HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Paul Sergeant Patrick Wouters Bart Meyns Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sergeant, P. Articles by Slabbaert, K. Search for Related Content PubMed PubMed Citation Articles by Sergeant, P. Articles by Slabbaert, K. Related Collections Coronary disease Extracorporeal circulation Minimally invasive surgery

Eur J Cardiothorac Surg 2004;25:779-785
© 2004 Elsevier Science NL

OPCAB versus early mortality and morbidity: an issue between clinical relevance and statistical significance

^a Department of Cardiac Surgery, Gasthuisberg University Hospital, Herestreet, 3000 Leuven, Belgium
^b Department of Anaesthesia, Gasthuisberg University Hospital, Leuven, Belgium
^c Bio statistical Department, Gasthuisberg University Hospital, Leuven, Belgium
^d Faculty of Medicine, Gasthuisberg University Hospital, Leuven, Belgium

Received 7 October 2003; received in revised form 19 January 2004; accepted 13 February 2004.

^* Corresponding author. Tel.: +32-163-44219; fax: +32-163-44616
e-mail: paul.sergeant{at}uz.kuleuven.ac.be

	Abstract

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
Objective: To evaluate the impact of OPCAB on major postoperative events in a large consecutive cohort of patients, covering the complete spectrum of risk. Methods: A consecutive series of 3333 CABG patients operated in a single institution (Jan/97–Jan/03) is analyzed after a complete (98%) midterm reengineering towards off-pump surgery (Oct/99). Patients in cardiogenic shock are excluded. The on- (N=1593) or off-pump (N=1740) datasets are comparable for most demographic and non-cardiac variability. The studied events are early mortality, early stroke, early infarct, early dialysis and hospital stay. Three methods adjust for possible patient selection: similar datasets, forced inclusion of a saturated OPCAB propensity score and finally multivariate correction. Results: Non-risk adjusted. The 3-month survival was 96.7±0.4% (OPCAB) and 95.9±0.5% (ECC) (P=0.2). The 8-day freedom from stroke was 99.4±0.2% (OPCAB) and 98.5±0.3% (ECC) (P=0.004). The prevalence of dialysis was 1.67% in OPCAB and 2.27% in ECC (P=0.2). The 8-day freedom from infarct was 98.4±0.2% (OPCAB) and 98.3±0.2% (ECC) (P=0.7). The freedom from hospital discharge day 15 was 17.6±0.9% (OPCAB) and 18.4±0.8% (ECC) (P=0.001). Propensity score corrected and adjusted for event-related variability. The survival effect remained non-significant (P=0.3), also for patients with a EuroSCORE>8 (P=0.9). The stroke effect became non-significant (P=0.2), but stayed significant for patients with severe internal carotid artery stenosis (P=0.02). The dialysis-effect remained non-significant (P=0.6), also for patients with an elevated creatinine (P=0.7). The early infarct-effect remained non-significant (P=0.8), also for the female patients (P=0.8). The hospital discharge was significantly influenced by the OPCAB approach for the total group (P=0.02) as well as for the patients with EuroSCORE>8 (P=0.01). Conclusions: The observed 20% reduction of mortality, 60% reduction of stroke and 20% reduction of dialysis were partly neutralized by the adjusting methods and demand, at least, larger datasets to obtain statistical significance. Subdatasets with fewer patients but higher risk identified risk-reducing effects for stroke. Hospital stay was shortened by the OPCAB approach. The interactions between risk, number of patients and the risk-reducing effect are the cornerstones of evidence generation for the OPCAB approach. These results were obtained through a very strict reengineering and cannot be extended to all OPCAB programs.

Key Words: Coronary artery bypass surgery • OPCAB • Propensity score

	1. Introduction

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
The effect of avoiding the cardiopulmonary bypass on the early clinical outcome events after CABG has been debated for some time. Several randomized trials [1–6] provided interesting information about minor clinical issues but were unable to clarify the issue of the major clinical events because of very selective patient inclusion, limited risk profile and biased observation intervals. Power calculations were difficult because of the ignorance about the expected improvements, if existent. Observational studies have often failed to convince, partly because of incomplete correction for propensity and variability between datasets. The aim of this study is to assess the effect, on major clinical events, of implementing a structured OPCAB approach in a large unselected consecutive CABG population.

	2. Materials and methods

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
2.1. Patient material
A consecutive series of 3333 CABG patients operated in a single institution (Jan/97–Jan/03) is analyzed after a complete (98%) midterm reengineering towards off-pump surgery. Patients in cardiogenic shock are excluded, acute infarcts with stable haemodynamics included. The patients are identified as ‘OPCAB’ patients if the procedure was started using this approach. A conversion towards the extra-corporeal circulation during the procedure did not change the patient's categorization. The propensity towards the OPCAB approach was extremely biased (see Table 1) until the midterm reengineering (Oct/99), limitedly biased in 2000 and unbiased in 2001 and 2002. The demographic, non-cardiac and cardiac variability of the on- (N=1593) or off-pump (N=1740) datasets is compared in Table 2.

2.3. The OPCAB approach and reengineering
The OPCAB approach was unstructured before the reengineering, as well for the enucleating, for the stabilizing as for the shunting aspects. The anesthesia was based on the cardiopulmonary bypass approach. The lack of global concept limited the applicability of this procedure.

The OPCAB reengineering consisted of a complete and structured change of all procedures, in the domains of surgery, anesthesia, nursing and operating room logistics. The surgical procedure was split-up into five distinctive and sequential elements: enucleation, visualization, stabilization, shunting and anastomosis. The enucleation was based on a single sling anchored with a tourniquet at the extreme right posterior pericardium. The ventricle was further reformatted and displaced laterally with an apical suction device. The anastomotic area was stabilized with a suction-stabilizing device. Two retracting stitches on both sides of the anastomotic area stabilized and realigned the coronary vessel. One proximal snare, in the diseased section of the coronary vessel, allowed the shunt insertion. The anastomosis was performed in a continuous backhand or forehand fashion using pen-type instruments. The anesthesia management was split-up similarly into distinctive components: conditioning, anticoagulation, monitoring, reconditioning and response management. This OPCAB conditioning included Lidoflazine^©, sometimes atrial pacing or additional ß-blocking, additional sedating and analgesic or vaso-active medication. The target activating clotting time was 400 s, with full 1-to-1 conversion at the end of the procedure. Automated ST-segment analysis, CVP as well as PAP pressure readings, associated with 2DTEE volumetry, informed about ischaemia and right- or left-sided filling. Positive inotropic medication was avoided. Any deviation of the optimal conditioning during the procedure induced a reconditioning, before continuation. The surgery–anesthesia communication and interaction became the pivotal element of the procedure.

2.4. The events
The diagnosis of early perioperative infarct was documented using repeated surface electrocardiogram and routine and repeated enzymatic measurements. The cut-off value for positive enzymes was a creatinine kinase MB fraction higher than 8% of total creatinine kinase, in the presence of elevated creatinine kinase. Routine and repeated troponine measurements performed after 2000 validated the previous diagnosis.

Patients with any neurological dysfunction, including disorientation, underwent routine evaluations by neurologists, including CT-scans of the brain.

Renal insufficiency is defined as dialysis or hemofiltration during hospital stay in non-dialysis patients.

Hospital discharge is defined as discharge to home from primary hospital, secondary hospital, rehabilitation or coma center.

2.5. The propensity score modeling
The propensity score [8–10] provided an estimate of the probability of starting the procedure off-pump versus on-pump. It is a classical logistic regression, not driven by an event but by the variability in datasets. First, a parsimonious stepwise logistic model selected 12 significant variables. The area under the curve of the ROC test was 0.77 with an R² of 0.19. The parsimonious model was then extended into a saturated propensity model including non-significant descriptive variables of morbidity or variability as well as first degree interactions. The final model included more than 400 variables. The area under the curve of the ROC test was 0.82 (See Appendix A.1). The propensity score was then calculated for every patient by solving the saturated logistic equation.

2.6. The follow up methodology and outcome analysis
An anniversary 3-month follow up was made of all the patients with a completeness of 100%. The 3-month interval was chosen for structuring the survival risk, thereby including the complete periprocedural risk and unbiased by the arbitrary hospital discharge. The early morbidity events were included till day 8 after surgery, irrelevant of hospital discharge.

A Cox proportional hazard analysis was made of the time-related events and stepwise logistic regression of the non-time related events. The study sample was the complete patient group (3333 patients). The different outcome events were analyzed first uncorrected for any variability between datasets.

In a second analytic step, the saturated propensity score was forced into the model, later the residual variability, in relationship to the studied event. This correction started with the demographic variability, followed by the non-cardiac variability, then the cardiac variability, finally the procedural variability. The P-value inclusion criterion was 0.05. Continuous variables were transformed (square, square root, log, exp, reciprocal and into nominal categories) in the search for the optimal relation with the studied outcome event. The missing ejection fractions and pulmonary functions were replaced by the mean values of the total population. The residual OPCAB effect was studied, once this process was terminated. This analysis is repeated for each event on an event-specific high-risk subgroup.

	3. Results

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
3.1. The 3-month survival
The 3-month survival, uncorrected for variability, was 96.3±0.4% for the total group of 3333 patients, 96.7±0.4% for the OPCAB group and 95.9±0.5% for the ECC group. The actual number of events in the ECC group was 65 and in the OPCAB group 57. The 20%, non-risk-adjusted difference in survival of the OPCAB approach, was not significant and remained so after correction for propensity and event-related variability (see Table 3).

3.2. The 8-day freedom from stroke
The 8-day freedom from stroke, uncorrected for variability, was 98.9±0.1% for the total group, 99.4±0.2% for the OPCAB group and 98.5±0.3% for the ECC group. The actual number of events in the ECC group was 25 and in the OPCAB group 10. The 60%, non-risk-adjusted difference in prevalence with the OPCAB approach, was very significant but lost statistical significance in the correction for propensity and event-related variability (see Table 4).

3.3. The hospital-stay hemofiltration or dialysis in non-dialysis patients
The prevalence of hemofiltration or dialysis in non-dialysis patients, uncorrected for any variability, was 1.96, 1.7% for the OPCAB group and 2.3% for the ECC group. The actual number of events in the ECC group was 36 and in the OPCAB group 29. The 26%, non-risk-adjusted difference in need for hemofiltration or dialysis, was not significant (P=0.2) and remained so (P=0.6) after correction for propensity and event-related variability (see Appendix A.3).

The prevalence of hemofiltration or dialysis, in patients with a preoperative creatinine level equal or above 1.5 mg/dl (N=337), was 11.6% in the ECC patients and 9.0% in the OPCAB patients. This 22%, non-risk-adjusted difference was non-significant (P=0.4) and remained so after propensity and variability correction (P=0.7).

3.4. The 8-day freedom from infarct
The 8-day freedom from infarct, uncorrected for any variability, was 98.4±0.2% for the OPCAB group and 98.3±0.2% for the ECC group. The actual number of events in the ECC group was 24 and in the OPCAB group 29. There was no difference (P=0.7) and this remained so (P=0.8) after correction for propensity and event-related variability (see Appendix A.4).

The 8-day freedom from infarct, in female patients (N=756), was 97.1±0.9% in the ECC patients and 97.5±0.8% in the OPCAB patients. This 14%, non-risk-adjusted difference was non-significant (P=0.8) and remained so after propensity and variability correction (P=0.8).

3.5. The hospital stay
The freedom from hospital discharge alive, uncorrected for patient- and surgical variability, at 10 and 15 days was 40.0±0.8 and 18.0±0.6%, respectively, for the total group, 38.6±1.1 and 17.6±0.9%, respectively, for the OPCAB group, and 41.6±1.2 and 18.4±0.8%, respectively, for the ECC group. The statistically significant (P=0.001) difference in favor of the OPCAB approach is reduced (P=0.02) after propensity and variability correction (see Table 5).

	4. Discussion

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
4.1. Datasets and patient variability
The overall dataset is a large, high risk and unbiased sequential series of primary and repeat CABG patients, representing the workload of several years for most large teaching hospitals. The CABG patients in cardiogenic shock and cardio-pulmonary resuscitation were excluded since they increased the analytical difficulty and since these patients were nearly all operated, for obvious reasons, using extra-corporeal circulation. The therapy-specific datasets are also large, but certainly biased. The event-specific high-risk subgroups are comparable in size to an annual practice of normal CABG patients for most centers. The propensity towards OPCAB before the reengineering was low, primarily driven by the absence of revascularization needs of the posterior, lateral or inferior regions of the heart. After October 15th 1999, the reengineering date, it was 93.4% till January 1st 2003, reaching even above 98% in 2000 and 2001. We could have deleted the OPCAB experience before the reengineering, this would have precluded even less differences between the datasets, but would have deleted maybe some learning curve morbidity and mortality. Even though conversion towards cardio-pulmonary bypass, after OPCAB start, was unrelated to early mortality (P=0.4) and morbidity (stroke P=0.1, prevalence of dialysis P=0.8 and hospital stay P=0.6), it was decided to analyze the patients according to the intention to treat.

Table 2 demonstrates that the two therapy-specific subgroups are similar for most risk-inducing co-morbidity, but some residual, usually non-risk-inducing variability will have to be corrected by the analytical methods. The strictest analytical method was therefore chosen to correct this possible variability. This consisted of saturated propensity score modeling, variable transformation and event-specific multivariate time-related correction. The purpose was to approximate as much as possible randomization without some of the limitations of randomization. It is even possible that the extreme corrections have somewhat overcorrected against an OPCAB effect, e.g. by forcibly correcting for surgeon.

A great deal of variability was explained in a rich availability of variables, but, as in most teaching hospitals, several patients had high-risk variability not expressed in selected variables (e.g. patients on a liver or renal transplant waiting list, patients with active, not-yet treated pulmonary tumors, patients in bowel obstruction). These were not excluded from the analysis.

4.2. The studied events
The ‘survival’ interval studied was 3 months. Extensive literature [11,12] has identified that early mortality is not restricted to hospital stay, not even the first 30 days. Most OPCAB literature continues to use these outdated observation intervals [13–15]. We had hoped, in addition, that by including all patients and by extending the interval, enough events would be available to obtain statistical significance. It will become clear that even good sample size datasets create analytical problems due to the limited prevalence of early events in current cardiac surgery. The data analysis identified a 20% benefit for the OPCAB approach before refined risk and propensity adjustment in these two EuroSCORE risk-comparable datasets. We failed in our intention, since this large difference was still non-significant. The dataset was clearly not powered to identify a 20% difference in mortality between the two therapeutic approaches. Several studies with shorter intervals and fewer patients at risk failed similarly [14,16]. Including fewer patients counteracted the selection process towards a higher risk population and more possibility for significant risk-reduction. This higher risk subgroup was not powered for identifying a 10% risk reduction. It can be concluded that larger datasets with correct observation intervals will be needed to identify early survival benefits, if they are present. A simple power calculation for a randomized trial, in the presence of a 3% risk and a 10% reduction of this risk with the usual (0.05) and ß (0.1) errors, identifies the need for 65,000 patients in each arm.

For the event ‘stroke’ an 8-day interval, unrelated to hospital stay, was chosen based on the hazard for stroke in a consecutive series of 10,016 patients (unpublished data). Most current literature uses in-hospital stroke. In this analysis the risk-reducing effect was enormous in the 60% range for the non-risk-adjusted analysis. The significance of this difference confirmed the presence of power. Further risk adjustment reduced this below the level of significance for the total population, but the benefit subsisted for patients with severe stenosis of the internal carotid artery. The possible bias in the presence and number of proximal anastomoses were corrected twice: once in the propensity score and once where appropriate in the multivariate analysis. This confirms other well-elaborated studies [17,18]. Cleveland [15] also identified a stroke benefit with the OPCAB approach in patients with known cerebro-vascular disease, but their risk-adjustment was incomplete and selection bias was obvious. Other studies [19] failed probably because of insufficient patients at risk.

We could not confirm the reduction of ‘renal failure requiring dialysis’ with the OPCAB approach, as seen by Sabik [14] and Arom [19]. The analysis was certainly not powered to identify differences of 20%. The decision for dialysis or hemofiltration is a human intervention, probably driven differently between institutions. Our analysis was certainly not as refined as the one described by Ascione [20], who observed significant differences in creatinine clearance between on-pump and off-pump approaches.

The absence of any difference in freedom from early infarct is probably due to the perfect enucleating and visualization method used, optimized by the routine shunting in all patients and vessels, creating a relaxed and stress-free environment. The close monitoring allowed real time detection of any anastomotic problem and its correction.

Since most patients are discharged from hospital alive, any identification of benefit becomes readily visible. Indeed hospital discharge, including secondary hospitals and revalidation centers, was shortened with OPCAB. Even though our hospital discharge was procedure-driven, a bias remains possible.

An alternative possibility is to combine events, as seen in interventional cardiology and vascular surgery publications. This has the possibility of reaching power, but the limitation of equalizing lethal and non-lethal events. We tested this option by combining, for all patients, all events (death, stroke, infarct and dialysis) but limited to the first 8 days. There was a non-risk adjusted 17% reduction of events using the OPCAB approach (5.34% for ECC versus 4.31% for OPCAB) but P-value 0.16. A second option combined, for all patients, all events (death, stroke, infarct and dialysis) but limited to the first 8 days for the last three and limited to 3 months for death. There was a non-risk adjusted 12% reduction of events using the OPCAB approach (7.03% for ECC versus 5.86% for OPCAB) but the P-value was 0.16. Both of these options were indicative of absences of power, even before refined correction for propensity and variability.

It was interesting to note that none of the major events was negatively influenced by the OPCAB approach.

	5. Conclusion

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
The interactions between risk, number of patients and the risk-reducing effect are the cornerstones of evidence generation for the OPCAB approach. The rules of statistics are stringent but immutable; the consequence is that for some time it will be an issue between clinical relevance and statistical significance. These results were obtained through a very strict reengineering and cannot be extended to all OPCAB programs.

	Footnotes

 
Presented at the joint 17th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 11th Annual Meeting of the European Society of Thoracic Surgeons, Vienna, Austria, October 12–15, 2003.

	Appendix A

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
Appendix A.1. The list of the variables included in the parsimonious propensity model
Weight before surgery, ejection fraction (and its transformation), presence of left ventricular hypertrophy.

Presence of unstable ST-segment at the start of surgery, number of anastomoses, presence and number of anastomoses on the lateral-posterior-inferior segment of the heart, presence and number of anastomoses on the aorta, redo procedure, EuroSCORE risk (additive model), surgeon

Appendix A.2. The list of the variables included in the saturated propensity model, (their first-degree interactions were added)
Age, weight before surgery, body mass, one second vital capacity (% of normal), history of renal failure (dialysis or creatinine>3 mg/dl), ejection fraction (and its transformation), presence of left ventricular hypertrophy, presence of unstable st-segment at the start of surgery, presence of triple vessel disease, number of anastomoses, presence and number of anastomoses on the lateral-posterior-inferior segment of the heart, presence and number of anastomoses on the aorta, redo procedure, acute (same day) infarct, EuroSCORE risk (additive model), surgeon, presence of disease on the ascending aorta, drug treated diabetes.

Appendix A.3. The list of variables included in the logistic regression correcting for variability in the analysis of the prevalence of dialysis and hemofiltration in the non-dialysis patients
Saturated propensity score (P=0.24), age (squared transformation P<0.0001), insulin treated diabetes (P<0.0001), creatinine before surgery (P<0.0001), unstable ST-segment at surgery (P=0.02) disease of ascending aorta (P=0.0001), EuroSCORE additive value>11 (P=0.03).

Appendix A.4. The list of variables included in the proportional hazard model correcting for variability in the analysis of the freedom from infarct
Saturated propensity model (P=0.2), female gender (P=0.007), body mass index (P=0.03), dialysis or creatinine >2 mg/dl (P=0.04), history of stroke (P=0.02), disease of ascending aorta (P=0.04).

	Appendix B. Conference discussion

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 
Dr J. Revuelta (Santander, Spain): We know that you are an expert in mathematics and statistics not only in surgery, but for a standard surgeon like me, would you please explain a little bit about this propensity scoring and saturated propensity scoring?

Dr Sergeant: There are two levels of propensity score correction. There is a parsimonious correction and there is a saturated propensity score correction. What is the propensity? The propensity is that what drives you towards one approach versus the other. So it is a mathematical model where the event that you study is not mortality or survival but the event is just a choice of OPCAB, yes or no.

A parsimonious model is where you only include these variables that are statistically significant. A saturated model is saturated with all kinds of variables, statistically significant or non-statistically significant. In this particular model there were more than 400 variables.

Dr R. El Oakley (Singapore): In the current era of evidence-based medicine, what class "A" evidence did you have to change your practice in favor of OPCAB in more than 90% of the cases?

Dr Sergeant: The shift towards more than 90% OPCAB was a mental shift based on the evidence that today coronary surgery, even in our institute, is done with a certain level of stroke and a certain level of risk for some patients, and we decided on one particular date that we would just stop having our patients submit to any risk. It was our intention to reduce mortality and morbidity to zero. We failed in that respect. We are lucky the way it has gone, but we have to understand that coronary surgery today is still not perfect.

Dr El Okaley: Does this mean that OPCAB on more than 90% of patients coming for CABG is not evidence-based?

Dr Sergeant: It is not evidence-based.

	References

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusion Appendix A Appendix B. Conference... References

 

1. Angelini G., Taylor F., Reeves B., Ascione R. Early and midterm outcome after off-pump and on-pump surgery in Beating Heart against Cardioplegic Arrest Studies (BHACAS 1 and 2): a pooled analysis of two randomized controlled trials. Lancet 2002;359:1194-1199.[CrossRef][Medline]
2. Nathoe H., van Dijk D., Jansen E., Suyker W., Diephuis J., van Boven W.J., Brutel de la Rivière A., Borst C., Kalkman C., Grobbee D., Buskens E., de Jaegere P. A comparison of on-pump and off-pump coronary bypass surgery in low-risk patients. N Engl J Med 2003;348:394-402.[Abstract/Free Full Text]
3. Puskas J., Williams W., Duke P., Staples J., Glas K., Marshall J., Leimbach M., Huber P., Garas S., Sammons B., McCall S., Petersen R., Bailey D., Chu H., Mahoney E., Weintraub W., Guyton R. Off-pump coronary artery bypass grafting provides complete revascularisation with reduced myocardial injury, transfusion requirements and length of stay: a prospective randomized comparison of two-hundred unselected patients undergoing off-pump versus conventional coronary artery bypass grafting. J Thorac Cardiovasc Surg 2003;125:797-808.[Abstract/Free Full Text]
4. Zamvar V., Williams D., Hall J., Payne N., Cann C., Young K., Karthikeyan S., Dunne J. Assessment of neurocognitive impairment after off-pump and on-pump techniques for coronary artery bypass graft surgery: prospective randomized controlled trial. Br Med J 2002;325:1268-1273.[Abstract/Free Full Text]
5. Diegeler A., Hirsch R., Schneider F., Schilling L., Falk V., Rauch T., Mohr F. Neuromonitoring and neurocognitive outcome in off-pump versus conventional coronary bypass operation. Ann Thorac Surg 2000;69:1162-1166.[Abstract/Free Full Text]
6. Van Dijk D., Jansen E., Hijman R., Nierich A., Diephuis J., Moons K., Lahpor J., Borst C., Keizer A., Nathoe H., Grobbee D., De Jaegere P., Kalkman C. Cognitive outcome after off-pump and on-pump coronary artery bypass graft surgery, a randomized trial. J Am Med Assoc 2002;287:1405-1412.[Abstract/Free Full Text]
7. Flameng W., Borgers M., Van der Vusse G., Demeyere R., Vandermeersch E., Thoné F., Suy R. Cardioprotective effects of lidoflazine in extensive aorta-coronary bypass grafting. J Thorac Cardiovasc Surg 1983;85:758-768.[Abstract]
8. D'Agostino R.B. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17(19):2265-2281.[CrossRef][Medline]
9. Grunkemeier G., Payne N., Ruyun J., Hnady J. Propensity score analysis of stroke after off-pump coronary artery bypass grafting. Ann Thorac Surg 2002;74:301-305.[Free Full Text]
10. Blackstone E. Comparing apples and oranges. J Thorac Cardiovasc Surg 2002;123:8-15.[Free Full Text]
11. Osswald B., Blackstone E., Tochtermann U., Schweiger P., Thomas G., Vahl C., Hagl S. The meaning of early mortality after CABG. Eur J Cardiothorac Surg 1999;15:401-407.
12. Sergeant P., Blackstone E., Meyns B., Leuven K.U. Coronary surgery program. Validation and interdependence with patient-variables of the influence of procedural variables on early and late survival after CABG. Eur J Cardiothorac Surg 1997;12:1-19.[Abstract]
13. Plomodon M., Cleveland J., Ludwig S., Grunwald G., Kiefe C., Grover F., Shroyer L. Off-pump coronary artery bypass is associated with improved risk-adjusted outcomes. Ann Thorac Surg 2001;72:114-119.[Abstract/Free Full Text]
14. Sabik J., Gillinov M., Blackstone E., Vacha C., Houghtaling P., Navia J., Smedira N., McCarthy P., Cosgrove D., Lytle B. Does off-pump coronary surgery reduce morbidity and mortality?. J Thorac Cardiovasc Surg 2002;124:698-707.[Abstract/Free Full Text]
15. Cleveland J., Shroyer L., Chen A., Peterson E., Grover F. Off-pump coronary artery bypass grafting decreases risk-adjusted mortality and morbidity. Ann Thorac Surg 2001;72:1282-1289.[Abstract/Free Full Text]
16. Al-Ruzzeh S., Nakamura K., Athanaiou T. Does off-pump coronary artery bypass surgery improve the outcome in high-risk patients?. Eur J Cardiothorac Surg 2003;23:50-55.[Abstract/Free Full Text]
17. Stamou S., Jablonski K., Pfister A., Hill P., Dullum M., Bafi A., Boyce S., Petro K., Corso P. Stroke after conventional versus minimally invasive coronary artery bypass. Ann Thorac Surg 2002;74:394-399.[Abstract/Free Full Text]
18. Patel N., Deodhar A., Grayson A., Pullan M., Keenan D., Hasan R., Fabri B. Neurological outcomes in coronary surgery: independent effect of avoiding cardiopulmonary bypass. Ann Thorac Surg 2002;74:400-406.[Abstract/Free Full Text]
19. Arom K., Flavin T., Emery R., Kshettry V., Janey P., Petersen R. Safety and efficacy of off-pump coronary artery bypass grafting. Ann Thorac Surg 2000;69:704-710.[Abstract/Free Full Text]
20. Ascione R., Lloyd C., Underwood M., Gomes W., Angelini G. On-pump versus off-pump coronary revascularization: evaluation of renal function. Ann Thorac Surg 1999;68:493-498.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Paul Sergeant Patrick Wouters Bart Meyns Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sergeant, P. Articles by Slabbaert, K. Search for Related Content PubMed PubMed Citation Articles by Sergeant, P. Articles by Slabbaert, K. Related Collections Coronary disease Extracorporeal circulation Minimally invasive surgery