13
Training in the Use of Fluoroscopy for Gastrointestinal Endoscopy

Douglas G. Adler¹ and Gandhi Lanke²

¹ University of Utah, School of Medicine, Huntsman Cancer Center, Salt Lake City, UT, USA

² Presbyterian‐Plains Regional Medical Center, Clovis, NM, USA

Introduction

Fluoroscopy plays a critical role in many endoscopic procedures and is a mainstay of most therapeutic endoscopic procedures such as ERCP, enteral stenting, pseudocyst drainage, etc. The proper use of fluoroscopy equipment and the ability to interpret fluoroscopic images correctly are core skills for gastroenterologists. Unfortunately, many gastrointestinal (GI) fellows receive no formal instruction in the proper use of fluoroscopy and are left to learn about this technology independently, which often results in significant gaps in knowledge. This chapter will review the fundamentals of fluoroscopic technique as it is used in GI endoscopy.

Training in fluoroscopy

While formal training in the use of fluoroscopy may be uncommon, all GI fellows have several potential routes to receive instruction in the proper use of fluoroscopy as it relates to the practice of clinical gastroenterology. Elective rotations in radiology, available to most GI fellows, allows direct experience working with radiologists and, perhaps more importantly, radiology technicians. These can be deep sources of knowledge regarding the use of fluoroscopy in general, the operation of fluoroscopic hardware, and the interpretation of fluoroscopic images. Outside of a formal elective, fellows requesting procedures on patients that require fluoroscopy should be encouraged to not simply order the test and read the radiologists report, but rather to attend the procedure itself and, if possible, be present for the radiologic interpretation of the images obtained. Lastly, when learning advanced procedures such as enteral stenting, ERCP, etc., fellows should also actively develop a good working understanding of how to generate and interpret adequate fluoroscopic images.

Hardware basics

Although gastroenterologists are infrequently called upon to directly operate a fluoroscope themselves, an understanding of how fluoroscopic hardware operates is very important. Many different types of fluoroscopy units are available, including and portable C‐arm units, fixed table models, and dedicated interventional units. All these devices operate under the same set of principles. Generally, the X‐ray beam passes from the X‐ray source (the cathode), usually located below the table and above the floor, upward through the patient and the examination table. Most people assume the beam passes in the opposite direction, but this is not the case.

The beam then is received by the image intensifier/receiver unit, which is typically suspended above the patient. As the X‐ray beam passes through the table and the patient, there is always some degree of scatter. As X‐rays encounter matter (such as the patient, the table, etc.), some of these rays are deflected from their initial direction of travel. In addition, scatter increases proportionately with the patient’s BMI (adipose tissue scatters X‐rays), which can make generation of adequate images difficult in obese patients [1].

The image intensifier should be placed as close to the patient as possible (within 1–2 inches). This will allow the procedure to be accomplished with less radiation usage overall and result in the generation of sharper, and often brighter, images. Placing the image intensifier further away from the patient will result in a darker, granier, and more magnified image and is generally not preferable. An image that has been too greatly magnified will limit the field of view and will be of lower resolution.

Protective garments

Radiation exposure to the patient and medical staff should be minimized and adhere to the ALARA principle (As Low As Reasonably Achievable). Personnel should always wear appropriate protective garments such as lead vests, skirts, aprons, and thyroid shields. Garments should fit properly and not be too large or too small for the wearer. Leaded eyewear is available as well, but many find such eyeglasses to be unacceptably heavy and/or uncomfortable and they are not universally used. Radiation badges, which can monitor a person’s exposure over a given time period, should be worn if they are available. The badge should be worn on the part of the body that is closest to the fluoroscope. If available, additional shielding in the form of leaded glass partitions should be placed between personnel and the fluoroscopy beam. These shields can be supported by an armature from the ceiling or can be stand‐alone objects that can be wheeled into place [2,3].

Most modern fluoroscope units are digital in design. These units are able to generate clear, high contrast images using lower radiation doses when compared to older analog units. Modern fluoroscopy units have two monitors. One monitor shows a “live” image when the fluoroscope is activated and often continues to display the last image captured once the fluoroscopy beam in deactivated. A second monitor is generally used as a reference monitor, upon which images previously obtained during the current or prior procedures may be recalled for study or comparison.

Most digital fluoroscopes can operate between 1 and 24 frames per second (fps). Radiation exposure can be significantly minimized by reducing the number of images per second (frame rate) that are generated. A common mistake is to set the frame rate too high, in the generally false assumption that this will allow better visualization. This results in excessive radiation exposure to the patient and personnel and often adds little to the procedure. With time and experience, almost all interventional GI procedures can be performed at very low frame rates. Often, no more than 3 fps are required to produce excellent visualization.

Clear, unambiguous communication between physicians and radiology technicians and support staff is important as it avoids confusion and can effectively serve to reduce radiation exposure. The use of phrases, such as “spot film” (a brief exposure to check position), “fluoroscopy on,” “fluoroscopy off,” and “full exposure” (to obtain an image that will be permanently saved) are very helpful for the fluoroscope operator.

Scout films

Prior to any fluoroscopic procedure, a baseline X‐ray image of the area to be intervened upon should be obtained. Such an image is known as a “scout film” (Figure 13.1). The scout film allows assessment of many features including the presence of prior GI contrast, the bowel gas pattern, the presence of any intra‐abdominal prostheses (feeding tubes, stents, drainage catheters, IVC filters, etc.), and the location of any air‐filled structures (the stomach, the airways, etc.). Scout films also allow for jewelry or inappropriately placed EKG leads, both of which may obscure areas of interest, to be removed or adjusted. The scout film can also be used as a baseline reference during a procedure to evaluate landmarks and to look for evidence of complications such as a perforation (which may manifest as extraluminal air or contrast).

ERCP

Of all the procedures in gastroenterology, none is as closely linked to the use of fluoroscopy as ERCP. While rare reports of ERCP being performed without X‐rays exist, these are generally in unusual situations. For all intents and purposes, ERCP cannot be properly performed without the use of fluoroscopy.

While in theory imaging and intervening upon the pancreatic and bile ducts via ERCP is straightforward, in practice it remains the most involved and complicated of all GI procedures. Pancreaticobiliary anatomy tends to be highly individualized with many variations. Distorted anatomies, either due to surgery, inflammation, or neoplastic processes, are commonly encountered. Furthermore, ERCP involves the use of a very wide range of endoscopic accessories, each of which has a specific fluoroscopic appearance and specialized radiopaque markers that the operator must become familiar with.

With regard to fluoroscopy, ERCP is most commonly performed with a mobile C‐arm, but some facilities use dedicated fixed table units or even more costly dedicated interventional fluoroscopes (of the kind more commonly used by interventional radiologists). In general, mobile C‐arms are the simplest to use but generate images of lesser quality, while fixed table units and interventional fluoroscopes generate higher‐quality images but at a higher financial cost.

ERCP is usually performed with patients in the prone position. Dye injected into the pancreaticobiliary tree produces pancreatograms and cholangiograms. Having the fluoroscope arranged vertically over the patient is often adequate to evaluate most pancreatograms given the anatomy of the pancreas. The biliary tree often has a more complex three‐dimensional shape, and rotation of the fluoroscope (“rainbowing”) is often required to fully interpret cholangiograms. Ducts may overlap (especially at or near the hepatic hilum), and rotation of the fluoroscope can help to image individual ducts more clearly. The biliary tree is also situated in a larger organ (the liver) than the pancreas and as such requires more contrast to fully opacify. Dependent ducts will fill first, and anti‐dependent ducts may require additional contrast or patient manipulation to be fully visualized.

The use of magnification with fluoroscopy during ERCP can expose staff and patients to increased risk of radiation. To avoid increased exposure, magnification should be used only when necessary. Also, use of fluoroscopy only when viewing the image and wire‐guided cannulation instead of injection‐cannulation techniques can minimize use of fluoroscopy during ERCP [4]. Factors that can contribute to increased fluoroscopy time (FT) and radiation dose during ERCP include anatomic location of the pathology [extrahepatic (mean FT 4.86 minutes), pancreatic (mean FT 5.98 minutes), and intrahepatic (mean FT 12.26 minutes)], difficult cannulation after 5 minutes, bile duct injury, biliary stent placement, removal of (trapped) internally migrated biliary stents, cannulation of the minor papilla, biliary stone extraction greater than 10 mm, pancreatic strictures, acute or recurrent pancreatitis, radiographer controlling fluoroscopy system, and benign strictures in the hilum and above [5,6]. Awareness of factors contributing to increased fluoroscopy and radiation dose can prepare the endoscopist with extra precautions. In addition to lead apron, thyroid shield, and protective lens wear, use of X‐ray shielding device (Hagoromo X‐ray Protective Curtain, Maeda & Co., Ltd, Tokyo, Japan: 0.125 mm lead equivalent) covering the X‐ray tube table can decrease exposure from scattered radiation to the endoscopist and the staff [7]. Also, reducing the distance between endoscopy and fluoroscopy screens can limit the time in rotating the endoscopists head and body to view the screens, which in turn can limit the FT, decreasing the amount of radiation [8].

ERCP in pregnancy can be challenging as there is risk of radiation to the fetus from fluoroscopy. Without fluoroscopy, residual stones or debris can be missed in common bile duct, which can potentially lead to recurrent cholangitis [9]. Modified technique including a supine position on the fluoroscopy table, avoiding use of hard copy radiographs, use of lead shield of 0.5–1.0 mm thickness in the lower abdomen and pelvis, and positioning of uterus outside the primary X‐ray beam can be used to decrease the radiation dose [10]. Thermoluminescent dosimeters (TLDs) can, if available, estimate fetal radiation exposure in pregnant women. TLDs can be placed on the skin (one pair on the abdomen over the uterus shielded by lead, one pair on the upper abdomen in the primary beam, one pair on the upper back in the primary, one pair on the lower back beneath the uterus shielded by lead) to estimate the radiation exposure to fetus [10]. There could be differences in actual and estimated exposure to radiation with fluoroscopy due to body habitus and accuracy of TLD reading. In general, ERCP is safe in pregnancy particularly after the first trimester.

Fluoroscopy and enteral stents

Enteral stents (those deployed in the esophagus, small bowel, and colon) are universally designed to be deployed under fluoroscopic guidance. Small and large bowel stents are usually deployed under combined endoscopic and fluoroscopic guidance as they utilize through‐the‐scope (TTS) technology, whereas esophageal stents are generally deployed using only fluoroscopic guidance. Esophageal stents utilize delivery catheters that are too thick to pass through the working channel of even a therapeutic endoscope [11,12].

Fluoroscopy is critical for the proper and safe placement of enteral stents in several respects. Prior to stenting a luminal stenosis, fluoroscopy aids in the assessment of the stenosis beyond what can be seen endoscopically. Injection of radiopaque contrast dye, usually through biliary catheters, provides valuable data regarding the length and geometry of the stricture and simplifies stent selection with regard to both length and diameter. Fluoroscopy allows precise localization of stents before, during, and after deployment. Most modern enteral stent delivery catheters, as well as the stents themselves, have built‐in radiopaque markers for better visualization under fluoroscopy.

Esophageal stents

There are different kinds of esophageal stents available, and they should be used judiciously based on the indications. These include removable fully covered self‐expandable metal stents (FC SEMS), removable partially covered self‐expandable metal stents (PC SEMS), and biodegradable stents (BDS) [13]. The various indications include inoperable malignant esophageal stricture, tracheoesophageal fistula, esophageal perforation, refractory benign esophageal stricture to balloon dilation, and extrinsic esophageal compression from primary or secondary mediastinal and lung tumors [14,15]. BDS can be used in refractory benign esophageal strictures as they have an advantage of prolonged dilatory effect when left in situ and undergo progressive degradation and less local tissue reaction because of biocompatible material [16].

All modern esophageal stents utilize relatively thick, semirigid delivery systems. None of these systems utilize TTS technology. Endoscopy is performed prior to stent placement to evaluate the stricture directly, to obtain tissue as needed, or to perform other evaluations such as endoscopic ultrasound (EUS) (to stage esophageal cancers) [17]. Esophageal stents are generally deployed under fluoroscopy and without endoscopic guidance. In cases where very precise control over stent deployment is needed and the use of fluoroscopy alone is felt to be inadequate, an upper endoscopy may be advanced alongside the esophageal stent delivery catheter to provide direct visual assessment.

All esophageal stents are deployed over a central guide wire. The guide wire must be advanced across the stricture, typically into the stomach, to help ensure that the stent is deployed within the lumen and to reduce the risk of a perforation. Guide wires that are readily visible under fluoroscopy (Savary wires or wires used for ERCP) work best, and in general stiffer wires are more commonly used, but some endoscopists prefer floppy wires. Once a guide wire is in proper position, fluoroscopy can be used to make sure that the wire does not inadvertently migrate proximally or distally during the remainder of the procedure.

There is no universal agreement as to how best to position patients on the fluoroscopy table for esophageal stent placement. Left lateral decubitus is often more comfortable for the patient but may make fluoroscopy more difficult if the fluoroscope itself cannot rotate (“rainbow”) around the patient. If the fluoroscopic view is a lateral one, relevant structures such as the trachea and the diaphragm may be more difficult to discern than if an anterior‐posterior (AP) view is obtained. The supine position allows for more en face fluoroscopic views to be obtained but is a less comfortable position for the physician. The decision of how best to position the patient is often made based on the available fluoroscopy unit and the patient and physician preference. When placing an esophageal stent, the endoscopists should take time to identify several key landmarks. These include the trachea and the bifurcation of the respiratory carina as well as the level of the diaphragm and the expected location of the esophageal stent after placement (Figure 13.2). In patients with esophageal tumors, the relationship between the tumor and the trachea is critical as airway compression due to esophageal stent expansion can occur. This is a rare event but has been reported. Most patients can have placement of an esophageal stent next to the airway without difficulty, but the physician should be aware of the relationship between these structures. For distal tumors, an understanding of the location of the diaphragm is important as these lesions typically extend to the gastroesophageal junction. In these cases, the distal end of the stent may need to be placed into the gastric fundus to ensure complete luminal patency [18–20]. The gastric lumen often contains air as well and can easily allow demarcation of the esophagogastric junction. Lastly, the inferior lung borders and the diaphragm are often visible on fluoroscopy and can serve as another rough marker for the level of the esophagogastric junction.

Photo depicts fluoroscopic image in a patient with a mid-to-distal esophageal adenocarcinoma. — **Figure 13.2** Fluoroscopic image in a patient with a mid‐to‐distal esophageal adenocarcinoma. The image reveals the left and right mainstem bronchi (white arrows) and the diaphragm with air in the gastric lumen. The film demonstrates that the airways are far from the expected location of the stent and that airway compression is unlikely to develop.

Prior to deployment of an esophageal stent, fluoroscopic markers may be used to help localize the proximal and distal extent of the stenosis to ensure proper stent placement. Several techniques to mark these locations have been developed [21].

If the lumen through the stenosis is wide enough to allow passage of a standard gastroscope, it is possible to mark both the proximal and distal portion of the esophagus to be stented. Endoscope passage through the stricture also allows accurate measurement of the length of the stricture and, along with information obtained via contrast injection, assists in the selection of an appropriately sized stent. The proximal and distal portion of the stricture can be fluoroscopically marked or labeled in several ways.

Submucosal radiocontrast dye injection

Using a standard needle catheter, contrast dye can be injected into the submucosal space above and below the stricture. One to two milliliters of contrast dye is usually adequate. This technique is inexpensive and simple to do. The primary disadvantage of this technique is that the submucosal radiocontrast dye is absorbed into the surrounding tissues in several minutes, limiting the time between injection and stent deployment. If the endoscope cannot pass through the stenosis, then the proximal end alone can be marked. This marker, along with the known length of the stenosis obtained via contrast injection, is often adequate for stent placement.

Endoscopic clips as markers

Endoscopic clips, which are made of metal and thus easy to see fluoroscopically, can be placed at the proximal and distal margins of the obstruction. This technique also requires endoscopic advancement through the stenosis if a distal marker is desired. An advantage of this technique is that, once placed, the clips remain in position and the need to rapidly place the stent (as is the case with submucosal dye injection) is obviated. A disadvantage is that clip placement in the esophagus, especially at the distal margin of the stenosis, may be challenging. To determine the exact location of the clip above the tumor, the length of the tumor is subtracted from the length of the stent and divided by two [22]. As a general rule, a good position for a stent should be 2–3 cm above and below the stenotic segment to allow enough room to prevent malposition as the exact location of the stent can change with movement and/or respiration of the patient. Although clips are helpful as markers, they are often not needed by experienced endoscopists.