent prospectively randomized trials, cholangioscopy‐guided lithotripsy was found to be more efficacious at stone clearance than mechanical lithotripsy as well as large balloon papillary dilation, however, with longer procedure times. [24, 25] It should be noted that cholangioscopy is actually required in only a small percentage of biliary stones. In a large single referral center experience where a majority of cases were referred for failed clearance using conventional methods, stone clearance efficacy for extrahepatic biliary stones approached 100% utilizing advanced techniques of papillary dilation and/or mechanical lithotripsy, with the use of cholangioscopy required in 40%. [26] Further prospective comparison trials are needed to help develop an algorithmic approach to the management of difficult biliary stones and how early cholangioscopy may play a role that allows for maximal efficacy, safety, and cost.

Cholangioscopy‐guided selective cannulation

Direct visualization of the bile duct has been described to facilitate cystic duct access, selective intrahepatic duct cannulation, and in cases of failed attempts to traverse malignant or benign strictures [27, 28]. For these cases of cholangioscopy‐guided selective cannulation, a long wire is encouraged, but not required, to ensure that full exchange of the cholangioscope can occur without risk of losing access.

Indications for cholangioscopy without fluoroscopy

Occasionally, biliary exploration is indicated when fluoroscopy is either unavailable or if radiation use is to be minimized. The ability to examine the bile duct without fluoroscopy guidance has proven useful in two clinical situations: (1) in patients during their first trimester of pregnancy and (2) in those patients too critically ill to be transported to a fluoroscopy unit. Recent interest in performing cholangioscopy without the use of fluoroscopy for noncomplex biliary stones has also been published potentially offering options when there is limited availability of fluoroscopy in more straightforward cases.

Only gold members can continue reading. Log In or Register to continue

Jul 31, 2022 | Posted by in GASTOINESTINAL SURGERY | Comments Off on Cholangioscopy and Pancreatoscopy
Premium Wordpress Themes by UFO Themes