Abstract
BACKGROUND AND AIMS: Current methods to diagnose malignant biliary strictures are of low sensitivity. Confocal endomicroscopy is a new approach that may improve the diagnosis of indeterminate biliary strictures. The purpose of this study was to evaluate indeterminate biliary strictures using probe-based confocal laser endomicroscopy and to understand the histologic basis for the confocal images. METHODS: Fourteen patients with indeterminate biliary strictures underwent endoscopic retrograde cholangiopancreatography with examination of their common bile duct with fluorescein-aided probe-based confocal laser endomicroscopy. Standard brushings and biopsies were performed. In parallel, rat bile ducts were examined either with conventional staining and light microscopy or with multiphoton microscopy. RESULTS: Earlier published criteria were used to evaluate possible malignancy in the confocal images obtained in the 14 patients. None of the individual criteria were found to be specific enough for malignancy, but a normal-appearing reticular pattern without other putative markers of malignancy was observed in all normal patients. Multiphoton reconstructions of intact rat bile ducts revealed that the reticular pattern seen in normal tissue was in the same focal plane but was smaller than blood vessels. Special stains identified the smaller structures in this network as lymphatics. CONCLUSIONS: Our limited series suggests that a negative confocal imaging study of the biliary tree can be used to rule out carcinoma, but there are frequent false positives using individual earlier published criteria. An abnormal reticular network, which may reflect changes in lymphatics, was never seen in benign strictures. Better correlation with known histologic structures may lead to improved accuracy of diagnoses.
🔬 Techniques
🔭 Microscopes
🧪 Sample Preparation
🏭 Microscope Brands
🧪 Reagent Suppliers
🔴 Lasers
🏛️ Research Organizations (ROR)
Affiliated research institutions:
📋 Methods
Fourteen patients with indeterminate biliary strictures underwent endoscopic retrograde cholangiopancreatography with examination of their common bile duct with fluorescein-aided probe-based confocal laser endomicroscopy. Standard brushings and biopsies were performed. In parallel, rat bile ducts were examined either with conventional staining and light microscopy or with multiphoton microscopy.
MATERIALS AND METHODS Patients
Fourteen consecutive patients referred to the Gastrointestinal Procedure
Center at our institution for ERCP to rule out biliary malignancy were recruited to participate in this study, which was approved by the Yale University Human Investigations Committee. Written informed consent was obtained before study participation. Inclusion criteria included age above 18 years and the ability to provide informed consent. Exclusion criteria included having a known allergy or a prior adverse reaction to fluorescent contrast agents or chromoendoscopy stains; below the age of 18 years or mentally or legally incapacitated or otherwise unable to give informed consent; are pregnant or breastfeeding; coagulopathy; and impaired renal function.
Confocal Imaging
A confocal miniprobe (Cellvizio, Mauna Kea Technologies, Paris, France) was used to examine the biliary mucosa in all cases. All 14 patients undergoing ERCP received up to 10 mL of 0.5% to 10% intravenous fluorescein approximately 2 to 3 minutes before imaging. Biliary strictures and bile ducts were imaged using either a CholangioFlex confocal miniprobe in most patients or a GastroFlex ultra-high definition (UHD) high-definition confocal miniprobe. The CholangioFlex confocal miniprobe is compatible with a 1.2 mm operating channel, has a lateral resolution of 3.5 μm, a depth of focus of 40 to 70 μm, and a field of view of 325 μm. The GastroFlex UHD confocal miniprobe was used in 1 patient. It is compatible with a 2.8 mm working channel, has a lateral resolution of 1 μm, a depth of focus of 55 to 65 μm, and field of view of 240 μm. The CholangioFlex miniprobe was inserted into the common bile duct (CBD) through a Spyglass (Boston Scientific, Natick, MA) catheter or Olympus swing tip catheter. The GastroFlex UHD miniprobe was inserted into the CBD using a freehand technique. The bile duct was cannulated with a standard ERCP catheter which was removed and a 0.035-inch guide wire left in place. The UHD probe was then inserted adjacent to the guidewire.
Show full methods section
Fourteen patients with indeterminate biliary strictures underwent endoscopic retrograde cholangiopancreatography with examination of their common bile duct with fluorescein-aided probe-based confocal laser endomicroscopy. Standard brushings and biopsies were performed. In parallel, rat bile ducts were examined either with conventional staining and light microscopy or with multiphoton microscopy.
MATERIALS AND METHODS Patients
Fourteen consecutive patients referred to the Gastrointestinal Procedure
Center at our institution for ERCP to rule out biliary malignancy were recruited to participate in this study, which was approved by the Yale University Human Investigations Committee. Written informed consent was obtained before study participation. Inclusion criteria included age above 18 years and the ability to provide informed consent. Exclusion criteria included having a known allergy or a prior adverse reaction to fluorescent contrast agents or chromoendoscopy stains; below the age of 18 years or mentally or legally incapacitated or otherwise unable to give informed consent; are pregnant or breastfeeding; coagulopathy; and impaired renal function.
Confocal Imaging
A confocal miniprobe (Cellvizio, Mauna Kea Technologies, Paris, France) was used to examine the biliary mucosa in all cases. All 14 patients undergoing ERCP received up to 10 mL of 0.5% to 10% intravenous fluorescein approximately 2 to 3 minutes before imaging. Biliary strictures and bile ducts were imaged using either a CholangioFlex confocal miniprobe in most patients or a GastroFlex ultra-high definition (UHD) high-definition confocal miniprobe. The CholangioFlex confocal miniprobe is compatible with a 1.2 mm operating channel, has a lateral resolution of 3.5 μm, a depth of focus of 40 to 70 μm, and a field of view of 325 μm. The GastroFlex UHD confocal miniprobe was used in 1 patient. It is compatible with a 2.8 mm working channel, has a lateral resolution of 1 μm, a depth of focus of 55 to 65 μm, and field of view of 240 μm. The CholangioFlex miniprobe was inserted into the common bile duct (CBD) through a Spyglass (Boston Scientific, Natick, MA) catheter or Olympus swing tip catheter. The GastroFlex UHD miniprobe was inserted into the CBD using a freehand technique. The bile duct was cannulated with a standard ERCP catheter which was removed and a 0.035-inch guide wire left in place. The UHD probe was then inserted adjacent to the guidewire.
Histology
Standard biopsy forceps and brushings were used to obtain cytology and tissue from the biliary tree during ERCP when clinically indicated. Biopsy specimens were collected in 10% buffered formalin (Val Tech Diagnostics, PA), were routinely processed, paraffin embedded, and sectioned at 3 μm. They were stained with hematoxylin and eosin (H&E) and examined using standard bright field (white light) microscopy.
Multiphoton Studies of Bile Duct Histology
Fresh, unfixed-rat bile ducts were examined with a multiphoton microscope to correlate confocal images with the corresponding histologic structure. Studies were approved by the Yale Animal Care and Use Committee. Adult male Sprague Dawley rats were anesthetized, subject to laparotomy to expose the inferior vena cava and CBD, and then injected with 1% fluorescein (1 mL/kg) in the inferior vena cava. Three minutes later, the CBD was removed and transferred to a dish containing cold saline and cut open lengthwise to expose the luminal surface, then the animal was euthanized. The duct was then placed in an incubation chamber fitted on the stage of a Zeiss LSM 510 NLO confocal/multiphoton microscope (Thornwood, NY). A mode-locked, femtosecond-pulsed Mai Tai laser (Spectra Physics, Mountain View, CA) tuned to 735 nm was used as a light source, and custom-built nondescanned (external) detectors were used to collect emission images at 320 to 380 nm, 410 to 490 nm, and 510 to 650 nm as described earlier. 15 Immunohistology of Bile Ducts In correlative separate studies of rat bile ducts, whole CBDs were collected in 10% buffered formalin, and then processed, paraffin embedded, and sectioned at 3 μm intervals. Sections were pretreated with 0.01 mol/L citrate buffer, and then endogenous peroxidase activity was blocked by 3% H 2 O 2 in absolute methanol. Each section was preincubated in rabbit serum for 10 minutes, then incubated with lymphatic tissue-specific mouse monoclonal antibody D2-40 (DAKO #M3619; Carpinteria, CA) diluted 1:500 for 60 minutes. Subsequently, sections were rinsed in phosphate-buffered saline, diaminobenzidine was used as a chromogen, and sections were counterstained with H&E and examined using standard bright field (white light) microscopy.
📊 Figures
FIGURE 1
Reticular pattern (arrows) in a normal bile duct. This pattern was observed by confocal imaging using the Cholangio-Flex confocal miniprobe.
FIGURE 2
An example of irregular epithelium. The epithelium has lost its normal structure of regular, adjacent cells. This pattern was observed by confocal imaging using the CholangioFlex confocal miniprobe.
FIGURE 3
Normal bile duct epithelium observed by confocal imaging using the CholangioFlex confocal miniprobe.
FIGURE 4
Normal bile duct observed by confocal imaging using the GastroFlex UHD high-definition confocal miniprobe. Note that blood vessels (arrows) can be identified as white (fluorescein-filled) structures t...
FIGURE 5
Serial optical sections of rat common bile duct examined by multiphoton microscopy. The duct was examined in a freshly isolated, unfixed, and unstained duct. The image is pseudocolored to reveal emiss...
Figure images are served from the NIH/NLM PubMed Central Open Access Subset or Europe PMC; copyright remains with the publishers and authors.
💬 Discussion
0 commentsNo comments yet. Be the first to start a discussion!
Leave a Comment