VESS Case Reports Pancreaticoduodenal Artery Aneurysms Secondary to Median Arcuate Ligament Syndrome May Not Need Celiac Artery Revascularization or Ligament Release Michael D. Sgroi,1,2 Nii-Kabu Kabutey,1,2 Mayil Krishnam,1,2 and Roy M. Fujitani,1,2 Orange, California
Introduction: Median arcuate ligament syndrome (MALS) is a rare disorder defined by compression and narrowing of the celiac artery by the median arcuate ligament. The increased blood flow through the pancreaticoduodenal arcade can lead to the aneurysmal formation within the vessel. We report 3 cases of pancreaticoduodenal arterial aneurysms (PDAAs) in patients with MALS whose aneurysms were occluded, but celiac artery revascularization was not performed. Methods: Case 1: Asymptomatic 61-year-old female with no past medical history was referred to vascular surgery for evaluation of a PDAA incidentally found on computed tomography (CT) scan. The patient was taken for laparoscopic division of the median arcuate ligament; however, the release was incomplete. This was followed by endovascular coil embolization of the PDAA without celiac revascularization. The patient tolerated the procedure well with no complications and the 1-year follow-up shows no signs of aneurysm recurrence. Case 2: A 61-year-old male found to have an incidental PDAA on CT scan. The patient was taken for coil embolization without median arcuate ligament release. At the 1-year follow-up, the patient continues to be asymptomatic with no recurrence. Case 3: A 56-year-old male presented with a ruptured PDAA. He was taken immediately for coil embolization of the ruptured aneurysm. Postoperatively, the patient was identified to have MALS on CT scan. Because of his asymptomatic history and benign physical examination before the rupture, he was not taken for a ligament release or celiac revascularization. He continues to be asymptomatic at his follow-up. Results: PDAAs secondary to MALS are very rare and most commonly diagnosed at the time of rupture, which has a mortality rate that reaches approximately 30%, making early identification and treatment necessary. Standard treatment would include exclusion of the aneurysm followed by celiac revascularization; however, these 3 cases identify an alternative approach to the standard treatment. Conclusion: Celiac revascularization may not be necessary in the asymptomatic patient with a PDAA who has close follow-up and serial imaging.
Median arcuate ligament syndrome (MALS), also known as celiac artery compression syndrome or Dunbar syndrome, is a rare cause of abdominal
pain and weight loss. This diagnosis of exclusion classically includes postprandial abdominal pain, weight loss, and an abdominal bruit. Celiac, superior
1 Division of Vascular Surgery, Department of Surgery, University of California, Irvine Medical Center, Orange, CA. 2 Department of Radiology, University of California, Irvine Medical Center, Orange, CA.
Ann Vasc Surg 2015; 29: 122.e1e122.e7 http://dx.doi.org/10.1016/j.avsg.2014.05.020 Published by Elsevier Inc. Manuscript received: March 5, 2014; manuscript accepted: May 18, 2014; published online: June 12, 2014.
Correspondence to: Michael D. Sgroi, MD, Division of Vascular and Endovascular Surgery, Department of Surgery, University of California, Irvine Medical Center, 333 The City Blvd. West, Ste 1600, Orange, CA 92868, USA; E-mail: [email protected]
122.e1
122.e2 Case reports
mesenteric, and pancreaticoduodenal arterial (PDA) aneurysms have also been shown to be associated with MALS. PDA aneurysms are associated with a high mortality rate if rupture occurs, particularly because unlike other splanchnic arteries, rupture occurs irrespective of aneurysm size.1 Multiple imaging techniques have been used to delineate significant celiac compression, with real-time cine magnetic resonance angiography (MRA) imaging being the most recent advancement. Historically, patients who have presented with MALS and PDA aneurysms have been treated with exclusion of the aneurysm, median arcuate ligament release, and celiac artery revascularization. We present 3 patients without chronic abdominal pain who were found to have PDA aneurysms that were treated with aneurysm exclusion without celiac artery revascularization or ligament release.
CASE REPORT 1 The patient was a 61-year-old Caucasian female with no significant past medical history who presented to the vascular surgery clinic to be evaluated for an incidental inferior PDAA. At her initial presentation, the patient was completely asymptomatic. She denied any postprandial abdominal pain, weight loss, nausea, or vomiting. Her physical examination was benign and an abdominal bruit was not appreciated. The patient was found to have microhematuria during a routine visit with her primary care physician, and because of a family history of bladder cancer, a computed tomography (CT) scan of the abdomen and pelvis was obtained. Evaluation of the CT scan demonstrated a 2.0-cm inferior PDA aneurysm that originated 4.7 cm from the ostium of the superior mesenteric artery (SMA). Dynamic magnetic resonance imaging (MRI) displayed significant stenosis at the celiac axis throughout the respiratory cycle. There was increased compression during the expiratory phase, confirming MALS. Because of the size of the aneurysm and the known risk of spontaneous rupture, the decision was made to perform a 2-staged procedure. The patient was initially taken to the operating room for laparoscopic division of the median arcuate ligament. The lesser sac was entered and the right crus was identified. The pre-aortic fascia was dissected down to the abdominal aorta, and then distally to the celiac artery origin. Because of extensive thickening of the ligament, complete circumferential dissection was unable to be performed safely using a laparoscopic approach. The patient was then brought back 2 weeks later for selective SMA aortography with exclusion of the PDA aneurysm. Selective aortography delineated the SMA, inferior pancreaticoduodenal artery, as well as the aneurysm. A 6-French Raabe sheath was delivered over a stiff 0.035-inch glidewire into the SMA. This was followed by an multipurpose A catheter
Annals of Vascular Surgery
(MPA). A 0.014-inch balanced middleweight wire was then introduced through the catheter and into the posterior inferior pancreaticoduodenal artery. Once the 6French Raabe sheath was advanced into proper position, the decision was made to place an Atrium iCast 6-mm 16-mm covered stent to exclude the aneurysm without the need for coiling. After multiple unsuccessful attempts were made to guide the stent graft into the appropriate position, the stent dislodged from the balloon in the inferior pancreaticoduodenal artery. Attempts were made to recanalize the stent with a 0.014-inch Pilot guidewire, but were unsuccessful. A wire was manipulated past the stent graft and the aneurysm was directly cannulated. A microcatheter was placed over the wire and coil embolization was performed using a Cook (Cook Medical; Bloomington, IN) MWCE-185e10/4Tornado (x5); Cook MWCE 6/2-Tornado (x3); and Cook MWCE-185-8/4-Tornado (x3). Following primary coiling, a Palmaz 5 mm 60 mm stent was placed over the aneurysm origin to entrap the coils within the aneurysm as well as compress the dislodged stent against the arterial wall. Repeat angiography showed patency of the inferior pancreaticoduodenal artery and exclusion of the aneurysm. Selective celiac angiography displayed a thrombosed celiac axis. Endovascular recannulation was attempted, but unsuccessful. After review of the imaging, it was determined that the patient had adequate collateralization as well as retrograde flow through the celiac axis; therefore, close observation was deemed a reasonable alternative to an open revascularization. One-year follow-up with serial ultrasound and CT imaging has demonstrated inferior pancreaticoduodenal artery patency and aneurysm exclusion.
CASE REPORT 2 A 61-year-old male presented to the emergency department with right-sided flank pain. The patient had a past medical history of nephrolithiasis. He denied any nausea, vomiting, diarrhea, jaundice, or weight loss. A CT scan of the abdomen and pelvis was ordered that demonstrated nephrolithiasis and an incidental calcification near the uncinate process of the pancreas. Because of concern for a malignancy, the patient was referred to the HepatoBiliary and Pancreas service for further evaluation. The patient had a mild elevation in his carbohydrate antigen 19-9 and alpha fetoprotein, and an endoscopic retrograde cholangiopancreatography with endoscopic ultrasound was ordered for further evaluation. Endoscopic ultrasound was able to identify the calcified lesion, and a percutaneous biopsy was performed. The biopsy returned gross blood, which instigated the ordering of a CT angiogram (CTA). It was determined on the repeat imaging that there was a calcified pancreatic artery aneurysm within the uncinate process, as well as stenosis at the celiac artery origin. The decision was made to take the patient to the angiography suite for coil embolization and possible median arcuate ligament release. After achieving brachial artery access, selective SMA aortography using an MPA
Vol. 29, No. 1, January 2015
diagnostic catheter (AngioDynamics, Inc., Latham, NY) was performed showing the dilatation of the inferior pancreaticoduodenal artery. The aneurysm was occluded using Cook MWCE-185-6/2 (x4) Toranado coils (Cook Medical). Postembolization angiography showed a patent artery and no flow within the aneurysm sac. The patient tolerated the coiling well with no complications. It was determined based on the large collateral vessels and retrograde flow through the celiac axis that a median arcuate ligament release and celiac artery revascularization would not be required. The patient recently had a 1-year followup CTA that demonstrated an occluded PDA aneurysm with patent surrounding vasculature and no further signs of aneurysm filling. The patient has remained asymptomatic despite not having a median arcuate ligament release or celiac artery revascularization.
CASE REPORT 3 A 52-year-old male with no significant medical history or chronic abdominal symptoms presented to the emergency department with an acute onset of epigastric pain. CT scan of the abdomen and pelvis showed that the patient had a PDAA with active extravasation. Attempted intervention was performed at the outside institution without success and therefore the patient was transferred to our facility. Because of an acute drop in the patient’s hemoglobin, he was transfused 1 U of packed red blood cells. He arrived at our institution in stable condition and taken urgently to the hybrid angiography suite. Selective SMA angiography from a brachial artery approach was performed identifying the PDAA with active extravasation. At the outside hospital, a femoral approach was attempted; however, because of the angulation at the SMA origin, they were unable to cannulate the vessel. A 5-French Ansel sheath was placed into the SMA orifice followed by a Renegade microcatheter (Boston Scientific; Natick, MA) that was used to cannulate the aneurysm. Coil embolization of the aneurysm, outflow, and inflow vessels were performed again using Cook Tornado coils (Cook Medical). Digital subtraction angiography demonstrated complete occlusion of the aneurysm and retrograde flow to the celiac axis. Postoperatively, a CT angiogram showed that the celiac axis demonstrated a classic upward hooking appearance consistent with MALS. Because of the fact that the patient denied any history of chronic abdominal pain, and continued to be symptom free, he was discharged home without celiac artery revascularization or surgical release of the median arcuate ligament. At 1-month follow-up, the patient continued to be without any abdominal pain.
DISCUSSION The anatomic structure of the median arcuate ligament was first described in 1917 by Lipshutz.2 It was almost 50 years later that the syndrome was described by Harjola and Dunbar.2,3 It has been
Case reports 122.e3
theorized that the reason for the delay between identification of the ligament and the syndrome is secondary to the fact that most individuals never experience symptoms. It is estimated that in 10e 24% of healthy, asymptomatic individuals the median arcuate ligament crosses anterior to the celiac artery, causing some degree of compression4,5; however, it is only severe enough to cause symptoms in approximately 1% of the population.5 Throughout the respiratory cycle, the aorta moves in the cephalad and caudal direction. With inspiration, the expansion of the chest leads to the movement of the aorta in the caudal direction. Conversely, during expiration, the aorta rises in the cephalad direction leading to increased compression by the median arcuate ligament. PDA aneurysms are a rare anomaly and are most commonly diagnosed at the time of rupture.6 When ruptured, mortality rates reach approximately 30%, making early identification and urgent treatment necessary.7 With compression of the celiac axis by the median arcuate ligament, there is decreased blood flow to the foregut. To compensate for this hypoperfusion, an increase in blood flow is directed through the SMA and its branches to the end organs of the foregut.1 Angiography has demonstrated robust arterial collateralization system in patients with celiac artery compression assisting with foregut perfusion. The increased blood volume propagating through these small visceral arteries leads to an increase in pressure as well as wall tension. Increased endoluminal pressure can eventually damage the vessel wall, leading to aneurysmal formation.8 This physiologic mechanism was originally proposed by Sutton and Lawton in 1973.9 Other causes of endoluminal damage may include atherosclerosis, trauma, fibromuscluar dysplasia, and pancreatitis.10 Both MALS and PDA aneurysms can be detected during conventional angiography (Fig. 1); however, with the advent of CTA and MRA, celiac compression can easily be identified and visceral artery aneurysms less than 1 cm in diameter are routinely detected.11 In regards to PDA aneurysms, most will be diagnosed either at the time of rupture or incidentally during a CT/MRI for some other indication. Attention should be taken to assure that 3-mm slices and iodinated contrast are ordered in addition to 3dimensional reconstruction to receive best visualization for preoperative planning.12 Imaging software improvements, as well as the technological hardware advancements of 16- and 64-slice scanners, have resulted in the ability to obtain high quality images of the abdominal aorta and its branches. Typically, the sagittal plane is optimal for visualizing the proximal portion of the celiac axis, and in many
122.e4 Case reports
Annals of Vascular Surgery
Fig. 1. Selective angiography demonstrating the visceral aneurysm as a branch of the superior mesenteric artery.
Fig. 2. Still image from dynamic magnetic resonance imaging demonstrating the compression of the celiac artery during expiration.
cases, 3-dimensional imaging also allows for identification of the median arcuate ligament.4 Dynamic MRA can aid in the diagnosis of MALS. This allows for observation of the celiac axis throughout the respiratory cycle, displaying constant stenosis of the artery, with augmentation of compression during the expiratory phase (Fig. 2). Focal narrowing with a hook appearance (Fig. 3) is a characteristic finding in patients with MALS, helping distinguish it from other forms of stenosis, such as atherosclerosis.11 Recommended treatment for patients with symptomatic MALS with a PDA aneurysm is 2-fold. Concomitant revascularization of the celiac axis and its branches after release of the median arcuate ligament has been the standard recommended treatment, particularly in symptomatic patients.13e15 Treatment of celiac stenosis in the face of a PDA aneurysm still remains a controversial subject. Reilly et al.16 was one of the first studies to identify that surgical decompression followed by revascularization had a significantly greater long-term result in relieving symptoms (76% vs. 56%, respectively). Additionally, the PDA aneurysm needs to be treated irrespective of the size because of the risk of
Fig. 3. MALS is often identified on sagittal imaging with the celiac artery in a classic ‘‘hook’’ appearance. The arrow identifies the ligament compressing the celiac axis.
rupture.17 The decision in our first case was to perform a 2-staged procedure, with the understanding that both would need to be done in a timely matter because of the high risk of visceral aneurysm rupture. We initially tried to perform a laparoscopic
Vol. 29, No. 1, January 2015
Case reports 122.e5
Table I. Review of the literature demonstrating other reported cases of PDAA treatment without celiac revascularization Author
Patients (n)
Treatment of aneurysm
Outcome/follow-up
1-year follow-up with no recurrence 4-year follow-up with no recurrence 0% mortality with mean follow-up of 37 months 0% mortality 1/5 (20%) stent thrombosis
Flood et al. (2013)
2
2/2 coil embolization
Nishiyama et al. (2013)
3
3/3 surgical resection
11
Ikenda et al. (2013)
Hildebrand et al. (2007) Ikeda et al. (2007)
2 3
5/11 coil embolization 5/11 coil embolization and celiac stenting 1/11 isolation only 2/2 surgical resection 3/3 coil embolization
Sugiyama and Takehara (2006) Bellosta et al. (2005) Ducasse et al. (2004) Imamura et al. (1998) Chiang et al. (1994) Grech et al. (1988)
2 1 1 1 1 1
2/2 observation Surgical resection Coil embolization Aorto-hepatic bypass Surgical resection N/A
N/A Short-term (6-month) follow-up showed no recurrence N/A 9 days no recurrence 4 months no recurrence N/A 4 months N/A
N/A, not applicable.
median arcuate ligament release, but because of anatomical constraints, a full release could not be performed safely. Decompression of the celiac artery involved dissection of the fibrous tissue on the anterior portion of the aorta down to the celiac axis. Identification of the proximal celiac axis is important to confirm that the median arcuate ligament has been divided. While open surgical release has been the standard method, laparoscopic release has been reported.18 There is an increased risk of arterial injury and major hemorrhage; however, studies have shown that laparoscopic treatment resulted in equivalent morbidity and mortality, as well as a shortened hospital stay and no difference in recurrence.19,20 It should be noted, however, that care must be taken to release all the thick fibrous tissues that make up the median arcuate ligament. Failure to fully release the ligament will result in continued external compression on the celiac stump, which will not alleviate the patient’s symptoms. Additionally, some authors have theorized that patient’s symptoms are secondary not only to arterial compression, but also overstimulation of the celiac sympathetic plexus.4,21e25 It is important that thorough external arteriolysis be performed circumferentially around the celiac trunk, to lyse these sympathetic nerves. Median arcuate ligament release may not be necessary if the patient is not experiencing chronic abdominal symptoms. As can be seen with our cases, all patients were free of chronic abdominal pain. In the first patient, a release was attempted, but
determined to be inadequate. The celiac axis was completely occluded, making the median arcuate ligament release irrelevant, as it would not have changed the antegrade blood flow through the celiac axis. The second and third patient also received coil embolization only without ligament release or revascularization. Angiography demonstrated that the foregut was receiving enough perfusion through retrograde blood flow and robust collateralization. It is our theory that the aneurysm likely formed before the evolving collateral vascular system within the foregut. Now that these patients have an extensive pancreaticoduodenal arcade to endure the increased blood flow, recurrent aneurysm formation is unlikely. In all cases, endovascular repair was performed to exclude the PDA aneurysm. Open surgical repair of mesenteric aneurysms can include a variety of surgical approaches; however, mortality rates for open repair have been reported to be as high as 19%.26 In patients who receive open repairs for visceral occlusive disease, complications such as bowel resections, infections, sepsis, acute renal failure, and death have been described.27e30 The endovascular approach allows for exclusion of the aneurysm with a covered stent graft or coil embolization. The anatomical location of the aneurysm may make cannulation challenging, but 0.014inch platforms and modern endovascular techniques can preserve the collateral blood flow successfully and occlude mesenteric aneurysms. Endovascular stenting or coil embolization is less
122.e6 Case reports
invasive than open surgery and can be safely applied with minimal complications. An endovascular approach allows one to selectively embolize the aneurysm, sparing other vessels.1 Landing zones will be crucial to identify before deployment of a stent as care must be taken not to occlude critical blood vessels. As seen with case 1, the anatomy may provide challenges that do not allow for adequate stent deployment. With improving endovascular techniques, percutaneous angioplasty and stenting of the celiac artery axis have been described and are performed by skilled endovascular interventionalists. Care should be taken with this approach, however, because angioplasty and stenting will not eliminate the external compression from the ligament. It is recommended that endovascular management not be performed if the median arcuate ligament is not released. Celiac artery revascularization may not be necessary in the asymptomatic patient. Though it seems logical that resolving the celiac stenosis/occlusion would decrease recurrence of aneurysms by improving celiac artery antegrade flow. To date, there are no documented cases of aneurysm recurrence within asymptomatic patients who did not receive ligament release or revascularization (Table I). PDA aneurysms secondary to MALS are a rare finding, but once identified should be treated urgently. This report demonstrates that celiac revascularization after treatment of the aneurysm may not be necessary, and the exclusion of the aneurysm may be the only necessary treatment. Short-term results demonstrate no recurrence of the aneurysm and the patients continue to be asymptomatic. Further studies and long-term follow-up are still necessary to further evaluate if new aneurysm formation will occur. It is recommended that follow-up imaging be performed at 6 months postoperatively followed by yearly imaging.
REFERENCES 1. Kallamadi R, Demoya MA, Kalva SP. Inferior pancreaticoduodenal artery aneurysms in association with celiac stenosis/occlusion. Semin Intervent Radiol 2009;26: 215e23. 2. Lipshutz B. A composite study of the celiac axis artery. Ann Surg 1917;65:159e69. 3. Harjola PT. A rare obstruction of the celiac artery: report of a case. Ann Chir Cynaecol Fenn 1963;52:547e50. 4. Horton KM, Talamini MA, Fishman EK. Median arcuate ligament syndrome: evaluation with CT angiography. Radiographics 2005;25:1177e82.
Annals of Vascular Surgery
5. Lindner HH, Kemprud E. A clinicoanatomical study of the arcuate ligament of the diaphragm. Arch Surg 1971;103: 600e5. 6. Santiagu S, Gananadha S, Harrington TJ, et al. Direct percutaneous puncture embolization of a peripancreatic pseudoaneurysm presenting with haemosuccus pancreaticus. J Med Imaging Radiat Oncol 2008;52:370e3. 7. Moore E, Matthews MR, Minion DJ, et al. Surgical management of peripancreatic arterial aneurysms. J Vasc Surg 2004;40:247e53. 8. Kobayashi T, Uenoyama S, Isogai S. Successful transcatheter arterial embolization of an inferior pancreaticoduodenal artery aneurysm associated with celiac axis stenosis. J Gastroenterol Hepatol 2004;19:599e601. 9. Sutton D, Lawton G. Coeliac stenosis or occlusion with aneurysm of the collateral supply. Clin Radiol 1973;24: 49e53. 10. Verta MJ Jr, Dean RH, Yao JS, et al. Pancreaticoduodenal arterial aneurysms. Ann Surg 1977;186:111e4. 11. Klonaris C, Psathas E, Katsargyris A, et al. Multiple aneurysms of the inferior pancreaticoduodenal artery: a rare complication of acute pancreatitis. Case Rep Vasc Med 2013;2013:621350. 12. Horiguchi A, Ishihara S, Ito M, et al. Multislice CT study of pancreatic head arterial dominance. J Hepatobiliary Pancreat Surg 2008;15:322e6. 13. Grotemeyer D, Duran M, Iskandar F, et al. Median arcuate ligament syndrome: vascular surgical therapy and follow-up of 18 patients. Langenbecks Arch Surg 2009;394:1085e92. 14. Takach TJ, Livesay JJ, Reul GJ Jr, et al. Celiac compression syndrome: tailored therapy based on intraoperative findings. J Am Coll Surg 1996;183:606e10. 15. Palmer OP, Tedesco M, Casey K, et al. Hybrid treatment of celiac artery compression (median arcuate ligament) syndrome. Dig Dis Sci 2012;57:1782e5. 16. Reilly LM, Ammar AD, Stoney RJ, et al. Late results following operative repair for celiac artery compression syndrome. J Vasc Surg 1985;2:79e91. 17. Nishiyama A, Hoshina K, Hosaka A, et al. Treatment strategies for a pancreaticoduodenal artery aneurysm with or without a celiac trunk occlusive lesion. Ann Vasc Dis 2013;6:725e9. 18. Duffy AJ, Panait L, Eisenberg D, et al. Management of median arcuate ligament syndrome: a new paradigm. Ann Vasc Surg 2009;23:778e84. 19. Tulloch AW, Jimenez JC, Lawrence PF, et al. Laparoscopic versus open celiac ganglionectomy in patients with median arcuate ligament syndrome. J Vasc Surg 2010;52: 1283e9. 20. Jimenez JC, Harlander-Locke M, Dutson EP. Open and laparoscopic treatment of median arcuate ligament syndrome. J Vasc Surg 2012;56:869e73. 21. Szillagyi DE, Rian RL, Elliott JP, et al. The celiac artery compression: does it exist? Surgery 1972;72:849e63. 22. Bech FR. Coeliac artery compression syndromes. Surg Clin North Am 1997;77:409e24. 23. Delis T, Gloviczki P, Altuwaijri M, et al. Median arcuate ligament syndrome: open celiac artery reconstruction and ligament division after endovascular failure. J Vasc Surg 2007;46:799e802. 24. Desmond CP, Roberts SK. Exercise-related abdominal pain as a manifestation of the median arcuate ligament syndrome. Scand J Gastroenterol 2004;39:1310e3. 25. Akatsu T, Hayashi S, Yamane T, et al. Emergency embolization of a ruptured pancreaticoduodenal artery aneurysm
Vol. 29, No. 1, January 2015
associated with the median arcuate ligament syndrome. J Gastroenterol Hepatol 2004;19:482e3. 26. Coll DP, Ierardi R, Kerstein MD, et al. Aneurysms of the pancreaticoduodenal arteries: a change in management. Ann Vasc Surg 1998;12:286e91. 27. Arthurs ZM, Titus J, Bannazadeh M, et al. A comparison of endovascular revascularization with traditional therapy for the treatment of acute mesenteric ischemia. J Vasc Surg 2011;53:698e704. discussion, 04e5.
Case reports 122.e7
28. Block TA, Acosta S, Bj€ orck M. Endovascular and open surgery for acute occlusion of the superior mesenteric artery. J Vasc Surg 2010;52:959e66. 29. Barakate MS, Cappe I, Curtin A, et al. Management of acute superior mesenteric artery occlusion. ANZ J Surg 2002;72: 25e9. 30. Ryer EJ, Kalra M, Oderich GS, et al. Revascularization for acute mesenteric ischemia. J Vasc Surg 2012;55: 1682e9.
Introduction: Median arcuate ligament syndrome (MALS) is a rare disorder defined by compression and narrowing of the celiac artery by the median arcuate ligament. The increased blood flow through the pancreaticoduodenal arcade can lead to the aneurysmal formation within the vessel. We report 3 cases of pancreaticoduodenal arterial aneurysms (PDAAs) in patients with MALS whose aneurysms were occluded, but celiac artery revascularization was not performed. Methods: Case 1: Asymptomatic 61-year-old female with no past medical history was referred to vascular surgery for evaluation of a PDAA incidentally found on computed tomography (CT) scan. The patient was taken for laparoscopic division of the median arcuate ligament; however, the release was incomplete. This was followed by endovascular coil embolization of the PDAA without celiac revascularization. The patient tolerated the procedure well with no complications and the 1-year follow-up shows no signs of aneurysm recurrence. Case 2: A 61-year-old male found to have an incidental PDAA on CT scan. The patient was taken for coil embolization without median arcuate ligament release. At the 1-year follow-up, the patient continues to be asymptomatic with no recurrence. Case 3: A 56-year-old male presented with a ruptured PDAA. He was taken immediately for coil embolization of the ruptured aneurysm. Postoperatively, the patient was identified to have MALS on CT scan. Because of his asymptomatic history and benign physical examination before the rupture, he was not taken for a ligament release or celiac revascularization. He continues to be asymptomatic at his follow-up. Results: PDAAs secondary to MALS are very rare and most commonly diagnosed at the time of rupture, which has a mortality rate that reaches approximately 30%, making early identification and treatment necessary. Standard treatment would include exclusion of the aneurysm followed by celiac revascularization; however, these 3 cases identify an alternative approach to the standard treatment. Conclusion: Celiac revascularization may not be necessary in the asymptomatic patient with a PDAA who has close follow-up and serial imaging.
Median arcuate ligament syndrome (MALS), also known as celiac artery compression syndrome or Dunbar syndrome, is a rare cause of abdominal
pain and weight loss. This diagnosis of exclusion classically includes postprandial abdominal pain, weight loss, and an abdominal bruit. Celiac, superior
1 Division of Vascular Surgery, Department of Surgery, University of California, Irvine Medical Center, Orange, CA. 2 Department of Radiology, University of California, Irvine Medical Center, Orange, CA.
Ann Vasc Surg 2015; 29: 122.e1e122.e7 http://dx.doi.org/10.1016/j.avsg.2014.05.020 Published by Elsevier Inc. Manuscript received: March 5, 2014; manuscript accepted: May 18, 2014; published online: June 12, 2014.
Correspondence to: Michael D. Sgroi, MD, Division of Vascular and Endovascular Surgery, Department of Surgery, University of California, Irvine Medical Center, 333 The City Blvd. West, Ste 1600, Orange, CA 92868, USA; E-mail: [email protected]
122.e1
122.e2 Case reports
mesenteric, and pancreaticoduodenal arterial (PDA) aneurysms have also been shown to be associated with MALS. PDA aneurysms are associated with a high mortality rate if rupture occurs, particularly because unlike other splanchnic arteries, rupture occurs irrespective of aneurysm size.1 Multiple imaging techniques have been used to delineate significant celiac compression, with real-time cine magnetic resonance angiography (MRA) imaging being the most recent advancement. Historically, patients who have presented with MALS and PDA aneurysms have been treated with exclusion of the aneurysm, median arcuate ligament release, and celiac artery revascularization. We present 3 patients without chronic abdominal pain who were found to have PDA aneurysms that were treated with aneurysm exclusion without celiac artery revascularization or ligament release.
CASE REPORT 1 The patient was a 61-year-old Caucasian female with no significant past medical history who presented to the vascular surgery clinic to be evaluated for an incidental inferior PDAA. At her initial presentation, the patient was completely asymptomatic. She denied any postprandial abdominal pain, weight loss, nausea, or vomiting. Her physical examination was benign and an abdominal bruit was not appreciated. The patient was found to have microhematuria during a routine visit with her primary care physician, and because of a family history of bladder cancer, a computed tomography (CT) scan of the abdomen and pelvis was obtained. Evaluation of the CT scan demonstrated a 2.0-cm inferior PDA aneurysm that originated 4.7 cm from the ostium of the superior mesenteric artery (SMA). Dynamic magnetic resonance imaging (MRI) displayed significant stenosis at the celiac axis throughout the respiratory cycle. There was increased compression during the expiratory phase, confirming MALS. Because of the size of the aneurysm and the known risk of spontaneous rupture, the decision was made to perform a 2-staged procedure. The patient was initially taken to the operating room for laparoscopic division of the median arcuate ligament. The lesser sac was entered and the right crus was identified. The pre-aortic fascia was dissected down to the abdominal aorta, and then distally to the celiac artery origin. Because of extensive thickening of the ligament, complete circumferential dissection was unable to be performed safely using a laparoscopic approach. The patient was then brought back 2 weeks later for selective SMA aortography with exclusion of the PDA aneurysm. Selective aortography delineated the SMA, inferior pancreaticoduodenal artery, as well as the aneurysm. A 6-French Raabe sheath was delivered over a stiff 0.035-inch glidewire into the SMA. This was followed by an multipurpose A catheter
Annals of Vascular Surgery
(MPA). A 0.014-inch balanced middleweight wire was then introduced through the catheter and into the posterior inferior pancreaticoduodenal artery. Once the 6French Raabe sheath was advanced into proper position, the decision was made to place an Atrium iCast 6-mm 16-mm covered stent to exclude the aneurysm without the need for coiling. After multiple unsuccessful attempts were made to guide the stent graft into the appropriate position, the stent dislodged from the balloon in the inferior pancreaticoduodenal artery. Attempts were made to recanalize the stent with a 0.014-inch Pilot guidewire, but were unsuccessful. A wire was manipulated past the stent graft and the aneurysm was directly cannulated. A microcatheter was placed over the wire and coil embolization was performed using a Cook (Cook Medical; Bloomington, IN) MWCE-185e10/4Tornado (x5); Cook MWCE 6/2-Tornado (x3); and Cook MWCE-185-8/4-Tornado (x3). Following primary coiling, a Palmaz 5 mm 60 mm stent was placed over the aneurysm origin to entrap the coils within the aneurysm as well as compress the dislodged stent against the arterial wall. Repeat angiography showed patency of the inferior pancreaticoduodenal artery and exclusion of the aneurysm. Selective celiac angiography displayed a thrombosed celiac axis. Endovascular recannulation was attempted, but unsuccessful. After review of the imaging, it was determined that the patient had adequate collateralization as well as retrograde flow through the celiac axis; therefore, close observation was deemed a reasonable alternative to an open revascularization. One-year follow-up with serial ultrasound and CT imaging has demonstrated inferior pancreaticoduodenal artery patency and aneurysm exclusion.
CASE REPORT 2 A 61-year-old male presented to the emergency department with right-sided flank pain. The patient had a past medical history of nephrolithiasis. He denied any nausea, vomiting, diarrhea, jaundice, or weight loss. A CT scan of the abdomen and pelvis was ordered that demonstrated nephrolithiasis and an incidental calcification near the uncinate process of the pancreas. Because of concern for a malignancy, the patient was referred to the HepatoBiliary and Pancreas service for further evaluation. The patient had a mild elevation in his carbohydrate antigen 19-9 and alpha fetoprotein, and an endoscopic retrograde cholangiopancreatography with endoscopic ultrasound was ordered for further evaluation. Endoscopic ultrasound was able to identify the calcified lesion, and a percutaneous biopsy was performed. The biopsy returned gross blood, which instigated the ordering of a CT angiogram (CTA). It was determined on the repeat imaging that there was a calcified pancreatic artery aneurysm within the uncinate process, as well as stenosis at the celiac artery origin. The decision was made to take the patient to the angiography suite for coil embolization and possible median arcuate ligament release. After achieving brachial artery access, selective SMA aortography using an MPA
Vol. 29, No. 1, January 2015
diagnostic catheter (AngioDynamics, Inc., Latham, NY) was performed showing the dilatation of the inferior pancreaticoduodenal artery. The aneurysm was occluded using Cook MWCE-185-6/2 (x4) Toranado coils (Cook Medical). Postembolization angiography showed a patent artery and no flow within the aneurysm sac. The patient tolerated the coiling well with no complications. It was determined based on the large collateral vessels and retrograde flow through the celiac axis that a median arcuate ligament release and celiac artery revascularization would not be required. The patient recently had a 1-year followup CTA that demonstrated an occluded PDA aneurysm with patent surrounding vasculature and no further signs of aneurysm filling. The patient has remained asymptomatic despite not having a median arcuate ligament release or celiac artery revascularization.
CASE REPORT 3 A 52-year-old male with no significant medical history or chronic abdominal symptoms presented to the emergency department with an acute onset of epigastric pain. CT scan of the abdomen and pelvis showed that the patient had a PDAA with active extravasation. Attempted intervention was performed at the outside institution without success and therefore the patient was transferred to our facility. Because of an acute drop in the patient’s hemoglobin, he was transfused 1 U of packed red blood cells. He arrived at our institution in stable condition and taken urgently to the hybrid angiography suite. Selective SMA angiography from a brachial artery approach was performed identifying the PDAA with active extravasation. At the outside hospital, a femoral approach was attempted; however, because of the angulation at the SMA origin, they were unable to cannulate the vessel. A 5-French Ansel sheath was placed into the SMA orifice followed by a Renegade microcatheter (Boston Scientific; Natick, MA) that was used to cannulate the aneurysm. Coil embolization of the aneurysm, outflow, and inflow vessels were performed again using Cook Tornado coils (Cook Medical). Digital subtraction angiography demonstrated complete occlusion of the aneurysm and retrograde flow to the celiac axis. Postoperatively, a CT angiogram showed that the celiac axis demonstrated a classic upward hooking appearance consistent with MALS. Because of the fact that the patient denied any history of chronic abdominal pain, and continued to be symptom free, he was discharged home without celiac artery revascularization or surgical release of the median arcuate ligament. At 1-month follow-up, the patient continued to be without any abdominal pain.
DISCUSSION The anatomic structure of the median arcuate ligament was first described in 1917 by Lipshutz.2 It was almost 50 years later that the syndrome was described by Harjola and Dunbar.2,3 It has been
Case reports 122.e3
theorized that the reason for the delay between identification of the ligament and the syndrome is secondary to the fact that most individuals never experience symptoms. It is estimated that in 10e 24% of healthy, asymptomatic individuals the median arcuate ligament crosses anterior to the celiac artery, causing some degree of compression4,5; however, it is only severe enough to cause symptoms in approximately 1% of the population.5 Throughout the respiratory cycle, the aorta moves in the cephalad and caudal direction. With inspiration, the expansion of the chest leads to the movement of the aorta in the caudal direction. Conversely, during expiration, the aorta rises in the cephalad direction leading to increased compression by the median arcuate ligament. PDA aneurysms are a rare anomaly and are most commonly diagnosed at the time of rupture.6 When ruptured, mortality rates reach approximately 30%, making early identification and urgent treatment necessary.7 With compression of the celiac axis by the median arcuate ligament, there is decreased blood flow to the foregut. To compensate for this hypoperfusion, an increase in blood flow is directed through the SMA and its branches to the end organs of the foregut.1 Angiography has demonstrated robust arterial collateralization system in patients with celiac artery compression assisting with foregut perfusion. The increased blood volume propagating through these small visceral arteries leads to an increase in pressure as well as wall tension. Increased endoluminal pressure can eventually damage the vessel wall, leading to aneurysmal formation.8 This physiologic mechanism was originally proposed by Sutton and Lawton in 1973.9 Other causes of endoluminal damage may include atherosclerosis, trauma, fibromuscluar dysplasia, and pancreatitis.10 Both MALS and PDA aneurysms can be detected during conventional angiography (Fig. 1); however, with the advent of CTA and MRA, celiac compression can easily be identified and visceral artery aneurysms less than 1 cm in diameter are routinely detected.11 In regards to PDA aneurysms, most will be diagnosed either at the time of rupture or incidentally during a CT/MRI for some other indication. Attention should be taken to assure that 3-mm slices and iodinated contrast are ordered in addition to 3dimensional reconstruction to receive best visualization for preoperative planning.12 Imaging software improvements, as well as the technological hardware advancements of 16- and 64-slice scanners, have resulted in the ability to obtain high quality images of the abdominal aorta and its branches. Typically, the sagittal plane is optimal for visualizing the proximal portion of the celiac axis, and in many
122.e4 Case reports
Annals of Vascular Surgery
Fig. 1. Selective angiography demonstrating the visceral aneurysm as a branch of the superior mesenteric artery.
Fig. 2. Still image from dynamic magnetic resonance imaging demonstrating the compression of the celiac artery during expiration.
cases, 3-dimensional imaging also allows for identification of the median arcuate ligament.4 Dynamic MRA can aid in the diagnosis of MALS. This allows for observation of the celiac axis throughout the respiratory cycle, displaying constant stenosis of the artery, with augmentation of compression during the expiratory phase (Fig. 2). Focal narrowing with a hook appearance (Fig. 3) is a characteristic finding in patients with MALS, helping distinguish it from other forms of stenosis, such as atherosclerosis.11 Recommended treatment for patients with symptomatic MALS with a PDA aneurysm is 2-fold. Concomitant revascularization of the celiac axis and its branches after release of the median arcuate ligament has been the standard recommended treatment, particularly in symptomatic patients.13e15 Treatment of celiac stenosis in the face of a PDA aneurysm still remains a controversial subject. Reilly et al.16 was one of the first studies to identify that surgical decompression followed by revascularization had a significantly greater long-term result in relieving symptoms (76% vs. 56%, respectively). Additionally, the PDA aneurysm needs to be treated irrespective of the size because of the risk of
Fig. 3. MALS is often identified on sagittal imaging with the celiac artery in a classic ‘‘hook’’ appearance. The arrow identifies the ligament compressing the celiac axis.
rupture.17 The decision in our first case was to perform a 2-staged procedure, with the understanding that both would need to be done in a timely matter because of the high risk of visceral aneurysm rupture. We initially tried to perform a laparoscopic
Vol. 29, No. 1, January 2015
Case reports 122.e5
Table I. Review of the literature demonstrating other reported cases of PDAA treatment without celiac revascularization Author
Patients (n)
Treatment of aneurysm
Outcome/follow-up
1-year follow-up with no recurrence 4-year follow-up with no recurrence 0% mortality with mean follow-up of 37 months 0% mortality 1/5 (20%) stent thrombosis
Flood et al. (2013)
2
2/2 coil embolization
Nishiyama et al. (2013)
3
3/3 surgical resection
11
Ikenda et al. (2013)
Hildebrand et al. (2007) Ikeda et al. (2007)
2 3
5/11 coil embolization 5/11 coil embolization and celiac stenting 1/11 isolation only 2/2 surgical resection 3/3 coil embolization
Sugiyama and Takehara (2006) Bellosta et al. (2005) Ducasse et al. (2004) Imamura et al. (1998) Chiang et al. (1994) Grech et al. (1988)
2 1 1 1 1 1
2/2 observation Surgical resection Coil embolization Aorto-hepatic bypass Surgical resection N/A
N/A Short-term (6-month) follow-up showed no recurrence N/A 9 days no recurrence 4 months no recurrence N/A 4 months N/A
N/A, not applicable.
median arcuate ligament release, but because of anatomical constraints, a full release could not be performed safely. Decompression of the celiac artery involved dissection of the fibrous tissue on the anterior portion of the aorta down to the celiac axis. Identification of the proximal celiac axis is important to confirm that the median arcuate ligament has been divided. While open surgical release has been the standard method, laparoscopic release has been reported.18 There is an increased risk of arterial injury and major hemorrhage; however, studies have shown that laparoscopic treatment resulted in equivalent morbidity and mortality, as well as a shortened hospital stay and no difference in recurrence.19,20 It should be noted, however, that care must be taken to release all the thick fibrous tissues that make up the median arcuate ligament. Failure to fully release the ligament will result in continued external compression on the celiac stump, which will not alleviate the patient’s symptoms. Additionally, some authors have theorized that patient’s symptoms are secondary not only to arterial compression, but also overstimulation of the celiac sympathetic plexus.4,21e25 It is important that thorough external arteriolysis be performed circumferentially around the celiac trunk, to lyse these sympathetic nerves. Median arcuate ligament release may not be necessary if the patient is not experiencing chronic abdominal symptoms. As can be seen with our cases, all patients were free of chronic abdominal pain. In the first patient, a release was attempted, but
determined to be inadequate. The celiac axis was completely occluded, making the median arcuate ligament release irrelevant, as it would not have changed the antegrade blood flow through the celiac axis. The second and third patient also received coil embolization only without ligament release or revascularization. Angiography demonstrated that the foregut was receiving enough perfusion through retrograde blood flow and robust collateralization. It is our theory that the aneurysm likely formed before the evolving collateral vascular system within the foregut. Now that these patients have an extensive pancreaticoduodenal arcade to endure the increased blood flow, recurrent aneurysm formation is unlikely. In all cases, endovascular repair was performed to exclude the PDA aneurysm. Open surgical repair of mesenteric aneurysms can include a variety of surgical approaches; however, mortality rates for open repair have been reported to be as high as 19%.26 In patients who receive open repairs for visceral occlusive disease, complications such as bowel resections, infections, sepsis, acute renal failure, and death have been described.27e30 The endovascular approach allows for exclusion of the aneurysm with a covered stent graft or coil embolization. The anatomical location of the aneurysm may make cannulation challenging, but 0.014inch platforms and modern endovascular techniques can preserve the collateral blood flow successfully and occlude mesenteric aneurysms. Endovascular stenting or coil embolization is less
122.e6 Case reports
invasive than open surgery and can be safely applied with minimal complications. An endovascular approach allows one to selectively embolize the aneurysm, sparing other vessels.1 Landing zones will be crucial to identify before deployment of a stent as care must be taken not to occlude critical blood vessels. As seen with case 1, the anatomy may provide challenges that do not allow for adequate stent deployment. With improving endovascular techniques, percutaneous angioplasty and stenting of the celiac artery axis have been described and are performed by skilled endovascular interventionalists. Care should be taken with this approach, however, because angioplasty and stenting will not eliminate the external compression from the ligament. It is recommended that endovascular management not be performed if the median arcuate ligament is not released. Celiac artery revascularization may not be necessary in the asymptomatic patient. Though it seems logical that resolving the celiac stenosis/occlusion would decrease recurrence of aneurysms by improving celiac artery antegrade flow. To date, there are no documented cases of aneurysm recurrence within asymptomatic patients who did not receive ligament release or revascularization (Table I). PDA aneurysms secondary to MALS are a rare finding, but once identified should be treated urgently. This report demonstrates that celiac revascularization after treatment of the aneurysm may not be necessary, and the exclusion of the aneurysm may be the only necessary treatment. Short-term results demonstrate no recurrence of the aneurysm and the patients continue to be asymptomatic. Further studies and long-term follow-up are still necessary to further evaluate if new aneurysm formation will occur. It is recommended that follow-up imaging be performed at 6 months postoperatively followed by yearly imaging.
REFERENCES 1. Kallamadi R, Demoya MA, Kalva SP. Inferior pancreaticoduodenal artery aneurysms in association with celiac stenosis/occlusion. Semin Intervent Radiol 2009;26: 215e23. 2. Lipshutz B. A composite study of the celiac axis artery. Ann Surg 1917;65:159e69. 3. Harjola PT. A rare obstruction of the celiac artery: report of a case. Ann Chir Cynaecol Fenn 1963;52:547e50. 4. Horton KM, Talamini MA, Fishman EK. Median arcuate ligament syndrome: evaluation with CT angiography. Radiographics 2005;25:1177e82.
Annals of Vascular Surgery
5. Lindner HH, Kemprud E. A clinicoanatomical study of the arcuate ligament of the diaphragm. Arch Surg 1971;103: 600e5. 6. Santiagu S, Gananadha S, Harrington TJ, et al. Direct percutaneous puncture embolization of a peripancreatic pseudoaneurysm presenting with haemosuccus pancreaticus. J Med Imaging Radiat Oncol 2008;52:370e3. 7. Moore E, Matthews MR, Minion DJ, et al. Surgical management of peripancreatic arterial aneurysms. J Vasc Surg 2004;40:247e53. 8. Kobayashi T, Uenoyama S, Isogai S. Successful transcatheter arterial embolization of an inferior pancreaticoduodenal artery aneurysm associated with celiac axis stenosis. J Gastroenterol Hepatol 2004;19:599e601. 9. Sutton D, Lawton G. Coeliac stenosis or occlusion with aneurysm of the collateral supply. Clin Radiol 1973;24: 49e53. 10. Verta MJ Jr, Dean RH, Yao JS, et al. Pancreaticoduodenal arterial aneurysms. Ann Surg 1977;186:111e4. 11. Klonaris C, Psathas E, Katsargyris A, et al. Multiple aneurysms of the inferior pancreaticoduodenal artery: a rare complication of acute pancreatitis. Case Rep Vasc Med 2013;2013:621350. 12. Horiguchi A, Ishihara S, Ito M, et al. Multislice CT study of pancreatic head arterial dominance. J Hepatobiliary Pancreat Surg 2008;15:322e6. 13. Grotemeyer D, Duran M, Iskandar F, et al. Median arcuate ligament syndrome: vascular surgical therapy and follow-up of 18 patients. Langenbecks Arch Surg 2009;394:1085e92. 14. Takach TJ, Livesay JJ, Reul GJ Jr, et al. Celiac compression syndrome: tailored therapy based on intraoperative findings. J Am Coll Surg 1996;183:606e10. 15. Palmer OP, Tedesco M, Casey K, et al. Hybrid treatment of celiac artery compression (median arcuate ligament) syndrome. Dig Dis Sci 2012;57:1782e5. 16. Reilly LM, Ammar AD, Stoney RJ, et al. Late results following operative repair for celiac artery compression syndrome. J Vasc Surg 1985;2:79e91. 17. Nishiyama A, Hoshina K, Hosaka A, et al. Treatment strategies for a pancreaticoduodenal artery aneurysm with or without a celiac trunk occlusive lesion. Ann Vasc Dis 2013;6:725e9. 18. Duffy AJ, Panait L, Eisenberg D, et al. Management of median arcuate ligament syndrome: a new paradigm. Ann Vasc Surg 2009;23:778e84. 19. Tulloch AW, Jimenez JC, Lawrence PF, et al. Laparoscopic versus open celiac ganglionectomy in patients with median arcuate ligament syndrome. J Vasc Surg 2010;52: 1283e9. 20. Jimenez JC, Harlander-Locke M, Dutson EP. Open and laparoscopic treatment of median arcuate ligament syndrome. J Vasc Surg 2012;56:869e73. 21. Szillagyi DE, Rian RL, Elliott JP, et al. The celiac artery compression: does it exist? Surgery 1972;72:849e63. 22. Bech FR. Coeliac artery compression syndromes. Surg Clin North Am 1997;77:409e24. 23. Delis T, Gloviczki P, Altuwaijri M, et al. Median arcuate ligament syndrome: open celiac artery reconstruction and ligament division after endovascular failure. J Vasc Surg 2007;46:799e802. 24. Desmond CP, Roberts SK. Exercise-related abdominal pain as a manifestation of the median arcuate ligament syndrome. Scand J Gastroenterol 2004;39:1310e3. 25. Akatsu T, Hayashi S, Yamane T, et al. Emergency embolization of a ruptured pancreaticoduodenal artery aneurysm
Vol. 29, No. 1, January 2015
associated with the median arcuate ligament syndrome. J Gastroenterol Hepatol 2004;19:482e3. 26. Coll DP, Ierardi R, Kerstein MD, et al. Aneurysms of the pancreaticoduodenal arteries: a change in management. Ann Vasc Surg 1998;12:286e91. 27. Arthurs ZM, Titus J, Bannazadeh M, et al. A comparison of endovascular revascularization with traditional therapy for the treatment of acute mesenteric ischemia. J Vasc Surg 2011;53:698e704. discussion, 04e5.
Case reports 122.e7
28. Block TA, Acosta S, Bj€ orck M. Endovascular and open surgery for acute occlusion of the superior mesenteric artery. J Vasc Surg 2010;52:959e66. 29. Barakate MS, Cappe I, Curtin A, et al. Management of acute superior mesenteric artery occlusion. ANZ J Surg 2002;72: 25e9. 30. Ryer EJ, Kalra M, Oderich GS, et al. Revascularization for acute mesenteric ischemia. J Vasc Surg 2012;55: 1682e9.
