DOI: 10.1016/j.jtcvs.2006.09.059 2007;133:428-434 J Thorac Cardiovasc Surg

Stiles, Benjamin B. Peeler, John A. Kern and Irving L. Kron T. Brett Reece, Curtis G. Tribble, Robert L. Smith, R. Ramesh Singh, Brendon M.

Central cannulation is safe in acute aortic dissection repair

http://jtcs.ctsnetjournals.org/cgi/content/full/133/2/428located on the World Wide Web at:

The online version of this article, along with updated information and services, is

2007 American Association for Thoracic Surgery Association for Thoracic Surgery and the Western Thoracic Surgical Association. Copyright ©

is the official publication of the AmericanThe Journal of Thoracic and Cardiovascular Surgery

on June 11, 2013 jtcs.ctsnetjournals.orgDownloaded from

CdTB

Surgery for Acquired Cardiovascular Disease Reece et al

4

ACD

entral cannulation is safe in acute aorticissection repair

. Brett Reece, MD, Curtis G. Tribble, MD, Robert L. Smith, MD, R. Ramesh Singh, MD, Brendon M. Stiles,

enjamin B. Peeler, MD, John A. Kern, MD, and Irving L. Kron, MD

Orsch

M1dafm

RbhopP4

Cpctm

Tci

fczbaiaoc

From the University of Virginia, Departmentof Surgery, Division of Thoracic and Cardio-vascular Surgery, Charlottesville, Va.

Read at the Thirty-second Annual Meetingof the Western Thoracic Surgical Associa-tion, Sun Valley, Idaho, June 21-24, 2006.

Received for publication June 17, 2006;revisions received Sept 9, 2006; acceptedfor publication Sept 20, 2006.

Address for reprints: T. Brett Reece, MD,University of Virginia Health System, De-partment of Surgery, PO. Box 801359, MR4Building, Room 3116, Charlottesville, VA22908 (E-mail: tbr5q@virginia.edu).

J Thorac Cardiovasc Surg 2007;133:428-34

0022-5223/$32.00

Copyright © 2007 by The American Asso-ciation for Thoracic Surgery

ldoi:10.1016/j.jtcvs.2006.09.059

28 The Journal of Thoracic and CardioDownl

bjective: The site of cannulation for the repair of ascending aortic dissectionemains controversial. It is not clear whether cannulation of the dissected vessel isafe or even preferred. We hypothesized that cannulation of the dissected aortaould be done safely with acceptable complication and mortality rates in thisigh-risk population.

ethods: The charts of repairs of acute ascending aortic dissections (n � 70) from996 to 2005 were reviewed. Cannulation was accomplished in 24 patients via theissected aorta (central) and in 46 patients through cannulation of the femoral orxillary artery (peripheral). All were converted to sidearm cannulation of the graftor reperfusion. Groups were compared on the basis of comorbidities in addition toortality, complications, hospital stays and final disposition.

esults: The groups were comparable on the basis of age and preoperative comor-idities. Similarly, there were no differences in bypass time, crossclamp time, orypothermic circulatory arrest time between groups. Hospital mortality and post-perative complications, including stroke, were similar between groups, but theeripheral group experienced more cardiac events (peripheral 15% vs central 0%;� .05) and higher mortality than the central group (peripheral 19.5% vs central

.2%; P � .05).

onclusions: Direct cannulation of the dissected aorta was safe compared witheripheral cannulation in these patients. Inasmuch as these data demonstrate thatannulation of the dissected ascending aorta is safe, this technique can be used toailor the cannulation approach to specific anatomic and patient characteristics thatight optimize postoperative outcomes in this disease entity.

he optimal management of acute dissection of the ascending aorta remainssurgical.1,2 Historically, morbidity from this entity has been reported as 30%or higher, but more recently several centers have published mortalities that

an approach 10%.3-6 Although outcomes of these surgically treated patients havemproved, the optimal approach to cannulation for these cases is not known.

Currently, three cannulation options exist. First, cannulation using the commonemoral artery is one standard option. Despite widespread use of this route, femoralannulation can carry some risk of critical organ malperfusion, retrograde emboli-ation, and femoral arterial injury. Second, axillary cannulation has more recentlyecome a widely used approach for arterial cannulation, especially in ascendingortic and arch surgery. There are also drawbacks to the use of this approach,ncluding the extra time that is required for sewing a graft or repairing the axillaryrtery. Third, direct cannulation of the dissected ascending aorta has been usedccasionally, but has been mainly reported as a bail-out technique when otherannulation options are not available.7-9 Although this technique allows for cannu-

ation of a part of the vessel that will be excised during the repair, some argue,

vascular Surgery ● February 2007 on June 11, 2013 jtcs.ctsnetjournals.orgoaded from

wos

cslFss

MAtcfpaaicaTo

mnencpawdm

cappmpicmpwvvun

dasIt

apcbwpw

Ftwec

Reece et al Surgery for Acquired Cardiovascular Disease

ACD

ithout data, that this technique may risk rupture, extensionf the dissection, or embolization of debris into cerebral orolid organ vasculature beds.

For the purpose of this study, we hypothesized that directannulation of the dissected ascending aorta was at least asafe as peripheral cannulation through the femoral or axil-ary arteries for cases of acute ascending aortic dissection.urthermore, on the basis of our data, we hope to propose aystem for choosing the optimal cannulation site given thepecific dissection characteristics and patient attributes.

aterials and Methodsfter gaining approval for this study from the Human Investiga-

ion Committee at the University of Virginia (with waiver ofonsent), all acute ascending aortic dissections were identifiedrom our retrospectively collected aortic database including caseserformed between July 1996 and July 2005. From more than 800ortic cases, 70 acute ascending aortic dissections that were oper-ted on were identified. These patients were then further dividednto two groups and compared on the basis of the initial site ofannulation, including central (cannulation of the ascending aorta)nd peripheral (cannulation of the femoral or right axillary artery).hese groups were compared on the basis of preoperative, intra-perative, and postoperative characteristics.

Although some variability occurred during the study period,ost patients were cared for in a similar fashion. Various diag-

ostic radiologic techniques were used including angiography,chocardiography, computed tomographic angiography, and mag-etic resonance angiography. More recently, all patients had aomputed tomographic angiogram for diagnosis and operativelanning. Criterion for site of cannulation varied over time andmong surgeons, but an operative plan for the site of cannulationas established before going to the operating room. The finaletermination of the suitability of the chosen cannulation site wasade in the operating room after direct inspection of the vessels.

Although the sites of cannulation varied, the approaches toooling and circulatory arrest were similar. The goal of ascendingortic dissection repair was prevention of proximal rupture andreservation of aortic valve competence. Thus, all procedures werelanned for replacement of the ascending aorta with repair/replace-ent of the aortic valve as needed. The general approach to these

atients included arterial cannulation of the chosen vessel andnitiation of cardiopulmonary bypass. Most patients were slowlyooled to a core body temperature of 18°C to allow 20 to 30inutes of circulatory arrest time. Only recently has antegrade

erfusion started being used (in some axillary cannulations), bute generally have employed retrograde cerebral perfusion as pre-iously described.10 Owing to fluctuations in practice over time, aariety of neuroprotective pharmacologic strategies have beensed during this period, whereas electroencephalography, transcra-ial Doppler, and cerebral oximetry have been used inconsistently.

Patients undergoing femoral artery cannulation (n � 31) un-erwent femoral cutdown. A purse-string suture was placed on thenterior surface of the femoral artery. Through the purse-stringuture, a cannula was placed by the Seldinger technique (Figure 1).n cases in which the vessel was calcified or appeared dissected,

he artery was secured with vessel loops and vascular clamps, and t

The Journal of Thoracicjtcs.ctsnetjouDownloaded from

transverse arteriotomy was made. The arteriotomies were closedrimarily at the end of the procedure. Patients undergoing axillaryannulation (n � 15) underwent axillary cutdown. The venousranches over the axillary artery were divided and the axillary veinas retracted out of the way. The axillary artery was looped andulled up for the application of a Satinsky clamp. The arteriotomyas started with a knife and completed with an aortic punch used

igure 1. Cannulation of the ascending aorta. A, Gaining accesso the lumen via the Seldinger technique with confirmation of

ire placement in the descending aorta by transesophagealchocardiography. B, Placement of the cannula high in the as-ending aorta with manual stabilization during cooling.

o make a circular hole to which an 8-mm Dacron graft was sewn.

and Cardiovascular Surgery ● Volume 133, Number 2 429 on June 11, 2013 rnals.org

Ob

itpioadrepawdaialdtactieafAcwt

ic

wdcktfttyu

RTccdffa

TC

ASDLCCPHAMRCHE

Tad rdial

TP

CNAVRA

Cgw

Surgery for Acquired Cardiovascular Disease Reece et al

4

ACD

nce peripheral cannulation had been achieved, cardiopulmonaryypass was initiated.

Cases involving central cannulation (n � 24) began with directnspection of the ascending aorta after sternotomy and creation ofhe pericardial well. The only specific contraindication to thisrocedure was extensive hematoma in the wall of the aorta, rang-ng from thrombosed false lumen to intramural hematoma. The sitef cannulation was determined from preoperative imaging as wells from intraoperative transesophageal echocardiography. Epicar-ial ultrasound was used if any question of clot in the vessel wallemained. Although cannulating the false lumen was not consid-red to be contraindicated, cannulation of the true lumen wasreferred. The technique involves placement of a wire into thescending aorta. Placement can be confirmed by identifying theire in the descending thoracic aorta by transesophageal echocar-iography. The site of cannulation was sufficiently high on thescending aorta that a crossclamp could be placed more proximallyn cases in which the vent was insufficient to overcome torrentialortic valve insufficiency, but low enough that the site of cannu-ation would be excised. Next, a percutaneous cannula was placedirectly in the ascending aorta over the wire by the Seldingerechnique with the cannulas loaded on a dilator (Figure 1). Inlmost all cases, the cannula was held in place by hand duringooling because of concerns about the ability of the dissected aortao hold a purse-string suture. Once cardiopulmonary bypass wasnitiated, flow in both lumina were confirmed by transesophagealchocardiography. In all cases, the offending area of the ascendingorta was excised, and the distal anastomosis was created with aelt strip and a gel- or albumin-impregnated graft with a sidearm.fter completion of this anastomosis, the sidearm of the graft was

annulated, cardiopulmonary bypass was restarted, and rewarmingas begun. The aortic valve was resuspended when indicated and

he proximal anastomosis was completed.For this study, the type of cannulation, operative times (includ-

ng hypothermic circulatory arrest time, crossclamp time, and

ABLE 1. Preoperative comorbiditiesomorbidity Central cannulation

ge 61.8 � 3.2ex (female) 8/24M 3/24ong-term steroid therapy 4/24HF class 2.9 � 0.2VA TIA 4/24VD 8/24istory of CAD 7/24rrhythmia 2/24arfan’s 1/24

heumatic 1/24RI 2/24D 1/24mergency 15/24

he groups were similar in terms of all preoperative variables evaluatedccident; TIA, transient ischemic attack; PVD, peripheral vascular disease;isease; emergency, hypotensive, significant valvular insufficiency, perica

ardiopulmonary bypass time), and any additional procedures a

30 The Journal of Thoracic and Cardiovascular Surgery ● Febrjtcs.ctsnetjournDownloaded from

ere recorded. Surgeon preferences dictated the adjunct proce-ures, including coronary bypass, which was performed mostommonly in cases of coronary dissection but also in cases ofnown significant coronary stenosis. Perioperative complica-ions, hospital and intensive care unit lengths of stay, andollow-up were recorded and compared between groups. All sta-istics were performed by an independent statistician. Variousechniques were used for comparing the groups, such as �2 anal-sis, the Fisher exact test, and the Student t test. The specific testsed is noted for each comparision or group of comparisions.

esultshe two groups were similar with regard to preoperativeomorbidities, as shown in Table 1. On the basis of theseomparisons, the groups are valid for comparison. Severalifferences were identified in terms of procedures per-ormed, as shown in Table 2. These differences may ariserom increasing use of central cannulation at the same times our increased comfort with valve preservation proce-

ipheral cannulation P value Statistics

65.2 � 2.1 .39 t test12/45 .56 �2 test1/46 .11 Fisher exact test1/46 .04 Fisher exact test

2.9 � 0.2 .84 t test10/46 .77 �2 test16/46 .90 �2 test22/46 .13 �2 test7/46 .41 �2 test4/45 .65 Fisher exact test0/46 .34 Fisher exact test2/44 .60 Fisher exact test0/46 .34 Fisher exact test

34/46 .32 �2 test

, Diabetes mellitus; CHF, congestive heart failure; CVA, cerebrovascular, coronary artery disease; CRI, chronic respiratory insufficiency; HD, hearteffusion, coronary insufficiency, and aortic rupture.

ABLE 2. Adjunct proceduresrocedure Central Peripheral P value

ABG 6/24 9/46 .6o. grafts/CABG 10/6 17/9 .8VR 4/24 10/46 .6alve resuspension 11/24 9/46 .03oot replacement 11/24 15/46 .3ny arch 2/25 12/46 .045*

ABG, Coronary artery bypass graft; AVR, aortic valve replacement. *Theroups had similar procedures performed except that the peripheral groupas significantly more likely to undergo replacement of some part of the

Per

. DMCAD

ortic arch and less likely to undergo aortic valve resuspension.

uary 2007 on June 11, 2013 als.org

da

whccnfitnnss

DTacaoslveccedse

sftccpIwaws

riiftIbt4aatsepitalcscfeirttcc

T

HCC

Ibc

T

C

IPRSAMAOAH3R

Tigch

Reece et al Surgery for Acquired Cardiovascular Disease

ACD

ures. Intraoperative times were similar between the groups,s demonstrated in Table 3.

In terms of complications and disposition, the groupsere similar except that the peripheral cannulation groupsad a siginificantly higher incidence of postoperative myo-ardial infarction as defined by significant troponin in-rease. These data are shown in Table 4. Similar rates ofeurologic complications and hospital mortality wereound, but significantly higher 30-day mortality was shownn the peripheral cannulation group compared with the cen-ral cannulation group. Of note, when comparing the smallumbers of axillary and femoral cannulations, axillary can-ulations had a higher 30-day mortality. These data showimilar need for placement in a rehabilitation facility orkilled nursing facility between the two groups.

iscussionhe optimal cannulation site for the repair of ascendingortic dissection is not known. The most popular sites forannulation in this setting are peripheral arteries such as thexillary artery or the femoral artery.11,12 A broad spectrumf approaches to these vessels has been published withatisfactory results. These variations include direct cannu-ation of the right axillary artery, sewing a graft to thisessel, and even perfusing via the carotid arteries.13-18 How-ver, the cannulation of the dissected ascending aorta, orentral cannulation, has not been used widely owing tooncerns over the fragility of the vessel and over distalmbolization. Interestingly, most cardiac surgical textbookso refer to central cannulation as the “fallback” cannulationite in the setting of failure of primary cannulation attemptslsewhere.7-9

The purpose of this study was twofold: first, to demon-trate that cannulation of the dissected ascending is not onlyeasible, but safe; second, to show that this technique addso the possible routes of cannulation for patients with as-ending aortic dissections. The study was not intended tolaim central cannulation to be the optimal approach in allatients presenting with dissection of the ascending aorta.nstead, by demonstrating that central cannulation was safe,e show that this approach could be added to the options

vailable, providing data to help develop an alogorithm forhich patient and dissection characteristics would be best

ABLE 3. Operative timesCentral Peripheral P value

CA time 35.1 � 3.8 28.9 � 12.8 .162rossclamp time 114.4 � 13.0 89.2 � 8.1 .158PB time 155.5 � 11.2 180.3 � 8.6 .089

n terms of operative times, there was no difference significant differenceetween groups regarding time on hypothermic circulatory arrest (HCA),rossclamp time, or cardiopulmonary bypass (CPB) time.

erved by each of the cannulation options. f

The Journal of Thoracicjtcs.ctsnetjouDownloaded from

Although case numbers remain relatively small, severaleports of cannulating the dissected ascending aorta, includ-ng this one, have now suggested that this approach isndeed safe. We and others have used central cannulationor nearly two decades, but the technique does not appear inhe ascending aortic dissection literature until 1998 from antalian group.19 This early experience suggested the feasi-ility of this technique. Minatoya and colleagues20 furtheredhis reported experience using central cannulation in 14 of1 ascending aortic dissections over a 2-year period. Theylso believed that this approach provided more natural flownd avoided extension of the dissection. Our data supporthe findings of these previously reported and unreportederies. In fact, these data imply that central cannulation mayven be superior to peripheral cannulation in terms of lowerostoperative mortality and fewer perioperative myocardialnfarctions. We have several theories as to why this may berue. First, as postulated in the repair of ascending aorticneurysms, the antegrade, or “natural,” flow pattern is moreikely to be preserved with central cannulation. Westaby andolleagues21 studied central cannulation in arch and de-cending aortic aneurysms using circulatory arrest. Theyoncluded that thoracoabdominal aortic perfusion through aemoral cannula predisposed patients to higher retrogradembolic risk. They suggested that cannulation of the ascend-ng aorta close to the brachiocephalic vessel decreased thisisk because of preservation of the natural blood flow pat-ern rather than a potentially more turbulent retrograde flowhat may lift and embolize plaque. These principles couldertainly translate to ascending aortic dissection. Moreover,entral cannulation may decrease the incidence of malper-

ABLE 4. Complications and disposition

omplication/dispositionCentral

cannulationPeripheral

cannulationP

value

nfection 21% (5/24) 20% (9/46) .8ulmonary 21% (5/24) 28% (13/46) .9enal 12.5% (3/24) 17% (8/46) .6troke/TIA 21% (5/24) 28% (13/46)rrhythmia 8% (1/24) 15% (7/46) .4I 0% (0/24) 15% (7/46) �.01*ny cardiac 12.5% (3/24) 30% (14/46) .07ther (ileus) 4% (1/24) 9% (4/46) .4ny 33% (8/24) 51% (24/46) .11ospital mortality 4% (1/24) 20% (9/46) .150-Day mortality 0% (0/24) 17% (8/46) .04*ehab/SNF placement 26% (6/23) 30% (11/37) .8

IA, Transient ischemic attack; MI, myocardial infarction; Rehab, rehabil-tation; SNF, skilled nursing facility. *Complications were similar betweenroups except that the peripheral cannulation group experienced signifi-antly more perioperative myocardial infarctions and had a significantlyigher 30-day mortality than the central cannulation group.

usion syndromes during cardiopulmonary bypass that can

and Cardiovascular Surgery ● Volume 133, Number 2 431 on June 11, 2013 rnals.org

bttmgippm

tcdpstatclwdtberttwsolcieAcaf

Mbma5rcicseisampacai

aiaatpwtwfdcompti

T

Surgery for Acquired Cardiovascular Disease Reece et al

4

ACD

e intrinsic to peripheral cannulation techniques. The insti-ution of antegrade flow into the true lumen should, inheory, reduce the possibilities of distal malperfusion inas-

uch as restoration of flow to the true lumen is the ultimateoal when treating complicated dissections of the descend-ng aorta. Even when the false lumen is cannulated, the flowatterns are similar to those in the dissected state. Theressure is lower when pulsatile perfusion is abolished,inimizing the ongoing progression of the dissection.Despite these favorable findings, these data are not meant

o advocate central cannulation approaches over peripheralannulation techniques. Instead, these data are intended toemonstrate that central cannulation is a safe option in someatients. The site of cannulation can be tailored to both thepecifics of the dissection and the patient. For instance,here are cases in which central cannulation should bevoided. The risk of embolus with cannulation throughhrombosed false lumen or intramural hematoma may makeentral cannulation prohibitive. Therefore, the presence andocation of clot needs to be carefully considered, usuallyith a combination of preoperative and intraoperative ra-iologic imaging, echocardiography, and direct visualiza-ion. If clot is present, another cannulation approach shoulde considered. Futhermore, in cases that may involve morextensive arch work or longer hypothermic circulatory ar-est time, the axillary approach may be more favorable thanhe central approach to allow antegrade cerebral perfusionechniques. Femoral cannulation may be relatively easy,ith an easy vascular repair at the end of the case. But a

ubset of patients with aortic dissections extending distallyr with extensive peripheral vascular disease may be moreikely to experience malperfusion or arterial injury at theannulation site.19,22,23 We try to avoid femoral cannulationn elderly patients with extensive aortic atheroma on preop-rative imaging to avoid potential retrograde emoblization.xillary cannulation has become increasingly popular re-

ently in both ascending aortic dissections and ascendingortic aneurysms, especially those that may require some

ABLE 5. Factors for choosing cannulation siteCannulation site Favorable characterist

Central Type A dissectionHemodynamic instability

Axillary Anticipated arch replacementDissected arch aneurysm

Femoral Type A dissection (confined to asHemodynamic instability

orm of circulatory arrest. Strauch and colleagues11 from s

32 The Journal of Thoracic and Cardiovascular Surgery ● Febrjtcs.ctsnetjournDownloaded from

ount Sinai have suggested that axillary cannulation mighte the optimal technique for reducing perfusion-relatedorbidity and adverse outcomes in both dissections and

therosclerotic aneurysms. Still, even in their deft hands,% of patients required alternative cannulation for variouseasons or had complications attributable to the axillaryannulation, supporting our thought that no single approachs ideal for all patients. Specific patients in whom axillaryannulation may need to be avoided are those whose dis-ections extend into the axillary artery, those with an ath-rosclerotic axillary artery that may be prone to iatrogenicnjury, those with a small axillary artery that may notupport sufficient perfusion flow, those with vascular anom-lies, and, finally, those with hemodynamic instability thatay require more urgent initiation of cardiopulmonary by-

ass.24 On the basis of the specific aspects of the dissectionnatomy and the patients’ comorbidities, favorable sites forannulation can be determined and unfavorable sites can bevoided to optimize potential outcomes, which are depictedn Table 5.

We acknowledge that this study has its limitations. It issingle institutional retrospective study that can carry with

t significant bias. Within the institution, though, variouspproaches were used, providing a mixture of techniquesnd philosophies for comparison. Despite these limita-ions, the cohorts did appear to be similar before therocedures. The total number of patients was small,hich could create the potential for a type II error, but

his is the largest number of cases published to date inhich this technique was used. The limited number also

orced comparison between central cannulation and twoifferent peripheral cannulation techniques. The peripheralannulation techniques described here are different fromne another. However, they are currently the most com-only accepted sites of cannulation to which we can com-

are central cannulation for overall safety. We also concedehat the strategies used for cerebral protection and monitor-ng, as well as radiographic evaluation over the time of this

Unfavorable characteristics

Thrombosed false lumenIntramural hematomaAnticipated arch replacementAxillary dissectedSmall or atherosclerotic vesselHemodynamic instability

ing aorta) Femoral dissectionExtensive peripheral vascular diseaseAortic atheromaObesity

ics

cend

tudy, were not consistent but were similar between groups.

uary 2007 on June 11, 2013 als.org

Fgmtfrtstirrplad

lsnfbwatdcutcspaspp

fr

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

DDpotc

tp

Reece et al Surgery for Acquired Cardiovascular Disease

ACD

inally, the adjunct procedures were similar between theroups overall, but the central cannulation group underwentore valve resuspensions and fewer procedures involving

he arch than the peripheral cannulation group. These dif-erences could potentially confound our interpretation of theesults. Since our intention was to demonstrate the safety ofhe technique rather than arguing that one approach isuperior, we believe our conclusion remains valid despitehese differences. Additionally, compared with other stud-es, our patients were more likely to undergo coronaryevascularization. However, both the number of patientsevascularized and the number of vessels revascularized peratient were similar between our groups. Despite theseimitations, we believe these data demonstrate that centralortic cannulation of the dissected ascending aorta can beone safely when approached cautiously and meticulously.

In summary, these data have shown that central cannu-ation of the dissected ascending aorta can be performedafely. In particular, we found not only similar rates ofeurologic complications, but also no difference in the needor placement in skilled nursing and rehabilitation facilitiesetween the groups. Of note, the central cannulation groupas less likely to have a postoperative myocardial infarction

nd had a lower 30-day mortality; however, hospital mor-ality was similar between groups. These results suggest thatevotion to a single approach for cannulation in these casesan be avoided. Although all three options can be safelysed, we believe that the site of cannulation should beailored to each specific patient on the basis of patientharacteristics and dissection anatomy. All three methodshould be considered to optimize the care of these difficultatients. Although this study does not advocate using thispproach on all cases of ascending aortic dissection, it doesuggest that central cannulation can be used as safely aseripheral cannulation, providing another option in the ap-roach to this complex pathologic condition.

We thank Kimberly Shockey for her statistical expertise usedor this study and Sandra Burks for facilitating the institutionaleview board protocol and its approval.

eferences

1. Hirst AE, Johns VJ, Kime SW. Dissecting aneurysm of the aorta: areview of 505 cases. Medicine (Baltimore). 1958;37:217-79.

2. Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ,Russman PL, et al. The International Registry of Acute Aortic Dis-section (IRAD): new insights into an old disease. JAMA. 2000;283:897-903.

3. Bavaria JE, Brinster DR, Gorman RC, Woo YJ, Gleason T, PochettinoA. Advances in the treatment of acute type A dissection: an integratedapproach. Ann Thorac Surg. 2002;74:S1848,52; discussion S1857-63.

4. Gleason TG. Contemporary management of aortic dissection: intro-duction. Semin Thorac Cardiovasc Surg. 2005;17:213.

5. Gallo A, Davies RR, Coe MP, Elefteriades JA, Coady MA. Indica-tions, timing, and prognosis of operative repair of aortic dissections.Semin Thorac Cardiovasc Surg. 2005;17:224-35.

6. Kazui T, Yamashita K, Washiyama N, Terada H, Bashar AH, SuzukiT, et al. Impact of an aggressive surgical approach on surgical outcome w

The Journal of Thoracicjtcs.ctsnetjouDownloaded from

in type A aortic dissection. Ann Thorac Surg. 2002;74:S1844,7;discussion S1857-63.

7. Kirklin JW, Kouchoukos NT. Kirklin/Barratt-Boyes cardiac surgery:morphology, diagnostic criteria, natural history, techniques, results,and indications. 3rd ed. Philadelphia: Churchill Livingstone; 2003. p.1938, I-83.

8. Cohn LH, Edmunds LH. Cardiac surgery in the adult. 2nd ed. NewYork: McGraw-Hill; 2003. p. 1573.

9. Kaiser LR, Kron IL, Spray TL. Mastery of cardiothoracic surgery.Philadelphia: Lippincott-Raven Publishers; 1998. p. 976.

0. Cope JT, Tribble RW, Komorowski B, Tribble CG. A simple tech-nique for retrograde cerebral perfusion during circulatory arrest.J Card Surg. 1996;11:65-7.

1. Strauch JT, Spielvogel D, Lauten A, Lansman SL, McMurtry K,Bodian CA, et al. Axillary artery cannulation: routine use in ascendingaorta and aortic arch replacement. Ann Thorac Surg. 2004;78:103-8.

2. Fusco DS, Shaw RK, Tranquilli M, Kopf GS, Elefteriades JA. Femoralcannulation is safe for type A dissection repair. Ann Thorac Surg.2004;78:1285-9.

3. Panos A, Murith N, Bednarkiewicz M, Khatchatourov G. Axillarycerebral perfusion for arch surgery in acute type A dissection undermoderate hypothermia. Eur J Cardiothorac Surg. 2006;29:1036-9.

4. Olsson C, Thelin S. Antegrade cerebral perfusion with a simplifiedtechnique: unilateral versus bilateral perfusion. Ann Thorac Surg.2006;81:868-74.

5. Veron S, Neri E, Buklas D, Pula G, Benvenuti A, Massetti M, et al.Cannulation of the extrathoracic left common carotid artery for tho-racic aorta operations through left posterolateral thoracotomy. AnnVasc Surg. 2004;18:677-84.

6. Reuthebuch O, Schurr U, Hellermann J, Pretre R, Kunzli A, Lachat M,et al. Advantages of subclavian artery perfusion for repair of acute typeA dissection. Eur J Cardiothorac Surg. 2004;26:592-8.

7. Mazzola A, Gregorini R, Villani C, Di Eusanio M. Antegrade cerebralperfusion by axillary artery and left carotid artery inflow at moderatehypothermia. Eur J Cardiothorac Surg. 2002;21:930-1.

8. Mazzola A, Gregorini R, Villani C, Di Eusanio M. Antegrade cerebralperfusion by axillary artery and left carotid artery inflow at moderatehypothermia. Eur J Cardiothorac Surg. 2002;21:930-1.

9. Lijoi A, Scarano F, Dottori V, Parodi E, Casali G, Bartolozzi F.Stanford type A aortic dissection. A new surgical approach. Tex HeartInst J. 1998;25:65-7.

0. Minatoya K, Karck M, Szpakowski E, Harringer W, Haverich A.Ascending aortic cannulation for Stanford type A acute aortic dissec-tion: another option. J Thorac Cardiovasc Surg. 2003;125:952-3.

1. Westaby S, Katsumata T, Vaccari G. Arch and descending aorticaneurysms: influence of perfusion technique on neurological outcome.Eur J Cardiothorac Surg. 1999;15:180-5.

2. Robicsek F, Zimmern SH, Howe HR. Subintimal retrograde perfusionduring repair of aortic dissection: a potential cause of disaster. AnnVasc Surg. 1988;2:298-302.

3. Parr GV, Manley NJ, Williams DR, Montesano RM. Obstruction ofthe true lumen during retrograde perfusion of type I aortic dissections:a simplified solution. Ann Thorac Surg. 1980;30:495-8.

4. Schachner T, Nagiller J, Zimmer A, Laufer G, Bonatti J. Technicalproblems and complications of axillary artery cannulation. Eur J Car-diothorac Surg. 2005;27:634-7.

iscussionr Robert C. Robbins (Stanford, Calif). I guess what yourresentation really means is that you cannot expect any differentutcome if you keep doing things the same way and so you arerying to make us think outside the box a little bit and considerentral cannulation. I have just a couple of brief questions.

Did the date of the operations differ across the distribution ofime, or was central cannulation evenly distributed across the studyeriod?

Dr Reece. It was not evenly distributed. Central cannulation

as used a little more in the more recent period, but Dr Tribble

and Cardiovascular Surgery ● Volume 133, Number 2 433 on June 11, 2013 rnals.org

tlpp

Tmao“y

ihwc

dubtahpwm

hyr

ceoei

Ipstta

hmtpapaw

Surgery for Acquired Cardiovascular Disease Reece et al

4

ACD

ells me that he has been using it since 1986. As some of the peopleike Dr Kern come on, they were using it a little more, androbably an increased number of the attendings used it as the studyeriod went on.

Dr Robbins. That leads to the next comment. I suspect that Drribble would probably be the one who uses it the most, Dr Kronaybe the least, and Dr Kern in the middle, so how does this

lgorithm really fit? Is there really a lot of preoperative planning,r would, say, Dr Tribble consider central cannulation as thedefault” and use it unless he could not do so for some reason thatou have alluded to?

Dr Reece. I think that for the most part all of them will uset. Dr Tribble is definitely more likely to use it as his default, bute is the one who came up with the algorithm and the idea thate could focus it on each particular patient and his or her

haracteristics.Dr. Robbins. It is too bad we are not going to have other

iscussants because I am sympathetic to this technique, havingsed it a few times first because I could not get a cannula upecause of severe atherosclerosis. I think it is actually a techniquehat can be used, but I think most everyone else would go to thexillary route. Particularly, as you alluded to, if you are going toave longer circulatory arrest cases, then the axillary route isreferable, but I would argue that the axillary is preferable any-ay. I have rarely seen an axillary artery dissected or have so

uch atherosclerosis that it was not usable. m

34 The Journal of Thoracic and Cardiovascular Surgery ● Febrjtcs.ctsnetjournDownloaded from

I have just one more comment about holding the cannula in. Iave found that it is pretty messy and you have to figure out howou can actually get in the true lumen. Speaking with Dr Kern, iteally probably does not matter if there is a large intimal tear.

The one real question is, you had no deaths and really very fewomplications in the central cannulated patients versus the periph-ral, so why do you think that was, particularly the large numberf perioperative myocardial infarctions that you saw in the periph-ral patients? I don’t know that you’ll have an answer for this, butt was interesting and I think the main issue of this paper.

Dr Reece. I cannot really explain it from the preoperative data.think it is probably more a function of the small number of

atients in the study. As we continue with this algorithm, we mayee the difference decline. I do not think there is anything intrinsico either type of cannulation that would make myocardial infarc-ion more likely to occur, but that we are seeing the limitations ofsmall study group.

Dr Robbins. I have just one last comment. Since Dr Tribbleas been doing it for 20 years now, I do not really see why intimalural hematoma would be a problem. If there is no communica-

ion and you know you are in the true lumen, then you can justerfuse down the true lumen and there is some advantage to goingntegradely. Similarly with the clot in the false lumen, I think it isrobably more problematic going retrogradely. Therefore, I wouldrgue that most people would say that the axillary artery is the bestay to go, and that is the way I would do it, but you could easily

ake an argument that you could go centrally for all cases.

uary 2007 on June 11, 2013 als.org

DOI: 10.1016/j.jtcvs.2006.09.059 2007;133:428-434 J Thorac Cardiovasc Surg

Stiles, Benjamin B. Peeler, John A. Kern and Irving L. Kron T. Brett Reece, Curtis G. Tribble, Robert L. Smith, R. Ramesh Singh, Brendon M.

Central cannulation is safe in acute aortic dissection repair

Continuing Medical Education Activities

http://cme.ctsnetjournals.org/cgi/hierarchy/ctsnetcme_node;JTCSSubscribers to the Journal can earn continuing medical education credits via the Web at

Subscription Information

http://jtcs.ctsnetjournals.org/cgi/content/full/133/2/428#BIBLThis article cites 21 articles, 13 of which you can access for free at:

Citations

http://jtcs.ctsnetjournals.org/cgi/content/full/133/2/428#otherarticlesThis article has been cited by 15 HighWire-hosted articles:

Subspecialty Collections

http://jtcs.ctsnetjournals.org/cgi/collection/great_vessels Great vessels

This article, along with others on similar topics, appears in the following collection(s):

Permissions and Licensing

http://www.elsevier.com/wps/find/obtainpermissionform.cws_home/obtainpermissionformreceipt, is available at: An on-line permission request form, which should be fulfilled within 10 working days of

. http://www.elsevier.com/wps/find/supportfaq.cws_home/permissionusematerialcan be found online at: General information about reproducing this article in parts (figures, tables) or in its entirety

on June 11, 2013 jtcs.ctsnetjournals.orgDownloaded from