- Email: [email protected]
Relation between preintervention angiographic evidence of coronary collateral circulation and clinical and angiographic outcomes after primary angioplasty or stenting for acute myocardial infarction
Relation Between Preintervention Angiographic Evidence of Coronary Collateral Circulation and Clinical and Angiographic Outcomes After Primary Angioplasty or Stenting for Acute Myocardial Infarction David Antoniucci, MD, Renato Valenti, MD, Guia Moschi, MD, Angela Migliorini, MD, Maurizio Trapani, MD, Giovanni Maria Santoro, MD, Leonardo Bolognese, MD, Giampaolo Cerisano, MD, Piergiovanni Buonamici, MD, and Emilio Vincenzo Dovellini, MD It is unknown if collateral circulation (CC) has a beneficial effect on outcomes of patients who undergo mechanical intervention in the first hours after onset of acute myocardial infarction (AMI). This study analyzes the relation between CC and outcome in patients with AMI who underwent primary angioplasty or stenting within 6 hours of symptom onset. The analysis was performed in a series of 1,164 consecutive patients. The contribution of clinical, angiographic, and procedural variables to the angiographic and clinical outcomes was evaluated by multivariate logistic regression analysis and the Cox proportional hazard model, respectively. Of 1,164 patients, 264 (23%) had angiographic evidence of CC. Patients with CC had a lower incidence of diabetes (11% vs 16%, p ⴝ 0.033), anterior AMI (41% vs 55%, p <0.001), cardiogenic shock (9% vs 14%, p ⴝ
0.029), anterograde TIMI grade flow >1 (10% vs 21%, p <0.001), and a greater incidence of preinfarction angina (43% vs 32%, p ⴝ 0.001), multivessel disease (59% vs 47%, p ⴝ 0.001), and total chronic occlusion (20% vs 10%, p <0.001). At 6 months, the mortality rate was lower in patients with CC compared with patients without CC (4% vs 9%, p ⴝ 0.011), whereas there were no differences in the incidence of reinfarction, target vessel revascularization, and angiographic restenosis. After multivariate analysis, CC did not emerge as a significant variable in relation to 6-month clinical and angiographic outcomes. CC does not exert a protective effect in patients who undergo mechanical intervention in the first 6 hours of AMI onset. 䊚2002 by Excerpta Medica, Inc. (Am J Cardiol 2002;89:121–125)
oronary collateral circulation (CC) may play an important role in maintaining viable myocardium C after abrupt coronary occlusion, and it is a primary
our institutions for all patients with AMI admitted within 6 hours of symptom onset or within 24 hours if there was evidence of continuing ischemia, without any restrictions based on age, sex, or clinical status on presentation. Criteria for enrollment for this study included: chest pain persisting for ⬎30 minutes associated with ST-T segment elevation of ⱖ0.1 mV in ⱖ2 contiguous electrocardiographic leads. Coronary angiography was required in all patients at 6 months after the procedure. Unscheduled angiography was allowed on the basis of clinical indication. Left ventriculography was performed in all patients if not contraindicated by cardiogenic shock or severe hemodynamic instability. Creatine kinase measurements were systematically performed on admission and every 3 hours for the first 24 hours, and then every 12 hours for 2 days. The peak value of creatine kinase and the time-to-peak of creatine kinase were estimated for each patient. Angiographic analysis: Coronary flow in the infarct artery was graded according to Thrombolysis In Myocardial Infarction (TIMI) study group classification.3 Collateral flow before direct angioplasty was graded using the classification developed by Rentrop et al4:
determinant of left ventricular function recovery after late mechanical infarct artery reperfusion.1,2 It is unknown whether early CC recruitment has a similar beneficial effect on clinical and angiographic outcomes in patients who undergo mechanical intervention in the first hours of onset of acute myocardial infarction (AMI). This study assesses the significance of preintervention angiographic evidence of CC in patients with AMI who undergo mechanical intervention within 6 hours of symptom onset.
METHODS
Patients and treatment protocol: Since January 1995, primary percutaneous transluminal coronary angioplasty (PTCA) has been the systematic treatment at From the Division of Cardiology, Careggi Hospital, Florence, Italy. Manuscript received July 2, 2001; revised manuscript received and accepted September 24, 2001. Address for reprints: David Antoniucci, MD, Division of Cardiology, Careggi Hospital, Viale Morgagni, I-50134, Florence, Italy. E-mail: [email protected]. ©2002 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 89 January 15, 2002
0002-9149/02/$–see front matter PII S0002-9149(01)02186-5
121
sion of the target lesion in association with objective evidence CC Group Non-CC Group of ischemia or viable myocardium. Variable (n ⫽ 264) (n ⫽ 900) p Value Statistical analysis: Categorical data are presented as absolute values Age (yrs) 63 ⫾ 12 64 ⫾ 12 0.146 Men 216 (82%) 690 (77%) 0.076 and percentages, whereas continuous Systemic hypertension 96 (36%) 290 (32%) 0.209 data are summarized as mean ⫾ SD. Diabetes mellitus 28 (11%) 143 (16%) 0.033 A 2-tailed Student’s t test was used Hypercholesterolemia (⬎200 mg/dl) 70 (26%) 229 (25%) 0.726 to test differences among continuous Previous myocardial infarction 40 (15%) 105 (12%) 0.132 Preinfarction angina pectoris 114 (43%) 292 (32%) 0.001 variables, whereas categorical variCardiogenic shock 23 (9%) 124 (14%) 0.029 ables were compared by chi-square Onset of symptoms to treatment (h) 3.9 ⫾ 2.3 3.6 ⫾ 1.9 0.067 analysis or Fisher’s exact test. The Anterior wall AMI 108 (41%) 491 (55%) ⬍0.001 contribution of clinical, angioLeft ventricular ejection fraction (%)* 48.1 ⫾ 11.9 48.3 ⫾ 12.2 0.907 graphic, and procedural variables to Infarct-related coronary artery† ⬍0.001 Left anterior descending 102 (39%) 486 (54%) the angiographic and clinical outRight 136 (51%) 276 (31%) comes was evaluated with multivarLeft circumflex 23 (9%) 126 (14%) iate logistic regression analysis and Left main 3 (1%) 12 (1%) the Cox proportional hazard model, Multivessel coronary disease 155 (59%) 426 (47%) 0.001 Chronic occlusion 52 (20%) 94 (10%) ⬍0.001 respectively. The odds ratio (OR) TIMI grade flow ⬎1 26 (10%) 188 (21%) ⬍0.001 and their 95% confidence intervals Reference vessel diameter (mm) 3.18 ⫾ 0.47 3.06 ⫾ 0.49 ⬍0.001 (CI) were calculated. The variables *CC group (n ⫽ 226) and non-CC group (n ⫽ 706). used for the analyses included age, † Included 2 coronary grafts in the CC group, and 5 coronary grafts in the non-CC group. gender, diabetes, previous myocardial infarction, preinfarction angina, anterior AMI, cardiogenic shock, grade 0 ⫽ no visible filling of any collateral channels; multivessel disease, total chronic occlusion, coronary 1 ⫽ filling of side branches of the infarct artery with CC, baseline infarct artery TIMI grade flow ⬎1, time no dye reaching the epicardial segment; 2 ⫽ partial to reperfusion, reference infarct artery diameter, infilling of the epicardial vessel; and 3 ⫽ complete farct artery stenting, multiple stents, failure of PTCA, filling of the epicardial vessel by collateral vessels. and postprocedure minimum lumen diameter. SurTwo observers assessed the coronary angiograms in a vival curves were generated using the Kaplan-Meier blinded fashion and reached a consensus regarding the method. Comparison between survival curves was TIMI flow grade and the collateral flow grade. Quan- performed using the log-rank test. A p value ⬍0.05 titative coronary angiography was performed with the was considered significant. Statistical tests were peruse of an automatic edge detection system (Siemens formed using package SPSS 7.0 (SPSS Inc., Chicago, Ancor, Solna, Sweden). Coronary occlusion was as- Illinois). signed a value of 0 mm for minimal luminal diameter and 100% for percent diameter stenosis. An optimal RESULTS angiographic result was defined as residual stenosis Patients: From January 1995 to December 2000, ⬍30% associated with TIMI grade 3 flow. An unsuc- 1,164 patients underwent primary mechanical intercessful procedure was defined as a procedure resulting vention for AMI within 6 hours of symptom onset. Of in TIMI grade 0 or 1 flow whatever the residual the 1,164 patients, 264 (23%) had angiographic evistenosis was. Angiographic restenosis was defined as dence of CC flow grade ⬎1 (CC group), whereas in ⬎50% stenosis of the target lesion on the scheduled or 900 patients, coronary angiography did not show corunscheduled follow-up angiogram. Left ventricular onary CC (non-CC group). Baseline clinical and anfunction was evaluated by contrast ventriculography giographic characteristics of the patient groups are in the right anterior oblique view. listed in Table 1. Patients with evidence of CC were Follow-up: Demographic, procedural, and fol- different from other patients in many ways. They had low-up data of the patients were collected and stored a lower incidence of anterior AMI, diabetes mellitus, in a prospective database. The following clinical cardiogenic shock, anterograde TIMI grade 2 or 3 events were considered: death, reinfarction, and repeat flow, and a greater incidence of preinfarction angina target vessel revascularization within 6 months of in the 24 hours before AMI onset, as well as a greater initial revascularization. Patients with ⬎1 event were incidence of multivessel disease and total chronic assigned the highest ranked event according to the occlusion. The infarct artery was the left anterior previously mentioned list. Recurrent ischemia was descending artery in 54% of patients without angiodefined as ischemic chest pain with either new ST- graphic evidence of CC, whereas it was the right segment or T-wave changes at rest or on exercise coronary artery in 51% of patients with angiographic testing. Reinfarction was defined as recurrent chest evidence of CC. Quantitative coronary angiography pain with ST-segment or T-wave changes and recur- showed larger reference infarct artery diameters in rent elevation of cardiac enzymes. Repeat target ves- patients with evidence of coronary CC. There was a sel revascularization was defined as PTCA or coro- trend toward longer delay from symptom onset to nary surgery performed due to restenosis or reocclu- recanalization in the CC patient group, whereas left TABLE 1 Baseline Clinical and Angiographic Characteristics of the Patients
122 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 89
JANUARY 15, 2002
TABLE 2 Procedural Characteristics Variable Coronary angioplasty failure Infarct artery stenting Multiple stents Intra-aortic balloon pump TIMI grade 3 flow Postprocedure minimum lumen diameter (mm) No-reflow Peak creatine kinase (U/L) Time-to peak creatine kinase (h)
CC Group (n ⫽ 264)
Non-CC Group (n ⫽ 900)
p Value
6 (2%) 186 (70%) 56 (30%) 19 (7%) 256 (97%) 3.20 ⫾ 0.49
15 (2%) 662 (74%) 153 (23%) 129 (14%) 868 (96%) 3.08 ⫾ 0.51
0.515 0.319 0.050 0.002 0.680 0.001
17 (6%) 2,283 ⫾ 1,964 7.7 ⫾ 4.2
65 (7%) 2,526 ⫾ 2,296 7.7 ⫾ 4.4
0.662 0.126 0.683
ter was larger in the CC group (3.20 ⫾ 0.49 mm vs 3.08 ⫾ 0.51 mm, p ⫽ 0.001). Intra-aortic balloon pump was used less frequently in the CC group. The incidence of no-reflow was almost identical in the 2 groups. There were no differences between the 2 groups in peak creatine kinase values or in the time-to peak creatine kinase values. Clinical and angiographic outcomes: Clinical follow-up data were
available for all patients and are listed in Table 3. At 6 months, the mortality rate was 4% in the CC group and 9% in the non-CC group TABLE 3 Six-Month Outcome (p ⫽ 0.011), whereas there were no Variable CC Group Non-CC Group p Value differences in reinfarction and target Clinical outcome n⫽264 n⫽900 vessel revascularization rates. There Death 11 (4%) 80 (9%) 0.011 were no differences in mortality beReinfarction 4 (2%) 16 (2%) 0.077 tween patients with cardiogenic Target vessel revascularization 43 (16%) 130 (14%) 0.459 shock and evidence of CC and those Coronary angioplasty 37 110 Coronary bypass 6 20 without evidence of CC (26% and All events 58 (22%) 226 (25%) 0.296 38%, respectively, p ⫽ 0.278). FigAngiographic outcome (n⫽205) (n⫽664) ure 1 shows survival curves for the 2 Follow-up rate 86% 86% patient groups. Multivariate analyses Infarct artery patency rate 93% 93% for death and composite major adRestenosis 66 (32%) 197 (30%) 0.491 Left ventricular ejection 54 ⫾ 13 55 ⫾11 0.529 verse events (death, reinfarction, refraction (%) peat target vessel revascularization) at 6 months are listed in Table 4, whereas the analysis of factors related to evidence of CC and to cardiogenic shock are listed in Table 5. The variables related to mortality were age, previous myocardial infarction, anterior AMI, cardiogenic shock, and primary failure, whereas angiographic evidence of CC did not emerge either as a significant variable in relation to the 6-month mortality or to the 6-month composite end point of death, reinfarction, and repeat target vessel revascularization. By multivariate analysis, the factors related to evidence of CC were female gender, preinfarction angina, anterior AMI, total chronic occlusion, cardiogenic shock, time-to-treatment, and baseline infarct artery TIMl flow FIGURE 1. Kaplan-Meier curves for survival for 264 patients with angiographic evigrade ⬎1. The variables independence of CC supplying the infarct area (CC), and for 900 patients without evidence dently related to cardiogenic shock of CC (non-CC). were age, anterior AMI, multivessel disease, total chronic occlusion, eviventricular ejection fraction was identical in the 2 dence of CC, and a baseline infarct artery TIMl flow grade ⬎1. The 6-month angiographic follow-up data groups. Procedural data: Table 2 lists the procedural data. were available for 869 of 1,013 eligible patients (folThe primary success rate was 98% in both groups. The low-up rate 86%) and are listed in Table 3. The incidence of stenting procedures was similar in the 2 6-month patency rate was identical in the 2 groups groups, but more patients of the CC group had mul- (93%), whereas the restenosis rates were similar (32% tiple stent implantation (30% vs 23%, p ⫽ 0.050). As in the CC group and 30% in the non-CC group, p ⫽ a consequence of the larger reference infarct artery 0.491). Multivariate analysis showed that age (OR diameter, the postprocedural minimum lumen diame- 1.04, 95% CI 1.02 to 1.05, p ⬍0.001), diabetes (OR CORONARY ARTERY DISEASE/COLLATERAL CIRCULATION AND PRIMARY ANGIOPLASTY
123
TABLE 4 Factors Related to Clinical Outcome Death HR (95% CI)
Variable Age (yrs) Diabetes mellitus Previous MI Acute anterior wall MI Cardiogenic shock Primary failure Primary stenting
1.05 (1.03–1.07)
⬍0.001
1.65 2.15 8.79 3.32
0.044 0.002 ⬍0.001 0.005
(1.01–2.68) (1.32–3.50) (5.73–13.49) (1.43–7.71)
MACE HR (95% CI)
p Value
1.02 (1.01–1.03) 1.45 (1.09–1.94)
⬍0.001 0.010
1.41 (1.10–1.81) 3.00 (2.28–3.93)
0.005 ⬍0.001
0.66 (0.51–0.85)
0.001
p Value
MACE ⫽ major adverse cardiac events; MI ⫽ myocardial infarction.
analysis. Consistent with the results of previous studies, the other variables related to CC were total chronic occlusion, a baseline infarct artery TIMl grade flow ⬎1, and cardiogenic shock.9 –13 This could be explained by the lower gradient across the coronary collateral network in patients with cardiogenic shock that may prevent the visualization of CC. Clinical and angiographic outcomes: The mortality rate was lower
in patients with coronary CC. However, after adjustement for clinical TABLE 5 Factors Related to Coronary Collateral Circulation (CC) and Cardiogenic and angiographic variables by multiShock variate analysis, CC was not predicCC Cardiogenic Shock tive of better survival. This may be explained considering the fact that in Variable OR (95% CI) p Value OR (95% CI) p Value the setting of AMI and ST-segment Age (yrs) 1.03 (1.01–1.04) 0.002 elevation, transmural myocardial Preangina pectoris 1.74 (1.30–2.34) ⬍0.001 ischemia subsequent to infarct artery Anterior wall AMI 0.60 (0.45–0.80) ⬍0.001 2.74 (1.84–4.08) ⬍0.001 Cardiogenic shock 0.52 (0.32–0.87) 0.012 occlusion is not apparently reduced Time to treatment (h) 1.07 (1.00–1.15) 0.038 by the early recruitment of collateral TIMI grade 2 or 3 flow 0.39 (0.25–0.61) ⬍0.001 0.43 (0.25–0.76) 0.004 vessels, whereas the early restoration CC 0.53 (0.32–0.87) 0.012 of the anterograde flow in the infarct Female gender 0.70 (0.48–0.99) 0.048 Multivessel coronary disease 1.53 (1.00–2.32) 0.048 artery provided by mechanical interTotal chronic occlusion 2.36 (1.58–3.50) ⬍0.001 3.31 (2.06–5.31) ⬍0.001 vention instantaneously overcomes the collateral flow in the area at risk, where CC could have a protective effect in maintaining myocardial vi1.73, 95% CI 1.12 to 2.67, p ⫽ 0.013), primary ability only if anterograde flow is persistently absent stenting (OR 0.56, 95% CI 0.37 to 0.85, p ⫽ 0.006), or severely reduced. Statistical analysis showed that 2 multiple stents (OR 2.05, 95% CI 1.37 to 3.06, p of the variables that were strong predictors of clinical ⬍0.001), and postprocedure minimum lumen diame- outcomes, such as anterior AMI and cardiogenic ter (OR 0.46, 95% CI 0.25 to 0.85, p ⫽ 0.014) were shock, were inversely related to the angiographic evrelated to the risk of restenosis. Angiograpbic evi- idence of coronary CC, and this association may exdence of coronary CC was not related to the risk of plain the difference in mortality rates as revealed in restenosis. Paired baseline and 6-month left ventricu- this study. Moreover, in survivors, CC was not assolograms were available for 932 patients. At 6 months, ciated with a lower peak creatine kinase release or left ventricular ejection fraction was identical in the 2 increased left ventricular function recovery at 6 groups, as was the improvement in left ventricular months, thereby supporting the hypothesis that early evidence of CC cannot have any significant effect in ejection fraction. patients who undergo infarct-related artery reperfusion within 6 hours of symptom onset. DISCUSSION Variables related to the angiographic evidence of coronary CC: This study is based on a large series of
patients with AMI who underwent coronary angiography within 6 hours of symptom onset. The incidence of CC supplying the infarct area was 23%. The incidence of recurrent episodes of angina in the 24 hours before AMI onset was higher in the CC group. This is consistent with the results of previous studies that have shown that sequential episodes of myocardial ischemia are a stimulating factor in the recruitment of collateral channels.5–9 Early recruitment of CC was more frequent in patients with acute right coronary occlusion, suggesting a more extensive collateralization from the left coronary system to the area supplied from the right coronary artery. Recruitment of collateral flow was also favored by a longer delay from symptom onset to treatment as found by multivariate 124 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 89
1. Sabia PJ, Powers ER, Jayaweera AR, Ragosta M, Kaul S. Functional signif-
icance of collateral blood flow in patients with recent acute myocardial infarction. A study using myocardial contrast echocardiography. Circulation 1992;85:2080 – 2089. 2. Sabia PJ, Powers ER, Ragosta M, Sarembock U, Burwell LR, Kaul S. An association between collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Engl J Med 1992;327:1825–1831. 3. The TIMl Study Group. The Thrombolysis In Myocardial Infarction (TIMl) trial: phase 1 findings. N Engl J Med 1985;312:932–936. 4. Rentrop KP, Cohen M, Blancke H, Phillips RA. Changes in collateral channel filling immediately after controlled coronary artery occlusion by an angioplasty balloon in human subjects. J Am Coll Cardiol 1985;5:587–592. 5. Takeshita A, Koiwaya Y, Nakamura M, Yamamoto K, Torii S. Immediate appearence of coronary collaterals during ergonovine-induced arterial spasm. Chest 1982;82:319 –322. 6. Tada M, Yamagishi M, Kodama K, Kuzuya T, Nanto S, Inoue M, Abe H. Transient collateral augmentation during coronary arterial spasm associated with ST-segment depression. Circulation 1983;67:693–698.
JANUARY 15, 2002
7. Inoue K, Ito H, Kitakaze M, Kuzuya T, Hori M, Iwakura K, Nishikawa N, Higashino Y, Fujii K, Minamino T. Antecedent angina pectoris as a predictor of better functional and clinical outcomes in patients with an inferior wall acute myocardial infarction. Am J Cardiol 1999;83:159 –163. 8. Cribier A, Korsatz L, Koning R, Rath P, Gamra H, Stix G, Merchant S, Chan C, Letac B. Improved myocardial ischemic response and enhanced collateral circulation with long repetitive coronary occlusion during angiopiasty: a prospective study. J Am Coll Cardiol 1992;20:578 –586. 9. Perez-Castellano N, Garcia EJ, Abeytua M, Soriano J, Serrano JA, Elizaga J, Botas J, Lopez-Sendon JL, Delcan JL. Influence of collateral circulation on in-hospital death from anterior acute myocardial infarction. J Am Coll Cardiol 1998;31:512–518. 10. Nitzberg WD, Nath HP, Rogers WJ, Hood WR Jr, Whitlow PL, Reeves R,
Baxley WA. Collateral flow in patients with acute myocardial infarction. Am J Cardiol 1985;56:729 –736. 11. Williams DO, Amsterdam EA, Miller RR, Mason DT. Functional significance of coronary collateral vessels in patients with acute myocardial infarction: relation to pump performance, cardiogenic shock and survival. Am J Cardiol 1976; 37:345–351. 12. Hansen JF. Coronary collateral circulation: clinical significance and influence on survival in patients with coronary artery occlusion. Am Heart J 1989;117: 290 –295. 13. Vyssoulis G, Kyriakidis M, Karpanou E, Kyriakidis C, Sfikakis P, Barbetseas J, Toutouzas P. A prospective angiographic study of the coronary collateral circulation in coronary artery disease. Int J Cardiol 1990;27:187–191.
CORONARY ARTERY DISEASE/COLLATERAL CIRCULATION AND PRIMARY ANGIOPLASTY
125
0.029), anterograde TIMI grade flow >1 (10% vs 21%, p <0.001), and a greater incidence of preinfarction angina (43% vs 32%, p ⴝ 0.001), multivessel disease (59% vs 47%, p ⴝ 0.001), and total chronic occlusion (20% vs 10%, p <0.001). At 6 months, the mortality rate was lower in patients with CC compared with patients without CC (4% vs 9%, p ⴝ 0.011), whereas there were no differences in the incidence of reinfarction, target vessel revascularization, and angiographic restenosis. After multivariate analysis, CC did not emerge as a significant variable in relation to 6-month clinical and angiographic outcomes. CC does not exert a protective effect in patients who undergo mechanical intervention in the first 6 hours of AMI onset. 䊚2002 by Excerpta Medica, Inc. (Am J Cardiol 2002;89:121–125)
oronary collateral circulation (CC) may play an important role in maintaining viable myocardium C after abrupt coronary occlusion, and it is a primary
our institutions for all patients with AMI admitted within 6 hours of symptom onset or within 24 hours if there was evidence of continuing ischemia, without any restrictions based on age, sex, or clinical status on presentation. Criteria for enrollment for this study included: chest pain persisting for ⬎30 minutes associated with ST-T segment elevation of ⱖ0.1 mV in ⱖ2 contiguous electrocardiographic leads. Coronary angiography was required in all patients at 6 months after the procedure. Unscheduled angiography was allowed on the basis of clinical indication. Left ventriculography was performed in all patients if not contraindicated by cardiogenic shock or severe hemodynamic instability. Creatine kinase measurements were systematically performed on admission and every 3 hours for the first 24 hours, and then every 12 hours for 2 days. The peak value of creatine kinase and the time-to-peak of creatine kinase were estimated for each patient. Angiographic analysis: Coronary flow in the infarct artery was graded according to Thrombolysis In Myocardial Infarction (TIMI) study group classification.3 Collateral flow before direct angioplasty was graded using the classification developed by Rentrop et al4:
determinant of left ventricular function recovery after late mechanical infarct artery reperfusion.1,2 It is unknown whether early CC recruitment has a similar beneficial effect on clinical and angiographic outcomes in patients who undergo mechanical intervention in the first hours of onset of acute myocardial infarction (AMI). This study assesses the significance of preintervention angiographic evidence of CC in patients with AMI who undergo mechanical intervention within 6 hours of symptom onset.
METHODS
Patients and treatment protocol: Since January 1995, primary percutaneous transluminal coronary angioplasty (PTCA) has been the systematic treatment at From the Division of Cardiology, Careggi Hospital, Florence, Italy. Manuscript received July 2, 2001; revised manuscript received and accepted September 24, 2001. Address for reprints: David Antoniucci, MD, Division of Cardiology, Careggi Hospital, Viale Morgagni, I-50134, Florence, Italy. E-mail: [email protected]. ©2002 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 89 January 15, 2002
0002-9149/02/$–see front matter PII S0002-9149(01)02186-5
121
sion of the target lesion in association with objective evidence CC Group Non-CC Group of ischemia or viable myocardium. Variable (n ⫽ 264) (n ⫽ 900) p Value Statistical analysis: Categorical data are presented as absolute values Age (yrs) 63 ⫾ 12 64 ⫾ 12 0.146 Men 216 (82%) 690 (77%) 0.076 and percentages, whereas continuous Systemic hypertension 96 (36%) 290 (32%) 0.209 data are summarized as mean ⫾ SD. Diabetes mellitus 28 (11%) 143 (16%) 0.033 A 2-tailed Student’s t test was used Hypercholesterolemia (⬎200 mg/dl) 70 (26%) 229 (25%) 0.726 to test differences among continuous Previous myocardial infarction 40 (15%) 105 (12%) 0.132 Preinfarction angina pectoris 114 (43%) 292 (32%) 0.001 variables, whereas categorical variCardiogenic shock 23 (9%) 124 (14%) 0.029 ables were compared by chi-square Onset of symptoms to treatment (h) 3.9 ⫾ 2.3 3.6 ⫾ 1.9 0.067 analysis or Fisher’s exact test. The Anterior wall AMI 108 (41%) 491 (55%) ⬍0.001 contribution of clinical, angioLeft ventricular ejection fraction (%)* 48.1 ⫾ 11.9 48.3 ⫾ 12.2 0.907 graphic, and procedural variables to Infarct-related coronary artery† ⬍0.001 Left anterior descending 102 (39%) 486 (54%) the angiographic and clinical outRight 136 (51%) 276 (31%) comes was evaluated with multivarLeft circumflex 23 (9%) 126 (14%) iate logistic regression analysis and Left main 3 (1%) 12 (1%) the Cox proportional hazard model, Multivessel coronary disease 155 (59%) 426 (47%) 0.001 Chronic occlusion 52 (20%) 94 (10%) ⬍0.001 respectively. The odds ratio (OR) TIMI grade flow ⬎1 26 (10%) 188 (21%) ⬍0.001 and their 95% confidence intervals Reference vessel diameter (mm) 3.18 ⫾ 0.47 3.06 ⫾ 0.49 ⬍0.001 (CI) were calculated. The variables *CC group (n ⫽ 226) and non-CC group (n ⫽ 706). used for the analyses included age, † Included 2 coronary grafts in the CC group, and 5 coronary grafts in the non-CC group. gender, diabetes, previous myocardial infarction, preinfarction angina, anterior AMI, cardiogenic shock, grade 0 ⫽ no visible filling of any collateral channels; multivessel disease, total chronic occlusion, coronary 1 ⫽ filling of side branches of the infarct artery with CC, baseline infarct artery TIMI grade flow ⬎1, time no dye reaching the epicardial segment; 2 ⫽ partial to reperfusion, reference infarct artery diameter, infilling of the epicardial vessel; and 3 ⫽ complete farct artery stenting, multiple stents, failure of PTCA, filling of the epicardial vessel by collateral vessels. and postprocedure minimum lumen diameter. SurTwo observers assessed the coronary angiograms in a vival curves were generated using the Kaplan-Meier blinded fashion and reached a consensus regarding the method. Comparison between survival curves was TIMI flow grade and the collateral flow grade. Quan- performed using the log-rank test. A p value ⬍0.05 titative coronary angiography was performed with the was considered significant. Statistical tests were peruse of an automatic edge detection system (Siemens formed using package SPSS 7.0 (SPSS Inc., Chicago, Ancor, Solna, Sweden). Coronary occlusion was as- Illinois). signed a value of 0 mm for minimal luminal diameter and 100% for percent diameter stenosis. An optimal RESULTS angiographic result was defined as residual stenosis Patients: From January 1995 to December 2000, ⬍30% associated with TIMI grade 3 flow. An unsuc- 1,164 patients underwent primary mechanical intercessful procedure was defined as a procedure resulting vention for AMI within 6 hours of symptom onset. Of in TIMI grade 0 or 1 flow whatever the residual the 1,164 patients, 264 (23%) had angiographic evistenosis was. Angiographic restenosis was defined as dence of CC flow grade ⬎1 (CC group), whereas in ⬎50% stenosis of the target lesion on the scheduled or 900 patients, coronary angiography did not show corunscheduled follow-up angiogram. Left ventricular onary CC (non-CC group). Baseline clinical and anfunction was evaluated by contrast ventriculography giographic characteristics of the patient groups are in the right anterior oblique view. listed in Table 1. Patients with evidence of CC were Follow-up: Demographic, procedural, and fol- different from other patients in many ways. They had low-up data of the patients were collected and stored a lower incidence of anterior AMI, diabetes mellitus, in a prospective database. The following clinical cardiogenic shock, anterograde TIMI grade 2 or 3 events were considered: death, reinfarction, and repeat flow, and a greater incidence of preinfarction angina target vessel revascularization within 6 months of in the 24 hours before AMI onset, as well as a greater initial revascularization. Patients with ⬎1 event were incidence of multivessel disease and total chronic assigned the highest ranked event according to the occlusion. The infarct artery was the left anterior previously mentioned list. Recurrent ischemia was descending artery in 54% of patients without angiodefined as ischemic chest pain with either new ST- graphic evidence of CC, whereas it was the right segment or T-wave changes at rest or on exercise coronary artery in 51% of patients with angiographic testing. Reinfarction was defined as recurrent chest evidence of CC. Quantitative coronary angiography pain with ST-segment or T-wave changes and recur- showed larger reference infarct artery diameters in rent elevation of cardiac enzymes. Repeat target ves- patients with evidence of coronary CC. There was a sel revascularization was defined as PTCA or coro- trend toward longer delay from symptom onset to nary surgery performed due to restenosis or reocclu- recanalization in the CC patient group, whereas left TABLE 1 Baseline Clinical and Angiographic Characteristics of the Patients
122 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 89
JANUARY 15, 2002
TABLE 2 Procedural Characteristics Variable Coronary angioplasty failure Infarct artery stenting Multiple stents Intra-aortic balloon pump TIMI grade 3 flow Postprocedure minimum lumen diameter (mm) No-reflow Peak creatine kinase (U/L) Time-to peak creatine kinase (h)
CC Group (n ⫽ 264)
Non-CC Group (n ⫽ 900)
p Value
6 (2%) 186 (70%) 56 (30%) 19 (7%) 256 (97%) 3.20 ⫾ 0.49
15 (2%) 662 (74%) 153 (23%) 129 (14%) 868 (96%) 3.08 ⫾ 0.51
0.515 0.319 0.050 0.002 0.680 0.001
17 (6%) 2,283 ⫾ 1,964 7.7 ⫾ 4.2
65 (7%) 2,526 ⫾ 2,296 7.7 ⫾ 4.4
0.662 0.126 0.683
ter was larger in the CC group (3.20 ⫾ 0.49 mm vs 3.08 ⫾ 0.51 mm, p ⫽ 0.001). Intra-aortic balloon pump was used less frequently in the CC group. The incidence of no-reflow was almost identical in the 2 groups. There were no differences between the 2 groups in peak creatine kinase values or in the time-to peak creatine kinase values. Clinical and angiographic outcomes: Clinical follow-up data were
available for all patients and are listed in Table 3. At 6 months, the mortality rate was 4% in the CC group and 9% in the non-CC group TABLE 3 Six-Month Outcome (p ⫽ 0.011), whereas there were no Variable CC Group Non-CC Group p Value differences in reinfarction and target Clinical outcome n⫽264 n⫽900 vessel revascularization rates. There Death 11 (4%) 80 (9%) 0.011 were no differences in mortality beReinfarction 4 (2%) 16 (2%) 0.077 tween patients with cardiogenic Target vessel revascularization 43 (16%) 130 (14%) 0.459 shock and evidence of CC and those Coronary angioplasty 37 110 Coronary bypass 6 20 without evidence of CC (26% and All events 58 (22%) 226 (25%) 0.296 38%, respectively, p ⫽ 0.278). FigAngiographic outcome (n⫽205) (n⫽664) ure 1 shows survival curves for the 2 Follow-up rate 86% 86% patient groups. Multivariate analyses Infarct artery patency rate 93% 93% for death and composite major adRestenosis 66 (32%) 197 (30%) 0.491 Left ventricular ejection 54 ⫾ 13 55 ⫾11 0.529 verse events (death, reinfarction, refraction (%) peat target vessel revascularization) at 6 months are listed in Table 4, whereas the analysis of factors related to evidence of CC and to cardiogenic shock are listed in Table 5. The variables related to mortality were age, previous myocardial infarction, anterior AMI, cardiogenic shock, and primary failure, whereas angiographic evidence of CC did not emerge either as a significant variable in relation to the 6-month mortality or to the 6-month composite end point of death, reinfarction, and repeat target vessel revascularization. By multivariate analysis, the factors related to evidence of CC were female gender, preinfarction angina, anterior AMI, total chronic occlusion, cardiogenic shock, time-to-treatment, and baseline infarct artery TIMl flow FIGURE 1. Kaplan-Meier curves for survival for 264 patients with angiographic evigrade ⬎1. The variables independence of CC supplying the infarct area (CC), and for 900 patients without evidence dently related to cardiogenic shock of CC (non-CC). were age, anterior AMI, multivessel disease, total chronic occlusion, eviventricular ejection fraction was identical in the 2 dence of CC, and a baseline infarct artery TIMl flow grade ⬎1. The 6-month angiographic follow-up data groups. Procedural data: Table 2 lists the procedural data. were available for 869 of 1,013 eligible patients (folThe primary success rate was 98% in both groups. The low-up rate 86%) and are listed in Table 3. The incidence of stenting procedures was similar in the 2 6-month patency rate was identical in the 2 groups groups, but more patients of the CC group had mul- (93%), whereas the restenosis rates were similar (32% tiple stent implantation (30% vs 23%, p ⫽ 0.050). As in the CC group and 30% in the non-CC group, p ⫽ a consequence of the larger reference infarct artery 0.491). Multivariate analysis showed that age (OR diameter, the postprocedural minimum lumen diame- 1.04, 95% CI 1.02 to 1.05, p ⬍0.001), diabetes (OR CORONARY ARTERY DISEASE/COLLATERAL CIRCULATION AND PRIMARY ANGIOPLASTY
123
TABLE 4 Factors Related to Clinical Outcome Death HR (95% CI)
Variable Age (yrs) Diabetes mellitus Previous MI Acute anterior wall MI Cardiogenic shock Primary failure Primary stenting
1.05 (1.03–1.07)
⬍0.001
1.65 2.15 8.79 3.32
0.044 0.002 ⬍0.001 0.005
(1.01–2.68) (1.32–3.50) (5.73–13.49) (1.43–7.71)
MACE HR (95% CI)
p Value
1.02 (1.01–1.03) 1.45 (1.09–1.94)
⬍0.001 0.010
1.41 (1.10–1.81) 3.00 (2.28–3.93)
0.005 ⬍0.001
0.66 (0.51–0.85)
0.001
p Value
MACE ⫽ major adverse cardiac events; MI ⫽ myocardial infarction.
analysis. Consistent with the results of previous studies, the other variables related to CC were total chronic occlusion, a baseline infarct artery TIMl grade flow ⬎1, and cardiogenic shock.9 –13 This could be explained by the lower gradient across the coronary collateral network in patients with cardiogenic shock that may prevent the visualization of CC. Clinical and angiographic outcomes: The mortality rate was lower
in patients with coronary CC. However, after adjustement for clinical TABLE 5 Factors Related to Coronary Collateral Circulation (CC) and Cardiogenic and angiographic variables by multiShock variate analysis, CC was not predicCC Cardiogenic Shock tive of better survival. This may be explained considering the fact that in Variable OR (95% CI) p Value OR (95% CI) p Value the setting of AMI and ST-segment Age (yrs) 1.03 (1.01–1.04) 0.002 elevation, transmural myocardial Preangina pectoris 1.74 (1.30–2.34) ⬍0.001 ischemia subsequent to infarct artery Anterior wall AMI 0.60 (0.45–0.80) ⬍0.001 2.74 (1.84–4.08) ⬍0.001 Cardiogenic shock 0.52 (0.32–0.87) 0.012 occlusion is not apparently reduced Time to treatment (h) 1.07 (1.00–1.15) 0.038 by the early recruitment of collateral TIMI grade 2 or 3 flow 0.39 (0.25–0.61) ⬍0.001 0.43 (0.25–0.76) 0.004 vessels, whereas the early restoration CC 0.53 (0.32–0.87) 0.012 of the anterograde flow in the infarct Female gender 0.70 (0.48–0.99) 0.048 Multivessel coronary disease 1.53 (1.00–2.32) 0.048 artery provided by mechanical interTotal chronic occlusion 2.36 (1.58–3.50) ⬍0.001 3.31 (2.06–5.31) ⬍0.001 vention instantaneously overcomes the collateral flow in the area at risk, where CC could have a protective effect in maintaining myocardial vi1.73, 95% CI 1.12 to 2.67, p ⫽ 0.013), primary ability only if anterograde flow is persistently absent stenting (OR 0.56, 95% CI 0.37 to 0.85, p ⫽ 0.006), or severely reduced. Statistical analysis showed that 2 multiple stents (OR 2.05, 95% CI 1.37 to 3.06, p of the variables that were strong predictors of clinical ⬍0.001), and postprocedure minimum lumen diame- outcomes, such as anterior AMI and cardiogenic ter (OR 0.46, 95% CI 0.25 to 0.85, p ⫽ 0.014) were shock, were inversely related to the angiographic evrelated to the risk of restenosis. Angiograpbic evi- idence of coronary CC, and this association may exdence of coronary CC was not related to the risk of plain the difference in mortality rates as revealed in restenosis. Paired baseline and 6-month left ventricu- this study. Moreover, in survivors, CC was not assolograms were available for 932 patients. At 6 months, ciated with a lower peak creatine kinase release or left ventricular ejection fraction was identical in the 2 increased left ventricular function recovery at 6 groups, as was the improvement in left ventricular months, thereby supporting the hypothesis that early evidence of CC cannot have any significant effect in ejection fraction. patients who undergo infarct-related artery reperfusion within 6 hours of symptom onset. DISCUSSION Variables related to the angiographic evidence of coronary CC: This study is based on a large series of
patients with AMI who underwent coronary angiography within 6 hours of symptom onset. The incidence of CC supplying the infarct area was 23%. The incidence of recurrent episodes of angina in the 24 hours before AMI onset was higher in the CC group. This is consistent with the results of previous studies that have shown that sequential episodes of myocardial ischemia are a stimulating factor in the recruitment of collateral channels.5–9 Early recruitment of CC was more frequent in patients with acute right coronary occlusion, suggesting a more extensive collateralization from the left coronary system to the area supplied from the right coronary artery. Recruitment of collateral flow was also favored by a longer delay from symptom onset to treatment as found by multivariate 124 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 89
1. Sabia PJ, Powers ER, Jayaweera AR, Ragosta M, Kaul S. Functional signif-
icance of collateral blood flow in patients with recent acute myocardial infarction. A study using myocardial contrast echocardiography. Circulation 1992;85:2080 – 2089. 2. Sabia PJ, Powers ER, Ragosta M, Sarembock U, Burwell LR, Kaul S. An association between collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Engl J Med 1992;327:1825–1831. 3. The TIMl Study Group. The Thrombolysis In Myocardial Infarction (TIMl) trial: phase 1 findings. N Engl J Med 1985;312:932–936. 4. Rentrop KP, Cohen M, Blancke H, Phillips RA. Changes in collateral channel filling immediately after controlled coronary artery occlusion by an angioplasty balloon in human subjects. J Am Coll Cardiol 1985;5:587–592. 5. Takeshita A, Koiwaya Y, Nakamura M, Yamamoto K, Torii S. Immediate appearence of coronary collaterals during ergonovine-induced arterial spasm. Chest 1982;82:319 –322. 6. Tada M, Yamagishi M, Kodama K, Kuzuya T, Nanto S, Inoue M, Abe H. Transient collateral augmentation during coronary arterial spasm associated with ST-segment depression. Circulation 1983;67:693–698.
JANUARY 15, 2002
7. Inoue K, Ito H, Kitakaze M, Kuzuya T, Hori M, Iwakura K, Nishikawa N, Higashino Y, Fujii K, Minamino T. Antecedent angina pectoris as a predictor of better functional and clinical outcomes in patients with an inferior wall acute myocardial infarction. Am J Cardiol 1999;83:159 –163. 8. Cribier A, Korsatz L, Koning R, Rath P, Gamra H, Stix G, Merchant S, Chan C, Letac B. Improved myocardial ischemic response and enhanced collateral circulation with long repetitive coronary occlusion during angiopiasty: a prospective study. J Am Coll Cardiol 1992;20:578 –586. 9. Perez-Castellano N, Garcia EJ, Abeytua M, Soriano J, Serrano JA, Elizaga J, Botas J, Lopez-Sendon JL, Delcan JL. Influence of collateral circulation on in-hospital death from anterior acute myocardial infarction. J Am Coll Cardiol 1998;31:512–518. 10. Nitzberg WD, Nath HP, Rogers WJ, Hood WR Jr, Whitlow PL, Reeves R,
Baxley WA. Collateral flow in patients with acute myocardial infarction. Am J Cardiol 1985;56:729 –736. 11. Williams DO, Amsterdam EA, Miller RR, Mason DT. Functional significance of coronary collateral vessels in patients with acute myocardial infarction: relation to pump performance, cardiogenic shock and survival. Am J Cardiol 1976; 37:345–351. 12. Hansen JF. Coronary collateral circulation: clinical significance and influence on survival in patients with coronary artery occlusion. Am Heart J 1989;117: 290 –295. 13. Vyssoulis G, Kyriakidis M, Karpanou E, Kyriakidis C, Sfikakis P, Barbetseas J, Toutouzas P. A prospective angiographic study of the coronary collateral circulation in coronary artery disease. Int J Cardiol 1990;27:187–191.
CORONARY ARTERY DISEASE/COLLATERAL CIRCULATION AND PRIMARY ANGIOPLASTY
125