Endoscopic ultrasound (EUS) is continuously stepping into the therapeutic arena, simultaneously evolving in different directions, such as the management of pancreatic and biliary diseases, celiac neurolysis, delivering local intratumoral therapy, and EUS-guided endosurgery. EUS-guided vascular procedures are also challenging, considering the variety of vascular pathology, proximity of the vascular structures to the GI tract wall, high resolution, and real-time guidance offering an attractive access route and precise delivery of the intervention. The literature on vascular therapeutic EUS demonstrates techniques for the management of upper GI variceal and nonvariceal bleeding, pseudoaneurysms, and coiling and embolization procedures, as well as the creation of intrahepatic portosystemic shunts. The paucity of studies, diversity of study designs, and the number of animal model studies hamper a systematic approach to the conclusion and decision making important to clinicians and healthcare policy makers. Nevertheless, theoretical benefits and findings up to date concerning technical feasibility, efficacy, and safety of the procedures drive further research and development in this rather young therapeutic arena.
Since its beginning in the 1980s, EUS has evolved into a powerful diagnostic tool used widely for a number of GI conditions, successfully replacing other “gold standard” diagnostic modalities. In recent years, EUS is expanding to the interventional arena, usurping the management of pathology that was traditionally in the domain of therapeutic endoscopy, interventional radiology, and surgery. Therapeutic endoscopic ultrasonography (T-EUS) is developing simultaneously in different areas, such as the management of pancreatic and biliary diseases (e.g., pseudocyst drainage, cholangiopancreatography, etc.), celiac neurolysis, delivering local intratumoral therapy, EUS-guided endosurgery, and vascular procedures [
PubMed/MEDLINE was searched to identify relevant publications in English. The following search string was used: {(EUS-guided or endosonography guided or endosonographic guidance or endoscopic ultrasound guidance or endoscopic ultrasound guided or echo-endosonography guided or echo-endosonographic guidance or endoscopic Doppler US guided or endoscopic Doppler ultrasound guided or endoscopic Doppler US guidance) AND (vascular or EUS-guided or varices or nonvariceal or pseudoaneurysm or Dieulafoy or Dieulafoy’s or porto-systemic shunt)}. The final search was launched on January 29, 2013, with no time restrictions. Additionally, the Cochrane Library was searched.
Our PubMed/MEDLINE search yielded 1291 publications that were assessed for relevance according to title and abstract by two reviewers, and potentially relevant papers were retrieved in full text. The bibliographies of publications identified as relevant were manually searched for potentially relevant titles, and additional publications were found. Finally, twenty published papers were included for analysis in this review. Abstracts presented at congresses were not included. We are aware that our search possibly missed relevant papers due to a relatively inconsistent and evolving terminology in this new therapeutic arena.
In this paper, the current literature on EUS-guided vascular procedures is summarized in a narrative form considering the paucity of the studies, great diversity of study designs (mostly pilot studies, case records, and small case series), and a number of animal model studies, hampering any systematic approach to the literature synthesis.
We were restricted to the literature on vascular procedures under real-time EUS guidance. There are various miniature ultrasound/Doppler probes available that can be introduced through a working channel of the standard endoscope [
Several areas of vascular procedures under the EUS-guidance could be identified throughout the available literature. For the purpose of this review, we have selected five areas of EUS-guided vascular procedures: EUS-guided management of nonvariceal upper GI bleeding; EUS-guided management of variceal bleeding; EUS-guided management of pseudoaneurysms; EUS-guided embolization of portal venous system; EUS-guided creation of portosystemic shunts.
In the next sections, each of these areas will be discussed overviewing the available literature.
Since a certain percentage of nonvariceal upper GI bleeding episodes are diagnostically challenging and refractory to a standard therapy (various endoscopic techniques, angiographic therapy), EUS-guided detection and management can offer an alternative in this group of patients. To date, EUS-guided therapy for nonvariceal upper GI bleeding has been described for the management of peptic ulcer disease, Dieulafoy’s lesions, and bleeding tumors. EUS-guided management of the pseudoaneurysms, which also can bleed, will be discussed in the following section of this review.
Peptic ulcer disease is responsible for more than 50% of upper GI bleeding admissions to hospitals and has high recurrent bleeding rates and high mortality rates. Standard treatment, including nonendoscopic modalities, and upper GI endoscopy with injection therapy, thermal therapy, and/or clipping are safe and effective. However, relatively high rates of unsuccessful bleeding stoppage and recurrent bleeding are reported, occurring in up to 20% of patients [
Levy et al. described a patient with recurrent upper GI bleeding from a duodenal ulcer previously treated with heater probe plus injection at two separate occasions. On EUS-examination, the authors visualized tortuous vessel branching from gastroduodenal artery to mucosa and, under the guidance of curved linear echoendoscope, they injected 3 mL of cyanoacrylate through a 22-gauge FNA needle pointing at the 1.5 mm wide branching vessel located within the ulcer. Success of the treatment was confirmed immediately by Doppler. There were no complications, and no rebleeding occurred during 14-month followup [
Dieulafoy’s lesion is a relatively rare vascular malformation that presents with an acute refractory often massive upper GI bleeding, with a rather unsatisfactory detection rate at repeated endoscopy in the case of nonactive bleeding at the time of endoscopy. Endoscopic ultrasound and/or Doppler ultrasound are highly sensitive for the detection of vascular structures in the GI tract wall, which is especially useful in the absence of an endoscopically visible lesion (e.g., ulcer or protruding vessel), giving theoretical advantages to EUS-guided detection and treatment over standard modalities.
Fockens et al. in their small study treated three patients with Dieulafoy lesions by EUS-guided sclerotherapy [
The current evaluation of EUS-guided interventions for the management of peptic ulcer bleeding and Dieulafoy’s lesions is limited to anecdotal case records, one case series and one animal study, thus hampering any comparison with standard treatment options, and therefore recommendations cannot be given. However, the theoretical benefits of direct EUS and Doppler visualization of the “culprit” vessel responsible for the recurrent bleeding are evident, allowing precise treatment delivery and therefore possibly higher successful treatment rate in the groups of patients unsuccessfully treated or with recurrent bleeding. The feasibility and safety demonstrated in these publications are encouraging, hoping that future larger scale studies with appropriate study designs will possibly find some benefit, especially in some groups of the patients (unsuccessful treatment, recurrent bleeding episodes).
Endoscopic injection therapy and band ligation have been widely used for the management of bleeding and nonbleeding upper GI tract varices with a high success rate and with known complication rates [
Treatment techniques under direct EUS guidance seem to be promising. Lahoti et al. were the first to demonstrate sclerotherapy for upper GI varices under real-time EUS guidance. In a small pilot study on five patients with esophageal varices, they injected sclerosant (sodium morrhuate) through a 2.5 mm catheter injector needle under EUS guidance directly to perforating vessels until “no flow” was detected by color Doppler [
Injection sclerotherapy for upper GI varices has been associated with local complication and embolic incidents [
(a) Endosonographic transesophageal view—a 19-gauge needle inserted into the varix (arrow). (b) Coil delivery (arrows) through the 19-gauge needle. (c) Injection of 1 mL of cyanoacrylate. (d) Fundal varices obliterated. C: crus muscle, F: fundus; MP: muscularis propria of stomach wall. (Images curtsey of [
EUS-guided management of the upper GI varices and variceal bleeding shares the same theoretical benefits with the EUS-guided management of nonvariceal upper GI bleeding; that is, the visualization of the “culprit” lesions—varices, perforating and collateral veins—allows thorough planning and precise delivery of treatment, as has been demonstrated through the procedures described in the cited papers. Again, in comparison to effective standard treatment modalities, EUS-guided management can offer additional benefit for the patients unsuccessfully treated and those with recurrent bleeding. The quantity of the publications in this area of EUS-guided vascular procedures indicates rapid evolvement, and probably more high quality studies might be expected for these techniques. The studies undertaken up to now underpin the postulated theoretical benefits of EUS-guided therapy over standard techniques, which is most robustly demonstrated in the elective treatment of patients with esophageal collateral veins.
Visceral pseudoaneurysms are quite rare and serious complications of pancreatitis or abdominal surgery, with high mortality rates when ruptured. Management of visceral pseudoaneurysms includes interventional radiology procedures and surgery, with considerable morbidity and mortality [
Gonzalez et al. in their case record described the patient with chronic pancreatitis having a pseudocyst scheduled for EUS drainage. During the drainage attempt, an intracystic pseudoaneurysm of the splenic artery was injured and massive intracystic hemorrhage occurred. Since this was a potentially life-threatening condition, immediate puncture and cyanoacrylate injection under EUS guidance was performed, embolizing the distal arm of the splenic artery, therefore stopping the bleeding [
Roach et al. appear to be the first to describe the use of thrombin for the EUS-guided management of visceral pseudoaneurysm. A 32-year-old man with a superior mesenteric artery (SMA) pseudoaneurysm after pancreatitis and recurrent bleeding episodes was considered for selective embolization during angiography and percutaneous US/CT-guided injection therapy, but these were rejected due to unsuccessful selective catheterization and anatomic positioning. Under the EUS guidance the pseudoaneurysm was punctured with a 22-gauge needle, and 500 IU of thrombin was injected immediately obliterating the sac of pseudoaneurysm leaving the SMA patent, which was confirmed by Doppler. At computed tomography 12 weeks later partial recanalization was verified, but no further treatment was undertaken, and spontaneous rethrombosis of the pseudoaneurysm was detected by computed tomography 28 and 42 weeks after the treatment [
Publications of rather anecdotal cases of EUS-guided treatment of visceral pseudoaneurysms demonstrated the feasibility and technical ease of performing such procedures and, together with the theoretical benefits over traditional methods, are promising. However, theoretical disadvantages such as bleeding, introduction of infections, unwanted thrombotic events, and efficacy and safety issues, in comparison to standard treatment modalities, should be further robustly evaluated in order to potentially set these methods as the standard.
There is some evidence (studies on animal models) that EUS-guided portal vein pressure measurement is feasible, suggesting the EUS-guided transhepatic puncture of portal vein to be safer [
Portal hypertension of any etiology is associated with substantial complications. Lowering portal vein pressure, thus reducing complication rates, can be achieved with drugs, transjugular intrahepatic portosystemic shunt (TIPSS) placement, or surgical shunts. TIPS is a widely used and effective technique but with a number of possible periprocedural complications, such as pneumothorax formation, cardiac arrhythmias, and injury to the blood vessels in the liver resulting in hemorrhage. The proximity of the portal and hepatic veins to the scope of the EUS offers a potentially more favorable route for shunt formation under EUS guidance, avoiding some of the complications such as pneumothorax and cardiac arrhythmias. However, some potential disadvantages of this approach have to be acknowledged, such as introduction of infection and potentially higher bleeding complications, in comparison to TIPSS.
Recently Buscaglia et al. in their animal model study demonstrated the feasibility of EUS guided creation of an intrahepatic portosystemic shunt (IPSS). After positioning in the plane where both portal and hepatic veins were endosonographically visualized, transhepatic puncture of selected hepatic vein was performed firstly and then advanced to intrahepatic branch of the portal vein with a 19-gauge needle under EUS-guidance aided by fluoroscopy. Hereafter, the stylet was withdrawn following guidewire insertion in the portal vein through the needle, and the needle was withdrawn out from endoscope. After measuring the distance between selected veins, an appropriate uncovered biliary metal stent was inserted over the guidewire, bridging the selected hepatic and portal veins, therefore forming a portosystemic shunt (Figure
Endosonographic views during the stent deployment. (a) The stent (one arrow) delivery over the guidewire (two arrows) into the hepatic vein (three arrows). (b) Deployment of the stent; the proximal end of the stent inside the hepatic vein (one arrow) and the distal end of the stent inside the portal vein (two arrows). (c) The stent fully deployed with its proximal end inside the hepatic vein (one arrow) and the distal end inside the portal vein (two arrows); the guidewire (three arrows) (images curtsey of [
The paucity of the studies undertaken to date and their quality cannot give answers to the questions posed by clinicians and healthcare policy makers. However, papers published to date can give directions for future research that needs, in a more robust scientific manner, to address technical feasibility, efficacy, and safety of the procedures, as well as cost benefits in this rather young therapeutic arena.
Standard treatment modalities for the management of variceal and nonvariceal GI bleeding as well as for management of visceral pseudoaneurysms and TIPS creation are relatively effective and safe, and as such, are being used worldwide. However, unsuccessful treatment in a percentage of patients and recurrent bleeding episodes, together with theoretical benefits of EUS-guided interventions over standard treatment, as well as feasibility and technical ease demonstrated in the papers summarized in this review, calls for future research.
Further development of EUS-guided access to the vessels could potentially replace the interventional radiology vascular interventions in the abdomen and provide more efficient and precise, as well as safer local application of drugs (chemotherapeutics, fibrinolytics, etc.), vascular embolization (sclerotherapy, coiling, etc.), endoprostheses placement, and creation of portosystemic shunts [
We all witnessed the “transformation” of ERCP from a purely diagnostic procedure at its beginnings to a mainly therapeutic procedure nowadays [