|
|
||||||||
a The University of Texas Medical School and Health Science Center, Houston, Texas
b Providence Health System, Portland, Oregon
Accepted for publication September 15, 2009.
* Address correspondence to Dr Letsou, 6410 Fannin St, Ste 450, Houston, TX 77030 (Email: george.v.letsou{at}uth.tmc.edu).
| Abstract |
|---|
|
|
|---|
Methods: During OPCAB, we selectively shunted the left anterior descending coronary artery (LAD) to provide myocardial protection when ischemia developed. We prospectively gathered data from 408 consecutive patients who underwent OPCAB. Data on shunt status were available for 386 of these patients. A "flow-through" shunt providing perfusion to the distal LAD was used whenever ST-segment elevations greater than 2 mm occurred or when patients had hemodynamic compromise unexplained by cardiac manipulation. Ejection fraction was assessed 1 to 3 days preoperatively and again 3 to 10 days postoperatively. Creatine kinase (muscle/brain) levels were assessed postoperatively for 24 hours.
Results: During OPCAB, 99 patients required shunting for presumed ischemia. Thirty-day cardiac mortality was 1.4% (4 of 296 without a shunt; 0 of 90 with a shunt). In patients without a LAD shunt, the mean peak creatine kinase index was 5.3; in patients who needed a shunt, the index was 5.2. Mean ejection fraction improved from 0.536 to 0.589 in the shunted patients and from 0.53.5 to 0.586 in the nonshunted patients.
Conclusions: Selective shunting of the LAD during OPCAB provides effective protection against myocardial ischemia when ischemia is detected by electrocardiogram or when hypotension develops that is unexplained by cardiac manipulation.
| Introduction |
|---|
|
|
|---|
We examined the effectiveness of selectively using "flow-through" shunts in the left anterior descending (LAD) coronary artery. In our series of 408 consecutive patients who underwent OPCAB, the LAD was shunted selectively when hypotension unrelated to cardiac positioning developed or when electrocardiographic (ECG) signs of myocardial ischemia were noted. We then analyzed postoperative release of cardiac enzymes and postoperative ejection fraction to detect any differences in myocardial performance and protection.
| Patients and Methods |
|---|
|
|
|---|
Patients
We reviewed prospectively collected observational demographic and operative data on coronary revascularization procedures performed by a single surgeon between August 2002 and April 2007. During the study period, 507 patients underwent CAB procedures; 408 underwent OPCAB, and 99 underwent coronary revascularization on the beating heart with cardiopulmonary bypass (CPB) support (BH-CAB). Of the 408 OPCAB patients, data on LAD shunt status were available for 386; these patients constituted the study group for this report.
All coronary revascularizations were begun with the intent to perform OPCAB. Emergency cases and redo operations were included in the database and in the analysis. During this period, 99 patients required additional CPB support for hemodynamic instability or evidence of ischemia (BH-CAB), Complete data were available for 81 of these patients, which were analyzed separately. No patient underwent CAB with cardiac arrest or aortic cross-clamping during this period; all underwent either OPCAB or BH-CAB. Excluded were 16 patients who underwent CAB but did not require revascularization of the LAD and 21 patients who underwent concomitant procedures, such as valve replacement.
Surgical Procedure and Data Collection
Whenever possible, surgical revascularization was done with a left internal mammary artery (LIMA) to LAD anastomosis. Such anastomoses were performed in 267 cases (96%). In the remaining 11, the LIMA was not available, and the right internal mammary artery (RIMA) was used in 6, a radial artery graft was used in 3 (all redo operations), and a saphenous vein graft was used in 2 (also both redo operations). All coronary arteries greater than 1.0 mm in diameter were grafted. CPB was used when necessary for safe exposure of the circumflex system.
The LAD was always the first vessel to be revascularized. A flow-through shunt of similar size (1.0, 1.5, 2.0, 2.5, or 3.0 cm in diameter) to the native coronary artery was used whenever myocardial ischemia was noted on ECG or when hypotension developed that was not due to myocardial positioning. If ECG changes occurred (ie, ST elevations of any significance in the V2 and AVF leads monitored during the operation), a flow shunt was promptly inserted. If mean blood pressure fell below 50 mm Hg and could not be increased by repositioning the heart to allow improved filling, volume loading to central venous pressure exceeding 10 mm Hg, pressor support with low-dose epinephrine (<0.05 µg/kg/min), or any combination of these, a flow-through shunt was also inserted. If mean blood pressure did not increase to more than 50 mm Hg after these interventions, CPB was instituted.
We used both Medtronic (Minneapolis, MN) and Maquet Cardiovascular (Boston Scientific, Natick, MA) shunts. Shunts were removed just before the anastomosis was completed.
The right coronary artery system and then the circumflex system were revascularized after the LAD. We routinely shunted the right coronary artery and selectively shunted posterior descending arteries, posterolateral arteries, and obtuse marginal arteries when hypotension developed that was unexplained by myocardial positioning or when ECG changes occurred. Shunting of the posterior descending artery or circumflex marginal vessels was necessary in less than 5% of cases. CPB was instituted if myocardial shunting did not lead to prompt resolution of ECG changes or restoration of mean arterial blood pressure to greater than 50 mm Hg.
Creatine kinase levels were obtained for 368 of the 386 OPCAB patients (95%) and for 81 of the 83 BH-CAB patients (98%) every 8 hours for 24 hours postoperatively. Creatine kinase (muscle/brain; CK-MB) indices were calculated.
Ejection fraction (EF) was obtained for 376 of the 386 OPCAB patients (97%) and for all 81 (100%) of the BH-CAB patients. EF was assessed preoperatively and postoperatively by using nuclear scanning, whenever possible. If nuclear scanning was not possible, EF was assessed echocardiographically. Nuclear scans were obtained preoperatively and within 3 to 10 days of operation in 91% of the OPCAB patients (351 of 386) and in 100% of the BH-CAB patients (81 of 81).
Statistical Analyses
Several variables were compared between shunted and nonshunted patients. Mann-Whitney U tests were used for statistical analysis of serum CK-MB indices and improvement in EF. Wilcoxon signed ranks tests and locally weighted scatterplot smoothing (LOWESS) curves were used for the preoperative and postoperative EF. Statistical analysis was performed by using SPSS 11.5 (SPSS Inc, Chicago, IL) and SPLUS 6.2 (Insightful Corp, Seattle, WA) software.
| Results |
|---|
|
|
|---|
|
|
|
|
|
BH-CAB was required in 83 patients. No CK-MB index data were collected for 2 patients, leaving 81 patients' data for analysis. Of these, the coronary revascularization procedure in 49 was initiated with CPB (ie, their procedures started out as BH-CAB) because of evidence of ongoing ischemia as manifested by chest pain, ECG changes, or hemodynamic instability on presentation to the operating room. None of these patients were subsequently shunted. During revascularization, 32 of the 81 patients required conversion from OPCAB to BH-CAB because of hemodynamic instability (mean blood pressure <50 mm Hg). Only five of the conversions occurred after insertion of a LAD shunt; 29 were done to permit full access to the circumflex system, which can be difficult to expose during OPCAB when left ventricular hypertrophy is present. The mean CK-MB index was not significantly different between shunted and nonshunted BH-CAB patients (Table 4).
|
| Comment |
|---|
|
|
|---|
Multiple observational reports document the presumed benefit of OPCAB in selected patient groups [4, 8]. Although such benefits are controversial, OPCAB may reduce mortality and morbidity rates in patients who are at high risk of renal failure, stroke, and, possibly, myocardial infarction [8]. Despite the continuing controversy, approximately 20% to 30% of all coronary artery operations are being performed off-pump [1]. Thus, OPCAB seems to be here to stay, which makes the question of how best to perform OPCAB important. The substantial body of literature concerning OPCAB lacks data concerning optimal techniques for the procedure. Should proximal anastomoses be performed before distal anastomoses? What are the optimal positions for myocardial stabilization? The question we addressed this article was whether LAD shunting is beneficial. This question has been addressed surprisingly infrequently [2].
In the 283 patients who did not have a LAD shunt, the mean peak CK-MB release of 5.3 was similar to the 5.2 in the 85 patients who required LAD shunting. (Unfortunately, our institution switched from measuring troponin I to troponin T midway through our study, obviating the value of analysis for either variable.) No significant difference in the shunted and nonshunted groups were noted in the 14 demographic variables recorded, including history of myocardial infarction (within the past 48 hours or the past 7 days) and the patients' operative priority (elective, urgent, or emergent/salvaged). Limited release of myocardial enzymes seems to follow all cardiac procedures, including angioplasty and traditional on-pump operations as well as OPCAB.
An interesting observation concerning OPCAB was that prolonged occlusion of the various vessels, and specifically the LAD in our study, did not seem to produce more cardiac enzyme release than angioplasty, which involves much shorter durations of coronary artery occlusion. One might theorize that the longer periods of coronary artery occlusion that involve snaring without shunting during OPCAB might lead to severe damage of the myocardium. However, this was not observed; indeed, none of our patients had significant myocardial damage that could be attributed to the shunt or to the nonshunting technique. Our patients' average CK-MB level was similar to levels observed in OPCAB patients by other authors and to levels recorded during on-pump operations and after angioplasty [9].
Thus, across multiple studies, the extent of myocardial injury seems to be similar during angioplasty, on-pump procedures, or OPCAB—with or without shunting. Flow from collaterals supplied by the circumflex and right coronary artery systems probably accounts for the lack of ischemic damage seen in OPCAB patients when no shunt is used. Only a short segment of the vessel is occluded in OPCAB; the portion of the vessel distal to the distal occluding snare is patent and perfused by collaterals that have formed downstream from the obstruction being bypassed.
Our study documents a significant improvement in EF both in patients who had a LAD shunt and in patients who did not require a LAD shunt. The magnitude of improvement in EF was not significantly different between these two groups of patients. Therefore, a policy of selective LAD shunting provides effective myocardial protection when the patient has hypotension unexplained by cardiac positioning or ECG changes consistent with myocardial ischemia. The collection of data on EF within 3 to 10 days after revascularization permits conclusions about the short-term effects of OPCAB. Because the long-term outcomes after OPCAB may not differ significantly from those of on-pump procedures, such short-term improvements in EF may hold the key to understanding the differences between OPCAB and traditional on-pump operations with ischemic arrest.
There are many issues to consider surrounding the use of intraluminal coronary artery shunts. Advocates of shunting cite several reasons to shunt routinely: improved myocardial protection, ease of use, and technical improvements in the anastomosis [10–13]. Advocates for performing coronary artery occlusion (with snares proximal and distal to the anastomotic site) without shunting cite several reasons to avoid shunting, including vessel endothelial damage caused by shunts, excellent clinical results without shunting, and the simplicity of the surgical technique [14–17]. These differing opinions have resulted in the mélange of techniques being used currently.
Other authors have observed that routine shunting of the LAD in all patients undergoing OPCAB leads to reduced myocardial enzyme release, that is, lower CK-MB isoenzyme and troponin levels after routine shunting [18]. In contrast, we found no difference in cardiac enzyme indices between our shunted and nonshunted patients. One possible explanation is that the patients in our study who did not undergo LAD shunt placement were those who did not manifest any signs of ischemia or hypotension and thus, presumably, had sufficient collateral flow to the distal LAD to prevent ischemia. In the other studies, all patients were shunted. The beneficial effects of LAD shunting in these other studies may have been confined to those patients in our study who were shunted (ie, those who manifested signs of cardiac ischemia). Our use of surface ECG to assess myocardial ischemia is not based on empiric evidence but seemed to be of benefit in our study and may be analogous to the use of electroencephalographic monitoring during carotid artery procedures. Further investigation into the most sensitive techniques for the detection of intraoperative myocardial ischemia would be beneficial.
The EF has been shown to improve after on-pump coronary revascularization involving hypothermic cardiac arrest [19, 20]; however, improvements in EF after OPCAB have not been specifically addressed. A study using magnetic resonance imaging showed no improvement in EF at 12 months postoperatively, although this was not a central finding of the study [21]. In our own study, EF improved after OPCAB whether or not LAD shunting was used. Similar improvement was documented in both groups when postoperative EF was compared with preoperative EF. Thus, our study supports the conclusion that a policy of selective shunting of the LAD does not influence myocardial function early after the operation. None of these studies, including our own, was randomized; a prospective, randomized trial of shunting would be of clear value. Nevertheless, the results of these observational studies suggest that myocardial preservation is as effective with shunting as it is without shunting when shunts are used in response to signs of myocardial ischemia.
The lack of improvement in EF noted by some investigators after OPCAB might be due to incomplete revascularization. Our policy of performing complete revascularization whenever possible, including revascularizing small vessels, may account for the improvements in EF seen in our patients.
We attempted to perform nuclear scanning both before and 3 to 10 days after the operation; however, nuclear scans were obtained in only 75% of patients. Multiple gated acquisition (MUGA) imaging is an objective and accurate method of measuring LVEF. We chose MUGA because its objectivity is particularly useful for minimizing interobserver variability and bias. When a MUGA image was not obtained, either because of patient refusal or oversight, echocardiography was performed. The correlation between EFs measured by echocardiographic and nuclear scanning is high but not perfect [22].
In conclusion, conversion from OPCAB to BH-CAB for LAD ischemia is uncommon when a policy of selective LAD shunting is used. Elevations in CK-MB levels are not significantly different from elevations seen after other cardiac procedures, such as angioplasty. Improvements in EF are of the same magnitude with and without shunting. Selective shunting is a safe and effective method of providing myocardial protection during OPCAB.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
30% Am J Cardiol 1997;79:1573-1578.[Medline]This article has been cited by other articles:
![]() |
G. V. Letsou, Y. X. Wu, G. Grunkemeier, M. M. Rampurwala, L. Kaiser, and A. L. Salaskar Off-pump coronary artery bypass and avoidance of hypothermic cardiac arrest improves early left ventricular function in patients with systolic dysfunction Eur J Cardiothorac Surg, July 1, 2011; 40(1): 227 - 232. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. P. Collison and Y. K. Mishra What is the Role of Intraluminal Shunts During Off-Pump Surgery? Ann. Thorac. Surg., October 1, 2010; 90(4): 1394 - 1394. [Full Text] [PDF] |
||||
![]() |
G. V. Letsou Reply Ann. Thorac. Surg., October 1, 2010; 90(4): 1394 - 1395. [Full Text] [PDF] |
||||
![]() |
G. V. Letsou Reply. Ann. Thorac. Surg., August 1, 2010; 90(2): 701 - 701. [Full Text] [PDF] |
||||
![]() |
M. P. Vallely and D. E. Ross Intracoronary Shunts and Off-Pump Surgery Ann. Thorac. Surg., August 1, 2010; 90(2): 700 - 701. [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 |