- Email: [email protected]
Six-year survival after coronary thrombolysis and early revascularization for acute myocardial infarction
Six-Year Survival After Coronary Thrombolysis and Early Revascularization for Acute Myocardial Infarction George J. Taylor, MD, H. Weston Moses, MD, Richard E. Katholi, MD, Cynthia Korsmeyer, MA, Paul Kolm, PhD, James T. Dove, MD, Frank L. Mikell, MD, Joseph M. Sutton, MD, Harry A. Wellons, MD, and Joel A. Schneider, MD Six-year follow-up was conducted in a consecutive series of 192 patients receiving thrombolytic therapy for acute myocardial infarction (AM) with STsegment elevation. Cardiac catheterization was performed within a day, and patients with an open infarct artery routinely had early revascularizatlon: 66 (67%) underwent coronary bypass surgery and 16 (12%) coronary angioplasty. With this treatment strategy, 6-year cardiac mortality was 14.5% 6% (12 patlents) in hospital and 9% (16 patients) for survivors of hospitalization. Multivarlate analysis showed that predictors of cardiac death among survivors of hospitalization were a closed infarct artery at catheterization (p
here have been few prospectivefollow-up studies of patients treated with thrombolytic therapy for acute myocardial infarction (AMI). Mathey et al1 describeda 4-year follow-up of 227 patients treated with streptokinaseearly during AMI, and reported improved survival with an open infarct artery. They also found that coronary artery bypassgrafting had an added survival benefit. In this report we present 6-year follow-up of patients having thrombolytic therapy for acute, transmural myocardial infarction and specifically examine clinical and angiographic variables that predict late mortality. We have found that a closed infarct artery soonafter thrombolytic therapy, diabetesand anterior location of AM1 were predictors of late cardiac death independent of left ventricular function.
T
METHODS
This study includes 192 consecutive patients with transmural AM1 treated with intracoronary or intravenous streptokinasewho were followed for >5 years. Selection criteria for thrombolytic therapy, contraindications, treatment protocols and early clinical course have been reported.*J Patients aged 170 years, with chest pain lasting <6 hours unresponsiveto sublingual nitroglycerin, and with 20.2 mV ST-segmentelevation,were treated. Intracoronary streptokinasewas usedduring the first 11 months of the study. Patients were brought to the cardiac catheterization laboratory as quickly as possible after diagnosisof transmural myocardial infarction and treated with an initial intracoronary streptokinasebolus of 30,000 U followed by 3,000 U of intracoronary streptokinase per minute for 60 minutes. Heparin was started immediately after intracoronary streptokinasein patients with arterial opening. During the subsequent 13 months of the study patients were treated with intravenous streptokinase.This was given in the emergencyroom as soonas the diagnosis of myocardial infarction was confirmed by the primary care physician. A “front loaded’ dosing regimen was used, with an initial doseof 500,000 U over 5 minutes followed by 200,000 U/hour for 4 hours. Heparin was then started at 1,000 U/hour, and the dose was From the Prairie CardiovascularCenter, Springfield, the Prairie Edu- adjusted to maintain activated partial thromboplastin cation & Research Cooperative, Springfield; and the Department of time 70 to 100 seconds.Patients treated with intraveMedicine, SouthernIllinois University Schoolof Medicine, Springfield, nous streptokinasein community hospitals were transIllinois. Manuscript received January 6, 1992;revised manuscript referred for angiography as soon after the initial streptoceived February 28,1992, and acceptedMarch 2. kinase bolus infusion as possible,and most had angiogAddress for reprints: George J. Taylor, MD, P.O. Box 19420, raphy within 24 hours.2,3 Heparin was discontinued Springfield, Illinois 62794-9420.
26
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70
JULY 1. 1992
with documentedabsenceof thrombolysis (patients with intravenous streptokinase) or at the time of revascularization (both intracoronary and intravenous streptokinase). Early revascularization was routinely recommended for patients with an open infarct artery and critical stenosis of the infarct artery (reduction of luminal diameter by 170%). Revascularization was performed during the samehospitalization as mn as it could be scheduled and after clotting function had returned to normal. Percutaneous transluminal coronary angioplasty or coronary artery bypass surgery was performed using standard techniques, and details of our early experience have been reported.2T3Radionuclide angiography was performed in 146 of the 180 survivors of myocardial infarction before hospital discharge (142 patients), or within 2 months (4 patients). No attempt was made to standardize medical therapy after discharge, although all patients were treated with daily aspirin. Follow-up was conducted between December 1988 and April 1989. This seriesincluded all patients treated with thrombolytic therapy before September 1983. Follow-up was achievedfor all 180 survivors of hospitalization in the seriesby a nurse researcherby telephone or clinic visit. The primary end point was cardiac death defined as sudden death, death after AMI, or death after hospitalization for congestiveheart failure. The effect of each clinical and angiographic variable on cardiac mortality for patients surviving hospitalization was analyzed by chi-square and t test for unpaired variables. Cox’s proportional hazards regressionwas subsequently used to analyze the relation (if any) among clinical and angiographic variables affecting survival4 The Cox model was also used to calculate an estimated survival function adjusted for the combination of variables found to be significant. An additional regressionanalysis was performed for a subset of patients having predischarge radionuclide angiograms. All values are expressedas mean f 1 standard deviation.
RESULTS During 24 months, 192 patients with transmural myocardial infarction were treated with intracoronary (n = 75) or intravenous (n = 117) streptokinase.Twelve patients (6%) died before hospital discharge (2,3). Follow-up data were obtained for all 180 survivors of hospitalization a minimum of 5 years after myocardial infarction. During follow-up, 21 of the 180 patients died (11.7%). Average follow-up for the 159 survivors was 74 f 7 months. There were 16 cardiac deaths (9%) (Figure 1); 7 had recurrent AMI, 6 had congestive heart failure, 2 died suddenly, and 1 died after coronary bypasssurgery. Subsequentmortality analysis will consider only these 16 patients with cardiac death. Clinical characttistii: The hospital course of the initial cohort of 192 patients has been descrk-12y3 The averagepatient in this follow-up study of 180 survivors was aged 55 f 10 years, and clinical characteristics are outlined in Table I. All patients had ST-segmentelevation and chest pain at the time of streptokinasetherapy and, as reported, all subsequentlyhad elevation of creatine kinase.2Streptokinase therapy was begun 202 f 92 minutes after onset of chest pain. An open infarct artery was identified at cardiac catheterization in 137 of the 180 survivors of hospitalization (76%). Early angiography was attempted in the 111 patients treated with intravenous streptokinase;the interval from initiation of therapy to catheterization was 1 to 4 hours for 46 (41%), 4 to 12 hours for 20 (18%), 12 to 24 hours for 19 (17%) 24 to 48 hours for 17 (15%) and 2 to 4 days for 9 (8%). Early revascularization while in the hospital after AM1 was performed an average of 3 f 2 days after streptokinase treatment in 111 of the 137 patients (81%) with an open infarct artery at catheterization; 94 (69%) had coronary artery bypass surgery and 17 (13%) had successfulpercutaneoustransluminal coronary angioplasty. Bypass patients received 3 f 1.4 grafts per patient. Angioplasty was attempted in 22 pa-
“0°%
0.80-
,x0.60-
0.20 I I 0.00
I 0
I 12
I 24
I 36 Months
I 48
I 60
SIX-YEAR SURVIVAL AFTER CORONARY THROMBOLYSIS
I 72
27
group for purposesof analysis), and 1 had angioplasty of a noninfarct artery. Univariate prediitors of survivab Clinical features No. of Cardiac Death and follow-up cardiac mortality are summarized in TaPts. p Value (%I ble I. Significant univariate predictors of survival in16 (9%) All patients 180 cluded location of myocardial infarction, follow-up ejec3 (9%) 32 NS* Age 2 65 years tion fraction, diabetes,Killip class on admission and an 5 (15%)/11 (7%) Women/men 331147 NS open infarct artery. Location of infarction and subse 11(9%) 119 NS* Cigarette uset quent left ventricular function were closely related, paNS* Systemic hypertension 63 7(11%) 27 7 (26%) <0.001* Diabetes mellitus tients with anterior infarction had a predischargeejec 32 5 (16%) NS* Prior AMI tion fraction of 41 f 15, and those with inferior infarc10 (15%) 66 Anterior AMI tion had a predischargeejection fraction of 58 f 11 (p 114 0.025 Inferior AMI 6 (5%)
28
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70
JULY 1, 1992
Closed, inferior
Closed, Anterior
gO.SOz Aa 2 2 a 0.40-
0.20-
0.00
1 0
I 12
I 24
I 36 Months
I 48
I 60
I 72
Closed, Inferior
Closed, Anterior
0.20-
0.00
I 0
I 12
passgrafting as an independent predictor of late survival, possibly becausemore of our patients had revasculark&on (68 vs 48%) and bypassgrafting (57 vs 22%) when comparedwith Mathey’s study. A higher percentage, if not most patients who might have benefited from bypass grafting, underwent it in our trial. Both of these studies identified an open infarct artery as a predictor of survival independentof changesin left ventricular function. Whereas the results support the “open artery hypothesis,” they do not prove it, as neither was a controlled study.5-9Furthermore, the high associationof an open infarct artery and revascularization in these studies makes it hard to determine which of the 2, reperfusion or revascularization, is most important in determining long-term outcome. Mathey claimed that bypasssurgery helped. Our results do not support that claim. On the other hand, the Thrombolysis in Myocardial Infarction trial, a controlled study of
I 24
I 36 Months
I 48
I 60
1
72
“revascularization” after reperfusion therapy, failed to show any short-term survival benefit.10But this study routinely usedearly angioplasty of the culprit lesion, not coronary bypasssurgery, the approach most often taken in our study and the Mathey trial. Our data provide an estimation of late prognosisafter AM1 using a treatment strategy of thrombolysis and early revascularization. There are no late survival data for patients having medical treatment or angioplasty as routine therapy after coronary thrombolysis. The present results should not be used to advise such patients about prognosis,since late mortality must be influenced both by thrombolysis and subsequenttherapy. REFERENCES 1. Mathey DG, Schafer J, SheehanFH, Krebber H, Justen M, Rodewald G, Dodge HT, Bleifeld W. Improved survival up to four years after early coronary thrombolysis. Am J Cmdiol 1988;61:524-529.
SIX-YEAR SURVIVAL AFTER CORONARY THROMBOLYSIS
29
2. Taylor GJ, Mike11FL, MosesHW, Dove JT, BatchelderJE, Thug A, Hansen S, Wellons HA, Schneider JA. Intravenous versus intracoronary streptokinase therapy for acute mywardiil infarction in community hospitals.Am J Cardiol 1984;54:256-260.
1. Sutton JM, Taylor GJ, Mike8 FL, Moses HW, Korsmeyer C, Dove JT, Batchelder JE, Wellons HA, SchneiderJA. Thrombolytic therapy followed by early revascularization for acute myocardiil infarction. Am J Cardiof 198657: 1227-1231. 4. Morrison DF. Multivariate Statistical Methods. New York: McGraw-Hill, 1%7:130-133.
5. Braunwald E. Myocardiil reperfusion,limitation of infarct size, reduction of left ventricular dysfunction, and improved survival: should the paradigm be expanded?Circulation 1989;79:441-444. 6. Califf RM, Topol EJ, Gersh BJ. From myocardiil salvageto patient salvagein acute myccardial infarction: the role of reperfusiontherapy (editorial). J Am CON
30
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70
Cardiol 1989;14:1382-1388. 7. KennedyJW, Ritchie JL, Davis KB, Fritz JK. WesternWashingtonintracoronary streptokinase in mywardial infarction trial. N Engl J Med 1983;309: 1477-1482. 8. Ritchie JL, Davis KB, Williams DL, Caldwell J, Kennedy W. Global and regional left ventricular function and tomographic radionuclide perfusion: The Western Washington randomized trial of intracoronary streptokinasein acute mvocardial infarction. Circulation 1984:70:867-875. 9: Dalen JE, Gore JM, Braunwald E, Borer J, Goldberg RJ, PassamaniER, Forman S, Knatterud G, and the TIMI Investigators.Six- and 1Zmonth followup of the phase I thrombolysis in myocardial infarction (TIMI) trial. Am J Cardiol 1988;62:179-185. 10. The TIMI Study Group. Comparisonof invasiveand conservativestrategies after treatment with intravenoustissueplasminogenactivator in acute myocardial infarction. N Engf J h&d 1989320~618-627.
JULY 1. 1992
here have been few prospectivefollow-up studies of patients treated with thrombolytic therapy for acute myocardial infarction (AMI). Mathey et al1 describeda 4-year follow-up of 227 patients treated with streptokinaseearly during AMI, and reported improved survival with an open infarct artery. They also found that coronary artery bypassgrafting had an added survival benefit. In this report we present 6-year follow-up of patients having thrombolytic therapy for acute, transmural myocardial infarction and specifically examine clinical and angiographic variables that predict late mortality. We have found that a closed infarct artery soonafter thrombolytic therapy, diabetesand anterior location of AM1 were predictors of late cardiac death independent of left ventricular function.
T
METHODS
This study includes 192 consecutive patients with transmural AM1 treated with intracoronary or intravenous streptokinasewho were followed for >5 years. Selection criteria for thrombolytic therapy, contraindications, treatment protocols and early clinical course have been reported.*J Patients aged 170 years, with chest pain lasting <6 hours unresponsiveto sublingual nitroglycerin, and with 20.2 mV ST-segmentelevation,were treated. Intracoronary streptokinasewas usedduring the first 11 months of the study. Patients were brought to the cardiac catheterization laboratory as quickly as possible after diagnosisof transmural myocardial infarction and treated with an initial intracoronary streptokinasebolus of 30,000 U followed by 3,000 U of intracoronary streptokinase per minute for 60 minutes. Heparin was started immediately after intracoronary streptokinasein patients with arterial opening. During the subsequent 13 months of the study patients were treated with intravenous streptokinase.This was given in the emergencyroom as soonas the diagnosis of myocardial infarction was confirmed by the primary care physician. A “front loaded’ dosing regimen was used, with an initial doseof 500,000 U over 5 minutes followed by 200,000 U/hour for 4 hours. Heparin was then started at 1,000 U/hour, and the dose was From the Prairie CardiovascularCenter, Springfield, the Prairie Edu- adjusted to maintain activated partial thromboplastin cation & Research Cooperative, Springfield; and the Department of time 70 to 100 seconds.Patients treated with intraveMedicine, SouthernIllinois University Schoolof Medicine, Springfield, nous streptokinasein community hospitals were transIllinois. Manuscript received January 6, 1992;revised manuscript referred for angiography as soon after the initial streptoceived February 28,1992, and acceptedMarch 2. kinase bolus infusion as possible,and most had angiogAddress for reprints: George J. Taylor, MD, P.O. Box 19420, raphy within 24 hours.2,3 Heparin was discontinued Springfield, Illinois 62794-9420.
26
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70
JULY 1. 1992
with documentedabsenceof thrombolysis (patients with intravenous streptokinase) or at the time of revascularization (both intracoronary and intravenous streptokinase). Early revascularization was routinely recommended for patients with an open infarct artery and critical stenosis of the infarct artery (reduction of luminal diameter by 170%). Revascularization was performed during the samehospitalization as mn as it could be scheduled and after clotting function had returned to normal. Percutaneous transluminal coronary angioplasty or coronary artery bypass surgery was performed using standard techniques, and details of our early experience have been reported.2T3Radionuclide angiography was performed in 146 of the 180 survivors of myocardial infarction before hospital discharge (142 patients), or within 2 months (4 patients). No attempt was made to standardize medical therapy after discharge, although all patients were treated with daily aspirin. Follow-up was conducted between December 1988 and April 1989. This seriesincluded all patients treated with thrombolytic therapy before September 1983. Follow-up was achievedfor all 180 survivors of hospitalization in the seriesby a nurse researcherby telephone or clinic visit. The primary end point was cardiac death defined as sudden death, death after AMI, or death after hospitalization for congestiveheart failure. The effect of each clinical and angiographic variable on cardiac mortality for patients surviving hospitalization was analyzed by chi-square and t test for unpaired variables. Cox’s proportional hazards regressionwas subsequently used to analyze the relation (if any) among clinical and angiographic variables affecting survival4 The Cox model was also used to calculate an estimated survival function adjusted for the combination of variables found to be significant. An additional regressionanalysis was performed for a subset of patients having predischarge radionuclide angiograms. All values are expressedas mean f 1 standard deviation.
RESULTS During 24 months, 192 patients with transmural myocardial infarction were treated with intracoronary (n = 75) or intravenous (n = 117) streptokinase.Twelve patients (6%) died before hospital discharge (2,3). Follow-up data were obtained for all 180 survivors of hospitalization a minimum of 5 years after myocardial infarction. During follow-up, 21 of the 180 patients died (11.7%). Average follow-up for the 159 survivors was 74 f 7 months. There were 16 cardiac deaths (9%) (Figure 1); 7 had recurrent AMI, 6 had congestive heart failure, 2 died suddenly, and 1 died after coronary bypasssurgery. Subsequentmortality analysis will consider only these 16 patients with cardiac death. Clinical characttistii: The hospital course of the initial cohort of 192 patients has been descrk-12y3 The averagepatient in this follow-up study of 180 survivors was aged 55 f 10 years, and clinical characteristics are outlined in Table I. All patients had ST-segmentelevation and chest pain at the time of streptokinasetherapy and, as reported, all subsequentlyhad elevation of creatine kinase.2Streptokinase therapy was begun 202 f 92 minutes after onset of chest pain. An open infarct artery was identified at cardiac catheterization in 137 of the 180 survivors of hospitalization (76%). Early angiography was attempted in the 111 patients treated with intravenous streptokinase;the interval from initiation of therapy to catheterization was 1 to 4 hours for 46 (41%), 4 to 12 hours for 20 (18%), 12 to 24 hours for 19 (17%) 24 to 48 hours for 17 (15%) and 2 to 4 days for 9 (8%). Early revascularization while in the hospital after AM1 was performed an average of 3 f 2 days after streptokinase treatment in 111 of the 137 patients (81%) with an open infarct artery at catheterization; 94 (69%) had coronary artery bypass surgery and 17 (13%) had successfulpercutaneoustransluminal coronary angioplasty. Bypass patients received 3 f 1.4 grafts per patient. Angioplasty was attempted in 22 pa-
“0°%
0.80-
,x0.60-
0.20 I I 0.00
I 0
I 12
I 24
I 36 Months
I 48
I 60
SIX-YEAR SURVIVAL AFTER CORONARY THROMBOLYSIS
I 72
27
group for purposesof analysis), and 1 had angioplasty of a noninfarct artery. Univariate prediitors of survivab Clinical features No. of Cardiac Death and follow-up cardiac mortality are summarized in TaPts. p Value (%I ble I. Significant univariate predictors of survival in16 (9%) All patients 180 cluded location of myocardial infarction, follow-up ejec3 (9%) 32 NS* Age 2 65 years tion fraction, diabetes,Killip class on admission and an 5 (15%)/11 (7%) Women/men 331147 NS open infarct artery. Location of infarction and subse 11(9%) 119 NS* Cigarette uset quent left ventricular function were closely related, paNS* Systemic hypertension 63 7(11%) 27 7 (26%) <0.001* Diabetes mellitus tients with anterior infarction had a predischargeejec 32 5 (16%) NS* Prior AMI tion fraction of 41 f 15, and those with inferior infarc10 (15%) 66 Anterior AMI tion had a predischargeejection fraction of 58 f 11 (p 114 0.025 Inferior AMI 6 (5%)
28
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70
JULY 1, 1992
Closed, inferior
Closed, Anterior
gO.SOz Aa 2 2 a 0.40-
0.20-
0.00
1 0
I 12
I 24
I 36 Months
I 48
I 60
I 72
Closed, Inferior
Closed, Anterior
0.20-
0.00
I 0
I 12
passgrafting as an independent predictor of late survival, possibly becausemore of our patients had revasculark&on (68 vs 48%) and bypassgrafting (57 vs 22%) when comparedwith Mathey’s study. A higher percentage, if not most patients who might have benefited from bypass grafting, underwent it in our trial. Both of these studies identified an open infarct artery as a predictor of survival independentof changesin left ventricular function. Whereas the results support the “open artery hypothesis,” they do not prove it, as neither was a controlled study.5-9Furthermore, the high associationof an open infarct artery and revascularization in these studies makes it hard to determine which of the 2, reperfusion or revascularization, is most important in determining long-term outcome. Mathey claimed that bypasssurgery helped. Our results do not support that claim. On the other hand, the Thrombolysis in Myocardial Infarction trial, a controlled study of
I 24
I 36 Months
I 48
I 60
1
72
“revascularization” after reperfusion therapy, failed to show any short-term survival benefit.10But this study routinely usedearly angioplasty of the culprit lesion, not coronary bypasssurgery, the approach most often taken in our study and the Mathey trial. Our data provide an estimation of late prognosisafter AM1 using a treatment strategy of thrombolysis and early revascularization. There are no late survival data for patients having medical treatment or angioplasty as routine therapy after coronary thrombolysis. The present results should not be used to advise such patients about prognosis,since late mortality must be influenced both by thrombolysis and subsequenttherapy. REFERENCES 1. Mathey DG, Schafer J, SheehanFH, Krebber H, Justen M, Rodewald G, Dodge HT, Bleifeld W. Improved survival up to four years after early coronary thrombolysis. Am J Cmdiol 1988;61:524-529.
SIX-YEAR SURVIVAL AFTER CORONARY THROMBOLYSIS
29
2. Taylor GJ, Mike11FL, MosesHW, Dove JT, BatchelderJE, Thug A, Hansen S, Wellons HA, Schneider JA. Intravenous versus intracoronary streptokinase therapy for acute mywardiil infarction in community hospitals.Am J Cardiol 1984;54:256-260.
1. Sutton JM, Taylor GJ, Mike8 FL, Moses HW, Korsmeyer C, Dove JT, Batchelder JE, Wellons HA, SchneiderJA. Thrombolytic therapy followed by early revascularization for acute myocardiil infarction. Am J Cardiof 198657: 1227-1231. 4. Morrison DF. Multivariate Statistical Methods. New York: McGraw-Hill, 1%7:130-133.
5. Braunwald E. Myocardiil reperfusion,limitation of infarct size, reduction of left ventricular dysfunction, and improved survival: should the paradigm be expanded?Circulation 1989;79:441-444. 6. Califf RM, Topol EJ, Gersh BJ. From myocardiil salvageto patient salvagein acute myccardial infarction: the role of reperfusiontherapy (editorial). J Am CON
30
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70
Cardiol 1989;14:1382-1388. 7. KennedyJW, Ritchie JL, Davis KB, Fritz JK. WesternWashingtonintracoronary streptokinase in mywardial infarction trial. N Engl J Med 1983;309: 1477-1482. 8. Ritchie JL, Davis KB, Williams DL, Caldwell J, Kennedy W. Global and regional left ventricular function and tomographic radionuclide perfusion: The Western Washington randomized trial of intracoronary streptokinasein acute mvocardial infarction. Circulation 1984:70:867-875. 9: Dalen JE, Gore JM, Braunwald E, Borer J, Goldberg RJ, PassamaniER, Forman S, Knatterud G, and the TIMI Investigators.Six- and 1Zmonth followup of the phase I thrombolysis in myocardial infarction (TIMI) trial. Am J Cardiol 1988;62:179-185. 10. The TIMI Study Group. Comparisonof invasiveand conservativestrategies after treatment with intravenoustissueplasminogenactivator in acute myocardial infarction. N Engf J h&d 1989320~618-627.
JULY 1. 1992