Case series
Cone-beam computed tomography – adding a new dimension to ERCP
Authors
Jochen Weigt1, Maciej Pech2, Arne Kandulski1, Peter Malfertheiner1
Institutions
1
submitted 11. November 2014 accepted after revision 31. December 2014
Six patients with biliary duct pathologies were investigated with three-dimensional endoscopic retrograde cholangiopancreatography (3D-ERCP). The technique of 3D-ERCP consists of c-arm rotation using a flat detector c-arm to acquire a complete set of images for 3 D visualization of the biliary system. Our case series demonstrates a
high level of accuracy with regard to the diagnosis of stenosis and biliary leakage. In two patients the diagnosis obtained from conventional ERCP changed significantly after 3D-ERCP. From these early observations we conclude that 3D-ERCP enhances the diagnostic performance of ERCP in a subset of patients with biliary tract disease.
Introduction
Patients and methods
!
!
Endoscopic retrograde cholangiopancreatography (ERCP) has a high diagnostic and therapeutic impact and is the gold standard for image-guided interventions of the biliary tract [1]. ERCP for diagnostic purposes has been replaced by magnetic resonance cholangiopancreatography (MRCP), which provides a clear three-dimensional (3 D) overview of the anatomy of the biliary tract. Previous studies have shown the benefit of 3 D imaging to guide therapeutic ERCP [1]. Newer fluoroscopy units are equipped with digital flat detectors which can be used to scan a 3 D set of images. This is known as cone-beam computed tomography (CBCT), flat-detector CT, or c-arm CT. With this modality, 3 D rendering is possible for better 3 D visualization along with multiplanar or curved reconstructions for additional soft tissue components. CBCT has been reported previously for percutaneous biliary interventions but never for ERCP [2 – 4]. It has been demonstrated that CBCT facilitates percutaneous biliary drainage or abscess drainage [2]. Moreover, CBCT can potentially lower the total radiation exposure because of improved planning and guidance of interventions. We describe the first clinical application of 3DERCP to facilitate endoscopic biliary interventions in a series of six patients.
This is a retrospective, observational, open-label, single-arm, consecutive case series conducted at a tertiary-care endoscopy center. We defined 3DERCP as the use of digital detectors to create a 3 D set of images during ERCP, including by 3D-image rendering. This technique was used for patients undergoing ERCP who had inconclusive anatomy of the biliary tract, leakage or strictures. All of the patients were sedated and investigated in the prone position. After the bile ducts had first been opacified with nondiluted contrast medium (Peritrast 60 %; Dr. Franz Köhler Chemie, Bernsheim, Germany) using standard catheters, papillotomes, or an inflated stone-retrieval balloon, the area of interest was centered in the fluoroscopy c-arm (Phillips Eleva Multi Purpose System; Philips Healthcare, Hamburg, Germany). Once in position, the c-arm was rotated 180° around the patient in a period of 6 seconds acquiring 15 images per second, so producing 90 images. Software on a dedicated 3 D workstation was used to render the images (Multi Diagnost Eleva 3D-RX-Option; Phillips Healthcare), a process taking approximately 2 minutes. The ERCP was then continued using the 3 D images obtained. The data were used to guide the procedures in different ways. First, the rotational scan was reviewed providing an overview of the 3 D positioning of the anatomic structures. Second, the biliary structures were rendered to a 3 D image similar to those seen during MRCP. Third, the full set of images was handled in a similar manner to a CT scan
Bibliography DOI http://dx.doi.org/ 10.1055/s-0034-1391483 Published online: 12.2.2015 Endoscopy 2015; 47: 654–657 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0013-726X Corresponding author Jochen Weigt, MD Dep. Gastroenterology, Hepatology and Infectious Diseases Otto-v.-Guericke University of Magdeburg Leipziger Str. 44 39120 Magdeburg Germany Fax: +49-391-6713105 [email protected]
Department of Gastroenterology, Hepatology and Infectious Diseases, Otto-von-Guericke University of Magdeburg, Germany 2 Department of Radiology and Nuclear Medicine, 2nd Radiology Department, University of Gdansk, Poland
Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Downloaded by: The University of Hong Kong. Copyrighted material.
654
Case series
655
using multiplanar reconstructions. With this mode, exact measurements could be taken and, in addition, non-biliary structures could be visualized.
Results Six consecutive patients (four men, two women; mean age 69, range 56 – 80) were investigated with 3D-ERCP. Detailed informa" Table e1. tion about the patients is given in ● The median total radiation exposure, expressed as the dose area product, was 3238 µGym2 (range 1666 – 6033 µGym2). The 3 D image helped in clinical decision-making for all of the patients. Patient #1 presented with hilar strictures due to infiltration of a cholangiocarcinoma. The exact size of the tumor correlated well with the measured stricture length in 3D-ERCP (both 35 mm). In contrast, the measurement during ERCP with standard projection underestimated the size of the tumor, revealing only half of the true length of the stenosis. Patient #2 also presented with hilar cholangiocarcinoma. The exact anatomy of the stricture was " Video 1). assessed in 3D-ERCP (● Another patient (#4) had an unclear hilar lesion, which was iden" Fig. 1). Drainage tified as a ruptured hilar cyst using 3D-ERCP (● of both hepatic lobes was performed endoscopically The patient improved and the size of the cyst decreased significantly. Another patient (#5) had dilated intrahepatic ducts. During standard ERCP the diagnosis of Klatskin tumor Bismuth type I was raised and 3D-ERCP was performed to explore the length of the stricture. However, 3D-ERCP in fact revealed a Klatskin type IV tumor, with the right hepatic duct originating from the stricture. This had a significant impact on the patient’s prognosis and further treatment. The suspected right-sided hepatic duct was cannulated with the help of a 0.025-inch angle-tip guidewire " Fig. 2; (VisiGlide; Olympus Europe, Hamburg, Germany) (● ●" Video 2). Both liver lobes were drained with plastic stents, which led to complete resolution of the patient’s jaundice. Postoperative bile leakage was detected and localized in two patients (#3 and #6). One of them (patient #6) had a variant of her biliary tree with segment 5 draining into the stump of the rem" Fig. 3; " Video 3). nant left hepatic duct (● ● In all of the patients, the image quality of the area of interest was good to excellent. In two of the six patients (#5 and #6), the initial diagnosis on the basis of the 2D-ERCP images was changed by the 3 D technique. In the remaining four patients, our experience was that the diagnosis was easier to prove and that localization of any stenosis or leakage was more precise.
Fig. 1 Three-dimensional endoscopic retrograde cholangiopancreatography (3D-ERCP) image in patient #4 showing a ruptured cyst in the liver hilum.
exposure by providing better guidance for the procedure. Poor visibility in terms of orientation during 2D-ERCP may result in high radiation exposure, too. In our series, the radiation exposure including the 3 D scan was not significantly higher when compared with other advanced biliary interventions [5]. One potential difficulty of using 3D-ERCP is the presence of possible artifacts that result from breathing movements. The system is not equipped with a breathing sensor, which is present in MRCP equipment, so strong breathing excursion can therefore lead to artifacts in the resulting images. The additional time required for the 3D-ERCP is low, being around 2 minutes in total. The overall investigation time was not recorded in our case series but the total investigation time may be shortened by the ability to perform more precisely targeted interventions. In the 3 D rendered images, only structures with high contrast are displayed. The use of multiplanar reconstructions with the em-
Video 1 Discussion !
This is the first study to describe the clinical use of 3D-ERCP with c-arm CT. We found that 3 D imaging provided additional valuable information to guide endoscopic biliary therapy. Two-dimensional investigations have known limitations and may lead to misinterpretation of anatomic structures, mainly due to the posture and projection plane of the c-arm system. The correct identification of anatomic variations, as shown in one of our patients, can play a crucial role in the treatment of biliary diseases. In addition, with 3D-ERCP, the exact extent of a stenosis can be assessed, as demonstrated in our series. The application of 3D-ERCP leads to an increased radiation exposure, but overall 3D-ERCP has the potential to lower the radiation
Online content including Three-dimensional endoscopic retrograde cholangiopancreatogra- video sequences viewable phy (3D-ERCP) of patient #2 showing at: www.thieme-connect.de a hilar stricture. The patient had undergone a Billroth II resection and the investigation was performed with a side-viewing endoscope. The filling defects in the common bile duct are caused by a stone-retrieval balloon.
Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Downloaded by: The University of Hong Kong. Copyrighted material.
!
Case series
Fig. 2 Images from patient #5 showing: a the standard endoscopic retrograde cholangiopancreatography (ERCP), which appeared to confirm the initial diagnosis of a Klatskin tumor Bismuth type I; b the 3D-ERCP, which revealed that only the left hepatic bile ducts were opacified; c rotated 3 D image (lateral projection; see pictogram in the lower corner), which revealed that the right hepatic duct was originating from the stricture (arrow); d the plastic stents positioned in both hepatic ducts after the right-sided hepatic duct had been cannulated.
Downloaded by: The University of Hong Kong. Copyrighted material.
656
Fig. 3 Images from patient #6 who had undergone previous left-sided hemihepatectomy showing: a, b different angles of view on the 3D-rendering model of the bile ducts demonstrating biliary leakage (*) from a small bile duct (arrow) that was draining into the remnant left hepatic duct; c planar reconstruction of the 3D-ERCP, which demonstrated the origin of the bile duct causing the biliary leak.
Video 2
Online content including Three-dimensional endoscopic retrograde cholangiopancreatogra- video sequences viewable phy (3D-ERCP) of patient #5 clearly at: www.thieme-connect.de demonstrating that only the left hepatic duct was initially opacified. The right hepatic duct is visible only as a small area of contrast to the right of the visible stricture. Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Video 3
Online content including Three-dimensional endoscopic retrograde cholangiopancreatogra- video sequences viewable phy (3D-ERCP) of patient #6 showing at: www.thieme-connect.de the leaking right-sided bile duct originating from the remnant left hepatic duct.
Case series
Competing interests: None
References 1 Hintze RE, Abou-Rebyeh H, Adler A et al. Magnetic resonance cholangiopancreatography-guided unilateral endoscopic stent placement for Klatskin tumors. Gastrointest Endosc 2001; 53: 40 – 46
2 Kapoor BS, Esparaz A, Levitin A et al. Nonvascular and portal vein applications of cone-beam computed tomography: current status. Tech Vasc Interv Radiol 2013; 16: 150 – 160 3 Nanashima A, Abo T, Sakamoto I et al. Three-dimensional cholangiography applying C-arm computed tomography in bile duct carcinoma: a new radiological technique. Hepatogastroenterology 2009; 56: 615 – 618 4 Shimazu M, Wakabayashi G, Tanabe M et al. [Three-dimensional cholangiography and angiography for hilar cholangiocarcinoma]. Nihon Geka Gakkai Zasshi 2000; 101: 393 – 398 5 Kruit AS, Vleggaar FP, van Erpecum KJ et al. No reduction of radiation dose following the introduction of dose-area product measurement in endoscopic retrograde cholangiopancreatography. Eur J Gastroenterol Hepatol 20. 05 2014: DOI 10.1097/MEG.0000000000000128
Table e1 online content viewable at: www.thieme-connect.de
Downloaded by: The University of Hong Kong. Copyrighted material.
phasis on soft tissues allows nonopacified biliary structures also to be visualized. In conclusion, 3D-ERCP is useful for the investigation of biliary pathologies and can facilitate ERCP in specific patients. The data are preliminary but, so far, our experience with 3D-ERCP encourages its use in complex conditions.
657
Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Cone-beam computed tomography – adding a new dimension to ERCP
Authors
Jochen Weigt1, Maciej Pech2, Arne Kandulski1, Peter Malfertheiner1
Institutions
1
submitted 11. November 2014 accepted after revision 31. December 2014
Six patients with biliary duct pathologies were investigated with three-dimensional endoscopic retrograde cholangiopancreatography (3D-ERCP). The technique of 3D-ERCP consists of c-arm rotation using a flat detector c-arm to acquire a complete set of images for 3 D visualization of the biliary system. Our case series demonstrates a
high level of accuracy with regard to the diagnosis of stenosis and biliary leakage. In two patients the diagnosis obtained from conventional ERCP changed significantly after 3D-ERCP. From these early observations we conclude that 3D-ERCP enhances the diagnostic performance of ERCP in a subset of patients with biliary tract disease.
Introduction
Patients and methods
!
!
Endoscopic retrograde cholangiopancreatography (ERCP) has a high diagnostic and therapeutic impact and is the gold standard for image-guided interventions of the biliary tract [1]. ERCP for diagnostic purposes has been replaced by magnetic resonance cholangiopancreatography (MRCP), which provides a clear three-dimensional (3 D) overview of the anatomy of the biliary tract. Previous studies have shown the benefit of 3 D imaging to guide therapeutic ERCP [1]. Newer fluoroscopy units are equipped with digital flat detectors which can be used to scan a 3 D set of images. This is known as cone-beam computed tomography (CBCT), flat-detector CT, or c-arm CT. With this modality, 3 D rendering is possible for better 3 D visualization along with multiplanar or curved reconstructions for additional soft tissue components. CBCT has been reported previously for percutaneous biliary interventions but never for ERCP [2 – 4]. It has been demonstrated that CBCT facilitates percutaneous biliary drainage or abscess drainage [2]. Moreover, CBCT can potentially lower the total radiation exposure because of improved planning and guidance of interventions. We describe the first clinical application of 3DERCP to facilitate endoscopic biliary interventions in a series of six patients.
This is a retrospective, observational, open-label, single-arm, consecutive case series conducted at a tertiary-care endoscopy center. We defined 3DERCP as the use of digital detectors to create a 3 D set of images during ERCP, including by 3D-image rendering. This technique was used for patients undergoing ERCP who had inconclusive anatomy of the biliary tract, leakage or strictures. All of the patients were sedated and investigated in the prone position. After the bile ducts had first been opacified with nondiluted contrast medium (Peritrast 60 %; Dr. Franz Köhler Chemie, Bernsheim, Germany) using standard catheters, papillotomes, or an inflated stone-retrieval balloon, the area of interest was centered in the fluoroscopy c-arm (Phillips Eleva Multi Purpose System; Philips Healthcare, Hamburg, Germany). Once in position, the c-arm was rotated 180° around the patient in a period of 6 seconds acquiring 15 images per second, so producing 90 images. Software on a dedicated 3 D workstation was used to render the images (Multi Diagnost Eleva 3D-RX-Option; Phillips Healthcare), a process taking approximately 2 minutes. The ERCP was then continued using the 3 D images obtained. The data were used to guide the procedures in different ways. First, the rotational scan was reviewed providing an overview of the 3 D positioning of the anatomic structures. Second, the biliary structures were rendered to a 3 D image similar to those seen during MRCP. Third, the full set of images was handled in a similar manner to a CT scan
Bibliography DOI http://dx.doi.org/ 10.1055/s-0034-1391483 Published online: 12.2.2015 Endoscopy 2015; 47: 654–657 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0013-726X Corresponding author Jochen Weigt, MD Dep. Gastroenterology, Hepatology and Infectious Diseases Otto-v.-Guericke University of Magdeburg Leipziger Str. 44 39120 Magdeburg Germany Fax: +49-391-6713105 [email protected]
Department of Gastroenterology, Hepatology and Infectious Diseases, Otto-von-Guericke University of Magdeburg, Germany 2 Department of Radiology and Nuclear Medicine, 2nd Radiology Department, University of Gdansk, Poland
Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Downloaded by: The University of Hong Kong. Copyrighted material.
654
Case series
655
using multiplanar reconstructions. With this mode, exact measurements could be taken and, in addition, non-biliary structures could be visualized.
Results Six consecutive patients (four men, two women; mean age 69, range 56 – 80) were investigated with 3D-ERCP. Detailed informa" Table e1. tion about the patients is given in ● The median total radiation exposure, expressed as the dose area product, was 3238 µGym2 (range 1666 – 6033 µGym2). The 3 D image helped in clinical decision-making for all of the patients. Patient #1 presented with hilar strictures due to infiltration of a cholangiocarcinoma. The exact size of the tumor correlated well with the measured stricture length in 3D-ERCP (both 35 mm). In contrast, the measurement during ERCP with standard projection underestimated the size of the tumor, revealing only half of the true length of the stenosis. Patient #2 also presented with hilar cholangiocarcinoma. The exact anatomy of the stricture was " Video 1). assessed in 3D-ERCP (● Another patient (#4) had an unclear hilar lesion, which was iden" Fig. 1). Drainage tified as a ruptured hilar cyst using 3D-ERCP (● of both hepatic lobes was performed endoscopically The patient improved and the size of the cyst decreased significantly. Another patient (#5) had dilated intrahepatic ducts. During standard ERCP the diagnosis of Klatskin tumor Bismuth type I was raised and 3D-ERCP was performed to explore the length of the stricture. However, 3D-ERCP in fact revealed a Klatskin type IV tumor, with the right hepatic duct originating from the stricture. This had a significant impact on the patient’s prognosis and further treatment. The suspected right-sided hepatic duct was cannulated with the help of a 0.025-inch angle-tip guidewire " Fig. 2; (VisiGlide; Olympus Europe, Hamburg, Germany) (● ●" Video 2). Both liver lobes were drained with plastic stents, which led to complete resolution of the patient’s jaundice. Postoperative bile leakage was detected and localized in two patients (#3 and #6). One of them (patient #6) had a variant of her biliary tree with segment 5 draining into the stump of the rem" Fig. 3; " Video 3). nant left hepatic duct (● ● In all of the patients, the image quality of the area of interest was good to excellent. In two of the six patients (#5 and #6), the initial diagnosis on the basis of the 2D-ERCP images was changed by the 3 D technique. In the remaining four patients, our experience was that the diagnosis was easier to prove and that localization of any stenosis or leakage was more precise.
Fig. 1 Three-dimensional endoscopic retrograde cholangiopancreatography (3D-ERCP) image in patient #4 showing a ruptured cyst in the liver hilum.
exposure by providing better guidance for the procedure. Poor visibility in terms of orientation during 2D-ERCP may result in high radiation exposure, too. In our series, the radiation exposure including the 3 D scan was not significantly higher when compared with other advanced biliary interventions [5]. One potential difficulty of using 3D-ERCP is the presence of possible artifacts that result from breathing movements. The system is not equipped with a breathing sensor, which is present in MRCP equipment, so strong breathing excursion can therefore lead to artifacts in the resulting images. The additional time required for the 3D-ERCP is low, being around 2 minutes in total. The overall investigation time was not recorded in our case series but the total investigation time may be shortened by the ability to perform more precisely targeted interventions. In the 3 D rendered images, only structures with high contrast are displayed. The use of multiplanar reconstructions with the em-
Video 1 Discussion !
This is the first study to describe the clinical use of 3D-ERCP with c-arm CT. We found that 3 D imaging provided additional valuable information to guide endoscopic biliary therapy. Two-dimensional investigations have known limitations and may lead to misinterpretation of anatomic structures, mainly due to the posture and projection plane of the c-arm system. The correct identification of anatomic variations, as shown in one of our patients, can play a crucial role in the treatment of biliary diseases. In addition, with 3D-ERCP, the exact extent of a stenosis can be assessed, as demonstrated in our series. The application of 3D-ERCP leads to an increased radiation exposure, but overall 3D-ERCP has the potential to lower the radiation
Online content including Three-dimensional endoscopic retrograde cholangiopancreatogra- video sequences viewable phy (3D-ERCP) of patient #2 showing at: www.thieme-connect.de a hilar stricture. The patient had undergone a Billroth II resection and the investigation was performed with a side-viewing endoscope. The filling defects in the common bile duct are caused by a stone-retrieval balloon.
Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Downloaded by: The University of Hong Kong. Copyrighted material.
!
Case series
Fig. 2 Images from patient #5 showing: a the standard endoscopic retrograde cholangiopancreatography (ERCP), which appeared to confirm the initial diagnosis of a Klatskin tumor Bismuth type I; b the 3D-ERCP, which revealed that only the left hepatic bile ducts were opacified; c rotated 3 D image (lateral projection; see pictogram in the lower corner), which revealed that the right hepatic duct was originating from the stricture (arrow); d the plastic stents positioned in both hepatic ducts after the right-sided hepatic duct had been cannulated.
Downloaded by: The University of Hong Kong. Copyrighted material.
656
Fig. 3 Images from patient #6 who had undergone previous left-sided hemihepatectomy showing: a, b different angles of view on the 3D-rendering model of the bile ducts demonstrating biliary leakage (*) from a small bile duct (arrow) that was draining into the remnant left hepatic duct; c planar reconstruction of the 3D-ERCP, which demonstrated the origin of the bile duct causing the biliary leak.
Video 2
Online content including Three-dimensional endoscopic retrograde cholangiopancreatogra- video sequences viewable phy (3D-ERCP) of patient #5 clearly at: www.thieme-connect.de demonstrating that only the left hepatic duct was initially opacified. The right hepatic duct is visible only as a small area of contrast to the right of the visible stricture. Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
Video 3
Online content including Three-dimensional endoscopic retrograde cholangiopancreatogra- video sequences viewable phy (3D-ERCP) of patient #6 showing at: www.thieme-connect.de the leaking right-sided bile duct originating from the remnant left hepatic duct.
Case series
Competing interests: None
References 1 Hintze RE, Abou-Rebyeh H, Adler A et al. Magnetic resonance cholangiopancreatography-guided unilateral endoscopic stent placement for Klatskin tumors. Gastrointest Endosc 2001; 53: 40 – 46
2 Kapoor BS, Esparaz A, Levitin A et al. Nonvascular and portal vein applications of cone-beam computed tomography: current status. Tech Vasc Interv Radiol 2013; 16: 150 – 160 3 Nanashima A, Abo T, Sakamoto I et al. Three-dimensional cholangiography applying C-arm computed tomography in bile duct carcinoma: a new radiological technique. Hepatogastroenterology 2009; 56: 615 – 618 4 Shimazu M, Wakabayashi G, Tanabe M et al. [Three-dimensional cholangiography and angiography for hilar cholangiocarcinoma]. Nihon Geka Gakkai Zasshi 2000; 101: 393 – 398 5 Kruit AS, Vleggaar FP, van Erpecum KJ et al. No reduction of radiation dose following the introduction of dose-area product measurement in endoscopic retrograde cholangiopancreatography. Eur J Gastroenterol Hepatol 20. 05 2014: DOI 10.1097/MEG.0000000000000128
Table e1 online content viewable at: www.thieme-connect.de
Downloaded by: The University of Hong Kong. Copyrighted material.
phasis on soft tissues allows nonopacified biliary structures also to be visualized. In conclusion, 3D-ERCP is useful for the investigation of biliary pathologies and can facilitate ERCP in specific patients. The data are preliminary but, so far, our experience with 3D-ERCP encourages its use in complex conditions.
657
Weigt Jochen et al. Cone-beam CT during ERC … Endoscopy 2015; 47: 654–657
