Coronary Blood Flow Assessment After Successful Angioplasty for Acute Myocardial Infarction Predicts the Risk of Long-Term Cardiac Events
Background— Analysis of coronary flow velocity (CFV) in the recanalized infarct-related coronary artery (IRA) with a Doppler guidewire is useful for predicting recovery of regional left ventricular function, in-hospital complications, and survival. We postulated that the CFV pattern after IRA reperfusion for acute myocardial infarction (AMI) would predict long-term adverse cardiac events.
Methods and Results— Sixty-eight consecutive patients with a first AMI underwent CFV measurement with a Doppler guidewire after successful reopening of the IRA by coronary angioplasty. At the end of follow-up, 3.8±1.7 years after AMI, 44 of the 65 surviving patients (67.7%) were free of long-term cardiac events. Univariate analysis showed that the following factors were predictive of an end point combining cardiac death, recurrent MI, and congestive heart failure: hypertension, age ≥65 years, time from onset of chest pain to PTCA ≥6 hours, peak creatine kinase >4000 IU/L, ejection fraction ≤50%, proximal left anterior descending artery occlusion, resting average peak velocity ≤10 cm/s, average systolic peak velocity ≤5 cm/s, a rapid diastolic deceleration time (≤600 ms), and early retrograde systolic flow. In the final multivariate model, only age ≥65 years (OR, 3.6; 95% CI, 1.1 to 11.8; P=0.03), time to PTCA ≥6 hours (OR, 2.9; 95% CI, 1.0 to 8.3; P=0.04), and a rapid diastolic deceleration time (OR, 5.4; 95% CI, 1.5 to 19.3; P=0.01) were independent predictors.
Conclusions— The CFV pattern appears to be an accurate predictor of long-term cardiac events in patients having undergone successful reopening of the IRA after AMI, identifying a subset of at-risk patients.
Received June 1, 2004; revision received July 21, 2004; accepted August 3, 2004.
Rapid coronary reperfusion during acute myocardial infarction (AMI) reduces mortality1 and preserves contractile function.2 The subsequent quality of coronary flow appears to be an important prognostic factor.3 However, angiographic success does not necessarily equate with adequate myocardial perfusion, even after early revascularization.4 A damaged microcirculation5 might explain this “no-reflow” phenomenon.6,7
Studies using a Doppler guidewire have shown that coronary blood flow velocity (CFV) measurement in recanalized infarcted arteries can predict recovery of regional left ventricular (LV) function,8–10 as well as in-hospital complications and survival.11 However, the relationship between the CFV pattern after reperfusion and long-term cardiac events is not fully clear. We postulated that qualitative and quantitative CFV alterations might be associated with the risk of in-hospital and long-term adverse clinical events. To test this hypothesis, we prospectively analyzed the risk of cardiac events according to CFV parameters, as measured with a Doppler guidewire, after reperfusion of the infarct-related coronary artery (IRA) in 68 patients with AMI.
Sixty-eight consecutive patients with a first AMI underwent coronary angioplasty and coronary flow measurement with a Doppler guidewire. All patients gave informed consent to CFV measurements during angioplasty, and our institutional review board approved the study protocol.
The diagnosis of AMI was based on chest pain lasting >30 minutes, ST-segment elevation of ≥2 mm on at least 2 contiguous ECG leads, and serum creatine kinase (CK) levels >3 times the upper limit of normal. Patients with aortic stenosis, hypertrophic cardiomyopathy, significant left main trunk stenosis, or malignant arrhythmias at the time of CFV determination were excluded.
Before angioplasty, all patients received intravenous aspirin (≥250 mg), an intracoronary heparin bolus (100 U/kg), and intracoronary linsidomine hydrochloride, a nitroglycerine-like drug (1 mg). After a 6F guiding catheter was placed in the coronary ostium, a 0.014-in-diameter Doppler guidewire was inserted for use as the primary angioplasty guidewire (FloWire, Cardiometrics, Inc). After documentation of angiographic success [residual stenosis of the infarct-related coronary artery (IRA) <40% and flow restoration] and determination of the TIMI flow grade, a final CFV recording was made distal to the culprit lesion 15 minutes after the last balloon inflation to avoid the influence of possible reactive hyperemia caused by the contrast medium. In the right coronary artery, the CFV measurement was made in the third segment or downstream of it to avoid the usual prominent positive systolic component that may prevent the detection of a retrograde (ie, negative) systolic signal.
The following parameters were computed: the time-averaged peak velocity normalized to the cardiac cycle (APV, cm/s), which is the time average of the spectral peak velocity waveform of 2 cardiac cycles; average systolic peak velocity (ASPV, cm/s); average diastolic peak velocity (cm/s); and the ratio of average diastolic to average systolic velocity. Spectral analysis of the signal and Doppler audio signals were videorecorded. Coronary velocity reserve (CVR) was calculated as the ratio of hyperemic (APVpeak) to resting APV (APVbase) after intracoronary injection of 18 μg adenosine in the last 26 patients. In accordance with previous findings,12 we defined microvascular injury (no-reflow phenomenon) as the presence of early systolic retrograde flow and a rapid deceleration time of diastolic velocity (≤600 ms).
Analysis of Coronary Angiographic Findings
Before coronary angioplasty, collateral vessels were classified according to the Rentrop classification.13 After angiographically successful coronary angioplasty, 2 cineangiograms were acquired in orthogonal projections. Two independent observers who were blinded to the patient’s other data evaluated the anterograde radio-contrast flow of the IRA using TIMI criteria and quantified the residual percentage coronary diameter stenosis using a commercially available quantitative cardiovascular angiographic software program and the guiding catheter (filmed without contrast) as a scaling tool.
An independent investigator blinded to coronary flow measurements carefully reviewed cardiac events. Major adverse cardiac events included death, nonfatal reinfarction, target lesion revascularization, and congestive heart failure (CHF). CHF was defined as the presence of dyspnea, Killip class 2, or evidence of pulmonary congestion on chest radiographs; target lesion revascularization was defined as repeated PTCA of the initial culprit lesion or coronary artery bypass surgery.
Continuous variables are expressed as mean±SD. Proportions were compared by use of the χ2 statistic, and Fisher’s exact test was used when appropriate. Factors associated with in-hospital and long-term cardiac events were first identified by univariate analysis. Parameters with values of P<0.05 were then included in a multivariate Cox regression model with stepwise selection to identify independent risk factors for cardiac events. For continuous variables, the optimal cutoff was that having the highest χ2 value in the Cox regression model. The variables examined included age ≥65 years, sex, hypertension, hyperlipidemia, diabetes, smoking, 24-hour preinfarction angina, the culprit lesion, multivessel disease, stenting, time from onset of chest pain to PTCA ≥6 hours, collaterals (Rentrop grade 2 and 3), peak serum CK level ≥4000 IU/L, final TIMI flow grade ≤2, APV ≤10 cm/s, ASPV ≤5 cm/s, a rapid deceleration time of diastolic velocity, systolic retrograde flow, and an ejection fraction ≤50%. ORs and relative risks were calculated with their 95% CIs. Cardiac event–free and overall survival curves were traced by use of the Kaplan-Meier method and compared with the log-rank test. All statistical analyses were performed with the Statistical Package for Social Scientists (SPSS Inc, release 10.2 for Windows). A value of P<0.05 was required for statistical significance.
The clinical characteristics of the population are described in Table 1. Sixty-eight patients underwent direct (n=64) or rescue (n=4) coronary angioplasty within 6 hours after chest pain onset in 45 cases, within 12 hours in 9 cases, and within 24 hours in 14 cases. All patients had Q-wave infarction in the anterior (n=32), lateral (n=6), or inferior (n=30) wall. The mean ejection fraction after PTCA was 49.9±11.0%.
Angiographic and Doppler Data
Angioplasty was performed successfully in 65 of the 68 patients (95.6%). Residual stenosis was <40% in 67 patients (19±8%). Stenting was performed in 47 patients because of suboptimal results after repeated balloon inflation (n=38) or large dissection (n=8) or for thrombotic recurrence (n=1).
The coronary occlusion was crossed by the Doppler guidewire in 66 patients; a conventional angioplasty guidewire was necessary in the other 2 patients. At the end of the angioplasty procedure, CFVs were recorded with the Doppler guidewire that had been positioned with a tracking catheter. Doppler signals were difficult to analyze in 5 patients because of tachycardia (n=2) or background noise (n=3). However, all recordings were analyzed on the basis of a consensus of 2 independent experts on Doppler flow mapping. Stented patients received ticlopidine and aspirin; nonstented patients received aspirin alone. Anti–glycoprotein IIb/IIIa agents were not used. Medical treatment was never decided on the basis of CFV measurements. Patients with and without abnormal CFV were treated identically.
Relation Between CFV and In-Hospital Cardiac Events
Fourteen patients had cardiac events during the initial hospital stay. There were 3 cardiac deaths (cardiac rupture in 2 cases, cardiogenic shock in 1 case), 1 nonfatal new MI, 3 CABG procedures, 2 repeated coronary angioplasties, and 9 cases of CHF. Patients with events had higher ST-segment elevations and CK-MB levels than patients without events (5.0±2.6 versus 3.2±2.1 mm, respectively; P=0.01; and 873±836 versus 411±341 IU/L; P=0.004).
Final APV and ASPV were significantly lower in patients with cardiac events than in the other patients (12.1±8.1 versus 18.8±9.0 cm/s, respectively; P=0.014; 6.0±4.2 versus 10.9±6.4 cm/s; P=0.002). APV >10 cm/s and ASPV >5 cm/s were associated with a significantly lower risk of CHF and of reaching the combined end point (cardiac death and/or reinfarction and/or CHF) (Table 2).
APV was similar in patients with and without stent implantation (16.8±8.4 and 19.6±10.4 cm/s, respectively). In the stenting subgroup, APV was also significantly lower in patients with in-hospital cardiac events (10.7±7.0 versus 18.1±8.2 cm/s; P=0.023).
Relation Between CFV and Long-Term Cardiac Outcome
At the end of follow-up, a mean of 3.8±1.7 years after initial AMI, 44 (67.7%) of the 65 surviving patients were free of long-term cardiac events. Among the remaining 21 patients (32.3%), 1 underwent heart transplantation, 3 had coronary artery bypass surgery, 7 had repeated PTCA, and 13 had CHF requiring hospital admission. Of the 5 patients who died, 4 died of cardiac causes.
APV >10 cm/s but not ASPV >5 cm/s was significantly associated with a lower risk of postdischarge cardiac death or heart transplantation (Table 2). Only the final ratio of average diastolic to average systolic velocity was significantly higher in patients with in-hospital or post-discharge cardiac events than in other patients (4.7±4.1 versus 2.7±1.9; P=0.012). During the entire follow-up period, a normal APV (>10 cm/s) after angioplasty was associated with a significantly lower risk of cardiac death and with a markedly but nonsignificantly lower risk of CHF (Table 2). ASPV >5 cm/s was associated with a significantly lower risk of CHF and of reaching the combined end point (cardiac death and/or reinfarction and/or CHF) (Table 2).
Relation Between the CFV Pattern and In-Hospital Cardiac Events
Twenty patients had early retrograde systolic flow, and 31 patients had a rapid deceleration in diastole (Figure 1). None of the patients free of early retrograde systolic flow died or had a new infarct while in hospital, whereas among the patients with early retrograde systolic flow, there were 3 deaths, 1 nonfatal recurrent MI, and 2 urgent revascularization procedures. Patients with a rapid diastolic deceleration time had a significantly higher risk of CHF and of reaching the combined end point of cardiac death and/or reinfarction and/or CHF (Table 3).
Relation Between the CFV Pattern and Long-Term Cardiac Events
Patients with early retrograde systolic flow had an increased risk of cardiac death and CHF during the entire follow-up period, whereas patients with a rapid diastolic deceleration time had an increased risk of CHF (Table 3).
Table 4 shows factors predictive of the combined end point of cardiac death, MI, and/or CHF during the postdischarge period and the entire follow-up period. In the final multivariate model, only age ≥65 years, hypertension, onset of chest pain to PTCA ≥6 hours, and a rapid diastolic deceleration time were independent predictors of postdischarge cardiac events. Except for hypertension, the same independent predictors were found for the entire follow-up period. As shown in Figure 2, the curve of cardiac event–free survival in the group of patients with rapid diastolic deceleration diverged drastically from that of the group of patients with a slower diastolic deceleration time from hospital discharge to the first year.
Previous studies have shown that using a Doppler guidewire offers a better quantitative assessment of CFV14 and microvascular injury9 in patients with reperfused AMI. Likewise, it has been reported that the CFV pattern immediately after primary coronary stenting may be predictive of LV functional recovery during the healing phase of AMI.12,15 More recently, the CFV pattern was found to be an accurate predictor of in-hospital complications and survival after AMI.11 Here, we show that both in-hospital and long-term adverse clinical events can be predicted by the CFV pattern both qualitatively and quantitatively immediately after primary or rescue PTCA, establishing a relationship between the risk of adverse clinical events and microcirculatory function measured by intracoronary Doppler examination.
In the early stages of AMI, it is important to identify individual patients likely to have poor in-hospital outcomes. Among our patients with APV ≤10 cm/s, ASPV ≤5 cm/s, or microvascular injury (rapid diastolic deceleration time), CHF was frequent during the initial hospital stay despite successful coronary recanalization. The infarct area was larger in these patients, establishing a no-reflow CFV pattern before reperfusion therapy and possibly accounting, at least in part, for the increased risk of early CHF. However, microvascular injury is determined not only by the size of the infarct but also by reperfusion injury,16 PTCA-induced thrombus, and atheroma fragmentation.17 Indeed, patients with severe microvascular injury immediately after PTCA have a higher frequency of ST-segment re-elevation11 or incomplete ST-segment resolution,18 and a continuously declining post-PTCA velocity has been linked to vessel reocclusion.19
Several studies have shown that patients with no reflow on myocardial contrast echocardiography20 or Doppler guidewire assessment9,11,12 have poor functional LV recovery and a high frequency of complications such as CHF, pericardial effusion, cardiac tamponade, and cardiac rupture. This link between the CFV pattern and in-hospital cardiac events is confirmed here.
We present the first evidence that no reflow, shown by Doppler guidewire measurement, has long-term prognostic value for fatal and nonfatal cardiac events. Univariate analysis of factors potentially associated with long-term cardiac events identified APV ≤10 cm/s, ASPV ≤5 cm/s, and microvascular injury (rapid diastolic deceleration time and early retrograde systolic flow). Multiple regression analysis showed that age, hypertension, a rapid diastolic deceleration time, and the interval between chest pain onset and reperfusion were the only independent predictors of long-term cardiac events. Thus, no-reflow demonstration with a Doppler guidewire reliably discriminates individual patients with poor long-term outcomes from patients with favorable outcome. Recently, Morishima et al21 reported that angiographic no reflow, defined as TIMI flow grade ≤2, for a mean period of 5.8 years was associated with malignant arrhythmias, a smaller ejection fraction, and more cardiac deaths after successful IRA reopening for AMI. In the study by Ito et al,7 all patients with TIMI flow grade 2 after successful PTCA showed defective myocardial perfusion, reflecting major microvascular damage. Our results support those of Morishima et al,21 although we studied patients successfully recanalized by PTCA (no obstructive lesions and mainly TIMI flow grade 3) with microvascular damage, which cannot be identified by angiographic TIMI grade assessment.
In our study, patients with a rapid diastolic deceleration time were more likely than patients with a normal diastolic deceleration time to develop CHF during long-term follow-up, resulting in a higher cardiac events rate. LV remodeling plays an important role in CHF onset and is strongly associated with adverse cardiac events. There is also evidence that the no-reflow phenomenon may have an adverse effect on LV remodeling after MI.22 Iwakura et al23 reported that the size of the risk area (culprit lesion in the proximal left anterior descending artery) and the severity of myocardial damage (peak serum CK) were determinants of the no-reflow phenomenon.
The study population was relatively small, and some predictors of long-term outcome may have been missed because of a lack of statistical power. Doppler signals were difficult to analyze in 5 patients (7.3%), a rate in keeping with previous studies of Doppler recordings in the acute phase of MI.11,18 We did not measure the cardiac function of long-term survivors. The impact of stenting on coronary microvascular function and the long-term risk of clinical events are not discussed because the number of stented patients was too small. Finally, this study was conducted before the era of systematic coronary stenting, use of platelet glycoprotein IIb/IIIa receptor inhibitors, and pharmacological treatment of no reflow.
No reflow after coronary reperfusion is associated with poor in-hospital and long-term clinical outcome. These high-risk patients should therefore be identified because they may benefit from adjunctive intracoronary therapies such as adenosine, verapamil, or nitroprusside.24
No reflow after PTCA, identified by Doppler guidewire measurement showing a rapid diastolic deceleration time, predicts adverse long-term outcomes after successful reopening of the IRA after AMI. Intracoronary Doppler guidewire–based demonstration of no reflow is therefore important for early clinical decision making to maintain the patency of both the epicardial coronary artery and the coronary microvasculature.
This work was supported in part by a grant from Fédération Française de Cardiologie.
↵*Drs Nguyen and Boulet contributed equally to this study.
Ito H, Tomooka T, Sakai N, et al. Lack of myocardial perfusion immediately after successful thrombolysis: a predictor of poor recovery of left ventricular function in anterior myocardial infarction. Circulation. 1992; 85: 1699–1605.
Piana RN, Paik GY, Moscucci M, et al. Incidence and treatment of “no-reflow” after percutaneous coronary intervention. Circulation. 1994; 89: 2514–2518.
Ito H, Okamura A, Iwakura K, et al. Myocardial perfusion patterns related to thrombolysis in myocardial infarction perfusion grades after coronary angioplasty in patients with acute anterior wall myocardial infarction. Circulation. 1996; 93: 1993–1999.
Iwakura K, Ito H, Takiuchi S, et al. Alternation in the coronary blood flow velocity pattern in patients with no reflow and reperfused acute myocardial infarction. Circulation. 1996; 94: 1269–1275.
Kawamoto T, Yoshida K, Akasaka T, et al. Can coronary blood flow velocity pattern after primary percutaneous transluminal coronary angiography predict recovery of regional left ventricular function in patients with acute myocardial infarction? Circulation. 1999; 100: 339–345.
Yamamuro A, Akasaka T, Tamita K, et al. Coronary flow velocity pattern immediately after percutaneous coronary intervention as a predictor of complications and in-hospital survival after acute myocardial infarction. Circulation. 2002; 106: 3051–3056.
Akasaka T, Yoshida K, Kawamoto T, et al. Relation of phasic coronary flow velocity characteristics with TIMI perfusion grade and myocardial recovery after primary percutaneous transluminal coronary angioplasty and rescue stenting. Circulation. 2000; 101: 2361–2367.
Kern MJ, Moore JA, Aguirre FV, et al. Determination of angiographic (TIMI grade) blood flow by intracoronary Doppler flow velocity during acute myocardial infarction. Circulation. 1996; 94: 1545–1552.
Rezkalla SH, Kloner RA. No-reflow phenomenon. Circulation. 2002; 105: 656–662.
Tanaka A, Kawarabayashi T, Nishibori Y, et al. No-reflow phenomenon and lesion morphology in patients with acute myocardial infarction; Circulation. 2002; 105: 2148–2152.
Feldman LJ, Coste P, Furber A, et al. Incomplete resolution of ST-segment elevation is a marker of transient microcirculatory dysfunction after stenting for acute myocardial infarction. Circulation. 2003; 107: 2684–2689.
Ito H, Maruyama A, Iwakura K, et al. Clinical implications of the no reflow phenomenon: a predictor of complications and left ventricular remodeling in reperfused anterior wall myocardial infarction. Circulation. 1996; 93: 223–228.
Gerber BL, Rochitte CE, Melin JA, et al. Microvascular obstruction and left ventricular remodeling early after acute myocardial infarction. Circulation. 2000; 101: 2734–2741.