- Email: [email protected]
Bedside markers of coronary artery patency and short-term prognosis of patients with acute myocardial infarction and thrombolysis
Bedside markers of coronary artery patency and short-term prognosis of patients with acute myocardial infarction and thrombolysis Ramón Corbalán, MD,a Juan C. Prieto, MD,a Eduardo Chavez, MD,a Carolina Nazzal, RN,a Francisco Cumsille, PHD,a and Mitchell Krucoff, MDb Santiago, Chile, and Durham, NC
Background In this study we have evaluated the prognostic power of noninvasive markers of coronary artery reperfusion in patients with acute myocardial infarction who were treated with intravenous streptokinase. Methods In 967 consecutive patients with acute myocardial infarction who were treated within 6 hours of symptoms, we analyzed the prognostic power of resolution of chest pain and ST-segment elevation >50% at 90 minutes, abrupt creatine kinase rise before 12 hours, and T-wave inversion in infarct-related electrocardiographic leads within the first 24 hours after thrombolysis.
Results Global in-hospital mortality rate was 12.0%. Each reperfusion marker was associated with improved outcome. Multivariate logistic regression analysis showed that 3 of the 4 markers of coronary artery reperfusion were significantly and independently associated to low in-hospital mortality rate. The presence of early T-wave inversion was associated with the lowest in-hospital mortality rate (odds ratio 0.25, confidence interval 0.10-0.56). When all markers of coronary artery reperfusion were included in the regression model, T-wave inversion (odds ratio 0.29, confidence interval 0.11-0.68) and abrupt creatine kinase rise (odds ratio 0.36, confidence interval 0.16-0.77) continued to be significantly associated with better outcome.
Conclusion A systemic analysis of noninvasive markers of coronary artery reperfusion can provide the clinician with an excellent tool to predict clinical outcomes when treating myocardial infarction. (Am Heart J 1999;138:533-9.)
Short- and long-term prognosis of patients with acute myocardial infarction (MI) treated with thrombolysis or coronary angioplasty is strongly related to achieving early patency of the culprit coronary artery. This has been well documented in multicenter trials in which coronary angiography has been performed shortly after thrombolytic therapy1-4 or when both primary coronary angioplasty and thrombolysis have been compared as alternative treatments for MI.5,6 The most commonly used bedside markers for coronary artery reperfusion after thrombolysis have been the relief of chest pain, resolution of ST-segment elevation, development of cardiac arrhythmias, and early creatine kinase (CK) rise.7-11 More recently, Twave inversion in the infarct-related electrocardiogram (ECG) leads within the first 24 hours has been shown as a reliable index of coronary artery reperfusion in patients with acute MI.12,13 If these noninvasive indexes are reliable in predicting coronary artery From athe GEMI Group, Santiago, Chile, and bDuke University Clinical Research Institute, Durham. Submitted August 24, 1998; accepted January 6, 1999. Reprint requests: Ramón Corbalán, MD, Departamento de Enfermedades Cardiovasculares, Hospital Clinico Pontificia, Universidad Católica de Chile, Marcoleta 367, Santiago, Chile. Copyright © 1999 by Mosby, Inc. 0002-8703/99/$8.00 + 0 4/1/97500
reperfusion, they should have an impact on short- and long-term prognosis after MI. In this study we prospectively analyzed the prognostic value of these indexes of coronary artery reperfusion in a series of 967 consecutive patients with acute MI who were treated with intravenous streptokinase within the first 6 hours of symptoms.
Methods In August 1993 we organized a network of 37 Chilean hospitals that started a prospective registry of cases with acute MI, thus originating the GEMI group (Grupo de Estudio Multicentrico del Infarto del Miocardio). At each hospital we had an investigator and a coinvestigator who were responsible for the study. In this study we included all patients with acute MI who were admitted within the first 6 hours of symptom onset and who were treated with 1.5 million U of intravenous streptokinase. Clinical diagnosis of MI was based on the presence of prolonged chest pain, characteristic ST-segment elevation in 2 or more contiguous ECG leads, and early CK rise. Between September 1993 and April 1995 2957 patients were registered with new MI. During this period 967 consecutive patients with new Q-wave MI presenting within the first 6 hours with acute chest pain and ST-segment elevation in the infarct-related ECG leads were intravenously treated with streptokinase 1.5 million U during a 60-minute period, oral aspirin 200 mg, and a bolus of 5000 U intravenous heparin. Mean time to treatment was 4.2 hours.
American Heart Journal September 1999
534 Corbalán et al
Table I. Baseline characteristics of patients who underwent thrombolysis (n = 967) Age (mean ± SD) Sex (M/F) Coronary risk factors (%) Hypertension Hypercholesterolemia Diabetes Smoking Obesity Family history Prior cardiac history (%) Previous MI Angina pectoris Heart failure CABG PTCA Admission characteristics (%) Anterior MI Inferior MI Killip class I-II Killip class III-IV In-hospital mortality rate
Table II. Adjunctive medical treatment during hospitalization in all patients (n = 967)
58.5 ± 12 80.3%/19.7% 39.6 19.5 15.9 48.1 18.9 15.6 9.6 37.7 3.2 1.8 0.6
% Medications Aspirin IV heparin β-blockers IV nitrates Oral nitrates ACE inhibitors Calcium antagonists Procedures Coronary angiography PTCA CABG
95 73 42 64 50 34 15 34 4.6 8.5
Abbreviations as in Table I.
50.0 48.8 92.7 7.3 12.0
CABG, Coronary artery bypass grafting; PTCA, percutaneous transluminal coronary angioplasty.
The following variables were analyzed in all patients. Coronary risk factors. We recorded hypertension, hyperlipidemia and diabetes mellitus, smoking, and family history of coronary artery disease, which was considered positive in the case of MI occurring before the age of 55 in a close relative. Clinical and demographic characteristics. Age and sex distribution, prior cardiac history, location of MI (anterior or inferior), and Killip class at admission were recorded. Baseline characteristics of patients treated with thrombolysis are shown in Table I. Cardiovascular events. Development of Killip class III or IV heart failure, supraventricular arrhythmias (sick sinus syndrome, atrial fibrillation or flutter), ventricular arrhythmias (ventricular tachycardia or fibrillation), conduction disturbances (new bundle branch block, second- or third-degree atrioventricular block), mechanical complications (acute mitral regurgitation, ventricular septal defect, or myocardial rupture), and in-hospital death were considered cardiovascular events.
out the hospital stay. A decrease of 50% of the pretreatment ST-segment elevation in the infarct-related ECG leads at 90 minutes of instituting therapy was considered a marker of reperfusion. Inversion of T-waves >0.5 mm below the baseline within the first 24 hours of thrombolytic therapy in all the corresponding infarct-related ECG leads, with previous ST-segment elevation, was considered a marker for coronary artery reperfusion. Interpretation of ECG changes was performed at each clinical center and submitted in a prespecified clinical record form to a central data base. Change in CK enzyme levels. Blood samples were taken at baseline, 6, 12, and 24 hours after admission. The elevation of CK in a proportion >60% with respect to its maximum value before 12 hours of starting thrombolytic therapy was also considered a marker of reperfusion. When there was no information available about each one of the reperfusion markers, or if this was uninterpretable, we considered this as missing data and these cases were excluded from the analysis. This occurred in 14 cases (1.4%) for relief of chest pain, in 23 cases (2.5%) for ST-segment resolution, in 63 cases (6.8%) for T-wave inversion, and in 64 cases (6.9%) for CK rise. The management of the patients after thrombolytic therapy and the performance of revascularization procedures did not differ among groups. The percentage of medications administered and coronary angiograms and revascularization procedures performed in all patients is shown in Table II.
Markers of coronary artery reperfusion
Statistical analysis
The noninvasive criteria for coronary artery reperfusion after thrombolysis were analyzed as follows. Pain relief. All patients were asked to grade the intensity of their chest pain on a scale from 0 to 10 before, during, and up to 90 minutes after thrombolytic therapy. Abrupt resolution of chest pain or more than 50% reduction of its intensity were considered criteria for reperfusion. When patients received opiates these data were considered uninterpretable.
Data are presented as mean ± SD and percentage. Categoric variables were analyzed by the χ2 test and through logistic regression model. The number of reperfusion criteria and mortality relation was analyzed by Cochran Armitage test of trend. A P value of <.05 was considered significant. SAS statistical software was used.
Changes in ST-segment elevation and in T-wave morphologic features. A 12-lead ECG was taken before starting throm-
Results
bolytic therapy and served as the pretreatment baseline ECG thereafter. The ECG lead with the most prominent ST-segment elevation was monitored and registered every 5 minutes during the first 90 minutes after starting thrombolysis. Thereafter, a complete ECG was recorded at 24 hours and daily through-
Noninvasive indexes of coronary artery reperfusion and cardiac death The overall in-hospital mortality rate was 12.0%. In the absence of one criteria for coronary artery reperfusion, cardiac mortality reached 32.1%. This mortality
American Heart Journal Volume 138, Number 3, Part 1
Corbalán et al 535
Figure 1
Comparison of mortality rate in absence or presence of reperfusion criteria. Cochran Armitage trend test shows a significant decrease in mortality rate up to presence of 3 reperfusion criteria. Number of patients in each category is shown in parentheses.
Table III. Predictive value of the reperfusion markers for in-hospital death Reperfusion markers Relief of chest pain Resolution of ST-segment elevation CK rise <12 hr Early T-wave inversion
OR (univariate analysis)
OR (multivariate analysis*)
OR (multivariate analysis†)
0.37 (0.24-0.56) 0.33 (0.21-0.50) 0.29 (0.18-0.45) 0.19 (0.11-0.34)
0.51 (0.27-0.95) 0.62 (0.33-1.17) 0.38 (0.19-0.76) 0.25 (0.10-0.56)
0.88 (0.37-2.12) 1.48 (0.37-2.12) 0.36 (0.16-0.77) 0.29 (0.11-0.68)
*Adjusted for age, sex, risk factors, Killip class, MI location, arrhythmias, conduction disturbances, mechanical complications. †Adjusted for age, sex, risk factors, Killip class, MI location, arrhythmias, conduction disturbances, mechanical complications, and other reperfusion criteria.
rate progressively and significantly (P < .001) decreased in relation to the presence of one more criteria and when all 4 coronary reperfusion criteria were present (Figure 1). In-hospital mortality rate was analyzed according to the presence or absence of each one of the reperfusion criteria. This was significantly lower every time one of the criteria was present. Thus persistence or resolution of chest pain was associated with a mortality rate of 17.3% versus 7.5%, persistence or resolution of >50% ST-segment elevation with a mortality rate of 17.3% versus 6.6%, late or early elevation of CK enzyme levels with a mortality rate of 19.0% versus 6.2%, and positive or negative T waves at 24 hours with a mortality rate of 17.6% and 3.8% respectively, (P < .001 for all 4 variables). The prognostic value of the different criteria for coronary reperfusion was examined through multivariate logistic regression analysis adjusted by age, sex, coronary risk factors, location of MI, development of heart
failure according to Killip class, arrythmias and/or conduction disturbances, and mechanical complications. We found that resolution of chest pain, early rise in CK enzyme levels, and inversion of the T wave within the first 24 hours were independently and significantly associated to lower mortality rate (Table III). Inversion of T waves within the first 24 hours was the most significant independent predictor for a lower in-hospital mortality rate. When the 4 markers of coronary artery reperfusion were included in the regression model we found that Twave inversion (odds ratio [OR] 0.29, confidence interval [CI] 0.11-0.68) and abrupt CK rise (OR 0.36, CI 0.16-0.77) continued to be significantly and independently associated with lower in-hospital mortality rate. There were no interactions between any of the adjunctive drug treatments given in the first 24 hours and the noninvasive markers of coronary artery reperfusion. Determinations of CK enzyme levels in patients who died before 12 hours and of T-wave inversion in
American Heart Journal September 1999
536 Corbalán et al
Figure 2
Incidence of Killip class III to IV heart failure in the presence or absence of specific noninvasive reperfusion criteria.
those who died before 24 hours were considered as missing data.
cantly lower for each one of the 4 reperfusion markers analyzed (Table IV).
Noninvasive indexes of coronary artery reperfusion and development of in-hospital heart failure
Discussion
Each one of the different criteria for coronary artery reperfusion, when present, was associated with a lower incidence of in-hospital Killip class III or IV heart failure (Figure 2).
Noninvasive indexes of coronary artery reperfusion and other in-hospital cardiovascular events All 4 indexes of coronary artery reperfusion were associated with a significantly lower incidence of mechanical complications. The incidence of supraventricular arrhythmias was significantly lower when there was resolution of chest pain and when T waves were inverted at 24 hours. The incidence of in-hospital ventricular arrhythmias was significantly lower when there was resolution of chest pain and of >50% ST-segment elevation. Finally, the incidence of mechanical complications was signifi-
It is highly desirable to have reliable, noninvasive methods to evaluate the prognosis of patients with acute MI. This is especially important in patients with MI who are undergoing thrombolytic therapy in whom the goal is early coronary artery reperfusion. Shah et al14 evaluated, through coronary angiography performed at 24 hours, the value of bedside markers of coronary artery reperfusion after thrombolysis. These authors found that pain resolution and reduction in the elevation of the ST segment equal or higher than 50%, together with an abrupt rise in CK enzyme levels, were associated to Thrombolysis In Myocardial Infarction grade 3 coronary blood flow. However, the same authors emphasized the fluctuating course of changes in chest pain and ST-segment elevation during thrombolysis. Therefore they proposed the need for continuous monitoring of these parameters. Krucoff et al15 proposed a continuous segment recovery analysis
American Heart Journal Volume 138, Number 3, Part 1
Corbalán et al 537
Table IV. Cardiovascular events and noninvasive indexes of coronary artery reperfusion Supraventricular arrhythmias (%) Persistent chest pain Resolution of chest pain P value Persistent ST elevation Resolution of ST elevation P value Late CK rise Early CK rise P value Positive T wave Negative T wave P value
19.0 13.7 <.04 16.8 15.1 NS 15.2 15.7 NS 15.8 14.2 NS
as an alternative method over serial ST-segment assessment to predict simultaneous angiographic patency. By using this technique, Veldkamp et al16 described different patterns of ST-segment changes during thrombolysis, depending mainly on the speed at which coronary artery reperfusion occurred. More recently, Krucoff et al17 have reported similar results. In addition, Schroeder et al18 have shown that the degree of ST-segment normalization has an important prognostic value in patients with acute MI after thrombolysis. According to these authors, patients with a partial (>70%) or complete normalization of ST-segment elevation have both significantly lower mortality rates and lower incidence of in-hospital complications during the course of MI. Previously, several authors had evaluated the importance of rapid release of CK enzymes after thrombolysis as a marker of coronary reperfusion. In all published reports this parameter has been found to be a highly sensitive marker of coronary reperfusion when its maximal value is reached within the first 12 hours of treatment.10,11,19,20 Inversion of the T wave in the infarct-related ECG leads within the first 24 hours after MI has been described as another marker of coronary artery reperfusion, although it becomes evident later than the others. Thus Matetzky et al,12 through a coronary angiogram performed between 36 and 48 hours after MI, found that T-wave inversion was associated with Thrombolysis In Myocardial Infarction grade 3 coronary blood flow in a series of 100 patients treated with thrombolysis. These authors explored the possibility that T-wave inversion at 3 and 12 hours after thrombolytic therapy could help predict coronary artery patency, but they failed to establish a correlation. Castro et al13 compared T-wave inversion with classic reperfusion indexes in patients treated with thrombolysis and found that Twave inversion at 24 hours was the most significant independent predictor of coronary artery patency.13
Ventricular arrhythmias (%) 17.9 11.8 <.002 17.5 12.1 <.03 14.6 14.2 NS 14.9 12.9 NS
Mechanical complications (%) 4.8 1.6 <.01 5.0 1.2 <.001 5.3 1.3 <.001 4.6 0.5 <.001
Based on this evidence, we found of interest to register the 4 coronary reperfusion criteria in the medical records of patients admitted to our hospital network (GEMI group). Our hypothesis was that patients treated with thrombolysis who had positive indexes of coronary artery reperfusion should have a more favorable prognosis after thrombolytic therapy. The analysis of 967 consecutive patients seen with Q-wave MI treated with streptokinase has validated this hypothesis. Although the overall in-hospital mortality rate of patients treated with streptokinase was 12.0%, it reached 32.1% when none of the coronary artery reperfusion criteria were present and fell to 4.2% when 3 or more indexes were present. Univariate analysis of in-hospital mortality rate in relation to the presence of each one of the coronary reperfusion indexes showed that resolution of chest pain and ST-segment elevation, as well as early rise of CK enzyme levels, were associated with mortality rates ranging from 6.1% to 7.6%. On the other hand, the inversion of the T wave in the infarct-related ECG leads at 24 hours was associated with an even lower mortality rate of only 3.9%. Interestingly the incidence of mechanical complications was significantly lower in the presence of each one of the reperfusion markers. In this regard the incidence of mechanical complications reached only 0.5% when there was early T-wave inversion. These findings suggest that both the incidence of death and mechanical complications derive benefits through similar mechanism from coronary artery reperfusion. Our results confirm recent findings of Schroeder et al.18 In the INJECT trial, in which the thrombolytic agents reteplase and streptokinase were compared, these authors analyzed the prognostic value of elevated ST segment at different levels of resolution. They found that complete resolution of ST-segment elevation was associated with a mortality rate of 2.5%, and when
American Heart Journal September 1999
538 Corbalán et al
there was only partial resolution of ST-segment elevation mortality rate was 4.3%. In turn, a persistently elevated ST segment was associated with a mortality rate of 17.5%. Differences in mortality rates were similar both in patients with anterior and inferior MI and in the presence of new bundle branch block. In addition, Krucoff et al21 have shown that continuous 12-lead ST-segment monitoring recovery analysis is more predictive of clinical outcomes than early coronary angiography after thrombolytic therapy in a meta-analysis of 4 different multicenter trials of patients with acute MI. On the other hand, Matetzki et al12 analyzed the value of early T-wave inversion as a marker of coronary reperfusion and found that its presence at 24 hours was significantly associated with a lower incidence of both complications and in-hospital death. This is similar to our findings. It is well known that different factors may influence MI death, such as older age, female sex, the presence of diabetes mellitus, anterior location of MI, development of cardiac failure, the presence of complex arrhythmias or conduction disturbances, and mechanical complications. A logistic regression analysis comparing coronary reperfusion indexes and death adjusted for these factors confirmed that early T-wave inversion has a significant independent prognostic value for a lower mortality rate. Resolution of chest pain and the early rise in CK enzyme levels also had an independent prognostic value. When the 4 markers of coronary artery reperfusion were included in the regression model, early Twave inversion and abrupt CK rise continued to be independently associated with a better outcome. Each one of the coronary reperfusion markers analyzed in our series has limitations. Adequate interpretation of changes in chest pain, because of its subjective nature, might be different, especially if analgesics or opiates have been used. Shah et al14 described the fluctuations that may occur in ST-segment elevation both during and after thrombolysis,14 which was confirmed by other authors.15,16 Rapid elevation of CK enzyme levels is fairly reliable as a reperfusion parameter, but it has the drawback of the delay by which the results are obtained. The same limitation applies to inversion of the T wave, which becomes evident 24 hours after treatment. Although this marker appears to have the most powerful prognostic value, its late occurrence could limit its clinical usefulness. In addition, we have previously described the finding of false-positive results, that is, patients with an obstructed coronary artery and negative T waves at 24 hours that may occur when collateral circulation develops at the infarcted area.13 An important limitation of our study was the absence of a core laboratory for the interpretation of serial ECG changes. For this reason we preferred to use more than 50% reduction of the pretreatment ST-segment elevation instead of the >70% reduction of ST-segment elevation,
which has been proposed by Schroeder et al.18 This method could explain why resolution of ST-segment elevation was not an independent predictor of death in the presence of other reperfusion markers. Another explanation could be that the ST-segment changes might have a fluctuating course after 90 minutes of thrombolytic therapy, reflecting alternating episodes of coronary artery reperfusion and coronary artery reocclusion. It would make sense, then, that reperfusion markers that occur late in time, such as CK rise and T-wave inversion, were independent predictors of death. According to our results the predictive information content of noninvasive markers of coronary artery reperfusion appears to be inversely related to the acuity with which they become available. Thus T-wave inversion is most predictive but not available until 24 hours; the CK enzyme peak is next predictive, but not available until 12 hours; the ST segment and chest pain are next, analyzed at 90 minutes after thrombolysis. It could be proposed, then, that the combined analysis of coronary artery reperfusion markers, over time, offers more relevant clinical information than the individual assessment of each marker alone. Thus ST-segment recovery can give a prognostic insight during the first 90 minutes of initiating therapy; CK determinations can amplify that information between 6 and 12 hours, and inversion of T waves can increase that prognostic information even further between 12 and 24 hours. The prognostic information, thus obtained, could contribute to optimize clinical decisions such as performing early or late coronary angiography, complementary medical therapy, and so on. It is possible that even more optimal combination and sequencing of new noninvasive markers could allow their prognostic information to enter the patient care decision-making environment. Availability of continuous ST-segment and T-wave monitoring during the first 24 hours of reperfusion therapy or implementation of new biochemical markers of myocardial injury such as determinations of troponins, CK mass, or myoglobin may help gain more precise prognostic information even earlier during the clinical course of MI. Systematic and careful analysis of noninvasive markers and their temporal sequence will certainly provide the clinician good tools at the bedside to predict clinical outcomes when treating MI.
References 1. Simes R, Topol EJ, Holmes DR, White HD, et al for the GUSTO-I investigators. Link between the angiographic substudy and mortality outcomes in a large randomized trial of myocardial reperfusion. Circulation 1995;91:1923-8. 2. Galvani M, Ottani F, Ferrini D, Sorbello F, Rusticali F. Patency of the infarct-related artery and left ventricular function as the major determinants of survival after Q wave acute myocardial infarction. Am J Cardiol 1993;71:1-7.
American Heart Journal Volume 138, Number 3, Part 1
3. De Boer MJ, Suryapranata H, Hoorntje JCA, et al. Limitation of infarct size and preservation of left ventricular function after primary angioplasty compared with intravenous streptokinase in acute myocardial infarction. Circulation 1994;90:753-61. 4. Grines CL, Browne MF, Marco J, et al. A comparison of immediate angioplasty with thrombolityc theraphy for acute myocardial infarction. N Engl J Med 1993:328:673-9. 5. White HD, Cross DB, Elliot JM, Morris RM, Yee TW. Long-term prognostic importance of patency of the infarct-related coronary artery after thrombolytic therapy for acute myocardial infarction. Circulation 1994;89:61-7. 6. Anderson JL, Karagounis LA, Becker LC, Sorensen SG, Menlove R, for the TEAM-3 Investigators. TIMI perfusion grade 3 but not grade 2 results in improved outcome after thrombolysis for myocardial infarction: ventricolographic, enzimatic and electrocardiographic evidence from the TEAM-3 study. Circulation 1993;873:1829-39. 7. Clemmensen P, Ohman EM, Sevilla DC, Peck S, Wagner NB. Changes in standard electrocardiographic ST segment elevation predictive of successful reperfusion in acute myocardial infarction. Am J Cardiol 1990;66:1407-11. 8. Zehender M, Utzolino S, Furtwängler A, et al. Time course and interrelation of reperfusion-induced ST changes and ventricular arrhythmias in acute myocardial infarction. Am J Cardiol 1991;68:1138-42. 9. Hackworthy RA, Vogel MB, Harris PH. Relationship between changes in ST segment elevation and patency of the infarct related coronary artery in acute myocardial infarction. Am Heart J 1986;112:279-84. 10. Gore JM, Roberts R, Ball S, et al. Peak creatine kinase as a measure of effectiveness of thrombolytic therapy in acute myocardial infarction. Am J Cardiol 1987;59:1234-8. 11. Garabedian HD, Gold HJ, Yasuda T, et al. Detection of coronay artery reperfusion with creatine kinase-MB determinations during thrombolytic therapy: correlation with angiography. J Am Coll Cardiol 1988;11:729-34. 12. Matetzky S, Barabash GI, Shahar A, et al. Early T wave inversion after thrombolytic therapy predicts better coronary perfusion: clinical and angiographic study. J Am Coll Cardiol 1994;2:378-83. 13. Castro P, Corbalan R, Kunstmann S, Rubio R, Marchant E, Pumarino JI, et al. Early T wave inversion: a bedside marker of coronary artery patency after thrombolysis in patients with acute myocardial infarction. Rev Ch Cardiol 1992;11:94-102. 14. Shah PK, Cercek B, Lew A, Ganz W. Angiographic validation of bedside markers of reperfusion. J Am Coll Cardiol 1993;21:55-61. 15. Krucoff MW, Croll MA, Pope JE, Pieper KS, Kanani PM, Granger CB, et al. Continuously updated 12-lead ST-segment recovery analysis for myocardial infarct artery patency assessment and its correlation wtih multiple simultaneous early angiographic observation. Am J Cardiol 1993;71:145-51. 16. Veldkamp R, Green C, Wilkins M, Pope W, et al. Comparison of continous ST-segment recovery analysis with methodos using static electrocardiograms for invasive patency assessment during acute myocardial infarction. Am J Cardiol 1994;73:1069-74. 17. Krucoff MW, Croll MA, Pope J, Granger CB, et al for the TAMI-7 study group. Continuous 12-lead ST-segment recovery analysis in the TAMI 7 Study. Performance of a noninvasive method for real-time detection of failed myocardial reperfusion. Circulation 1993;88:437-46. 18. Schroder R, Wegsscheider K, Schroder K, Dissmann R, MeyerSabellek W, for the INJECT trial group. Extent of early segment elevation resolution: a strong predictor of outcome in patients with acute myocardial infarction and a sensitive measure to compare thrombolytic regimens. J Am Coll Cardiol 1995;27:1657-64.
Corbalán et al 539
19. Lewis B, Ganz W, Laramee P, et al. Usefulness of a rapid initial increase in plasma creatine kinase activity as a marker of reperfusion during thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1988;62:20-4. 20. Ong L, Coromilas J, Zimmerman JM, et al. A physiologically based model of creatine kinase-MB release in reperfusion of acute myocardial infarction. Am J Cardiol 1989;64:11-5. 21. Krucoff MW, Green CL, Trollinger KM, and Pope JE. Angiography “gold” standard or “tin” standard of reperfusion? A comparison with continuous 12 lead ST-segment recovery for prediction of clinical outcomes after thrombolytic therapy for acute myocardial infarction [abstract]. Circulation 1996;94(suppl):2572.
Appendix Gemi Group Participants (Grupo De Estudio Multicentrico Del Infarto, Sociedad Chilena De Cardiologia) Hospital Regional Arica: Ivan Criollo, Mario Gatica, Hernan Figueroa. Hospital Regional Iquique: Patricio Sierralta, Ignacio Auger, Virginia Araya. Hospital Regional Antofagasta: Juan Antonio Cotoras. Clinica Antofagasta: Guillermo Illanes. Hospital Regional Copiapo: Guillermo Ananias, Walter Evans. Hospital Regional De Coquimbo: Carlos Saldias, Hans Gonzalez. Hospital Regional De La Serena: Carlos Echeverria, Claudio Bugueño. Hospital Naval Viña Del Mar: Fernando Cardenas, Marcos Opazo. Hospital Gustavo Fricke: Mauricio Aninat, Carlos Raffo, Jorge Bartolucci. Clinica Reñaca: Rienzi Diaz, Monica Carvajal. Clinica Miraflores: Pedro Chadid. Regional Rancagua: Leopoldo Manriquez. Hospital Fusatde Rancagua: Enrique Buzeta, Rafael Castillo. Hospital Regional Talca: Enrique Mercadal, Patricio Vildosola. Hospital Herminda Martin De Chillan: Gonzalo Marin, Gustavo Yanine. Hospital Guillermo Grant Concepcion: Carlos Otero. Hospital Higueras Talcahuano: Hernan Diaz, Nelson Alburquerque. Hospital Naval De Talcahuano: Alejandro Dapelo, Carlos Nicolich. Hospital Regional Temuco: Fernando Lanas, Mario Santibañez, Luis Sepulveda. Hospital Base Valdivia: Eduardo Garces, Herminia Riquelme. Hospital Regional Osorno: Sergio Potthoff. Hospital Regional Puerto Montt: Luis Felipe Del Campò, Jame Venegas. Hospital Regional Punta Arenas: Guillermo Araneda, Bogdam Liberon. Hospital Asistencia Publica: Eduardo Chavez, Mario Ratkhmap, Marcela Argandoña. Hospital Salvador: Alvaro Puelma, Francisco Manzur. Hospital Jose Joaquin Aguirre: Jorge Yovanovich, Ruben Aguayo. Hospital Universidad Catolica: Ivan Godoy, Carlos Almendares. Hospital Barros Luco Trudeau: Ligia Gallardo. Hospital Sotero Del Rio: Claudio Fernandez, Ximena Muñoz. Hospital San Borja Arriaran: Rolando Ricciulli. Hospital Del Profesor: Mario Hassi. Clinica Alemana: Rene Etchegaray, Alejandro Abufele. Clinica Santa Maria: Sonia Kunstmann, Hernan Chamorro. Clinica Las Lilas: Alberto Gonzalez. Hospital Dipreca: Barbara Clericus, Milton Alcaino.
Background In this study we have evaluated the prognostic power of noninvasive markers of coronary artery reperfusion in patients with acute myocardial infarction who were treated with intravenous streptokinase. Methods In 967 consecutive patients with acute myocardial infarction who were treated within 6 hours of symptoms, we analyzed the prognostic power of resolution of chest pain and ST-segment elevation >50% at 90 minutes, abrupt creatine kinase rise before 12 hours, and T-wave inversion in infarct-related electrocardiographic leads within the first 24 hours after thrombolysis.
Results Global in-hospital mortality rate was 12.0%. Each reperfusion marker was associated with improved outcome. Multivariate logistic regression analysis showed that 3 of the 4 markers of coronary artery reperfusion were significantly and independently associated to low in-hospital mortality rate. The presence of early T-wave inversion was associated with the lowest in-hospital mortality rate (odds ratio 0.25, confidence interval 0.10-0.56). When all markers of coronary artery reperfusion were included in the regression model, T-wave inversion (odds ratio 0.29, confidence interval 0.11-0.68) and abrupt creatine kinase rise (odds ratio 0.36, confidence interval 0.16-0.77) continued to be significantly associated with better outcome.
Conclusion A systemic analysis of noninvasive markers of coronary artery reperfusion can provide the clinician with an excellent tool to predict clinical outcomes when treating myocardial infarction. (Am Heart J 1999;138:533-9.)
Short- and long-term prognosis of patients with acute myocardial infarction (MI) treated with thrombolysis or coronary angioplasty is strongly related to achieving early patency of the culprit coronary artery. This has been well documented in multicenter trials in which coronary angiography has been performed shortly after thrombolytic therapy1-4 or when both primary coronary angioplasty and thrombolysis have been compared as alternative treatments for MI.5,6 The most commonly used bedside markers for coronary artery reperfusion after thrombolysis have been the relief of chest pain, resolution of ST-segment elevation, development of cardiac arrhythmias, and early creatine kinase (CK) rise.7-11 More recently, Twave inversion in the infarct-related electrocardiogram (ECG) leads within the first 24 hours has been shown as a reliable index of coronary artery reperfusion in patients with acute MI.12,13 If these noninvasive indexes are reliable in predicting coronary artery From athe GEMI Group, Santiago, Chile, and bDuke University Clinical Research Institute, Durham. Submitted August 24, 1998; accepted January 6, 1999. Reprint requests: Ramón Corbalán, MD, Departamento de Enfermedades Cardiovasculares, Hospital Clinico Pontificia, Universidad Católica de Chile, Marcoleta 367, Santiago, Chile. Copyright © 1999 by Mosby, Inc. 0002-8703/99/$8.00 + 0 4/1/97500
reperfusion, they should have an impact on short- and long-term prognosis after MI. In this study we prospectively analyzed the prognostic value of these indexes of coronary artery reperfusion in a series of 967 consecutive patients with acute MI who were treated with intravenous streptokinase within the first 6 hours of symptoms.
Methods In August 1993 we organized a network of 37 Chilean hospitals that started a prospective registry of cases with acute MI, thus originating the GEMI group (Grupo de Estudio Multicentrico del Infarto del Miocardio). At each hospital we had an investigator and a coinvestigator who were responsible for the study. In this study we included all patients with acute MI who were admitted within the first 6 hours of symptom onset and who were treated with 1.5 million U of intravenous streptokinase. Clinical diagnosis of MI was based on the presence of prolonged chest pain, characteristic ST-segment elevation in 2 or more contiguous ECG leads, and early CK rise. Between September 1993 and April 1995 2957 patients were registered with new MI. During this period 967 consecutive patients with new Q-wave MI presenting within the first 6 hours with acute chest pain and ST-segment elevation in the infarct-related ECG leads were intravenously treated with streptokinase 1.5 million U during a 60-minute period, oral aspirin 200 mg, and a bolus of 5000 U intravenous heparin. Mean time to treatment was 4.2 hours.
American Heart Journal September 1999
534 Corbalán et al
Table I. Baseline characteristics of patients who underwent thrombolysis (n = 967) Age (mean ± SD) Sex (M/F) Coronary risk factors (%) Hypertension Hypercholesterolemia Diabetes Smoking Obesity Family history Prior cardiac history (%) Previous MI Angina pectoris Heart failure CABG PTCA Admission characteristics (%) Anterior MI Inferior MI Killip class I-II Killip class III-IV In-hospital mortality rate
Table II. Adjunctive medical treatment during hospitalization in all patients (n = 967)
58.5 ± 12 80.3%/19.7% 39.6 19.5 15.9 48.1 18.9 15.6 9.6 37.7 3.2 1.8 0.6
% Medications Aspirin IV heparin β-blockers IV nitrates Oral nitrates ACE inhibitors Calcium antagonists Procedures Coronary angiography PTCA CABG
95 73 42 64 50 34 15 34 4.6 8.5
Abbreviations as in Table I.
50.0 48.8 92.7 7.3 12.0
CABG, Coronary artery bypass grafting; PTCA, percutaneous transluminal coronary angioplasty.
The following variables were analyzed in all patients. Coronary risk factors. We recorded hypertension, hyperlipidemia and diabetes mellitus, smoking, and family history of coronary artery disease, which was considered positive in the case of MI occurring before the age of 55 in a close relative. Clinical and demographic characteristics. Age and sex distribution, prior cardiac history, location of MI (anterior or inferior), and Killip class at admission were recorded. Baseline characteristics of patients treated with thrombolysis are shown in Table I. Cardiovascular events. Development of Killip class III or IV heart failure, supraventricular arrhythmias (sick sinus syndrome, atrial fibrillation or flutter), ventricular arrhythmias (ventricular tachycardia or fibrillation), conduction disturbances (new bundle branch block, second- or third-degree atrioventricular block), mechanical complications (acute mitral regurgitation, ventricular septal defect, or myocardial rupture), and in-hospital death were considered cardiovascular events.
out the hospital stay. A decrease of 50% of the pretreatment ST-segment elevation in the infarct-related ECG leads at 90 minutes of instituting therapy was considered a marker of reperfusion. Inversion of T-waves >0.5 mm below the baseline within the first 24 hours of thrombolytic therapy in all the corresponding infarct-related ECG leads, with previous ST-segment elevation, was considered a marker for coronary artery reperfusion. Interpretation of ECG changes was performed at each clinical center and submitted in a prespecified clinical record form to a central data base. Change in CK enzyme levels. Blood samples were taken at baseline, 6, 12, and 24 hours after admission. The elevation of CK in a proportion >60% with respect to its maximum value before 12 hours of starting thrombolytic therapy was also considered a marker of reperfusion. When there was no information available about each one of the reperfusion markers, or if this was uninterpretable, we considered this as missing data and these cases were excluded from the analysis. This occurred in 14 cases (1.4%) for relief of chest pain, in 23 cases (2.5%) for ST-segment resolution, in 63 cases (6.8%) for T-wave inversion, and in 64 cases (6.9%) for CK rise. The management of the patients after thrombolytic therapy and the performance of revascularization procedures did not differ among groups. The percentage of medications administered and coronary angiograms and revascularization procedures performed in all patients is shown in Table II.
Markers of coronary artery reperfusion
Statistical analysis
The noninvasive criteria for coronary artery reperfusion after thrombolysis were analyzed as follows. Pain relief. All patients were asked to grade the intensity of their chest pain on a scale from 0 to 10 before, during, and up to 90 minutes after thrombolytic therapy. Abrupt resolution of chest pain or more than 50% reduction of its intensity were considered criteria for reperfusion. When patients received opiates these data were considered uninterpretable.
Data are presented as mean ± SD and percentage. Categoric variables were analyzed by the χ2 test and through logistic regression model. The number of reperfusion criteria and mortality relation was analyzed by Cochran Armitage test of trend. A P value of <.05 was considered significant. SAS statistical software was used.
Changes in ST-segment elevation and in T-wave morphologic features. A 12-lead ECG was taken before starting throm-
Results
bolytic therapy and served as the pretreatment baseline ECG thereafter. The ECG lead with the most prominent ST-segment elevation was monitored and registered every 5 minutes during the first 90 minutes after starting thrombolysis. Thereafter, a complete ECG was recorded at 24 hours and daily through-
Noninvasive indexes of coronary artery reperfusion and cardiac death The overall in-hospital mortality rate was 12.0%. In the absence of one criteria for coronary artery reperfusion, cardiac mortality reached 32.1%. This mortality
American Heart Journal Volume 138, Number 3, Part 1
Corbalán et al 535
Figure 1
Comparison of mortality rate in absence or presence of reperfusion criteria. Cochran Armitage trend test shows a significant decrease in mortality rate up to presence of 3 reperfusion criteria. Number of patients in each category is shown in parentheses.
Table III. Predictive value of the reperfusion markers for in-hospital death Reperfusion markers Relief of chest pain Resolution of ST-segment elevation CK rise <12 hr Early T-wave inversion
OR (univariate analysis)
OR (multivariate analysis*)
OR (multivariate analysis†)
0.37 (0.24-0.56) 0.33 (0.21-0.50) 0.29 (0.18-0.45) 0.19 (0.11-0.34)
0.51 (0.27-0.95) 0.62 (0.33-1.17) 0.38 (0.19-0.76) 0.25 (0.10-0.56)
0.88 (0.37-2.12) 1.48 (0.37-2.12) 0.36 (0.16-0.77) 0.29 (0.11-0.68)
*Adjusted for age, sex, risk factors, Killip class, MI location, arrhythmias, conduction disturbances, mechanical complications. †Adjusted for age, sex, risk factors, Killip class, MI location, arrhythmias, conduction disturbances, mechanical complications, and other reperfusion criteria.
rate progressively and significantly (P < .001) decreased in relation to the presence of one more criteria and when all 4 coronary reperfusion criteria were present (Figure 1). In-hospital mortality rate was analyzed according to the presence or absence of each one of the reperfusion criteria. This was significantly lower every time one of the criteria was present. Thus persistence or resolution of chest pain was associated with a mortality rate of 17.3% versus 7.5%, persistence or resolution of >50% ST-segment elevation with a mortality rate of 17.3% versus 6.6%, late or early elevation of CK enzyme levels with a mortality rate of 19.0% versus 6.2%, and positive or negative T waves at 24 hours with a mortality rate of 17.6% and 3.8% respectively, (P < .001 for all 4 variables). The prognostic value of the different criteria for coronary reperfusion was examined through multivariate logistic regression analysis adjusted by age, sex, coronary risk factors, location of MI, development of heart
failure according to Killip class, arrythmias and/or conduction disturbances, and mechanical complications. We found that resolution of chest pain, early rise in CK enzyme levels, and inversion of the T wave within the first 24 hours were independently and significantly associated to lower mortality rate (Table III). Inversion of T waves within the first 24 hours was the most significant independent predictor for a lower in-hospital mortality rate. When the 4 markers of coronary artery reperfusion were included in the regression model we found that Twave inversion (odds ratio [OR] 0.29, confidence interval [CI] 0.11-0.68) and abrupt CK rise (OR 0.36, CI 0.16-0.77) continued to be significantly and independently associated with lower in-hospital mortality rate. There were no interactions between any of the adjunctive drug treatments given in the first 24 hours and the noninvasive markers of coronary artery reperfusion. Determinations of CK enzyme levels in patients who died before 12 hours and of T-wave inversion in
American Heart Journal September 1999
536 Corbalán et al
Figure 2
Incidence of Killip class III to IV heart failure in the presence or absence of specific noninvasive reperfusion criteria.
those who died before 24 hours were considered as missing data.
cantly lower for each one of the 4 reperfusion markers analyzed (Table IV).
Noninvasive indexes of coronary artery reperfusion and development of in-hospital heart failure
Discussion
Each one of the different criteria for coronary artery reperfusion, when present, was associated with a lower incidence of in-hospital Killip class III or IV heart failure (Figure 2).
Noninvasive indexes of coronary artery reperfusion and other in-hospital cardiovascular events All 4 indexes of coronary artery reperfusion were associated with a significantly lower incidence of mechanical complications. The incidence of supraventricular arrhythmias was significantly lower when there was resolution of chest pain and when T waves were inverted at 24 hours. The incidence of in-hospital ventricular arrhythmias was significantly lower when there was resolution of chest pain and of >50% ST-segment elevation. Finally, the incidence of mechanical complications was signifi-
It is highly desirable to have reliable, noninvasive methods to evaluate the prognosis of patients with acute MI. This is especially important in patients with MI who are undergoing thrombolytic therapy in whom the goal is early coronary artery reperfusion. Shah et al14 evaluated, through coronary angiography performed at 24 hours, the value of bedside markers of coronary artery reperfusion after thrombolysis. These authors found that pain resolution and reduction in the elevation of the ST segment equal or higher than 50%, together with an abrupt rise in CK enzyme levels, were associated to Thrombolysis In Myocardial Infarction grade 3 coronary blood flow. However, the same authors emphasized the fluctuating course of changes in chest pain and ST-segment elevation during thrombolysis. Therefore they proposed the need for continuous monitoring of these parameters. Krucoff et al15 proposed a continuous segment recovery analysis
American Heart Journal Volume 138, Number 3, Part 1
Corbalán et al 537
Table IV. Cardiovascular events and noninvasive indexes of coronary artery reperfusion Supraventricular arrhythmias (%) Persistent chest pain Resolution of chest pain P value Persistent ST elevation Resolution of ST elevation P value Late CK rise Early CK rise P value Positive T wave Negative T wave P value
19.0 13.7 <.04 16.8 15.1 NS 15.2 15.7 NS 15.8 14.2 NS
as an alternative method over serial ST-segment assessment to predict simultaneous angiographic patency. By using this technique, Veldkamp et al16 described different patterns of ST-segment changes during thrombolysis, depending mainly on the speed at which coronary artery reperfusion occurred. More recently, Krucoff et al17 have reported similar results. In addition, Schroeder et al18 have shown that the degree of ST-segment normalization has an important prognostic value in patients with acute MI after thrombolysis. According to these authors, patients with a partial (>70%) or complete normalization of ST-segment elevation have both significantly lower mortality rates and lower incidence of in-hospital complications during the course of MI. Previously, several authors had evaluated the importance of rapid release of CK enzymes after thrombolysis as a marker of coronary reperfusion. In all published reports this parameter has been found to be a highly sensitive marker of coronary reperfusion when its maximal value is reached within the first 12 hours of treatment.10,11,19,20 Inversion of the T wave in the infarct-related ECG leads within the first 24 hours after MI has been described as another marker of coronary artery reperfusion, although it becomes evident later than the others. Thus Matetzky et al,12 through a coronary angiogram performed between 36 and 48 hours after MI, found that T-wave inversion was associated with Thrombolysis In Myocardial Infarction grade 3 coronary blood flow in a series of 100 patients treated with thrombolysis. These authors explored the possibility that T-wave inversion at 3 and 12 hours after thrombolytic therapy could help predict coronary artery patency, but they failed to establish a correlation. Castro et al13 compared T-wave inversion with classic reperfusion indexes in patients treated with thrombolysis and found that Twave inversion at 24 hours was the most significant independent predictor of coronary artery patency.13
Ventricular arrhythmias (%) 17.9 11.8 <.002 17.5 12.1 <.03 14.6 14.2 NS 14.9 12.9 NS
Mechanical complications (%) 4.8 1.6 <.01 5.0 1.2 <.001 5.3 1.3 <.001 4.6 0.5 <.001
Based on this evidence, we found of interest to register the 4 coronary reperfusion criteria in the medical records of patients admitted to our hospital network (GEMI group). Our hypothesis was that patients treated with thrombolysis who had positive indexes of coronary artery reperfusion should have a more favorable prognosis after thrombolytic therapy. The analysis of 967 consecutive patients seen with Q-wave MI treated with streptokinase has validated this hypothesis. Although the overall in-hospital mortality rate of patients treated with streptokinase was 12.0%, it reached 32.1% when none of the coronary artery reperfusion criteria were present and fell to 4.2% when 3 or more indexes were present. Univariate analysis of in-hospital mortality rate in relation to the presence of each one of the coronary reperfusion indexes showed that resolution of chest pain and ST-segment elevation, as well as early rise of CK enzyme levels, were associated with mortality rates ranging from 6.1% to 7.6%. On the other hand, the inversion of the T wave in the infarct-related ECG leads at 24 hours was associated with an even lower mortality rate of only 3.9%. Interestingly the incidence of mechanical complications was significantly lower in the presence of each one of the reperfusion markers. In this regard the incidence of mechanical complications reached only 0.5% when there was early T-wave inversion. These findings suggest that both the incidence of death and mechanical complications derive benefits through similar mechanism from coronary artery reperfusion. Our results confirm recent findings of Schroeder et al.18 In the INJECT trial, in which the thrombolytic agents reteplase and streptokinase were compared, these authors analyzed the prognostic value of elevated ST segment at different levels of resolution. They found that complete resolution of ST-segment elevation was associated with a mortality rate of 2.5%, and when
American Heart Journal September 1999
538 Corbalán et al
there was only partial resolution of ST-segment elevation mortality rate was 4.3%. In turn, a persistently elevated ST segment was associated with a mortality rate of 17.5%. Differences in mortality rates were similar both in patients with anterior and inferior MI and in the presence of new bundle branch block. In addition, Krucoff et al21 have shown that continuous 12-lead ST-segment monitoring recovery analysis is more predictive of clinical outcomes than early coronary angiography after thrombolytic therapy in a meta-analysis of 4 different multicenter trials of patients with acute MI. On the other hand, Matetzki et al12 analyzed the value of early T-wave inversion as a marker of coronary reperfusion and found that its presence at 24 hours was significantly associated with a lower incidence of both complications and in-hospital death. This is similar to our findings. It is well known that different factors may influence MI death, such as older age, female sex, the presence of diabetes mellitus, anterior location of MI, development of cardiac failure, the presence of complex arrhythmias or conduction disturbances, and mechanical complications. A logistic regression analysis comparing coronary reperfusion indexes and death adjusted for these factors confirmed that early T-wave inversion has a significant independent prognostic value for a lower mortality rate. Resolution of chest pain and the early rise in CK enzyme levels also had an independent prognostic value. When the 4 markers of coronary artery reperfusion were included in the regression model, early Twave inversion and abrupt CK rise continued to be independently associated with a better outcome. Each one of the coronary reperfusion markers analyzed in our series has limitations. Adequate interpretation of changes in chest pain, because of its subjective nature, might be different, especially if analgesics or opiates have been used. Shah et al14 described the fluctuations that may occur in ST-segment elevation both during and after thrombolysis,14 which was confirmed by other authors.15,16 Rapid elevation of CK enzyme levels is fairly reliable as a reperfusion parameter, but it has the drawback of the delay by which the results are obtained. The same limitation applies to inversion of the T wave, which becomes evident 24 hours after treatment. Although this marker appears to have the most powerful prognostic value, its late occurrence could limit its clinical usefulness. In addition, we have previously described the finding of false-positive results, that is, patients with an obstructed coronary artery and negative T waves at 24 hours that may occur when collateral circulation develops at the infarcted area.13 An important limitation of our study was the absence of a core laboratory for the interpretation of serial ECG changes. For this reason we preferred to use more than 50% reduction of the pretreatment ST-segment elevation instead of the >70% reduction of ST-segment elevation,
which has been proposed by Schroeder et al.18 This method could explain why resolution of ST-segment elevation was not an independent predictor of death in the presence of other reperfusion markers. Another explanation could be that the ST-segment changes might have a fluctuating course after 90 minutes of thrombolytic therapy, reflecting alternating episodes of coronary artery reperfusion and coronary artery reocclusion. It would make sense, then, that reperfusion markers that occur late in time, such as CK rise and T-wave inversion, were independent predictors of death. According to our results the predictive information content of noninvasive markers of coronary artery reperfusion appears to be inversely related to the acuity with which they become available. Thus T-wave inversion is most predictive but not available until 24 hours; the CK enzyme peak is next predictive, but not available until 12 hours; the ST segment and chest pain are next, analyzed at 90 minutes after thrombolysis. It could be proposed, then, that the combined analysis of coronary artery reperfusion markers, over time, offers more relevant clinical information than the individual assessment of each marker alone. Thus ST-segment recovery can give a prognostic insight during the first 90 minutes of initiating therapy; CK determinations can amplify that information between 6 and 12 hours, and inversion of T waves can increase that prognostic information even further between 12 and 24 hours. The prognostic information, thus obtained, could contribute to optimize clinical decisions such as performing early or late coronary angiography, complementary medical therapy, and so on. It is possible that even more optimal combination and sequencing of new noninvasive markers could allow their prognostic information to enter the patient care decision-making environment. Availability of continuous ST-segment and T-wave monitoring during the first 24 hours of reperfusion therapy or implementation of new biochemical markers of myocardial injury such as determinations of troponins, CK mass, or myoglobin may help gain more precise prognostic information even earlier during the clinical course of MI. Systematic and careful analysis of noninvasive markers and their temporal sequence will certainly provide the clinician good tools at the bedside to predict clinical outcomes when treating MI.
References 1. Simes R, Topol EJ, Holmes DR, White HD, et al for the GUSTO-I investigators. Link between the angiographic substudy and mortality outcomes in a large randomized trial of myocardial reperfusion. Circulation 1995;91:1923-8. 2. Galvani M, Ottani F, Ferrini D, Sorbello F, Rusticali F. Patency of the infarct-related artery and left ventricular function as the major determinants of survival after Q wave acute myocardial infarction. Am J Cardiol 1993;71:1-7.
American Heart Journal Volume 138, Number 3, Part 1
3. De Boer MJ, Suryapranata H, Hoorntje JCA, et al. Limitation of infarct size and preservation of left ventricular function after primary angioplasty compared with intravenous streptokinase in acute myocardial infarction. Circulation 1994;90:753-61. 4. Grines CL, Browne MF, Marco J, et al. A comparison of immediate angioplasty with thrombolityc theraphy for acute myocardial infarction. N Engl J Med 1993:328:673-9. 5. White HD, Cross DB, Elliot JM, Morris RM, Yee TW. Long-term prognostic importance of patency of the infarct-related coronary artery after thrombolytic therapy for acute myocardial infarction. Circulation 1994;89:61-7. 6. Anderson JL, Karagounis LA, Becker LC, Sorensen SG, Menlove R, for the TEAM-3 Investigators. TIMI perfusion grade 3 but not grade 2 results in improved outcome after thrombolysis for myocardial infarction: ventricolographic, enzimatic and electrocardiographic evidence from the TEAM-3 study. Circulation 1993;873:1829-39. 7. Clemmensen P, Ohman EM, Sevilla DC, Peck S, Wagner NB. Changes in standard electrocardiographic ST segment elevation predictive of successful reperfusion in acute myocardial infarction. Am J Cardiol 1990;66:1407-11. 8. Zehender M, Utzolino S, Furtwängler A, et al. Time course and interrelation of reperfusion-induced ST changes and ventricular arrhythmias in acute myocardial infarction. Am J Cardiol 1991;68:1138-42. 9. Hackworthy RA, Vogel MB, Harris PH. Relationship between changes in ST segment elevation and patency of the infarct related coronary artery in acute myocardial infarction. Am Heart J 1986;112:279-84. 10. Gore JM, Roberts R, Ball S, et al. Peak creatine kinase as a measure of effectiveness of thrombolytic therapy in acute myocardial infarction. Am J Cardiol 1987;59:1234-8. 11. Garabedian HD, Gold HJ, Yasuda T, et al. Detection of coronay artery reperfusion with creatine kinase-MB determinations during thrombolytic therapy: correlation with angiography. J Am Coll Cardiol 1988;11:729-34. 12. Matetzky S, Barabash GI, Shahar A, et al. Early T wave inversion after thrombolytic therapy predicts better coronary perfusion: clinical and angiographic study. J Am Coll Cardiol 1994;2:378-83. 13. Castro P, Corbalan R, Kunstmann S, Rubio R, Marchant E, Pumarino JI, et al. Early T wave inversion: a bedside marker of coronary artery patency after thrombolysis in patients with acute myocardial infarction. Rev Ch Cardiol 1992;11:94-102. 14. Shah PK, Cercek B, Lew A, Ganz W. Angiographic validation of bedside markers of reperfusion. J Am Coll Cardiol 1993;21:55-61. 15. Krucoff MW, Croll MA, Pope JE, Pieper KS, Kanani PM, Granger CB, et al. Continuously updated 12-lead ST-segment recovery analysis for myocardial infarct artery patency assessment and its correlation wtih multiple simultaneous early angiographic observation. Am J Cardiol 1993;71:145-51. 16. Veldkamp R, Green C, Wilkins M, Pope W, et al. Comparison of continous ST-segment recovery analysis with methodos using static electrocardiograms for invasive patency assessment during acute myocardial infarction. Am J Cardiol 1994;73:1069-74. 17. Krucoff MW, Croll MA, Pope J, Granger CB, et al for the TAMI-7 study group. Continuous 12-lead ST-segment recovery analysis in the TAMI 7 Study. Performance of a noninvasive method for real-time detection of failed myocardial reperfusion. Circulation 1993;88:437-46. 18. Schroder R, Wegsscheider K, Schroder K, Dissmann R, MeyerSabellek W, for the INJECT trial group. Extent of early segment elevation resolution: a strong predictor of outcome in patients with acute myocardial infarction and a sensitive measure to compare thrombolytic regimens. J Am Coll Cardiol 1995;27:1657-64.
Corbalán et al 539
19. Lewis B, Ganz W, Laramee P, et al. Usefulness of a rapid initial increase in plasma creatine kinase activity as a marker of reperfusion during thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1988;62:20-4. 20. Ong L, Coromilas J, Zimmerman JM, et al. A physiologically based model of creatine kinase-MB release in reperfusion of acute myocardial infarction. Am J Cardiol 1989;64:11-5. 21. Krucoff MW, Green CL, Trollinger KM, and Pope JE. Angiography “gold” standard or “tin” standard of reperfusion? A comparison with continuous 12 lead ST-segment recovery for prediction of clinical outcomes after thrombolytic therapy for acute myocardial infarction [abstract]. Circulation 1996;94(suppl):2572.
Appendix Gemi Group Participants (Grupo De Estudio Multicentrico Del Infarto, Sociedad Chilena De Cardiologia) Hospital Regional Arica: Ivan Criollo, Mario Gatica, Hernan Figueroa. Hospital Regional Iquique: Patricio Sierralta, Ignacio Auger, Virginia Araya. Hospital Regional Antofagasta: Juan Antonio Cotoras. Clinica Antofagasta: Guillermo Illanes. Hospital Regional Copiapo: Guillermo Ananias, Walter Evans. Hospital Regional De Coquimbo: Carlos Saldias, Hans Gonzalez. Hospital Regional De La Serena: Carlos Echeverria, Claudio Bugueño. Hospital Naval Viña Del Mar: Fernando Cardenas, Marcos Opazo. Hospital Gustavo Fricke: Mauricio Aninat, Carlos Raffo, Jorge Bartolucci. Clinica Reñaca: Rienzi Diaz, Monica Carvajal. Clinica Miraflores: Pedro Chadid. Regional Rancagua: Leopoldo Manriquez. Hospital Fusatde Rancagua: Enrique Buzeta, Rafael Castillo. Hospital Regional Talca: Enrique Mercadal, Patricio Vildosola. Hospital Herminda Martin De Chillan: Gonzalo Marin, Gustavo Yanine. Hospital Guillermo Grant Concepcion: Carlos Otero. Hospital Higueras Talcahuano: Hernan Diaz, Nelson Alburquerque. Hospital Naval De Talcahuano: Alejandro Dapelo, Carlos Nicolich. Hospital Regional Temuco: Fernando Lanas, Mario Santibañez, Luis Sepulveda. Hospital Base Valdivia: Eduardo Garces, Herminia Riquelme. Hospital Regional Osorno: Sergio Potthoff. Hospital Regional Puerto Montt: Luis Felipe Del Campò, Jame Venegas. Hospital Regional Punta Arenas: Guillermo Araneda, Bogdam Liberon. Hospital Asistencia Publica: Eduardo Chavez, Mario Ratkhmap, Marcela Argandoña. Hospital Salvador: Alvaro Puelma, Francisco Manzur. Hospital Jose Joaquin Aguirre: Jorge Yovanovich, Ruben Aguayo. Hospital Universidad Catolica: Ivan Godoy, Carlos Almendares. Hospital Barros Luco Trudeau: Ligia Gallardo. Hospital Sotero Del Rio: Claudio Fernandez, Ximena Muñoz. Hospital San Borja Arriaran: Rolando Ricciulli. Hospital Del Profesor: Mario Hassi. Clinica Alemana: Rene Etchegaray, Alejandro Abufele. Clinica Santa Maria: Sonia Kunstmann, Hernan Chamorro. Clinica Las Lilas: Alberto Gonzalez. Hospital Dipreca: Barbara Clericus, Milton Alcaino.