| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Eastern Maine Medical Center, Bangor, Maine
b The Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, New Hampshire
c Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
Accepted for publication July 11, 2007.
* Address correspondence to Dr Hernandez, Cardiothoracic Surgery, Eastern Maine Medical Center, 417 State St, Suite 421, Bangor, ME 04401 (Email: fhernandez{at}emh.org).
Presented at the Poster Session of the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
| Dr Hernandez discloses that he has a financial relationship with Genzyme.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods: We prospectively randomly assigned 201 patients undergoing nonemergent isolated coronary artery bypass graft surgery to conventional coronary artery bypass graft surgery (n = 102) or off-pump coronary artery bypass graft surgery (n = 99). The primary end points of the study were neurocognitive function assessed using a 19-test neurocognitive battery at baseline, discharge, and 6 months. Neurocognitive deficit was defined as a 20% or greater reduction from baseline in at least 20% of the tests. Secondary end points included index admission mortality, stroke, low-output cardiac failure, return to the operating room for bleeding, and postoperative troponin release. Risk ratios and 95% confidence intervals were calculated based on intention-to-treat analysis.
Results: There was no difference in neurocognitive deficit at discharge (discharge versus preoperative: risk ratio, 0.83; 95% confidence interval, 0.65 to 1.07) or at 6 months (6 months versus preoperative: risk ratio, 0.94; 95% confidence interval, 0.70 to 1.28). There was no significant difference in mortality or morbidity between the two groups. The off-pump coronary artery bypass graft group had fewer patients with troponin release than the conventional coronary artery bypass graft group.
Conclusions: Off-pump coronary artery bypass graft surgery did not result in decreased frequency of neurocognitive deficit. Off-pump coronary artery bypass graft surgery was associated with substantially lower levels of troponin release after surgery.
Coronary artery bypass graft (CABG) surgery plays a central role in the management of patients with ischemic heart disease [1, 2]. Despite improvements in surgical and anesthetic techniques, the conventional operation, performed with the use of cardiopulmonary bypass (CPB) and myocardial arrest, remains associated with an incidence of significant morbidity and mortality [3]. There has been substantial interest in the performance of on-pump versus off-pump CABG surgery with respect to improved outcomes [4].
Previous work, albeit predominantly retrospective in nature, has questioned the benefit, if any, to CABG performed without CPB (OPCAB), relative to conventional CABG (CCAB); as well as raised concern regarding graft patency and longer-term outcomes in patients undergoing OPCAB procedures [5–8]. The available prospective clinical data has been limited to studies both small in number and scope. Overall, OPCAB has been proven to be a safe alternative to CABG, but no clear advantage has yet been shown over CABG [9]. At most cardiac surgery centers, CCAB remains the standard of care.
Neurocognitive deficit has been reported in 40% to 70% of patients undergoing CCAB [10]. It is believed by many investigators that the neurocognitive changes seen after CCAB may be the area in which the largest potential advantage of OPCAB lies [11–13]. We designed a prospective, randomized study to compare the two techniques with respective to neurocognitive function and common secondary end points among 204 patients undergoing CABG.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
= 0.05; power = 0.80). We enrolled 10% more patients than our preliminary calculations determined. Inclusion criteria were as follows: patients aged 40 to 80 years requiring elective or urgent CABG surgery. Exclusion criteria were as follows: patients requiring concomitant valve or carotid artery procedures, emergency operation, reoperation, heavily calcified ascending aorta, deep intramyocardial left anterior descending coronary artery, preoperative inotropic agent (eg, dobutamine, epinephrine, amrinone, milrinone, or dopamine greater than 3 µg · kg–1
· min–1) or cardiac-assist device (intraaortic balloon pump) support for hemodynamic instability, or patients from whom written informed consent was not obtained. From January 2001 through January 2004, 102 patients were randomly assigned to CCAB and 99 to OPCAB.
Research Design
Patients who satisfied the inclusion criteria and consented to participate were randomly assigned to undergo CCAB or OPCAB operations. Randomization was conducted using a computer and sealed card system. The study coordinator notified the surgeon of the study number, and the surgeon opened the corresponding envelope. The surgeon, perfusion team, and operating room staff were informed of the treatment assignment immediately before the start of the case.
Anesthesia
All patients were medicated preoperatively according to standard protocol. Central venous, Swan-Ganz, and radial artery catheters were inserted before the induction of general anesthesia. Near-infrared spectroscopy was used to detect oxyhemoglobin desaturation (Somanetics Corporation, Troy, MI). Patches were applied before induction of general anesthesia. Cerebral oximetry measurements were maintained within the normal range (>92% saturated), when possible, by maintaining perfusion pressure, hematocrit, and oxygen saturation at optimal levels. Anesthesia was induced and maintained according to standard protocol. Hypotension was managed by the administration of intravenous fluids or vasopressor agents. Cardiac index was maintained (before and after CPB in the CABG group) at greater than 2.2 L · min–1
· m–2 by means of the administration of intravenous fluids or inotropic agents when appropriate.
Surgical Procedure
Conventional Coronary Artery Bypass Graft Group
Conventional CABG surgery was performed through a full median sternotomy incision. The left internal mammary artery was used routinely and taken down in standard fashion. Patients were anticoagulated before CPB with 300 IU/kg heparin sodium supplemented with additional doses as needed to maintain an activated clotting time of greater than 400 seconds during CPB. Cardiopulmonary bypass was instituted by cannulating the ascending aorta and right atrium with use of a membrane oxygenator and centrifugal pump. The left ventricle was vented through the right superior pulmonary vein. A flow rate of 2.4 L · min–1
· m–2 and mean arterial pressure of 60 to 80 mm Hg was targeted. Moderate hypothermia with core temperatures of 28° to 32°C was used. Cold-blood hyperkalemic cardioplegia (4:1 ratio) was used for myocardial arrest according to standard protocol using both antegrade and retrograde administration. Distal and proximal anastomoses were performed using continuous suturing technique. A cross-clamp was used for the distal anastomoses followed by a partial occlusion clamp for the proximal anastomoses under CPB. The acceptability of the grafts was assessed intraoperatively using an ultrasonic flowmeter. Heparin neutralization after termination of CPB was performed by the administration of protamine sulfate at a dose of 1 mg/100 IU of estimated active heparin. To ensure a return to preheparinization baseline, Hepcon (Medtronic, Minneapolis, MN) was monitored.
Off-Pump Coronary Artery Bypass Graft Group
Surgical incision, use of the left internal mammary artery, and performance of distal and proximal anastomoses were carried out in an identical fashion as for the CABG group. Patients were anticoagulated with 300 IU/kg heparin sodium before performing the distal coronary anastomoses and supplemented with additional doses as needed to maintain an activated clotting time of greater than 400 seconds until the proximal anastomoses were completed. Exposure of the target coronary arteries was achieved by means of the placement of deep pericardial traction sutures and table rotation. Stabilization of the target arteries was accomplished with either the Genzyme OPCAB Elite (Genzime Surgical, Fall River, MA) or Medtronic Octopus system (Medtronic, Minneapolis, MN) depending on surgeon preference. Proximal anastomoses were performed after the distal anastomoses using a partial occlusion clamp. Heparin neutralization was accomplished after measuring flows in the grafts using the same reversal dose schedule of protamine sulfate used in the CCAB group.
Postoperative Management
Treatment of patients in both groups followed identical standardized protocols. Intraoperative need for using vasoactive agents or cardiac-assist devices were determined cooperatively between the surgeon and anesthesiologist. Intravenous fluids and vasoactive agents were administered and weaned in the intensive care unit according to standardized protocols designed to maintain designated hemodynamic variables. Protocols for the administration of blood products (packed red blood cells, fresh-frozen plasma, platelets, and cryoprecipitate) were used. Additional criteria for weaning patients from the ventilator, removal of central catheters and chest tubes, and transfer from the intensive care unit to an intermediate-level cardiac ward with subsequent care until the time of discharge were standardized.
Evaluation of Efficacy
Primary end points: Neurocognitive analysis
Neurocognitive testing was completed on the day before surgery (baseline), on the day of discharge, and at 6 months postoperatively. Nine neuropsychological tests were administered spanning six neurocognitive domains. Altogether, the tests yielded nineteen scores for analysis. Z scores are reported for each test at each time with the mean and standard deviation. Conventional CABG surgery and OPCAB surgery were compared with respect to the absolute change in z score from both postoperative scores relative to the preoperative scores. The 19-test neurocognitive battery included Trails Digit Span, VIGIL (hit rate, omissions, and commissions), grooved pegboard, Rey-Osterrieth complex, controlled oral word association, Hopkins verbal learning, oral reading test WRAT-3 (Wide Range Achievement Test, 3rd edition), Brixton Spatial Anticipation Test, and Beck Depression Inventory and Spielberger State-Trait Anxiety Inventory scores [14]. We measured mood (Beck Depression Inventory and Spielberger State-Trait Anxiety Inventory scores) as a potential confounder of any cognitive changes. The neurocognitive battery was administered by a trained psychometrist, blinded to patient treatment assignment.
Secondary end points
Mortality was defined as the index admission mortality at patient discharge. Stroke was defined as a cerebrovascular accident documented by a physician and defined by the following: new focal neurologic deficit that appears and is still at least partially evident more than 24 hours after its onset, occurring during or after the CABG procedure, and established before discharge. Low-output cardiac failure was defined as the need for intraoperative or postoperative intraaortic balloon pump insertion or the need for inotropic agents to maintain cardiac index greater than 2.1 L · min–1
· m–2 at 4 hours after the operation. Return to the operating room for bleeding was defined for any cause. Cardiac troponin T leak was defined as 24-hour postoperative cardiac troponin T greater than 0.2 ng/mL.
Postoperative support requirements included the need for and duration of inotropic or intraaortic balloon pump support to maintain cardiac output, need for pressor support to maintain mean arterial pressure, duration of dependence on mechanical ventilation, length of intensive care unit stay, and length of hospital stay. Morbidities included troponin greater than 0.2 ng/mL, mediastinitis or deep sternal wound infection, renal insufficiency or hemodialysis, pulmonary complications (acute respiratory distress syndrome, pneumonia, or respiratory failure), postoperative bleeding and transfusion requirements (chest tube output), mediastinal reexploration for bleeding, perioperative (packed red blood cell and clotting factor) transfusion requirements, and postoperative arrhythmias (treated supraventricular and ventricular arrhythmias).
Statistical Analysis
Neurocognitive deficit was defined as a 20% or greater decline in at least 20% of the neurocognitive tests [13, 15, 16]. Z scores were calculated from the age, hand dominance, and gender-adjusted norms using a standard set of norms and calculated in the traditional method (raw test score minus group mean divided by the standard deviation) [11, 17]. Probability values were generated from Student’s t test comparing the change in z score from the postoperative value relative to baseline. Probability values were also generated using the Wilcoxon rank-sum test to confirm significance using nonparametric methods; both tests were in agreement.
Statistical analysis was performed using STATA 9.0 software (STATA Corporation, College Station, TX). We compared baseline data between two groups with the following specifications to assess randomization. The specific variables that were assessed are listed in Table 1. Categorical variables were analyzed using the Mantel-Haenszel 
2 test except for small cell sizes, in which the Fisher’s exact test was used. Continuous data were analyzed using the two-sample Student’s t test. Statistical significance was defined at a probability value of less than 0.05.
|
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
There were no differences between the groups in patient age, percent female, body mass index, estimated IQ, comorbidities, ejection fraction, number of diseased vessels, or severity of coronary artery disease. The CCAB group had a higher proportion of patients who underwent urgent surgery (CCAB, 69.6%; versus OPCAB, 54.6%). There were no differences in the number of distal anastomoses, number of distal anastomoses per number of diseased vessels, or frequency of internal mammary artery use between the two groups. Eight patients were crossed over to CCAB by the surgeon because of an intramyocardial left anterior descending coronary artery.
Overall cognitive decline (
20% decline in 20% of the tests; Table 2) was not statistically different between CCAB (61.8%) and OPCAB (51.5%) groups at 4 days postoperatively (risk ratio [RR], 0.83; 95% confidence interval [CI], 0.65 to 1.07) or at 6 months (CCAB, 47.1%; OPCAB, 44.4%; RR, 0.94; 95% CI, 0.70 to 1.28).
|
Table 3 shows neurocognitive test performance across times and procedures. The probability values at discharge and 6 months reflect the difference in the change in z score between CCAB and OPCAB. Two tests, the Trails A (psychomotor) and the State anxiety (mood) tests, demonstrated OPCAB patients had a significant decline in neurocognitive function at discharge compared with CCAB patients. The State anxiety test showed CCAB patients on average improved in mood, whereas OPCAB patients’ mood declined at discharge. One test (Brixton spatial anticipation test) showed a statistically significant difference at 6 months. At 6 months, the Brixton demonstrated that CCAB patients showed improved function, whereas OPCAB patients declined relative to baseline; this may be related to a slight learning effect. All other tests showed no significant difference between CCAB and OPCAB patients with respect to neurocognitive function at discharge and 6 months after the procedure.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In our initial series, we compared our results with 332 OPCAB patients and compared risk factors and outcomes with 445 patients who underwent CCAB [18]. We did see a lower incidence of perioperative myocardial infarctions (0.3% OPCAB versus 7.2% CCAB) and postoperative low-output cardiac failure (0.60% OPCAB versus 3.15% CCAB) in the OPCAB group. We confirmed this result on a regional level, showing OPCAB did not result in worse morbidity or mortality and was associated with lower incidence of postoperative intraaortic balloon pump use [9]. We later reported a statistically significant reduction in myocardial necrosis evidenced by 24-hour postoperative cardiac troponin T leak after OPCAB compared with CCAB [19].
van Dijk and associates [20] reported the results of a randomized trial comparing neurocognitive outcomes for patients undergoing OPCAB versus CCAB in 2002. Neurocognitive outcomes were not significantly different at 3 months (RR, 0.65; 95% CI, 0.36 to 1.16) or at 1 year (RR, 0.88; 95% CI, 0.52 to 1.49) between the two groups. The computed standard change score showed a significant improvement for OPCAB compared with CCAB at 3 months (p = 0.03), but not at 1 year (p = 0.09). In a second report in 2004, van Dijk and colleagues [21] reported a 49% incidence of neurocognitive decline in OPCAB patients and a 57% incidence in CCAB patients at 4 days after surgery. These numbers are very similar to our observations (51.5% OPCAB versus 61.8% CCAB). They concluded that they found no significant effect of CPB on early neurocognitive outcome. In their latest report, van Dijk and associates [22] extended their observations to 5 years postoperatively with the same conclusion. After applying a more conservative definition of neurocognitive decline that attempted to control for the normal variation in cognitive performance observed in a short period in a control population of healthy volunteers, their conclusion was unchanged. Lee and coworkers [23] demonstrated that OPCAB patients experience improvement in verbal learning at 2 weeks and 1 year, but CCAB patients did not experience this improvement. The authors concluded that OPCAB may result in improved cognitive function compared with CCAB. Other nonrandomized trials reported cognitive benefits and quality of life associated with OPCAB [24]. Jensen and colleagues [25], in a substudy of a randomized trial, found no difference in postoperative cognitive function between CCAB and OPCAB patients. Our trial agrees with those of van Dijk and associates [20–22] that there is no significant difference between CCAB and OPCAB patients with respect to cognitive decline. Our data support the notion that the effect of CPB on postoperative neurocognitive decline may not be as significant as previously thought, and other factors, such as anesthesia or the major inflammatory response associated with major surgical procedures, may play a role.
Recently, the American Heart Association released a scientific statement on the practice of CCAB and OPCAB and a summary of clinical evidence [26]. They report that there is a general trend toward better myocardial protection and less cognitive dysfunction when using OPCAB. Our data demonstrate the former but do not show statistically significant improvement in the frequency of neurocognitive decline in patients undergoing OPCAB surgery. This trend has been reported elsewhere [19, 27], and suggests the need for a directional improvement in myocardial protection when using CPB.
Our study has limitations. First, it is not fully blinded. Only the neurocognitive testing technician and psychologists were unaware of the operative procedure that had been performed. In addition, the sample size calculation was based on achieving a one-third reduction in cognitive dysfunction in the OPCAB group. Therefore, a more-modest benefit cannot be excluded based on our data.
There was no statistically significant difference in neurocognitive decline at discharge or at 6 months between CCAB and OPCAB procedures. There were 1 death and three strokes in the CCAB group and none in the OPCAB group. There was no significant difference in low-output cardiac failure or return to the operating room for bleeding. However, there were significantly more patients with troponin release grater than 0.2 ng/mL in the CCAB group as opposed to the OPCAB group. Further work is needed to elucidate the relationship between the mechanisms producing these injuries and the processes of care.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Takagi, N. Kawai, and T. Umemoto Neurocognitive decline after off-pump versus on-pump coronary artery bypass. Ann. Thorac. Surg., August 1, 2008; 86(2): 690 - 690. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
