JRSM >
Volume 97, Number 7 >
Pp. 313-316
Bristol Heart Institute, University of Bristol, Bristol Royal Infirmary, Bristol BS2 8HW, UK
Correspondence to: Professor G D Angelini
E-mail:
G.D.Angelini{at}bristol.ac.uk
Over the past decade coronary artery bypass grafting (CABG) on the beating heart (off-pump or OPCAB) has become a standard procedure in many centres worldwide; however, the practice of OPCAB ranges from 0% to 100% of cases, even between surgeons operating in the same unit. This variation is the result of several interrelated factors, the most important of which is the question of graft patency. Some surgeons express concern about the quality of distal anastomoses performed on a moving field, and question the need to alter their practice from the perceived gold standard of conventional coronary artery surgery with cardiopulmonary bypass and cardioplegic arrest. This scepticism is self perpetuating, since a lack of familiarity reinforces the perception that OPCAB is technically too demanding or gives worse outcomes. Another major consequence is a paucity of adequate training in OPCAB techniques. In truth, the evidence to date indicates that OPCAB, when compared with conventional CABG, is associated with at least equivalent clinical outcome at lower cost.1-11 Furthermore, training in OPCAB techniques, as part of a structured programme, has been shown to be safe and successful.12,13 Existing evidence challenges the paradigm that conventional on-pump CABG represents the gold standard in surgical revascularization. This has important implications for the future practice of cardiac surgery. In particular, are current cardiothoracic training programmes in the UK and elsewhere capable of training the next generation of surgeons in OPCAB techniques?
Outcomes after off-pump surgery
In the randomized clinical trials of off-pump versus on-pump surgery, the number of vessels treated has generally been equivalent in the two groups, with short-term outcomes either better or similar for off-pump surgery. In the Beating Heart against Cardioplegic Arrest Studies (BHACAS) we randomized 401 patients to either on or off pump multivessel CABG.1 In BHACAS-1 patients were excluded if they required grafts to the distal marginal branches, had poor left ventricular function or had recent myocardial infarction, although these were included in BHACAS-2. A meta-analysis of these studies showed that postoperative inotropic requirements, postoperative blood loss and consequent transfusion requirement, intubation time, frequency of postoperative chest infections and arrhythmias, intensive care unit stay and hospital length of stay were all significantly lower for off-pump CABG.1
The Octopus Study Group randomized 281 patients in a multicentre study,2 and Puskas and colleagues at Emory University3 randomized 200 patients, to either on or off pump multivessel CABG. In both trials, those undergoing off-pump surgery had shorter ventilation times, were discharged from hospital earlier and bled less, requiring fewer transfusions. Results of PRAGUE-4 (400 patients randomized) likewise favoured off-pump surgery, with the exception of number of distal anastomoses performed, which was lower; total hospital costs were about 25% less for off-pump surgery.4 Two other trials, by Légaré5 (300 randomized) and Gerola6 (160) and co-workers, revealed no significant differences in short-term outcomes.
The patients recruited in these randomized trials were usually selected in some way—e.g. for low risk or limited vascular disease. Only in PRAGUE-4 were they specified to be unselected. But registry data, which are more inclusive and therefore more representative of overall practice, likewise point to advantages for off-pump CABG.8,9 Cleveland and co-workers,9 in a multicentre risk-adjusted analysis of 118 140 patients undergoing isolated CABG (11 717 off pump) drawn from the National Adult Cardiac Surgery Database of the Society of Thoracic Surgeons, showed significant benefits associated with off pump—lower operative mortality (2.3% versus 2.9%) and fewer major complications such as deep sternal wound infection, bleeding, renal failure and need for prolonged ventilation (overall complication rate 10.6% versus 14.1%). Lower morbidity with OPCAB is attributed to avoidance of the haemostatic and systemic inflammatory activation and the gaseous and particulate emboli that result from cardiopulmonary bypass. These benefits are especially pronounced in those at higher risk of cardiopulmonary-bypass-associated morbidity—notably, people of advanced age and those with poor left ventricular function, renal impairment, obesity or diabetes.14 In the largest contemporary multicentre comparison (17 969 OPCAB patients) of outcomes in multivessel CABG performed with and without cardiopulmonary bypass, subgroup analysis of propensity-matched groups identified off-pump survival benefits in patients with previous CABG (odds ratio 0.53), diabetes (0.66), left ventricular ejection fraction 30-50% (0.75), and age 66-75 years (0.80); also, the odds ratio in women was 0.79.14
Graft patency
The success of coronary artery bypass grafting is ultimately measured by long-term graft patency. The difficulties of operating on the beating heart, especially when grafting vessels on the lateral or inferior wall, might compromise the quality of the distal anastomoses and result in inferior long-term graft patency. So, might the improvements in short-term outcome be at the expense of long-term benefits? Large observational studies report excellent angiographic patency following OPCAB, although these are prone to reporting bias. Puskas et al.15 assessed 421 grafted arteries in 167 OPCAB patients and reported early graft patency rates of 98.8%. Similarly in an observational study of 257 patients undergoing total arterial grafting, early postoperative angiography revealed 97.8% (264/270) patency of internal thoracic arteries and 97.9% (512/523) patency of radial arteries in 238 patients following off-pump surgery.16 In contrast, Khan and colleagues,17 in a randomized study of 104 CABG patients (angiographic follow-up incomplete), reported graft patency rates at three months of 98% and 88% in on and off pump groups, respectively—a highly significant difference (P50.002). The two surgeons in this trial only occasionally performed off-pump surgery in their normal practice, however (13% of all CABG cases in the preceding two years), and it is difficult to see how this cannot have introduced a bias into the result. Analysis of 16 988 consecutive patients operated on in seventy-two US centres demonstrated that OPCAB performed in low as opposed to high volume OPCAB units was associated with significantly greater cardiac specific as well as overall mortality.18 Angiographic follow-up of a subgroup of patients from the multicentre Octopus Study Group trial, where operators were more familiar with OPCAB techniques, demonstrated one-year patency of 93% and 91% for on and off pump, respectively (absolute difference 2%; 95% confidence interval -6.5 to +10.4).10
Neither of these randomized trials was powered to detect differences in graft patency, and adequately sized angiographic follow-up of randomized trials of on versus OPCAB have yet to be performed. Clinical follow-up of several large randomized studies does not, however, support the notion that graft patency is inferior following OPCAB. Lower graft patency ultimately results in higher mortality and frequency of cardiac related events, an earlier return of symptoms, lower quality of life, and an increased requirement for repeat revascularization procedures. Midterm follow-up of the BHACAS-1 and BHACAS-2 patients demonstrated no difference in mortality or cardiac events at a mean follow-up of 25 months for BHACAS-1 and 14 months for BHACAS-2.1 33 (17%) of 200 patients in the off-pump group died or had a cardiac-related event, compared with 42 (21%) of 201 in the on-pump group (odds ratio 0.78; 95% confidence interval 0.49 to 1.22). Similarly, one-year follow-up from the Octopus trial10 revealed no difference in the rate of freedom from death, stroke, myocardial infarction, and coronary reintervention—91% on-pump versus 88% off-pump (absolute difference 2.6%; 95% confidence interval -4.6 to +9.8). This study also revealed no difference in quality of life between the two groups; in both groups EuroQol summary scores increased from 0.65 at baseline to 0.84 three months postoperatively. Analysis of disease specific and generic quality of life in BHACAS-1 and -2 patients at a median of three years' follow-up likewise revealed no difference between on and off pump groups.11
Training and OPCAB
Whilst some practising surgeons remain to be convinced of the merits of OPCAB, this is not the case with the next generation of surgeons. In a survey of cardiothoracic trainees in the UK, 96% felt that training in OPCAB was essential (S Bays, GD Angelini, unpublished). Scepticism on the part of trainers has had a negative effect on the widespread introduction of OPCAB techniques, however, and only 51% of UK trainees surveyed had gained experience of OPCAB in their training programme. Bristol has led the way in training in OPCAB techniques, and over the past five years we have successfully trained four of our residents in off-pump surgery, three of whom are now consultants. As OPCAB surgery was integrated into the training programme at our institution the proportion of OPCAB procedures performed by trainees increased from 18% to 62% between the years 1999 and 2001.13 The initial phase was aimed at making the trainees familiar with handling the beating heart and learning the technique of exposing and stabilizing the target vessels. Once they had acquired proficiency with this they were allowed to perform anastomoses on the anterior surface of the heart (mostly left internal mammary to left anterior descending or grafting of the diagonal branches) before gradually moving to posterior descending and circumflex grafting. In 2001, we reported our experience of the early and mid-term clinical outcome of OPCAB procedures performed by supervised trainees or senior surgeons. Multivariate analysis demonstrated that the risk of morbidity or mortality was not increased if the OPCAB procedure was performed by a trainee.12 Thereafter, trainees with experience of at least 50 supervised OPCAB cases were allowed to perform procedures without supervision. In a new analysis, clinical outcomes of OPCAB procedures performed by trainees differed little whether they operated independently or under consultant supervision.13 Furthermore the results of OPCAB performed by trainees (n=422) were superior to those they performed on pump (n=547) over the same period.13 The Emory University group has similarly demonstrated that the learning curve in OPCAB can be traversed without any increase in patient morbidity19—an achievement they attributed to careful case selection such that high-risk patients with poor left ventricular function, left main stem or complex three-vessel disease were excluded from OPCAB until sufficient operator experience (over 200 cases) had been obtained. Analysis of the Bristol patient registry demonstrates a similar trend.13 In a separate analysis of those high-risk OPCAB patients operated on by trainees or consultants there was no significant difference between the groups for morbidity or mortality (the frequency of perioperative myocardial infarction was actually higher in the consultant group) after adjustment for potentially confounding preoperative factors.20 Recently, continuous performance monitoring has been introduced to complement our training programme by monitoring changes in performance over time and alerting trainers to suboptimal performance. This has shown that cumulative performance (as assessed by variable life-adjusted displays and risk-adjusted probability ratio tests) by trainees becomes equal to, or better than, that of a consultant after approximately 100 OPCAB cases. In addition, performance for all trainees in off-pump cases was equivalent to or better than their performance with conventional on-pump CABG.21 From this evidence we conclude that, where senior surgeons are proficient in OPCAB techniques, a structured cardiothoracic training programme that includes OPCAB surgery is both possible and safe for patients.
Conclusion
As the clinical evidence in support of OPCAB mounts, a consensus must be reached as to the indications for OPCAB surgery. What is clear from the current evidence is that, in certain patients, OPCAB offers advantages over conventional CABG and that the existing paradigm must change. Cardiothoracic training in the UK is not yet addressing this change.
REFERENCES
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
