Author + information
- Received November 12, 2007
- Revision received December 30, 2007
- Accepted January 26, 2008
- Published online June 10, 2008.
- Alexander Hirsch, MD⁎,†,⁎ (, )
- Robin Nijveldt, MD†,‡,
- Joost D.E. Haeck, MD⁎,
- Aernout M. Beek, MD‡,
- Karel T. Koch, MD, PhD⁎,
- José P.S. Henriques, MD, PhD⁎,
- Rene J. van der Schaaf, MD⁎,
- Marije M. Vis, MD⁎,
- Jan Baan Jr, MD, PhD⁎,
- Robbert J. de Winter, MD, PhD⁎,
- Jan G.P. Tijssen, PhD⁎,
- Albert C. van Rossum, MD, PhD‡ and
- Jan J. Piek, MD, PhD⁎
- ↵⁎Reprint requests and correspondence:
Dr. Alexander Hirsch, Department of Cardiology, Academic Medical Center, Meibergdreef 9, P.O. Box 22660, 1100 DD, Amsterdam, the Netherlands.
Objectives We studied the relation between presence and severity of microvascular obstruction (MO), measured by cardiovascular magnetic resonance (CMR) and intracoronary Doppler flow measurements, for assessment of myocardial reperfusion in patients with acute anterior myocardial infarction (MI) treated by primary percutaneous coronary intervention (PCI).
Background Cardiovascular magnetic resonance has been used to detect and quantify MO in patients after acute MI but has never been compared with coronary blood flow velocity patterns.
Methods Twenty-seven patients with first anterior ST-segment elevation MI successfully treated with primary PCI were included. Coronary blood flow velocity was measured during recatheterization 4 to 8 days after primary PCI. These measurements were related to MO determined by late gadolinium-enhanced (LGE) CMR performed the day before recatheterization.
Results Early systolic retrograde flow was observed in 0 of 8 patients without MO on LGE CMR and in 10 (53%) of 19 patients with MO (p = 0.01). The extent of MO correlated with the diastolic-systolic velocity ratio (r = 0.44; p = 0.02), diastolic deceleration time (r = −0.61; p = 0.001), diastolic deceleration rate (r = 0.75; p < 0.0001), and coronary flow velocity reserve of the infarct-related artery (r = −0.44; p = 0.02). Furthermore, multivariate regression analyses, including extent of MO, infarct size, and transmural necrosis on LGE CMR, revealed that extent of MO was the only independent factor related to early systolic retrograde flow and diastolic deceleration rate.
Conclusions Assessment of microvascular injury by LGE CMR corresponds well to evaluation by intracoronary Doppler flow measurements. By means of CMR, quantification of myocardial function, infarct size, and microvascular injury can accurately be performed with a single noninvasive technique in patients with acute MI.
Survival and prognosis of patients with an acute myocardial infarction (MI) have improved substantially using therapies aiming at early and sustained reperfusion of the myocardium at risk (1). However, successful restoration of epicardial flow does not necessarily translate into optimal reperfusion at the myocardial tissue level, and may result in microvascular obstruction (MO), also known as the “no-reflow” phenomenon (2). Experimental and clinical studies have demonstrated that MO is a common phenomenon and is associated with increased infarct size, reduced myocardial function, left ventricular (LV) remodeling, and worse clinical outcome (3–11). As a consequence, MO has important prognostic significance in patients after acute MI.
Earlier studies have suggested that both measurements of coronary blood flow velocity patterns (12–15) and cardiovascular magnetic resonance (CMR) can be used to assess and quantify microvascular injury in patients after acute MI (3,5,7). Characteristics of coronary blood flow velocity patterns that are associated with no-reflow are the presence of early systolic retrograde flow (SRF), rapid deceleration of diastolic flow, and reduced coronary flow velocity reserve (CFVR) (12,15). Cardiovascular magnetic resonance allows direct visualization and quantification of microvascular injury with transmural extent using gadolinium-enhanced imaging, which corresponds to anatomically defined areas of no-reflow (3,5,16). To our knowledge, no data are available with respect to the relationship between both techniques in patients after acute MI. Therefore, the aim of the present study was to determine whether the presence and severity of microvascular injury determined by gadolinium-enhanced CMR was related to intracoronary Doppler flow measurements for assessment of myocardial reperfusion.
Patients and study protocol
This study comprised 27 consecutive patients presenting with a first acute anterior ST-segment elevation MI. All patients were treated by successful primary percutaneous coronary intervention (PCI), defined as Thrombolysis In Myocardial Infarction (TIMI) flow grade ≥2 and stent implantation with a residual stenosis of <50%. Inclusion criteria were PCI within 12 h after onset of symptoms, >10-fold increase in serum creatine kinase-MB levels, and wall motion abnormalities in ≥3 segments observed on resting echocardiogram. Patients with a previous MI, 3-vessel disease, cardiogenic shock, (relative) contraindications for CMR, or significant comorbidities were excluded from the study. Patients were treated with aspirin, heparin, abciximab, clopidogrel, statins, beta-blockade, and angiotensin-converting enzyme inhibitors, according to European Society of Cardiology practice guidelines (17).
All patients underwent CMR and recatheterization for intracoronary flow measurements at least 48 h but within 8 days after primary PCI. All CMR studies were supervised and analyzed by 1 operator who was blinded to intracoronary flow measurements, and vice versa.
The Institutional Review Boards of the Academic Medical Center and VU University Medical Center approved the protocol for study measurements. The study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from each patient.
The CMR examination was performed on a 1.5-T clinical scanner (Sonata, Siemens, Erlangen, Germany) using electrocardiographic gating and a phased array cardiac receiver coil. Cine images were acquired to measure LV volumes. A steady-state free precession pulse sequence during repeated breath-holds of approximately 10 seconds, in multiple short axis views every 10 mm covering the entire left ventricle. Typical in plane resolution was 1.6 × 1.9 × 6.0 mm3 (repetition time/echo time = 3.2/1.6 ms, flip angle 60°, matrix 256 × 156, temporal resolution 35 to 50 ms).
The late gadolinium-enhanced (LGE) images were acquired to determine infarct size and MO size and extent. A 2-dimensional segmented inversion recovery gradient-echo pulse sequence was used 12 to 15 min after administration of a gadolinium-based contrast agent (0.2 mmol/kg Dotarem, Guerbet, Roissy, France), with slice positions identical to the cine images. Typical in-plane resolution was 1.4 × 1.7 × 6.0 mm3 (repetition time/echo time = 9.6/4.4 ms, flip angle 25°, triggering to every other heart beat). The inversion time was set to null the signal of viable myocardium and ranged from 240 to 300 ms.
CMR data analysis and definitions
All CMR data were analyzed on a separate workstation using dedicated software (Mass version 2006 beta, Medis, Leiden, the Netherlands). On all short-axis cine slices, the endocardial and epicardial borders were outlined manually on end-diastolic and -systolic images to measure LV volumes and calculate ejection fraction. Segmental LV function was determined by dividing each short-axis slice into 12 equiangular segments, starting at the posterior septal insertion of the right ventricle. Systolic wall thickening was calculated as the difference between end-diastolic and -systolic wall thickness divided by end-diastolic wall thickness.
Infarct size and regions of MO were determined on LGE images as previously described, using a standardized and pre-defined definition of hyperenhancement (11,18). In short, total infarct size was calculated by automatic summation of all slice volumes of hyperenhancement (signal intensity >5 SD above the mean signal intensity of remote myocardium) and expressed as a percentage of LV mass. The extent of transmural necrosis was calculated in each patient and expressed as the sum of segments with >75% transmural hyperenhancement as a percentage of the total number of segments scored. Microvascular obstruction was defined as hypoenhanced regions within the hyperenhanced infarcted area and was included in the calculation of total infarct size. The total size of MO was calculated by summation of all slice volumes of hypoenhancement and expressed as a percentage of total LV mass. The extent of MO was defined as the number of segments in which MO was detected (irrespective of the absolute size of MO in each segment) and expressed as a percentage of the total number of segments scored.
For analysis of segmental myocardial function, segmental extent of transmural necrosis, and extent of MO, the 2 most basal and 2 most distal slices were excluded, because segmental evaluation at these levels is not reliable, owing to the LV outflow tract and partial volume effect, respectively. It was ensured that the excluded basal slices for segmental analysis were not affected by infarct or MO.
Recatheterization and Doppler flow measurements
Cardiac catheterization was performed following routine procedures. A bolus of 0.1-mg nitroglycerin was administered before flow measurements. Intracoronary flow was measured with a 0.014-inch Doppler-tipped guidewire (FloWire, Volcano Corporation, Rancho Cordova, California) that was positioned distal to the previously implanted stent. After optimization of the Doppler signal, velocity recordings were obtained at rest and after induction of maximal hyperemia with an intracoronary bolus of 20 to 40 μg adenosine. Doppler measurements were repeated at least 3 times and recorded continuously on videotape (FloMap, Jomed, Rancho Cordova, California). Coronary flow was also measured in an angiographically normal reference artery (the left circumflex artery [n = 25] unless there was >30% diameter stenosis, in which case the right coronary artery was used [n = 2]).
Analysis of Doppler flow measurements
Doppler flow velocity spectra were analyzed offline to determine the following parameters: baseline average and maximum peak flow velocity, diastolic and systolic average peak flow velocity, diastolic–systolic flow velocity ratio, and diastolic deceleration time. The rate of decline in flow velocity in diastole was calculated as the diastolic deceleration rate (DDR). The SRF was defined as retrograde peak velocity ≥10 cm/s and duration of >60 ms, as previously described (12). Absolute CFVR was calculated as the ratio of hyperemic to baseline average peak flow velocity and the relative CFVR as the ratio of the absolute CFVR in the infarct-related vessel to the absolute CFVR in the reference artery.
Values are reported as mean ± SD or median (25th to 75th percentile) for continuous variables and as frequency with percentage for categorical variables. Dichotomous variables were compared with the Fisher exact test and continuous variables with the Mann-Whitney U test. Correlations between extent and size of MO and Doppler flow parameters were calculated with simple linear regression analysis. To identify parameters independently associated with the extent and size of MO determined by CMR, multivariable linear regression analysis with a forward selection procedure was used. Variables were entered if p < 0.10. Multivariate logistic regression analysis was used to assess the relation between CMR parameters (extent or size of MO, infarct size, and transmural necrosis) and the presence of SRF. Similar analysis was performed using multivariate linear regression for the relationship with DDR. All statistical tests were 2-tailed, and a p value of <0.05 was considered to be statistically significant. All calculations were generated by Statistical Package for Social Sciences software (SPSS 12.0 for Windows, SPSS, Chicago, Illinois).
Of the 27 patients included in the study (mean age 53 ± 11 years), 23 were men (85%), 11 (41%) had hypercholesterolemia, 2 (7%) had diabetes mellitus, 7 (26%) had a history of hypertension, and 13 (48%) were smokers. Mean time from onset of symptoms to reperfusion was 203 ± 123 min. After primary PCI, TIMI flow grade 2 was observed in 5 patients (19%) and TIMI flow grade 3 in the other 22 patients (81%).
Patients underwent CMR and intracoronary flow measurements within 8 days after acute MI (4 ± 1 days and 5 ± 1 days, respectively). No CMR images or flow measurements were excluded from analysis because of insufficient quality. There was no clinical, electrocardiographic, or enzymatic evidence of reinfarction in any patient between primary PCI, CMR, and recatheterization. Before flow measurements, TIMI flow grade 3 was observed in all patients.
Early systolic retrograde flow was detected in 10 of 27 patients (37%). No differences in clinical and angiographic findings were observed between patients with or without SRF (Table 1). Microvascular obstruction was present on the LGE images in 9 patients (53%) without SRF and in all patients with SRF (p = 0.01). Quantification of MO showed that both extent and size of MO were significantly higher in patients with SRF compared with patients without SRF (p = 0.001 and p = 0.003, respectively). Infarct size was slightly higher in the SRF group, though not statistically significantly (25.9 ± 6.1% vs. 21.4 ± 5.6%; p = 0.07), and SRF was associated with more extensive transmural necrosis (p = 0.02). No differences were found in end-diastolic and -systolic volumes, LV ejection fraction, and systolic wall thickening in the infarct area. Figure 1 shows examples of coronary blood flow velocity patterns and corresponding short-axis LGE images of patients with and without microvascular injury.
Eight patients showed no presence of MO on the LGE images, and the remaining patients (n = 19) were divided into 2 subgroups according to the extent of MO: patients with mild MO (1% to 16% of the segments; n = 10) and severe MO (>16% of segments; n = 9). Although differences between patients with and without presence of MO were calculated, a subdivision in severity of MO was used to show possible trends of different variables in relation to the extent of MO. Subdivision was based on an arbitrary cutoff value.
Clinical, angiographic, and CMR data are presented in Table 2. Patients with MO had larger infarct size and more transmurally infarcted segments. Doppler flow velocity data showed that the presence of MO was associated with higher baseline maximum peak flow velocity, diastolic average peak flow velocity, and DDR; shorter diastolic deceleration time; and lower CFVR of the infarct-related artery (Table 3). Early systolic retrograde flow was observed in 0 of 8 patients without MO and in 10 (53%) of 19 patients with MO (p = 0.01).
The extent of MO correlated with the diastolic–systolic flow velocity ratio (r = 0.44; p = 0.02), diastolic deceleration time (r = −0.61; p = 0.001), DDR (r = 0.75; p < 0.0001) (Fig. 2), and CFVR of the infarct-related artery (r = −0.44; p = 0.02). Similar relationships were observed between the size of MO and diastolic–systolic flow velocity ratio (r = 0.46; p = 0.01), diastolic deceleration time (r = −0.53; p = 0.004), DDR (r = 0.56; p = 0.002), and CFVR (r = −0.36; p = 0.07). There was no relationship between relative CFVR and the extent or size of MO (r = −0.12; p = 0.54, and r = 0.03; p = 0.89, respectively). Stepwise multivariate linear regression analysis including Doppler flow parameters identified SRF (beta = 7.23; SE = 3.20; p = 0.03), DDR (beta = 0.08; SE = 0.02; p = 0.003), and CFVR (beta = −8.98; SE = 4.18; p = 0.04) as independent factors related to the extent of MO determined by CMR (adjusted R2 = 0.64). For the relationship with the size of MO, only SRF (beta = 2.29; SE = 0.66; p = 0.002) and CFVR (beta = −1.83; SE = 1.01; p = 0.08) remained as independent factors at multivariate stepwise regression analysis (adjusted R2 = 0.37).
MO, infarct size, and transmural necrosis in relation to SRF and DDR
The extent of MO by CMR was related to infarct size and extent of transmural necrosis (r = 0.60; p = 0.001, and r = 0.51; p = 0.007, respectively). The predictive values of the extent of MO, infarct size, and transmural necrosis with respect to SRF and DDR were compared relative to each other by multivariate analyses. Multivariate regression analyses showed that the extent of MO was the only independent factor related to both SRF and DDR (Tables 4 and 5).⇓ Both the odds ratio and the regression coefficient of the extent of MO did not substantially change when infarct size and transmural necrosis were added to the models. In fact, infarct size and extent of transmural necrosis were no longer associated with the flow parameters after adjusting for extent of MO. Similar results were found regarding the size of MO, but the relationship between MO size and both SRF and DDR was not as strong as its extent (data not shown).
We investigated whether the extent and size of microvascular injury measured by LGE CMR was related to intracoronary Doppler flow velocity measurements in 27 patients with acute MI treated by primary PCI. Previous Doppler flow studies have established that SRF, rapid deceleration of the diastolic flow velocity, and reduced CFVR are typical findings of the no-reflow phenomenon (12,13,19). In the present study, we showed that the extent and size of microvascular injury as assessed by LGE CMR correlated well with these parameters. Furthermore, after adjusting for MO (extent or size), infarct size and extent of transmural necrosis were no longer associated with SRF and DDR, and MO remained the only factor independently related to these flow parameters. To our knowledge, this is the first study that directly compared both techniques in humans.
Microvascular abnormalities are often present in myocardium exposed to prolonged ischemia, owing to a multitude of processes, including tissue edema, platelet plugging, neutrophil adhesion, myonecrosis, and intracapillary red blood cell stasis (20). This pathological process may be enhanced by reperfusion injury within the early stages after reperfusion therapy (21). The Doppler flow characteristics of “no reflow” have been established by direct comparison of flow patterns with perfusion deficits on myocardial contrast echocardiography. Iwakura et al. (12) were the first to describe the association between the presence of SRF, a short diastolic deceleration time, and microvascular dysfunction. They suggested that those findings may be explained by an increase of microvascular impedance and a decrease of intramyocardial blood pool volume. Microvascular obstruction with subsequent high impedance results in the inability to squeeze blood forward into the venous circulation during systole, and consequently blood will be pushed back into the epicardial coronary artery, reflected by SRF. Furthermore, the reduced intramyocardial blood pool, which fills rapidly during diastole, has been used to explain the rapid decline of diastolic velocity (12). The CFVR measures the functional status of the distal microvascular bed and depends on multiple factors, including myocardial resistance, metabolic demands, neurohormonal activation, filling pressures, and vascular resistance of small and large coronary arteries (15,22).
In the present study, SRF was observed in only about one-half of the patients with MO on LGE imaging. This percentage is similar to studies using myocardial contrast echocardiography (23). Early systolic retrograde flow will only appear in the epicardial coronary artery if there is a substantial amount of blood that is pushed back and reflects more severe microvascular injury (14). Cardiac magnetic resonance is probably more sensitive in detecting MO. To our knowledge, no data exist on which parameters have greater clinical relevance for identifying patients with worse functional and clinical outcome.
The technique to visualize regions of microvascular damage with CMR has been validated in several experimental studies and corresponds to anatomically defined areas of “no reflow” (3,5,16). Regions of MO in LGE CMR appear as hypoenhanced regions within the hyperenhanced infarcted area. In the area with microvascular damage, there is delayed contrast penetration owing to several factors, such as reduced functional capillary density, capillary compression and obstruction, and hemorrhage (2,24). Therefore, the signal intensity of these regions is not enhanced by the contrast agent and appears as dark zones initially, but slowly becomes hyperenhanced over time as well, due to wash-in by diffusion of the contrast agent from surrounding regions with intact microcirculation (16,25). Thus, timing of LGE image acquisition influences the incidence and extent of MO as detected by CMR. We determined MO on LGE images acquired 12 to 15 min after contrast injection. The optimal time point for determining MO in relation to its predictive value on LV remodeling and functional recovery needs further investigation.
The present results suggest that the relationship between intracoronary flow parameters and the extent of MO (% segments affected by microvascular damage) might be slightly closer than those parameters and the size of MO (% myocardium with microvascular damage). It is conceivable that the coronary blood flow velocity measured in an epicardial vessel is influenced more by a large proportion of the area at risk affected than by a small area with extensive MO. However, these results may also be influenced by different speeds of contrast diffusion in areas with microvascular injury between patients and by differences in timing of image acquisition after contrast administration.
In line with the study of Bogaert et al. (26), we found a close relationship between MO, infarct size, and infarct transmurality. After adjustment for MO, infarct size and transmural necrosis were no longer associated with SRF and DDR. These findings suggest that the changes in coronary blood flow velocity patterns are specific for areas of microvascular damage within the infarct and do not merely reflect larger and more transmural infarctions associated with MO.
In this study, no relationship was observed between the time from symptom onset to PCI and the presence of MO. Data from experimental studies suggest that the extent of microvascular injury worsened with longer durations of ischemia (20). However, other clinical studies also did not find this relationship in patients treated with primary PCI (4,26,27). Iwakura et al. (4) showed that the development of the no-reflow phenomenon was related to severity of myocardial damage, the size of the area at risk, and the occlusion status of infarct-related artery. Probably other factors than time from symptom onset to reperfusion are predominant in the clinical setting.
We performed CMR and recatheterization >48 h but within 8 days after primary PCI, because microvascular function and coronary flow may change, particularly in the first 48 h after revascularization (5,7). Experimental studies have shown that infarct size and the area of MO increase in the first 48 h after reperfusion and stabilize between 2 and 9 days (3,5). On the other hand, Doppler flow studies with serial measurements of coronary flow observed an increase in CFVR, normalization of diastolic deceleration rate and time, and disappearance of SRF within 24 to 48 h after reperfusion in some of the patients, indicating recovery of myocardial microcirculation (5,19,23,28). Because of these early serial changes, assessment of the microvascular integrity between 2 days and 1 week after reperfusion might have better predictive accuracy for LV recovery and remodeling than measurements immediately after reperfusion (19,28).
Microvascular injury is associated with greater infarct size and worse myocardial function (5,11,26). However, we did not find a relationship between the extent of MO and LV function in the present study. This is probably explained by the fact that we selected only patients with large MIs. Owing to this selection, the incidence of MO was high (70%) compared with earlier studies. Thus, it was possible to compare the presence or absence of microvascular injury and to correlate the severity of MO between both techniques.
An increasing number of studies are published relating microvascular injury to LV remodeling and clinical outcome in patients after reperfused acute MI (2). We have shown for the first time in vivo that functional measurements of microvascular injury by Doppler flow velocity correspond to anatomic measurements of microvascular injury by LGE CMR. Assessment of the microvascular integrity with LGE CMR has several advantages. Cardiovascular magnetic resonance allows a complete and accurate assessment of LV status in patients after acute MI in a single noninvasive examination without the use of ionizing radiation. Additionally, information can be obtained on cardiac structure, global and regional myocardial function, and infarct size with a high spatial resolution.
Evaluation of MO is useful for identifying high-risk patients with an acute MI after successful PCI and may facilitate decision making regarding the necessity of additional interventions, such as cell therapy. Theoretically, improvement of microvascular perfusion may have beneficial effects on infarct healing, ventricular remodeling, and collateral formation, which may ultimately lead to better outcome. Finally, besides selecting patients for adjunctive therapy, measurements of MO by CMR can be very useful in evaluating the efficacy of novel treatment strategies aiming at preserving the microvasculature when treating acute MI in both clinical and experimental studies.
Several potential limitations of this study need to be addressed. First, our results were derived from a selected population of patients with anterior MI. Previously, it has been reported that blood flow velocity patterns in patients with acute MI differ between the right and left coronary arteries (12) and, therefore, our findings may not be representative for all patients with acute MI. The number of patients included was limited, and we did not perform serial measurements. We cannot conclude whether quantification of MO by LGE imaging is accurate enough to detect changes over time and whether these changes correspond to changes in blood flow velocity parameters. Nevertheless, several studies have shown that “no reflow” detected by either of the techniques is associated with poor LV remodeling and outcome (6–10,13–15,19,22,26–28). Microvascular obstruction was assessed 12 to 15 min after contrast administration rather than during the first 3 min as in previous studies (3,5). Because contrast diffuses into the infarct area and MO regions over time, the true extent and especially size of MO may have been underestimated. This might have influenced the observed relationship of CMR MO and intracoronary Doppler flow velocity measurements.
In conclusion, the extent and size of MO as assessed by LGE CMR correlated well with coronary blood flow velocity characteristics of microvascular injury, such as the presence of SRF, rapid deceleration of diastolic flow, and a reduced CFVR. This relation was independent of the size and transmural extent of the infarcted myocardium. By using CMR, quantification of myocardial function, infarct size, and microvascular injury can be performed accurately with a single noninvasive technique in patients with acute MI treated by primary PCI.
The authors thank the technicians, research nurses, and staff of the cardiology and radiology departments of the Academic Medical Center and VU University Medical Center for their skilled assistance.
Supported by the Netherlands Heart Foundation, grant no. 2003B126. The first two authors contributed equally to this work.
- Abbreviations and Acronyms
- coronary flow velocity reserve
- cardiovascular magnetic resonance
- diastolic deceleration ratio
- late gadolinium-enhanced
- left ventricular
- myocardial infarction
- microvascular obstruction
- percutaneous coronary intervention
- early systolic retrograde flow
- Thrombolysis in Myocardial Infarction
- Received November 12, 2007.
- Revision received December 30, 2007.
- Accepted January 26, 2008.
- American College of Cardiology Foundation
- Rochitte C.E.,
- Lima J.A.,
- Bluemke D.A.,
- et al.
- Iwakura K.,
- Ito H.,
- Kawano S.,
- et al.
- Wu K.C.,
- Kim R.J.,
- Bluemke D.A.,
- et al.
- Wu K.C.,
- Zerhouni E.A.,
- Judd R.M.,
- et al.
- Gerber B.L.,
- Rochitte C.E.,
- Melin J.A.,
- et al.
- Furber A.P.,
- Prunier F.,
- Nguyen H.C.,
- Boulet S.,
- Delepine S.,
- Geslin P.
- Kawamoto T.,
- Yoshida K.,
- Akasaka T.,
- et al.
- Hombach V.,
- Grebe O.,
- Merkle N.,
- et al.
- Iwakura K.,
- Ito H.,
- Takiuchi S.,
- et al.
- Lepper W.,
- Hoffmann R.,
- Kamp O.,
- et al.
- Montisci R.,
- Chen L.,
- Ruscazio M.,
- et al.
- Judd R.M.,
- Lugo-Olivieri C.H.,
- Arai M.,
- et al.
- Van de Werf F.,
- Ardissino D.,
- Betriu A.,
- et al.,
- Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology
- Nohtomi Y.,
- Takeuchi M.,
- Nagasawa K.,
- et al.
- Bax M.,
- de Winter R.J.,
- Schotborgh C.E.,
- et al.
- Lepper W.,
- Sieswerda G.T.,
- Franke A.,
- et al.
- Kim R.J.,
- Chen E.L.,
- Lima J.A.,
- Judd R.M.
- Lima J.A.,
- Judd R.M.,
- Bazille A.,
- Schulman S.P.,
- Atalar E.,
- Zerhouni E.A.
- Yamamuro A.,
- Akasaka T.,
- Tamita K.,
- et al.