Obesity has become a worldwide epidemic, not only in the developed world but also in developing countries.1 In fact, after the United States, China and India have the second- and third-largest obese population, respectively. Obesity leads to many comorbid conditions such as cardiovascular disease, diabetes mellitus, sleep apnea, and stroke; therefore, there is a critical need to find solutions to this widespread problem, which is responsible for one-fifth of all deaths in the United States.2 Obesity is defined as a body mass index (BMI) > 30 kg/m2, and morbid obesity is defined as a BMI > 40 kg/m2 (candidacy for bariatric surgery). Interestingly, these definitions are different for people of Asian ethnicity (BMI > 25 kg/m2 for obesity and > 35 kg/m2 for candidacy for bariatric surgery).3,4

Produced by cells located mostly (> 90%) in the mucosa of the fundus of the stomach, ghrelin is the only known orexigenic hormone. Gastric sleeve surgery causes a marked sustained reduction in ghrelin production (approximately 40%–50% at 5 years),5 and it is believed that this may be a contributing factor to weight loss in gastric sleeve surgery (other than the actual reduction in stomach volume). Therefore, it is logical that embolization of the fundus of the stomach could result in marked appetite suppression and, thus, weight loss. This article reviews study results of initial animal and human studies of bariatric embolization and their limitations, procedural technique, and challenges to conducting trials of gastric artery embolization (GAE) for obesity to confirm its safety and efficacy.


GAE has been used since the 1970s to treat life-threatening gastric hemorrhage.6,7 This is commonly accepted as standard of care, has been life saving for thousands of patients, and yet has been safe and effective enough to be used empirically in the setting of angiographically negative life-threatening hemorrhage.8

Recent animal in porcine and canine models have shown that GAE suppresses ghrelin levels, resulting in weight loss. Arepally et al first described the technique of GAE to reduce weight gain in a controlled study. Sodium morrhuate was used in a porcine model with resultant lower ghrelin levels and significantly blunted weight gain in otherwise rapidly growing young swine.9 Paxton et al described using 40-µm microsphere particle embolization in a similar porcine model, which also resulted in decreased ghrelin levels, reduced weight gain, and no duodenal upregulation of ghrelin.10,11 Bawudun et al described a technique of left gastric embolization using a mixture of bleomycin and Lipiodol (Guerbet USA) versus polyvinyl alcohol (PVA) 500- to 700-µm particles in a canine model. At the end of the 8-week trial, there was significant reduction in plasma ghrelin, body weight, and subcutaneous fat without gastric ulceration (unlike the swine studies where there was ulceration).12

Human Experience

A recent retrospective case-controlled study found a 7.9% average decrease in body weight at 3 months in 15 patients who underwent left GAE for life-threatening hemorrhage compared to 1.2% for age-matched controls who underwent embolization other than the left gastric artery for upper gastrointestinal bleeding (P = .001).13 In a prospective, single-arm study of five patients in the Georgian Republic, Kipshidze et al showed an average total body weight loss of 16% and 17% at 6 and 24 months, respectively, as well as reduced ghrelin levels after left GAE using 300- to 500-µm Bead Block (BTG International) compressible microspheres.14 In the US Food and Drug Administration (FDA)–supervised investigational device exemption (IDE) studies, Weiss et al reported 9% mean excess weight loss at 3 months using 300- to 500-µm Embosphere Microspheres (Merit Medical),15 and Syed et al reported a mean 17.2% excess weight loss at 6 months using 300- to 500-µm Bead Block.16 However, these prospective human studies have only had short-term follow-up with a very limited number of patients.

This stage of early prospective human experience has shown that the procedure appears to be safe in the intermediate term. In animal studies, when more than one gastric fundal artery was embolized, 40% to 100% of animals developed gastric ulcers using 40-µm particles (embolized to stasis for five cardiac cycles) at 3 to 5 weeks even with a proton pump inhibitor and sucralfate. Kipshidze et al reported no ulceration at 24 hours and at 1-week postprocedure after using Bead Block (300–500 µm, embolized to stasis of the distal branches) without gastroprotective agents.14 Syed et al reported superficial ulcerations in 75% of four patients at day 3 postprocedure endoscopy (all healed by 30 days) after use of 300- to 500-µm Bead Block (embolized to stasis of the distal branches for at least five cardiac cycles, with proton pump inhibitor and sucralfate).16 Superficial ulcerations were described by Weiss et al in one of five patients based on endoscopy at approximately 3 weeks, even with use of gastroprotective agents, but they were healed at the time of a 3-month endoscopy. Additionally, there was one episode of transient pancreatitis.15


Although intermediate-term results are promising in small studies, efficacy of GAE for obesity has not yet been established, and the mechanism of action has not been precisely determined. Ghrelin is the obvious player, but other gastrointestinal neuropeptides may be involved. Ghrelin secretion is reduced after bariatric surgery, but other gut hormones/neuropeptides are also stimulated, including glucagon-like peptide-1, peptide tyrosine-tyrosine, and oxyntomodulin.17-19 There may also be some type of mechanical effect in addition to appetite suppression and weight loss. Surgical placebo effect has also not been excluded.

It should also be noted that embolizing the left gastric artery may preclude the use of gastric sleeve surgery in the future for the patient, because the left gastric artery becomes the only supply to the gastric remnant in gastric sleeve surgery. In addition, gastric sleeve surgery and gastric bypass have level 1 evidence to support their efficacy in the morbidly obese population unlike bariatric embolization, and there is no surgical backup procedure available if bariatric embolization fails. Some preliminary animal studies suggest that it may be feasible to perform bariatric embolization solely using the gastroepiploic arteries (EMBARGO procedure), which would have the theoretical advantage of allowing sleeve gastrectomy to be safely performed afterward. Although decreased ghrelin levels were noted at 3 weeks in these swine studies, blunting of weight gain or weight loss was not mentioned.20,21

Future Directions

To date, only patients with a BMI > 40 kg/m2 have been allowed to participate in clinical trials per the FDA and European trials. A large population of patients with BMI < 40 kg/m2 may potentially benefit from this procedure if its safety and efficacy become further established (eg, patients with a BMI > 35 kg/m2 and comorbidities). Obese diabetic patients are another subgroup that should be studied further, because dysglycemia and frank diabetes are directly related to insulin resistance as a result of visceral fat deposition (abdominal obesity). Anecdotally, one diabetic patient underwent bariatric embolization in our GET LEAN study,16 and hemoglobin A1c levels normalized at 3 months and remained normal at follow-up at 6 months.


Standard microcatheters, guidewires, and embolic agents that are commonly utilized in visceral interventions are also used for GAE. Bead Block compressible microspheres (300–500 µm), PVA particles (500–700 µm), and Embosphere Microspheres (300–500 µm) have been given IDEs for clinical trials by the FDA.

Similar to the uterus, the stomach has a rich collateral arterial supply, which helps prevent complications such as gastric infarction or nonhealing ischemic ulceration. However, visceral vessels are prone to spasm, and therefore, intra-arterial vasodilators are extremely helpful during the procedure.

In addition, anatomic variation commonly occurs in these visceral vessels, and it is critical to have carefully mapped out the anatomy to ensure the target vessel is identified properly. The anatomy of the left gastric artery typically has an S-curve configuration, which makes catheterization rather difficult. Often, a reverse curve catheter (from the femoral approach) is used to leverage the tip into a vertically oriented superior arterial branch off of the celiac artery (whereas the main celiac artery trunk may be horizontal or even inferiorly oriented forming an acute angle anteriorly with the aorta). This can be challenging if one is not familiar with this procedure.

Figure 1. Selective left celiac angiogram in the anteroposterior projection demonstrates angiographic anatomy prior to selection. The left gastric artery is highlighted (arrow).

Figure 2. Preembolization selective left gastric arteriogram reveals normal anatomy prior to embolization.

Figure 3. Postembolization selective left gastric arteriogram demonstrates “pruning” of the arterial branches consistent with successful embolization.

Procedure Description

The steps for performing GAE have been previously described by Syed et al.16 Ultrasound-guided vascular access is either obtained to the right common femoral artery or left radial artery. A pigtail catheter is then utilized to perform an angiogram of the abdominal aorta at the level of the superior margin of T12. The celiac artery is then selected utilizing a 4- or 5-F Simmons 1 catheter or any other appropriate type of reverse curve catheter, which allows visualization of the left gastric artery (Figure 1). A coaxial microcatheter is used to further select the left gastric artery (Figure 2). At this point, the embolic agent is prepared according to the package insert and typically mixed with contrast. Embolic particles are then injected into the left gastric artery and its branches until there is complete cessation of flow (stasis) within the left gastric artery and its branches. Stasis is defined as visualization of contrast within the main left gastric artery for at least five cardiac cycles. Cessation of flow is then confirmed with follow-up contrast injection into the left gastric artery (Figure 3). Celiac arteriography is then performed to confirm occlusion of the left gastric artery and lack of gross nontarget embolization. Patient photographs before and after bariatric embolization are shown (Figure  4 and Figure 5).

Figure 4. Patient photograph preembolization of the left gastric artery.

Figure 5. Photograph of the same patient in Figure 4, 2 years after embolization of the left gastric artery (after 33 kg of weight loss).


Conducting Clinical Trials

Several variables need to be considered when undertaking a trial for GAE in obese patients. First, in the United States, an IDE is required from the FDA to conduct clinical research using experimental devices that may have significant risk, even if the device is already FDA approved for other clinical indications. Presently, embolization particles for bariatric embolization require an IDE for research in human subjects. It is also likely that for trials conducted in Europe, the equivalent European country competent authority approval would have to be obtained. Additionally, investigational review board approval must be obtained worldwide.

One of the unique challenges of designing and constructing a GAE trial is the ethical dilemma of embolizing an otherwise normal organ with no actual pathology. This is far different from any other interventional radiology procedure where a diseased or damaged organ is typically treated. Consequently, any trial would have to have a low threshold for stopping points. The potential for irreversible injury to an otherwise normal organ can be intimidating for any physician. Therefore, a trial must meet and preferably exceed the FDA's expectations for safety; this includes accounting for backup plans for all conceivable worst-case scenarios. Additionally, the FDA will likely mandate the use of a data safety monitoring board for oversight, and the European equivalent may also require a data monitoring committee.

At the same time, there is a conflicting drive to demonstrate efficacy, which would require taking a small amount of risk to “push the envelope,” as well as concern for a having a backup plan if a catastrophic complication (eg, stomach perforation/infarction) were to occur that required hospital admission/emergency surgery and/or resulted in significant morbidity or mortality. Costs related to these complications should be considered, as well as insurance coverage in general, as health insurance may not cover experimental procedures. Embolization endpoints also need to be determined; we used stasis of contrast during intra-arterial injection lasting at least five cardiac cycles in the GET LEAN study.16

Patient selection and recruitment can also be an issue. One must determine patient readiness and motivation, as well as exclude patients with eating disorders and those who eat for psychological comfort, which can drastically affect study results. Unlike animals, humans often eat when they are not hungry. Thus, inclusion of psychologists can be beneficial for patient selection. Additionally, interventional radiologists do not have a bariatric practice, and partnering with a bariatric surgeon/surgery department, although essential, may present its own challenges, such as (1) ethical dilemmas from referring surgeons due to the potential for excluding a patient from an effective and arguably life-saving procedure, especially if bariatric embolization fails; and (2) apprehension on the part of the surgeon regarding how to manage unforeseen and unknown complications should they arise.

Ensuring patient compliance, particularly with dietary follow-up, can be a factor in GAE trials. None of the patients in the GET LEAN trial followed up with the dietician regularly,16 which is similar to the experience in the BEAT Obesity study group.15 Another unique challenge is how to manage postembolization syndrome of the stomach, which has never been described in the literature to our knowledge. Postembolization syndrome is usually self-limited and is a combination of symptoms including nausea, vomiting, pain, leukocytosis, and fever. We typically see a similar pattern following embolization of the stomach.

There is also the challenge of how to control for confounding factors of change in diet and exercise patterns (or lack thereof), as well as how to handle unforeseen complications. Results of serum ghrelin can also be a confounding factor because these levels can fluctuate up to 116% within 3 to 4 hours, especially with respect to timing of usual food intake.22 Determining a set number of calories ingested prior to testing and having blood testing for ghrelin performed at a precise time after a meal challenge could help balance this confounding variable.

Psychologic factors include depression, which must be continuously assessed even if the patient has no prior history of depression. It is known that 50% of morbidly obese patients suffer from depression.23 Surgical placebo effect is another confounding factor but may not be able to be excluded early in trial experience. There is no way to quantify desired decrease in hunger, and it can only be measured subjectively or indirectly through measurement of dietary intake (that could be inaccurate).

Multidisciplinary Team Considerations

Team formation can also be difficult in terms of finding a gastroenterologist, bariatric medicine specialist, and anesthesia provider. A dietician must be motivated enough to encourage patients to be seen for regular follow-up visits, which can be difficult in this patient population. Explaining a potentially risky procedure to a skeptical referring primary care physician may also be difficult. In our study, one challenge we encountered was that patients often did not mention use of antidepressants and/or a history of prior depression so as not to be excluded from the study. Therefore, records from a primary care physician are critical, and communication with a primary care physician before and after the procedure is also very important. Coordination of the numerous tests, studies, and physician visits before and after the procedure can be complex, and involvement of a conscientious study coordinator is also a must.


The only proven treatment for obesity when medical therapy fails is bariatric surgery. Currently, there are three main surgeries performed in the United States: laparoscopic gastric banding, laparoscopic sleeve gastrectomy, and Roux-en-Y gastric bypass. Current practice patterns are favoring laparoscopic sleeve gastrectomy as the first-line surgical treatment. Moreover, recent study results are now showing long-term outcome of laparoscopic sleeve gastrectomy is comparable to Roux-en-Y gastric bypass without as many complications.24 It is estimated that only 0.1% of the obese population undergoes bariatric surgery.25 Thus, there is a tremendous opportunity for an alternative to bariatric surgery.

However, bariatric embolization is still in its infancy. The procedure itself may need to be further refined. This includes factors such as appropriate patient selection, particle size and composition/structure of embolic agents, and technique. Additionally, long-term confirmation of safety and efficacy (phase 2 clinical trials) will be necessary. Until this happens, randomized controlled trials, preferably with a sham procedure, cannot be considered. Intense skepticism and resistance by bariatric surgeons to bariatric embolization can also be expected, even if it becomes a proven modality.

It would be a feat for bariatric embolization to become a proven alternative to the current gold standard of bariatric surgery for treatment of morbidly obese patients. Therefore, long-term results in the 5- to 10-year range may be necessary. Insurance and reimbursement will also be a key factor, and interventional radiologists will need to identify their ideal role in the patient's overall care continuum, which could evolve along with the procedure. Knowledge and certification in obesity medicine would be de rigueur for any interventionist interested in practicing this procedure. n

1. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384:766-781.

2. Masters RK, Reither EN, Powers DA, et al. The impact of obesity on US mortality levels: the importance of age and cohort factors in population estimates. Am J Public Health. 2013;103:1895-1901.

3. Remedios C, Bhasker AG, Dhulla N, et al. Bariatric nutrition guidelines for the Indian population. Obes Surg. 2016;26:1057-1068.

4. Stegenga H, Haines A, Jones K, Wilding J; Guideline Development Group. Identification, assessment, and management of overweight and obesity: summary of updated NICE guidance. BMJ. 2014;349:g6608.

5. Yousseif A, Emmanuel J, Karra E, et al. Differential effects of laparoscopic sleeve gastrectomy and laparoscopic gastric bypass on appetite, circulating acyl-ghrelin, peptide YY3-36 and active GLP-1 levels in non-diabetic humans. Obes Surg. 2014;24:241-252.

6. Rösch J, Dotter CT, Brown MJ. Selective arterial embolization. A new method for control of acute gastrointestinal bleeding. Radiology. 1972;102:303-306.

7. Bookstein JJ, Chlosta EM, Foley D, Walter JF. Transcatheter hemostasis of gastrointestinal bleeding using modified autogenous clot. Radiology. 1974;113:277-285.

8. Morris DC, Nichols DM, Connell DG, Burhenne HJ. Embolization of the left gastric artery in the absence of angiographic extravasation. Cardiovasc Intervent Radiol. 1986;9:195-198.

9. Arepally A, Barnett BP, Patel TH, et al. Catheter-directed gastric artery chemical embolization suppresses systemic ghrelin levels in porcine model [published erratum appears in Radiology. 2008;249:1083]. Radiology. 2008;249:127-133.

10. Paxton BE, Kim CY, Alley CL, et al. Bariatric embolization for suppression of the hunger hormone ghrelin in a porcine model. Radiology. 2013;266:471-479.

11. Paxton BE, Alley CL, Crow JH, et al. Histopathologic and immunohistochemical sequelae of bariatric embolization in a porcine model. J Vasc Interv Radiol. 2014;25:455-461.

12. Bawudun D, Xing Y, Liu WY, et al. Ghrelin suppression and fat loss after left gastric artery embolization in canine model. Cardiovasc Intervent Radiol. 2012;35:1460-1466.

13. Kipshidze N, Archvadze A, Bertog S, et al. Endovascular bariatrics: first in humans study of gastric artery embolization for weight loss. JACC Cardiovasc Interv. 2015;8:1641-1644.

14. Weiss CR, Akinwande O, Paudel K, et al. Clinical safety of bariatric arterial embolization: preliminary results of the BEAT Obesity trial. Radiology. 2017 Jan 17:160914.

15. Syed MI, Morar K, Shaikh A, et al. Gastric artery embolization trial for the lessening of appetite nonsurgically (GET LEAN): six-month preliminary data. J Vasc Interv Radiol. 2016;27:1502-1508.

16. Gunn AJ, Oklu R. A preliminary observation of weight loss following left gastric artery embolization in humans. J Obes. 2014;2014:185349.

17. Perry B, Wang Y. Appetite regulation and weight control: the role of gut hormones. Nutr Diabetes. 2012;2:e26.

18. Cummings DE, Overduin J. Gastrointestinal regulation of food intake. J Clin Invest. 2007;117:13-23.

19. Murphy KG, Bloom SR. Gut hormones in the control of appetite. Exp Physiol. 2004;89:507-516.

20. Diana M, Pop R, Beaujeux R, et al. Embolization of arterial gastric supply in obesity (EMBARGO): an endovascular approach in the management of morbid obesity. proof of the concept in the porcine model. Obes Surg. 2015;25:550-558.

21. Diana M, Halvax P, Pop R, et al. Gastric supply manipulation to modulate ghrelin production and enhance vascularization to the cardia: proof of the concept in a porcine model. Surg Innov. 2015;22:5-14.

22. Shiiya T, Nakazato M, Mizuta M, et al. Plasma ghrelin levels in lean and obese humans and the effect of glucose on ghrelin secretion. J Clin Endocrinol Metab. 2002;87:240-244.

23. Youdim A. The Clinician's Guide to the Treatment of Obesity. New York: Springer; 2015.

24. Gagner M, Hutchinson C, Rosenthal R. Fifth International Consensus Conference: current status of sleeve gastrectomy. Surg Obes Relat Dis. 2016;12:750-756.

25. Angrisani L, Santonicola A, Iovino P, et al. Bariatric Surgery Worldwide 2013. Obes Surg. 2015;25:1822-1832.

Mubin I. Syed, MD, FSIR, FACR Dayton Interventional Radiology Boonshoft School of Medicine Wright State University Dayton, Ohio mubinsyed@aol.com Disclosures: Partial funding for the GET LEAN trial provided by the SIR Foundation Pilot Research Grant and by the Radiological Society of North America Research Seed Grant.

Azim Shaikh, MD, MBA
Dayton Interventional Radiology
Dayton, Ohio
Disclosures: Partial funding for the GET LEAN trial provided by the SIR Foundation Pilot Research Grant and by the Radiological Society of North America Research Seed Grant.

Sumeet Patel
Dayton Interventional Radiology
Dayton, Ohio
Disclosures: None.

Ilyas Ikramuddin, DO
Boonshoft School of Medicine
Wright State University
Dayton Gastroenterology
Dayton, Ohio
Disclosures: None.

Kamal Morar, MD, MBA
Dayton Interventional Radiology
Boonshoft School of Medicine
Wright State University
Dayton, Ohio
Disclosures: Partial funding for the GET LEAN trial provided by the SIR Foundation Pilot Research Grant and by the Radiological Society of North America Research Seed Grant.