Abstract
PURPOSE
We aimed to evaluate weight changes after left gastric artery (LGA) embolization in a retrospective cancer-naive cohort.
METHODS
A retrospective study was conducted to identify patients who underwent LGA embolization for gastrointestinal bleeding (GI). Patients with known cancer diagnoses at the time of LGA embolization were excluded. Pre- and postprocedure weights were assessed. Statistical analysis was performed using paired t-test and Wilcoxon signed-rank test.
RESULTS
A total of 39 patients were identified. In 21 patients who had documented pre- and postprocedural weights, a median of 16.3 kg weight loss (P = 0.045) was observed over a median time of 12 months (range, 2–72). In patients who had pre- and postprocedure endoscopies (n=6), 2 had worsening ulcers following LGA embolization and 4 had stable or no abnormal findings.
CONCLUSION
Our preliminary observation suggests that LGA embolization is well tolerated and results in unintended weight loss. Larger studies are needed to confirm these preliminary findings.
Obesity is an epidemic that affects more than a third of adults in the United States (1, 2). Despite advances in medical research and surgical techniques, obesity remains one of the largest drivers of chronic disease and healthcare costs in the United States (2, 3). In response to the growing obesity epidemic, left gastric artery (LGA) embolization has evolved as a promising procedure for weight loss therapy (4, 5). LGA embolization, also termed bariatric embolization when used for weight loss, decreases blood supply to the gastric fundus, which has the highest concentration of cells responsible for releasing ghrelin, a potent appetite-stimulating hormone (6). LGA embolization has been shown to suppress serum ghrelin levels and modulate weight in animal models (4, 5, 7). Promising results in animal models prompted investigators to retrospectively evaluate patients who underwent LGA embolization, an established procedure in interventional radiology for treatment of gastrointestinal (GI) bleeding. Early retrospective studies, though limited by small sample sizes and potential confounding factors, demonstrated significant weight loss as a nonintended side effect of LGA embolization and showed promise for reproducibility in humans. Results from clinical trials investigating the utility of LGA embolization to modulate weight have demonstrated promising results (4–6, 8–15). However, studies have been limited to small sample sizes and long-term data are lacking.
This retrospective cohort affords us a unique opportunity to assess a potential treatment population where weight loss is not intended and also to evaluate adverse events such as mucosal ulcers in the absence of strict standardized GI prophylaxis regimens implemented by current ongoing clinical trials. Drawing from the current literature on this topic, our hypothesis is that LGA embolization for GI bleeding will result in weight loss as an unintended side effect.
Methods
Patient selection
This single institution retrospective study was approved by the Institutional Review Board. Patients who underwent LGA embolization for treatment of acute gastrointestinal (GI) bleeding from 2001 to 2016 were identified through a search of the institutional database. Patients had to be at least 18 years of age and have no known cancer diagnoses at the time of procedure. All of the patients reviewed underwent LGA embolization in the inpatient or emergency setting. Twenty-one patients had documented pre- and postprocedure weights in their electronic medical record.
LGA embolization technique
While variations in technique existed depending on user preference, a generalized institutional approach to LGA embolization demonstrated the following technique: After sterile preparation of the right groin, the right common femoral artery was accessed with an 18-gauge needle. A Bentson wire was advanced into the aorta. The needle was exchanged for a 6 French (F) sheath and connected to a heparinized saline drip. A 5 F SOS catheter was advanced into the celiac artery. Selective catheterization was performed using a Progreat catheter and GT wire. The choice of embolic used depended on the type and location of the vascular lesion as well as user preference. The most common embolic agents used included gelfoam slurry followed by coil embolization and polyvinyl alcohol (PVA) particle embolization.
Outcome measures
Patients were assessed for any treatment-related adverse events. Primary adverse events were defined as the presence of gastric mucosal abnormalities ranging from superficial mucosal abnormalities to bleeding gastric ulcers on esophagogastroduodenoscopy. Medical records were evaluated for postprocedure complications, rebleeding events, or complaints of worsening abdominal pain within 30 days following the procedure. Height (m) and weight (kg) data were obtained through electronic medical records. Postprocedure weights were obtained based on availability in subsequent clinic notes. Changes in weight as well as changes in body mass index (BMI) as measured by weight in kilograms (kg) per height in meters squared (m2) were calculated.
Statistical analysis
Continuous variables were summarized using means (±SD) or median (25%, 75%) and categorical variables using frequencies (percent). Pre and post BMI and weight were compared using a paired t-test. All tests were two-sided and conducted in SAS v9.4. A P value of 0.05 was considered statistically significant.
Results
Patients ranged from 33–92 years of age. The median preprocedure weight and BMI of the patients evaluated in this study were 93.4 kg and 29.9 kg/m2, respectively. Additional demographics and clinical characteristics of all patients are displayed in Table 1.
Table 1.
All LGA embolization patients (n=39) | LGA embolization patients with pre- and postprocedure weights (n=21) | |
---|---|---|
Age (years), median (range) | 57 (41–92) | 57 (33–92) |
| ||
Age (years), mean (SD) | 60.5 (14.7) | 56.3 (15.6) |
| ||
Gender, n (%) | ||
Male | 27 (69.2) | 12 (57.1) |
Female | 12 (30.8) | 9 (42.9) |
| ||
Race, n (%) | ||
Caucasian | 30 (76.9) | 17 (81.0) |
African American | 9 (23.1) | 4 (18.2) |
| ||
Follow-up time intervals, n (%) | ||
1–3 months | 4 (19.1) | |
4–6 months | 3 (14.3) | |
7–12 months | 4 (19.1) | |
12–24 months | 5 (23.8) | |
24+ months | 5 (23.8) |
LGA, left gastric artery; SD, standard deviation.
The method of arterial embolization included coils (n=6), gelfoam slurry (n=19), PVA particles ranging from 100–300 μm, 300–500 μm, and 500–700 μm (n=5), and combination embolics (n=9). Additional data regarding embolization techniques are displayed in Table 2.
Table 2.
All LGA embolization patients (n=39) | LGA embolization patients with pre- and postprocedure weights (n=21) | |
---|---|---|
Embolic used, n (%) | ||
Coil | 6 (15.4) | 4 (19.1) |
Gelfoam | 19 (48.7) | 11 (52.4) |
PVA | 5 (12.8) | 0 (0.0) |
PVA + coils | 3 (7.7) | 2 (9.5) |
Gelfoam + coils | 6 (15.4) | 4 (19.5) |
| ||
Embolic status, n (%) | ||
Permanent | 19 (48.7) | 10 (47.6) |
Temporary | 20 (51.3) | 11 (52.4) |
| ||
Vessels embolized, n (%) | ||
LGA | 34 (87.1) | 18 (85.7) |
LGA + splenic artery | 3 (7.7) | 2 (9.5) |
LGA + right gastroepiploic artery | 1 (2.6) | 1 (4.8) |
LGA + phrenic artery | 1 (2.6) | 0 (0.0) |
LGA, left gastric artery; PVA, polyvinyl alcohol. |
Following LGA embolization, median weight decreased from 93.4 kg to 77.1 kg and BMI decreased from 29.9 kg/m2 to 27.1 kg/m2 (n=21; P = 0.045) over a median time of 12 months (range, 2–72 months).
Pre- and postprocedure endoscopies were performed in 6 of the 39 patients. Of these patients, 2 had worsening ulcers following LGA embolization and 4 had stable or no abnormal findings. The majority of patients who did not have endoscopies were lost to follow-up or had no documentation of GI associated complications. One patient with a complex medical history requiring lifelong anticoagulation for multiple prosthetic heart valves was found to have complications with ischemic segment of jejunum requiring surgical intervention two days postprocedure. Given the site of ischemia, these findings are favored to represent thrombus related to holding anticoagulation rather than being related to LGA embolization. Another patient died within 60 days following the procedure from preexisting complex medical complications related to heart failure, again not directly related to LGA embolization.
Discussion
Studying a retrospective cohort of patients for weight loss provides an opportunity to assess patients without confounding factors such as intrinsic motivation, diet modification or other weight loss strategies. Gunn et al. (16) reported on a small retrospective cohort (n=22) where they compared patients with LGA embolization for GI bleeding against those that had embolization elsewhere (extra-gastric) for GI bleeding. The group that received LGA embolization demonstrated significant nonintended weight loss (7.3% at 3 months) compared with the control group; however, this study was limited due to the inclusion of patients with a cancer diagnosis which introduces a significant confounding variable. A subsequent retrospective study was performed by Anton et al. (17), where they investigated patients with LGA embolization for GI bleeding in patients without cancer diagnoses (n=10). Their study also demonstrated significant weight loss (9.8% at one month and 11.7% at 4 months) compared with baseline weight.
Our study involves a retrospective cohort that does not include patients with a cancer diagnosis at the time of procedure. The patients in the study experienced a median 16.3 kg (17.4%) of unintended weight loss following LGA embolization for GI bleeding, over a median time interval of 12 months. Our cohort had an average BMI of 29.9 kg/m2 which classifies them as overweight, whereas current clinical trials have been performed on morbidly obese individuals. This finding may provide insight into the effectiveness of LGA embolization on individuals who are not morbidly obese; however, clinical trials are needed to validate this claim.
Current clinical trials implement GI prophylaxis regimens both pre and post LGA embolization. While many of the patients received proton pump inhibitors on admission for GI bleeding, there was no rigorous standardized GI prophylaxis regimen before or after the procedure, as defined in clinical trials. Current clinical trials administer protracted dosing of sucralfate and proton-pump inhibitors to reduce the rate of gastric ulceration after LGA embolization (5). Our findings suggest that even in the absence of standardized preprocedure GI prophylaxis, the majority of patients appear to tolerate the procedure with minimal complications following LGA embolization. Of note, the authors do not advocate eliminating GI prophylaxis from clinical trials as the embolization techniques are quite different.
Our study has several limitations. Most notably, without a large sample size and low statistical power, our findings regarding association between weight change related to LGA embolization and various endpoints such as embolic type and follow-up time interval serve as observations, and true efficacy of this procedure cannot be inferred from this observation. Furthermore, the retrospective nature of this study introduces inevitable selection bias. Second, is an inability to determine whether the mechanism of weight loss was due to appetite suppression or other confounding factors such as weight loss due to illness, as is usually present in a hospitalized patient. Third, is lack of a control group which will make it difficult to assess the net effect of this LGA embolization. Fourth, is the nonstandardized embolization technique with no consistent embolic endpoint.
In conclusion, our findings suggest that LGA embolization for GI bleeding results in weight loss as an unexpected side effect. This procedure appears to be well tolerated and may be a promising technique for modulating weight in obese adults. However, further larger and prospective studies are needed to confirm these preliminary findings.
Main points.
Left gastric artery embolization was evaluated retrospectively for unintended weight loss.
Left gastric artery embolization results in unintended weight loss.
Left gastric artery embolization appears to be well tolerated in patients.
Footnotes
Conflict of interest disclosure
Dr Akinwande is supported by an institutional (Washington University School of Medicine) and departmental (Mallinckrodt Institute of Radiology) grant. Siemens Healthcare, Merit Medical, BTG, Clear Guide Medical, and NIH NIH/NIBIB R01EB017615 provide grant support for Dr. Weiss. Other authors have nothing to disclose.
References
- 1.Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA. 2002;288:1723–1727. doi: 10.1001/jama.288.14.1723. https://doi.org/10.1001/jama.288.14.1723. [DOI] [PubMed] [Google Scholar]
- 2.Mitchell NS, Catenacci VA, Wyatt HR, Hill JO. Obesity: overview of an epidemic. Psychiatr Clin North Am. 2011;34:717–732. doi: 10.1016/j.psc.2011.08.005. https://doi.org/10.1016/j.psc.2011.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cawley J, Meyerhoefer C. The medical care costs of obesity: an instrumental variables approach. J Health Econ. 2012;31:219–230. doi: 10.1016/j.jhealeco.2011.10.003. https://doi.org/10.1016/j.jhealeco.2011.10.003. [DOI] [PubMed] [Google Scholar]
- 4.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. doi: 10.1148/radiol.12120242. https://doi.org/10.1148/radiol.12120242. [DOI] [PubMed] [Google Scholar]
- 5.Weiss CR, Gunn AJ, Kim CY, Paxton BE, Kraitchman DL, Arepally A. Bariatric embolization of the gastric arteries for the treatment of obesity. J Vasc Interv Radiol. 2015;26:613–624. doi: 10.1016/j.jvir.2015.01.017. https://doi.org/10.1016/j.jvir.2015.01.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K. Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature. 1999;402:656–660. doi: 10.1038/45230. https://doi.org/10.1038/45230. [DOI] [PubMed] [Google Scholar]
- 7.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. doi: 10.1016/j.jvir.2013.09.016. https://doi.org/10.1016/j.jvir.2013.09.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Arepally A, Barnett BP, Montgomery E, Patel TH. Catheter-directed gastric artery chemical embolization for modulation of systemic ghrelin levels in a porcine model: initial experience. Radiology. 2007;244:138–143. doi: 10.1148/radiol.2441060790. https://doi.org/10.1148/radiol.2441060790. [DOI] [PubMed] [Google Scholar]
- 9.Arepally A, Barnett BP, Patel TH, et al. Catheter-directed gastric artery chemical embolization suppresses systemic ghrelin levels in porcine model. Radiology. 2008;249:127–133. doi: 10.1148/radiol.2491071232. https://doi.org/10.1148/radiol.2491071232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bawudun D, Xing Y, Liu WY, et al. Ghrelin suppression and fat loss after left gastric artery embolization in canine model. Cardiovasc Interv Radiol. 2012;35:1460–1466. doi: 10.1007/s00270-012-0362-8. https://doi.org/10.1007/s00270-012-0362-8. [DOI] [PubMed] [Google Scholar]
- 11.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. doi: 10.1007/s11695-014-1535-0. https://doi.org/10.1007/s11695-014-1535-0. [DOI] [PubMed] [Google Scholar]
- 12.Kipshidze N, Archvadze A, Bertog S, Leon MB, Sievert H. Endovascular bariatrics: first in humans study of gastric artery embolization for weight loss. JACC Cardiovasc Interv. 2015;8:1641–1644. doi: 10.1016/j.jcin.2015.07.016. https://doi.org/10.1016/j.jcin.2015.07.016. [DOI] [PubMed] [Google Scholar]
- 13.Pasciak AS, Bourgeois AC, Paxton BE, et al. Bariatric radioembolization: a pilot study on technical feasibility and safety in a porcine model. J Vasc Interv Radiol JVIR. 2016;27:1509–1517. doi: 10.1016/j.jvir.2016.05.020. https://doi.org/10.1016/j.jvir.2016.05.020. [DOI] [PubMed] [Google Scholar]
- 14.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 JVIR. 2016;27:1502–1508. doi: 10.1016/j.jvir.2016.07.010. https://doi.org/10.1016/j.jvir.2016.07.010. [DOI] [PubMed] [Google Scholar]
- 15.Weiss CR, Akinwande O, Paudel K, et al. Clinical safety of bariatric arterial embolization: preliminary results of the BEAT obesity trial. Radiology. 2017;283:598–608. doi: 10.1148/radiol.2016160914. https://doi.org/10.1148/radiol.2016160914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gunn AJ, Oklu R. A preliminary observation of weight loss following left gastric artery embolization in humans. J Obes. 2014;2014:185349. doi: 10.1155/2014/185349. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Anton K, Rahman T, Bhanushali A, Nadal L, Pierce G, Patel AA. Weight loss following left gastric artery embolization in a human population without malignancy: a retrospective review. Obes Weight Loss Ther. 2015;5 [Google Scholar]