Hospital Outcome Analysis After Different Techniques of Left Internal Mammary Grafts Harvesting

DOI: 10.1016/j.athoracsur.2007.06.024 2007;84:1912-1919 Ann Thorac Surg

Mastroroberto and Attilio Renzulli Francesco Onorati, Antonio Esposito, Francesco Pezzo, Antonio di Virgilio, Pasquale

Grafts HarvestingHospital Outcome Analysis After Different Techniques of Left Internal Mammary

http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912located on the World Wide Web at:

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

Print ISSN: 0003-4975; eISSN: 1552-6259. Southern Thoracic Surgical Association. Copyright © 2007 by The Society of Thoracic Surgeons.

is the official journal of The Society of Thoracic Surgeons and theThe Annals of Thoracic Surgery

by on May 31, 2013 ats.ctsnetjournals.orgDownloaded from

http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912

http://ats.ctsnetjournals.org

HTHFAC

cnlpso

tJtvraigs(

t

Ligasdsse[acds

wodA

A

AI

©P

CA

RD

IOV

ASC

ULA

R

ospital Outcome Analysis After Differentechniques of Left Internal Mammary Graftsarvesting

rancesco Onorati, MD, Antonio Esposito, MD, Francesco Pezzo, MD,ntonio di Virgilio, MD, Pasquale Mastroroberto, MD, and Attilio Renzulli, MD, PhD

ardiac Surgery Unit, Magna Graecia University of Catanzaro, Naples, Italy

apebctfllhgsp0

mnho

Background. Although accepted worldwide as the first-hoice conduit for myocardial revascularization (coro-ary artery bypass graft surgery), it is still questionable if

eft internal mammary arteries (LIMAs) do better asedicled or skeletonized grafts. Recent reports haveuggested that the skeletonized technique improves theutcomes.Methods. In all, 516 consecutive isolated coronary ar-

ery bypass graft surgery patients admitted betweenanuary 2003 and February 2007 were grouped accordingo the harvesting technique of LIMAs (387 pedicledersus 129 skeletonized). Intraoperative transit time flowesults, hospital outcome, perioperative troponin I leak-ge, and echocardiographic results were analyzed. Max-mum, mean, and minimum flows, pulsatility index, andraft flow reserve were stratified according to the type ofurgery (on pump and off pump) and subgroups at risksevere left ventricular hypertrophy).

Results. The transit time flow results were unsatisfac-
ory in 4 patients (0.8%), all 4 of whom underwent redo
csMdkitLcpfl

pavpstsTd

address correspondence to Dr Onorati, Viale dei Pini, 28, Napoli, 80131,

taly; e-mail: [email protected].

2007 by The Society of Thoracic Surgeonsublished by Elsevier Inc

ats.ctsnetjournDownloaded from

nastomosis. Maximum, mean, and minimum flows andulsatility index values were comparable between skel-tonized and pedicled LIMAs (p � not significant), foroth on-pump and off-pump surgeries (p � not signifi-ant). Graft flow reserve was comparable between thewo groups (p � not significant), as well as transit timeow results in severe ventricular hypertrophy. Troponin

eakage, postoperative echocardiographic recovery, andospital outcome were comparable between the tworoups (p � not significant at all time points). Onlyuperficial wound complications proved higher in theedicled group (2.6% versus skeletonized group: 0%; p �.05).Conclusions. Pedicled LIMA carries a higher risk forinor wound complications. Neverthless, the two tech-

iques showed similar in-vivo functional behavior andospital outcome. Therefore, there is no reason to preferne technique over the other.

(Ann Thorac Surg 2007;84:1912–9)
© 2007 by The Society of Thoracic Surgeons
eft internal mammary arteries (LIMAs) have becomethe first-choice conduit for myocardial revascular-

zation [1, 2]. Given its long-term survival benefit, sur-eons must harvest the LIMA in the best possible way tochieve both short-term and long-term advantages. De-pite the popularity of the LIMA, however, it is stillebated whether to use it as a pedicled graft or as akeletonized graft. A pedicled graft is the more rapid,impler, and safer method to harvest the conduit. Skel-tonization of the graft is more technically demanding3]. However, the literature has often reported somedvantages of the skeletonized LIMA over the pedicledonduit, ranging from the achievement of a longer con-uit [4], to a higher free flow [4–6], to the reduced risk ofternal wound complications [7, 8].

Although the survival advantage of LIMAs comparedith other grafts is accepted worldwide, no clear benefitsf one harvesting technique over the other have beenefinitely demonstrated. In particular, a recent review bythanasiou and coworkers [3] has shown the only ac-

ccepted for publication June 6, 2007.

epted evidence, grade A, is the induction of a significantternal devascularization with bilateral pedicled IMAs.

oreover, despite a number of studies that have ad-ressed the superiority of the skeletonized LIMA, little isnown about the in-vivo functional behavior of skeleton-

zed and pedicled grafts. Only one study analyzed transitime flow (TTF) results of skeletonized and pedicledIMAs [9]. Few studies have described the free-flowapacity of the graft, often after intragraft injection ofapaverine [5, 6, 10]. No study has reported on the graftow reserve (GFR) of these conduits.Finally, some authors have shown that the risk of

ostoperative hypoperfusion syndrome, a rare perioper-tive condition associated with low cardiac output, leftenticular failure, and cardiac arrest, is higher whenedicled LIMAs are used [10, 11]. The major pathophy-ologic factor for that syndrome seems to be a dispropor-ion between LIMA flow and myocardial demand due toevere venticular hypertrophy [10, 11]. IntraoperativeTF analysis is, therefore, able to detect early such areadful postoperative complication [9].Therefore, the aim of this study was to systematically

nalyze our experience with TTF and in-hospital results

0003-4975/07/$32.00doi:10.1016/j.athoracsur.2007.06.024

by on May 31, 2013 als.org


omtph

M

TsoLftp

mw

bepLcdpsccc

APras

SAswlL

T

ASEDHDCAALLPPPPLCRACNBT

ET

1913Ann Thorac Surg ONORATI ET AL2007;84:1912–9 SKELETONIZED VERSUS PEDICLED MAMMARY GRAFTS

CA

RD

IOV

ASC

ULA

R

f pedicled versus skeletonized LIMAs in first-timeyocardial revascularization, stratifying the results by

he surgical technique employed (on pump and offump), and associated risk factor (severe ventricularypertrophy).

aterial and Methods

he present study evaluates clinical and flowmetric re-ults of a prospective series of patients undergoing cor-nary artery bypass graft surgery (CABG) with pedicledIMA (group A) or skeletonized LIMA (group B), per-

ormed either off pump (OPCABG) or on pump, duringhe last 4 years at a single academic institution. The studyrotocol was approved by the institution’s Ethical Com-

able 1. Preoperative and Intraoperative Data

Grou

geex (male)uroSCORE (mean � SD)iabetes mellitisypertensionyslipidemiahronic obstructive pulmonary diseasecute coronary syndromecute myocardial infarction � 4 weekseft main stem diseaseeft ventricular hypertrophy (ILVM � 125 g/m2)reoperative intra-aortic balloon pumpreoperative EF � 30%reoperative EF 31%–49%reoperative EF � 50%eft anterior descending artery TIMI scoreircumflex artery TIMI scoreight coronary artery TIMI scoreortic cross-clamp timeardiopulmonary bypass timeumber of anastomoses/patientilateral internal mammary artery graftingotal arterial grafting

Abbreviations and Acronyms

CABG � coronary artery bypass graft surgeryGFR � graft flow reserveIABP � intra-aortic balloon pumpingLAD � left anterior descending arteryLIMA � left internal mammary arteryOPCABG � off-pump CABGPI � pulsatility indexTTF � transit time flow

F � ejection fraction; EuroSCORE � European System for Cardiac Operatihrombolysis in Myocardial Infarction study.


ittee/Institutional Review Board, and informed consentas obtained.Five hundred-twenty consecutive patients, admitted

etween January 2003 and February 2007 for first-timelective CABG, were enrolled and grouped to receiveedicled LIMA (group A, 389 patients) or skeletonizedIMA (group B, 131 patients) according to the surgeon’shoice. Additional cardiac or vascular surgical proce-ures and severe systemic comorbidities (dyalisis, he-atic failure, cancer, autoimmune disease) were exclu-ion criteria. Because the different geometry for theonstruction of a Y graft or T graft influenced TTF results,oronary revascularization with these grafts were ex-luded from the analysis.

ngiographyreoperative coronary angiography of each patient waseviewed by two cardiologists, blinded toward the study,nd Thrombolysis in Myocardial Infarction study (TIMI)cores were calculated.

urgeryll surgical procedures were performed by the same

urgeons (A.R., P.M., A.dV., F.O.). In all patients, CABGas performed through a median sternotomy, and the

eft anterior descending artery always grafted using theIMA.The pedicled LIMA was harvested together with the

387 Patients) Group B (129 Patients) p Value

� 9.8 70.0 � 7.5 0.275(64.6%) 84 (65.1%) 0.502� 2.92 3.48 � 2.66 0.426

(56.1%) 83 (64.3%) 0.061(71.6%) 101 (78.3%) 0.083(58.7%) 66 (51.2%) 0.151(66.1%) 83 (64.3%) 0.393(50.9%) 70 (54.3%) 0.288(28.9%) 42 (32.6%) 0.251(36.2%) 39 (30.2%) 0.131(38.5%) 52 (40.3%) 0.396(36.7%) 51 (39.5%) 0.317(17.1%) 30 (23.3%) 0.077(56.1%) 72 (55.8%) 0.520(26.9%) 27 (20.9%) 0.109� 0.70 0.77 � 0.62 0.186� 0.65 1.02 � 0.59 0.268� 0.64 0.85 � 0.52 0.217� 22.7 46.8 � 25.9 0.708� 19.6 85.9 � 24.6 0.716� 0.57 3.77 � 0.82 0.819

(11.4%) 18 (14.0%) 0.262(26.4%) 36 (27.9%) 0.406

p A (

67.82503.5421727722725619711214014914266

2171040.851.060.9043.582.33.4644

102

ve Risk Evaluation; IVLM � indexed left ventricular mass; TIMI �



smwfifpuo

ntfvmpalthvw

dabiguhTwss

D

T

IIIP1247

GG

GG

a

E ore in

TS

L

L

TSO

O

C

C

1914 ONORATI ET AL Ann Thorac SurgSKELETONIZED VERSUS PEDICLED MAMMARY GRAFTS 2007;84:1912–9

CA

RD

IOV

ASC

ULA

R

urrounding veins, muscle, and fascia. Low-voltage (20V) cautery was always used, and the side branchesere clipped. Proximal dissection was always above therst rib. The width of pedicled LIMAs always ranged

rom 1 to 1.5 cm. To avoid any distortion to the graft, theedicle was always secured with two epicardial stitchessing 6-0 polypropylene on both sides after completionf distal anstomoses.When the LIMA conduit was skeletonized, the tech-

ique was as follows. After having dissected the reflec-ion of the mediastinal pleura from the endothoracicascia, the mammary artery and both satellite veins wereisualized. The fascia was incised medially to the medialammary vein for the whole length of the vessel, and

ulled down on the pleura. A blunt dissection of thertery from the chest wall was performed by means ofow-voltage (10 mV) cautery or cold tip; the sternal andhe anterior intercostal branches were occluded withemoclips. Collaterals were always divided using low-oltage cautery, and its mammary side similarly occludedith small hemostatic clips.Both pedicled and skeletonized LIMAs were cut at the

able 2. Perioperative Inotropic Support, Troponin I Leakage,

Group A

notropes low doses 300/387 (77.5%)notropes medium doses 76/387 (19.6%)notropes high doses 11/387 (2.8%)reoperative TnI (�g/l) 0.04 � 0.092-hour TnI (�g/l) 1.63 � 0.804-hour TnI (�g/l) 1.26 � 1.418-hour TnI (�g/l) 1.17 � 1.142-hour TnI (�g/l) 1.10 � 2.12

Preoperative EF

roup A 44.9 � 9.9%roup B 43.6 � 9.6%

Preoperative WMSI

roup A 1.44 � 0.34roup B 1.54 � 0.36

p value between groups; b p value within a single group.

F � ejection fraction; TnI � troponin I; WMSI � wall motion sc

able 3. Transit Time Flow Results of Pedicled Versuskeletonized LIMA Grafts in the Global Population

Group A(387 Patients)

Group B(129 Patients) p Value

IMA (387 versus129 grafts)

Maximum flow 54.5 � 35.1 52.0 � 22.3 0.441Mean flow 29.9 � 21.3 27.2 � 12.4 0.174Minimum flow 3.1 � 10.6 3.9 � 5.5 0.391PI 2.10 � 0.98 2.15 � 0.62 0.593

IMA � left internal mammary artery; PI � pulsatility index.lb


istal end 1 cm proximal to bifurcation, and checked forgood pulsatile free flow (at least � 20 mL/minute)

efore their use. No vasodilators or other drugs werenjected inside the LIMA at any time. Two LIMAs inroup A and 2 skeletonized LIMAs in group B were notsed as conduits because of unsatisfactory free flow afterarvesting. These grafts were excluded from the study.herefore, 387 patients in group A and 129 in group Bere ultimately enrolled in the protocol. Distal anstomo-

es were always constructed using 8-0 polypropyleneutures.

A standard cardiopulmonary bypass circuit was used: aideco (Mirandola, Modena, Italy) tubing set, which

Perioperative Echocardiographic Results

Group B p Valuea p Valueb

92/129 (71.3%) 0.09633/129 (25.6%) 0.0974/129 (3.1%) 0.5400.03 � 0.04 0.2841.61 � 0.82 0.8721.29 � 0.87 0.9100.92 � 0.84 0.2700.82 � 1.24 0.477

operative EF

.3 � 9.1% 0.248 � 0.01

.8 � 7.8% � 0.01

operative WMSI

1.30 � 1.11 0.228 � 0.011.29 � 1.08 � 0.01

dex.

able 4. Transit Time Flow Results of Pedicled andkeletonized LIMA Grafts Stratified by Off-Pump andn-Pump Surgery

Group A Group Bp

Value

PCABG, LIMA (105 versus42 grafts)


PBCABG, LIMA (271versus 87 grafts)


PBCABG � on-pump coronary artery bypass graft surgery; LIMA �

and

Post

5252

Post

eft internal mammary artery; OPCABG � off-pump coronary arteryypass graft surgery; PI � pulsatility index.



iImMtpottMw

FIupppswiflomm[p

GAv(sbodt(wtipcsfd

PIr[rwtd6dwta

BDpkufnga

EAdpssa

SSftcnittteos

T

L

a

G


CA

RD

IOV

ASC

ULA

R

ncluded a 40 �m filter, a Stockert roller pump (Stockertnstrumente, Munich, Germany), and a hollow fiber

embrane oxygenator (Dideco D903 Avant; Mirandola).yocardial protection was achieved by using blood po-

assium cardioplegia as previously reported [12]. Of 516atients enrolled, 116 (29.9%) of group A and 42 (32.5%)f group B underwent OPCABG. Exposure and stabiliza-ion were achieved with the Octopus-IV with or withouthe Starfish-II tissue stabilizer (Medtronic, Minneapolis,

innesota). Intracoronary shunts were used routinely,ith a success rate approximating 100%.

lowmetric Analysisn both groups, assessment of each graft was performednder stable hemodynamic conditions (mean arterialressure at 70 to 80 mm Hg), generally 30 minutes afterrotamine administration. Flowmetry of the grafts waserformed with a transit-time flowmeter (HT313 Tran-onic; Transonic Systems, Ithaca, New York). The curvesere always coupled with the electrocardiographic trac-

ng to correctly differentiate the systolic from the diastolicow. The TTF measurements were interpreted as previ-usly suggested by D’Ancona and colleagues [13]. Theaximum, minimum, and mean flows were reported asL/min, and pulsatility index (PI) as an absolute number

13]. Data from LIMA conduits were recorded and com-ared between the two groups.

raft Flow Reservell patients with ongoing unstable angina despite intra-enous nitrates with poor left ventricular ejection fraction� 40%) with or without severe narrowing of the left maintem or ST-segment elevation underwent intra-aorticalloon pump (IABP) before anesthetic induction, inrder to improve myocardial protection during the in-uction of anesthesia and the early phases of the opera-

ion. In particular, 142 patients in group A (36.7%) and 5139.5%) in group B underwent preoperative IABP, andere, therefore, the objective of GFR study. According to

he previous demonstration that IABP recruits GFR dur-ng assistance [14], mean flow and PI were recorded in allatients during IABP support and during temporaryessation. Both measurements were accomplished undertable hemodynamic conditions. The GFR was calculatedrom the mean flow assessed during 1:1 IABP supportivided by mean flow at baseline (IABP off) [15].

able 5. Transit Time Flow Results During 1:1 Intra-Aortic B

Group A Group B

IMA

Mean flow 58.3 � 23.2 51.1 � 27.9

PI 3.75 � 1.12 3.88 � 1.06

GFR 1.94 � 0.75 1.87 � 1.05

p � 1:1 IABP versus baseline (IABP off). b p � pedicled versus skele

FR � graft flow reserve; LIMA � left internal mammary artery; PI �


ostoperative Carenotropes were started immediately after aortic cross-clampemoval with enoximone at a dosage of 5 �g · kg�1 · min�1

12, 15]. The need to further increase inotropes wasecorded. Inotropic support was defined as low dosehen enoximone was administered at a dosage lower

han or equal to 5 �g · kg�1 · min�1; medium dose wasefined when enoximone was used at a dosage betweenand 10 �g · kg�1 · min�1 or dobutamine was added at aosage between 5 and 10 �g · kg�1 · min�1; high dose washen enoximone or dobutamine infusion was greater

han 10 �g · kg�1 · min�1 or epinephrine at any dose wasdded.

iochemical Analysiseterminations of cardiac troponin I were conductedreoperatively and at 12, 24, 48, and 72 hours. Diagnosticits provided by Beckman Coulter for troponin I weresed (Access Immunoassay System; AccuTnI). Criteria

or perioperative myocardial infarction were peak tropo-in I greater than 3.7 �g/L, and troponin I concentrationreater than 3.1 �g/L at 12 hours or greater than 2.5 �g/Lt 24 hours [12, 15].

chocardiographyll studies were performed using transthoracic echocar-iography with an Acuson Sequoia C256 (Acuson Cor-oration, Mountain View, California), always done by theame physician in a blinded manner, at hospital admis-ion and before discharge. Left ventricle ejection fractionnd wall motion score index were recorded.

tatistical Analysistatistical analysis was performed by the SPSS program

or Windows, version 10.1 (SPSS, Chicago, Illinois). Con-inuous variables are presented as mean � SD, andategorical variables are presented as either absoluteumbers or percentages. Data were checked for normal-

ty before statistical analysis. Normally distributed con-inuous variables were compared using the unpaired test, whereas the Mann-Whitney U test was used forhose variables that were not normally distributed. Cat-gorical variables were analysed using either the �2 testr Fisher’s exact test. Comparisons were consideredignificant if the p value was less than 0.05.

n Pump (IABP) Assistance and Graft Flow Reserve

p Valuea p Valueb

0.001 Group A0.529

0.001 Group B0.01 Group A

0.7160.01 Group B

— — 0.673

d.

alloo

tonize

pulsatility index.



R

Ta

gt2egfov

nt[appI

et[A0dbatt2h6ehgt(

d[ssLtesrapbattpm

u

cecgmssbfch

sfmBataortmgtwm

C

Latatiigcvttfi

TMiP

MMMPG


CA

RD

IOV

ASC

ULA

R

esults

he two groups demonstrated comparable demographicnd intraoperative data (Table 1).Hospital mortality was comparable: 10 patients from

roup A (2.6%) and 3 from group B (2.3%) died duringheir in-hospital stay. Of these, 3 patients in group A and

patients in group B had low-output state after periop-rative myocardial infarction. Another 4 patients fromroup A and 1 patient from group B had multiorganailure after pneumonia. Two patients from group A diedf stroke. Finally, 1 patient from group A died of rightentricular rupture after deep sternal wound infection.When wound healing complications were considered,

o differences were recorded between the two groups inerms of deep sternal wound infection (group A, 5 of 3871.5%] versus group B, 1 of 129 [0.8%]; p � 0.533),lthough superficial wound infection was higher in theedicled group (10 of 387 [2.6%] versus group B, 0 of 129;� 0.05). None of these patients underwent bilateral

MA grafting.No diffferences were recorded between the two groups

ither in terms of postoperative acute myocardial infarc-ion (group A, 5 of 387 [1.3%] versus group B, 2 of 1291.6%]; p � 0.556) or need for postoperative IABP (group, 11 of 387 [2.8%] versus group B, 6 of 129 [4.7%]; p �

.232). No IABP-related complications were registereduring the period of assistance, except for a patientelonging to group A who had transient limb ischemia,nd recovered after IABP withdrawal. Perioperative ino-ropic support and troponin I proved to be similar in thewo groups (Table 2). Intensive care unit stay (group A,.1 � 1.6 days versus group B, 1.7 � 0.7; p � 0.137) andospital stay (group A, 6.7 � 0.6 days versus group B,.9 � 1.0; p � 0.606) also proved to be similar. Whenchocardiography was considered, except for patientsaving perioperative acute myocardial infarction, the tworoups demonstrated comparable recovery of left ven-ricular ejection fraction and wall motion score indexTable 2).

According to graft flowmetry, 4 LIMA to left anteriorescending artery (LAD) anastomoses (group A, 2 of 387

0.5%] versus group B, 2 of 129 [1.6%]; p � 0.261) demon-trated unsatisfactory resuts. Two patients of group Ahowed high PIs, low flows, and systolic patterns ofIMA to LAD anastomoses. In 1 case, that was due to a

orsion of the anastomosis secondary to its fixation to thepicardial fat, and TTF results completely recovered byimply relieving the fixation stitch. In 1 patient, graftevision demonstrated an early thrombosis of the heel,nd TTF results completely recovered after revision. Oneatient belonging to group B undergoing off-pump dou-le CABG demonstrated low maximum flow (9 mL/min)nd mean flow (3 mL/min), and high PI (5) of the LIMAo LAD anastomosis. The anastomosis was redone withhe aid of an intravascular shunt, and TTF results com-letely recovered (maximum flow, 48 mL/min; mean, 38L/min; PI 2.0, diastolic pattern of the curve).All 3 patients undergoing graft revision because of

nacceptable TTF results demonstrated an uncompli- t


ated postoperative course with optimal postoperativelectrocardiographic, echocardiographic, and biochemi-al (troponin I) results. Another patient belonging toroup B demonstrated a diastolic pattern but low maxi-um (13 mL/min) and mean flows (6 mL/min), with a

ufficient PI (4.2), compatible with the poor run-off of aeverely diseased LAD. The anastomosis was not redone,ut the patient had perioperative acute myocardial in-

arction requiring IABP assistance. The perioperativeourse was then uneventful, and the patient was dischargedome in healthy condition on postoperative day 13.When TTF values were considered, both pedicled and

keletonized LIMA showed comparable results (Table 3),or on-pump and off-pump surgery (Table 4). Further-

ore, 193 patients (142 belonging to group A, 51 to group) underwent preoperative IABP because of unstablengina despite maximal intravenous nitrates administra-ion. To rule out differences in GFR between pediclednd skeletonized IMAs, these patients underwent intra-perative TTF analysis with either 1:1 IABP and tempo-ary cessation. The 1:1 IABP support recruited GFR in allhese patients, and no differences were detected in graft

ean flow, pulsatility index, and GFR between the tworoups (Table 5). Finally, when patients with preopera-ive severe ventricular hypertrophy at echocardiographyere analyzed, no differences in LIMA to LAD TTFeasurements were ruled out (Table 6).

omment

eft internal mammary arteries are accepted worldwides the first-choice conduits for myocardial revasculariza-ion because they demonstrate higher patency, survival,nd freedom from cardiac events [1–3], improved resis-ance to the development of atherosclerosis [16, 17], andeal coronary-to-conduit size match [18], and a function-ng arterial endothelium [4, 18, 19]. However, despite theeneral acceptance of the LIMA’s superiority to otheronduits, controversy still exists on the preferential har-esting technique. In particular, reports have advocatedhe superiority of the skeletonized technique over theraditional pedicled technique [5, 9, 16, 18], ranging fromunctional advantages as higher free flows [4, 5] andn-vivo TTF results [9], to better angiographic results in

able 6. Transit Time Flow Results of Left Internalammary Artery to Left Anterior Descending Artery Graft

n Patients With Severe Left Ventricular Hypertrophy atreoperative Echocardiography

Group A(149 Grafts)

Group B(52 Grafts) p Value

aximum flow 54.9 � 35.3 49.4 � 16.9 0.283ean flow 30.3 � 21.9 25.8 � 9.7 0.152inimum flow 3.5 � 10.8 2.3 � 6.7 0.103

ulsatility index 2.09 � 0.97 2.15 � 0.47 0.639raft flow reserve 1.67 � 0.86 1.88 � 1.08 0.257

erms of vascular resistance [9], to lower postoperative



pa

rnsfidlcwLirpbpmaw

emaswadmneaestdpfriitm

ictoi[vttewvleo

mu

ptfr(ialLetmgcosfgtttwpcgsldes

oafwbet

niwFt1s2sid[trdbt


CA

RD

IOV

ASC

ULA

R

ain [20], lower incidence of wound complications [3, 16],nd longer available conduits [16, 21].Conversely, clinical results and angiographic patency

ates appear to be comparable betwen the two tech-iques [3, 16, 18]. However, a recent review by Athana-iou and coworkers [3] showed that the only evidencerom the current literature is a higher sternal devascular-zation with the pedicled technique, which was clearlyemonstrated by the studies of Lorberboym and col-

eagues [22], Cohen and colleagues [7], and Parish andoworkers [8]. We found a higher incidence of superficialound complications in patients undergoing pedicledIMA harvesting, but comparable deep sternal wound

nfections in the two groups. It can be argued that theelatively low incidence of such complication in ourractice may explain such discrepancy. Moreover, it cane hypothesized that the relatively small width of theedicle (1 to 1.5 cm) and the low prevalence of bilateralammary grafting in the study period may further

ccount for these results. However, further studies arearranted to better define such topics.It has to be kept in mind that very few studies have

valuated the functional behavior of the conduits, [4–6, 9],ost focusing on the free-flow capacity [4, 5]. Both Deja

nd coworkers [4] and Wendler and coworkers [5] foundkeletonized LIMAs to have higher free flow comparedith pedicled conduits. Both studies measured flows

fter intra-arterial papaverine injection, which has beenemonstrated to cause endothelial damage to the mam-ary artery [23]. Accordingly, we never used intralumi-

al papaverine on LIMA or on the other arterial conduitsmployed. Furthermore, Wendler and coworkers [5] an-lyzed only a limited number of grafts (40 patients inach group), initially showing comparable pedicled andkeletonized graft free flows, which reverse in favor ofhe skeltonized technique after papaverine injection; weo not know how this different behavior could be ex-lained, especially when Deja and coworkers [4], who

ound similar response to papaverine, reported a compa-able functional behavior when the acetylcholine-nduced vasodilation was studied [4]. Finally, these stud-es did not report if differences existed between the twoechniques in the caliber of LIMA, which seems to be the

ost important factor for free-flow measurements.However, surgeons now have the possibility of study-

ng the functional “in-vivo” behavior of the grafts, afterompletion of distal anastomoses, through the aid of TTFechnology. The TTF method is fast, easy, and predictivef graft occlusion at short-term angiographies, and it

s less affected by artifacts than are other techniques13, 24]. In particular, not only are the absolute flowalues but also the flow curve patterns and the PI neededo correctly use this technology. In particular, accordingo D’Ancona and coworkers [13], and based on ourxperience, we take down and redo the anastomoses inhich systolic spiky patterns of the curves, low flow

alues, and PIs of 4 or more are reported by TTF, whereasow flows with preserved diastolic pattern and good PI,specially when associated with intraoperative findings
f a small and severely diseased coronary bed, do not g

andate graft revision. However, any unexpected andnexplained TTF finding should suggest graft revision.Only Takami and coworkers [9] used TTF to compare

edicled and skeletonized LIMA, and found skeletoniza-ion to give higher mean flows, but comparable PI. Apartrom this discrepancy, the study did not stratify theesults by grafted vessel, did not reported the qualityrun-off) of the grafted vessel (of paramount importancen TTF interpretation), and, most importantly, did notnalyze GFR. Moreover, Takami and coworkers [9] foundarger juxta-anastomotic diameters of skeletonizedIMAs, which can be, accordingly, the result of a differ-nt run-off of the grafted vessels in the two groups. Onhe contrary, we found comparable maximum, mean, and

inimum flows and PIs in pedicled and skeletonizedrafts, either after off-pump or on-pump CABG, withomparable TIMI scores of the grafted territories. More-ver, instead of papaverine injection, we chose IABP totudy GFR by a more physiologic approach [14, 15], andound comparable GFR in pedicled and skeletonizedrafts. These data confirm the reported comparable ace-ylcholine-induced vasorelaxation [4], and may explainhe reported comparable long-term angiographic pa-ency, survival, and freedom from cardiovascular eventsith these two techniques [3, 16, 18]. Accordingly, com-arable in-vivo behavior of pedicled and skeletonizedonduits resulted in similar in-hospital results in the tworoups; although the optimal flow values we found in thekeletonized group must be also attributed to the use ofow-current cautery, minimizing the risk of endothelialamage [25], and to the absence of a learning curve in ourxperience because only experienced surgeons employedkeletonization.

One potential disadvantage of the pedicled technique,ften underestimated by the authors, is the possibility ofpedicled torsion on its axis, leading to severe hypoper-

usion of a well-harvested pedicled graft. As in the casee reported, such a complication can be easily detectedy TTF, and should always be suspected when unexpect-dly poor TTF results are encountered. Obviously, ex-reme caution in securing a pedicled LIMA is mandatory.

Another supposed advantage of the skeltonized tech-ique is the direct visual inspection of the vessel to

dentify vessel wall injuries, hematomas, and dissections,hich, if unnoticed, may jeopardize the outcome [16].

irst, it has to be considered that the pedicled harvestingechnique is the only real “no-touch” technique [3, 10, 16,8]. Similarly, although skeletonization has been demon-trated to preserve the mammary function [1-6, 10, 18, 19,6], different pathologic studies show a tendency forkeletonized grafts to contain a larger number of lesionsn the endothelium, microscopic intimal dissection andetachment, and injuries to the external elastic lamina

23, 27]. It is well known, however, that intimal dissec-ions as well as any vessel wall injury and hematomas areesponsible for, at least, arterial spasm, leading to re-uced pulsation and free flow of the conduit, which maye easily recognized in the operating theater. Unsatisfac-

ory pulsatile flow of the LIMA may negate its use as a
raft, or at least induce the surgeon to inspect the graft


aeec

isflttlpvcsOcn

ssipfns

LTrcfTrmFi

iostiiidgnpafst

ttwale

ngrids

R

1

1

1

1

1

1

1

1


CA

RD

IOV

ASC

ULA

R

nd maybe skeletonize it if harvested as a pedicle, toxclude vessel wall damages. With such an approach, wexcluded 2 pedicled LIMAs during the last 4 years oflinical practice.

Finally, it has been suggested that skeletonization maymprove clinical results, reducing the risk of hypoperfu-ion syndrome due to a disproportion between LIMAow and myocardial demand because of a severe ven-

icular hypertrophy [10, 11]. Our results demonstratedhat GFR was recruited in a comparable manner, regard-ess of the harvesting technique employed. Furthermore,atients with echocardiographic evidence of severe leftentricular hypertrophy, who are at higher risk for suchatastophic complication, showed comparable TTF re-ults and GFR with pedicled and skeletonized LIMAs.ur data are consistent with those of Athanasiou and

oworkers [3], who demonstrated that skeletonization didot affect the incidence of hypoperfusion syndrome.We conclude that skeletonized and pedicled LIMA

howed a comparable in-vivo functional behavior, givingimilar TTF results and recruitment of GFR, comparablen-hospital results, similar troponin I leakage, and similarostoperative echocardiographic improvement. Apart

rom the risk of higher sternal devascularization or theeed for longer conduits, there is no reason to preferkeletonized to pedicled mammary grafts.

imitations of the Studyhe main limitation of the study is the absence ofandomization of the two surgical techniques. An idealomparison between two treatments should be per-ormed on the basis of a prospective randomized study.he study has been prospectively designed, but no realandomization has been performed, although such a goalay be very difficult to reach in current clinical practice.

urthermore, surgeons harvesting IMAs, and surgeonsmplanting LIMAs did not vary during the study period.

The study lacks angiographic control. Angiographys still the gold standard to evaluate long-term resultsf two different surgical strategies in coronary arteryurgery. However, it was not the purpose of the studyo assess graft patency, but to evaluate functionaln-vivo behavior of the LIMA grafts—with the aid ofntraoperative TTF as well as major perioperative clin-cal and subclinical results. Furthermore, it has beenemonstrated that the TTF method is predictive ofraft occlusion at angiographic follow-up [13, 24]. Fi-ally, different and complex mechanisms affect graftatency during follow-up, first of all the progression oftherosclerosis and comorbid conditions. It can there-ore be a mistake to only attribute graft patency to aingle surgical variable, such as the harvestingechnique.

Finally, this is a single-center study, which guaran-ees uniformity of the perioperative management ofhe patient population. On an intention-to-treat basis,e excluded patients undergoing sequential or T-graft

nd Y-graft construction, first of all because of theimited number of sequential LIMA grafting in our
xperience and also because the different hemody-

amic rules underlying sequential and compositerafts might mislead TTF results. Certainly, largerandomized studies including sequential and compos-te graft construction may help to better define theifferences in the functional properties of pedicled andkeletonized mammary grafts.

eferences

1. Lytle BW, Blackstone EH, Loop FD, et al.. Two internalthoracic artery grafts are better than one. J Thorac Cardio-vasc Surg 1999;117:855–72.

2. Buxton BF, Komeda M, Fuller JA, Gordon I. Bilateral internalthoracic artery grafting may improve outcome of coronaryartery surgery. Risk-adjusted survival. Circulation 1998;98(Suppl 2):1–6.

3. Athanasiou T, Crossman MC, Asimakopoulos G, et al.Should the internal thoracic artery be skeletonized? AnnThorac Surg 2004;77:2238–46.

4. Deja MA, Wos S, Golba KS, et al. Intraoperative and labo-ratory evaluation of skeletonized versus pedicled internalthoracic artery. Ann Thorac Surg 1999;68:2164–8.

5. Wendler O, Tscholl D, Huang Q, Schäfers HJ. Free flowcapacity of skeletonized versus pedicled internal thoracicartery grafts in coronary artery bypass grafts. Eur J Cardio-thorac Surg 1999;15:247–50.

6. Walpoth BH, Monadjer A, Gersbach P, Rogulenko R, Wal-poth BN, Althaus U. Intraoperative internal mammary ar-tery transit-flow measurements: comparative evaluation oftwo surgical pedicle preparation techniques. Eur J Cardio-thorac Surg 1996;10:1064–70.

7. Cohen AJ, Lockman J, Lorberboym M, et al. Assessment ofsternal vascularity with single photon emission tomographyafter harvesting of the internal thoracic artery. J ThoracCardiovasc Surg 1999;118:496–502.

8. Parish MA, Asai T, Grossi EA, et al. The effects of differenttechniques of internal mammary artery harvesting onsternal blood flow. J Thorac Cardiovasc Surg 1992;104:1303–7.

9. Takami Y, Ina H. Effects of skeletonization on intraoperativeflow and anastomosis diameter of the internal thoracicarteries in coronary artery bypass grafting. Ann Thorac Surg2002;73:1441–5.

0. Calafiore AM, Vitolla G, Iaco AL, et al. Bilateral internalmammary artery grafting: midterm results of pedicledversus skeletonized conduits. Ann Thorac Surg 1999;67:1637– 42.

1. Jones EL, Lattouf OM, Weintraub WS. Catastrophic conse-quences of internal mammary artery hypoperfusion. J Tho-rac Cardiovasc Surg 1989;98:902–7.

2. Onorati F, Renzulli A, De Feo M, et al. Does antegrade bloodcardioplegia alone provide adequate myocardial protectionin patients with left main stem disease? J Thorac CardiovascSurg 2003;126:1345–51.

3. D’Ancona G, Karamanoukian HL, Ricci M, Schmid S, Bergs-land J, Salerno TA. Graft revision after transit time flowmeasurement in off-pump coronary artery bypass grafting.Eur J Cardiothorac Surg 2000;17:287–93.

4. Tedoriya T, Kawasuji M, Sakakibara N, Takemura H, Wa-tanabe Y, Hetzer R. Coronary bypass flow during use ofintraaortic balloon pumping and left ventricular assist de-vice. Ann Thorac Surg 1998;66:477–81.

5. Onorati F, Olivito S, Mastroroberto P, et al. Perioperativepatency of coronary artery bypass grafting is not influ-enced by off-pump technique. Ann Thorac Surg 2005;80:2132– 40.

6. Raja SG, Dreyfus GD. Internal thoracic artery: to skeletonizeor not to skeletonize? Ann Thorac Surg 2005;79:1805–11.

7. Ruengsakulrach P, Sinclair R, Komeda M, Raman J, Gordon
I, Buxton B. Comparative histopathology of radial arteryversus internal thoracic artery and risk factors for develop-


1

1

2

2

2

2

2

2

2

2

MA

Thmpsaap

ktbtad


©P

CA

RD

IOV

ASC

ULA

R

ment of intimal hyperplasia and atherosclerosis. Circulation1999;100(Suppl 2):139–44.

8. Ascione R, Underwood MJ, Lloyd CT, Jeremy JY, Bryan AJ,Angelini GD. Clinical and angiographic outcome of differentsurgical strategies of bilateral internal mammary arterygrafting. Ann Thorac Surg 2001;72:959–65.

9. Pearson PJ, Evora PRB, Schaff HV. Bioassay of EDRF frominternal mammary arteries: implications for early and latebypass graft patency. Ann Thorac Surg 1992;54:1078–84.

0. Bar-El Y, Gilboa B, Unger N, Pud D, Eisenberg E. Skeleton-ized versus pedicled internal mammary artery: impact ofsurgical technique on post CABG surgery pain. Eur J Car-diothorac Surg 2005;27:1065–9.

1. Tector AJ, Kress DC, Downey FX, Schmahl TM. Completerevascularization with internal thoracic artery grafts. SeminThorac Cardiovasc Surg 1996;8:29–46.

2. Lorberboym M, Medalion B, Bder O, et al. 99mTc-MDP boneSPECT for the evaluation of sternal ischaemia followinginternal mammary artery dissection. Nucl Med Commun

nnals

In lieu of the above procedure, if you have forgotten your

Cang

aaiAHfjCb

a

2007 by The Society of Thoracic Surgeonsublished by Elsevier Inc


3. Noera G, Pensa P, Lodi R, Lammara M, Biagi B, Guelfi P.Influence of different harvesting techniques on the arterialwall of the internal mammary artery graft: microscopicanalysis. Thorac Cardiovasc Surg 1993;41:16–20.

4. Takami Y, Ina H. Relation of intraoperative flow measure-ment with postoperative quantitative angiographic assess-ment of coronary artery bypass grafting. Ann Thorac Surg2001;72:1270–4.

5. Yoshida H, Wu MH, Kouchi Y, Onuki Y, Shi Q, Sauvage LR.Comparison of the effect of monopolar and and bipolarcauterization on skeletonized, dissected internal thoracicarteries. J Thorac Cardiovasc Surg 1995;110:504–10.

6. Gaudino M, Trani C, Glieca F, et al. Early vasoreactiveprofile of skeletonized versus pedicled internal thoracicartery grafts. J Thorac Cardiovasc Surg 2003;125:638–41.

7. Ueda T, Taniguchi S, Kawata T, Mizuguchi K, Nakajima M,Yoshioka A. Does skeletonization compromise the integrityof internal thoracic artery grafts? Ann Thorac Surg 2003;75:

2002;23:47–52. 1429–33.

ember and Individual Subscriber Access to the Online

he address of the electronic edition of The Annals isttp://ats.ctsnetjournals.org. If you are an STS or STSAember or a non-member personal subscriber to the

rint issue of The Annals, you automatically have aubscription to the online Annals, which entitles you toccess the full-text of all articles. To gain full-textccess, you will need your CTSNet user name andassword.Society members and non-members alike who do not

now their CTSNet user name and password should followhe link “Forgot your user name or password?” that appearselow the boxes where you are asked to enter this informa-

ion when you try to gain full-text access. Your user namend password will be e-mailed to the e-mail address youesignate.

TSNet username and/or password, you can always sendn email to CTSNet via the feedback button from the leftavigation menu on the homepage of the online Annals oro directly to http://ats.ctsnetjournals.org/cgi/feedback.We hope that you will view the online Annals and take

dvantage of the many features available to our subscriberss part of the CTSNet Journals Online. These includenter-journal linking from within the reference sections ofnnals’ articles to over 350 journals available through theighWire Press collection (HighWire provides the platform

or the delivery of the online Annals). There is also cross-ournal advanced searching, eTOC Alerts, Subject Alerts,ite-Track, and much more. A listing of these features cane found at http://ats.ctsnetjournals.org/help/features.dtl.We encourage you to visit the online Annals at http://

ts.ctsnetjournals.org and explore.

Ann Thorac Surg 2007;84:1919 • 0003-4975/07/$32.00



DOI: 10.1016/j.athoracsur.2007.06.024 2007;84:1912-1919 Ann Thorac Surg

Mastroroberto and Attilio Renzulli Francesco Onorati, Antonio Esposito, Francesco Pezzo, Antonio di Virgilio, Pasquale

Grafts HarvestingHospital Outcome Analysis After Different Techniques of Left Internal Mammary

& ServicesUpdated Information

http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912including high-resolution figures, can be found at:

References http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912#BIBL

This article cites 27 articles, 23 of which you can access for free at:

Citations

shttp://ats.ctsnetjournals.org/cgi/content/full/84/6/1912#otherarticleThis article has been cited by 3 HighWire-hosted articles:

Subspecialty Collections

http://ats.ctsnetjournals.org/cgi/collection/coronary_disease Coronary disease

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

Permissions & Licensing

[email protected]: orhttp://www.us.elsevierhealth.com/Licensing/permissions.jsp

in its entirety should be submitted to: Requests about reproducing this article in parts (figures, tables) or

Reprints [email protected]

For information about ordering reprints, please email:

by on May 31, 2013 ats.ctsnetjournals.orgDownloaded from

http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912

http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912#BIBL

http://ats.ctsnetjournals.org/cgi/content/full/84/6/1912#otherarticles

http://ats.ctsnetjournals.org/cgi/collection/coronary_disease

http://www.us.elsevierhealth.com/Licensing/permissions.jsp


Hospital Outcome Analysis After Different Techniques of Left Internal Mammary Grafts Harvesting

Documents

Transcript of Hospital Outcome Analysis After Different Techniques of Left Internal Mammary Grafts Harvesting