OPCAB surgery: a critical review of two different categories of pre-operative ejection fraction (original) (raw)

Abstract

Objective: Literature review found little information on off-pump coronary artery bypass (OPCAB) procedure in patients with poor left ventricular function and there was no information comparing the low EF and normal EF patients undergoing OPCAB procedure. Methods: Between 1/1/1998 and 6/30/1999, 387patients had surgery performed utilizing the off-pump technique and 45 of these patients had pre-operative left ventricular function of equal to or less than 30% (LVEF ≦30). The two groups (LVEF ≦30 and LVEF≫30) were compared using univariate analysis. Patients in LVEF ≦30 were older and more female gender. LVEF≪30 had more NYHA class IV patients (64 vs. 50%) and more symptoms related to depressed left ventricular function. The mean pre-operative left ventricular function was 25% in LVEF ≦30 and 56% in LVEF≫30. Pre-operative predicted risk was 6.4±5.5% in LVEF ≦30 and 2.7±4.5% in LVEF≫30 (P≪0.001). Most (≫95%) of the patients in both groups were elective status, and LVEF ≦30 patients had increased incidence of redo (11 vs. 6%, P=0.2). In LVEF≫30, 84% of the patients had stable angina while only 69% in LVEF ≦30 (P=0.009). Results: Intra-operatively no significant differences were measured in number of grafts per patient (2.7 vs. 2.8), amount of blood loss, peak CK-MB, skin-to-skin time, or OR time. Patients with LVEF ≦30 have more frequent utilization IABP during pre, intra and post-operative period. The statistical analysis yields no significance in post-operative major neurological deficit between these two groups; and are comparative to the nationally reported incidence of neurological deficit for on-pump patients. The operative mortality in the low EF group was 4.4 and 1.8% in LVEF≫30 group (P=0.23). Conclusions: Given the clinical presentation of the low EF group, higher prediction risk, longer pre-operative stay, and length of ventilation (24 vs. 8 h P=0.12) a longer surgery to discharge stay (8 vs. 6 days, P=0.02) is anticipated. Short-term clinical outcomes for both groups of OPCAB patients encouraged us to continue to offer this approach to this broad base of patient population.

1 Introduction

Earlier studies have concluded that risks of on-pump surgery and operative mortality are prohibitive for patients with severe left ventricular dysfunction (30%) [1]. Management of patients with ventricular function impairment is also unclear, and there is reluctance on the part of some cardiologists to refer these patients for surgical intervention [1,2]. A recent study suggested that overall comparison of operative mortality and in-hospital complications between on-pump and off-pump patients, the off-pump group has the same outcomes when compared to the low and medium risk on-pump groups. Comparing the high-risk patient groups, it was the off-pump group that significantly lowered operative mortality [3].

Off-pump coronary artery bypass (OPCAB) surgery through a full median sternotomy has recently gained renewed interest for multi-vessel revascularization [4,5]. To be able to bypass the circumflex system and the postero-lateral branches of the right, one must expose the postero-lateral aspect of the heart by displacing the heart vertically. Displacement of the beating heart in animals and patients with preserved left ventricular function may be well tolerated. In an ischemic heart or critically impaired left ventricle, any hemodynamic instability may lead to increased complications or death [3].

Literature review found little information on OPCAB procedure in patients with poor left ventricular function, and no information comparing the low EF and normal EF patients undergoing OPCAB procedure. The aim of this study was to: (a) analyze our own experience in patients with depressed left ventricular function (EF ≦30%) who underwent off-pump multi-vessel bypass surgery; and (b) provide group comparison of patients with EF of ≦30% to those patients with an EF ≫30% using univariate analysis.

2 Material and methods

Our cardiac surgical registry was queried and resulted in 2303 consecutive patients having had full sternotomy and multi-vessel coronary artery bypass graft (CABG) performed between 1/1/1998 and 6/30/1999. Three hundred and eighty-seven of these patients had surgery done utilizing off-pump technique. Forty-five of these patients had pre-operative left ventricular function of ≦30% as calculated by angiographic ventriculogram (LVEF ≦30). Three hundred and forty-two of these patients had left ventricular function ≫30% (LVEF ≫30). The pre-operative demographics, pre-operative co-morbidities, operative factors, pre and post-operative variables, and post-operative complications and mortality of these groups were analyzed.

Pre-operative, intra-operative, and post-operative variables as listed in the STS National Cardiac Surgery Database (see appendix) were used. The continuous data were presented as mean±SD. The two groups were compared using univariate analysis (X2, Fischer's exact test) and Student's unpaired _t_test. LVEF≫30 was deemed the control group in all statistical analysis. The variables were analyzed and only those with _P-_values of ≦0.05 are discussed.

2.1 Surgical techniques

Off-pump coronary artery bypass operation was carried out through a full sternotomy incision with or without taking down of the left internal mammary artery in the usual fashion. Three deep pericardial traction stitches were placed near the left upper and lower pulmonary veins and to the left of the inferior vena cava, thereby achieving vertical displacement of the apex of the heart. With perfectly placed stitches and aggressive traction, the apex of the heart should be elevated to approximately 90 degrees. To further assist in providing good presentation of the target arteries on the lateral and inferior aspect of the heart, the patients were placed in a gentle right decubitus Trendelenburg position. The stabilization of the target arteries in the early phase of the study has been accomplished by using the CTS tissue stabilizer (Cardiothoracic Systems Inc., Cupertino, CA) and more recently with the Octopus II stabilizer (Medtronic Inc., Minneapolis, MN). With the addition of suction capability on the Octopus II device, presentation and stabilization of the remote target arteries near the circumflex trunk were feasible. Two stay sutures of 4-0 prolene were placed proximal and distal to the area planned for arteriotomy. The artery was dissected out and the stabilizer was applied. After the suction has been applied, the arm of the stabilizer could be moved to get rid off the compression on the heart and augment the presentation of the anastomotic area. Arteriotomy was made prior to the application of gentle traction on the 4-0 prolene stay sutures.

Treatment of all patients in both groups had followed the standard care and processes from operation through discharge. This included the admission to the intensive care unit from the operating room with subsequent transfer to an intermediate care ward within 24 h or as dictated by the patient's clinical status. A hospital designed extubation protocol was followed and targeted for 4 h post return from the operative suite.

3 Results

Demographic comparison between the two groups is listed in Table 1 . Patients in LVEF ≦30 were older (70.2 vs. 66.5 years) and more female gender. There were more NYHA class IV patients (64 vs. 50%), history of CHF (13 vs. 2%) and CHF present at the time of surgery (22 vs. 4%) in the low EF group. Close review of the groups indicate that the clinical presentation and pharmacological management of low EF group represented greater symptoms related to depressed left ventricular function. The mean pre-operative left ventricular function was 25±5.2% in LVEF ≦30 and 56±10% in LVEF≫30 (P≪0.001).

Demographics

Pre-operative predicted risk was 6.4±5.5% in LVEF ≦30 and 2.7±4.5% in LVEF≫30 (P≪0.001). Most (≫95%) of the patients in both groups were elective status, and LVEF ≦30 patients had increased the incidence of redo (11 vs. 6%, P=0.2). (Table 1) Otherwise, there was no difference in the incident of co-morbidities such as diabetes (DM), hypertension (HTN), stroke (CVA), vascular disease (PVD), chronic obstructive lung disease (COPD), smoker or obesity. Eighty-four percent of the patients in LVEF≫30 had stable angina but only 69% in LVEF ≦30 (P=0.009) (Table 2) .

Preoperative co-morbidities

Table 2

Preoperative co-morbidities

Intra-operatively, left internal mammary have been used in more than 90% of the cases in each group. There were no significant differences measured in number of grafts per patient (2.7 vs. 2.8), the amount of blood lost, peak CK-MB, skin-to-skin time, into/out of the OR time (Table 3) . The use of intra aortic balloon pump (IABP) deserves more emphasis, the IABP was used more frequently in the low EF group as noted, 29 vs. 5% pre-op (P≪0.001), 8 vs. 1% intra-op (P=0.01) and 11 vs. 2% post-op (P=0.2). Higher utilization of pre-operative intra-aortic balloon pump represents attention to the preventative management of the compromised left ventricular patient. Recently, the use of pre-operative and intra-operative IABP has been decreased significantly.

Operative factors

Post-operatively, there was no difference in the number of patients that required re-operation for bleeding or graft occlusion. There was no incidence of neurological deficit, either from stroke or TIA in the low EF group. The high EF group had incidence of 1% TIA and 2% of stroke. The statistical analysis yields no statistical significance in neurological deficit between these two groups; and both results were in the range of previously reported incidence of neurological deficit in on-pump patients. The incidence of POMI, however, was less than 1% in both groups. The incidence of new renal failure was higher in LVEF ≦30 group (11 vs. 5%, P=0.08) while the incidence of new AF was similar (13 vs. 16%, P=0.6). Operative mortality in the low EF group was 4.4 and 1.8% in LVEF≫30 group (P=0.23). Given the clinical presentation of the low EF group, higher prediction risk, longer pre-operative stay, the length of ventilation (24 vs. 8 h, P=0.12) and surgery to discharge (8 vs. 6 days, P=0.02) is anticipated (Tables 4 and 5) .

Peri- and post-operative variables

Table 4

Peri- and post-operative variables

Post-operative morbidity & mortality

Table 5

Post-operative morbidity & mortality

4 Comment

Indication to perform the off-pump procedure during this 18-month review period (January 1, 1998 through June 30, 1999) was dependent on a physician or patient specific choice and not specifically directed by a set of inclusion or exclusion criteria. Although for the OPCAB approach there was preference towards non-calcified coronary vessels, non-intramuscular arteries and target arteries of ≧1.5 mm in size. Patients were less likely to be considered for the OPCAB approach when they presented with severe left ventricular hypertrophy. During that same time period, a total of 2303 coronary artery bypass surgeries were performed with the study subset (387 OPCAB patients) representing 16.8% of the total. One hundred and seventy-seven of the total coronary artery surgery population (7.6%) had ejection fractions of ≦30% and of these 45 had the OPCAB approach and 132 patients had the conventional on-pump surgical approach. A comparison between the on-pump and off-pump patients with EF of ≦30% was carried out and reported elsewhere [3].

Our previous observation comparing the off-pump and on-pump patients has shown that there was no difference in operative mortality or complications in the low or medium risk group. The high risk group (pre-operative predicted risk of 10% or higher according to the risk model of the STS NCSDB), operative mortality of the off-pump patients was 7.7 compared to 28.5% for the on-pump group [4]. Additionally, those off-pump patients with ejection fractions of ≦30% had a 4.4% operative mortality when compared to the 7.5% operative mortality of the on-pump population. Regression analysis carried out in this study also showed that after the effects of the variable CPB on outcomes was adjusted for gender, age, and predicted risk, the surgical risk associated with on-pump increased [3].

The limitation of this study is that it is non-randomized, comparing two similar surgical approaches to cardiac revascularization and a small cohort of patients that may not lead to meaningful statistical comparison. All of the patients included in this study, however, were consecutive and selected solely as a preference of each surgeon.

5 Conclusion

Short-term clinical outcomes for both groups of OPCAB patients encouraged us to continue to offer this approach to this broad base of patients. In spite of compromised heart function and co-existing physiological risk factors associated with low ejection fraction, the mortality and morbidity results are equivalent to or less than the nationally reported norms. This confirms our belief that OPCAB could be a viable alternative to conventional coronary artery bypass surgery particularly for patients with impaired left ventricular function. It is imperative that continued review and evaluation of off-pump results be monitored. Future long-term follow-up results comparing re-intervention rates, patency and functional status of on-pump to off-pump must be carried out. Although the off-pump procedure is gaining widespread acceptance, it is a procedure that warrants a skilled-team approach to achieve quality outcomes, especially on the cardiac compromised patient.

Pre-operative data: gender, age, predicted risk, ejection fraction, New York Heart Association class IV, status, incident, angina (stable or unstable), smoker, diabetes, chronic obstructive lung disease, morbid obesity, peripheral vascular disease, hypertension, congestive heart failure (history and present), and cardiovascular accident.

Intra-operative data: operation room skin-to-skin times, operation room in-room to out-of-room times, estimated blood loss in the operating room, estimated blood loss in first 24 h, creatine phosphokinase peak and myocardial band enzymes, paced in the first 24 h, intra-aortic balloon usage (pre, intra, and post insertion), intubation, and intensive care unit times.

Post-operative data: re-operation for bleed and graft occlusion, deep sternal wound infection, permanent stroke, transient ischemic attack, peri-operative myocardial infarction, thoracentesis, new renal failure and atrial fibrillation, return to intensive care, operative mortality, and readmission to the hospital within 30-days of discharge.

Dr L.F. Sridhar (Chennai, India): The real problem comes intraoperatively when you try to displace these patients with a low ejection fraction to look at the posterior vessels. I mean, you haven't talked much about that. Do you encounter that at all and how do you circumvent it?

Dr Arom: Was the question about the posterior vessels correct?

Dr Sridhar: The real problem in patients with a low ejection fraction dilated out is when you try to displace and look at the posterior vessels and graft them. You haven't talked much about the technical aspect of it. Do you encounter it at all and how do you go about it?

Dr Arom: I agree with you that it is most difficult to do the posteriolateral vessels such as the OMs and posteriolateral branches of the right. Displacement of the heart until the apex of the heart is in the vertical position before starting the anastomosis is essential, and is even more important in the low ejection fraction group. We have learned that the heart has to be up to 90 degrees, which is done by placing three deep pericardial traction stitches, and observed until there is no hemodynamic instability before placing the stabilizer. Because the stabilizer has suction capability, we don't have to compress the heart all the time. When hemodynamic instability occurs after placing the stabilizer device, you can then grab the arm of the stabilizer and pull it up. By doing this, you minimize the compression on the heart and improve the hemodynamics. It is amazing to see the cardiac output come back to near normal, which this allows us to complete the anastomosis.

Dr F. Wellens (Aalst, Belgium): There is quite a lot of pressure from the industry to use all kind of circulatory support systems in OPCAB and mainly in this cohort of patients. What is the philosophy of your team?

Dr Arom: My philosophy is that since the revisit of OPCAB is to do cost effective surgery; in other words, do surgery as cheap as possible without compromising the outcomes. With this in mind, I do not believe that I need to use the right ventricular assist device or right heart bypass even in people with this low EF. Our late results have confirmed that the number of grafts per sites are comparable with the on-pump patients. Industry doesn't like me when I say this, but I feel strongly that if I have to put the patient on the pump, I will use total cardiopulmonary bypass. And if I have to use an apical suction device for better exposure, which has not happened to me yet, I would rearrange my deep pericardial traction stitches first. It does not make sense to me at all to try to do OPCAB and then spend more money for additional devices.

Dr A. Youhana (Swansea, UK): I totally agree with you. OPCAB came about just to get rid of the bypass and the sort of inflammatory response. I think if we go back again to partial heparinization, then the idea of OPCAB is taken out.

But what do you think about using the intra-aortic balloon pump routinely for this type of patients? Do you think that may improve the outcome of these poor ventricles, just like the cardiologists use it for their angioplasties?

Dr Arom: I think we did put the balloon in more than we should. Intra-operatively I put the balloon in sometimes for hemodynamic instability during surgery, but not for poor LV function. And there were times I was not sure that was necessary. Going back and looking at our practice, I believe we overused the IABP. We put the balloon in less now. My answer to your question is that my partners and I who do this off-pump surgery use less IABP, even in the low EF group.

References

[1]

Risk factors in coronary artery bypass surgery

,

Arch Surg

,

1972

, vol.

105

6

(pg.

918

-

923

)

[2]

Comparison of medical and surgical treatment for unstable angina pectoris: results of a Veterans Administration Cooperative Study

,

N Engl J Med

,

1987

, vol.

316

16

(pg.

977

-

984

)

[3]

Is low ejection fraction safe for off-pump coronary artery bypass operation?

,

Ann Thorac Surg

,

2000

, vol.

70

(pg.

1021

-

1025

)

[4]

Safety and efficacy of off-pump coronary artery bypass grafting

,

Ann Thorac Surg

,

2000

, vol.

69

(pg.

704

-

710

)

[5]

Emery RW, Arom KV, Flavi TF, Kshettry VR. Minimally invasive cardiac surgery: first thousand cases. Presented to IMSICS Annual Meeting, June 8–10, 2000 in Atlanta, GA, USA.

© Elsevier Science B.V. 2001