Results of the angiographic substudy of the Hirudin for Improvement of Thrombolysis (HIT)-4 trial

Transcription

1 European Heart Journal (2001) 22, doi: /euhj , available online at on Non-invasive detection of early infarct vessel patency by resolution of ST-segment elevation in patients with thrombolysis for acute myocardial infarction Results of the angiographic substudy of the Hirudin for Improvement of Thrombolysis (HIT)-4 trial U. Zeymer 1, R. Schröder 2, U. Tebbe 3, G. P. Molhoek 4, K. Wegscheider 2 and K.-L. Neuhaus 1 1 Medizinische Klinik II, Klinikum Kassel, Kassel, Germany; 2 Universitätsklinikum Benjamin Franklin, Berlin, Germany; 3 Medizinische Klinik II, Klinikum Lippe-Detmold, Detmold, Germany; 4 Cardiology Research, Medisch Spectrum Enschede, Twente, The Netherlands Aims The purpose of this study was to validate ST segment resolution as a non-invasive marker for patency of the infarct-related artery 90 min after the start of streptokinase therapy in patients with acute myocardial infarction. Methods and Results In the HIT-4 angiographic substudy, 447 patients with acute myocardial infarction 6h received 1 5 million IU streptokinase. Angiograms of the infarct vessel were obtained after 90 min and 12-lead ECGs at baseline and after 90 min. The best cut-off points for a correct prediction of 90 min infarct vessel patency (TIMI 2/3 flow) and complete patency (TIMI 3) were 30% ST resolution and 40%, respectively (specificity 68% and 69%, sensitivity 76% and 75%). Prediction of infarct vessel patency by in steps of 10% displayed a gradual increase in patency rates. Patients with 70% (n=70) had a 92% probability of TIMI 2/3 flow, while <30% (n=172) was associated with the absence of TIMI 3 flow in 84% of patients. Conclusions Despite fairly good sensitivities and specificities the prediction of infarct vessel patency by in the individual patient is limited. However, patients with 70% are likely to have a patent infarct artery and <30% predicts epicardial vessel occlusion or, since persistent ST elevation reflects the existing ischaemic myocardial injury, absence of myocardial perfusion. (Eur Heart J 2001; 22: , doi: /euhj ) 2001 The European Society of Cardiology Key Words: Acute myocardial infarction, thrombolysis, angiography,. See page 722 for the Editorial comment on this article Introduction In patients with acute myocardial infarction, coronary artery patency 90 min after initiation of thrombolytic therapy is believed to be the most powerful independent predictor of survival [1,2]. However, the only direct diagnostic method of assessing coronary reperfusion Revision submitted 22 May 2000, and accepted 24 May The HIT-4 Study was supported by the Behringwerke AG (AVENTIS), Marburg, Germany. Correspondence: Dr Uwe Zeymer, Klinikum Kassel, Medizinische Klinik II, Mönchebergstrasse 41-43, D Kassel, Federal Republic of Germany X/01/ $35.00/0 emergency angiography is cumbersome, expensive, and unavailable in most hospitals. To overcome these drawbacks, several methods of electrocardiographic STsegment analysis have been developed which are aimed to be practical, non-invasive approaches in the assessment of coronary reperfusion [3 8]. However, most studies have only been performed in a limited number of patients. This is the case with non-invasive determination of reperfusion status by ST-segment elevation resolution (), which needs to be validated by a study of sufficient size. Early infarct-related artery patency status, as the gold standard prognosticator, has been challenged by the notion that the extent of STsegment elevation resolution 90 and 180 min after the 2001 The European Society of Cardiology

2 770 U. Zeymer et al. start of thrombolytic therapy is also a strong predictor of outcome in patients with acute myocardial infarction [9 11]. We therefore conducted an angiographic substudy in conjunction with the Hirudin for Improvement of Thrombolysis (HIT)-4 trial [11], in which in addition to a 90 min angiogram, ECGs were recorded at baseline and 90 min after the start of thrombolysis. The purpose was to prospectively assess the validity of, as a non-invasive determinant of reperfusion, in a well defined large cohort of patients. Methods Study patients The HIT-4 trial was a multicentre, randomized, doubleblind trial that compared r-hirudin (HBW 023) and heparin as adjuncts to thrombolysis with streptokinase (1 5 million units over 60 min) in 1208 patients with acute myocardial infarction of less than 6 h duration. The trial was approved by the Ethics Committee of the University of Göttingen. Details of the rationale, design and the main results will be available in the original report [12]. Eligible patients had ST-segment elevation of 0 1 mv in at least two limb leads and/or 0 2 mvin two contiguous precordial leads. All 447 patients who enrolled in selected hospitals for the angiographic part of the HIT-4 trial were considered for this analysis. Angiographic analysis Coronary angiograms of the right and left coronary arteries were obtained 90 min after the initiation of streptokinase therapy, starting with the presumed infarct-related artery. Angiograms were evaluated in a core laboratory by two independent investigators in a blinded fashion. The patency status of the infarct vessel was determined at the first contrast injection and graded according to the classification of the Thrombolysis in Myocardial Infarction (TIMI) flow grading [13]. of the infarct vessel was defined as TIMI grade 2 or 3 flow. Lesion location was defined as proximal or nonproximal in relation to the first septal perforator branch of the left anterior descending artery, the first obtuse marginal branch of the left circumflex artery or the first right ventricular branch at the right coronary artery, respectively. Electrocardiographic analysis A 12 lead ECG was recorded at baseline and immediately before the 90 min angiogram, which was obtained 90 min after the start of streptokinase therapy. The sum of the ST-segment elevation was measured 20 ms after the end of the QRS complex from leads I, avl, and V 1 V 6 for anterior acute myocardial infarction, and leads II, III, avf, V 5 and V 6 for inferior acute myocardial infarction. The ST segment deviation was evaluated centrally, independent of and blinded to the other study data or results or treatment assignment. Data analysis and statistics The sum ST-segment elevation resolution, immediately before the 90-min angiogram, was used for non-invasive prediction of the coronary artery patency status. Sensitivity and specificity to predict vessel patency correctly was calculated at various (X) cut-off points in steps of 10%, from 90% to no. Sensitivity represents the proportion of patent artery patients classified as patent by >X%. Specificity represents the proportion of patients with occluded vessels classified as occluded by <X% ST resolution. The patency by is the proportion of patients with a patent artery classified as such by >X%. The occlusion is the proportion of patients with an occluded artery classified as such by <X%. To assess the optimal cut-off point that best predicts correctly the coronary artery patency status, the squared standardized log odds ratio statistic was applied [14,15]. As a function of the hypothetical cut-off points in sum of ST segment resolution it measures the differences in angiographic patency rates above and below the cut-off points in each sample of between 100% to less than 0%. The optimal cut-off point corresponds to the maximum squared log odds ratio. For univariate analyses, only non-parametric methods were applied. Comparisons between two groups were performed using the Mann Whitney U test for continuous variables and Fisher s exact tests or chi square tests for dichotomous variables. Simultaneous comparisons for more than two groups were performed using the Kruskal Wallis test. Results There were no significant differences, with regard to medical history or conditions at randomization (Table 1), between the patients enrolled in the clinical or the angiographic parts of the substudy of the HIT-4 trial. The time between symptom onset and treatment was slightly longer in the angiographic patients because the time to randomization was longer. Of the 447 patients enrolled in the angiographic substudy, for various reasons 19 did not have a 90 min angiogram, and in 10 patients the infarct-related coronary artery could not be identified. Of the evaluable 418 patients, the infarct-related artery was the right coronary artery in 177 patients (42%), the left main coronary artery in one patient, the left anterior descending coronary artery in 160 (38%) patients, the circumflex

3 Results of the HIT trial 771 Table 1 Clinical characteristics of patients enrolled in the angiographic substudy (n=447) or the clinical part (n=761) of the HIT-4 Study Angio-substudy n=447 Other patients n=761 P-value Age (years)* ns Men 76 7% (343) 75 4% (574) ns Anterior MI 43 0% (192) 38 5% (293) ns Medical history Hypertension 35 6% (159) 37 8% (287) ns Hyperlipidaemia 25 5% (114) 24 4% (185) ns Diabetes 13 2% (59) 13 1% (100) ns Current smoker 42 1% (188) 45 5% (345) ns Previous MI 12 3% (55) 13 9% (106) ns Conditions at randomization Heart rate* ns Systolic blood pressure* ns Killip class >1 12 8% (57) 11 4% (87) ns Time to treatment* Number in parentheses is number of patients. *Mean value SD; minutes. MI=myocardial infarction. coronary artery in 60 patients (14%), and other arteries in 20 patients (5%). A TIMI grade 0/1, 2, and 3 flow at 90 min was observed in 39%, 24%, and 37% of patients, respectively. groups could be assessed at 90 min in 355 patients. The factors not allowing for assessment of ST resolution were: insufficient ST elevation at baseline (37 patients), no 90-min ECG (15 patients), idioventricular rhythm (11 patients) and a pacemaker ECG (one patient). The sum of ST elevation at baseline for anterior acute myocardial infarction patients was 1 67 mv, for the subgroup of patients with a proximal coronary artery lesion 1 64 mv, and with a non-proximal lesion 1 71 mv. ST elevation in inferior acute myocardial infarction patients was 0 93 mv and in proximal or non-proximal lesions 0 89 mv or 0 95 mv, respectively. Prediction of infarct vessel patency by related to vessel patency The spectra of sensitivities and specificities with their 95% confidence intervals for correct prediction of infarct vessel patency (TIMI grade 2/3 flow) 90 min after the start of streptokinase infusion, by resolution of sum ST segment elevation, are shown in Fig. 1(a). The sensitivity and specificity curves intersect at a level slightly below 30%. Diagnostic test data, provided at the level of 30%, are displayed in Table 2. Since TIMI 3 flow is considered the primary goal of reperfusion therapy the spectra for specificities and sensitivities for TIMI grade 3 flow are shown in Fig. 1(b). Compared to Fig. 1(a), the sensitivity curve is shifted somewhat upwards and the specificity curve downwards with an intersect slightly below 40% ST resolution. Diagnostic test data for complete perfusion (TIMI grade 3 flow) at the level of 40% are shown in Table 3. Evaluation of the optimal cut-off point for TIMI 2 or 3 flow by the squared standardized log odds ratio statistic revealed two maxima, one at 30%, and the second around 20%. Separate evaluation identified the 30% cut-off point, as related to inferior acute myocardial infarction, and the 20% cut-off point to anterior acute myocardial infarction. In Table 4, the angiographic patency rates and statistical data for non-invasive prediction of coronary artery patency are presented. For proximal lesions, patency rates were lower than for non-proximal lesions. However, sensitivity and specificity tended to be better for proximal lesions. Vessel patency related to The patency s for TIMI grade 2 and 3 flow are depicted in 10% steps of in Fig. 2. There was a gradual increase in patency rates from <10% to >80%. For TIMI 3 patency the major step was from 36% to 53% with a >40, which has been identified as the best cut-off point for sensitivity and specificity. The proportions of patients with various TIMI flow grades related to no (<30%), partial (30 70) and complete (>70%) [9 11] are shown in Table 5. Data for patients with anterior or inferior infarct location were very similar. Patients with complete had a 92% probability of TIMI 2/3 patency, but of all patients who had TIMI 2/3 flow only 30% had complete. With no,

4 772 U. Zeymer et al. % % (a) % resolution of the sum of ST elevations at baseline (b) % resolution of the sum of ST elevations at baseline Figure 1 (a) Cut-off point of dependent sensitivity ( ) and specificity ( ) curves of at 90 min for correct prediction of a patent infarct vessel (TIMI 2/3 flow) at the 90 min angiogram in 355 patients. The best cut-off point for detection of TIMI 2/3 patency is about 30% (220 patients had a TIMI 2/3 flow, 135 patients had TIMI 0/1 flow). (b) Cut-off point of dependent sensitivity ( ) and specificity ( ) curves of ST resolution at 90 min for correct prediction of TIMI grade 3 flow at the 90 min angiogram in 355 patients. The best cut-off point for detection of TIMI 3 patency is about 40% (132 patients had TIMI 3 flow; 223 patients had TIMI 0/1/2 flow). 84% of patients had no TIMI 3 flow, but 43% of patients with >30% also showed absence of TIMI 3 flow. Discussion The present report is part of the HIT-4 Study [12] and its ST Segment Resolution substudy [11] and represents the largest series comparing analysis and 90 min angiography after thrombolytic therapy for acute myocardial infarction. As a simple method, easy to obtain and allowing immediate interpretation in all patients with acute myocardial infarction presenting with ST elevation, was assessed 90 min after the start of the streptokinase infusion. To be consistent with all previous studies, infarct vessel patency was defined as TIMI flow grade 2 or 3. However, since TIMI grade 3 flow is considered to be the primary goal of thrombolytic therapy we looked separately for this TIMI flow grade. Prediction of infarct vessel patency by Most previous studies on the value of ECG monitoring for non-invasive prediction of patency suffered from small sample sizes, particularly from low numbers of patients with occluded coronary arteries [3,5 8,16]. Confidence limits on the predictive performance, when reported, were fairly wide [8,17]. Excellent accuracies have been reported [3,5,6,8,16] but when the same criteria were applied prospectively in somewhat larger patient series, the results are less impressive [4,17,18]. In studies aimed to predict vessel patency by ECG criteria, results are usually presented as in our Table 4 and interpreted as a clinically useful non-invasive predictor of the patency status of the infarct vessel. Although these data may appear sufficient for larger patient groups, for the individual patient, prediction of epicardial vessel patency is limited. The relatively large number of patients in our series, including 135 patients who had an occluded infarct-related coronary artery, allowed sensitivity and specificity curves to be constructed for correct prediction of vessel patency with relatively narrow 95% confidence limits over the whole range of percent (Fig. 1(a) and (b)). There is important clinical information in the spectra of sensitivities and specificities, as plotted in Fig. 1(a) and (b) and in the graded relationship between and infarct vessel patency in Fig. 2, which provide useful tools for a better understanding of the dynamic interactions between vessel patency, ST elevation recovery, and early mortality. In previous studies on the prognostic implications of we could demonstrate the strong prognostic power of a three-tiered scheme with complete, partial and no for stratifying outcome in acute myocardial infarction [9 11], which was also confirmed by others [19]. As the specificity curve in Fig. 1(a) depicts, less than 5% of patients who had an occluded infarct artery have complete ( 70%) and, according to the patency and as shown in Fig. 2, 92% of patients who have complete ( 70%) had a patent infarct-related artery and 69% had TIMI grade 3 flow (Table 5). The high specificities, however, could only be obtained at the expense of the sensitivities. A considerable number of patients who had TIMI grade 2 or 3 flow have <70%. At the other end, no (<30%) is associated with absence of TIMI 3 flow (84% of patients), rather than absence of TIMI 2/3 flow. Sensitivity of <30% ST resolution to correctly predict absence of TIMI 3 flow was 65% and specificity 79% (Fig. 1(b), Table 5). For patients with partial (30 <70%), reliable prediction of epicardial patency is not possible (Fig. 2).

5 Results of the HIT trial 773 Table 2 90 min in relation to 90 min angiographic vessel patency (TIMI grade 2 or 3 flow). Chi-square 2 2 table for calculation of sensitivity, specificity, patency, and occlusion with the 95% confidence intervals in parentheses 30% <30% TIMI 2/ Sensitivity 68% (62 74) TIMI 0/ Specificity 76% (67 82) Occlusion 82% (76 87) 59% (52 67) Table 3 90 min in relation to 90 min angiographic complete vessel patency (TIMI grade 3 flow). Chi-square 2 2 table for calculation of sensitivity, specificity, patency, and occlusion with the 95% confidence intervals in parentheses 40% <40% TIMI Sensitivity 69% (63 75) TIMI 0/1/ Specificity 75% (67 82) Occlusion 62% (51 73) 80% (71 88) Table 4 Angiographic TIMI grade 2 and 3 patency rates and sensitivity, specificity, patency and occlusion for non-invasive prediction of coronary artery patency by sum ST segment elevation resolution 90 min after start of streptokinase infusion No. of patients rate predicted by ST cut point Sens. Spec. PPV OPV P value Anterior MI % 20% 72% 72% 82% 60% Proximal 48 58% 20% 79% 80% 85% 73% Non prox % 20% 70% 67% 79% 55% Inferior MI % 30% 72% 73% 81% 63% Proximal 67 52% 30% 80% 75% 78% 77% Non prox % 30% 70% 72% 83% 54% For proximal (non proximal) coronary artery lesion location: patients with lesion in venous bypass grafts are excluded. MI=myocardial infarction; OPV=occlusion ; PPV=patency ; Sens.=sensitivity; Spec.=specificity. Because there is a gradual increase in patency rates with the increasing amount of (Fig. 2) on average half of these patients have TIMI grade 3 flow. Epicardial vessel patency and myocardial perfusion In evolving myocardial infarction, ST segment elevation quite accurately reflects the existing ischaemic myocardial injury [20]. In the few patients who had complete ( 70%) despite persistent occlusion of the infarct vessel, effective early collateral circulation may have attenuated myocardial ischaemia [16 18,21]. Persistent ST elevation, despite a patent epicardial infarct vessel, is usually considered to be false negative. This is correct as far as epicardial vessel patency is concerned. However, partial or even no despite a patent epicardial vessel probably relates to impairment of perfusion at the myocardial level [22,23].

6 774 U. Zeymer et al. % STR 92% 70% n = 50 91% 85% 82% 67% 58% 55% 53% 72% 69% 36% 55% % <80 70 <70 60 <60 50 <50 40 <40 30 <30 20 <20 <10% 10 any increase When the heart is exposed to ischaemia and reperfusion, damage occurs not only to the myocytes but also to the microvasculature [24]. Despite an angiographically successful opening of an infarct-related artery within a 6 h time frame, in most patients microvascular damage impedes myocardial blood supply, at least to some extent. The spectrum of reperfusion at the myocardial level ranges from no reflow or partial reperfusion to normal perfusion [22]. Myocardial no-reflow is associated with a similar worse prognosis: persistent epicardial vessel occlusion [22,23,25 27]. Impaired microvascular perfusion, as evidenced by persistent ST elevation after successful mechanical recanalization, was associated with more extensive infarctions and an unfavourable clinical outcome [25 27]. Therefore, <30% in patients with TIMI 3 flow in the infarct vessel most likely identifies patients with no myocardial reflow, which means that these patients have a higher risk for an adverse outcome, and may therefore not be defined as false negative for reperfusion, but only false negative for patency of the epicardial coronary artery. Clinical implications 43% 37% 24% 21% 13% Figure 2 Incidence of TIMI grade 2 ( )and3( ) flow by in steps of 10% (STR). There was a steady increase in patency rates with the increase in. Table 5 Proportion of patients with various TIMI flow grades 90 min after start of streptokinase infusion related to the three groups of for stratifying outcome in acute myocardial infarction TIMI grade No (<30%) n=172 Partial (30 70%) n=112 Complete ( 70%) n=71 0/1 59% (102) 24% (27) 8% (6) 0/1/2 84% (144) 52% (57) 31% (22) 2 24% (42) 27% (30) 23% (16) 3 16% (28) 49% (55) 69% (49) 2/3 41% (70) 76% (85) 92% (65) Presented are percentages with number of patients in parentheses. For the individual patient prediction of the patency status of the epicardial vessel is limited. However, in evolving acute myocardial infarction the extent of STsegment elevation reflects the actual existing ischaemic myocardial injury, regardless of whether blood flow is impaired at the epicardial or myocardial level. The extent of after epicardial vessel recanalization is probably related to the intensity and duration of impairment of myocardial blood flow. Complete ST resolution is highly predictive of an unimpaired flow both at the epicardial and myocardial level. The present study was too small to draw any relevant conclusions with respect to mortality, therefore we did not present these data. As demonstrated with a larger sample size in our corresponding clinical HIT-4 substudy [11], complete at 90 min is associated with a 30-day cardiac mortality rate of only 0 9%. Thus, outcome is so good that any prophylactic intervention is unlikely to improve it. On the other hand, no ST resolution at 90 min indicates persistent epicardial vessel occlusion or myocardial no-reflow, and is associated with a higher mortality risk, particularly in patients with anterior acute myocardial infarction [11]. These patients may benefit from emergency angiography and rescue PTCA [28]. However, randomized studies are necessary to prove this concept, since in our study recovery of ST-segment elevation 180 min after start of therapy was better in patients treated conservatively than in those treated with rescue PTCA at 90 min, and there was a trend towards a higher mortality in patients treated interventionally [29]. In patients who have persistent ST elevation despite epicardial patency, strategies that attenuate the process of microvascular reperfusion injury may be helpful to improve outcome [30]. In patients with partial, infarct vessel patency rates vary and prediction is uncertain. Since early mortality is relatively low [11], patients with partial might be handled conservatively, if they are free of symptoms and haemodynamically stable. Watchful waiting and elective angiography appears appropriate. References [1] Vogt A, von Essen R, Tebbe U, Feuerer W, Appel K-F, Neuhaus K-L. Impact of early perfusion status of the infarctrelated artery on short-term mortality after thrombolysis for acute myocardial infarction: Retrospective analysis of four German multicenter studies. J Am Coll Cardiol 1993; 21: [2] The GUSTO Angiographic Investigators. The effects of tissue plasminogen activator, streptokinase, or both on coronaryartery patency, ventricular function, and survival after acute myocardial infarction. N Engl J Med 1993; 329: [3] Clemmensen P, Ohman EM, Sevilla DC et al. Changes in standard electrocardiographic ST-segment elevation predictive of successful reperfusion in acute myocardial infarction. Am J Cardiol 1990; 66: [4] Zabel M, Hohnloser SH, Parussel A, Just H. ST segment analysis for assessment of coronary artery patency: comparison of surface ECG and Holter recordings. Eur Heart J 1992; 13: [5] Krucoff MW, Croll MA, Pope JE et al. Comparison of continuous ST-segment recovery analysis for myocardial infarct artery patency assessment and its correlation with

7 Results of the HIT trial 775 multiple simultaneous early angiographic observations. Am J Cardiol 1993; 71: [6] Dellborg M, Riha M, Swedberg K. Dynamic QRS-complex and ST-segment monitoring in acute myocardial infarction during recombinant tissue type plasminogen activator therapy. Am J Cardiol 1991; 67: [7] Doevendans PA, Gorgels AP, van der Zee R, Partouns J, Bär FW, Wellens HJJ. Electrocardiographic diagnosis of reperfusion during thrombolytic therapy in acute myocardial infarction. Am J Cardiol 1995; 75: [8] Fernandez AR, Sequeira RF, Chakko S et al. ST segment tracking for rapid determination of patency of the infarctrelated artery in acute myocardial infarction. J Am Coll Cardiol 1995; 26: [9] Schröder R, Dissmann R, Brüggemann T et al. Extent of early ST segment elevation resolution: A simple but strong predictor of outcome in patients with acute myocardial infarction. J Am Coll Cardiol 1994; 24: [10] Wegscheider K, Neuhaus KL, Dissmann R, Tebbe U, Zeymer U, Schröder R. Impact of prediction of outcome by early ST segment changes in acute myocardial infarction. Herz 1999; 24: [11] Schröder R, Zeymer U, Wegscheider K, Neuhaus KL. Comparison of the of ST segment elevation resolution at 90 and 180 minutes after start of streptokinase in acute myocardial infarction. Results of the Hirudin for the Improvement of Thrombolysis (HIT)-4 Study. Eur Heart J 1999; 20: [12] Neuhaus KL, Molhoek GP, Zeymer U et al. for the HIT-4 Investigators. Recombinant hirudin (lepirudin) for the improvement of thrombolysis with streptokinase in patients with acute myocardial infarction. Results of the HIT-4 Study. J Am Coll Cardiol 1999; 34: [13] TIMI Study Group. The thrombolysis in Myocardial Infarction (TIMI) trial: phase I findings. N Engl J Med 1985; 312: [14] Fleiss J. Statistical methods for rates and proportions, 2nd edn. New York: Wiley, 1981: [15] Lausen B, Schumacher M. Maximally selected rank statistics. Biometrics 1992; 48: [16] Saran RK, Furniss SS, Hawkins T, Reid DS. Reduction in ST segment elevation after thrombolysis predicts either coronary reperfusion or preservation of left ventricular function. Br Heart J 1990; 64: [17] Krucoff MW, Croll MA, Pope JE et al. forthetami7study Group. Continuous 12-lead ST-segment recovery analysis in the TAMI 7 Study. Circulation 1993; 88: [18] Dellborg M, Steg PG, Simoons M et al. Vectorcardiographic monitoring to assess early vessel patency after reperfusion therapy for acute myocardial infarction. Eur Heart J 1995; 16: [19] Anderson RD, White HD, Ohman EM et al. Resolution of ST-segment elevation 90 minutes after thrombolysis for acute myocardial infarction predicts outcome: A GUSTO-III Substudy (Abstr). J Am Coll Cardiol 1998; 31 (Suppl A): 371A. [20] Muller JE, Maroko PR, Braunwald E. Precordial electrocardiographic mapping: a technique to assess the efficacy of interventions designed to limit infarct size. Circulation 1978; 57: [21] Charney R, Cohen M. The role of coronary collateral circulation in limiting myocardial ischemia and infarct size. Am Heart J 1993; 126: [22] Kenner MD, Zajac EJ, Kondos GT et al. Ability of the no-reflow phenomenon during an acute myocardial infarction to predict left ventricular dysfunction at one-month follow-up. Am J Cardiol 1995; 76: [23] Ito H, Maruayama A, Iwakura K et al. Clinical implications of the no-reflow phenomenon. Circulation 1996; 93: [24] Kloner RA. Does reperfusion injury exist in humans? J Am Coll Cardiol 1993; 21: [25] van t Hof AW, Liem A, de Boer MJ, Zijlstra F, for the Zwolle Myocardial Infarction Study Group. Clinical value of 12-lead electrocardiogram after successful reperfusion therapy for acute myocardial infarction. Lancet 1997; 350: [26] Somitsu Y, Nakamura M, Degawa T, Yamaguchi T. Prognostic value of slow resolution of ST-segment elevation following successful direct percutaneous transluminal coronary angioplasty for recovery of left ventricular function. Am J Cardiol 1997; 80: [27] Claeys MJ, Bosmans J, Veenstra L et al. Determinants and prognostic implications of persistent ST-segment elevation after primary angioplasty for acute myocardial infarction. Circulation 1999; 99: [28] Ellis SG, Ribeiro da Silva E, Heyndrick G et al. for the Rescue Investigators. Randomized comparison of rescue angioplasty with conservative management of patients with early failure of thrombolysis for acute myocardial infarction. Circulation 1994; 90: [29] Zeymer U, Schröder R, Tebbe U et al. Führt die Rescue- PTCA 90 Minuten nach Beginn der Streptokinaselyse im Vergleich zur konservativen Therapie zu einer Verbesserung der myokardialen Durchblutung oder Senkung der Sterblichkeit (Abstr). Z Kardiol 1998; 87 (Suppl 5): 57. [30] De Lemos J, Antman EM, Gibson M et al. Abciximab improves both epicardial flow and myocardial reperfusion in ST elevation myocardial infarction. Observation from the TIMI 14 Trial. Circulation 2000; 101: