You are on page 1of 60

CHAPTER 150 Renovascular and Aortic Developmental


Figure 150-10 A, Suprarenal coarctation (bracket) with superior mesenteric artery stenosis (arrow). B,
Subsequent thoracoabdominal bypass (broad arrow) with aortic implantation of superior mesenteric artery (arrow).
(From Stanley JC, et al: Abdominal aortic coarctation: surgical treatment of 53 patients with a thoracoabdominal
bypass, patch aortoplasty, or interposition aortoaortic graft. J Vasc Surg 48:1073-1082, 2008.)

Figure 150-11 Complex aortic,

1 splanchnic, and renal arterial reconstruction: 1, Celiac artery (CA) stenosis; 2,
superior mesenteric artery (SMA) stenosis; 3, suprarenal midabdominal
aortic coarctation; 4, right renal artery
ostial stenosis; 5, left segmental renal artery stenosis; 6, CA implanted onto aorto-SMA bypass (with autogenous
internal iliac artery graft); 7, reconstructed SMA; 8, thoracoabdominal aortic bypass; 9, right renal artery
2 10, left segmental
reimplantation onto aorta;
renal artery implantation onto adjacent segmental renal artery.
(From Upchurch GR, Jr, et al: Pediatric splanchnic arterial occlusive disease: clinical relevance and operative
treatment. J Vasc Surg 35:860-867, 2002.)


Figure 150-12 A, Absence of the abdominal aorta and agenesis of the proximal main renal arteries, with
reconstitution of the distal renal arter- ies and iliac arteries by way of extensive collateral vessels. B, Complex
thoracic aortobi-iliac bypass with bilateral renal artery reconstructions.

Renal Artery Repair

Renal artery repair of renal artery stenoses found independently of an aortic narrowing or in conjunction with an
aortic coarctation is most often undertaken to prevent longterm sequelae of hypertension. The surgical options and
reconstructive configurations should be considered carefully
to minimize kidney ischemic times and facilitate
unobstructed and durable renal artery reconstructions
A 2,7
(Figs. 150-13 to 150-15).
There should be a degree of caution regarding the timing
of operation. Significant technical challenges exist in reconstructing arteries in very small infants. Uncontrolled severe
hypertension or renal failure in a rare instance may force the
surgeon to undertake an operation in the very young.
However, renal revascularizations are more likely to be

implantations. Note the residual origins of the native
renal arteries. (From Stanley JC, et al: Pedi- atric
renovascular hypertension: a thirty-year experience of
operative treatment. J Vasc Surg 21:212-227, 1995.)

successful after age 3 years, and deferring reconstructive

procedures in infants when possible is reasonable and
Determining the optimal surgical therapy depends on the
specific character of the renal artery disease being treated.
Considerations may include whether or not both renal arteries are involved, the length of the stenotic disease, and
whether segmental renal branches are also narrowed. The
B common method of surgical reconstruction has been
renal artery implantation into the aorta or an undiseased
adjacent artery.2,7 Use of the internal iliac artery for aortorenal bypass grafts when needed has proven to be a suitable
conduit and is currently recommended if re-implantation of
the vessel cannot be performed.2 An aortorenal bypass with
a vein graft, as utilized in the past, is no longer
recommended due to aneurysmal degeneration seen in
more than half of these grafts at follow-up.48
The common coexistence of splanchnic arterial narrowings in patients with renal artery stenoses and the propensity
for the CA to develop stenosis are reasons why splenorenal
reconstructions with a direct anastomosis of the involved
vessels are not favored. Using this type of reconstruction
may perpetuate or result in a recurring hypertensive state.
Less common renal artery reconstructive procedures
include resection of the stenotic segment of artery with a
primary re-anastomosis, a focal arterioplasty, or an open
oper- ative dilation. It may be necessary to consider
nephrectomy in patients with non-reconstructable disease.
Indications for nephrectomy include multiple intrarenal
stenoses not ame- nable to open in-situ or endovascular
treatment, or diminu- tive nonfunctioning kidneys 2 to 3 cm
in size. Nephrectomy is only appropriate when the
contralateral kidney is assumed to be sufficient in preventing
the patient from going into renal failure.

Figure 150-14 A, Focal

segmental renal artery with
renal artery implantation
into the distal main renal
Stanley JC, et al: Pediatric
hypertension: a thirty-year
experience of operative
treatment. J Vasc Surg
21:212-227, 1995.)

Large clinical experiences with pediatric-aged renovascular hypertension are uncommon.7,49-55 The published University of Michigan series included 97 children (39 girls and 58
boys), ages 3 months to 17 years, who underwent operation
from 1963 to 2006.2 All but one patient had refractory hypertension not responsive to medical therapy. Renal artery

Figure 150-15 Aortorenal bypass with an internal iliac

artery graft. (From Stanley JC, et al: Pediatric

renovascular hypertension: a thirty-year experience of operative

treatment. J Vasc Surg 21:212-227, 1995.)

stenoses in this series were developmental in

80% of cases. Mesenteric arterial occlusive
coarctations affected 33%. Primary renal artery
operations were performed 132 times, with
procedures includ- ing resection beyond the
stenosis and implantation into the aorta in 49
cases, renal artery in 7, or SMA in 3 cases; aortorenal and iliorenal bypasses with vein or iliac
artery grafts in 40 cases; focal arterioplasty in
10 cases; resection with re-anastomosis in 4;
operative dilation in 4; splenorenal bypass in 2;
and primary nephrectomy in 13 when arterial
complexity of this disease is underscored by the
fact that 32 children required bilateral renal
artery operations, 17 patients underwent a
simultaneous CA or SMA reconstruction, 19
had a patch aortoplasty, and 11 underwent a
thoracoabdominal aorto-aortic bypass. Thirty
secondary renal artery procedures were

performed in 19 of this series patients, including 9

nephrectomies. Hypertension was cured in 68 children
(70%), improved in 26 (27%), and was unchanged in 3 (3%).
There was no perioperative mortality. No patient required
dialysis during a mean follow-up of 4.2 years.
Another large experience was reported from the Hospital
Beaujon in France.50 This series included 78 children (35
boys and 43 girls), ranging in age from 1.4 to 18 years. There
were 91 primary revascularization procedures and 15
nephrecto- mies. These numbers were quite similar to the
University of Michigan series. A third series from the
Cleveland Clinic involved 56 children (23 boys and 33
girls), ranging in age from 8 months to 21 years.52 They
reported 46 primary opera- tions and 10 primary
nephrectomies. A fourth series was from both Vanderbilt
University and the Childrens Hospital of Philadelphia.53 This
experience included 50 children (24 boys and 26 girls),
ranging in age from 5 months to 16 years. The authors
reported on 28 primary reconstructive proce- dures and an
additional 12 primary nephrectomies. The


latter two series, unlike the experiences at the University

of Michigan or Hospital Beaujon, did not contain many
patients treated for aortic or splanchnic arterial disease.
When blood pressure control outcomes were analyzed in
the three major reports on pediatric renovascular hypertension from the University of Michigan, Vanderbilt UniversityChildrens Hospital of Philadelphia, and the Cleveland
Clinic, a great deal of similarity was found in the results.
Respective cure rates were 70%, 70%, and 66%; reported
improvement rates were 27%, 26%, and 23%; and failure
rates were 3%, 4%, and 11%. Each of these reports reflected
the open surgical management of renal artery stenoses in

Splanchnic Artery Repair

Splanchnic artery repair depends on whether obstructions of
the CA and SMA are symptomatic. Any mandate to reconstruct the CA or SMA applies only to symptomatic cases. A
relative indication to prophylactically reconstruct these
vessels exists when performance of an aortoplasty or renal
revascularization would make a subsequent CA or SMA
revascularization exceedingly difficult (Fig. 150-16).56 Aortic
reconstructions remote from the mesenteric vessels make
prophylactic mesenteric revascularization less likely to be
Operative treatment of pediatric splanchnic arterial
occlu- sive disease is complex. Internal iliac artery grafts
are used to

treat lengthy CA or SMA stenoses in young patients. Anastomoses are fashioned with interrupted sutures when reconstructing small arteries 2 to 3 mm in diameter. In older
adolescents, a continuous suture may be used to complete
anastomoses of larger splanchnic arteries. In other children,
aortic implantation of the CA or SMA after spatulation of
the transected vessel beyond its stenotic origin is favored
over an aortosplanchnic bypass. Patients treated by both
methods are quickly able to gain weight and are rapidly freed
of abdominal discomfort. Developmental CA or SMA
stenoses in adults may be reconstructed with either vein
grafts or synthetic conduits. Percutaneous transluminal
balloon angio- plasty with or without stenting in children for
these splanch- nic ostial lesions is not appropriate and illadvised.
Splanchnic arterial occlusive disease encountered in
childhood has an uncertain natural history. Long-term
follow-up of these children with periodic imaging studies in
those operated upon seems appropriate, and regular clinical
assessment of those not subjected to surgical therapy is recommended. With regard to the latter, it is important to
remember that most CA and SMA stenoses are asymptomatic
because of the inferior mesenteric artery serving effectively
as a source of collateral circulation. Unoperated patients
should be aware of their splanchnic arterial anatomy and be
able to pass this information to a surgeon performing a later
abdominal operation. Inadvertently interrupting the collateral circulation might cause catastrophic intestinal ischemia
in these individuals.


implantation into infrare- nal aorta (arrow), as
treatment of intestinal ischemia due to an ostial
stenosis of this artery, in a child with near occlusion of
the supraceliac abdominal aorta who underwent a
concomitant thoracoabdominal bypass to the distal
aorta (not visualized on this lateral aortogram).

Case reports and small series exist in support of the technical

feasibility of endovascular treatment of abdominal aortic narrowings. Endoluminal interventions may have utility in the
treatment of select focal abdominal aortic coarctations
distant from the renal and splanchnic branches. This therapy
has been most commonly employed for the treatment of
stenotic disease due to an aortitis.57-61 Developmental aortic
narrow- ings are prone to significant recoil from the
hypoplastic and fibrotic segments, and stenting may be
required to overcome this.62-64 Long-term benefits of
endovascular interventions are unproven in children with
developmental aortic narrowings, and subsequent operations
following stent failure are consid- erably more difficult.
The role of endovascular interventions in the treatment
of pediatric renal artery stenoses remains controversial.2,65,66
Some anticipated causes of failure following angioplasty in
these cases include excessive elastic tissue in many stenoses,
resulting in early postdilation recoil, as well as the minute
caliber of developmentally narrowed vessels that might lead
to their disruption. However, a few recent reports suggested
success with catheter-based interventions.67,68 It is more
logical to expect successful endovascular therapy for the
treatment of a quiescent inflammatory aortoarteritis than for
the treatment of a developmentally hypoplastic renal artery.
The management of developmental aortic, renal, and
splanchnic artery disease remains a complex process that is

handled best by the cooperation of multiple specialties.

Nephrology, vascular surgery, diagnostic and interventional
radiology, and cardiology all may routinely contribute valuable insights into the diagnosis, follow-up, and management
of these patients. A disease specific, multidisciplinary,
patient- and family-centered focus will afford the most
salutary outcomes.


Castelli PK, Dillman JR, Smith EA, Vellody R, Cho K, Stanley JC: Imaging
of renin-mediated hypertension in children. Am J Roentgen 200:W662
672, 2013.
State of contemporary imaging.
ONeill JA, Jr: Long-term outcome with surgical treatment of renovascular
hypertension. J Ped Surg 33:106111, 1998.
Early operative experience from one of the largest childrens hospital in
the United States.
Shroff R, Roebuck DJ, Gordon I, Davies R, Stephens S, Marks S, Chan M,
Barkovics M, McLaren CA, Shah V, Dillon MJ, Tullus K: Angioplasty for
renovascular hypertension in children: 20-year experience. Pediatrics
118: 268275, 2006.

An extensive experience with catheter-based interventions performed at one

of the largest childrens hospital in Western Europe.
Stanley JC, Criado E, Eliason JL, Upchurch GR, Jr, Berguer R, Rectenwald
JE: Abdominal aortic coarctation: surgical treatment of 53 patients with
a thoracoabdominal bypass, patch aortoplasty, or interposition aortoaortic
graft. J Vasc Surg 48:10731082, 2008.
Largest reported surgical experience with abdominal aortic coarctations.
Stanley JC, Criado E, Upchurch GR, Jr, Brophy PD, Cho KJ, Rectenwald
JE, Kershaw DB, Williams DM, Berguer R, Henke PK, Wakefield TW:
Pediatric renovascular hypertension: 132 primary and 30 secondary
opera- tions in 97 children. J Vasc Surg 44:12191228; discussion 1228
1229, 2006.
Largest reported surgical experience with developmental renovascular
Upchurch GR, Jr, Henke PK, Eagleton MJ, Grigoryants V, Sullivan VV,
Wakefield TW, Jacobs LA, Greenfield LJ, Stanley JC: Pediatric
splanchnic arterial occlusive disease. Clinical relevance and operative
treatment. J Vasc Surg 35:860867, 2002.
Largest reported surgical experience with developmental splanchnic arterial
The reference list can be found on the companion Expert Consult website


CHAPTER 150 Renovascular and Aortic Developmental Disorders


1. Stanley JC, et al: Abdominal aortic coarctation: surgical treatment of
53 patients with a thoracoabdominal bypass, patch aortoplasty, or interposition aortoaortic graft. J Vasc Surg 48:10731082, 2008.
2. Stanley JC, et al: Pediatric renovascular hypertension: 132 primary
and 30 secondary operations in 97 children. J Vasc Surg 44:12191228;
discussion 12191228, 2006.
3. Sen PK, et al: The middle aortic syndrome. Br Heart J 25:610618,
4. Messina LM, et al: Middle aortic syndrome. Effectiveness and
durability of complex arterial revascularization techniques. Ann Surg
204:331339, 1986.
5. Arnot RS, et al: The anatomy of the posterior wall of the abdominal
aorta. Its significance with regard to hypoplasia of the distal aorta. S
Afr Med J 47:899902, 1973.
6. Stanley JC, et al: Developmental occlusive disease of the abdominal
aorta and the splanchnic and renal arteries. Am J Surg 142:190196,
7. Stanley JC, et al: Pediatric renovascular hypertension: a thirty-year
experience of operative treatment. J Vasc Surg 21:212226; discussion
212226, 1995.
8. Connolly JE, et al: Middle aortic syndrome: distal thoracic and abdominal coarctation, a disorder with multiple etiologies. J Am Coll Surg
194:774781, 2002.
9. Delis KT, et al: Middle aortic syndrome: from presentation to contemporary open surgical and endovascular treatment. Perspect Vasc Surg
Endovasc Ther 17:187203, 2005.
Delis KT, et al: Neurofibromatosis type 1: from presentation and
diag- nosis to vascular and endovascular therapy. Perspect Vasc Surg
Endovasc Ther 18:226237, 2006.
11.West CA, et al: Middle aortic syndrome: surgical treatment in a child
with neurofibromatosis. J Vasc Surg 42:1236, 2005.
Flynn PM, et al: Coarctation of the aorta and renal artery
stenosis in tuberous sclerosis. Ped Radiol 14:337339, 1984.
Hall EK, et al: A case report of rapid progressive coarctation and
severe middle aortic syndrome in an infant with Williams syndrome.
Congen Heart Dis 4:373377, 2009.
Radford DJ, et al: The middle aortic syndrome: an important
feature of Williams syndrome. Cardiol Young 10:597602, 2000.
Salerno AE, et al: Vascular involvement in tuberous sclerosis.
Pediatr Nephrol 25:15551561, 2010.
Shefler AG, et al: Middle aortic syndrome in a boy with
arteriohepatic dysplasia (Alagille syndrome). Pediatr Cardiol 18:232
234, 1997.
Friedman JM, et al: Cardiovascular disease in neurofibromatosis
1: report of the NF1 cardiovascular task force. Genet Med 4:105111,
Hamilton SJ, et al: Insights into the pathogenesis of
neurofibromatosis 1 vasculopathy. Clin Genet 58:341344, 2000.
Lin AE, et al: Cardiovascular malformations and other
cardiovascular abnormalities in neurofibromatosis 1. Am J Med Genet
95:108117, 2000.
Oderich GS, et al: Vascular abnormalities in patients with
neurofibro- matosis syndrome type I: clinical spectrum, management,
and results. J Vasc Surg 46:475484, 2007.
Easton DF, et al: An analysis of variation in expression of
neurofibroma- tosis (NF) type 1 (NF1): evidence for modifying genes.
Am J Hum Genet 53:305313, 1993.
Degterev A, et al: The role of NF1 factors in regulation of elastin
gene transcription. Matrix Biol 18:295307, 1999.
Esterly JR, et al: Vascular lesions in infants with congenital
Circulation 36:544554, 1967.
Limbacher JP, et al: Hypoplasia of the abdominal aorta associated
with rubella syndrome. South Med J 72:617619, 1979.
Singer DB, et al: Pathology of the congenital rubella syndrome. J
71:665675, 1967.
Ogino H, et al: Overview of late outcome of medical and surgical
treat- ment for Takayasu arteritis. Circulation 118:27382747, 2008.


Adelman RD, et al: Coarctation of the abdominal aorta and

renal artery stenosis related to an umbilical artery catheter placement
in a neonate. Pediatrics 106:E36, 2000.
Seifert ME, et al: Acquired coarctation of the abdominal aorta
related to umbilical artery catheterization. Conn Med 73:273276,
Hahn D, et al: A review of Takayasus arteritis in children in
Gauteng, South Africa. Pediatr Nephrol 12:668675, 1998.
Kanitkar M: Renovascular hypertension. Indian Pediatr 42:47
54, 2005.
McCulloch M, et al: Angiographic features of 26 children with
Takayasus arteritis. Pediatr Radiol 33:230235, 2003.
Soto ME, et al: Takayasu arteritis: clinical features in 110 Mexican
Mestizo patients and cardiovascular impact on survival and prognosis.
Clin Exper Rheumatol 26:S915, 2008.
Criado E, et al: Abdominal aortic coarctation, renovascular,
hyperten- sion, and neurofibromatosis. Ann Vasc Surg 16:363367,
Fossali E, et al: Renovascular disease and hypertension in children
with neurofibromatosis. Pediatr Nephrol 14:806810, 2000.
Lehrnbecher T, et al: Neurofibromatosis presenting as a severe
systemic vasculopathy. Eur J Pediatr 153:107109, 1994.
Bartosh SM, et al: Childhood hypertension. An update on etiology,
diagnosis, and treatment. Pediatr Clin North Am 46:235252, 1999.
37.Ingelfinger JR: Renovascular disease in children. Kidney Int 43:493
505, 1993.
Silverstein DM, et al: Treatment of primary and secondary
hypertension in children. Pediatr Nephrol 21:820827, 2006.
Bayazit AK, et al: Reno-vascular hypertension in childhood: a
nation- wide survey. Pediatr Nephrol 22:13271333, 2007.
Lewis VE, et al: Hypertension and facial palsy in middle aortic
syn- drome. Arch Dis Child 85:240241, 2001.
Hanevold C, et al: The effects of obesity, gender, and ethnic
group on left ventricular hypertrophy and geometry in hypertensive
children: a collaborative study of the International Pediatric
Hypertension Associa- tion. Pediatrics 113:328333, 2004.
The fourth report on the diagnosis, evaluation, and treatment of
high blood pressure in children and adolescents. Pediatrics 114:555
576, 2004.
Deshpande PV, et al: Hypertension: a cause of growth impairment.
J Hum Hypertens 16:363366, 2002.
Knecht S, et al: How much does hypertension affect cognition?
Explained variance in cross-sectional analysis of non-demented
community-dwelling individuals in the SEARCH study. J Neurol Sci
283:149152, 2009.
Singh-Manoux A, et al: High blood pressure was associated with
cogni- tive function in middle-age in the Whitehall II study. J Clin
Epidemiol 58:13081315, 2005.
Castelli PK, et al: Imaging of renin-mediated hypertension in
Am J Roentgenol 200:W662672, 2013.
Marks SD, et al: Update on imaging for suspected renovascular
hyper- tension in children and adolescents. Curr Hypertens Rep
14:591595, 2012.
Stanley JC, et al: Fate of 100 aortorenal vein grafts:
characteristics of late graft expansion, aneurysmal dilatation, and
stenosis. Surgery 74: 931944, 1973.
Kimura H, et al: Surgical treatment and long-term outcome of
renovas- cular hypertension in children and adolescents. Eur J Vasc
Endovasc Surg 39:731737, 2010.
Lacombe M: Surgical treatment of renal artery lesions in children.
Chirurgie 124:264271 (in French), 1999.
Lacombe M: Surgical treatment of renal artery stenosis in
children: 60 surgically treated cases. Chirurgie 117:661666 (in
French), 1991.
Martinez A, et al: Improved results of vascular reconstruction in
pedi- atric and young adult patients with renovascular hypertension.
J Urol 144:717720, 1990.
ONeill JA, Jr: Long-term outcome with surgical treatment of
renovas- cular hypertension. J Ped Surg 33:106111, 1998.


Piercy KT, et al: Renovascular disease in children

and adolescents. J Vasc Surg 41:973982, 2005.



Renovascular Disease

Stadermann MB, et al: Results of surgical treatment for

renovascular hypertension in children: 30 year single centre experience.
Nephrol Dial Transplant 25:807813, 2010.
Upchurch GR, Jr, et al: Pediatric splanchnic arterial occlusive
disease. Clinical relevance and operative treatment. J Vasc Surg
35:860867, 2002.
Gu ZM, et al: Transluminal catheter angioplasty of abdominal
aorta in Takayasus arteritis. Acta Radiol 29:509513, 1988.
Khalilullah M, et al: Percutaneous transluminal balloon
angioplasty of the aorta in patients with aortitis. Circulation 76:597
600, 1987.
Rao SA, et al: Takayasu arteritis: initial and long-term follow-up
in 16 patients after percutaneous transluminal angioplasty of the
descending thoracic and abdominal aorta. Radiology 189:173179,
Sawada S, et al: Treatment of Takayasus aortitis with selfexpanding metallic stents (Gianturco stents) in two patients. Cardiovasc
Intervent Radiol 17:102105, 1994.
Tyagi S, et al: Balloon angioplasty of the aorta in Takayasus
arteritis: initial and long-term results. Am Heart J 124:876882, 1992.


Eliason JL, et al: Durability of percutaneous angioplasty and stent

implantation for the treatment of abdominal aortic coarctation: a case
report. Vasc Surg 35:397401, 2001.
Fava MP, et al: Percutaneous transluminal angioplasty in patients
with Takayasu arteritis: five-year experience. J Vasc Intervent Radiol
4:649 652, 1993.
Hallett JW, Jr, et al: Coarctation of the abdominal aorta: current
options in surgical management. Ann Surg 191:430437, 1980.
Booth C, et al: Management of renal vascular disease in
neurofibroma- tosis type 1 and the role of percutaneous transluminal
angioplasty. Nephrol Dial Transplant 17:12351240, 2002.
McTaggart SJ, et al: Evaluation and long-term outcome of
pediatric renovascular hypertension. Ped Nephrol 14:10221029, 2000.
Casalini E, et al: Two-year clinical follow-up of children and
adolescents after percutaneous transluminal angioplasty for
renovascular hyperten- sion. Invest Radiol 30:4043, 1995.
Shroff R, et al: Angioplasty for renovascular hypertension in
children: 20-year experience. Pediatrics 118:268275, 2006.


Mesenteric Vascular Disease:

General Considerations

Based on a chapter in the seventh edition by Juan Carlos Jimenez and William J. Quinones-Baldrich

Mesenteric ischemia occurs when perfusion of the visceral

organs fails to meet normal metabolic requirements. This
disorder is categorized as either acute or chronic on the basis
of the duration of symptoms. Acute mesenteric ischemia
(AMI) occurs rapidly during hours to days and frequently
leads to acute intestinal infarction requiring resection (see
Chapter 153). The most common causes are embolization to
the mesenteric arteries and acute thrombosis related to a
preexisting plaque. Chronic mesenteric ischemia (CMI) is a
more insidious process and progresses during weeks to several
months (see Chapter 152). The most common cause is progressive occlusive disease of the visceral arteries, usually
related to atherosclerosis. Because CMI is relatively uncommon, it is frequently misdiagnosed as a gastrointestinal disorder. Patients typically have undergone an extensive workup
for other potential causes before being diagnosed with CMI.
The first case of AMI was diagnosed and treated successfully with intestinal resection and reanastomosis by Elliott in
1895. Goodman first described chronic intestinal angina as a
clinical disorder in 1918.1 Dunphy, a surgical resident at the
Peter Bent Brigham Hospital, in 1936 reported a case of a
patient with weight loss and pain out of proportion to abdominal findings who subsequently died and was found to have
mesenteric occlusive disease on autopsy. After reviewing the
medical records of 12 patients who died of intestinal angina,
he found that 7 of the 12 (58%) had a history of chronic
abdominal pain, thus introducing the potential for early
intervention to prevent disease progression and death.2
Warren and Eberhard were the first to describe mesenteric
venous thrombosis (MVT) as a distinct cause of intestinal
infarction in 1935, differentiating it from occlusion of the
mesenteric arteries.3
In the modern era, Klass4 performed the first superior
mes- enteric artery (SMA) embolectomy in 1950, avoiding
intes- tinal resection. Although the patient died several
days later of a heart-related condition, autopsy revealed
the bowel to be normal. In 1958, Shaw and Maynard5
performed the first successful thromboendarterectomy of the
SMA at the Massachusetts General Hospital. Morris et al 6
performed the first successful retrograde bypass graft from
the infrarenal

aorta to the SMA in 1962. Stoney and Wylie 7 at the University of California, San Francisco, first described antegrade
aortovisceral bypass and transaortic visceral thromboendarterectomy in 1966. Ushering in the current era of percutaneous treatment of visceral arterial occlusive disease, Furrer8
and Novelline9 separately published the initial reports of
endovascular dilatation of the SMA in 1980, and Finch 10 first
reported the treatment of a celiac artery stenosis with a
Palmaz stent.


The primitive dorsal aorta gives rise to the abdominal aorta
during fetal development. Ventral segmental arteries emerge
from the primitive ventral aorta, which disappears around the
fourth week of gestation. Multiple segmental branches from
the primitive ventral aortathe 10th, 13th, and 21st
persist and develop into the celiac artery, SMA, and inferior
mesenteric artery (IMA), respectively. Disparity in the
regres- sion of the primitive ventral aorta and its segmental
branches infrequently causes deviations in the visceral
arterial anatomy. The celiac artery arises from the abdominal
aorta just caudal to the diaphragm at the level of L1 and is
bordered by the median arcuate ligament at the aortic
hiatus superiorly and the superior border of the pancreas
inferiorly. Tradition- ally, the three branches from this
common trunk include the left gastric, splenic, and common
hepatic arteries. However, multiple variations of the true
trifurcation can exist. Most frequently, the common
hepatic artery and its branches arise
from the SMA or directly from the abdominal aorta.
Exposure of the celiac trunk is best achieved through a
midline transabdominal incision, which also allows visual
assessment of bowel viability during surgical revascularization. The celiac trunk and its branches are surrounded by the
celiac plexus of nerves, which must be divided for proximal
exposure. A midline laparotomy is performed, and the triangular ligament is divided. The gastrohepatic ligament is then
divided longitudinally to the level of the posterior parietal
peritoneum. The liver is carefully retracted to the right of
midline with a self-retaining retractor. Placement of a


nasogastric tube facilitates identification of the esophagus.

The posterior peritoneum overlying the diaphragmatic crus
is divided sharply to expose the celiac trunk. The first vessel
encountered is usually the common hepatic artery traversing
to the right of midline toward the liver. The hepatic artery
can be exposed back to the origin of the celiac trunk, which
is covered by lymphatic tissue and the celiac nerve plexus.
The diaphragmatic crus is divided to expose the celiac origin
and the supraceliac aorta.
The SMA arises a few centimeters caudal to the celiac
trunk, and its origin is crossed by the neck of the pancreas
and the splenic vein. It arises at a less acute and downwardsloped angle than the celiac trunk and lies superior to both
the uncinate process of the pancreas and the third portion of
the duodenum. The superior mesenteric vein runs parallel
adjacent to the artery, usually along its right border. The first
important branch of the SMA is usually the inferior pancreaticoduodenal artery, which supplies collateral circulation
with the celiac artery through the gastroduodenal and superior pancreaticoduodenal arteries. The second major branch
of the SMA is frequently the middle colic artery, which arises
at the inferior border of the pancreas. The right colic, ileocolic, and third-order mesenteric branches arise distally,
supplying the small bowel within the mesentery.
During surgical exposure, the SMA can be approached
either anteriorly at the base of the transverse colon
mesentery or lateral to the fourth portion of the
duodenum. Anterior SMA exposure involves lifting the
transverse colon superi- orly to clearly expose the base of
its mesentery. The small intestine is covered in a moist
towel or a bowel bag and retracted to the right. A
horizontal incision is made through the posterior peritoneum
at the base of the mesentery at the level of the proximal
jejunum and extended to the right of midline. The middle
colic artery can be used as a landmark within the transverse
colon mesentery and to localize the main SMA trunk.
Palpation often aids in localizing the SMA. The superior
mesenteric vein is often visualized first, and the SMA can be
palpated adjacent and to the left of it. This approach provides
excellent exposure of the SMA. Exposure of the more
proximal SMA to the left and lateral to the fourth portion
of the duodenum can also be achieved. The ligament of
Treitz is divided, and the lateral wall of the duo- denum is
mobilized off the anterior surface of the aorta. The SMA can
be identified just distal to its origin from the aorta. The IMA
is usually located 3 to 4 cm cephalad to the aortic
bifurcation, just to the left of midline, and usually arises at
the level of the third lumbar vertebra. The main trunk
frequently divides into sigmoidal branches and the left colic
artery. The ascending left colic artery forms the inferior
marginal artery of Drummond, which is the major collateral arcade between the SMA and IMA. The SMA and
IMA are also linked by the meandering mesenteric artery (of
Moskowitz), known historically as the arc of Riolan. This
vessel runs more centrally, medial to the mesenteric border
of the colon and through the middle of the mesenteric arcade
near the inferior mesenteric vein. The meandering mesenteric artery is likely to be produced by the dilation of a

Mesenteric Vascular Disease: General

collateral vessel in response to significant stenosis or occlusion of the SMA or IMA.11 Sigmoidal branches lead to the
left and right superior rectal arteries, which collateralize with
branches of the hypogastric arteries in the pelvis.


Normal intestinal function and nutrient absorption rely on
adequate perfusion and oxygenation to the microvascular
splanchnic circulation. Various autoregulatory mechanisms
ensure adequate gut circulation through both vasoconstriction and relaxation of arterial smooth muscle. The degree of
visceral artery dilatation and constriction determines the
relatively large fluctuations in splanchnic blood flow during
fasting, postprandial states, and periods of extreme stress.
Visceral blood flow can vary dramatically, ranging from 10%
of cardiac output in the setting of shock or hypovolemia, to
20% to 25% at rest or while fasting, and up to 35% after a
large carbohydrate meal.12,13 Seventy percent to 80% of mesenteric blood flow supplies the mucosal and submucosal
layers. The severely diminished blood flow observed in
patients with nonocclusive mesenteric ischemia results from
severe vasospasm related to this process. Duplex studies
dem- onstrate moderate to high arterial resistance in the
SMA circulation, with low diastolic flow and slight flow
reversal during fasting states. In the postprandial period, lowresistance signals are noted throughout both systole and
diastole, indic- ative of dilated splanchnic arteriolar beds;
flow reversal does not occur. In contrast, low arterial
resistance signals are noted in the celiac artery circulation
regardless of feeding, probably because of the influence of
the low-resistance hepatic vascu- lar bed. Perko et al14 also
noted that in fasting subjects per- forming submaximal
exercise, splanchnic vascular resistance doubled and
exhibited a 50% reduction in hepatosplenic blood flow and a
25% reduction in mesenteric blood flow.
Multiple mechanisms are responsible for regulating mesenteric arteriolar smooth muscle tone, and they are often
interdependent. Extrinsic factors include sympathetic efferent nerves in the prevertebral celiac and mesenteric ganglia,
which initiate stimuli for arterial vasoconstriction. Hormonal
pathways also contribute to extrinsic regulation of splanchnic
blood flow. The renin-angiotensin feedback mechanism
causes mesenteric vasoconstriction through the direct action
of angiotensin II during hypovolemic states. Low-volume
states and hyperosmolarity stimulate the neurohypophysis,
which releases vasopressin,15 a hormone that causes splanchnic vasoconstriction, reduction in portal venous pressure, and
venodilatation.15 In addition, when subjected to shear stress,
activation of the nitric oxide synthase enzymes on the
surface of red blood cell membranes occurs, leading to
significant dilation of mesenteric arteries under hypoxic
Intrinsic regulation also occurs through metabolic and
myogenic feedback mechanisms.15 In the metabolic pathway,
mucosal ischemia prompts the release of metabolic byproducts, causing vasodilatation in arteriolar smooth muscle and
preferentially shunting increased blood flow to the intestinal
mucosa. In the myogenic pathway, which dominates the


regulation of blood flow in the small intestines, 17 abrupt

decreases in perfusion pressure are sensed by arteriolar baroreceptors, which respond by decreasing arteriolar wall
tension to maintain blood flow.18 Together, these mechanisms
main- tain mucosal perfusion and integrity during periods of
relative ischemia.

Asymptomatic occlusive disease of the visceral arteries is a
common finding in elderly patients. Wilson et al15 demonstrated that 17.5% of 553 consecutive patients older than 65
years examined with duplex ultrasonography (DUS) had a
critical stenosis of at least one visceral vessel. In addition,
autopsy studies have estimated the prevalence of atherosclerosis involving the mesenteric arteries to be between 6%
and 10%.19
Despite an aging population, hospitalizations for AMI in
the United States have declined from 9.6 to 6.7 per 100,000
from 1998 to 2010.20 There is some evidence that this decline
may be related to the increasing and widespread use of
statins and the efficacy of warfarin in the prevention of
thromboem- bolic events in patients with atrial fibrillation. 21
Probably in part because of differences in longevity, AMI
and CMI dis- proportionately affect women, with a femaleto-male ratio of approximately 3 : 1.22
Because the majority of patients with mesenteric
occlusive disease manifest no symptoms, the exact incidence
of CMI is not known. However, admissions for CMI
account for less than 1 per 100,000 admissions23 and have
been increasing steadily in recent years in the United
States.22 Whether these figures represent an actual increasing
incidence or simply increased re-intervention due to
restenosis from endovascular therapy remains to be
determined. Despite the high preva- lence of individuals
with asymptomatic mesenteric arterial occlusive disease,
patients usually demonstrate involvement of two or more
mesenteric vessels before symptoms arise owing to the
development of extensive collateralization over time. In fact,
in a study by Thomas et al,24 who observed 980 consecutive
patients with asymptomatic significant (50%) stenosis of at
least one mesenteric artery, only four patients developed
mesenteric ischemia and all of them had signifi- cant threevessel disease after follow-up of 1 to 6 years. The variability
of symptoms in patients with chronic abdominal pain often
makes the diagnosis challenging, resulting in treat- ment
delays and increased morbidity.

Chronic Mesenteric Ischemia
Atherosclerosis is the most common cause of CMI, and
patients frequently have a history of smoking, hypertension,
and hyperlipidemia. They may also have evidence of atherosclerotic disease in other vascular beds, particularly
coronary, cerebrovascular, renal, aortoiliac, and other
peripheral arter- ies. Although it is much more uncommon,
CMI may also be

seen in association with vasculitis and other inflammatory

conditions, such as lupus, Buergers disease, and radiation
arteritis. Median arcuate ligament syndrome is a separate
entity that may lead to symptoms of CMI and is caused by
the compression of the celiac artery by the median arcuate
ligament. Symptoms are classically exacerbated by full

Acute Mesenteric Ischemia

Arterial emboli are the most common cause of AMI,
repre- senting 40% to 50% of cases.25 The proximal
source of the embolus is frequently intracardiac mural
thrombus that develops in patients with atrial
tachyarrhythmias, myocardial infarction, cardiomyopathy,
structural heart defects, and cardiac tumors. Endocarditis
can result in septic emboli from affected valve leaflets.
Mural thrombus in proximal aneu- rysms in the thoracic or
proximal abdominal aorta can also serve as an embolic
source, as can atheromatous plaque even in the absence of
aneurysm. The SMA is the most common final destination
for mesenteric emboli, perhaps because of its relative size
and the decreased angle of takeoff from the abdominal
aorta compared with the other mesenteric vessels. In
addition, such emboli tend to lodge several centimeters
from the vessels origin, usually distal to the middle
colic artery (Fig. 151-1A). The angiographic hallmark of an
embolic occlusion seen on computed tomographic
angiogra- phy (CTA) or magnetic resonance angiography
(MRA) is an abrupt cutoff just beyond the origin of the
middle colic artery.26

Arterial Thrombosis

Arterial thrombosis constitutes the next most common cause

of AMI and occurs in 20% to 35% of cases.27,28 Preexisting
atherosclerotic plaque affecting all visceral vessels is the
most common finding. Hypercoagulability syndromes can
also pre- dispose to acute visceral artery thrombosis. The
affected segment of artery is usually its origin at the level of
the aorta. Patients with acute arterial thrombosis frequently
have preexisting symptoms of CMI. Schoots et al 29 reviewed
45 observational studies encompassing 3692 patients with
AMI and found that mortality from acute thrombosis of a
mesen- teric artery was 77.4%, compared with 54.1% for
patients with acute arterial embolism to visceral vascular
beds. This increased mortality is likely due to the
involvement of larger segments of bowel because occlusions
are typically within the first 2 cm of the SMA origin (Fig.
151-1B). This theory was corroborated in a Swedish study,
in which 213 patients with acute thromboembolic occlusion
of the SMA and intestinal infarction were examined post
mortem.30 The cause of occlu- sion was embolic in 57.3% of
patients and thrombotic in 41.3% (indeterminate in 1.4%).
The extent of intestinal infarction was significantly greater
in patients with SMA thrombosis compared with embolus.
Acute extension of an aortic dissection can also serve as a
mechanism for abrupt mesenteric vessel occlusion and



Mesenteric Vascular

Figure 151-1 A, Operative findings typical of acute mesenteric ischemia secondary to an embolus. Note that the
first portion of the jejunum is spared because the embolus lodged distal to the middle colic artery. B, Operative
findings typical of acute mesenteric ischemia secondary to arterial thrombosis. Note that the entire bowel is
affected. C, Appearance of the small bowel at second-look surgery after revascularization. Note the hemorrhagic
changes in the mesentery.

Nonocclusive Mesenteric Ischemia

Impaired intestinal perfusion in the absence of thromboembolic occlusion is termed nonocclusive mesenteric ischemia
(NOMI) and makes up approximately 5% to 15% of cases.31
Visceral ischemia can occur from a low-flow state, which is
exacerbated by the presence of any intestinal atherosclerotic
disease. It is theorized that in such circumstances, in an effort
to maintain cardiac and cerebral perfusion, excessive sympathetic output results in mesenteric vasospasm. NOMI most
commonly occurs secondary to cardiac disease, particularly
severe congestive heart failure, and in patients who have
undergone cardiac surgery. Atrial fibrillation, commonly a
cause of cardiac thrombi and visceral embolization, can also
induce NOMI by reducing left ventricular function and
causing low cardiac output. Other risk factors for NOMI
include age, hypovolemia, systemic vasoconstrictors,
vasoac- tive drugs (e.g., digoxin, -adrenergic agents, receptor blocking agents, cocaine), aortic insufficiency,
cardiopulmo- nary bypass, abdominal and cardiovascular
surgery, abdomi- nal compartment syndrome, and liver
In recent years, NOMI has been increasingly observed in
patients on hemodialysis. Quiroga et al33 reported incidence
rates more than 40 times greater among patients on hemodialysis compared with the general population. Using the
Taiwan National Health Insurance Research Database, Li
et al34 similarly reported a 44-fold higher risk of mesenteric
ischemia in patients on hemodialysis or peritoneal dialysis
compared with the general population. Older age and a
heavier burden of cardiovascular risk factors undoubtedly
contribute to these disparities. In addition, patients with endstage renal disease are frequently taking erythropoietin,
which elevates red cell mass and can lead to relative hyperviscosity. However, hypotension during dialysis has been
implicated as the most important and immediate precipitating risk factor for development of NOMI. 35 Therefore,
special care should be made to avoid dialysis-related
hypotension in patients at particularly high risk. High-risk
patients include those who are older, are taking
erythropoietin, have a longer history of dialysis, and are

Mesenteric Venous Thrombosis

MVT constitutes 5% to 15% of all cases of mesenteric
ischemia.36,37 Involvement is usually limited to the superior
mesenteric vein, but the inferior mesenteric, splenic, and
portal veins can also be involved. MVT is classified as
either primary (idiopathic) or secondary. Secondary MVT
occurs when an underlying disease process is present; this
type accounts for 90% of all patients with this disorder.
Conditions associated with MVT can be categorized into
three major categories: direct injury, local venous stasis or
congestion, and thrombophilia. Causes of direct injury
include abdominal surgery, trauma, and local inflammation,
such as that seen with inflammatory bowel disease, pancreatitis, and diverticulitis.38,39 Splanchnic venous stasis may
occur in the case of increased intra-abdominal pressure,
obesity, and intraoperative manipulation of the mesenteric
vessels. Inherited or acquired hypercoagulable diseases,
including protein C and protein S deficiency, myeloproliferative disorders, antithrombin III deficiency, antiphospholipid antibody syndrome, and factor V Leiden mutation,
are frequent causes of thrombophilia and thus also
predispose to MVT.40
The extent of bowel ischemia depends largely on the
degree of venous involvement and the presence of collateral
vessels. The transition from normal to ischemic intestine
is slower with MVT than with arterial occlusive disease.25
Edema and hemorrhage of the intestinal wall are frequently
seen, followed by focal sloughing of the mucosa. 25 The
origin of thrombosis varies, depending on the etiologic
process. When an intra-abdominal process is the cause,
thrombosis begins in the larger mesenteric veins and progresses to involve the smaller venous arcades and arcuate
channels.41 MVT caused by hypercoagulable conditions
usually begins in the smaller mesenteric veins. Symptomatic
acute MVT is associated with a 20% to 50% mortality
rate.41 A Swedish population-based autopsy study showed
that MVT-related death is more likely in patients with
portal vein thrombosis, systemic venous thromboembolism,
and obesity.42


Acute Mesenteric Ischemia
The most common symptom of AMI associated with arterial
thromboembolic disease is the sudden onset of abdominal
pain. Because of a lack of collateral flow to the visceral
organs, the presentation of AMI is more dramatic and severe,
often with rapid clinical deterioration. Nausea, vomiting,
diarrhea, emptying symptoms, and distention can also occur.
Classi- cally, the pain is out of proportion to the findings on
physical examination. Initially, bowel sounds are hyperactive
as the failure to relax the bowel smooth muscle leads to
emptying symptoms. Bowel sounds are typically diminished
in the later stages. Abdominal guarding and rebound
tenderness are absent in the early stages of AMI; however,
as intestinal ischemia and infarction progress, these signs
become more pronounced. They are typically late findings,
so their absence should not delay the diagnosis and treatment
of AMI. Other late findings include fever, oliguria,
dehydration, confusion, tachycardia, and shock.25 Metabolic
abnormalities can include leukocytosis, metabolic acidosis,
hyperamylasemia, elevated liver function values, and lactic
Patients with NOMI or MVT typically present with a
slower, more insidious clinical course. Frequently, patients
with NOMI are critically ill, hospitalized, intubated patients
who experience a sudden deterioration in their clinical condition. These patients are often administered intravenous
pressors, worsening mesenteric vasoconstriction and thus
decreasing splanchnic perfusion. In patients with MVT,
fever, abdominal pain and distention, nausea and vomiting,
and bloody stools are the most common findings.
Dehydration and profound fluid shifts lead to bloody ascites
and a hypo- volemic state, causing further propagation of
venous throm- bosis.25 Although nonspecific, normal Ddimer levels may help rule out MVT.43
To develop better diagnostic criteria to reduce delays in
diagnosis, Mitsuyoshi et al44 reviewed their 13-year experience treating 22 patients with NOMI. Multidetector row
computed tomography (MDCT) was unavailable at the time
the first 13 patients were treated, and 9 subsequently died
of intestinal necrosis. These first 13 cases were used to
devise four criteria for determining which patients
warranted MDCT evaluation once this technology
became available:
(1) ileus or abdominal pain, (2) catecholamine
(3) episode of hypotension, and (4) gradual rise in serum
transaminase level. If three of the four criteria were present,
patients received MDCT and underwent treatment with
high-dose intravenous prostaglandin E1. Of the nine patients
treated with this algorithm, only one died.

Chronic Mesenteric Ischemia

Postprandial abdominal pain and progressive weight loss are
the most common symptoms in patients with CMI. Pain is
often described as dull and crampy and located in the midepigastric region. Pain often occurs 15 to 45 minutes after a

Figure 151-2 Typical patient with chronic mesenteric

ischemia with significant weight loss and cachexia.
The patient is placed in a right semilateral decubitus
position in preparation for retroperitoneal trans- aortic

meal, and the severity varies according to the size and

type of meal. Patients typically develop food fear and
decrease their oral intake in anticipation of severe pain
after meals. Consequently, weight loss is a common
finding and in fact, when present, can help distinguish CMI
from other func- tional bowel disorders when the

diagnosis is in question.45 Changes in bowel habits, nausea,

and vomiting are less common findings. CMI is seen more
frequently in elderly women, who represent 70% of
patients. The variable nature of symptoms often makes the
diagnosis confusing and can result in delayed treatment.
The traditional risk factors for atherosclerosis are usually
present. A heavy smoking history is frequently obtained, and
the majority of patients also have a history of symptomatic
manifestations in other vascular beds, most commonly
cerebrovascular, coronary, and periph- eral arteries.46
Physical examination findings are usually nonspecific.
Patients are commonly undernourished and cachectic (Fig.
151-2). An abdominal bruit can sometimes be auscultated
but is not always present. Bowel sounds are frequently
hyper- active. Guarding and rebound tenderness are usually
absent. Low prealbumin and albumin levels are often seen
owing to the patients chronic malnourished state.

Noninvasive Evaluation
DUS is a useful tool for the early, noninvasive diagnosis of
visceral ischemic syndromes. Color Doppler scanning can be
used to assess the flow velocities and resistance index in the
splanchnic arteries and their arterial beds as well as to evaluate end-organ vascularity. The intestinal wall can also be
assessed with a high degree of accuracy by high-resolution


transabdominal ultrasound. Transmural hemorrhage, inflammation, and necrotic thickening in the bowel wall can be
imaged sonographically. Asymmetrical wall thickening with
associated ileus as well as ascites and free peritoneal air can
be seen in patients with AMI.48 DUS combined with expiratory maneuvers is also an excellent screening examination
for median arcuate ligament syndrome.
Moneta et al49 performed blinded DUS studies in 100
patients who previously underwent arteriography of the
celiac trunk and SMA. They hypothesized that lack of flow
or a peak systolic velocity (PSV) in the SMA of greater
than 275 cm/s, or no flow or a PSV of greater than 200 cm/s
in the celiac trunk, was a reliable indicator of 70% or greater
angiographic stenosis.50 With use of these criteria, duplex
sensitivity for detection of lesions in the SMA and celiac
artery was 92% and 87%, respectively. Overall accuracy for
detection of a 70% lesion in the SMA and celiac artery
was 96% and 82%, respectively. The IMA is not generally
assessed because it is generally difficult to visualize and is
of lower clinical importance. However, patency of the IMA
can usually be determined.
Limitations of transabdominal duplex scanning include
the wide variation in examination quality, which is operator
dependent. Patient-related factors, such as obesity, excessive
intraluminal bowel gas, variation in local anatomy, and
effects of respiration, can affect image quality.51 Care must be
taken to clearly define the origin of each vessel to avoid
inac- curate measurements. DUS is generally not
recommended in the workup of AMI.
DUS is also the primary imaging modality used for surveillance after both bypass and stenting. However, no
specific criteria exist for determining restenosis. Baker et al 52
evalu- ated the accuracy and utility of DUS for detection of
in-stent stenosis in 23 patients who underwent successful
(<20% residual stenosis on completion angiography)
mesenteric stenting for CMI (20 SMA alone, 3 both SMA
and celiac artery). Preprocedure DUS was performed in 13
patients with a mean PSV of 464 cm/s. Initial surveillance
DUS was per- formed in 21 patients at a mean of 0.9
month after revas- cularization. Mean PSV at this time
was 335 cm/s, and for 12 of these patients, the first
postoperative PSV in success- fully stented vessels was
higher than the 275 cm/s threshold used to diagnose highgrade native SMA stenosis. In addi- tion, there was no
correlation between surveillance PSV and the degree of
angiographic stenosis seen at the time of re-intervention.
In their review of 107 patients who underwent endovascular therapy for CMI, Schoch et al53 similarly reported that
although 83% of patients had recurrent stenosis on surveillance DUS, 53% of patients remained asymptomatic and
required no further intervention. These findings therefore
mandate obtaining an early baseline DUS against which
future surveillance scans can be compared. They also stress
the importance of considering the clinical context and
symptom recurrence in making the decision to re-intervene.
The predictive value of DUS for detection of disease
recurrence after mesenteric bypass has also been studied.

Mesenteric Vascular Disease: General


Liem et al54 evaluated 167 duplex examinations used in the

surveillance of 43 bypass grafts in 38 patients. On
comparison of antegrade and retrograde bypass
configurations, differences in
demonstrated only at the inflow artery. There were no
differences in PSV at the proximal or distal anastomosis, in
the midgraft, or in the outflow artery. Between the first
postoperative DUS examination and the latest follow-up
scan (on average 38 months), PSV measurements did not
change significantly. Graft failure occurred in two
patients, both of whom had no findings to suggest impending
occlusion on the DUS examinations immediately preceding
failure. No predictors of graft thrombosis were identified on
multivariable analysis.
Computed tomography (CT) is an accurate,
noninvasive imaging modality for diagnosis of mesenteric
ischemia. Modern MDCT enables imaging with excellent
spatial and temporal resolution. A meta-analysis of six
studies published between 1996 and 2009 on the
diagnostic accuracy of MDCT in AMI showed a pooled
sensitivity of 93% and specificity of 96%.55 Advantages
over conventional angiography include the relative ease and
speed of performance; the rapid infu- sion of contrast
agent through peripheral intravenous lines; and the ability
to simultaneously image the mesenteric arteries, veins,
and visceral organs. CTA is also useful for evaluating
patency of previously placed grafts and stents. Common
radiographic findings in the bowel wall related to AMI
include increased thickening, dilatation, and attenu- ation,
which can be easily detected with CT. Pneumatosis
intestinalis, portal venous air, mesenteric edema, and
ascites can also be detected. During the arterial phase of
contrast infusion, the mesenteric vessels can be evaluated
for throm- bosis, embolus, dissection, and aneurysm.
Venous engorge- ment and subtle intestinal findings
may be identified in cases of mesenteric venous
occlusion. A target sign may be seen in the superior
mesenteric vein, with thrombus in the center of the lumen
peripherally. The portal venous phase is more accurate for
diagnosis of MVT, and images during the venous phase
may be acquired if the diagnosis is still in question or for
surveillance of clot burden or extension in patients with
MVT. In many centers, a biphasic scan, which includes a
delayed venous phase in addition to an arterial phase,
has been used in the diagnosis of AMI because it enables
detection not only of arterial occlusion but also of MVT
and improves the ability to detect changes in the bowel
wall as well.
Disadvantages of CT include radiation, risk of contrast
nephropathy, and hypersensitivity reactions to iodinated
contrast agents. Inaccurate timing of contrast infusion
during the arterial phase may provide indeterminate
images and delay diagnosis. Because calcification at the
vessel origins enhances in a similar fashion to intravenous
contrast mate- rial, it is possible to underestimate the
degree of stenosis. Therefore, the noncontrast-enhanced
images should be reviewed. In addition, with careful
adjustment of the window and level, calcification can


236 SECTION 25 Mesenteric Vascular

be distinguished from intravascular
contrast material. Last, CT serves strictly as a

diagnostic modality; treatment must be performed through a

separate angiographic procedure or laparotomy.
MRA is useful for diagnosis of mesenteric occlusive
disease. Although MRA takes significantly longer to perform
than CTA, it avoids the radiation exposure associated with
CTA. Patients with hypersensitivity to iodinated contrast
agents may also benefit from MRA. However, CTA has
superior spatial resolution and allows better visualization of
the IMA, peripheral splanchnic vessels, calcified plaque, and
previously placed stents, giving it a distinct advantage. 26 For
these same reasons, CTA is also preferred to MRA for
surveillance after revascularization. Our standard workup
includes DUS to screen those thought to have CMI. CTA
is used routinely before intervention for AMI, CMI, and

Invasive Evaluation
Conventional angiography remains the gold standard in
the diagnosis of mesenteric ischemia. Anteroposterior and
lateral views of the visceral aorta (Fig. 151-3) as well as
selective catheterization of the celiac trunk, SMA, and
IMA provide the most accurate and specific localization of
stenotic and occlusive lesions. Therapeutic alternatives such
as balloon angioplasty, stenting, and thrombolysis and percutaneous thrombus extraction can all be used to restore
luminal visceral blood flow. These options are discussed in
more detail in the next section.
Gastric tonometry, or the measurement of PCO2 levels in
the gastric, jejunal, or colonic mucosa through a nasogastric
catheter, has been shown to be a useful adjunct for the diagnosis of mesenteric ischemia56,57 but is not currently widely
used in the United States.



Figure 151-3 A, Anteroposterior angiogram of a patient
with chronic mesenteric ischemia. Note the
meandering mesenteric collateral
vessels. B, Lateral aortogram of a patient with chronic



The goal of therapy for patients with mesenteric ischemia is
the prompt restoration of blood flow to the visceral organs.
Specifics of the treatment of AMI and CMI are also
discussed in Chapters 152 and 153.

Medical Treatment
Preventive risk factor modification helps control the progression of atherosclerosis in the mesenteric circulation, but
medical treatment alone is not effective in patients with
symptomatic mesenteric ischemia. Patients with known risks
for inheritable hypercoagulable disorders should undergo
screening and should be treated with systemic anticoagulation if indicated.
Before operation, aggressive fluid resuscitation with
resto- ration of adequate urine output is required owing to
the frequent finding of severe dehydration on presentation.
Elec- trolyte abnormalities and metabolic acidosis should
also be corrected. Patients with CMI are frequently
malnourished, so albumin, prealbumin, transferrin, and Creactive protein

ischemia. Lesions are typically located at the origin of

the vessel and often protrude into the aortic lumen.
SMA, Superior mesenteric artery. C, Meandering
mesenteric collateral vessels in a patient with chronic
mesenteric ischemia at the time of exploration. In
patients undergoing surgery for other reasons, this
finding should raise the suspicion of significant
mesenteric occlusive disease.

levels should be checked before revascularization. Preoperative total parenteral nutrition or enteral nutrition should be
considered in severely malnourished patients. Finally,
because the intestinal mucosa is damaged during periods of
prolonged ischemia, bacterial translocation can occur,
contributing to systemic sepsis. Broad-spectrum intravenous
antibiotics with aggressive fluid resuscitation can lead to
decreased mortality in these patients.58 Treatment against
gram-negative and anaerobic organisms is especially
A recent Cochrane review of 22 randomized controlled
trials investigating the use of different antiplatelet regimens
after endovascular treatment of peripheral vascular disease
concluded that there is limited evidence to support the use

of antiplatelet drugs in the reduction of restenosis and reocclusion.59

No comparable studies for mesenteric

and stenting have been published of which we are aware.

Unless it is contraindicated, our practice is to administer an
initial 300-mg clopidogrel load after the procedure, followed
by 75 mg daily for 1 to 3 months of treatment. Patients are
also prescribed aspirin indefinitely.

Endovascular Treatment
General Principles
Advances in endovascular techniques have greatly expanded
the role of percutaneous interventions for patients with mesenteric ischemia in recent years. Balloon angioplasty with
stenting has surpassed open surgery as the dominant method
of revascularization for CMI,22 and an endovascular approach
is now generally accepted as primary therapy.60-62 In contrast,
the adoption of endovascular modalities for treating AMI has
been slower, and a decline in the number of open revascularizations has not been observed.22 This is likely because most
patients with AMI have some degree of bowel ischemia
and many require laparotomy for adequate evaluation and
potential resection. In those with short-segment stenoses,
cardiac and pulmonary comorbidities, prior abdominal
surgery, coagulopathy, or malnutrition, endovascular therapy
is often favored.

Acute Mesenteric Ischemia. Because AMI occurs
so infre- quently and most revascularizations for AMI until
recently have been performed open, there are relatively few
retrospec- tive reviews describing outcomes after
endovascular treat- ment of AMI. Arthurs et al63 at the
Cleveland Clinic reported their experience with
revascularization in 70 consecutive patients with AMI from
1998 to 2008. Over time, their institution adopted an
endovascular-first approach, such that 81% of the patients
in their series were treated with endo- vascular therapy.
The etiology was more often thrombotic (endovascular
72% vs open 36%) than embolic (endovascu- lar 28% vs
open 64%), and age was older in patients undergo- ing
endovascular-first therapy (65 years vs 60 years), but there
were no other significant differences in comorbidities or
clini- cal presentation between the treatment groups.
Technical success was 87%, and 31% of patients were
able to avoid laparotomy. Compared with patients
undergoing open revas- cularization, those who underwent
endovascular revascular- ization had shorter lengths of
bowel resected, developed fewer complications, and had
lower perioperative mortality (39% vs 50%; P < .05).
Ryer et al64 also published their institutions 2-decade
(1990-2010) experience with revascularization for 93
patients with AMI. In addition, they performed a subgroup
revascularization in the first decade compared with the 48
treated in the second decade. More patients underwent
endovascular revasculariza- tion in the latter decade (7% vs
17%), but this difference was not significant, and the
majority (88%) of all patients still underwent open
revascularization. Improvements in mortal- ity between the
first and second decade were modest and

nonsignificant (30-day mortality: 27% vs 17%, P = .28; 1year mortality: 51% vs 31%, P = .11).
Chronic Mesenteric Ischemia. A large body of
literature now exists documenting early success and
favorable outcomes for patients with CMI undergoing
percutaneous endovascular therapy,53,65-71 and several
institutions have recently published their midterm outcomes.
Between 1995 and 2007, Dias et al72 successfully stented 47
of 49 mesenteric vessels in 43 patients with CMI. No
technical failures occurred after a 3-year learn- ing curve.
Intraoperative complications occurred in seven (15%)
patients, but all were successfully managed with endovascular techniques. No patient died within 30 days, and
87% of patients experienced symptom relief. Three-year
survival was 76%, and re-interventionfree survival was
Peck et al62 also published their intermediate outcomes
with similar results. From 2002 to 2008, 66 mesenteric arteries were treated in 49 patients. All patients had at least one
mesenteric vessel successfully treated, but five technical failures occurred (four mesenteric occlusions could not be
crossed and one IMA could not be stented). There was one
in-hospital death due to a myocardial infarction in a patient
who under- went a peripheral bypass during the same
admission. Symptom relief was achieved in 90%, and major
complications occurred in 16%. Three-year primary patency
was 64%, and freedom from symptomatic recurrence was
Compared with traditional open approaches, endovascular
revascularization results in lower or similar perioperative
mortality, fewer complications, and shorter hospital stays but
is associated with higher rates of recurrence of symptoms,
restenosis, and re-intervention73-75 Oderich et al73 performed
a risk-stratified comparison of 146 patients who underwent
open and 83 patients who underwent endovascular revascularization for CMI. Operative risk was assessed by the
Society for Vascular Surgery scores, and patients were
classified as high risk by the presence of at least one highrisk
revascularization were sig- nificantly older (71 years vs 65
years; P < .5) and higher risk (58% vs 31%; P < .001).
Angiographic features and extent of disease were similar
between the endovascular and open treatment groups. For
both low- and high-risk patients, there were no differences in
procedure-related mortality, but patients in the endovascular
group had fewer complications and shorter intensive care
unit and hospital stays. At 5-year follow-up, patients who
had undergone open revasculariza- tion had superior
recurrence-free survival (89% vs 51%) and primary patency
(88% vs 41%). Patients undergoing endo- vascular
revascularization had 5.1-fold higher odds of reste- nosis and
4.3-fold higher odds of re-intervention than did patients
undergoing open revascularization.
In a study by Atkins et al, 74 42 mesenteric vessels treated
with angioplasty (with stenting in 87%) were compared with
88 vessels treated with open revascularization. Mean followup was 15 months in the percutaneous transluminal
angioplasty (PTA) group and 42 months in the open surgery
group. No difference was noted in major morbidity or


mortality. Radio- graphic primary patency and

primary-assisted patency in the

PTA group at 1 year were significantly lower (58% and

65%, respectively) compared with the open surgery group
(90% and 96%, respectively). PTA with stenting was also
associ- ated with the need for earlier re-intervention.
Surprisingly, the rates of symptomatic recurrence
requiring re-intervention were high in both groups, with
no significant difference noted.
No randomized studies comparing open and endovascular
repair have been published to date. However, van Petersen
et al76 performed a systematic review of the literature
between 1988 and 2009. In a pooled comparison of 412 open
repairs and 227 endovascular repairs, they found similar
technical success, immediate symptom relief, and early
mortality (in-hospital and 30-day) between the two treatment
groups. Patients undergoing open revascularization
experienced higher in-hospital and 30-day morbidity (32%
vs 11%; P <
.0001) and were more likely to receive intensive care unit
care (88% vs 6%; P < .0001) but had lower rates of
recurrent stenosis (15% vs 37%; P < .0001), recurrent
symptoms (13%
vs 30%; P < .0001), and re-intervention (9% vs 20%; P
.0004) and higher primary patency (86% vs 51%; P < .
0001). These differences were seen at a mean follow-up of
35 months for the open group and 20 months for the
endovascular group. Using a large national administrative
database with 22,413 patients undergoing mesenteric
revascularization, we found that mortality was lower after
angioplasty and stenting com- pared with bypass for both
CMI (3.7% vs 13%; P < .01) and AMI (16% vs 28%; P < .
01).22 Bowel resection was shown to be a significant risk
factor for perioperative mortality (bypass: 45% vs 17%, P <
.001; angioplasty and stenting: 29% vs 11%, P < .001).
More patients undergoing bypass underwent bowel resection,
indicating greater disease severity among this
Mesenteric angioplasty and stenting have been associated
with high restenosis and re-intervention rates. Fortunately,
outcomes after re-intervention have been associated with
low mortality and excellent symptom improvement. Tallarita
et al77 studied re-interventions among 157 patients with
CMI who initially underwent mesenteric artery angioplasty
and stenting. In-stent restenosis (defined as PSV >330 cm/s
or angiographic stenosis >60%) developed in 36% of
patients, and 30 patients underwent re-intervention (87% by
repeated endovascular and 13% by open revascularization).
Among patients undergoing repeated endovascular
revascularization, 13 bare metal stents and 4 covered
stents were used. One patient undergoing repeated stenting
died, and 27% expe- rienced perioperative complications. Of
the 24 patients who had developed recurrent symptoms, 92%
had symptom improvement after re-intervention. At a
mean follow-up of 29 months, 50% of patients developed a
second reste- nosis and 23% required another reintervention. Of note, however, none of the patients in
whom covered stents were used experienced recurrence of
symptoms, restenosis, or re-intervention.

Lesions most amenable to endovascular treatment are

short, focal stenosis/occlusions with less calcification and
thrombus. More complex lesions and complete arterial

occlusions traditionally are better treated with open

revascu- larization. However, feasibility of endovascular
recanalization of complete occlusions has been reported.
Sharafuddin et al78 published the largest series to date on
the endovascular man- agement of totally occluded
mesenteric arteries and reported a technical success rate of
85% for the recanalization of 27 nonembolic total
occlusions of the SMA and celiac arteries. All four failures
were in SMA occlusions. Interestingly, tech- nical success
was not related to presence of stump (an appro- priate
probing point used to initiate the recanalization sequence
and to establish a stable catheter position), ostial plaque,
intraluminal versus subintimal route, occlusion length, or
vessel diameter. Instead, success often depended on the
ability to identify an appropriate probing point, for which
preoperative cross-sectional imaging was crucial, and a
stable catheter position, which frequently necessitated a
brachial approach. As a result, the access site
complication rate was relatively high (22%).

performing patch angioplasty. The sheath is then placed

through the distal end of the patch for retrograde access to
the SMA. Hand-injected lateral aortog- raphy is performed,
and a wire is placed retrograde across the lesion into the
aorta. A balloon-expandable stent is

Retrograde Mesenteric Stenting

Techniques for retrograde mesenteric stenting during laparotomy for AMI were developed by Milner et al and Wyers
et al, and success with this method has since been
duplicated and adapted (Fig. 151-4).79-83 During laparotomy,
the SMA is dissected at the base of the transverse
mesocolon, and a sheath is inserted in the vessel either
percutaneously or by arteriotomy. Wyers et al80 described
placing a longitudinal arteriotomy in the SMA, performing
a local thromboendar- terectomy if required, and

Figure 151-4 A, Retrograde superior mesenteric

arteriogram during exploration in a patient with acute
mesenteric ischemia secondary to arterial thromboses.
Note the proximal occlusion of the superior mes- enteric
artery. B, Intraoperative placement of a proximal superior
mes- enteric artery stent for acute revascularization of
the small bowel.

positioned and deployed to allow protrusion of the stent 1 to

2 mm into the aorta. In the review by Wyers et al,80 they
compared patients treated in this manner with patients who
underwent antegrade stenting and traditional surgical bypass
for AMI. Mortality was 17% for the patients who underwent
retrograde stenting compared with 100% for the antegrade
stenting group and 80% for those who underwent surgical
bypass. The technical success rate for the retrograde stenting
group was 100%.
In a study by Moyes et al, 81 four patients were treated
similarly with retrograde stenting of the SMA for AMI. Two
of the patients were alive at 2 years. One patient experienced
stent thrombosis on day 14 and required surgical bypass. The
remaining patient died on postoperative day 6 due to multiorgan failure. Stout et al83 also used the technique in three
female patients, one of whom required bowel resection.
After a mean follow-up of 8.4 months, the primary patency
rate was 100%. Pisimisis et al82 described adapting the
procedure by performing endarterectomy and patch
angioplasty after stent- ing rather than before. Advantages to
this alternative include improved ability to flush and to
inspect the lumen for residual thrombus and avoidance of a
stenosis with primary closure. In addition, the authors
reported that inflating the angio- plasty balloon within the
stent allowed excellent proximal control during the
endarterectomy and anastomosis.
Although early results demonstrate good
success, the long-term efficacy of retrograde mesenteric
stent- ing has not been established. As with antegrade
stenting, the main advantage of this technique is reduced
invasiveness and morbidity. However, retrograde stenting
also provides the ability to inspect the bowel before and after
reperfusion and can potentially reduce operative times
because the expo- sure of proximal inflow vessels (aorta,
iliac) and the harvest of autogenous conduit for surgical
bypass are avoided. Ret- rograde SMA stenting may also
be useful in patients if a good source of inflow is
unavailable for bypass or as a tem- porizing measure until
definitive bypass can be accomplished (e.g., in the case of
peritoneal contamination). Potential disadvantages include
inadvertent injury to the vessel or aortic dissection during
wire manipulation and restenosis due to intimal

Endovascular therapy should be the treatment of choice in
high-risk patients with CMI. High technical success rates
and decreased patient morbidity and mortality rates have
been reasonably well established in such individuals.
However, in patients who are good surgical candidates,
the advantage is not so clear. Restenosis and symptomatic
recurrence rates remain relatively high, as documented in
the current litera- ture, and re-intervention is often
required earlier and more frequently than with open
surgery. Placement of stents in the mesenteric arteries
may also complicate future surgical inter- vention,
especially in the celiac trunk, which is relatively short
before branching. However, anastomosis to the hepatic
artery is usually not compromised by prior celiac stent
placement. Because the origins of the visceral arteries are

angulated downward, especially the SMA, percutaneous

access from the femoral arteries may be difficult and result in
suboptimal stent placement. Thus, a brachial approach is
often preferred. One study found a significant improvement
in primary patency in patients who underwent treatment
through the brachial artery compared with a femoral
approach.84 In contrast, Turba et al85 did not find a significant
association of femoral versus brachial access to primary
patency of the celiac, SMA, or IMA when reviewing their
28-year experience treating 166 patients with CMI. Furthermore, as noted before, brachial access is associated with
higher rates of access siterelated complications.78 Technologic advances, such as deflecting tip sheaths, may improve
results with femoral access.
The risk of distal embolization is reported at approximately 8% without embolic protection devices. 86 With
increasingly aggressive treatment of more complex stenotic
lesions and total occlusions, the risk of distal embolization
theoretically increases. However, Sharafuddin et al 78 detected
no angiographically significant embolic events in their experience with endovascular recanalization of 27 nonembolic
total occlusions, but the possibility of microembolization
could not be excluded. Therefore, distal protection, such as
that used during carotid stenting, can be considered, especially for patients with complete occlusions, long lesions, or
severe calcification. Other complications described in the
literature include access site problems, renal insufficiency,
dissection, stent thrombosis, dislodgement or embolization,
and vessel perforation. Oderich et al.87 studied complications
that developed in 156 patients who underwent angioplasty
and stenting for CMI. On univariate analysis, antiplatelet
therapy was associated with lower rates of distal
embolization or vessel thrombosis. There were no significant
multivariable predictors. Stent fracture producing restenosis
and recurrent ischemia has been documented.88,89
Reperfusion hemorrhage, traditionally associated with open
surgical revascularization, has also been described after
balloon angioplasty.90 Despite these limitations, the role of
endovascular techniques will likely continue to expand for
patients with mesenteric ischemia. Early restenosis rates
are likely to improve with the development of new
balloons, drug-eluting stents, and stent-grafts.

At our institution, we perform mesenteric angiography for all
patients with CMI and attempt primary stenting as first-line
therapy. The brachial approach has been preferred for selective catheterization of the SMA owing to its acute downward
angle at its origin. However, recently we have started using
a deflecting tip sheath from a femoral approach with good
results in selected patients. Patients with isolated stenosis of
the celiac artery due to extrinsic compression by the median
arcuate ligament and appropriate symptoms undergo laparoscopic lysis of the median arcuate ligament first (Fig. 151-5).
Balloon angioplasty and stenting are avoided initially in
these patients but may be used after median arcuate ligament
release in patients with persistent abdominal symptoms and


no stenosis


necrosis after embolus may be less extensive and spare the

proximal jejunum and transverse colon distribution from
ischemia (see Fig. 151-1A). Acute thrombosis usually occurs
at the orifice of the SMA, and a more extensive pattern of
hypoperfusion and necrosis is present owing to the involvement of proximal SMA branches (see Fig. 151-1B).
After vascular reconstruction and reperfusion, the bowel
should be reassessed. Segments of bowel that otherwise
would have been resected may be spared owing to
improvement after revascularization. A wide range of
techniques to assess bowel viability have been described,
including pulse oxime- try, measurement of oxygen tension,
spectrophotometry, Doppler ultrasound, fluorescein dye, and
examination with Woods lamp, laser Doppler flowmetry,
and infrared imaging.91 At our institution, we favor a
combination of clinical assess- ment (using criteria such as
serosal color, bowel peristalsis, and vessel palpation) and
angiography after vascular reconstruction may be performed
to accurately assess the restoration of mesen- teric flow.
Most patients should undergo second-look lapa- rotomy in
24 hours to reassess bowel viability and the need for further
resection (see Fig. 151-1C). The decision to perform a
second-look laparotomy should be made during the initial
procedure and should not be changed on the basis of
subsequent clinical improvement. For those in whom bowel
resection is performed at initial exploration, the ends are
stapled and no anastomosis is performed until second-look

Acute Mesenteric Ischemia


Figure 151-5 A and B, Angiographic documentation
of extrinsic compression of the celiac artery by the
median arcuate ligament. Note that the stenosis is
more severe on expiration.

Detailed descriptions of the techniques of surgical visceral

revascularization for AMI are highlighted in Chapter 153.
SMA Embolectomy. Perfusion of the mesenteric
arteries is assessed by palpation and Doppler evaluation.
In cases in

duplex- or CTA-demonstrated persistent stenosis. We

conduct duplex surveillance 1 month after the initial
proce- dure and then at 6-month intervals to assess for
continued vessel patency.

Surgical Treatment
General Considerations
Laparotomy with visceral revascularization can be used to
treat patients with both AMI and CMI. Patients presenting
with signs and symptoms of AMI require urgent abdominal
exploration, assessment of bowel viability, and revascularization. Several techniques for the restoration of intestinal perfusion are available to the vascular surgeon, and familiarity
with a variety of options is crucial. Before revascularization,
large segments of both small and large intestine may appear
dusky, ischemic, or necrotic. Because SMA emboli typically
lodge distal to its origin, the middle colic artery and ileocolic
branches may remain patent. Thus, the pattern of bowel

Figure 151-6 Doppler examination of the intestine

should be done both at the mesentery and at the
antimesenteric border. This is most useful after

which the obstruction is caused by an embolus, a proximal

SMA pulse is often appreciated. Systemic heparinization is
established. If the artery feels relatively soft and free of atherosclerotic disease, a transverse arteriotomy is performed
distal to the area of obstruction, and the arterial lumen is
assessed for thrombus. Balloon-tipped embolectomy
catheters are gently passed proximally and distally until no
more clot can be removed. Care must be taken not to
overinflate the balloons and dissect the arterial intima.
Distally, mesenteric vessels are very thin, and overinflation
can result in rupture and intramesenteric extravasation. The
transverse arteriot- omy is then closed primarily with simple
interrupted Prolene sutures if no endarterectomy is
necessary. In cases in which a flow-limiting plaque is
present, the arteriotomy is converted to a longitudinal one,
and a local thromboendarterectomy is performed. Patch
angioplasty with autogenous vein is the preferred method of
revascularization owing to potential con- tamination from
concomitant bowel resection. The arteri- otomy site can also
be used for distal anastomosis of an antegrade or retrograde
bypass if necessary.

Chronic Mesenteric Ischemia

Detailed descriptions of the techniques of surgical visceral
revascularization for CMI are highlighted in Chapter 152.
Transaortic Endarterectomy. Stoney and Wylie
first described the trap-door approach for transaortic
endarterectomy (Fig. 151-7) in 1966 at the University of
California, San Fran- cisco.7 The procedure involves medial
visceral rotation, dis- section of the SMA, and longitudinal
aortotomy. It has been performed during the past 4 decades
with acceptable morbid- ity and mortality rates and proven
durability.92 Although it is rarely indicated anymore, this
approach may be useful in patients who have failed to
respond to endovascular therapy or in those with hostile
abdomens. Advantages of this opera- tion include removal
of atheroma from the aorta and both visceral arteries
simultaneously. Limitations include the need for extended
exposure of the upper abdominal aorta and incomplete
plaque removal if the atheroma extends to the distal
artery or if transmural calcification is present.

Antegrade Mesenteric Bypass. The majority of

open mesen- teric revascularizations consist of bypass
with a bifurcated synthetic graft from the supraceliac aorta
to the celiac artery and SMA. Historically, antegrade bypass
has demonstrated efficacy and durability.93-95 Jimenez et al95
reviewed the results of 47 patients with CMI who underwent
antegrade synthetic aortoceliac and aortomesenteric bypass
during a 14-year period. The in-hospital mortality rate was
11%, and the mean length of hospital stay was 32 30 days.
At a mean follow-up of 37 months, 100% of surviving
patients were symptom free, and 86% had significant weight
gain. Primary, primary- assisted, and secondary graft
patency rates at 5 years were 69%, 94%, and 100%,
respectively. Actuarial survival was 74%. Although
antegrade bypass was associated with signifi- cant
perioperative morbidity and mortality, it was also associated with excellent durability and symptom-free survival. 95
Cunningham and Reilly found similar results in their review
of 26 antegrade bypasses for CMI.93,96 They reported a
slightly lower perioperative mortality rate (7.7%). The
percentage of patients who were symptom free at 1 and 5
years was 96% and 86%, respectively. English et al 97
performed antegrade revascularization for CMI in 80
mesenteric vessels. Their in-hospital mortality rate was 29%.
The presence of intestinal gangrene correlated significantly
with perioperative death and escalation of symptoms before
bypass (acute-on-chronic). Symptom-free survival at 70
months was 57%.
Retrograde Mesenteric Bypass. In retrograde
mesenteric bypass, the inflow is typically from the
infrarenal aorta or either of the iliac arteries, although case
reports have docu- mented the success of bypasses
originating from the splenic artery.98 Wylie et al99 first
described retrograde mesenteric bypass (Fig. 151-8), and the
debate continues about its effi- cacy compared with
antegrade bypass. Retrograde bypass is favored in patients
who are elderly, are cachectic, or have severe cardiac,
pulmonary, and renal comorbidity or those who have
extensive plaque and calcification in the suprace- liac
aorta.100 Despite this controversy, no convincing evi- dence
favors one approach over the other, and no randomized
controlled trials have been conducted. Foley et al19 from

L renal

L renal
Figure 151-7 A, Transaortic endarterectomy
using a trap-door incision in a patient with severe aortic, mesenteric,
and renal occlusive disease. B, Appearance of the visceral vessel orifice after transaortic endarterectomy. The
aortotomy is usually closed with pledgeted
sutures because the arterial wall is very thin after endarterectomy. C,


Typical specimen after transaortic endarterectomy of an orificial lesion. Note the tapered end- point. Patients
selected for this technique should have lesions limited to the proximal segment of the artery. Failure to obtain an
endpoint can be corrected by intraoperative placement of a stent under direct vision. L, Left; SMA, superior
mesenteric artery.


Right iliomesenteric bypass for
mesenteric artery (SMA). The graft is passed between
the leaves of the base of the mesentery to avoid contact
R common
with the intestines.
iliac artery

Oregon Health Sciences University examined their results

with 50 bypass grafts to the SMA alone for both AMI and
CMI. All grafts originated from either the infrarenal aorta or
the iliac artery. Perioperative mortality was 3% in patients
with CMI and 24% in patients with AMI. All survivors had
relief of symptoms in the postoperative period. There were
two late deaths related to graft occlusion. Nine-year primaryassisted graft patency was 79%, and the 5-year survival
was 61%.
In a combined study from Beth Israel Deaconess Medical
Center and University of California, Los Angeles, the results
of 39 mesenteric bypass procedures were reviewed during a
9-year period.101 Symptom-free survival after antegrade (n
21) and retrograde (n = 18) bypass was compared by
the Kaplan-Meier life-table method. No significant
difference was found between the two techniques. Although
the overall incidence of postoperative complications was
higher in the antegrade group, the number of major
complications in both groups was similar. The 30-day
mortality was higher in the antegrade group (14.3%) than in
the retrograde group (0%), and none of the patients who
died were revascularized for acute ischemia.

Outcomes of Open
Acute Mesenteric Ischemia. Although AMIrelated mortal- ity after open revascularization has declined
from 50% in the 1990s to 30% in the 2000s, 22 mortality
remains high in patients with AMI even after successful
surgical revascular- ization. One of the contributing factors is
ischemia- reperfusion intestinal injury. Although the goal of
therapy is prompt restoration of mesenteric blood flow,
revasculariza- tion may lead to a paradoxical exacerbation of
tissue damage. Although the exact mechanism is not

Mesenteric Vascular Disease: General

play an important role.102 Ischemia-reperfusion injury has

been linked to postoperative myocardial depression,103,104
sepsis,102 acute lung injury,105 and multiorgan failure.106
Immediately after reperfusion, direct injury to the intestinal
wall occurs and triggers a series of events, including the
release of various inflammatory mediators, activation and
aggregation of neutrophils, and bacterial translocation.103
The intestine may take several weeks to months to recover,
and patients may experience prolonged diarrhea and periods
of malabsorption.106
Several translational studies have reported promising
strategies to attenuate the effects of ischemia-reperfusion
injury. In murine models, when revascularization after a prolonged period of SMA occlusion is preceded by brief SMA
clamping (termed preconditioning), epithelial narrowing and
hyperpermeability, leukocyte rolling and adherence, and bacterial translocation are all reduced.107-109 Similarly, postconditioning, or brief periods of reocclusion in the first few
minutes after revascularization, has also been shown to
decrease the extent of damage caused by oxidative stress and
cytokine release.110 The protective effects of conditioning are
thought to be mediated through the upregulation of heat
shock proteins and antioxidant enzymes and the downregulation of apoptosis and the inflammatory response.18
Aside from ischemia-reperfusion effects, other factors can
contribute to worse outcomes after mesenteric revascularization. Kougias et al111 reviewed the records for 72 patients
who underwent surgical visceral revascularization for AMI.
The overall 30-day mortality was 31%. Factors associated
with increased mortality were renal insufficiency, age older
than 70 years, metabolic acidosis, symptom duration, and
need for bowel resection during second-look operations.
Increased age and prolonged symptom duration were independent predictors of mortality, with relative risks of 3.6 and
4.6, respectively.
In their 20-year experience with revascularization for
AMI encompassing 93 patients, the group at the Mayo Clinic
found that Society for Vascular Surgery comorbidity score,
conges- tive heart failure, and chronic kidney disease
predicted higher early mortality, whereas advanced age and
connective tissue disease predicted long-term mortality.64
Thirty-day mortality was 22%, and the primary causes of
death were ongoing mesenteric ischemia, multiorgan failure,
adverse cardiac events, and withdrawal of care by the health
care proxy.

reviewed 265
Chronic Mesenteric Ischemia.
Oderich et al
completely under- stood, the production of toxic oxygen
radicals is thought to

SECTION 25 Mesenteric Vascular

vessels (143
SMA, 113 celiac, and 9 IMA) in 146 patients

with CMI who underwent open revascularization (93%

bypass, 7% transaortic endarterectomy). Symptoms were
improved in 97%, and procedure-related morbidity and
mor- tality were 36% and 3%, respectively. At 5 years,
overall survival was 72% and symptom-free survival was

89%. All patients with recurrent symptoms underwent reintervention. Half of these (n = 4) required emergency
operation for acute thrombosis. The other half underwent
angioplasty and stent- ing. Table 151-1 presents the results
of revascularization for CMI from selected series.

Table 151-1

Review of Contemporary Series of Elective Revascularization for Chronic

Mesenteric Ischemia
Results (%)

No. of







No. of

et al73
et al112
Atkins et al74
et al97
Park et al46
et al95
Cho et al113





































NOMI represents an insidious disease process that is distinct
from thromboembolic AMI but has a similarly high
mortality. Cardiac surgery and low-flow states are the
most common causes of this disorder, and treatment is
directed toward improving circulatory support and increasing
cardiac output. Selective mesenteric angiography remains
the best invasive diagnostic modality, which can be followed
by catheter-based interventions. These include direct infusion
of intra-arterial vasodilators, such as papaverine and
prostaglandin E1, as well as angioplasty and stenting if
necessary.28 Intravenous
prostaglandin E1 has also demonstrated efficacy in treatment
of vasospasm associated with NOMI. 44 In a model of cardiac
tamponadeinduced acute
NOMI, Kang et al114
demonstrated that low doses of intra-arterial iloprost
(prostacyclin), a potent inhibitor of platelet aggrega- tion
with fibrinolytic activity, exhibited a significant vasodilatory effect on mesenteric blood flow.

A detailed discussion of the diagnosis and treatment of MVT
is contained in Chapter 154. The mainstay for treatment of
acute and subacute MVT is the prompt initiation of systemic
anticoagulation, which improves survival and reduces the
risk of recurrence.41 Heparin should be initiated with a bolus
fol- lowed by a continuous infusion to keep the partial
throm- boplastin time between 50 and 70 seconds.
Intravenous antibiotics should be administered to decrease
bacterial trans- location from the intestinal mucosa.
Aggressive fluid resusci- tation and circulatory support
should be performed because of severe bowel edema and
shifting of fluid into the peritoneal cavity. Nasogastric
decompression, bowel rest, and adminis- tration of total
parenteral nutrition are also indicated.

Mean Follow-Up

Patients with suspected peritonitis, severe gastrointestinal

bleeding, or intestinal stricture require abdominal exploration and resection of nonviable bowel. Bowel resection
should be conservative in patients with MVT. We frequently
perform a second-look laparotomy 24 hours after the initial
operation, which helps avoid resection of viable bowel
during the initial operation. Several case reports have been
pub- lished describing the successful use of direct
percutaneous thrombolytic infusion into the mesenteric
veins through the transjugular115 or transhepatic route116 or
indirectly through the SMA.117 However, Grisham et al118
reviewed the outcomes of 24 patients with MVT and noted a
sig- nificantly higher mortality rate in those treated with
throm- bolytic therapy compared with those receiving
systemic anticoagulation alone. There was no significant
difference in length of hospital stay between the two
groups. In con- trast, Semiz-Oysu et al119 performed multiple
percutaneous interventions for prehepatic MVT, including
venous recana- lization (n = 19), thrombolysis (n = 1), and
mechanical thrombectomy (n = 5). They noted an
improvement in symptoms in 83% of patients treated,
with relatively low
mortality (13.6%). Other endovascular treatment options
include mechanical or aspiration thrombectomy, angioplasty,
and portosystemic shunt.
Dentali et al120 performed the first large, multicenter, retrospective review to determine the efficacy of oral
anticoagu- lant therapy for prevention of MVT recurrence.
After a median follow-up of 36 months, reported recurrence
rates were 45.9 events/1000 patient-years for patients who
had ceased anticoagulation therapy and 10.5 events/1000
patient- years for those with ongoing treatment. Major
bleeding events occurred in 2.6% while receiving
anticoagulation. In light of such findings, lifelong
anticoagulation with vitamin K antagonists is recommended
for patients with MVT because recurrence is common (up to
36% without anticoagulation) and potentially catastrophic.
Patients may be transitioned from heparin to a vitamin K
antagonist once bowel function has normalized.




Mesenteric Vascular


Acosta S: Epidemiology of mesenteric vascular disease: clinical
Semin Vasc Surg 23:48, 2010.
One of the few large-scale, population-based autopsy studies describing
the epidemiology of mesenteric ischemia.
Arthurs ZM, Titus J, Bannazadeh M, Eagleton MJ, Srivastava S, Sarac TP,
Clair DG: A comparison of endovascular revascularization with traditional therapy for the treatment of acute mesenteric ischemia. J Vasc Surg
53:698704; discussion 704705, 2011.
A description of contemporary outcomes after open and endovascular repair
for acute mesenteric ischemia.
Baker AC, Chew V, Li CS, Lin TC, Dawson DL, Pevec WC, Hedayati N:
Application of duplex ultrasound imaging in determining in-stent stenosis
during surveillance after mesenteric artery revascularization. J Vasc Surg
56:13641371, 2012.
Provides evidence that surveillance duplex ultrasound findings have poor correlation with in-stent restenosis after mesenteric angioplasty and stenting.
Foley MI, Moneta GL, Abou-Zamzam AM, Jr, Edwards JM, Taylor LM, Jr,
Yeager RA, Porter JM: Revascularization of the superior mesenteric artery
alone for treatment of intestinal ischemia. J Vasc Surg 32:3747, 2000.

Landmark paper supporting the acceptability of results of retrograde

mesenteric bypass for chronic mesenteric ischemia.
Jimenez JG, Huber TS, Ozaki CK, Flynn TC, Berceli SA, Lee WA, Seeger
JM: Durability of antegrade synthetic aortomesenteric bypass for chronic
mesenteric ischemia. J Vasc Surg 35:10781084, 2002.
Provides evidence supporting the excellent durability and symptom-free
survival associated with antegrade mesenteric bypass.
Oderich GS, Bower TC, Sullivan TM, Bjarnason H, Cha S, Gloviczki P:
Open versus endovascular revascularization for chronic mesenteric ischemia: risk-stratified outcomes. J Vasc Surg 49:14721479.e3, 2009.
Comparison of open versus endovascular treatment for chronic mesenteric
isch- emia documenting similar mortality and efficacy and greater morbidity for
open revascularization but higher rates of recurrence and reintervention for
endovascular revascularization.
Wyers MC, Powell RJ, Nolan BW, Cronenwett JL: Retrograde mesenteric
stenting during laparotomy for acute occlusive mesenteric ischemia. J
Vasc Surg 45:269275, 2007.
Description of technique and outcomes of retrograde mesenteric stenting.
The reference list can be found on the companion Expert Consult website

1. Bacelli F, cited by Goodman EH: Angina abdominis. Am J Med Sci
155:524528, 1918.
2. Dunphy JE: Abdominal pain of vascular origin. Am J Med Sci
192:109 113, 1936.
3. Warren S, et al: Mesenteric venous thrombosis. Surg Gynecol Obstet
61:102121, 1935.
4. Klass AA: Embolectomy in acute mesenteric occlusion. Ann Surg
134:913917, 1951.
5. Shaw RS, et al: Acute and chronic thrombosis of the mesenteric arteries associated with malabsorption; a report of two cases successfully
treated by thromboendarterectomy. N Engl J Med 258:874878, 1958.
6. Morris GC, Jr, et al: Revascularization of the celiac and superior mesenteric arteries. Arch Surg 84:95107, 1962.
7. Stoney RJ, et al: Recognition and surgical management of visceral
ischemic syndromes. Ann Surg 164:714722, 1966.
8. Furrer J, et al: Treatment of abdominal angina with percutaneous dilatation of an arteria mesenterica superior stenosis. Preliminary communication. Cardiovasc Intervent Radiol 3:4344, 1980.
9. Novelline RA: Percutaneous transluminal angioplasty: newer applications. AJR Am J Roentgenol 135:983988, 1980.
10.Finch IJ: Use of the Palmaz stent in ostial celiac artery stenosis. J Vasc
Interv Radiol 3:633635; discussion 636637, 1992.
11.Lin PH, et al: Embryology, anatomy, and surgical exposure of the great
abdominal vessels. Surg Clin North Am 80:417433, xiv, 2000.
12.Gallavan RH, et al: Pathophysiology of the gastrointestinal system,
Bethesda, Md, 1989, American Physiological Society.
Rosenblum JD, et al: The mesenteric circulation. Anatomy and
physi- ology. Surg Clin North Am 77:289306, 1997.
Perko MJ, et al: Mesenteric, coeliac and splanchnic blood flow in
humans during exercise. J Physiol 513(Pt 3):907913, 1998.
Wilson DB, et al: Clinical course of mesenteric artery stenosis in
elderly Americans. Arch Intern Med 166:20952100, 2006.
Ulker P, et al: Nitric oxide generated by red blood cells following
exposure to shear stress dilates isolated small mesenteric arteries under
hypoxic conditions. Clin Hemorheol Microcirc 2012 [Epub ahead of
Hansen MB, et al: Profile of neurohumoral agents on mesenteric
and intestinal blood flow in health and disease. Physiol Res 47:307
327, 1998.
Vollmar B, et al: Intestinal ischemia/reperfusion: microcirculatory
pathology and functional consequences. Langenbecks Arch Surg
396:13 29, 2011.
Foley MI, et al: Revascularization of the superior mesenteric
artery alone for treatment of intestinal ischemia. J Vasc Surg 32:3747,
Lo RC, et al: The decline of mesenteric ischemia-related mortality
in the last decade. Abstract presented at the 41st Annual Symposium
of the Society for Clinical Vascular Surgery, Miami, Fla., March 16,
Go AS, et al: Anticoagulation therapy for stroke prevention in
atrial fibrillation: how well do randomized trials translate into clinical
prac- tice? JAMA 290:26852692, 2003.
Schermerhorn ML, et al: Mesenteric revascularization:
management and outcomes in the United States, 1988-2006. J Vasc
Surg 50:341348. e1, 2009.
Mitchell EL, et al: Mesenteric duplex scanning. Perspect Vasc
Surg Endovasc Ther 18:175183, 2006.
Thomas JH, et al: The clinical course of asymptomatic mesenteric
arterial stenosis. J Vasc Surg 27:840844, 1998.
Bradbury AW, et al: Mesenteric ischaemia: a multidisciplinary
Br J Surg 82:14461459, 1995.
Shih MC, et al: CTA and MRA in mesenteric ischemia: part 1,
role in diagnosis and differential diagnosis. AJR Am J Roentgenol
188:452 461, 2007.
Mansour MA: Management of acute mesenteric ischemia. Arch
134:328330; discussion 331, 1999.
Oldenburg WA, et al: Acute mesenteric ischemia: a clinical

Arch Intern Med 164:10541062, 2004.


Schoots IG, et al: Systematic review of survival

after acute mesenteric ischaemia according to disease
aetiology. Br J Surg 91:1727, 2004.
Acosta S, et al: Clinical implications for the
management of acute thromboembolic occlusion of the
superior mesenteric artery: autopsy findings in 213
patients. Ann Surg 241:516522, 2005.
Acosta S: Epidemiology of mesenteric vascular
disease: clinical impli- cations. Semin Vasc Surg 23:48,
Bjorck M, et al: Nonocclusive mesenteric
hypoperfusion syndromes: recognition and treatment.
Semin Vasc Surg 23:5464, 2010.
Quiroga B, et al: Detection of patients at high
risk for non-occlusive mesenteric ischemia in
hemodialysis. J Surg Res 180:5155, 2013.
Li SY, et al: Mesenteric ischemia in patients with
end-stage renal disease: a nationwide longitudinal study.
Am J Nephrol 35:491497, 2012.
Archodovassilis F, et al: Nonocclusive mesenteric
ischemia: a lethal complication in peritoneal dialysis
patients. Perit Dial Int 27:136141, 2007.
Grendell JH, et al: Mesenteric venous thrombosis.
82:358372, 1982.
37.Kumar S, et al: Mesenteric venous thrombosis. N Engl J
Med 345:1683 1688, 2001.
Jackson CS, et al: Mesenteric vascular
thromboembolism in inflamma- tory bowel disease: a
single center experience. J Gastrointest Surg 15:97
100, 2011.
Gonzelez HJ, et al: Splanchnic vein thrombosis in
severe acute pan- creatitis: a 2-year, single-institution
experience. HPB (Oxford) 13:860 864, 2011.
Boley SJ, et al: History of mesenteric ischemia.
The evolution of a diagnosis and management. Surg
Clin North Am 77:275, 1997.
Deitch EA, et al: Effect of hemorrhagic shock on
bacterial transloca- tion, intestinal morphology, and

intestinal permeability in conven- tional and antibioticdecontaminated rats. Crit Care Med 18:529536, 1990.
Acosta S, et al: Mesenteric venous thrombosis with transmural
intestinal infarction: a population-based study. J Vasc Surg 41:5963,
Acosta S, et al: D-dimer testing in patients with suspected acute
throm- boembolic occlusion of the superior mesenteric artery. Br J
Surg 91: 991994, 2004.
Mitsuyoshi A, et al: Survival in nonocclusive mesenteric
ischemia: early diagnosis by multidetector row computed tomography
and early treatment with continuous intravenous high-dose
prostaglandin E1. Ann Surg 246:229235, 2007.
White CJ: Chronic mesenteric ischemia: diagnosis and management.
Prog Cardiovasc Dis 54:3640, 2011.
Park WM, et al: Current results of open revascularization for
chronic mesenteric ischemia: a standard for comparison. J Vasc Surg
35:853 859, 2002.
Dietrich CF, et al: Sonographic assessment of splanchnic arteries
and the bowel wall. Eur J Radiol 64:202212, 2007.
Rhee RY, et al: Mesenteric venous thrombosis: still a lethal
disease in the 1990s. J Vasc Surg 20:688697, 1994.
Moneta GL, et al: Mesenteric duplex scanning: a blinded
prospective study. J Vasc Surg 17:7984; discussion 8586, 1993.
Moneta GL, et al: Duplex ultrasound criteria for diagnosis of
splanch- nic artery stenosis or occlusion. J Vasc Surg 14:511518;
discussion 518520, 1991.
Geelkerken RH, et al: Pitfalls in the diagnosis of origin stenosis
of the coeliac and superior mesenteric arteries with transabdominal
color duplex examination. Ultrasound Med Biol 22:695700, 1996.
Baker AC, et al: Application of duplex ultrasound imaging in
deter- mining in-stent stenosis during surveillance after mesenteric
artery revascularization. J Vasc Surg 56:13641371, 2012.
Schoch DM, et al: Management of chronic mesenteric vascular
insuf- ficiency: an endovascular approach. J Am Coll Surg 212:668
675; dis- cussion 675677, 2011.
Liem TK, et al: Duplex scan characteristics of bypass grafts to
mesen- teric arteries. J Vasc Surg 45:922927; discussion 927928,




Mesenteric Vascular Disease

Menke J: Diagnostic accuracy of multidetector CT in acute

mesenteric ischemia: systematic review and meta-analysis. Radiology
256:93101, 2010.
Otte JA, et al: Jejunal tonometry for the diagnosis of
gastrointestinal ischemia. Feasibility, normal values and comparison
of jejunal with gastric tonometry exercise testing. Eur J Gastroenterol
Hepatol 20:62 67, 2008.
Sana A, et al: Radiological imaging and gastrointestinal tonometry
add value in diagnosis of chronic gastrointestinal ischemia. Clin
Gastroen- terol Hepatol 9:234241, 2011.
Jamieson WG, et al: Effect of antibiotic and fluid resuscitation
upon survival time in experimental intestinal ischemia. Surg Gynecol
Obstet 167:103108, 1988.
Robertson L, et al: Antiplatelet and anticoagulant drugs for
prevention of restenosis/reocclusion following peripheral endovascular
treatment. Cochrane Database Syst Rev 8:CD002071, 2012.
Fioole B, et al: Percutaneous transluminal angioplasty and stenting
as first-choice treatment in patients with chronic mesenteric
ischemia. J Vasc Surg 51:386391, 2010.
Pecoraro F, et al: Chronic mesenteric ischemia: critical review and
guidelines for management. Ann Vasc Surg 27:113122, 2013.
Peck MA, et al: Intermediate-term outcomes of endovascular
treat- ment for symptomatic chronic mesenteric ischemia. J Vasc Surg
51:140147.e1-2, 2010.
Arthurs ZM, et al: A comparison of endovascular
revascularization with traditional therapy for the treatment of acute
mesenteric isch- emia. J Vasc Surg 53:698704; discussion 704705,
64.Ryer EJ, et al: Revascularization for acute mesenteric ischemia. J Vasc
Surg 55:16821689, 2012.
Matsumoto AH, et al: Percutaneous transluminal angioplasty and
stenting in the treatment of chronic mesenteric ischemia: results and
longterm followup. J Am Coll Surg 194(Suppl):S22S31, 2002.
Sharafuddin MJ, et al: Endovascular treatment of celiac and
mesenteric arteries stenoses: applications and results. J Vasc Surg
38:692698, 2003.
Landis MS, et al: Percutaneous management of chronic mesenteric
ischemia: outcomes after intervention. J Vasc Interv Radiol 16:1319
1325, 2005.
Silva JA, et al: Endovascular therapy for chronic mesenteric
J Am Coll Cardiol 47:944950, 2006.
Sarac TP, et al: Endovascular treatment of stenotic and occluded
vis- ceral arteries for chronic mesenteric ischemia. J Vasc Surg
47:485491, 2008.
Brown DJ, et al: Mesenteric stenting for chronic mesenteric
J Vasc Surg 42:268274, 2005.
van Wanroij JL, et al: Endovascular treatment of chronic
splanchnic syndrome. Eur J Vasc Endovasc Surg 28:193200, 2004.
Dias NV, et al: Mid-term outcome of endovascular
revascularization for chronic mesenteric ischaemia. Br J Surg 97:195
201, 2010.
Oderich GS, et al: Open versus endovascular revascularization for
chronic mesenteric ischemia: risk-stratified outcomes. J Vasc Surg
49:14721479.e3, 2009.
Atkins MD, et al: Surgical revascularization versus endovascular
therapy for chronic mesenteric ischemia: a comparative experience.
J Vasc Surg 45:11621171, 2007.
Davies RS, et al: Surgical versus endovascular reconstruction for
chronic mesenteric ischemia: a contemporary UK series. Vasc
Endovas- cular Surg 43:157164, 2009.
van Petersen AS, et al: Open or percutaneous revascularization
for chronic splanchnic syndrome. J Vasc Surg 51:13091316, 2010.
Tallarita T, et al: Reinterventions for stent restenosis in patients
treated for atherosclerotic mesenteric artery disease. J Vasc Surg
54:14221429. e1, 2011.
Sharafuddin MJ, et al: Endovascular recanalization of total
occlusions of the mesenteric and celiac arteries. J Vasc Surg 55:1674
1681, 2012.


Milner R, et al: Superior mesenteric artery angioplasty and stenting

via a retrograde approach in a patient with bowel ischemiaa case
report. Vasc Endovascular Surg 38:8991, 2004.


Wyers MC, et al: Retrograde mesenteric stenting during

laparotomy for acute occlusive mesenteric ischemia. J Vasc Surg
45:269275, 2007.
81. Moyes LH, et al: Intraoperative retrograde mesenteric
angioplasty for acute occlusive mesenteric ischaemia: a case series.
Eur J Vasc Endovasc Surg 36:203206, 2008.
82. Pisimisis GT, et al: Technique of hybrid retrograde superior
mesenteric artery stent placement for acute-on-chronic mesenteric
ischemia. Ann Vasc Surg 25:132.e711, 2011.
83. Stout CL, et al: Retrograde open mesenteric stenting for acute
mesen- teric ischemia is a viable alternative for emergent
revascularization. Vasc Endovascular Surg 44:368371, 2010.
84. Sivamurthy N, et al: Endovascular versus open mesenteric
revascular- ization: immediate benefits do not equate with shortterm functional outcomes. J Am Coll Surg 202:859867, 2006.
85. Turba UC, et al: Chronic mesenteric ischaemia: 28-year
experience of endovascular treatment. Eur Radiol 22:13721384,
86. Oderich GS, et al: Anatomic measurements and factors
associated with embolic events during superior mesenteric artery
stenting: implications for use of embolic protection devices.
Abstract book of the 37th Society for Clinical Vascular Surgery
Annual Symposium, 2009.
87. Oderich GS, et al: Mesenteric artery complications during
angioplasty and stent placement for atherosclerotic chronic
mesenteric ischemia. J Vasc Surg 55:10631071, 2012.
88. Klepczyk L, et al: Superior mesenteric artery stent fracture
producing stenosis and recurrent chronic mesenteric ischemia: case
report. Vasc Endovascular Surg 42:7981, 2008.
89. Schellekens JF, et al: Superior mesenteric artery stent fracture
leading to recurrent mesenteric ischemia. Vasc Endovascular Surg
45:654659, 2011.
90. Moore M, et al: Reperfusion hemorrhage following superior
mesenteric artery stenting. Cardiovasc Intervent Radiol 31(Suppl
2):S57S61, 2008.
91. Urbanavicius L, et al: How to assess intestinal viability during
surgery: a review of techniques. World J Gastrointest Surg 3:5969,


Mell MW, et al: Outcomes after endarterectomy for chronic

mesenteric ischemia. J Vasc Surg 48:11321138, 2008.
93. Bergman RT, et al: The role of intravenous fluorescein in the
detection of colon ischemia during aortic reconstruction. Ann Vasc
Surg 6:7479, 1992.
94. Rapp JH, et al: Durability of endarterectomy and antegrade grafts
in the treatment of chronic visceral ischemia. J Vasc Surg 3:799
806, 1986.
95. Jimenez JG, et al: Durability of antegrade synthetic
aortomesenteric bypass for chronic mesenteric ischemia. J Vasc Surg
35:10781084, 2002.
96. Cunningham CG, et al: Chronic visceral ischemia. Three decades
of progress. Ann Surg 214:276287; discussion 287288, 1991.
97. English WP, et al: Chronic visceral ischemia: symptom-free
survival after open surgical repair. Vasc Endovascular Surg 38:493
503, 2004.
98. Mukherjee D, et al: Splenic arterytosuperior mesenteric artery
bypass for chronic mesenteric ischemiaa case report. Vasc
Endovascular Surg 38:465468, 2004.
99. Wylie EJ, et al: Manual of vascular surgery, New York, 1980,
Springer- Verlag.
100. Oderich GS, et al: Open surgical treatment for chronic mesenteric
ischemia in the endovascular era: when it is necessary and what is the
preferred technique? Semin Vasc Surg 23: 3646, 2010.
101. Kansal N, et al: A comparison of antegrade and retrograde
mesenteric bypass. Ann Vasc Surg 16:591596, 2002.
102. Horton JW, et al: Oxygen radicals, lipid peroxidation, and
permeability changes after intestinal ischemia and reperfusion. J
Appl Physiol 74: 15151520, 1993.
103. Horton JW, et al: Free radical scavengers prevent intestinal
ischemia- reperfusionmediated cardiac dysfunction. J Surg Res
55:282289, 1993.
104. Horton JW, et al: Cardiac contractile injury after intestinal
ischemia- reperfusion. Am J Physiol 261(Pt 2):H1164H1170, 1991.
105. Gerkin TM, et al: Intestinal ischemia-reperfusion injury causes
pulmo- nary endothelial cell ATP depletion. Ann Surg 217:4856,

106. Schoenberg MH, et al: Oxygen radicals in intestinal ischemia and
reperfusion. Chem Biol Interact 76:141161, 1990.
107. McCallion K, et al: Ischemic preconditioning ameliorates
hyperpermeability in rats. Shock 14:429434, 2000.
108. Aksoyek S, et al: Intestinal ischemic preconditioning protects the
intestine and reduces bacterial translocation. Shock 18:476480, 2002.
109. Moore-Olufemi SD, et al: Intestinal ischemic preconditioning
after ischemia/reperfusion injury in rat intestine: profiling global gene
expression patterns. Dig Dis Sci 55:18661877, 2010.
110. Liu KX, et al: Immediate postconditioning during reperfusion
attenu- ates intestinal injury. Intensive Care Med 35:933942, 2009.
111. Kougias P, et al: Determinants of mortality and treatment outcome
following surgical interventions for acute mesenteric ischemia. J Vasc
Surg 46:467474, 2007.
112. Kruger AJ, et al: Open surgery for atherosclerotic chronic
mesenteric ischemia. J Vasc Surg 46:941945, 2007.
113. Cho JS, et al: Long-term outcome after mesenteric artery
reconstruc- tion: a 37-year experience. J Vasc Surg 35:453460, 2002.
114. Kang H, et al: Intravenous iloprost increases mesenteric blood
flow in experimental acute nonocclusive mesenteric ischemia. Crit
Care Med 30:25282534, 2002.

Mesenteric Vascular Disease: General


115. Wang MQ, et al: Acute symptomatic mesenteric venous

thrombosis: treatment by catheter-directed thrombolysis with
transjugular intrahe- patic route. Abdom Imaging 36:390398,
116. Goldberg MF, et al: Treatment of acute superior mesenteric vein
throm- bosis with percutaneous techniques. AJR Am J Roentgenol
181:1305 1307, 2003.
117. da Motta Leal Filho JM, et al: Infusion of recombinant human
tissue plasminogen activator through the superior mesenteric artery
in the treatment of acute mesenteric venous thrombosis. Ann Vasc
Surg 25:840.e14, 2011.
118. Grisham A, et al: Deciphering mesenteric venous thrombosis:
imaging and treatment. Vasc Endovascular Surg 39:473479, 2005.
119. Semiz-Oysu A, et al: Interventional radiological management of
prehepatic obstruction of [corrected] the splanchnic venous system.
Cardiovasc Intervent Radiol 30:688695, 2007.
120. Dentali F, et al: Natural history of mesenteric venous
thrombosis in patients treated with vitamin K antagonists: a multicentre, retrospec- tive cohort study. Thromb Haemost 102(3):501504,
doi: 10,1160/ TH08-12-0842, 2009.



Based on a chapter in the seventh edition by Thomas S. Huber and W. Anthony Lee

Vascular Pathophysiology
Chronic Ischemia
he first clinical and anatomic descriptions of intestinal
ischemia were recognized by Chienne in 1869 and
Councilman in 1894.1,2 Goodman in 1918 associated the
symptoms of postprandial abdominal pain with those of
patients with angina pectoris.3 Dunphy from the Peter Bent
Brigham Hospital reported in 1936 the correlation between
recurrent abdominal pain and fatal intestinal infarction
from occlusive mesenteric arterial disease.4 In that report,
60% of patients who died of intestinal infarction had a
history of recurrent abdominal pain that preceded the fatal
event by weeks, months, or years. Since then, the term
intestinal angina has been coined to describe the classic
symptom of chronic abdominal pain that occurs after meals,
which is the cardinal symptom of chronic mesenteric

Current estimates indicate that chronic mesenteric ischemia
accounts for less than 1 per 100,000 hospital admissions
in the United States and less than 2% of all admissions
for gastrointestinal conditions.5 Since the first successful
mesenteric endarterectomy by Shaw and Maynard in 1958,
techniques of revascularization have greatly evolved.6
Advances in diagnostic imaging, medical therapy, surgical
techniques, and endovascular technology resulted in
improved outcomes. Balloon angioplasty was reported for
treatment of mesenteric arterial stenoses by Uflacker, Furrer,
Gruntzig, and colleagues in 1980.7,8 During the last decade,
mesenteric angioplasty and stenting gained widespread
acceptance and became the most frequently used treatment
of chronic mesenteric ischemia, relegating open surgery to
patients who fail to respond to endovascular therapy or
who have complex lesions unsuitable to it. 9 This chapter
provides a comprehensive review of the pathophysiology,
clinical presentation, indications, techniques, and outcomes
of revascularization in patients with chronic mesenteric

Approximately 20% of the cardiac output goes through

the mesenteric arteries under normal conditions.10 Blood
S. ODERICHtract increases shortly after the
to the gastrointestinal
ingestion of a meal, remaining elevated at levels
approaching 100% to 150% of normal (2000 mL/min)
during the next 3 to 6 hours.11 This normal hyperemic
postprandial response is mediated by cardiovascular
changes that accompany the ingestion and digestion of
food. These changes start with anticipation of a meal and
increase considerably with pres- ence of food in the
stomach and intestine, including an increase in cardiac
output, heart rate, and blood pressure. Mesenteric
vasodilatation starts 3 to 5 minutes after food enters the
intestine, reaching its maximum 30 to 90 minutes later
and lasting 4 to 6 hours. The latency and duration of these
responses depend on the type and quantity of a meal, with
high-fat and protein-containing foods producing the most
profound and sustained intestinal hyperemia.12
Postprandial mesenteric hyperemia is confined to organs
in which digestion is occurring but is not shared equally
within the same mesenteric arterial territory. The increased
blood flow in the superior mesenteric artery (SMA) territory
elicited by food in the intestine is associated with little or no
change in blood flow to the stomach, pancreas, and colon.
In conscious humans, Moneta et al 13 have shown marked
increase in end-diastolic velocity in the SMA, with minimal
change in velocities in the celiac axis, presumably because of
the relatively low resistance on the splenic and hepatic circulations at baseline. Portal venous and hepatic blood flow
also increase. At the level of the intestinal wall, blood flow
distribution favors the mucosa rather than the submucosa and
Patients with mesenteric ischemia fail to achieve the postprandial hyperemic response that is required to supply
oxygen for the metabolic processes of secretion and
absorption and for increased peristaltic activity.15 Just as in
the patient with ischemic cardiomyopathy, in whom angina
pectoris occurs from inadequate supply of oxygen, intestinal
angina results from the relative imbalance between tissue



Mesenteric Vascular

Figure 152-1 Mesenteric artery circulation and

common collateral pathways in patients with
severe occlusive mesen- teric artery disease.
Note severe disease at the celiac axis, superior
mesenteric artery (IMA). Common collateral
pathways include the arc of Riolan between the
left colic artery (IMA) and middle colic artery
(SMA). The celiac axis and SMA have
collateralization via the pancreaticoduodenal
arcade (arc of Buhler) and the gas- troduodenal

demand for oxygen and other metabolites. At the tissue and

cellular level, the lack of adenosine triphosphate metabolism
affects intestinal mucosa, muscularis, and visceral nerves,
causing failure of most intestinal mucosal transport pathways
and contracture of the muscle layer with inadequate relaxation, resulting in malabsorption and abdominal pain.16,17
The mesenteric circulation is rich in collateral networks
between the three main visceral artery territories (celiac
axis, SMA, and inferior mesenteric artery [IMA]) and
the internal iliac arteries (Fig. 152-1). Direction of blood
flow is contingent on the location of significant stenoses.
The gastroduodenal and pancreaticoduodenal arteries
provide collateralization between the celiac axis and SMA.
The marginal artery of Drummond and the arc of Riolan
connect the left colic artery (IMA) to the middle colic artery
(SMA). Meandering mesenteric or central anastomotic
artery describes marked enlargement that occurs in the arc of
Riolan in patients with high-grade stenosis or occlusion of
the SMA
and collateralization via a patent IMA. This artery lies in
the mesentery close to the inferior mesenteric vein. Inadvertent ligation, division, or thrombosis of this important
collateral artery during aortic exposure or other operative
procedures may result in acute ischemia or bowel gangrene.

Because of the extensive collateral network, the majority

of patients with symptoms of chronic mesenteric ischemia
have significant stenosis or occlusion of at least two of the
three mesenteric arteries. In the last Mayo Clinic review of
229 mesenteric arteriograms, 98% of patients with chronic
mesenteric ischemia had two- or three-vessel involvement,
with occlusion or critical stenosis of the SMA in 92%.19
However, contrary to what has been propagated in many
surgical textbooks, this is not an absolute requirement.20,21
The clinical significance of ischemia correlates not only to
the extent of disease but also to the adequacy of collateral
pathways, acuteness of symptoms, and presence of arterial
steal; approximately 2% to 10% of patients with chronic
mesenteric ischemia have single-vessel disease, which
affects primarily the SMA in patients with poorly
developed col- laterals or more acute presentation, as might
be predicted from the postprandial hyperemic response.19


The internal iliac arteries provide a collateral pathway via

the hemorrhoidal branches.

The most common cause of chronic mesenteric ischemia is

atherosclerotic disease, accounting for more than 90% of
cases in most series. Atherosclerotic lesions usually affect


Mesenteric Vascular Disease: Chronic

origin or the proximal 2 to 3 cm of the mesenteric
arteries, frequently with associated plaque in the aorta
and renal

arteries. The prevalence of mesenteric atherosclerotic disease

ranges from 6% to 10% in autopsy studies to 14% to 24% in
patients undergoing aortography for investigation of
coronary or peripheral arterial disease. Imaging studies
typically dem- onstrate stigmata of atherosclerosis (e.g.,
calcified plaque, atheromatous debris) in multiple vascular
beds, notably the coronary, carotid, renal, and aortoiliac
Nonatherosclerotic lesions can also affect the mesenteric
arteries and tend to occur in younger patients. A variety of
diagnoses merit investigation, including vasculitis (giant cell
arteritis, Takayasus disease, and polyarteritis nodosa), systemic lupus erythematosus, Buergers disease, mesenteric or
neurofibromatosis, radiation arteritis, mesenteric venous
stenosis or occlusion, and drug-induced arteriopathy from
cocaine or ergot use. Abdominal aortic coarctation or
midaortic syndrome can also be manifested with symptoms
of mesenteric ischemia.

Natural History
The natural history of mesenteric arterial disease has not
been completely defined. It is generally accepted that asymptomatic lesions carry a benign course, not justifying prophylactic revascularization. Nonetheless, the observation that
15% to 50% of patients who present with bowel gangrene
have thrombosis of preexisting lesions with no antecedent
warning signs suggests that these lesions are not entirely
benign.22,23 The likelihood of symptom progression seems to
depend on the extent of disease. Wilson et al24 reported a
large prospective cohort study of 553 elderly patients who
were screened for mesenteric artery disease with duplex
ultra- sound. The prevalence of mesenteric stenosis greater
than 70% or occlusion was 18%. After a follow-up period
of 7 years, none of the patients developed symptoms of
mesenteric ischemia, but most had single-vessel disease,
only 14% had SMA stenoses, and none had three-vessel
involvement. Thomas et al25 reviewed 980 aortograms and
found 60 patients (6%) with significant mesenteric artery
disease (>50% stenosis). Of these, 15 had involvement of
all three visceral arteries. During follow-up of 2.6 years, four
patients (27%) developed symptoms, three had successful
revascular- ization, but one died of acute ischemia.
The natural history of patients with symptoms of chronic
mesenteric ischemia is even less well understood because
revascularization is typically recommended. There are no
cohort studies with a control or medical treatment arm.
It is generally accepted that once a patient develops symptoms of chronic ischemia, there is considerable risk of
progression to cachexia or bowel gangrene, and revascularization is indicated.

The diagnosis of chronic mesenteric ischemia is suggested
by clinical history and confirmed by one or more diagnostic
studies, such as duplex ultrasonography, magnetic resonance

24-hour or exercise tonometry, if it is available in centers

with established protocols, can be useful in the differential
diagnosis of patients with atypical symptoms, single-vessel
disease, or median arcuate ligament syndrome. Tests of intestinal absorptive and excretory function have not been

Clinical Presentation


The typical patient with chronic mesenteric ischemia is

female with a median age of 65 years, with age ranging from
40 to 90 years.9,19,27 Most studies quote a 3 : 1 or 4 : 1 femaleto-male ratio. A survey of the Nationwide Inpatient Sample
examining the outcomes of mesenteric revascularizations in
22,413 patients reported a mean age of 66 years and that
76% of the patients were women.9 In patients with
nonatheroscle- rotic lesions, the female-to-male ratio varies
from 1 : 1 or 1 : 2 for coarctation and neurofibromatosis to
5 : 1 for vasculitis. Most patients with nonatherosclerotic
disease are in their third or fourth decade of life.28-30
The classic symptoms of chronic mesenteric ischemia
include abdominal pain, weight loss, and food fear. The
abdominal pain is often postprandial and begins within a few
minutes to 30 minutes after meals, persisting for as long as 5
to 6 hours. It is usually midabdominal in location and
crampy or dull. Patients may describe intolerance to certain
types of food, and consequently they alter their eating habits
to avoid foods that precipitate symptoms. In fact, some
patients do not report any pain at time of presentation
because of their adap- tive strategies to reduce or to relieve
it. Unintentional weight loss progresses to malnutrition and
cachexia, which is often present at the time of intervention.
In the last review of the Mayo Clinic, which included 357
consecutive patients treated by revascularization since 1990,
abdominal pain was present in 96% (postprandial in 74%)
and weight loss in 84%. Duration of symptoms, or time
delay, before revascular- ization averages 15 months. The
pattern of symptoms often changes and progresses in
intensity, frequency, and severity. Subacute mesenteric
ischemia is characterized by progression of pain from
intermittent to unremitting or continuous during the course
of days or a few weeks, warranting immediate
revascularization. Physical examination is nonspecific, does
not commonly reveal any pathognomonic findings, but can
point to the diagnosis. Many individuals are thin at the onset
of their symptoms and progress to cachexia, particularly
those with significant delay at time of diagnosis. These
patients have obvious signs of malnutrition, muscle wasting,
and a flat or scaphoid abdomen. Pain may be present but not
localized or aggravated by abdominal palpation; the finding
of pain out of proportion to physical findings should point
toward the diagnosis of mesenteric ischemia. An abdominal
bruit may be
noted in up to 50% of patients; it differs from the bruit caused
angiography (MRA), computed tomography angiography (CTA),
conventional aortography. Endoscopy with


by compression of the celiac axis by the median

arcuate liga- ment, which is elicited by deep
expiration and elevation of the diaphragm. A
complete vascular examination may elicit a

diminished ankle-brachial index or absent peripheral pulses

and bruit in other vascular beds (e.g., carotid, subclavian, and
iliacs). Laboratory test results are nonspecific or unremarkable in the absence of acute symptoms but may demonstrate

malnutrition (decreased serum albumin, transferrin, and prealbumin levels), systemic inflammation (e.g., vasculitis), and
elevation of L-lactate and D-dimer after meal challenge.31
The clinical presentation can be less specific in some
patients. Vague abdominal pain, nausea, vomiting, or change
in bowel habits, without the classic postprandial component
to the pain, can make the diagnosis difficult to ascertain; in
these patients, 24-hour gastric tonometry has been useful if
it is available.32-39 Liver function abnormalities or endoscopic
evidence of diffuse small ulcerations in the stomach or proximal duodenum or patchy areas of ischemia in the colon are
not uncommon. A previous history of smoking and the diagnoses of hypertension and hyperlipidemia are documented in
60% to 70%.19,40,41 Consequently, patients often have other
manifestations of atherosclerotic disease affecting the coronary (50% to 70%), cerebrovascular (20% to 45%), and
peripheral (20% to 35%) arteries. Concomitant renal artery
disease with difficult to control hypertension or ischemic
nephropathy is not uncommon.42 The presence of severe
aortic or peripheral arterial occlusive disease represents a
challenge in terms of source of inflow for bypass, access for
percutaneous procedures, or sequence of revascularization in
patients with limb-threatening ischemia.

Diagnostic Tests
It is not infrequent for patients with chronic mesenteric
ischemia to experience significant delay in diagnosis or to
undergo an extensive evaluation to rule out other causes of
chronic abdominal pain and weight loss. The differential
diagnosis is extensive, including inflammatory, infectious,
and malignant disease. The investigation often includes
upper and lower gastrointestinal endoscopy and crosssectional imaging of the abdomen with either computed
tomography or magnetic resonance imaging. Often, the
finding of mesenteric artery stenosis in an imaging study is
the first clue to the diagnosis. If chronic mesenteric ischemia

is suspected on clinical examination, mesenteric duplex

ultrasound is the most frequently used screening study. Nonetheless, the author recommends obtaining additional crosssectional imaging of the abdomen and mesenteric arteries
before proceeding with an intervention. Because abdominal
pain and unintentional weight loss are common presentations of other conditions (e.g., cancer, inflammatory bowel
disease, infections), imaging of the abdomen and pelvis may
help rule out other diagnoses and identify nonatherosclerotic
causes of mesenteric disease. Most important, anatomic
detail about the number of vessels affected and lesion
characteristics (diameter, length, presence of occlusion,
calcification, throm- bus, or tandem lesions) are key factors
that affect selection of the type of revascularization.

Mesenteric Duplex Ultrasound

Mesenteric duplex ultrasound is an excellent screening study
in patients with abdominal pain or an epigastric bruit in
whom mesenteric artery disease is suspected (Fig. 152-2). A
negative duplex ultrasound study essentially excludes the
diagnosis of mesenteric artery disease. In preparation for the
study, patients undergo a minimum 6- to 8-hour fast. A complete study should include B-mode images, spectral analysis,
and velocity measurements. In the celiac axis, velocity measurements are also obtained with deep inspiration and deep
expiration; an increase in peak systolic velocity with deep
expiration is indicative of compression of the celiac axis by
the median arcuate ligament. Mesenteric ultrasound can be
technically challenging because of body habitus, bowel gas,
or unusual anatomy.
Since the original work by Nicholls et al 43 from the University of Washington, the criteria for significant mesenteric
artery stenosis have been established by retrospective studies
and validated by prospective comparison with mesenteric
angiography.44-47 In 1993, Moneta et al48 from Oregon Health
Science University reported the first prospective validation
of mesenteric diagnostic criteria in 100 patients. In that

Figure 152-2 Mesenteric duplex ultrasound with normal and abnormal waveform patterns during the fasting state.
IMA, Inferior mesenteric artery; PSV, peak systolic velocity; SMA, superior mesenteric artery.

the most accurate criteria to indicate a stenosis of 70% or

greater were a peak systolic velocity of more than 275 cm/s
for the SMA and more than 200 cm/s for the celiac axis (see
Fig. 152-2).48 As a criterion for SMA stenosis of 70% or
greater, a peak systolic velocity of more than 275 cm/s had a
sensitivity of 92%, specificity of 96%, positive predictive
value of 80%, and negative predictive value of 99%. A
veloc- ity of 200 cm/s in the celiac axis as a predictor of a
stenosis of 70% or greater had somewhat lower values. The
Bowersox criteria,45 later validated prospectively by the
Dartmouth group,47 are based on 50% or greater arterial
stenosis. For the SMA, an end-diastolic flow velocity of 45
cm/s or higher had 90% sensitivity, 91% specificity, 90%
positive predictive value, 91% negative predictive value, and
91% accuracy. For the celiac axis, reversal of blood flow in
the hepatic artery was 100% predictive of stenosis. An enddiastolic velocity higher than 55 cm/s or no flow signal had
the best overall accuracy (95%) with high sensitivity (93%)
and specificity (94%).
The IMA is the smallest of the mesenteric arteries and
usually gains significance in advanced cases of chronic mesenteric ischemia, in which it provides collateral flow into the
SMA via the meandering mesenteric artery. Because of its
small size, difficult visualization, and minimal clinical
impor- tance, this vessel has gained little attention in the
mesenteric duplex literature. The evaluation of the IMA is
indicated in situations in which colonic ischemia is the
predominant clinical finding. AbuRahma reported the duplex
ultrasound criteria for greater than 50% stenosis in 85
patients with paired
angiography.49 The most

tomography angiography with
reconstruction in a patient with
severe three- vessel mesenteric
disease. A, Note
occlusion of the celiac axis and
su- perior mesenteric artery
(SMA), with collat- eral flow via a
large inferior mesenteric artery
(IMA) and meandering artery.
Collat- eralization from the IMA to
SMA via arc of Riolan (curved
arrow) and marginal artery of
Drummond (arrowhead) and from
the SMA to celiac axis via a large

gastroduodenal artery (straight arrow). B,

Axial view of the SMA demonstrates
occlusion with a small stump (double arrow).

accurate peak systolic velocity was higher than

250 cm/s, with sensitivity of 90%, specificity of
96%, and overall accu- racy of 95%. A ratio of
IMA to aortic peak systolic velocity above 4.0
had overall accuracy of 93%. There have been
no studies with prospective validation of duplex
criteria for the IMA.
Comparison of preprandial and postprandial
blood flow velocities is not specific enough to
distinguish physiologi- cally appropriate from
inappropriate blood flow responses. Despite a
large quantity of data from healthy individuals,
there is little evidence that postprandial
testing adds to the fasting duplex examination.13
Gentile et al50 compared fasting and postprandial
duplex ultrasound velocities with angiographic

stenosis measurements in the SMA of healthy controls and

those with peripheral arterial disease. Combin- ing fasting
and postprandial duplex results increased specific- ity and
positive predictive value slightly but did not improve overall

Multidetector Computed Tomography and

Magnetic Resonance Angiography
Although there is little question about the accuracy of mesenteric duplex ultrasound as a screening study, crosssectional imaging of the abdomen is often needed to rule
out other causes of abdominal pain and to provide anatomic
analysis to plan revascularization.51-53 The choice of CTA or
MRA is somewhat related to individual expertise at the
institution. Most centers use CTA, which allows analysis
with three- dimensional reformatting techniques (Fig. 1523), maximum


intensity projection, and volume rendering. Images can be

analyzed in multiple planes (sagittal, coronal, and axial), and
centerline of flow measurements can be obtained for
accurate measurement of lengths. Multidetector computed
tomogra- phy technology is readily available in most centers,
combin- ing multiple rows of detection with narrow
collimation.54 Multidetector computed tomography has the
highest spatial resolution and finest image detail and is
considered by most the best study to evaluate anatomic
characteristics (calcifica- tion, thrombus, diameters, and
lengths) that are important to plan mesenteric interventions
(Fig. 152-4). In patients with classic symptoms, CTA may
supplement or even replace duplex ultrasound, and it is often
the only imaging study that is obtained before intervention.
It is also useful to objectively assess patency of grafts and
stents. Limitations include radia- tion exposure, cost, risk of
complications related to the con- trast agent (e.g.,
nephropathy, allergic reactions), and, in some cases, difficult
assessment of luminal diameter due to significant vessel
calcification or prior stents.
For patients who present with subacute or acute
symptoms, multidetector computed tomography has the
advantage of

Focal Disease

showing additional imaging signs that are not visible with

catheter angiography. Bowel wall changes indicative of ischemia include circumferential thickening with low (edema)
or high (hemorrhage) attenuation, increased or delayed
enhancement in the bowel wall due to hyperemia, and frank
bowel infarction or perforation with pneumatosis intestinalis,
portal venous air, or pneumoperitoneum. Another potential
application of multidetector computed tomography is the
evaluation of venous thrombosis or narrowing affecting the
mesenteric veins and portal venous system.
Gadolinium-enhanced MRA has advanced in recent years
to provide improved imaging resolution with shorter acquisition times. This imaging modality is used to a lesser
extent but affords many of the same advantages as CTA.55,56
The overall sensitivity and specificity for detection of significant mesenteric artery disease is more than 95%
compared with contrast angiography.57 MRA provides
functional infor- mation by integrating flow dynamics and
blood oxygen saturation techniques with the anatomic
detail, which can be useful in patients with questionable
diagnosis of chronic mesenteric ischemia.58,59 The study is
limited in patients

Complex Disease

Figure 152-4 Computed tomography angiography is the most useful imaging study to plan revascularization.
Anatomic characteristics of the superior mesenteric artery can be used to identify patients with focal disease (A),
for which angioplasty and stenting are favored, and patients with complex disease (B), for which endovascular
therapy is technically more challenging. Lesions with unfavorable anatomy for stenting include heavily calcified
occlusions, long-segment occlusions, and long-segment stenosis involving multiple branches.

with prior stents or excessive calcification. Until recent

reports of complications from gadolinium, MRA was the
preferred imaging modality in patients with chronic kidney
disease. The complication of systemic fibrosis due to gadolinium has significantly decreased its use in this patient
population. Time-of-flight imaging, without gadolinium, has
proved to be inaccurate to estimate the degree of mesenteric

Contrast Arteriography
A catheter-based arteriography is considered the
gold standard diagnostic study for evaluation of mesenteric
artery disease in patients with chronic mesenteric ischemia.
During the last decade, its role as a confirmatory test and for
planning revascularization diminished in favor of the aforementioned noninvasive modalities. Since 2002, the use of
contrast arteriography to plan mesenteric reconstructions at
the Mayo Clinic decreased from 97% to 57%, with an
increase in the use of CTA (55% to 88%) and MRA (12% to
33%).60 Mesenteric arteriography is rarely needed to confirm
the diag- nosis, and it typically does not add anatomic
detail to plan an intervention. More frequently, angiography
is obtained in conjunction with a planned endovascular
intervention. Exceptions are patients with suboptimal
imaging studies and

those with extensive calcification, small vessels, or multiple

prior stents causing metallic artifact.
A diagnostic study includes biplane abdominal aortography with anterior-posterior and lateral projections. A lateral
view is imperative to provide visualization of the origins
of the celiac axis and SMA. A right anterior oblique view
(Fig. 152-5) demonstrates the origin of the IMA. For better
visualization of ostial lesions and quantification of the degree
of stenosis, selective catheterization of the visceral arteries
is necessary. In patients who had prior CTA or MRA,
abdominal aortography may be avoided, unless it is needed
to identify the origin of a vessel before selective catheterization. However, if CTA or MRA has not been performed,
biplane aortography is recommended to define the location,
severity, and extent of visceral artery involvement, to
identify the presence of concomitant lesions in the renal or
iliac arteries, and to indicate the suitability of the
supraceliac or infrarenal aorta as an inflow site if open
reconstruction is contemplated.
Selective catheterization is often performed to provide
anatomic detail about the extent of disease, presence of
tandem lesions, aberrant anatomy, and collateral patterns. It
is not infrequent to find small, flow-related aneurysms in collateral branches, specifically the pancreaticoduodenal

Figure 152-5 Abdominal aortogram with right anterior oblique view demonstrates a large patent inferior mesenteric
artery (IMA). Selective IMA angiography confirms collateralization to the superior mesenteric artery via the arc of
Riolan (arrow) and collateralization to the celiac axis via the gastroduodenal artery (arrow).




Mesenteric Vascular

combined with occlusion of the celiac axis and extensive

collateralization via the SMA.

clinical deterioration, bowel infarction, and risk of sepsis

from catheter-related complications.71,72

Endoscopy is often obtained as part of the investigation of
abdominal pain. It can demonstrate inflammatory and ischemic changes, most noticeable in the stomach, duodenum, or
right colon. Erosive ischemic gastritis, gastroduodenitis, or
ischemic colitis noted on endoscopy has been described in
association with chronic mesenteric ischemia. 61 In a study of
55 patientsAtreated for mesenteric ischemia, endoscopy demonstrated ischemic duodenitis in 38% of patients and ischemic colitis in 57%.62 Clinical experience from European
centers has shown that endoscopy can be useful to assess
mucosal perfusion, to confirm ischemia, and to differentiate
causes of abdominal pain in patients with atypical symptoms
of chronic mesenteric ischemia.32-34,36-39,63-70

Gastric Tonometry
Gastric tonometry has been shown to be a valuable
diagnostic test to assess intestinal perfusion. The initial
concept was described in 1965 and later developed into
clinical protocols32-34,36-39,61
in the late
The study is based on the premise
that PCO2 levels rise above normal from reduced carbon
dioxide washout in the ischemic tissue. Higher PCO2 levels
measured in the gastric, jejunal, or colonic mucosa correlate
with poor mucosal perfusion. Tonometry can be performed as
part of a 24-hour monitoring study during the fasting and
postprandial states or as an exercise test using a small nasogastric tonometry catheter with serial PCO2 measurements in
the stomach, duodenum, or upper jejunum.33,35,38 Most
recently, jejunal tonometry has been used with additional
diagnostic value.37 For exercise tonometry, the patient uses a
bicycle ergometer and the gastric-arterial P CO2 gradient is
measured at incremental workloads.36 Excessive or
inadequate exercise can result in false-positive or falsenegative results. Whereas several studies have shown the
clinical utility of tonometry for diagnosis of ischemia, this
modality has not gained widespread acceptance in the
United States.32-39

Oxygen Light Spectroscopy

Visible light spectroscopy is a new technique that enables
noninvasive measurements of mucosal capillary hemoglobin
oxygen saturation during endoscopy with use of white light
from a fiberoptic probe. The technique relies on differences
in absorption spectra of oxygenated and deoxygenated hemoglobin; oxygen saturation reflects mucosal perfusion. Pilot
studies and recent clinical correlation have shown promising

There is no role for a conservative approach with chronic
parenteral nutrition and noninterventional therapy in patients with symptomatic mesenteric artery disease. Excessive
delays in proceeding with definitive revascularization and

Indications for Revascularization

Revascularization is indicated in all patients with symptoms
of chronic mesenteric ischemia. Treatment goals are to
relieve symptoms, to restore normal weight, and to
prevent bowel infarction. The indication of prophylactic
in patients with asymptomatic disease
remains controversial. Based on the report by Thomas et
al,25 there may be a role for prophylactic revascularization
in patients with severe three-vessel disease, particularly for
those with difficult access to medical care who live in remote
or underserved areas. Our approach in these patients has
been close surveillance and counseling about symptoms of
mesenteric ischemia, with a low threshold to proceed with
revascularization if any gastro- intestinal symptoms (e.g.,
bloating, diarrhea, atypical pain) arise. Revascularization
has been advised in asymptomatic patients with severe threevessel disease undergoing aortic reconstructions for other

Choice of Open versus

Treatment selection has evolved in most centers. The number
of mesenteric revascularizations has increased 10-fold in the
United States in the last decade, largely because of improved
diagnosis and decreased morbidity of endovascular therapy.
In most centers, angioplasty with stenting surpassed open
bypass as the first option and is currently used in more than
70% to 80% of the patients treated for chronic mesenteric
ischemia.9,19,51 These changes in treatment paradigm have
occurred despite the lack of prospective randomized
compari- sons between the two techniques. Endovascular
revascular- ization has been associated with decreased
morbidity, length of stay, and convalescence time but similar
mortality com- pared with open repair in single-institution
studies.19,73 A larger population-based study suggests a
possible mortality benefit with endovascular treatment, but
this may be limited by selection bias.9 Mesenteric bypass
offers improved patency, with lower rates of re-interventions
and better freedom from recurrent symptoms.9,19,41,73-83
In most centers, including the authors, mesenteric angioplasty with stenting is currently the first choice of treatment
in patients with chronic mesenteric ischemia who have suitable lesions, independent of their clinical risk (see Fig. 1524). A careful review of preprocedure CTA with attention to
anatomic factors determines selection of the open or endovascular approach. The SMA is the primary target for revasthe use of parenteral nutrition alone have been associated

cularization, and as such the anatomy of the SMA is the

most important determinant of choice of therapy. The ideal
lesion for angioplasty and stenting is a short, focal stenosis
or occlu- sion with minimal to moderate calcification or


Mesenteric Vascular Disease: Chronic

Ischemia (see Fig. 152-4). For celiac axis lesions,

angioplasty with stenting carries a higher rate of restenosis,84
and it should not be per- formed if there is active
compression by the median arcuate

ligament unless this has been surgically released. We found

no benefit with two-vessel stenting.84 The technical difficulty
of endovascular procedures is increased by the presence
of severe eccentric calcification, flush occlusion, longer
lesions, small vessels, and tandem lesions affecting branches.
Although these anatomic features do not contraindicate the
use of stents, the technical result is often not optimal, with
higher rates of arterial complications (e.g., distal embolization, dissection) and restenosis.85,86 Our preference in lower
risk patients has been to offer open revascularization if the
anatomy is unfavorable for angioplasty and stenting. 87,88 Mesenteric bypass has also been increasingly performed in
patients for whom percutaneous intervention has failed
because of flush occlusion or stent occlusion and in patients
with recur- rent in-stent stenosis for whom multiple reinterventions have failed. Open reconstruction has been
preferred as the first option to treat most patients with
neurofibromatosis, and midaortic syndrome.28,29 Finally, for
patients who are not good candi- dates for open repair
because of severe comorbidities or cachexia, stenting can be
used as a bridge to open surgical bypass in those with
complex lesions prone to restenosis.



Endovascular Revascularization
Endovascular mesenteric revascularization carries definitive
risk. The average 30-day mortality in a recent systematic
review was 6% (0% to 21%), surpassing the mortality
reported for other types of endovascular interventions,
including aortic, renal, and carotid procedures.73 Even though
most interventions are done with local anesthesia, these
patients typically undergo a comprehensive medical
evaluation to identify and to optimize cardiovascular risk
factors and their nutritional status.

Preprocedure Evaluation
Many of the comorbidities may require medical therapy to be
started before or after the intervention, depending on their
severity. Revascularization should not be excessively
delayed. Patients who present with deterioration of
symptoms should be admitted, prescribed intravenous
heparin, and treated urgently within 24 to 48 hours. Patients
with allergy to iodin- ated contrast agents should be
premedicated with a steroid and antihistamine preparation.
Those with chronic kidney disease who have a serum
creatinine level above 1.5 to
2.0 mg/dL (133 to 177 mmol/L) undergo intravenous hydration with sodium bicarbonate and oral acetylcysteine,
starting the day before intervention. Review of preprocedure
imaging (CTA, MRA, or conventional angiography) is key to
selec- tion of the ideal approach based on the angle of origin
of the mesenteric vessels in relation to the aorta, the amount
of calcium and thrombus load, and the presence of

important collaterals or unusual anatomy (e.g., replaced hepatic) in

proximity to the target lesion.

Angiography. Diagnostic angiogra- phy is
most often done immediately before a planned
inter- vention, through either the femoral or
brachial approach. Access is established with
ultrasound guidance and a 0.035- inch guide
wire system. A 5F sheath is positioned in the
external iliac artery and a 5F diagnostic flush
catheter is advanced to T12 level over a 0.035inch guide wire.
heparinization (40 units/kg) is recommended
before selective catheterization of the mesenteric
arteries. The use of a low-osmolarity contrast
agent (e.g., Visipaque) minimizes abdominal
discomfort during selective injections. Choice of
catheter shape depends on access site, angle
of origin, and individual preference. A
multipurpose catheter is ideal for selective
catheterization through a brachial approach,
whereas a secondary curve catheter (e.g., SOS or
Simmons) or a catheter with a more acute curve
(e.g., Cobra 2) can be used for interventions
done through a femoral approach. A complete
study includes abdominal aortography with
anterior- posterior and lateral views to define
the location, severity, and extent of visceral
artery involvement and to identify concomitant
lesions in the aorta and renal or iliac arteries.
The optimal projection to display the proximal
celiac axis and SMA is a lateral view; for the
origin of the IMA, it is a 15-degree right lateral
oblique view. Selective angiography is necessary
to confirm the severity of disease and to

identify tandem lesions and collateral patterns. In patients

with ques- tionable lesions, pressure gradients can be
measured by pres- sure wire, pullback, or simultaneous
pressure measurement techniques.89

The primary goal of percutaneous treatment is to restore
antegrade flow into at least one of the three mesenteric arteries, preferentially the SMA. First reports described
successful results with balloon angioplasty alone, but elastic
recoil and restenosis limited its utility for ostial lesions. 81,90-99
Although there are no prospective comparisons between
angioplasty alone and primary stenting, most agree that
routine stenting is indicated, given that mesenteric lesions
resemble renal artery stenoses.84,100-114 Although there are no
randomized comparisons between SMA and celiac stent
placement, ret- rospective studies suggest that celiac
stenting is associated with more recurrences in the first
year after treatment.84 In patients with compression of the
celiac axis by the median arcuate ligament, there is risk of
stent fracture and compres- sion. The role of two-vessel
stenting remains controversial. Two retrospective studies by
the Massachusetts General Hospital group109 and by Silva et
al106 have shown a nonsig- nificant trend toward less
recurrence with two-vessel stent- ing. Malgor et al84 from the
Mayo Clinic reported nearly identical recurrence rates at 2
years in patients treated by SMA stents (78%) compared
with two-vessel stenting of the SMA and celiac axis (60%).
Two-vessel mesenteric interven- tions may have a role in
select patients with severe gastric ischemia who do not have
a good collateral network between the celiac axis and SMA.
However, there is no proven benefit that routine two-vessel
stenting provides more durable relief,




Mesenteric Vascular

and a second intervention adds cost and potential risk of

Celiac axis intervention may be considered in higher risk
patients for whom attempted recanalization of the SMA has
failed or in those in whom an SMA intervention is thought
to have a low yield for success because of excessive
calcifica- tion or long-segment occlusion. In these patients,
celiac stenting may be considered a bridge to open bypass or
retro- grade SMA stenting.115 Angioplasty of the IMA in our
experience carries a higher risk of rupture, dissection, or
embolization and is not advised, with rare exceptions.
A brachial artery approach is preferred for patients with a
very angulated origin of the SMA off the aorta and in those
with occlusions or longer lesions. The authors preference is
to use the brachial artery approach whenever possible (Fig.
152-6). This offers excellent support with small-profile
systems and precise stent deployment in patients with an
acute SMA angle. Because the risk of puncture-related complications is higher with a total percutaneous technique,
another option is to use a small 1- to 2-cm incision under
local anesthesia to expose and to repair the brachial artery.
Less frequently, a radial approach has been used.

Percutaneous access is established by a 0.018-inch micropuncture set with ultrasound guidance, after which the
system is exchanged for a 0.035-inch guide wire system. Full
systemic heparinization (80 mg/kg) is administered before

catheter manipulations to achieve an activated clotting

time above 250 seconds. A 6F or 7F 90-cm hydrophilic
sheath is positioned in the descending thoracic aorta
above the celiac axis origin. A 5F multipurpose catheter is
ideal for selective catheterization of the mesenteric
arteries through the brachial approach, whereas an SOS
or VS1 catheter can be used from the femoral approach.
The initial selective angiography should demonstrate the
origin of the vessel from the aortic wall and the severity
of the stenosis, and it should document the distal
branches for comparison with postintervention views.
The target lesion is initially crossed with a 0.035-inch
soft angled glide wire, which is exchanged for the interventional wire of choice after confirmation of true lumen access.
The authors preference is to use a small-profile (0.014- or
0.018-inch) stiff guide wire for most interventions. Most
recently, our practice has changed to covered stents on the
basis of a recent report that indicates superior patency rates
compared with bare metal stents.116 The tip of the guide
wire should be visualized and positioned within the main
trunk of the SMA rather than within small jejunal branches,
which are prone to perforation or dissection (Fig. 152-7).
Embolic protection may be useful in select patients with
occlusions, long lesions (>30-mm length), severe calcification, thrombus, and acute or subacute symptoms; the authors
preference is to use a 320-cm working length 0.014-inch
filter wire (SpiderRX, Covidien, Plymouth, Minn). Alternatively, Brown et al101 described the use of temporary balloon

Figure 152-6 Angioplasty and stenting of a focal stenosis of the superior mesenteric artery (SMA) by a brachial
approach. After selective angi- ography (A), the lesion is crossed and a 0.014-inch SpiderRX filter wire is
deployed in the main trunk of the SMA (B), avoiding jejunal branches. The entire lesion is treated by a balloonexpandable stent (C), which is extended 1 to 2 mm into the aorta and flared proximally (D). Completion
angiography demonstrates patency of the stent without embolization or dissection (E).

Figure 152-7 An important technical point is

to visu- alize the tip of the guide wire
during the intervention and to position the
guide wire in the main trunk of the superior
mesenteric artery (curved black arrow) as
opposed to distal jejunal branches (curved
black arrow), which are prone to perforation
resulting in mesenteric hematoma (straight
white arrow).

occlusion and aspiration with the GuardWire (Medtronic,

Minneapolis, Minn). If a 0.035-inch stent is selected, a twowire technique can be used by combining a 0.014-inch filter
wire with a 0.018-inch buddy wire; the stent is introduced
via both wires for better support and to facilitate subsequent
retrieval of the embolic protection device (Fig. 152-8). Predilatation is recommended for tight stenoses, occlusions, and
severe calcification and to size stents. A balloon-expandable

Figure 152-8 Technique of recanalization

and primary stenting of a total superior
mesenteric artery (SMA) occlusion. In these
cases, a stiff support system is built with
combination of a 7F 90-cm hydrophilic
sheath, 7F 100-cm multipurpose guide
catheter, and 5F 125-cm multipurpose
catheter. The stump of the occluded SMA is
combination (A); the lesion is crossed with
a straight glide wire. After true lumen
access is confirmed, a 0.014-inch filter wire
and a 0.018-inch buddy wire are deployed

stent with diameters ranging from 5 to 8 mm is used in

more than 95% of cases, allowing precise deployment and
greater radial force. The stent is positioned under protection
of the sheath, covering slightly more than the entire length
of the lesion. Positioning the stent so that it extends 1 to
2 mm into the aortic lumen is critical to avoid missing the
proximal portion of the lesion (see Fig. 152-6). Ideally, the
stent should be flared gently into the aorta, which minimizes


into the SMA via 0.035-inch catheter (B);

the lesion is predilated (C) and stented with
a balloon-expandable stent (D).

the potential to miss disease at the ostia and facilitates

repeated catheterization if needed. On occasion, a selfexpandable stent is needed to treat a nonostial lesion or
segments with excessive tortuosity extending beyond the
angulated portion of the SMA.
Recanalization of Mesenteric Occlusions.
The technique is slightly modified in patients with difficult
occlusions. In these cases, it is of paramount importance to
use the brachial approach and a stiff support system, which
is accomplished by combining a 7F sheath, 7F
multipurpose guide catheter, and 5F multipurpose catheter
(see Fig. 152-8). In the authors opinion, attempting a
difficult recanalization from the femoral approach adds time,
contrast material, and catheter manipu- lations and is fraught
with exceedingly high failure rates. Ideally, the tip of the
multipurpose catheter is used to engage the stump of the
occluded SMA (Fig. 152-9), and sufficient support is
provided by the combination of the sheath and guide
catheter. The lesion is crossed with a straight-tip,
hydrophilic, soft 0.035-inch glide wire but also with 0.018inch and 0.014-inch guide wires if needed. It is ideal to avoid
the subintimal plane, which is best achieved by use of
straight- tip guide wires. A Quick-Cross (Spectranetics,
Colorado Springs, Colo) or an alternative support catheter
or even a small coronary balloon may be needed to cross a
tight lesion. Once the lesion is crossed, access into the true
lumen should be confirmed. Our preference has been to
use an embolic protection device (e.g., SpiderRX) with a
two-wire technique routinely in cases of total occlusion.

After deployment and flaring of the stent, the embolic

protection device is retrieved with careful attention to avoid
entrapment into the stent. The basket is examined for debris.
Formal completion angiography should be performed,
includ- ing a focal magnified view of the stent with the
sheath into the aorta to demonstrate the vessel origin and
a panoramic view of the entire SMA and its branches to rule
out emboliza- tion or perforation. The stiff guide wire should
be retracted, and nitroglycerin may be administered through
the sheath to minimize spasm or kinks caused by the guide
wire tip. It is particularly important to note the presence of
distal emboliza- tion, dissection, thrombus, or branch
perforation. These com- plications occur in 5% to 10% of
patients and remain a major source of morbidity and
mortality if they are not immediately recognized.85
Adjunctive Techniques. A number of adjunctive
techniques can be used to optimize results of mesenteric
stents in patients with complex lesions, but the author
acknowledges that these techniques are anecdotal or
supported by a limited number of case reports. An acute or
subacute symptom presentation suggests fresh thrombus or
complicated plaque. In these cases, local administration of
tissue plasminogen activator into the diseased segment 20
to 30 minutes before stent placement may improve
technical success. For eccentric, calcified lesions,
percutaneous atherectomy has been carefully used in very
select cases.117 It is critical to have an appreciation of the
limitations of this technique when it is applied as an offlabel use in the mesenteric arteries.

Figure 152-9 Recanalization of superior mesenteric artery (SMA) occlusion by the technique described in Figure
152-8. After the stump is engaged by the catheter, guide catheter, and sheath (A, arrow), the lesion is crossed
(B) and stented with use of embolic protection (C). Note that the balloon is used to flare the proximal part of the
stent (C, arrow). Completion angiography shows a flared, widely patent SMA stent (D).


stent retrograde by direct puncture. This option may be used

if there is no good source of inflow or if there is peritoneal
The most common causes of death after mesenteric stenting
are cardiac events, gastrointestinal bleeding, and bowel ischContemporary reports from large-volume centers have
emia. Bowel ischemia is typically associated with intraproceshown
that mesenteric bypass can be performed with mordural complications, such as distal embolization, thrombosis,
tality rates of less than 3%. Improvements in the outcomes
or dissection.
C of mesenteric reconstructions
D can be attributed to several
treated by SMA stents without embolic protection, with
factors, including technical refinements, better patient selechigher rates among patients with subacute symptoms, occlu87
tion, and advances in medical, anesthetic, and critical care
sion, long lesions (>30 mm), and severe calcification.
management. In the first two Mayo Clinic reports from
Therefore, there may be a role for selective use of embolic
1981 and 1992, more than 50% of the patients had threeprotection in these patients. The most commonly reported
vessel revascularization, and concomitant aortic reconstruccomplications are access-related problems in 2% to 15%,
tion was performed in 20% to 30% of patients. The operative
renal insufficiency in 5% to 12%, acute bowel ischemia in
mortality was 10% in both reports. Since then, our practice
1% to 5%, gastrointestinal bleeding in 1% to 4%, cardiac
has evolved from complete retrograde revascularization and
events in 1% to 3%, and respiratory complications in 3%.
a low threshold for concomitant aortic reconstruction to
Postprocedure Management
a preference for antegrade mesenteric reconstructions based
on the supraceliac aorta whenever possible. Aortic reconThe postprocedure care after mesenteric interventions is
struction is reserved for the rare patient who needs it for
comparable to that of other peripheral endovascular procean inflow source or in whom aortic disease necessitates
dures. All patients are admitted for observation overnight.
Worsening abdominal pain after the procedure is unusual and
Reconstruction of the celiac axis and the SMA with a
warrants evaluation to rule out thrombosis, embolization, or
polyester graft originating from the supraceliac
a mesenteric hematoma from jejunal branch perforation (see
more than 80% of open mesenteric reconFig. 152-7). Patients are allowed to resume a regular diet
is preferentially selected in lower
within 6 to 8 hours. Antiplatelet therapy is typically started
ideal candidates for endovascular
before the intervention with acetylsalicylic acid and contintreatment
disease, with no evidence of
ued indefinitely thereafter. Clopidogrel is started the day of
or debris. Other
the intervention with a loading dose of 300 mg and continsources
or the iliac
ued for 6 to 8 weeks as a dual antiplatelet agent, after which
patients continue with acetylsalicylic acid alone. The
Cauthors preference is to obtain a duplex ultrasound scan
before dis- charge or within the first few days after the
laparotomy to expose the SMA combined with retrograde
procedure to serve as a baseline for future comparison. The
placement of an SMA stent has been proposed by the Unipresence of an ele- vated velocity on duplex ultrasound may
versity of Pennsylvania and Dartmouth groups. This option
be due to inadequate stenting with incomplete treatment of
avoids the need for extensive dissection, vein harvesting, and
the lesion proximal or distal to the stent. Follow-up
use of a prosthetic graft; it may be selected for patients with
includes clinical examination and duplex ultrasound every
extensive aortoiliac disease and no adequate source of inflow
6 months during the first year and annually thereafter.
or for those with acute mesenteric ischemia, bowel gangrene,
and contamination. Hybrid, retrograde stenting provides one
Open Revascularization
of the most expeditious methods of revascularization in
patients with difficult occlusions. Transaortic endarterectomy
is rarely indicated, but it may be considered in patients for
A variety of open surgical techniques have been described
whom endovascular therapy has failed or in patients who are
to reconstruct the mesenteric arteries. Selection involves
not candidates for endovascular therapy and have bacterial
the type of incision (transperitoneal vs retroperitoneal),
contamination or perforated bowel, previous abdominal irraconduit (vein vs prosthetic), graft configuration
diation, extensive abdominal wall hernias, or other hostile
(antegrade vs retro- grade), source of inflow (aortic vs
iliac), and number of vessels to be reconstructed (single vs
multiple). The type of open reconstruction is selected on
Preoperative Evaluation
the basis of the anatomy and the patients clinical risk
Preoperative evaluation should focus on a critical review of
assessment. Elderly patients and those with cachexia or
surgical risk, nutritional status, and anatomic factors that
severe cardiac, pulmonary, and renal dysfunction are
affect the choice of reconstruction. A comprehensive evalufrequently not good candidates for an aortic- based
ation of cardiac, pulmonary, and renal performance is crucial
procedure. We have found that the iliac artery is a good
to optimize patient selection. These operations are often
source of inflow in high-risk patients or those who have
indicated in patients with multiple comorbidities who are
dis- eased or calcified aortas. Alternatively, a stent can be
not candidates for endovascular therapy or for whom
placed with surgical exposure of the SMA and
introduction of the


angioplasty and stenting have failed. The evaluation should

include noninvasive cardiac stress testing (dobutamine stress
echocardiography or sestamibi study), pulmonary function
tests, carotid ultrasound, and noninvasive lower extremity
arterial studies. These patients have a high incidence of
coronary artery disease as demonstrated by the landmark
report of Hertzer et al.118 Routine cardiac catheterization is
unnecessary, and cardiac evaluation and perioperative management are guided by the recommendations of the
American College of Cardiology for patients undergoing
major non- cardiac surgery.119 The Society for Vascular
Surgery clinical comorbidity score system can be used to
stratify operative risk, but the criteria have not been
validated prospectively in patients undergoing mesenteric
reconstruction. The risk of major noncardiac surgery is
increased in patients with unstable angina, symptomatic or
poorly controlled ectopy, recurrent congestive heart failure,
ejection fraction below 25%, recent myocardial infarction
(<6 months), vital capac- ity of less than 1.8 L, forced
expiratory volume in 1 second below 800 mL, diffusing
capacity of lung for carbon mon- oxide below 30%,
resting PO2 of less than 60 mm Hg, PCO2 of less than 50 mm
Hg, serum creatinine concentration above 2.5 mg/dL, or
ongoing renal replacement therapy. Optimal medical therapy
in these patients ideally should include cessation of cigarette
smoking and initiation of ace- tylsalicylic acid, beta
blockade, and a
cholesterol-lowering medication,
preferentially a statin.
Equally important is an evaluation of the nutritional
status. In patients with severe cachexia who have
stable symptoms and cannot tolerate an enteral diet, a
period of total parenteral nutrition may be beneficial.
However, it is critical to avoid excessive delays,
particularly in patients with symptom deterioration. Those
with chronic kidney disease and serum creatinine level
above 1.5 to 2.0 mg/ dL (133 to 177 mmol/L) are
admitted to the hospital the day before for intravenous
hydration. Gentle bowel prepara- tion may be used in the
patient with stable symptoms but should be avoided in
those with severe or subacute ischemia.

Antegrade Supraceliac Aortato Celiacand
SMABypass. The distal thoracic or supraceliac aorta is
often spared from severe atherosclerotic disease. Bypass with
an antegrade graft con- figuration based in the supraceliac or
lower thoracic aorta120 offers a potential hemodynamic
advantage while avoiding kinks that can occur with grafts
placed in a retrograde fashion. Reconstruction of the celiac
axis and the SMA with a bifur- cated polyester graft
originating from the supraceliac aorta compromises more
than 80% of open mesenteric reconstruc- tions.19,121 This
approach is preferentially selected in lower risk patients who
have multivessel disease and no evidence of significant
supraceliac aortic calcification or atheromatous debris.
The operation is performed through a transperitoneal
upper midline or bilateral subcostal incision, depending on
the patients body habitus and costal cartilage flare. Abdominal exploration includes an evaluation for other intraabdominal disease and careful inspection of the small bowel
for unsuspected ischemic perforations. A third-arm retractor
assists with exposure of the upper abdomen. Alternatively,
an Omni self-retaining retractor can be used. The lesser
omentum is opened, and the left lobe of the liver is retracted
after division of the left triangular ligament. The esophagus
is retracted toward the patients left side with a nasogastric
tube in place, and the stomach is gently retracted caudally.
The diaphragmatic crura are divided longitudinally, exposing
the supraceliac aorta (Fig. 152-10). Approximately 5 to
10 cm of supraceliac aorta is dissected free in preparation
for clamping. While doing so, care should be taken not to
enter the pleural cavity on either side of the aorta. Usually
only the celiac trunk and the proximal hepatic and splenic
arteries need isolation; but with more extensive disease,
the common hepatic artery may be a better target for the
anastomosis. The left gastric artery is often small and may
be divided without sequelae, which facilitates celiac anastomosis and makes tunneling of the SMA graft behind the
pancreas easier.

Figure 152-10 The supraceliac aorta is

exposed after division of the diaphragmatic
crura (A). After suprace- liac aortic crossclamping (B), an oblique aortotomy is
made for the proximal anastomosis of the
bifurcated graft. Note that the graft is gently
beveled and that the left limb is tunneled
anastomosis to the superior mesenteric
artery (C). The right graft limb, which is

anastomosed to the celiac axis or hepatic

The transverse mesocolon is retracted cephalad and the

root of the mesentery is incised longitudinally over the SMA
just below the pancreas. Several lymphatic and small venous
tributaries require meticulous ligation. The SMA and several
jejunal branches are dissected free and carefully controlled
with Silastic vessel loops. Excessive traction on these vessel
loops can easily result in avulsion of small branches and
should be avoided. In most cases, 4 to 5 cm of the SMA is
dissected distal to the lower border of the pancreas, but additional dissection may be required in patients with extensive
disease. In these cases, one or more crossing venous tributaries from the superior mesenteric vein may need to be ligated
and divided to allow exposure of the SMA.
Patients are given systemic heparin (60 to 80 mg/kg),
and diuresis is induced with mannitol before placement of
the supraceliac clamp. Partial aortic cross-clamping can be
performed with a Satinsky or a multipurpose clamp; alternatively, total aortic occlusion with two cross-clamps affords
better exposure for the proximal aortic graft anastomosis.
A straight or angled aortic clamp (Cherry supraceliac clamp)
and a Wylie hypogastric clamp work well. Placed appropriately, occlusion of the lumbar vessels is achieved as well.
A slightly oblique or vertical aortotomy is made. A 12
7-mm knitted polyester graft is beveled in an oblique
fashion with a short main body and is anastomosed to
the supra- celiac aorta in an end-to-side fashion with running
4-0 Prolene suture. Aortic cross-clamp time rarely exceeds
20 minutes and most often ranges from 12 to 15 minutes.
The risk of renal ischemia or embolization is low when
patients are properly selected and have a relatively disease
free supra- celiac aorta.
The left limb of the bifurcated graft is positioned slightly
posterior and is tunneled in a retropancreatic position and
anterior to the left renal vein toward the SMA (see Fig. 15210). The anastomosis to the SMA is performed in an end-toside fashion when the anastomosis is performed at the base
of the mesentery below the pancreas. If the SMA is extensively diseased or the patient has had prior stents, the lumen
may require modest endarterectomy or removal of stent
struts. Any endarterectomy should be done carefully to avoid
tearing of an excessively thin arterial wall remnant that
cannot be reconstructed. In a few cases with extensive
plaque, the distal graft can be beveled into a long patch or
anasto- mosed end to side into a bovine pericardium patch. It
is important to relax retraction when cutting the graft limbs
to length to avoid angulation or kinking, whether antegrade
or retrograde bypass is done. Next, the celiac axis
anastomosis is performed in an end-to-end fashion, or more
frequently the anastomosis may be performed end to side
to the common hepatic artery (Fig. 152-11).
Retrograde Iliac Artery or Infrarenal Aorta
to SMA Bypass. Supraceliac origin grafts are not
ideal in patients with compromised cardiac or pulmonary
function or those with extensive atherosclerosis or
circumferential calcification of the supraceliac aorta. In these
cases, other sources of inflow, such as the infrarenal aorta, a
prior infrarenal aortic graft, or

the iliac arteries, are preferred (Fig. 152-12). In general, we

reconstruct only one artery (SMA) if a retrograde graft is
used. The infrarenal aorta may be replaced, if it is diseased,
with retrograde reconstruction of the SMA based on the
aortic graft. Nonetheless, concomitant aortic reconstruction
increases operative mortality and should be avoided unless it
is absolutely necessary.19 The proximal anastomosis is performed to the anterolateral wall of the aorta and can be done
with either two cross-clamps or a partial-occlusion clamp,
depending on the aortic size and the presence of atherosclerosis or calcification within that segment. A 6-mm coronary
punch can be used to remove a portion of the aortic wall.
The Oregon group has reported on the use of the distal infrarenal aorta or the infrarenal aortaright common iliac artery
junction as the preferred site for the proximal anastomosis.122
Our preference in the higher risk patient is to select the
common iliac artery as a source of inflow whenever possible,
avoiding aortic cross-clamping.
The key to avoidance of graft elongation, angulation, or
kinking is to cut it to length with the SMA in a nearly anatomic position. This is done by relaxing the retractors before
cutting the graft and after the proximal or distal anastomosis
is done. A large 8- or 10-mm graft should be used for a retrograde aortomesenteric graft, and it is imperative to perform
the distal anastomosis first and then push the mesentery close
to the aorta and select the appropriate aortic site for the
anastomosis. The Oregon group has favored a C-shaped graft
configuration.122 We have also preferred a C-shaped graft
when the iliac artery is the source of inflow. In these cases,
the proximal anastomosis may be done first to the iliac artery
or distal aorta, followed by the distal anastomosis to the
SMA. In our most recent analysis, retrograde grafts worked
as well as antegrade grafts, although the patients with retrograde grafts were older and had reduced lifespans. 19,121 There
are some patients who have extensive circumferential aortic
calcification but soft common or external iliac arteries that
can serve as good donor vessels. Either the right or left
common iliac artery can be chosen for inflow, depending on
the orientation of that artery to the normal anatomic position
of the SMA. In general, the right iliac artery lays better if
both vessels are suitable. Two-vessel reconstructions can also
be accomplished with retrograde grafts by performing a sideto-side anastomosis to the SMA and an end-to-side anastomosis to the common hepatic artery. These grafts may be
passed on top of or beneath the pancreas and curved in a
C-shaped fashion toward the hepatic artery.
Retrograde Hybrid Revascularization. A
hybrid approach of a midline laparotomy to expose the SMA
and retrograde placement of an SMA stent has been
by Milner et al123 from the University of
Pennsylvania and Wyers et al124 from Dartmouth. This
option avoids the need for extensive dissection, vein
harvesting, and use of a prosthetic graft. It may be selected
in patients with extensive aortoiliac disease and no adequate
source of inflow or in those with acute mes- enteric ischemia
with bowel gangrene and contamination (Fig. 152-13).


Figure 152-11 Bifurcated supraceliac aorta to common hepatic artery (A) and superior mesenteric artery bypass (B).
Computed tomography angiogram (C) demonstrates a widely patent bypass

The SMA is dissected below the pancreas as

previously described. Several jejunal branches are
controlled with Silas- tic vessel loops and occluded before
manipulation to avoid distal embolization. Retrograde
SMA access is established by use of a micropuncture set
with a 0.018-inch guide wire. This is exchanged for a
0.035-inch guide wire system, and a 6F to 7F sheath is
advanced into the SMA. Retrograde angiography is
obtained, and the SMA occlusion or stenosis is crossed,
predilated, and stented with a balloon-expandable
stent. Before antegrade flow is restored to the SMA, the
sheath is flushed to prevent distal embolization. The
puncture site may be closed with interrupted sutures or
opened longitudinally and closed with a patch if it is
severely diseased.
Transaortic endar- terectomy is rarely indicated but may be
B in patients for whom endovascular therapy
failed and in patients who have bacterial contamination,
perforated bowel, previous abdominal irradiation,
extensive abdominal wall hernias, or other hostile
conditions. The most recent experience by the
University of Wisconsin group reported an operative

mortality of 3.8% among 80 patients treated by transaortic

Our preference has been to approach the
paravisceral aorta through a full-length midline
abdominal or subcostal incision or a thoracoabdominal
incision for patients who have narrow costal flares or are
truly obese. Exposure by an abdominal incision alone in
the latter patients is suboptimal because access to the
origins of the visceral arteries is restricted, orientation
from which to perform the endarterec- tomy is poor, and
adequate retraction of the costal margins is difficult.
The aorta is exposed by medial visceral rotation with
the left kidney left in its bed, and dissection is carried
anterior to the renal vein. The diaphragmatic crus is
transected longitu- dinally, allowing exposure of the left
anterior lateral wall of the aorta and origins of the SMA
and celiac axis (Fig. 152- 14). The SMA is dissected free
for several centimeters. After systemic heparinization and
induced diuresis, the supraceliac aorta and infrarenal aorta
are clamped. A longitudinal or trap-door aortotomy is
performed, starting at the level of the renal arteries and
extending up to just above the celiac axis