Our Experiences for Off-Pump Coronary ArteryBypass Grafting to the Circumflex SystemTomoaki Suzuki, MD, Manabu Okabe, MD, PhD, Fuyuhiko Yasuda, MD, PhD,Yoichiro Miyake, MD, Mitsuteru Handa, MD, and Takazumi Nakamura, MD, PhDKochi Municipal Hospital and Izumino Hospital, Kochi, Japan

Background. Complete revascularization has been dif-ficult in off-pump coronary artery bypass grafting(OPCAB). Hemodynamic deterioration often preventsaccess to the circumflex territory. This study presentsinstrumentation for accessing the circumflex territory,and our clinical experience.

Methods. From August 1999 through December 2002,140 patients underwent OPCAB via sternotomy in ourinstitution. The 114 requiring reconstruction of the cir-cumflex artery are the subjects of this study. There wereno exclusion criteria. A series of techniques and instru-ments were developed to provide access to the circumflexarea while hemodynamic stability was preserved, includ-ing the left pericardial traction technique, compression ofthe right pericardium, a right sternal retractor, and a typeof shunt tube.

Results. Patients received an average of 3.2 grafts(range, 2 to 6). Complete revascularization was achieved

in 95% of the cases. Complications included respiratoryinsufficiency (0.8%), renal dysfunction (7%), and sternalwound infection (0.8%). Blood transfusions were re-quired in 10 patients (8%). No patient suffered perioper-ative myocardial infarction or stroke. No operation wasconverted to cardiopulmonary bypass. There was nooperative death. Predischarge angiography demonstrateda 99% patency rate.

Conclusions. With our techniques and instruments,off-pump coronary revascularizaion of the circumflexarea may be performed safely to achieve complete revas-cularization. Early clinical results are excellent, but long-term longitudinal follow-up is required to assess thefuture effectiveness of OPCAB procedure with our tech-niques.

(Ann Thorac Surg 2003;76:2013–6)© 2003 by The Society of Thoracic Surgeons

Off-pump coronary artery bypass grafting (OPCAB)has been gaining acceptance among cardiovascular

surgeons because it eliminates problems associated withcardiopulmonary bypass (CPB) [1–7). However, completerevascularization cannot be achieved in every patient byOPCAB as hemodynamic deterioration can occur duringpositioning and stabilization of the heart and snaring ofthe coronary artery. Attempts at revascularization in thecircumflex coronary artery territory often results in he-modynamic compromise [8–11). Thus, accessing the cir-cumflex (CX) system, without compromising hemody-namic stability, is a crucial step towards completerevascularization by OPCAB. We have developed tech-niques and instrumentation that permit complete revas-cularization by OPCAB in virtually all patients. In thisarticle, we describe our methods and instruments forexposing the circumflex territory and present the resultsof an uncontrolled series of patients who underwentcircumflex reconstruction by OPCAB.

Material and Methods

From August 1999 through December 2002, 140 patientsunderwent OPCAB grafting via sternotomy in our insti-

tution. The 114 (81.4%) patients who required revascular-ization of the circumflex artery are the subjects of thisstudy. Minimally invasive coronary artery bypass graft-ing (CABG) cases and emergent cases were not includedin this material. There were 81 men and 33 women, aged45 to 82 years (mean, 67.1 � 8.7 years) (Table1). Weperformed OPCAB for all elective CABG, including pa-tients with unstable angina: there were no exclusioncriteria.

Anesthetic ManagementPremedication consisted of intramuscular midazolam(0.04 mg/kg). A standard anesthetic technique was usedfor all patients. The induction of anesthesia was achievedwith fentanyl citrate (10 to12 �g/kg), thiopental (3 to 5mg/kg), and vecronium bromide (0.1 mg/kg). Anesthesiawas maintained with fentanyl, propofol (2 to 3 mg/kg),and low concentrations of sevoflene as necessary. Anti-coagulation was achieved with heparin (1 mg/kg) afterthe internal thoracic artery (ITA) graft was harvested. Theactivated clotting time was maintained at � 250 seconds.Heparin was reversed with protamine after completion ofthe anastomoses.

Standard intraoperative monitoring techniques wereused. Pulmonary artery floatation catheters were usedroutinely and they provided continuous evaluation ofcardiac output. Transesophageal echocardiography wasnot used routinely. Red blood cell saving devices were

Accepted for publication June 25, 2003.

Address reprint requests to Dr Suzuki, Department of CardiovascularSurgery, Kochi Municipal Hospital, 1-7-45 Marunouti, Kochi 780-0850,Japan; e-mail: suzukit@bronze.ocn.ne.jp.

© 2003 by The Society of Thoracic Surgeons 0003-4975/03/$30.00Published by Elsevier Inc doi:10.1016/S0003-4975(03)01326-2

CA

RD

IOV

AS

CU

LA

R

used. A standby without a primed CPB circuit was held inabeyance for all cases.

Operative TechniqueAll procedures were performed through a median ster-notomy. The conduits (one or both ITAs, the radial artery,right gastroepiploic artery (RGEA), and saphenous vein)were harvested and skeletonized. Pericardiotomy is es-sential to preserve hemodynamics stability while expos-ing the circumflex territory. Following is our technique:After midline pericardiotomy, the left side of the pericar-dium is peeled away from the pleura and two transverseleft pericardiotomies are created. First, the left side of thepericardium is incised from the apex to the back of thepericardial sac to the anterior of the phrenic nerve, inorder that it not be injured. The other pericardiotomy isat the level of the left atrial appendage. The right side ofthe pericardium is incised longitudinally around thesuperior vena cava (SVC). This maneuver prevents theSVC from being compressed by the pericardium andprevents disruption of venous return when the heart islifted. Finally, a vertical pericardiotomy is created andcarried out toward the inferior vena cava (IVC) at thediaphragmatic surface. The incised left side of the peri-cardium is pulled up and sutured to the skin, andretracted by sternal retractor. Our sternal retractor is ourown design, and elevates the right sternum to create afree space beneath it. Furthermore, the right side of thepericardium is compressed down between the right up-per and lower pulmonary veins by a pericardial retractorthat we designed. The pericardial retractor is fixed by thesurgical arm that is fixed on the operating table (Figs 1, 2,and 3). As a result, the cardiac base is rotated right sidedown and left side up and the apex moves into the freespace beneath the right sternum. Placing the patient in

the right decubitus Trendelenburg position improvesexposure of the circumflex area. We also use the deeppericardial sutures in conjunction with our techniques incases with cardiomegaly or a too deep pericardial sac.

We use a suction-type mechanical stabilizer (Octopus3, Medtronic, Minneapolis, MN) to immobilize the targetcoronary artery. Soft silicon loops are used to perform thecoronary occlusion test and control coronary blood flow.Ischemic preconditioning was performed only if hemo-dynamic deterioration occurred during test occlusion ofthe coronary artery. Intracoronary shunt tubes were usedroutinely in all target vessels. These shunts create abloodless surgical field, assure distal coronary bloodflow, and provide hemodynamic stability during anasto-mosis. Our shunt tube (JMS Bypass Tube, JMS, Hiro-shima, Japan) was designed with a very flexible shaft thatmakes it easy to insert and remove without intimalinjury. The JMS bypass tubes are available from 1.25 mmto 2.5 mm in 0.25 mm increments. The distal anastomosis

Table 1. Patient Characteristics

Total number of patients 114Gender

Male 81Female 33

Age (years) 67.1 � 8.7NYHA class

I 0II 38 (33.3%)III 61 (53.5%)IV 15 (13.1%)Left main 48 (42.1%)EF (.40) 13 (11.4%)

Previous MI 63 (55.2%)Urgent 16 (14.0%)Diabetes mellitus 40 (35.1%)Chronic renal failure 10 (8.8%)Unstable angina 20 (17.5%)Prior CVA 26 (22.8%)Peripheral vascular disease 8 (7.0%)

NYHA � New York Heart Association; MI � myocardial infarction;CVA � cerebrovascular accident.

Fig 1. (A) Our sternal retractor and right pericardial retractor. Thepericardial retractor is fixed with the surgical arm that is fixed onthe operating table. (B) The retractor for compressing the right sideof the pericardium.

Fig 2. The surgical view with our technique. The cardiac apexmoves into the free space beneath the right sternum without beingcompressing.

2014 SUZUKI ET AL Ann Thorac SurgOUR OPCAB TECHNIQUES FOR THE CX 2003;76:2013–6

CA

RD

IOV

AS

CU

LA

R

was performed using 8-0 polypropylene by a standardtechnique.

Results

An average of 3.20 � 0.31 grafts per patient (range, 2 to 6grafts per patient) were performed. The total number ofdistal anastomosis was 365, and sequential anastomosiswas 42 (two anastomoses in 39 patients and three in 3patients). The left internal thoracic artery (LITA) wasused in 111 (97%) patients, the right internal thoracicartery in 28 (25%, in situ graft in 23, free graft in 5), radialartery in 72 (63%), RGEA in 41 (36%), and the saphenousvein in 48 (42.5%). The grafted vessel was the left anteriordescending artery (LAD) in 97% of patients, a diagonalbranch in 34%, the oblique marginal (OM) in 65%, theposterior descending artery (PDA) in 78%, the postero-lateral branch in 69%, and the right coronary artery(RCA) in 8%. Complete revascularization was achieved in108 (95%) patients. Seven patients experienced hemody-namic deterioration, which occurred during snaring ofthe coronary artery (LAD, 3; CX, 2; PDA, 2). One patientdeveloped frequent ventricular premature contractionduring test occlusion of the LAD. The other patientsdeveloped hemodynamic deterioration as decrease ofsystemic pressure and cardiac output and as increase ofpulmonary pressure and right atrial pressure. However,hemodynamic stability was achieved in all cases, afterischemic preconditioning and prompt insertion of thecoronary shunt tube. No patient was converted fromoff-pump to on-pump. The length of stay in the intensivecare unit was 1 day in 112 patients and 2 days in 2patients. Only 10 patients (8.7%) needed bloodtransfusions.

Morbidity and MortalityNo patient suffered low cardiac output or perioperativemyocardial infarction. A sternal wound infection oc-curred in one patient (0.8%). Respiratory insufficiencynecessitating prolonged (� 24 hours) respiratory supportoccurred in one patient (0.8%), new renal dysfunction(serum creatinine � 2.0 mg/dL) that did not requirehemodialysis occurred in 3 patients (2.5%), and no pa-tient had a stroke. There was no operative or hospitaldeath. Postoperative angiography was performed in 113(99%) patients before discharge (1 to 2 weeks after theoperation). Of 376 anastomoses studied, 370 were patent.

Graft occlusion was documented in 2 cases; in 1 case,the SVG was occluded at the anastomotic site to the RCAand, in 1 case, the sequential RGEA graft was occluded atthe side-to-end anastomotic site to PDA. Graft stenosisoccurred in 4 cases. In 2 cases, SVG to PDA and to OMshowed stenosis of the anastomotic site, in the other 2cases, graft stenosis occurred at the side-to-end anasto-motic site of the sequential SVGs to OM–PL and tomidRCA–PDA.

Comment

As interest in OPCAB surgery has grown the technologysupporting it has matured. Refinements in the mechan-ical stabilizer, especially the suction type (Octopus seriesMedtronic, Minneapolis, MN) provide a very stable sur-gical field with minimum hemodynamic change [12]. Thedeep pericardial suture technique is very effective forproviding to the lateral and posterior walls of the heart.Lima [13] first described a method employing a series ofsutures to expose the circumflex vessels on the beatingheart. With greater experience, the deep pericardial su-ture technique was modified and is now performeduniversally [14]. Bergsland and colleagues [15] modifiedthe Lima sutures into a “single suture” technique anddocumented the simplicity and effectiveness of thismethod. Ricci and colleagues [16] reported a single su-ture technique with vaginal tape. This is very simple anduseful to expose all areas of the heart without the otherinstrument. However, this maneuver sometime com-presses the heart by the vaginal tape. The point of ourtechnique is to expose the circumflex area without com-pressing the heart. Our technique, making a free spacebeneath the right sternum by elevating the right sternumand compressing the right side of pericardium, is veryeffective for comfortable contraction for the right ventri-cle without compressing.

We have developed several devices and techniques toimprove the quality of the surgical field when accessingthe circumflex territory. Rotation of the cardiac base,right side down and left side up, creates wide and stableexposure of the circumflex territory. The incised left sideof the pericardium traction technique we presented ef-fectively exposes the circumflex area in conjunction withthe deep pericardial suture technique, even in case ofcardiomegaly. Compression of the right side of the peri-cardium by the pericardial retractor enhances rotation of

Fig 3. (1) The left side of the pericardium is peeled and separatedaway from the pleura. (2) The left side of the pericardium is pulledup and sutured to the skin. It makes the left side cardiac base rotateup. (3) The right sternum is lifted up by the sternal retractor, whichwe originally designed, and the result in the free space is made be-neath the sternum. (4) The right side of the pericardium is com-pressed down between the upper and lower pulmonary veins by thepericardium retractor, which we originally designed. It makes theright-side cardiac base rotate down and the apex move into the freespace beneath the right sternum.

2015Ann Thorac Surg SUZUKI ET AL2003;76:2013–6 OUR OPCAB TECHNIQUES FOR THE CX

CA

RD

IOV

AS

CU

LA

R

the cardiac base. Finally, the right sternal retractor cre-ates a free space beneath the right sternum that allowsthe heart to be displaced without compromisingcontraction.

The coronary shunt tube prevents transient myocardialdysfunction during OPCAB [17]. It maintains coronaryperfusion, creates a bloodless surgical field, and prevents“back walling” while creating the anastomosis [18]. He-modynamic deterioration can occur during snaring of thecoronary artery, with either the LAD or the circumflexartery. Therefore, we routinely use a coronary shunt tubein all target vessels. Occasionally it can be very difficultand risky to insert the shunt tube into a coronary arterythat is calcified, narrow, or serpentine. We developed acoronary shunt tube that is quite flexible, facilitatinginsertion and removal without coronary injury. We havenever been unable to insert or remove a shunt, nor hasany coronary artery been injured through shunt use.

The important factors in achieving stable hemodynam-ics during circumflex artery anastomosis are as follows:(1) the cardiac base should be rotated right side downand left side up; (2) a free space should be createdbeneath the right sternum to allow displacement of theheart; (3) systemic and pulmonary venous return shouldnot be impeded; and (4) coronary blood flow should bemaintained using a coronary shunt tube.

Routine complete revascularization in all patients byOPCAB surgery is the goal of surgery for myocardialischemia. The technique presented here allows surgeonseasy access to the circumflex territory.

References

1. Atsushi A, Hitoshi H, Akihito T, et al. Off-pump coronaryartery bypass. Jpn J Thorac Cardiovasc Surg 2001;49:67–78.

2. Arom KV, Flavin TF, Emery RW, et al. Safety and efficacy ofoff-pump coronary artery bypass grafting. Ann Thorac Surg2000;69:704–710.

3. Cleveland JC Jr, Shroyer ALW, Chen AY, et al. Off-pumpcoronary artery bypass grafting decreases risk-adjustedmortality and morbidity. Ann Thorac Surg 2001;72:1282–89.

4. Chamberlain MH, Ascione R, Reeves BC, et al. Evaluation ofthe effectiveness of off-pump coronary artery bypass grafting

in high-risk patients: An observational study. Ann ThoracSurg 2002;73:1866–73.

5. Kshettry VR, Flavin TF, Emwry RW, et al. Does multivessel,off-pump coronary artery bypass reduce postoperative mor-bidity? Ann Thorac Surg 2000;69:1725–31.

6. Beti HS, Suri A, Kalkat MS, et al. Global myocardial revas-cularization without cardiopulmonary bypass using innova-tive techniques for myocardial stabilization and perfusion.Ann Thorac Surg 2000;69:156–64.

7. Calafiore AM, Di Mauro M, Contini M, et al. Myocardialrevascularization with and without cardiopulmonary bypassin multivessel disease: impact of the strategy on earlyoutcome. Ann Thorac Surg 2001;72:456–63.

8. Cartier R, Blain R. Off-pump revascularization of the circum-flex artery: technical aspect and short-term results. AnnThorac Surg 1999;68:94–9.

9. Mathison M, Edgerton JR, Horswell JL, et al. Analysis ofhemodynamic changes during beating heart surgical proce-dure. Ann Thorac Surg 2000;70:1355–61.

10. Vassiliades TA, Nielsen JL, Lonquist JL. Hemodynamic col-lapse during off-pump coronary artery bypass grafting. AnnThorac Surg 2002;73:1874–9.

11. Stamou SC, Bafi AS, Boyce SW, et al. Coronary revascular-ization of the circumflex system: different approaches andlong-term outcome. Ann Thorac Surg 2000;70:1371–7.

12. Jansen EWL, Borst C, Lahpor JR, et al. Coronary arterybypass grafting without cardiopulmonary bypass using theOctopus method: results in the first one hundred patients.J Thorac Cardiovasc Surg 1998;116:60–7.

13. Lima R. Revascularization a o da arteria circunflexa semauxilio da CEC. In: XII encontro dos discipulos do dr. EJZerbini, Curitiba, 1995. Sessa ode videos. Curitiba, Parana,Sociedade dos discipulos do dr. EJ Zerbini Outtubro de 1995,6.b.

14. Baumgartner FJ, Gheissari A, Capouya ER, et al. Technicalaspects of total revascularization in off-pump coronary by-pass via sternotomy. Ann Thorac Surg 1999;67:1653–8.

15. Bergsland J, Karamanoukian HL, Soltoski PR, et al. “Singlesuture” for circumflex exposure in off-pump coronary arterybypass grafting. Ann Thorac Surg 1999;68:1428–30.

16. Ricci M, Karamanoukian HL, D’Ancona G, et al. Exposureand mechanical stabilization in off-pump coronary arterybypass grafting via sternotomy. Ann Thorac Surg 2000;70:1736–40.

17. Yeatman M, Caput M, Narayan P, et al. Intracoronary shuntsreduce transient intraoperative myocardial dysfunction dur-ing off-pump coronary operations. Ann Thorac Surg 2002;73:1411–7.

18. Rvetti LA, Gandra SMA. Initial experience using intralumi-nal shunt during revascularization of the beating heart. AnnThorac Surg 1997;63:1742–7.

2016 SUZUKI ET AL Ann Thorac SurgOUR OPCAB TECHNIQUES FOR THE CX 2003;76:2013–6

CA

RD

IOV

AS

CU

LA

R