Ann Thorac Surg 2003;75:1400-1405
© 2003 The Society of Thoracic Surgeons
Original article: cardiovascular
Emergency surgery after unsuccessful coronary angioplasty: a review of 15 years’ experience
Michael S. Barakate, FRACSa,
Paul G. Bannon, PhDa,
Clifford F. Hughes, AO, FRACSa*,
Matthew D. Horton, FRACSa,
Ann Callaway, MMgta,
Tara Hurst, MAStata
a Cardiothoracic Surgical Unit, Royal Prince Alfred Hospital, Baird Institute for Heart and Lung Surgical Research, University of Sydney, Sydney, Australia
Accepted for publication December 12, 2002.
* Address reprint requests to Dr Hughes, Cardiothoracic Surgical Unit, Level 8, Page Chest Pavilion, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, NSW 2050, Australia.
e-mail: clifford.hughes{at}email.cs.nsw.gov.au
 |
Abstract
|
|---|
BACKGROUND: Emergency coronary artery bypass grafting (CABG) is occasionally necessary for failed percutaneous transluminal coronary angioplasty (PTCA). The aim of this study was to assess the outcome of patients receiving emergency CABG after unsuccessful PTCA over a 15-year study period.
METHODS: From January 1982 through December 1996, 74 patients underwent emergency CABG after unsuccessful PTCA (crash group). This group was compared with a matched group of 74 patients having primary elective CABG (control group).
RESULTS: All 74 crash group patients were to have PTCA of one coronary system. After PTCA failure, 58 patients (78.3%) developed electrocardiographic changes of evolving acute myocardial infarction (AMI). The overall rate of AMI was 8.1% for the crash group and 2.7% for the control group. Two patients in the crash group died, with no deaths in the control group. There was no significant difference between mean in-hospital length of stay.
CONCLUSIONS: With prompt, aggressive, and complete myocardial revascularization, patients who required emergency CABG after PTCA failure had an outcome not significantly different from that of patients having elective CABG.
|
Introduction
|
|---|
Emergency coronary artery bypass grafting (CABG) is occasionally necessary for failed percutaneous transluminal coronary angioplasty (PTCA). Emergency CABG after PTCA failure has a reported acute myocardial infarction (AMI) rate of 8.9% to 51% and a mortality rate of 3.8% to 14% [1–5]. In comparison, elective coronary artery surgery is a relatively low risk procedure with a reported mortality rate of 1.3% and an AMI rate of 5.4% [6]. This study reports 15 years of experience in emergency coronary bypass after angioplasty failure at the Royal Prince Alfred Hospital (RPAH), from January 1982 through December 1996 by retrospective review. The aim of this study was to assess the outcome of patients receiving emergency CABG after unsuccessful PTCA. We compared these outcomes with patients having primary elective CABG.
|
Material and methods
|
|---|
Over the 15-year study period 4,146 PTCA procedures were performed at RPAH, and of these, 74 patients (1.8%) required emergency CABG within 24 hours of PTCA failure (crash group). Angiographic findings of PTCA failure included acute arterial dissection and coronary artery closure that could not be redilated. Clinical findings associated with PTCA failure included persistent chest pain, electrocardiograph (ECG) changes of evolving myocardial infarction, persistent hypotension, unstable arrhythmias, or cardiac arrest. Specific cardiac catheterization data examined for the crash group included degree of coronary artery disease (significant stenosis was defined as greater than 50% reduction in luminal diameter or estimated 75% reduction in cross-sectional area), the cause of angioplasty failure, the culprit vessel, and whether a reperfusion catheter was inserted in the catheterization laboratory. Reperfusion catheters were available from 1989 onward. The decision to use a reperfusion catheter was made by the cardiologist before involvement of the cardiothoracic surgical unit.
This paper reports all the patients for whom the surgeons were called. Patients who required salvage techniques in the catheterization laboratory were not the focus of this paper. Coronary artery stenting was introduced at our institution in 1994. Stenting did not make a major impact until 1995 with the introduction of second-generation stents. GpIIbIIIa inhibitors were not available at our institution until 1998, well after this study was completed. The fact that there were only approximately 5 patients each year who required emergency CABG after PTCA failure represents the skill of our cardiologists and not the use of salvage techniques.
During the study period 11,909 patients underwent primary elective CABG and had been entered in a prospective database. From this database a matched group of 74 patients (control group) was selected by computer search. Further patient data were obtained from the hospital’s cardiology reporting system, the cardiothoracic unit’s clinical reporting system, medical records, and the surgeons’ records. Control group data were matched to the crash group data for the year of operation, the number of coronary systems diseased, and number of bypass grafts performed. We then checked that there were no statistically significant differences in preoperative risk factors. Preoperative risk factors assessed for the two study groups included left ventricular ejection fraction, coexisting valvular heart disease, diabetes mellitus, obesity, previous stroke, preexisting chronic renal impairment, chronic obstructive pulmonary disease, peripheral vascular disease, and family history of ischemic heart disease. Patients who had previous CABG were excluded from both groups. Preoperative comparison is shown in Table 1.
Statistics
Comparison between the crash and control groups was performed using the SPSS statistical software package (SPSS Inc, Chicago, IL). Fisher’s exact tests were calculated for two-by-two tables. Mann-Whitney U tests were calculated for continuous data, as appropriate.
|
Results
|
|---|
Emergency surgery was necessary in 74 patients having PTCA (1.8%). The PTCA failure rate leading to emergency CABG was 4.2% over the first half of the study (January 1982 to June 1989) and was 1.3% over the latter half of the study (July 1989 to December 1996 inclusive). Comparisons of preoperative criteria studied are shown in Table 1. Of the 74 crash CABG patients, all were intended to have angioplasty of one system, usually a single culprit vessel. The angiograms had been reported before intervention, and on review of these reports it was found that 47 of 74 patients (63.5%) had significant stenosis of one coronary system, 19 of 74 (25.7%) had two-system disease, and 8 of 74 (10.8%) had three-system disease. Two patients had significant left main stem disease. Details regarding coronary arteries diseased, coronary systems grafted, mean cardiopulmonary bypass (CPB) times, and number of distal anastomoses performed for both study groups are shown in Table 2.
PTCA failures were due to arterial dissection with or without complete occlusion in 40 patients (54%) and complete occlusion without dissection in 34 patients (46%). Vessels intended to be dilated were the left anterior descending in 39 patients (52.7%), the right coronary artery in 26 patients (35.1%), and the left circumflex in 9 patients (12.2%). Only 1 patient had two-vessel dilatation for disease of the left anterior descending and first diagonal arteries. GpIIbIIIa inhibitors were not available at our institution until 1998. Intraaortic balloon pumps (IABP) were not inserted preoperatively. A coronary artery stent was inserted in 3 patients (4%), indicated by arterial dissection in 2 and acute arterial closure in 1 patient. Of these 3, the patient with acute arterial closure sustained an AMI but made a good recovery after emergency CABG. The remaining 2 patients did well without complication after emergency CABG. A reperfusion catheter was inserted in 24 crash group patients (32%). For details regarding outcomes after emergency CABG after unsuccessful PTCA with and without reperfusion catheter usage see Figure 1.
View larger version (22K):
[in this window]
[in a new window]
|
Fig 1. Details regarding outcomes after emergency coronary artery bypass graft (CABG) surgery after unsuccessful percutaneous transluminal angioplasty (PTCA) with and without reperfusion catheter usage. (AMI = acute myocardial infarction despite emergency CABG; h = hours; LIMA = left internal mammary artery usage at CABG; m = minutes; OR = operating room.)
|
|
Angioplasty failure was accompanied by angina in 63 patients (85%), and ECG changes of evolving myocardial infarction in 58 patients (78%). Hemodynamic instability occurred in 24 patients (33%) after the procedure, with prolonged hypotension in 13 patients (18%), and unstable ventricular arrhythmias in 8 patients (11%). Cardiopulmonary resuscitation was required for 3 patients (4%) with evolving AMI and shock who were transported directly to the operating theater. These 3 patients were massaged onto CPB, fortunately all of whom survived.
The average time from termination of the PTCA to commencement of CPB was 3 hours 22 minutes for 59 patients (80%) in whom the decision to operate was made in the catheterization laboratory. After emergency CABG the AMI and mortality rates for these patients were 4 of 59 (6.8%) and 2 of 59 (3.4%) respectively. For the remaining 15 patients time from termination of PTCA to commencement of CPB ranged from 6 to 24 hours. These patients had all developed intractable angina pectoris and signs of evolving myocardial infarction that necessitated emergency operative revascularization. After emergency CABG the AMI rate for these patients was 2 of 15 (13.3%) with no mortality. The overall mortality rate for the crash group was 2 of 74 (2.7%).
Crash group patients received St. Thomas II cardioplegic solution (crystalloid buffered with bicarbonate without additives; Mayne Pharma, Melbourne, Victoria). This was administered antegrade through the aortic root in 65 patients. Nine patients received cardioplegia through other means: retrograde through the coronary sinus in 2, antegrade and retrograde in 2, antegrade and retrograde and down the grafts in 1, and antegrade and down the grafts in 4 patients. These 9 patients were operated on over the last 7 years of the study (1991 to 1996 inclusive) and none of these patients suffered myocardial infarction or died.
There was no significant difference in the CPB times of the two groups. Although single system angioplasty was initially intended for all patients, at operation the average number of systems grafted was 1.5 (see Table 2). The left internal mammary artery (LIMA) was used for bypass grafting in 25 patients (34%) in the crash group (4 sustained AMI, with no mortality) and 51 patients (69%) in the control group (2 sustained AMI, with no mortality). The average number of distal anastomoses for patients in the crash group was 2.1 (range, 1 to 7), and for the matched control group was 2.2 (range, 1 to 5). During the same period the average number of distal anastomoses was 4.1 per patient for those who underwent primary coronary revascularization at our unit.
Differences in morbidity for the crash group between the first and second half of the study and comparison with the control group are shown in Table 3.
For the crash group over the first half of the study there was a higher rate of low cardiac output, postoperative hemorrhage greater than 1.5 L, myocardial infarction, and 30-day mortality. There were no significant differences between the two study groups in incidences of postoperative AMI or 30-day mortality. In the crash group AMI developed in 6 patients, 2 of whom died. In the control group 2 patients developed AMI, with no mortality. There were no significant differences in requirement for postoperative catecholamines, reoperation for hemorrhage, nor the incidence of postoperative arrhythmias. There was no significant difference between mean in-hospital length of stay.
View this table:
[in this window]
[in a new window]
|
Table 3. Differences in Morbidity for the Crash Group Between the First and Second Half of the Study and Comparison Between the Whole Crash and Control Groups
|
|
|
Comment
|
|---|
The reported incidence of emergency CABG after PTCA failure ranges from 0.32% to 7% [1–5, 7]. Emergency CABG after PTCA failure is now required infrequently, however, owing to increased operator experience in PTCA techniques and intracoronary stent usage [4, 7]. The consequences of PTCA failure can be serious and include AMI and death. The international literature indicates that the patient immediately enters a higher risk group after PTCA failure, with reported mortality rates for subsequent emergency CABG of 3.8% to 14% [1–5]. In contrast, the 30-day mortality in this series was 2 of 74 (2.7%). The explanation for this is multifactorial and demands a multidisciplinary approach to management after PTCA failure. In our experience the key factors in achieving a satisfactory outcome include simultaneous resuscitation and management in the catheterization laboratory, minimizing the time to surgical revascularization (with the operating rooms in the same building as the catheterization laboratory in our hospital), and complete myocardial revascularization at surgery.
In the present study the overall rate of emergency CABG after PTCA failure was 1.8%, with a rate of 4.2% for the first half of the study period compared with 1.3% during the latter half. The cause of PTCA failure in the present study was arterial dissection (54%) and acute arterial occlusion (46%), and these are the most commonly reported reasons for PTCA failure necessitating emergency CABG [1–3, 8–11]. Other reported indications for emergency surgery include arterial spasm and arterial perforation into the pericardial cavity [1, 10–12]. More recently stent complications have been reported as the cause of PTCA failure in as many as 37% of cases [4]. When PTCA failure causes coronary artery damage, surgical options for that artery may be difficult depending upon the extent of the injury (ie, coronary artery dissection). As a result of the failure 33% of patients in this series were in critical condition before surgery. That is consistent with 25% to 38% of patients in other reported series being unstable before emergency CABG after PTCA failure [2, 5, 8], highlighting the high risk involved in operating in these circumstances.
Our devised policy was to get the patient to the operating room and on CPB as quickly as possible. For the crash group the AMI rate was 6.8% for those in whom the decision to operate was made in the catheterization laboratory, compared with 13.3% for those patients held until intractable angina pectoris and signs of evolving myocardial infarction developed (see Results). The reported risk of AMI and death is proportional to the duration of myocardial ischemia [1, 4, 5] and it was our surgical aim to minimize the time to revascularization. Measures to decrease ischemia may be used after PTCA failure [2]. The use of reperfusion catheters after PTCA failure has been shown to help reverse the ECG changes seen and reduce the incidence of AMI [7, 12–14]. In our experience reperfusion catheter usage resulted in a longer delay to the commencement of CPB (4 hours 6 minutes versus 3 hours 18 minutes) and subsequently higher AMI (12.5% versus 5.3%) and mortality (4.2% versus 2.6%) rates for those patients in whom the decision to operate was made in the catheterization laboratory (see Fig 1). Reperfusion catheter use after unsuccessful angioplasty may delay time to operative revascularization and lead to a false sense of security, potentially resulting in worse outcomes. This area needs further investigation.
Other techniques reported to successfully decrease the rate of AMI in this setting include the use of the IABP, intracoronary nitroglycerin, and coronary stenting. Percutaneous CPB may minimize myocardial damage after unsuccessful PTCA but only when the patient regains a stable cardiac rhythm [15]. In the early experience IABP use meant an open procedure under difficult nonsterile circumstances and our view was that the risk outweighed the benefit of urgent CABG once the patient crashed. Since the advent of percutaneous IABP this is now possible in the catheterization laboratory but use should not delay transfer to the operating room and institution of CPB. Our experience has shown that IABP use (along with new techniques) is not critical provided there is no delay with prompt operative revascularization after unsuccessful PTCA.
Ischemic time has been defined differently in previous reports, making comparison difficult. Parsonnet and colleagues [1] reported time for transit from the catheterization room to the operating room for 59 of 67 patients (88% of their study group) averaging 26 minutes, with 8 patients (12%) held for observation until sudden acceleration of symptoms and signs. In the post-PTCA group the rate of AMI was 28% and mortality of 12% [1]. Greene and colleagues [2] reported the time for revascularization from angioplasty failure to when the patient came off bypass. This time averaged 3 hours 4 minutes for 53 patients, with an AMI rate of 51% and a mortality rate of 3.8% [2]. Borkon and colleagues [5] reported 73 of 91 patients (80%) went directly from the catheterization room to the operating room without specifying the actual time while the remaining 18 patients (20%) had development of symptoms within 24 hours after PTCA that necessitated emergency CABG. The AMI rate was 29% with a mortality rate of 12.1% [5]. The differences in outcomes between these studies may in part be explained by a different method of estimating ischemic time, which may inaccurately reflect the actual duration of myocardial ischemia. Therefore we recommend standardization of reported ischemic times. We have employed the time from the onset of ischemia (by clinical and angiographic features) to the commencement of CPB. The time for onset of CPB was a consistently reported time point for all patients and represents the point at which the myocardium was rested.
Previous reports have employed varied methods for obtaining control groups to compare with patients undergoing emergency CABG after unsuccessful PTCA [1, 2, 5, 8, 10, 12]. We chose a control group who underwent elective CABG matched for the year of operation, the number of coronary systems diseased, and number of bypass grafts performed. We then checked that there were no statistically significant differences in preoperative risk factors as shown in Table 1. We were unable to confirm the validity of the data for smoking history and presence of hypertension because these factors were not consistently measured over the study period. Smoking history was not always reliably recorded in the emergency situation. Year of operation was matched to ensure that similar surgical techniques were performed. In this way we have compared the outcomes of an emergency procedure after unsuccessful PTCA intervention with a similar but elective procedure. It is interesting to note that the PTCA (crash) group had a significantly shorter duration of symptoms before intervention was offered (see Table 1). It is also interesting that except for 1, all patients were intended to have single-vessel PTCA targeting the culprit vessel. At surgery, however, an average 2.1 coronary grafts (range, 1 to 7) were performed after PTCA failure (see Table 2). That compares with an average of approximately two grafts per patient having emergency revascularization after failed PTCA in the international literature [1–3, 5, 8]. It is only because of very aggressive and complete coronary artery surgery that we have been able to achieve results that are equivalent to planned surgery (see Table 3). We have not addressed long-term outcomes in this paper. This study specifically addresses the immediate results with aggressive CABG. The long-term outcomes in these circumstances would be complex but nevertheless of great interest.
Recent data from Reinecke and colleagues [16] analyzed significant differences between survivors and nonsurvivors of emergency CABG after failed PTCA. In their study survivors were significantly younger (58.2 versus 65.4 years, p < 0.01), had greater mean body surface area (1.93 m2 versus 1.73 m2, p < 0.001), had lower mean Cleveland score (7.06 versus 8.86, p < 0.001), more frequently received complete operative revascularization (80% versus 36%, p < 0.001), and had faster mean bypass times (56 versus 91 minutes, p < 0.001). Furthermore this paper stated that "non survivors were more frequently female (64% versus 24%, p < 0.01), had a moderately or severely reduced left ventricle (29% versus 9.4%, p < 0.05), more frequently required intensive treatment (cardiocompression, defibrillation, IABP insertion etc., 93% versus 33%, p < 0.001), and interestingly had faster mean time from PTCA end to start of CABG (57 versus 94 minutes, p < 0.05)" [16]. In our series 6 patients sustained AMI after PTCA failure and of these, 2 died. Of the remaining 68 patients, none died. Although the relatively small number of patients in our study who died precluded further analysis of mortality risk factors, reported data indicates that the risk of death is greatest for those who experience ongoing myocardial ischemia and AMI despite surgical revascularization [1, 4, 5]. Other investigators have found additional risk factors for mortality including advanced age, low left ventricular ejection fraction, multivessel disease, female sex, PTCA of unfavorable stenoses, multiple vessel PTCA, and prior CABG [5, 8, 10, 17, 18].
In summary, this series reports a low rate of AMI (8.1%) and death (2.7%) for patients who underwent emergency CABG after PTCA failure, results that were not significantly different when compared with those of a matched group who underwent elective CABG (see Table 3). These results compare well with reports from units with onsite surgical backup [1, 2, 5] and certainly compare favorably with those from units with offsite surgical backup, which cite high mortality rates of 14% [4]. The results in this series have been achieved by simultaneous resuscitation and management in the catheterization laboratory, minimizing the time to surgical revascularization, and complete myocardial revascularization at surgery. Our salvage rate highlights the need for a coordinated effort between cardiac surgeons and invasive cardiologists. The era of having an operating room open with surgeons and a team "standing by" is over. Yet acute closures and dissections do still occur. A system that allows early admission of trouble, notification of surgeon and operating room team, and a coordinated effort to get the patient to the operating room and on bypass will clearly give the best chances of survival.
|
Acknowledgments
|
|---|
The authors thank the surgeons Matthew S. Bayfield, FRACS, Bruce G. French, FRACS, Nick Hendel, FRACS, Brian C. McCaughan, FRACS, and Duncan S. Thomson, FRACS, for their contribution to the clinical work that formed the basis for this research.
|
References
|
|---|
- Parsonnet V., Fisch D., Gielchinsky I., et al. Emergency operation after failed angioplasty. J Thorac Cardiovasc Surg 1988;96:198-203.[Abstract]
- Greene M.A., Gray L.A., Slater A.D., et al. Emergency aortocoronary bypass after failed angioplasty. Ann Thorac Surg 1991;51:194-199.[Abstract]
- Ladowski J.S., Dillon T.A., Deschner W.P., et al. Durability of emergency coronary artery bypass for complications of failed angioplasty. Cardiovasc Surg 1996;4:23-27.[Medline]
- Loubeyre C., Morice M.-C., Berzin B., et al. Emergency coronary artery bypass surgery following coronary angioplasty and stenting. Results of a French multicenter registry. Cathet Cardiovasc Intervent 1999;48:441-448.
- Borkon A.M., Failing T.L., Piehler J.M., et al. Risk analysis of operative intervention of failed coronary angioplasty. Ann Thorac Surg 1992;54:884-891.[Abstract]
- Pocock S.J., Henderson R.A., Rickards A.F., et al. Meta-analysis of randomised trials comparing coronary angioplasty with bypass surgery. Lancet 1995;346:1184-1189.[Medline]
- Heikkinen L., Virtanen K., Heikkila J., et al. Surgical treatment of acute myocardial ischaemia related to coronary angioplasty with special reference to use of perfusion balloon catheter and long-term outcome. J Cardiovasc Surg 1997;38:101-106.[Medline]
- Carey J.A., Davies S.W., Balcon R., et al. Emergency surgical revascularization for coronary angioplasty complications. Br Heart J 1994;72:428-435.[Abstract/Free Full Text]
- Celermajer D.S., Bailey B.P., Beetson R., et al. Emergency coronary artery surgery following angioplasty—favourable medium term outcome after eight years’ experience. Aust NZ J Med 1991;21:211-216.[Medline]
- Boylan M.J., Lytle B.W., Taylor P.C., et al. Have PTCA failures requiring emergent bypass operation changed?. Ann Thorac Surg 1995;59:283-287.[Abstract/Free Full Text]
- Cameron D.E., Stinson D.C., Greene P.C., Gardner T.J. Surgical standby for percutaneous transluminal coronary angioplasty: a survey of patterns of practice. Ann Thorac Surg 1990;50:35-39.[Abstract]
- Lazar H.L., Faxon D.P., Paone G., et al. Changing profiles of failed coronary angioplasty patients: impact on surgical results. Ann Thorac Surg 1992;53:269-273.[Abstract]
- Ferguson T.B., Hinohara T., Simpson J., Stack R.S., Wechsler A.S. Catheter reperfusion to allow optimal coronary bypass grafting following failed transluminal coronary angioplasty. Ann Thorac Surg 1986;42:399-405.[Abstract]
- Ferguson T.B., Muhlbaier L.H., Salai D.L., et al. Coronary bypass grafting after failed elective and failed emergent percutaneous angioplasty. J Thorac Cardiovasc Surg 1988;95:761-772.[Abstract]
- Grambow D.W., Deeb G.M., Pavlides G.S., Margulis A., O’Neill W.W., Bates E.R. Emergent cardiopulmonary bypass in patients having cardiovascular collapse in the cardiac catheterisation laboratory. Am J Cardiol 1994;73:872-875.[Medline]
- Reinecke H., Fetsch T., Roeder T., et al. Emergency coronary artery bypass grafting after failed coronary angioplasty: what has changed in a decade?. Ann Thorac Surg 2000;70:1997-2003.[Abstract/Free Full Text]
- Lazar H.L., Haan C.K. Determinants of myocardial infarction following emergency coronary artery bypass for failed percutaneous coronary angioplasty. Ann Thorac Surg 1987;44:646-650.[Abstract]
- Lazar H.L., Jacobs A.K., Aldea G.S., et al. Factors influencing mortality after emergency coronary artery bypass grafting for failed percutaneous transluminal coronary angioplasty. Ann Thorac Surg 1997;64:1747-1752.[Abstract/Free Full Text]
Top
Abstract
Introduction
