| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;64:1747-1752
© 1997 The Society of Thoracic Surgeons
Department of Cardiothoracic Surgery and Section of Cardiology, The Boston Medical Center, Boston, Massachusetts
Accepted for publication June 18, 1997.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
Methods. From 1981 through 1995, 117 patients at our institution underwent emergency coronary artery bypass grafting after failed percutaneous transluminal coronary angioplasty, with an in-hospital mortality rate of 13.6%. Univariate and multivariate analyses were used to identify the factors that influenced the risk of death.
Results. Univariate analysis revealed that patients who died more often were women and had chronic renal failure, lower ejection fractions, and more diffuse coronary artery disease; less often received an internal mammary artery graft or an antegrade perfusion catheter; required inotropic support in the cardiac catheterization laboratory; and experienced myocardial infarction. Multivariate analysis demonstrated that the need for inotropic support in the cardiac catheterization laboratory was the best predictor of perioperative death.
Conclusions. Patients with a reduced ejection fraction in whom percutaneous transluminal coronary angioplasty fails, antegrade perfusion does not produce a response, and myocardial infarction occurs are more likely to die after coronary artery bypass grafting. The risk appears to be highest for patients who require inotropic support in the cardiac catheterization laboratory.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
This study was undertaken to determine the factors that influence the risk of death after emergency CABG for failed PTCA. We hoped to identify the preoperative risk profiles, coronary anatomy, catheterization interventions, and surgical techniques that were most likely to result in a surgical death. These data then could be used to develop new strategies for selecting patients for PTCA and managing them more effectively when emergency CABG is necessary.
|
Material and Methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Acute ischemia during PTCA was evaluated immediately by coronary angiography to determine whether the vessel was occluded or dissected. In the event of a coronary occlusion, attempts were made to redilate the vessel or to maintain patency using a reperfusion catheter. "Stenting" of the coronary artery, now a common practice at our institution, was not performed in this series of patients. Myocardial ischemia was documented using 12-lead electrocardiograms. Patients with hemodynamic instability and evidence of ischemia received an intraaortic balloon pump (IABP). Inotropic agents were used to maintain a systolic blood pressure of 90 mm Hg or greater. Percutaneous bypass was instituted in the event of cardiopulmonary arrest and to maintain a systolic blood pressure of 90 mm Hg or greater when the IABP and inotropic support were unsuccessful.
Patients were transported rapidly to the operating room, cannulated, and placed on cardiopulmonary bypass. The decision to use an internal mammary artery (IMA) graft was made by the individual surgeon on the basis of the hemodynamic stability of the patient, the coronary anatomy, and the segmental wall motion of the region to be grafted. All vessels with a lesion of 50% or greater were grafted. Distal anastomoses were performed during one period of ischemic arrest. The first anastomosis was performed to the vessel involved during the PTCA. Myocardial protection was provided with antegrade crystalloid, antegrade blood, and antegrade/retrograde blood cardioplegic techniques, and it varied according to the preference of the individual surgeon.
A perioperative myocardial infarction (MI) was diagnosed by the appearance of new Q waves and by elevation of the myocardial fraction of creatine kinase (CK-MB) to greater than 50 IU in the immediate 24-hour period after operation. Mortality was defined as any in-hospital death or death within 30 days of CABG.
Univariate comparisons between survivors and nonsurvivors were performed using the Fisher exact test or 
2 analysis for discrete variables and the unpaired Student's t test for continuous data. All variables found to be of significance (p < 0.05) or marginal significance (p < 0.10) with univariate testing were entered into a multivariate analysis using forward stepwise regression analysis to assess the independent contributions of multiple variables on operative mortality.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
and Figure 1. From 1981 through 1995, 117 patients at our institution underwent emergency CABG after failed PTCA, with an in-hospital mortality rate of 13.6% (16/117). Patients were divided further into three groups on the basis of the year their PTCA was performed (1981–1985, 1986–1989, and 1990–1995) (see Table 5). The mortality rate remained constant during these three periods.
|
|
|
|
|
|
|
. There was no significant difference in age or the incidence of diabetes, hypertension, peripheral vascular disease, or unstable angina between the two groups. Patients who died were more likely to be women (26/101 versus 9/16; p < 0.02), and they had a higher incidence of chronic renal dysfunction (creatinine 
1.8 mg/dL; 0/101 versus 8/16; p < 0.0001). Although ejection fraction tended to decrease in all patients who underwent PTCA from 1981 through 1995 (Table 1), patients who died had significantly lower ejection fractions (0.54 ± 0.01 versus 44 ± 0.02; p < 0.03).
Table 2 summarizes the coronary anatomy and angioplasty outcomes. Although the incidence of attempted multivessel PTCA was similar in survivors and nonsurvivors, patients who died tended to have more vessels with 70% or greater stenosis (1.74 ± 0.08 versus 2.18 ± 0.08; p < 0.05). Patients who died also were more likely to have had a total occlusion of the angioplastied vessel (44/101 versus 11/16; p < 0.05) as opposed to a dissection or a combination of an occlusion and a dissection. The most common culprit vessel involved was the left anterior descending artery. The distribution of culprit vessels was similar in both groups. Patients who died were more likely to have had electrocardiographic changes after PTCA (43/101 versus 14/16; p < 0.001), and to have required inotropic support (11/101 versus 12/16; p < 0.0001). Although the incidence of preoperative IABP placement was similar in both groups, patients who died were less likely to have had a reperfusion catheter inserted across the culprit lesion (33/101 versus 0/16; p < 0.01). The use of a reperfusion catheter did not appear to alter the incidence of perioperative MI from 1981 to 1995; however, there appeared to be a decrease in the use of the IABP as the use of reperfusion catheters increased (see Table 5; Fig 1).
Table 3 reviews the operative data. The mean number of vessels bypassed, the duration of cross-clamping, and the type and delivery of cardioplegia was similar in both groups. The use of an IMA graft increased during the various periods (see Table 5). Nevertheless, patients who died were less likely to have received an IMA graft (44/101 versus 1/16; p < 0.05).
Clinical outcomes are summarized in Table 4. Patients who died had a significantly higher incidence of MI (25/101 versus 15/16; p < 0.001). Although the incidence of reoperation for bleeding, respiratory insufficiency, sternal infection, cerebrovascular accident, sepsis, and intestinal ischemia was not statistically different between the two groups, the overall incidence of complications was statistically higher in patients who died (14/101 versus 10/16; p < 0.02). Patients who died also were more likely to have required postoperative hemodialysis (10/101 versus 5/16; p < 0.05), although 2 of these patients already were receiving long-term hemodialysis before operation.
Most patient deaths were related to cardiac complications (11/16). Three patients died after episodes of ventricular tachycardia that degenerated into ventricular fibrillation that was unresponsive to electrical cardioversion. All 3 had sustained large MIs. Eight other patients died as a result of low output syndrome. Two patients had percutaneous bypass devices inserted; both died. Other causes of death included sepsis (2 patients), intestinal ischemia (1 patient), liver failure (1 patient), and massive cerebrovascular accident (1 patient).
All 8 patients who had a preoperative creatinine level of 1.8 mg/dL or higher died. Two of these patients had been receiving preoperative long-term hemodialysis, and 4 required preoperative inotropic support. Of these 8 patients, 2 died of sepsis, 1 of liver failure, and 5 of low cardiac output syndromes. Multivariate analysis using stepwise regression analysis showed that the best predictor of perioperative death after failed PTCA was the need for preoperative inotropic support (p = 0.01) (see Table 6).
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
In this, as in other series, patients with multivessel disease were more likely to die after CABG performed for unsuccessful PTCA [1, 5–7]. Wang and colleagues [1] noted that the operative mortality rate increased from 4% to 21% in patients with multivessel disease after failed PTCA. Furthermore, shock developed in twice as many patients with multivessel disease after PTCA. In contrast, preoperative cardiogenic shock had no effect on the mortality rate in patients with single-vessel disease. Using multivariate analysis in a series of patients in whom acute coronary closure developed after failed PTCA, Ellis and co-workers [6] found that the presence of multivessel disease was an independent predictor of death. They suggested that the presence of jeopardized collaterals resulting from multivessel disease makes this group of patients more susceptible to low output syndromes and extensive infarctions after unsuccessful PTCA. Greene and associates [7] also noted that patients with multivessel disease had a significantly higher need for IABP support and antiarrhythmic therapy than did patients with single-vessel disease.
In our study, the mortality rate was doubled for women. Ellis and colleagues [6] also found female sex to be an independent predictor of death after unsuccessful PTCA. Although it is unclear why women have a higher mortality rate, age, higher risk profiles, smaller vessels, and a higher incidence of left ventricular hypertrophy may explain their poorer outcomes. Although Boylan and co-workers [2] found a higher mortality rate associated with PTCA of the circumflex and obtuse marginal branches, we and others [1, 3] have found that the culprit vessel did not affect the clinical outcome. However, we did find, as have others [1], that patients with lower ejection fractions tended to have higher mortality rates.
Preoperative hemodynamic instability and cardiogenic shock continue to be the major determinants of operative death [1–4, 6, 8–10]. Boylan and associates [2] observed an increase in the operative mortality rate from 1.4% to 28% in patients with preoperative shock. In our series, multivariate analysis demonstrated that the need for inotropic support was a very significant predictor of operative death. Although IABP support was used liberally in our patients with hypotension and cardiogenic shock, it had no effect in reducing mortality in this or other series [2, 4]. It had been hoped that percutaneous bypass might be able to stabilize patients who were in cardiogenic shock in preparation for CABG. However, both patients who were placed on percutaneous bypass in our series died, and other centers also have reported poor results [2]. Although percutaneous bypass may support the systemic circulation, our experimental studies show that it does not reverse myocardial ischemia in the presence of an acute coronary occlusion [11]. Because most patients in our series and those of others in whom PTCA failed died because of an MI or other cardiac-related event [1–4], this would explain why percutaneous bypass has failed to decrease the mortality rate significantly in these patients. Furthermore, the presence of peripheral vascular disease and limited femoral access may make it difficult to cannulate the femoral vessels in some patients.
Our experimental studies have shown that although percutaneous bypass may have a limited role in reversing acute coronary ischemia, the combination of percutaneous bypass and coronary retroperfusion significantly decreases infarct size [12]. It was hoped that the use of antegrade perfusion catheters also might limit myocardial necrosis during unsuccessful PTCA. Ferguson and co-workers [13] noted that coronary artery perfusion catheters helped to reverse acute electrocardiographic changes and decreased the incidence of MI. Sundram and colleagues [14] noted that patients who were managed with a perfusion catheter had a significantly lower incidence of Q-wave infarctions, and that these patients were more likely to receive an IMA graft. A study by Douglas and associates [15] as well as earlier studies by our group [4, 16], could not demonstrate a decreased incidence of MI with the use of these catheters. However, our current study shows that the use of antegrade perfusion catheters was associated with a lower mortality rate. Further, the increased use of these "bail-out" catheters was associated with decreased use of the IABP. Borkon and co-workers [3] reported similar results. In their study, reperfusion catheters were placed in 33% of patients. Although there was no associated improvement in the mortality rate or the incidence of MI, these catheters did result in greater hemodynamic stability and a decrease in ST segment elevation. Caes and Van Nooten [17] also observed no improvement in the mortality rate or the incidence of MI with reperfusion catheters. However, these catheters resulted in a significant decrease in the use of antiarrhythmic agents and more hemodynamic stability in preparation for operation. On the basis of these results and our own experience, we recommend placing antegrade perfusion catheters whenever technically possible. Although the incidence of MI appears to be unchanged, these patients seem to be more hemodynamically stable. Further, studies have shown that the higher levels of CK-MB seen in patients who undergo CABG after failed PTCA actually may reflect better washout of previously underperfused areas of the myocardium, leading to an overestimation of the actual infarct size [18]. Hence, the CK-MB level may not be a sensitive measure of the extent of an infarct when perfusion catheters are used.
Although our previous experimental studies have shown that retrograde cardioplegia significantly decreases infarct size after an acute coronary occlusion, we could not demonstrate any clinical improvement with any particular cardioplegia technique [19]. Substrate-enhanced cardioplegia (aspartate and glutamate) was not used on a regular basis in these patients. It is conceivable that this cardioplegic additive might have lowered the incidence of perioperative MI. We suspect that our clinical parameters of assessing myocardial necrosis are not as sensitive as the histochemical staining techniques used in the research laboratory. Previous studies also have failed to document any clinical improvement associated with a particular cardioplegia technique, cross-clamp time, or number of vessels bypassed [1, 2, 4, 7]. The lack of improvement with a particular cardioplegia technique lends further support to the observation that most ischemic damage after failed PTCA occurs before the period of cardioplegic arrest, and that interventions aimed at decreasing ischemic damage in the cardiac catheterization laboratory are more likely to affect clinical outcomes.
In our series, 36% of all patients received an IMA graft. Patients who died were less likely to have received an IMA graft. There are several potential reasons why an IMA graft may be unsuitable in these patients. Harvesting the IMA may be more time-consuming than harvesting vein grafts. However, if the patient is unstable, it still may be possible to harvest the IMA after the patient has been placed on cardiopulmonary bypass. There are concerns about the adequacy of flow in the IMA when it is grafted to a large area of ischemic myocardium. To overcome any concerns about inadequacy of flow, Caes and Van Nooten [17] often supplemented the IMA graft with a vein graft to the same vessel. However, the potential for competitive flow from this vein graft to affect the patency of the IMA graft remains uncertain. Another area of concern involves sewing a fragile IMA to a vessel that has been dissected. We have not had any problems grafting IMAs to dissected vessels; however, the process can be technically more demanding if the IMA is small and thin. Using the IMA also can affect the delivery of cardioplegia to the area of ischemic damage. Our experimental studies have shown that this can be ameliorated by using retrograde cardioplegia, especially if the left anterior descending artery is the culprit vessel [19]. Clearly, surgical judgment comes into play in deciding when to use arterialized grafts in these patients. We try to use an IMA graft whenever possible, but tend to use only vein grafts when the patient is unstable after anesthetic induction or when the flow through the IMA is thought to be inadequate for the amount of myocardium to be revascularized.
One of the major findings of this study was the poor outcome of patients with preoperative renal insufficiency. All 8 patients with a preoperative creatinine level of 1.8 mg/dL or greater died after emergency CABG performed for failed PTCA. It is not surprising that this group of patients should do so poorly. In addition to the immediate problems involved in the management of fluids and electrolytes after cardiopulmonary bypass, patients with chronic renal failure also have several conditions that predispose them to increased operative morbidity and mortality rates. These include their inability to metabolize certain medications, bleeding diatheses secondary to coagulation defects and platelet dysfunction, and susceptibility to infection. In a series of 775 patients who were undergoing cardiac operations with preexisting renal dysfunction. Zanargo and colleagues [20] found a significantly higher mortality rate (3.4% versus 17.1%), a higher incidence of respiratory complications (8.8% versus 21.9%), and a higher incidence of infections (2.3% versus 9.7%). Other series also have documented higher mortality and morbidity rates after CABG in patients with renal failure, especially when the procedure was performed under emergency conditions and in older patients with reduced ejection fractions [21–23]. The strategy of using PTCA to revascularize patients with chronic renal dysfunction is questionable. Koyanagi and associates [24] compared the clinical outcomes of CABG to those of PTCA in patients undergoing renal dialysis. Patients who underwent CABG had 100% patency of arterial grafts and 85% patency of vein grafts. In contrast, there was only a 76% success rate for the initial PTCA. The 5-year event-free rate was 70% for the CABG group, but only 18% for the PTCA group (p < 0.001). Although PTCA is technically feasible and provides relief of angina, aggressive restenosis limits the long-term benefit in this patient population. Further, when failed PTCA results in emergency CABG in these patients, their mortality and morbidity rates are excessive.
The results of this study have helped us to select patients better for percutaneous revascularization and to treat them more efficaciously when ischemia develops in the cardiac catheterization laboratory. Patients with multivessel disease are judged to be candidates for PTCA not only on the basis of the ease by which the lesion can be dilated, but also on the amount of myocardium at risk and the amount of myocardial reserve in the nondilated segments. Patients with multivessel disease and decreased ejection fraction in whom the culprit lesion serves a large territory of viable myocardium may be served better by surgical revascularization, because they are at higher risk for death after CABG if the vessel becomes occluded during PTCA. Further, because arterialized grafts are less likely to be used in a "crash" situation, the choice of conduit that can be used becomes an important issue in selecting the method of revascularization, especially in younger patients. Patients with preexisting renal dysfunction must be screened carefully for PTCA. Because the benefits of PTCA appear to be limited in these patients, surgical revascularization under stable conditions appears to be a safer procedure in this high-risk group.
Our clinical findings appear to confirm our experimental results, which show that techniques that are instituted immediately after abrupt coronary occlusion are more likely to limit necrosis and result in more favorable clinical outcomes [11, 12]. Because most patients die of MI after CABG for failed PTCA, every effort should be made to protect the myocardium in the cardiac catheterization laboratory. Antegrade perfusion ("bail-out") catheters are helpful in decreasing the extent of electrocardiographic changes and result in more hemodynamic stability. If these catheters cannot be inserted across the occlusion, an IABP should be placed to alter more favorably the balance of supply and demand. If electrocardiographic changes persist despite antegrade perfusion and insertion of an IABP, patients should be transported rapidly to the operating room before hemodynamic instability occurs. Multiple attempts at dilation are less likely to be successful and may compromise the patient further. Our results are similar to those of previous studies showing that the condition of the patient on leaving the cardiac catheterization laboratory probably is the most important factor in determining the clinical outcome [1–3, 6, 7].
The continuing improvement of PTCA techniques and the recent use of stents have decreased the incidence of emergency CABG after failed PTCA to less than 3% [1–4]. Nevertheless, as our study and other series have demonstrated, morbidity and mortality rates in this subset of patients remain high. The results of this study will help us to identify those patients selected for PTCA who are at higher risk for complications and death, and to treat them more effectively when emergency CABG is required. These strategies may reduce further the morbidity and mortality rates in this high-risk group of patients.
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. K. Haan, S. O'Brien, F. H. Edwards, E. D. Peterson, and T. B. Ferguson Trends in emergency coronary artery bypass grafting after percutaneous coronary intervention, 1994-2003. Ann. Thorac. Surg., May 1, 2006; 81(5): 1658 - 1665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. H. Yang, R. J. Gumina, R. J. Lennon, D. R. Holmes Jr, C. S. Rihal, and M. Singh Emergency Coronary Artery Bypass Surgery for Percutaneous Coronary Interventions: Changes in the Incidence, Clinical Characteristics, and Indications From 1979 to 2003 J. Am. Coll. Cardiol., December 6, 2005; 46(11): 2004 - 2009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Lotfi, K. Mackie, V. Dzavik, and P. H. Seidelin Impact of delays to cardiac surgery after failed angioplasty and stenting J. Am. Coll. Cardiol., February 4, 2004; 43(3): 337 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Barakate, P. G. Bannon, C. F. Hughes, M. D. Horton, A. Callaway, and T. Hurst Emergency surgery after unsuccessful coronary angioplasty: a review of 15 years' experience Ann. Thorac. Surg., May 1, 2003; 75(5): 1400 - 1405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Albes, M. Gross, U. Franke, J. Wippermann, T. U. Cohnert, R. Vollandt, and T. Wahlers Revascularization during acute myocardial infarction: risks and benefits revisited Ann. Thorac. Surg., July 1, 2002; 74(1): 102 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Reinecke, T. Fetsch, N. Roeder, C. Schmid, A. Winter, M. Ribbing, E. Berendes, M. Block, H. H. Scheld, G. Breithardt, et al. Emergency coronary artery bypass grafting after failed coronary angioplasty: what has changed in a decade? Ann. Thorac. Surg., December 1, 2000; 70(6): 1997 - 2003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. L. Lazar, Y. Bao, D. Lancaster, O. M. Shapira, G. S. Aldea, and R. J. Shemin Favorable impact of stents after emergent coronary artery bypass for failed angioplasty Ann. Thorac. Surg., November 1, 1999; 68(5): 1644 - 1647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Gurevitch, Y. Paz, I. Shapira, M. Matsa, A. Kramer, D. Pevni, O. Lev-Ran, Y. Moshkovitz, and R. Mohr Routine use of bilateral skeletonized internal mammary arteries for myocardial revascularization Ann. Thorac. Surg., August 1, 1999; 68(2): 406 - 411. [Abstract] [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 |
