DOI: 10.1016/j.ejcts.2010.02.029 2010;38:380-386 Eur J Cardiothorac Surg

Cirillo, Federico Brunelli, Eugenio Quaini and Giovanni Troise Antonio Messina, Emmanuel Villa, Zean Mhagna, Margherita Dalla Tomba, Marco

trainee surgeonsMedium-term results of systematic off-pump coronary surgery performed by

This information is current as of September 8, 2010

http://ejcts.ctsnetjournals.org/cgi/content/full/38/3/380located on the World Wide Web at:

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

ISSN: 1010-7940. European Association for Cardio-Thoracic Surgery. Published by Elsevier. All rights reserved. Printfor Cardio-thoracic Surgery and the European Society of Thoracic Surgeons. Copyright © 2010 by The European Journal of Cardio-thoracic Surgery is the official Journal of the European Association

by Emmanuel Villa on September 8, 2010 ejcts.ctsnetjournals.orgDownloaded from

www.elsevier.com/locate/ejctsEuropean Journal of Cardio-thoracic Surgery 38 (2010) 380—386

Medium-term results of systematic off-pump coronary surgeryperformed by trainee surgeons§

Antonio Messina a,*, Emmanuel Villa a, Zean Mhagna a, Margherita Dalla Tomba a,Marco Cirillo a, Federico Brunelli a, Eugenio Quaini b, Giovanni Troise a

aCardiovascular Department, Cardiac Surgery Operating Unit, Poliambulanza Foundation Hospital, Via Bissolati 57, 25124 Brescia, ItalybCardiac Surgery Operating Unit, Ospedali Riuniti, Bergamo, Italy

Received 30 September 2009; received in revised form 8 February 2010; accepted 16 February 2010; Available online 25 March 2010

Abstract

Objective: Our unit has used off-pump coronary artery bypass (OPCAB) surgery since 1998, and has consequently developed teaching methodsfor surgical trainees. This study aimed to compare the medium-term results of OPCAB performed by experts or supervised trainees.Methods:Weretrospectively analysed the data relating to 1333 OPCAB operations performed between January 1998 and January 2006 (mean patient age:65.3 � 13; M/F ratio: 2.9), and compared the medium-term outcomes of the 977 (73.3%) carried out by three expert surgeons (group A) with theremaining 356 (26.7%) carried out by four supervised trainees (group B). Results: There were no preoperative differences in patient age, gender,angina class, operative priority, extent of coronary artery disease, the presence of a recent myocardial infarction or left main stenosis orEuropean System for Cardiac Operative Risk Evaluation (EuroSCORE) between the two groups. Thirty-day mortality was 1% in group A and 0.6% ingroup B (p = 0.43), and 4-year actuarial survival, respectively, 97.4 � 1.1% and 94.3 � 4.1% (p = 0.41); the freedom from new re-vascularisationrates in the two groups were, respectively, 96 � 0.7% and 95.3 � 1.4% ( p = 0.3). Conclusions: The results of this study reflect our unit’s longexperience of OPCAB surgery and that its successful re-engineering towards the systematic use of OPCABwas feasible. They also show that, in thiscontext, teaching OPCAB surgery is safe in a non-selected cohort of patients, and that the medium-term outcomes of the patients operated on bytrainee or expert surgeons are similar.# 2010 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.

Keywords: Cardiac surgical procedures; Coronary disease; Off-pump coronary artery bypass; Training

1. Introduction

Off-pump coronary artery bypass (OPCAB) surgery hasbecome increasingly more widespread, over the last 20years, and is now used by many surgeons [1].

However, despite the considerable evidence that it leadsto similar graft patency, reduces the risk of postoperativemorbidity, shortens intensive care unit and hospital stays anduses fewer resources than coronary surgery with a cardio-pulmonary bypass (CPB) [2—7], <25% of all coronary arterybypass grafting (CABG) operations are carried out using theOPCAB technique [8—10].

OPCAB is undoubtedly technically demanding as surgeonsare faced with a beating heart rather than a still andbloodless surgical field. A lack of experience may not only put

§ Presented at the 23rd Annual Meeting of the European Association forCardio-thoracic Surgery, Vienna, Austria, October 18—21, 2009.* Corresponding author. Tel.: +39 030 3518534; fax: +39 030 3515244.E-mail addresses: messina-antonio@poliambulanza.it,

anto.mex@tiscali.it (A. Messina).

1010-7940/$ — see front matter # 2010 European Association for Cardio-Thoracic Sdoi:10.1016/j.ejcts.2010.02.029

byejcts.ctsnetjournals.orgDownloaded from

off even the most experienced conventional CABG surgeon,but also dissuade experienced OPCAB surgeons from teachingthe technique.

It is always difficult to teach new operative techniques,not least because a trainee’s learning curve should notdisadvantage any patient; furthermore, only an institution’scommitment to OPCAB can lead to its systematic use andallow the development of the standardised surgical,anaesthesiology and nursing procedures necessary toimprove trainee tutoring without compromising outcomes[11—13]. Systematic usemeans considering all of the patientsscheduled for isolated coronary bypass surgery, except forthose with ST-elevation myocardial infarction (MI) andcardiomegaly with ventricular dysfunction, as candidatesfor OPCAB.

A few studies have shown that teaching the OPCABtechnique to young surgeons is possible and safe, but there isstill a lack of longer-term outcome data [14—18].

This study aimed to compare the hospital and medium-term outcomes of OPCAB surgery performed by seniorsurgeons and trainees.

urgery. Published by Elsevier B.V. All rights reserved.

Emmanuel Villa on September 8, 2010

A. Messina et al. / European Journal of Cardio-thoracic Surgery 38 (2010) 380—386 381

2. Materials and methods

2.1. Patient population

As OPCAB surgery has been used in our unit since 1998 andit now accounts for up to 95% of all operations for isolatedcoronary artery disease, we retrospectively analysed the1333 consecutive patients who underwent isolated OPCABthrough a midline sternotomy between January 1998 andJanuary 2006; those who underwent limited access opera-tions were excluded.

Nine hundred and seventy-seven of the procedures(73.3%) were carried out by three expert surgeons (groupA) and 356 (26.7%) by four supervised trainees (group B).

All of the data had been prospectively entered in our unit’soff-pump database and an audit has shown that there are<2% missing data points.

Fig. 1 shows the percentage of OPCAB operations inrelation to all of the coronary artery surgical proceduresperformed during the study period.

2.2. Anaesthetic and surgical technique

Anaesthesia was induced by means of a 2-min bolus ofremifentanil 1.5 mg kg�1 min�1 and one bolus of midazolam0.2 mg kg�1 and neuromuscular blockade by means of a 0.15-mg kg�1 bolus of cisatracurium. Maintenance was obtainedby infusing propofol 3 mg kg�1 h�1, ramifentanil 0.05—0.5 mg kg�1 min�1 and cisatracurium 1.5 mg kg�1 min�1.

Anticoagulation was achieved by administering heparin300 IU kg�1 before sectioning the internal mammary arteries(IMAs), and additional doses to maintain an activated clottingtime of >300 s.

The proximal right coronary and left anterior descending(LAD) arteries were directly accessed using very little heartmanipulation. In order to gain access to the posteriordescending and obtuse marginal arteries (PDA and OM), theheart was positioned vertically by applying two or threetraction stitches to the posterior pericardium, between theleft inferior pulmonary vein and the inferior vena cava.

A longwet stripofgauzewas securedtothedeepest tractionstitch by means of a tourniquet, and the lateral and posteriorvessels were exposed by handling the two ends of the gauze,tilting the table head-down and turning it towards the surgeon.

The coronary vessels were stabilised by means of areusable stainless steel compressive stabiliser developed at

Fig. 1. The percentage of OPCAB operations in relation to all of the coronaryartery surgical procedures performed during the study period.

by ejcts.ctsnetjournals.orgDownloaded from

our institution, which can be fitted to the supports of theCosgrove self-retaining mitral retractor (Kapp Surgical Inc.,Cleveland, OH, USA).

A filtered blower/mister was used to perform theanastomosis under perfect vision, avoiding any trauma tothe arterial walls (ClearView, Medtronic Inc. Minneapolis,MN, USA), and a soft and flexible intracoronary shunt(Anastaflow, Edwards Lifesciences LLC, Irvine, CA, USA)was used to avoid transient ischaemia and reduce bleedingfrom the anastomotic site [19].

In the case of redo surgery, adhesiolysis and re-vascularisation were carried out progressively, starting fromthe anterior branches; similar sequential re-vascularisationwas used for patients with critical left main-stem stenosis ordepressed left ventricular function, in whom carefulmanipulation of the heart is required to reach the posteriorvessels.

In cases of unstable angina unresponsive to maximalintravenous therapy, or in the presence of severe leftventricular dysfunction, the off-pump operation was per-formed after the prophylactic insertion of an intra-aorticballoon pump (IABP).

We routinely constructed an arterial conduit using theright skeletonised IMA as a free graft anastomosed termino-laterally to the left skeletonised IMA (Y-graft) in order to re-vascularise the left coronary system. The right coronarysystem was re-vascularised by means of a saphenous veingraft (usually) or radial artery (rarely) anastomosed to theproximal aorta.

In a few cases, in order to avoid aortic manipulation, theproximal stump of the right IMA was used as an intrathoracicarterial source of flow for the saphenous vein graft [20].

2.3. Definitions

A trainee case was defined as a procedure performed from‘skin to skin’ by a trainee, with an expert surgeon acting asthe first assistant or supervising while another person directlyassisted the resident.

Pre-existing renal dysfunction was defined as a preopera-tive creatinine level >1.5 mg dl�1.

Carotid artery disease was defined as the presence of a>50% stenosis as assessed by means of vascular ultrasono-graphy.

Thirty-day mortality was defined as any death occurringwithin 30 days of the operation (inside or outside thehospital), or any in-hospital death at any time if the patienthad remained in hospital until his or her death.

Perioperative cerebrovascular accident (CVA) was definedas a global or focal neurological deficit lasting less (transientischaemic attack) or more (stroke) than 24 h, that wasapparent after the patient emerged from anaesthesia,diagnosed by a neurologist and confirmed by a braincomputed tomography (CT) scan [21,22].

2.4. Training method

All of the three expert surgeons performed >90% of theirre-vascularisation surgical procedures using OPCAB. All of thetrainees were introduced to OPCAB irrespective of theirsurgical coronary re-vascularisation background, although

Emmanuel Villa on September 8, 2010

A. Messina et al. / European Journal of Cardio-thoracic Surgery 38 (2010) 380—386382

Table 1Patients’ demographic and preoperative characteristicsa.

Variables Expert surgeons (n = 977) Trainee surgeons (n = 356) p-valueb

n % n %

Age (years) 65.3 � 13.2 65.2 � 15.3 0.6Gender (male/female) 769/173 277/63 0.9COPD 106 10.8 51 14.3 0.08Pre-existing renal dysfunction 57 5.8 19 5.3 0.3Carotid artery disease 190 19.4 53 14.8 0.048DM 243 24.8 97 27.2 0.3

PriorityElective 740 75.7 277 77.8 0.6Urgent 197 20.2 63 17.7 0.3Emergent 40 4.1 16 4.5 0.3

Left main stenosis 309 31.6 98 27.5 0.1

Extent of coronary artery diseaseSingle vessel 72 7.4 29 8.1 0.3Double vessel 278 28.4 105 29.5 0.7Triple vessel 627 64.1 222 62.4 0.5

Ejection fractionPoor (<30%) 22 2.3 10 2.8 0.3Moderate (30—50%) 297 30.4 111 31.2 0.3Good (>50%) 658 67.3 235 66 0.3

Redo surgery 23 2.3 3 0.84 0.03Recent MI (<90 days) 467 47.8 178 50 0.4Unstable angina 194 19.9 66 18.5 0.6Preoperative IABP 22 2.3 5 1.4 0.3Median additive EuroSCORE 2 (1—3) 2 (1—2) 0.7

COPD: chronic obstructive pulmonary disease; IABP: intra-aortic balloon pump; MI: myocardial infarction; DM: diabetes mellitus; and EuroSCORE: European Systemfor Cardiac Operative Risk Evaluation.

a Values shown as frequencies and percentages, means � SD or medians (25th and 75th centiles).b x2 test and Fisher’s exact test or the Mann—Whitney U-test.

complete competence in establishing and managing a CPBwas required.

The trainees performed all of the technical aspects of theoperation, such as the exposure and stabilisation of vessels,graft construction and anastomoses. They were also involvedin constant intra-operative dialogue with the anaesthetist inorder to determine the progress of the operation and assesshaemodynamic status, changes in heart rate, variations inmyocardial kinesis during heart handling and coronaryshunting and anastomoses.

2.5. Follow-up

The cross-sectional follow-up was carried out betweenDecember 2008 and March 2009 by reviewing the hospitalrecords, making contact with the patients or their physiciansand assessing cardiac events: that is, recurrent angina, MI,cardiac catheterisation, percutaneous transluminal coronaryangioplasty (PTCA) and repeated CABG.

The primary clinical end points were survival and freedomfrom new re-vascularisation (PTCA or repeated CABG).

The additive European System for Cardiac Operative RiskEvaluation (EuroSCORE) was used to stratify the patients onthe basis of the predicted risk of operative mortality [23].

2.6. Statistical analysis

The continuous data were assessed for normality, and arepresented as mean values � standard deviation (SD), ormedian values and interquartile ranges: the normally

byejcts.ctsnetjournals.orgDownloaded from

distributed variables were tested using an independentsample t-test, and the skewed data by means of the non-parametric Mann—Whitney U-test.

The categorical data are presented as frequencies andpercentages, and were analysed using the x2 test and Fisher’sexact test. Follow-up survival and freedom from new re-vascularisation were determined using the Kaplan—Meiermethod and the log-rank test. p-values of <0.05 wereconsidered statistically significant.

All of the data were analysed using the Statistical Packagefor Social Sciences (SPSS) statistical software version 11.0(SPSS, Chicago, IL, USA) and MedCalc version 9.3.9.0(MedCalc Software, Mariakerke, Belgium).

3. Results

Table 1 shows the demographic and preoperativecharacteristics of the two cohorts. The two groups werehomogeneous in terms of patient age (65.3 vs 65.2 years),gender, chronic obstructive pulmonary disease, chronicrenal failure and diabetes mellitus, but differed in terms ofepi-aortic vessel disease ( p = 0.04) and redo surgery( p = 0.03).

The ejection fraction values obtained by means ofechocardiography were divided into three classes on thebasis of the EuroSCORE definition (poor, moderate and good)in order to highlight any differences: there was no significantdifference between the expert and trainee surgeon groups inany of the classes.

Emmanuel Villa on September 8, 2010

A. Messina et al. / European Journal of Cardio-thoracic Surgery 38 (2010) 380—386 383

Table 2Intra-operative dataa.

Variables Expert surgeons (n = 977) Trainee surgeons (n = 356) p-valueb

n % n %

Grafts/patient 2.46 � 0.82 2.37 � 0.85 0.1OPCAB time (min) 221 � 54 220 � 58 0.3Conversion 30 3 15 4 0.3LIMA 924 98.1 329 96.8 0.1RIMA 677 71.9 229 67.4 0.1Radial artery 32 3.4 5 1.5 0.1Saphenous vein 405 42.9 158 46.5 0.2Sequential graft 213 22.6 74 21.8 0.8

Anterior system 977 100 356 100LAD artery 973 99.6 353 99.2 0.3Diagonal artery 192 20 60 17 0.3

Lateral system 691 73.4 232 68 0.06Obtuse marginal 583 62 183 53.8 0.01PLA from circumflex 70 7.4 40 11.7 0.02

Posterior system 386 41 140 41.2 0.9PDA 233 24 95 27 0.1PLA from RCA 160 17 46 13.5 0.1

LAD: left anterior descending coronary artery; PDA: posterior descending coronary artery; PLA: postero-lateral coronary artery; RCA: right coronary artery; LIMA: leftinternal mammary artery; and RIMA: right internal mammary artery.

a Values shown as frequencies and percentages, means � SD or medians (25th and 75th centiles).b x2 test and Fisher’s exact test or the Mann—Whitney U-test.

The two groups were also similar in terms of surgicalpriorities (elective, urgent or emergent) and the extent ofthe coronary artery disease, but the expert surgeon groupincluded more patients with carotid artery disease( p = 0.048).

The median EuroSCORE-predicted risk of mortality was 2in both the groups ( p = 0.7).

Table 2 shows the intra-operative results. The meannumber of grafts per patient was 2.4 � 0.8 in the expertsurgeon group and 2.3 � 0.8 in the trainee surgeon group; thedifference is not statistically significant ( p = 0.1).

All of the patients received a graft on the anterior system;grafts on the lateral system were performed in 73.4% of thepatients in the expert surgeon group and 68% of those in thetrainee surgeon group ( p = 0.06); grafts on the posterior

Table 3Postoperative dataa.

Variables Expert surgeons (n = 977)

n %

30-day deaths 10 1Perioperative MI 23 2.4Postoperative IABP 11 1Atrial fibrillation 227 23Reopening for bleeding 11 1Haemofiltration/dialysis 5 1Perioperative CVA 4 0.4CK (UI l�1) 651 (266; 711)CK-Mb (ng ml�1) 19 (13; 28)Total bleeding (cc) 530 (410; 680)Intubation time (h) 12.3 � 26ICU stay (days) 1.4 � 2.5Hospital stay (days) 4.7 � 3.4

MI: myocardial infarction; ARF: acute renal failure; CVA: cerebrovascular accident;a Values are shown as frequencies and percentages, means � SD or medians (25tb x2 test and Fisher’s exact test or the Mann—Whitney U-test.

by ejcts.ctsnetjournals.orgDownloaded from

systemwere performed in, respectively, 41% and 41.2% of thepatients ( p = 0.9).

The use of a CPB was necessary in 30 patients in the expertsurgeon group (3%) and 15 in the trainee surgeon group (4%)( p = 0.3).

Table 3 shows that there was no significant between-groupdifference in terms of perioperative morbidity or mortality.There were ten 30-day deaths in the expert surgeon group(1%) and two in the trainee group (0.6%) ( p = 0.4), for a totalhospital mortality rate of<1% (12/1333). The incidence of MI(2.4% vs 1%; p = 0.4), acute renal failure (1% vs 1%; p = 0.3)and neurological complications (0.4% vs 0.3%; p = 0.4) wassimilar in the two groups.

The follow-up was 99.8% complete, and had a medianduration of 4 years (interquartile range: 3—4 years; Table 4).

Trainee (n = 356) p-valueb

n %

2 0.6 0.44 1 0.29 3 0.3

92 26 0.24 1 0.33 1 0.31 0.3 0.4

412 (273; 650) 0.921 (14; 30) 0.06

550 (420; 710) 0.311.7 � 29 0.31.3 � 1.7 0.34.5 � 2.4 0.3

CK: creatine kinase; and CK-MB: myocardium brain.h and 75th centiles).

Emmanuel Villa on September 8, 2010

A. Messina et al. / European Journal of Cardio-thoracic Surgery 38 (2010) 380—386384

Table 4Follow-up resultsa.

Variables Expert surgeons (n = 967) Trainee surgeons (n = 354) p-valueb

n % n %

Completed follow-up 965 99.8 353 99.7Years of follow-up 4 (3; 4) 4 (3; 4)

Deaths during follow-up 34 3.5 15 4.2 0.6Cardiac 10 1 4 1.1 0.4Stroke 8 0.8 — 0.04Neoplastic 7 0.7 5 1.4 0.2Others 9 0.9 6 1.9 0.4

Recurrence of angina 32 3 13 3.7 0.7Resumed work/daily activities 902 93.4 337 95.4 0.6New angiography 71 7.3 22 6.2 0.4New re-vascularisation 25 2.6 12 3.4 0.4

a Values are shown as frequencies and percentages, means � SD or medians (25th and 75th centiles).b x2 test and Fisher’s exact test or the Mann—Whitney U-test.

Death during follow-up occurred in 3.5% of the patientsin the expert surgeon group and 4.2% of those in thetrainee surgeon group ( p = 0.6); there were moredeaths due to stroke in the expert surgeon group( p = 0.04).

Almost all of the patients in both the groups resumed theirworking or daily activities ( p = 0.6). The recurrence of anginawas <4% in both groups ( p = 0.7). A new angiographicexamination was performed in 7.3% of the patients in theexpert surgeon group and 6.2% of those in the trainee group( p = 0.4); 25 (2.6%) and 12 patients (3.4%) underwent new re-vascularisation.

Fig. 2 shows that Kaplan—Meier 1-, 2- and 4-year event-free survival was similar in the expert and trainee surgeongroups: 98.5% � 0.4% versus 99.7% � 0.2%, 97.6% � 0.5%versus 98.8% � 0.6% and 97.4 � 1.1% versus 94.3% � 4.1%( p = 0.4); the same was true of new re-vascularisation-freesurvival: 99.2% � 0.2% versus 98% � 0.7%, 98.6% � 0.3%versus 97.1% � 0.8% and 96.6 � 0.7% versus 95.3% � 1.4%( p = 0.3; Fig. 3).

Fig. 2. Medium-term survival of patients operated on by expert or traineesurgeons. Log rank = 0.4.

byejcts.ctsnetjournals.orgDownloaded from

4. Discussion

The widespread acceptance of a new surgical techniquedepends on its reproducibility and the feasibility of teachingit to the next generation of surgeons. Conventional CABGusing CPB and cardioplegic arrest has been the gold standardtreatment of ischaemic heart disease for several decades,and provides surgeons with a still and bloodless operatingfield. However, in order to avoid the potentially negativeeffects of extracorporeal circulation, the performance ofcoronary bypass surgery has evolved over the last few yearswith the use of beating-heart techniques. The results showthat coronary surgery without CPB is at least as safe asconventional CABG, and it offers the additional benefits ofless myocardial and cerebral damage, fewer transfusions,lower costs and a shorter hospital stay [7,9].

Although these positive results encourage its wideradoption, OPCAB surgery is not yet performed throughoutthe world, and junior surgeons are only sporadically taughthow to operate on a beating heart. The academic pathway of

Fig. 3. Actuarial freedom from new re-vascularisation in patients operated onby expert or trainee surgeons. Log rank = 0.3.

Emmanuel Villa on September 8, 2010

A. Messina et al. / European Journal of Cardio-thoracic Surgery 38 (2010) 380—386 385

coronary surgery is traditionally based on the sequentialsteps of proximal vein anastomosis, distal vein—coronaryanastomosis on an arrested heart, distal left internalmammary artery (LIMA) anastomosis on an arrested heart,the assembly of composite conduits or sequential grafts andanastomosis on a beating heart; however, although OPCABsurgery may require a different approach because some ofthe typical CABG skills are more relevant than others, only afew surgical schools have changed their training schedule[24].

This resistance to change is mainly due to the sporadic useof OPCAB surgery by the training institute itself. However, ifcoronary surgery is systematically performed without CPB (ashas been the case in our hospital, where up to 95% of surgicalre-vascularisations are carried out using OPCAB techniques),it is necessary to modify training as well as surgical,anaesthesiology and nursing procedures and technologies.

After we introduced the techniques of heart positioning,coronary shunting and graft construction at the end of the1990s, someoftheexperiencedsurgeons inourunittaughttheiryounger colleagues how to perform beating-heart operationseven though our hospital is not a formally established teachinghospital. A graded approach was scheduled for each trainee,irrespective of any previous experience with conventionalCABG — and they often used OPCAB surgery for their very firstmyocardial re-vascularisation.

In addition to undergoing a long period of assiduouslyassisting with OPCAB cases, the trainees were taught how toprepare and handle a skeletonised composite arterial graft inorder to make them familiar with arterial anastomoses(mainly Y-grafts using IMAs), shunt insertion, LAD anasto-moses (mainly using the LIMA), heart positioning andstabilisation for OM and PDA as single anastomoses and thensequential anastomoses.

In general, we believe that coronary shunts are anessential acquisition of modern surgical practice and thatcoronary snaring has to be abandoned: modern shunts seemto be truly atraumatic, avoidmyocardial ischaemia and act asa mould that guarantees patency (which also helps beginnersin coronary surgery). Haste and imprecision are usually theenemies of surgeons. Moreover, suturing a coronary artery invivo (rather than the empty and collapsed structures ofcardioplegic arrest) improves haemostasis, and we haveexperienced a negligible number of anastomoses requiringadditional haemostatic sutures.

The results of this retrospective study show that OPCABsurgery can be taught to surgical trainees with good clinicaland angiographic results that are maintained during medium-term follow-up. Unlike those in studies, the patients in thisseries were not selected for OPCAB or for trainees, butrepresent a consecutive sample of the ongoing workload ofour unit. This explains the homogeneous distribution of thepreoperative data in the two groups (there was only fewstatistically significant differences between them), andsuggests that both the expert and trainee surgeons werefaced with similar cases and that the corresponding outcomewas not influenced by patient selection.

The median operative risk score (assessed using theadditive EuroSCORE) was 2 in both groups (with a 75thpercentile of 2 in the trainee and 3 in the expert surgeongroup), and there were no differences in the extent of re-

by ejcts.ctsnetjournals.orgDownloaded from

vascularisation. However, because of the complexity of thesurgery involved, the expert surgeons performed more re-operations than the trainees.

The main causes of conversion to CPB (necessary in<5% ofthe cases in both groups) were deep intra-myocardialcoronary vessels, new-onset and persistent myocardialischaemia and the prohibitive exposure of posterior orlateral coronary vessels due to cardiomegaly.

The rates of morbidity andmortality were low, with 30-daymortality being recorded in only 0.6% of the 356 patients whowere operated on by a trainee surgeon. The major complica-tions involved four patients who suffered a perioperative MI,ninewho required the use of an IABPand onewho experienceda perioperative CVA. The overall incidence of perioperativeCVA was 5/1333 (0.37%), thus supporting the well-knownneuroprotective role of OPCAB, which can reduce or eliminatethe need for aortic manipulation during surgery [25].

The incidence of angiographic examinations in bothgroups was higher than the percentage of patients withangina, probably because they were not always carried outbecause of the presence of spontaneous symptoms, but alsobecause of those due to scheduled ergometric tests.

Four-year new re-vascularisation-free survival was similarin both the groups: 96.6% � 0.7% in the patients operated onby an expert surgeon, and 95.3% � 1.4% in those operated onby a trainee.

The results of this study reflect the long experience of ourunit in OPCAB surgery, and provides further data concerningmedium-term outcomes to supplement those published byother authors who have analysed the feasibility of training inOPCAB surgery [14—18].

The study presents many intrinsic limitations primarilydue to its retrospective analyses nature and involvement ofonly a single institution. A further confounding factor may bethe fact that the operations were performed by ‘four hands’and, particularly in the early training phases, an expertsurgeon may have intervened to help the trainee to positionthe heart, find a deep vessel or perform a more-difficult-than-expected anastomosis. However, the prevalent surgeonlabelled the case. Finally, we did not consider the issuesrelating to the trainees’ background in conventional CABG ortheir learning curves.

In conclusion, our data show that OPCAB procedures canbe safely and effectively performed by cardiothoracictrainees with good results that are maintained over time.In our opinion, systematic use of the technique is the condiciosine qua non for the development of a successful trainingprogramme.

Acknowledgements

We appreciate the work of colleagues, nurses andtechnicians of the entire Cardiovascular Department at thePoliambulanza Foundation Hospital.

References

[1] Yacoub M. Off-pump coronary bypass surgery in search of an identity.Circulation 2001;104:1743—5.

Emmanuel Villa on September 8, 2010

A. Messina et al. / European Journal of Cardio-thoracic Surgery 38 (2010) 380—386386

[2] Cheng DC, Bainbridge D, Martin JE, Novick RJ. Evidence based perio-perative clinical outcomes research group. Does off-pump coronaryartery bypass reduce mortality, morbidity, and resource utilization whencompared with conventional coronary artery bypass? A meta-analysis ofrandomized trials. Anesthesiology 2005;102:188—203.

[3] Wijeysundera DN, Beattie WS, Djaiani G, Rao V, Borger MA, Karkouti K,Cusimano RJ. Off pump coronary artery surgery for reducing mortalityand morbidity: meta-analysis of randomized and observational studies. JAm Coll Cardiol 2005;46:872—82.

[4] Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD.Economic outcome of off-pump coronary artery bypass surgery: a pro-spective randomized study. Ann Thorac Surg 1999;68:2237—42.

[5] Puskas JD, Williams WH, Mahoney EM, Huber PR, Block PC, Duke PG,Staples JR, Glas KE, Marshall JJ, Leimbach ME, McCall SA, Petersen RJ,Bailey DE, Weintraub WS, Guyton RA. Off-pump vs. conventional coronaryartery bypass grafting: early and 1-year graft patency, cost, and quality-of-life outcomes: a randomized trial. J Am Med Assoc 2004;291:1841—9.

[6] Nathoe HM, van Dijk D, Jansen EWL, Suyker WJ, Diephuis JC, van BovenWJ, de la Riviere AB, Borst C, Kalkman CJ, Grobbee DE, Buskens E, deJaegere PP, Octopus Study Group. A comparison of on-pump and off-pumpcoronary bypass surgery in low risk patients. N Engl J Med 2003;348:394—402.

[7] Angelini GD, Culliford L, Smith DK, Hamilton MC, Murphy GJ, Ascione R,Baumbach A, Reeves BC. Effects of on- and off-pump coronary arterysurgery on graft patency, survival, and health-related quality of life: long-term follow-up of 2 randomized controlled trials. J Thorac CardiovascSurg 2009;137:295—303.

[8] Cleveland Jr JC, Shroyer AL, Chen AY, Peterson E, Grover FL. Off-pumpcoronary artery bypass grafting decreases risk-adjusted mortality andmorbidity. Ann Thorac Surg 2001;72:1282—8.

[9] Abu-Omar Y, Taggart DP. The present status of off-pump coronary arterybypass grafting. Eur J Cardiothorac Surg 2009;36:312—21.

[10] Song HK, Petersen RJ, Sharoni E, Guyton RA, Puskas JD. Safe evolutiontowards routine off-pump coronary artery bypass: negotiating the learn-ing curve. Eur J Cardiothorac Surg 2003;24:947—52.

[11] Troise G, Brunelli F, Cirillo M, Dalla Tomba M, Tasca G, Mhagna Z, Danzi GB,Amari B, Quaini E. Off-pump coronary surgery in a single center experi-ence: from selective to systematic use. Ital Heart J 2002;3:446—54.

[12] Villa E, Cirillo M, Troise G. Beating heart coronary surgery: significance ofsystematic use. G Ital Cardiol 2006;7:709—10.

[13] Alex J, Rao VP, Cale AR, Griffin SC, Cowen ME, Guvendik L. Surgical nurseassistants in cardiac surgery: a UK trainee’s perspective. Eur J Cardi-othorac Surg 2004;25:111—5.

[14] Caputo M, Bryan AJ, Capoun R, Mahesh B, Ciulli F, Hutter J, Angelini GD.The evolution of training in off-pump coronary surgery in a singleinstitution. Ann Thorac Surg 2002;74:S1403—7.

[15] Caputo M, Chamberlain MH, Ozalp F, Underwood MJ, Ciulli F, Angelini GD.Off-pump coronary operations can be safely taught to cardiothoracictrainees. Ann Thorac Surg 2001;71:1215—9.

[16] Ascione R, Reeves BC, Pano M, Angelini GD. Trainees operating on high-risk patients without cardiopulmonary bypass: a high-risk strategy? AnnThorac Surg 2004;78:26—33.

[17] Karagounis A, Asimakopoulos G, Niranjan G, Valencia O, ChandrasekaranV. Complex off-pump coronary artery bypass surgery can be safely taughtto cardiothoracic trainees. Interact Cardiovasc Thorac Surg 2006;5:222—6.

[18] Asimakopoulos G, Karagounis AP, Valencia O, Rose D, Niranjan G, Chan-drasekaran V. How safe is it to train residents to perform off-pumpcoronary artery bypass surgery? Ann Thorac Surg 2006;81:568—72.

[19] Gurbuz A, Emrecan B, Yilik L, Ozsoyler I, Kestelli M, Ozbek C, Karahan N.Intracoronary shunt reduces postoperative troponin leaks: a prospectiverandomized study. Eur J Cardiothorac Surg 2006;29:186—9.

[20] Cirillo M, Messina A, Dalla Tomba M, Brunelli F, Mhagna Z, Villa E, DettoriE, Troise G. A new no-touch aorta technique for arterial-source, off-pump

byejcts.ctsnetjournals.orgDownloaded from

coronary surgery. Ann Thorac Surg 2009;88:e46—7. doi: 10.1016/j.athor-acsur.2009.07.045.

[21] Calafiore AM, Di Giammarco G, Teodori G, Di Mauro M, Iaco AL, Bivona A,Contini M, Vitolla G. Late results of first myocardial revascularization inmultiple vessel disease: single versus bilateral internal mammary arterywith or without saphenous vein grafts. Eur J Cardiothorac Surg2004;26:542—8.

[22] Calafiore AM, Di Mauro M, Teodori G, Di Giammarco G, Cirmeni S, ContiniM, Iaco AL, Pano M. Impact of aortic manipulation on incidence ofcerebrovascular accidents after surgical myocardial revascularization.Ann Thorac Surg 2002;73:1387—93.

[23] Nashef S, Roques F, Michael P, Gauducheau E, Lemeshow S, Salamon R.European System for Cardiac Operative Risk Evaluation (EuroSCORE). EurJ Cardiothorac Surg 1999;16:9—13.

[24] Sergeant P, de Worm E, Meyns B, Wouters P. The challenge of depart-mental quality control in the reengineering towards off-pump coronaryartery bypass grafting. Eur J Cardiothorac Surg 2001;20:538—43.

[25] Patel NC, Pullan MD, Fabri BM. Does off-pump total arterial revascular-ization without aortic manipulation influence neurological outcome? Astudy of 226 consecutive, unselected cases. Heart Surg Forum 2001;5:28—32.

Appendix A. Conference and discussion

Dr M. Shrestha (Hannover, Germany): I fully agree with you that onlycomplete commitment of an institution can lead to any full implementation ofany new techniques such as OPCAB. Moreover, any new technique, howeverbeneficial or better than the conventional technique, will only be broadlyaccepted if it is reproducible and can be taught to the next generation ofsurgeons. This has also been shown very successfully in the case of OPCAB by DrSergeant’s team in Leuven over many years. But I do have some questionsregarding your presentation.

First of all, although in the patient characteristics you say that they weresimilar in both groups operated by the senior surgeons and the residents, howwere the patients selected? Was it blindly, randomised, or by selection?

Dr Messina: Patients were not selected for OPCAB and also were notselected for trainee. They represent a consecutive sample of the ongoingworkload of our unit during 8 years.

Dr Shrestha: In 40% of your patients with triple-vessel disease, the rightcoronary was re-vascularised with a venous graft or a radial artery graft as afree graft with a proximal anastomosis to the aorta. As we all know, one of thebiggest advantages of OPCAB is the so-called ‘no-touch’ technique of theaorta. Since you are doing the proximal anastomosis on the aorta, how do youexplain the supposed advantage of this technique against a normal techniquewith CPB, because if you touch the aorta, then you can as well do it with CPB.

Dr Messina: Well, since ’98 we use both mammaries arranged in a ‘Y’ graftto perform a re-vascularisation of the anterior and lateral coronary system anda saphenous vein graft proximally anastomosed to the aorta for the otheranastomoses. Four years ago we analysed our results, and noted that thepatients who didn’t receive a proximal anastomosis showed less incidence ofcerebrovascular accidents. So recently we developed a new technique to avoidaortic clamping. We performed a proximal anastomosis of the saphenous graftto the right mammary artery stump, termino-terminal anastomosis, to avoidthe tangential clamp and to reduce cerebrovascular accidents as discussed thismorning in the SYNTAX study.

Dr Shrestha: I have a final short question. Did you look into the patientswho were re-vascularised again and did you find out if the reoperations werebecause of new stenosis or because of problems in the grafts?

Dr Messina: Well, the analysis was conducted on new re-vascularisationsoverall. Also patients who received a new re-vascularisation in the coronarythat were not treated with an off-pump bypass were included in the statisticalanalysis to overestimate the need of new re-vascularisation.

Emmanuel Villa on September 8, 2010

DOI: 10.1016/j.ejcts.2010.02.029 2010;38:380-386 Eur J Cardiothorac Surg

Cirillo, Federico Brunelli, Eugenio Quaini and Giovanni Troise Antonio Messina, Emmanuel Villa, Zean Mhagna, Margherita Dalla Tomba, Marco

trainee surgeonsMedium-term results of systematic off-pump coronary surgery performed by

This information is current as of September 8, 2010

& ServicesUpdated Information

http://ejcts.ctsnetjournals.org/cgi/content/full/38/3/380including high-resolution figures, can be found at:

References

http://ejcts.ctsnetjournals.org/cgi/content/full/38/3/380#BIBL

This article cites 25 articles, 20 of which you can access for free at:

Subspecialty Collections

ionhttp://ejcts.ctsnetjournals.org/cgi/collection/extracorporeal_circulat

Extracorporeal circulation http://ejcts.ctsnetjournals.org/cgi/collection/coronary_disease

diseaseCoronary http://ejcts.ctsnetjournals.org/cgi/collection/education

Education http://ejcts.ctsnetjournals.org/cgi/collection/cardiac_other

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

Permissions & Licensing

http://ejcts.ctsnetjournals.org/misc/Permissions.shtmlor in its entirety can be found online at: Information about reproducing this article in parts (figures, tables)

Reprints http://ejcts.ctsnetjournals.org/misc/reprints.shtml

Information about ordering reprints can be found online:

by Emmanuel Villa on September 8, 2010 ejcts.ctsnetjournals.orgDownloaded from