Author + information
- Received October 11, 2010
- Revision received February 15, 2011
- Accepted February 24, 2011
- Published online September 1, 2011.
- Georg Fuernau, MD⁎,⁎ (, )
- Ingo Eitel, MD⁎,
- Vinzenz Franke, BSc⁎,
- Lysann Hildebrandt, BSc⁎,
- Josefine Meissner, BSc⁎,
- Suzanne de Waha, MD⁎,
- Philipp Lurz, MD⁎,
- Matthias Gutberlet, MD†,
- Steffen Desch, MD⁎,
- Gerhard Schuler, MD⁎ and
- Holger Thiele, MD⁎
- ↵⁎Reprint requests and correspondence:
Dr. Georg Fuernau, University of Leipzig—Heart Center, Department of Internal Medicine—Cardiology, Strümpellstraße 39, 04289 Leipzig, Germany
Objectives The objective of this study was to assess the area at risk (AAR) in ST-segment elevation myocardial infarction with 2 different cardiac magnetic resonance (CMR) imaging methods and to compare them with the validated angiographic Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease Score (APPROACH-score) in a large consecutive patient cohort.
Background Edema imaging with T2-weighted CMR and the endocardial surface area (ESA) assessed by late gadolinium enhancement have been introduced as relatively new methods for AAR assessment in ST-segment elevation myocardial infarction. However, data on the utility and validation of these techniques are limited.
Methods A total of 197 patients undergoing primary percutaneous coronary intervention in acute ST-segment elevation myocardial infarction were included. AAR (assessed with T2-weighted edema imaging and the ESA method), infarct size, and myocardial salvage (AAR minus infarct size) were determined by CMR 2 to 4 days after primary angioplasty. Angiographic AAR scoring was performed by use of the APPROACH-score. All measurements were done offline by blinded observers.
Results The AAR assessed by T2-weighted imaging showed good correlation with the angiographic AAR (r = 0.87; p < 0.001), whereas the ESA showed only a moderate correlation either to T2-weighted imaging (r = 0.56; p < 0.001) or the APPROACH-score (r = 0.44; p < 0.001). Mean AAR by ESA (20.0 ± 11.7% of left ventricular mass) was significantly (p < 0.001) smaller than the AAR assessed by T2-weighted imaging (35.6 ± 10.9% of left ventricular mass) or the APPROACH-score (27.9 ± 10.5% of left ventricular mass) and showed a significant negative dependence on myocardial salvage index. In contrast, no dependence of T2-weighted edema imaging or the APPROACH-score on myocardial salvage index was seen.
Conclusions The AAR can be reliably assessed by T2-weighted CMR, whereas assessment of the AAR by ESA seems to be dependent on the degree of myocardial salvage, thereby underestimating the AAR in patients with high myocardial salvage such as aborted infarction. Thus, assessment of the AAR with the ESA method cannot be recommended. (Myocardial Salvage and Contrast Dye Induced Nephropathy Reduction by N-Acetylcystein [LIPSIA-N-ACC]; NCT00463749)
Acute occlusion of a coronary artery leads to a process of myocyte necrosis, which spreads from the subendocardium of the perfusion bed to the subepicardial layers, the so-called wave front phenomenon (1). Without reperfusion or relevant collaterals, this process ends in complete necrosis of the cardiomyocytes within the perfusion bed. Early reopening of the infarct-related artery can interrupt this process and lead to tissue salvage within the area at risk (AAR), albeit reperfusion itself might induce myocyte damaging processes known as reperfusion injury (2). In general, the resulting necrotic zone in reperfused acute myocardial infarction (AMI) is significantly smaller than the AAR before reperfusion. With angiographic jeopardy scores such as the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease Score (APPROACH-score) the AAR can be calculated (3). However, for direct measurement of the myocardial AAR, few techniques have been developed. In recent studies (4,5), AAR has often been measured by single-photon emission computed tomography (SPECT), which has limitations because of availability and handling in AMI. Two newer technologies for AAR assessment in AMI using cardiac magnetic resonance (CMR) have been recently introduced (6–8). The most widely used method for assessment of the AAR is T2-weighted CMR. After initial validation in animals (6,9), the ability of this method to detect myocardial edema and, thereby, the AAR after ischemia has been demonstrated in humans (7,10,11). Furthermore, it has been demonstrated that edema can be seen as early as 30 min after ischemia onset (12), seems to be stable over 1 week (13), and can be assessed with good reproducibility between days 2 and 3 (14). Moreover, the amount of myocardial salvage as defined by this approach also predicts patient outcome (15).
Another method for estimating the AAR is the endocardial surface area (ESA) method, which is measured on contrast-enhanced magnetic resonance images. The ESA uses the hypotheses that necrosis first spreads to a final subendocardial extent and then only improves in transmurality according to the wave front phenomenon (8). The AAR is calculated by the relation of endocardial extent of late gadolinium enhancement (LGE) to the total endocardial circumference.
The ESA method has been applied in several trials (8,10,16). However, a comparison of T2-weighted imaging versus the ESA method in a large consecutive patient cohort with ST-segment elevation myocardial infarction (STEMI) is lacking (10,13,16,17).
Therefore, the aim of the current trial was to study and compare the performance of T2-weighted imaging versus ESA for assessment of the AAR and to validate these two CMR approaches against the angiographic APPROACH-score.
This prospective trial is a subanalysis of the LIPSIA-N-ACC (Myocardial Salvage and Contrast Dye Induced Nephropathy Reduction by N-Acetylcystein) trial, which compared high-dose N-acetylcysteine versus placebo for reperfusion injury prevention in STEMI patients and did not show a difference between the treatment groups. The detailed design and main results of the trial have been previously published (18). In brief, between November 2006 and February 2008, 251 consecutive patients were enrolled and underwent primary percutaneous coronary intervention for STEMI (Fig. 1). Patients were eligible, if symptoms lasted <12 h and if ST-segment elevation of ≥0.1 mV in ≥2 extremity leads or ≥0.2 mV in ≥2 precordial leads was present. The study had been approved by the local ethics committee and all patients gave written informed consent.
Primary percutaneous coronary intervention
Primary angioplasty was performed as described previously (18). In brief, the use of bare-metal or drug-eluting stents was left to the discretion of the interventional cardiologist. Additional use of thrombectomy was recommended depending on relevant thrombus. All patients were treated with aspirin, heparin, and clopidogrel. The use of glycoprotein IIb/IIIa inhibitors, angiotensin-converting enzyme inhibitors, beta-blockers, and statins was strongly recommended.
Angiographic AAR assessment was performed offline by 2 blinded observers using the modified APPROACH-score as described previously (8). This system is based on a score, which divides the left ventricle (LV) into regions according to pathological studies in humans evaluating the relative proportion of myocardium perfused by each coronary artery (19,20). Considering the location (proximal, mid, or distal) of the culprit lesion, vessel dominance, site of occlusion, and size of major branches of the infarcted artery, the score calculates the jeopardized myocardium for a given site of occlusion (Table 1,Fig. 2A) (3,8).
Cardiac magnetic resonance
CMR was performed using a 1.5-T scanner (Intera CV, Philips Medical Systems, Best, the Netherlands). The detailed scan protocol and sequence parameters have been described previously (18). In brief, visualization of myocardial edema was performed using a T2-weighted triple inversion recovery breath-hold pulse sequence (Figs. 2B and 3A) (6,7,21). Early and LGE images covering the whole ventricle were acquired 1 and 15 min after intravenous administration of gadolinium-chelate (Gadovist, Bayer Schering Pharma, Berlin, Germany) with an inversion recovery gradient echo sequence. Inversion times were individually adjusted to optimize nulling of apparently normal myocardium (typical values 200 to 300 ms).
All measurements were performed by fully blinded operators at the CMR core laboratory, which has proven excellent reproducibility for infarct size (22), myocardial salvage, and microvascular obstruction assessment (14). Infarct size and AAR by T2 edema were assessed by manual planimetry and expressed as percentage of LV mass (14,22,23). A central core of hypointense signal within the area of increased signal intensity, which is deemed to be intramyocardial hemorrhage (24), was included in the AAR assessment. Care was taken to exclude increased signal intensity from the blood pool adjacent to the endocardium due to slow flow. AAR by ESA was measured using ImageJ version 1.43u (National Institutes of Health, Bethesda, Maryland) as previously described (Fig. 3B) (8).
The following calculations were applied:
AAR assessed by T2-weighted CMR = volume edema/volume LV mass × 100
AAR assessed by ESA = summed endocardial infarct length/total LV endocardial length × 100
%infarct size = volume infarct/volume LV mass × 100
myocardial salvage = area at risk − infarct size
myocardial salvage index = (area at risk − infarct size)/area at risk × 100
Categorical data are presented as counts or proportions with the corresponding percentages. All quantitative data are expressed as mean ± SD or median and interquartile range (IQR) on the basis of whether they had a normal distribution or not. Bland-Altman analyses were applied to compare the 3 AAR measurement methods. Correlation analyses were done by Pearson or Spearman tests, as indicated. Given the paired nature of the data, a repeated-measures analysis of variance with post hoc analysis and the use of the Bonferroni correction to account for multiple testing was used to detect differences among angiographic AAR, ESA, and T2-weighted imaging. A trend analysis using analysis of variance was used to analyze a possible influence of myocardial salvage index (MSI) on AAR assessment with the 3 methods.
Statistical analysis was performed using commercially available software (SPSS 17.0, SPSS Inc., Chicago, Illinois). A 2-tailed p value <0.05 was considered statistically significant.
Of 251 eligible STEMI patients, this study included 197 patients. The main reasons for exclusion from the study are listed in Figure 1. Demographic and clinical characteristics are shown in Table 2. The culprit vessel was the left anterior descending coronary artery in 87 (44.2%) patients, the left circumflex coronary artery in 12 (6.1%), and the right coronary artery in 98 (49.7%) patients. Time from symptom onset to angioplasty ranged from 45 to 720 min (median: 231 [IQR: 153 to 290]). The median time between reperfusion and CMR was 3 days (IQR: 2 to 4).
Comparison of AAR assessed by T2-weighted imaging with LGE infarct size
All patients had a region of increased transmural signal intensity on T2-weighted images in the territory of the corresponding culprit artery (Figs. 2 and 3). The T2-weighted AAR (35.6 ± 10.9 %LV) was significantly larger than infarct size (18.2 ± 11.7 %LV; p < 0.001). In total, 11 patients had regional edema, but no LGE consistent with an aborted infarction (25) and 19 patients had evidence of prior infarction (with LGE in a different vascular territory).
The calculated myocardial salvage was 17.5 ± 11.6 %LV and the myocardial salvage index 49.3 ± 27.3.
Comparison of AAR assessed by T2-edema versus APPROACH-score
AAR assessed with T2-weighted imaging showed good correlation with the angiographic AAR (r = 0.87; p < 0.001). However, as shown by Bland-Altman analyses (Fig. 4A) there was a certain bias toward an overestimation of the AAR by T2-weighted CMR in comparison to angiographic scoring (35.6 ± 10.9 %LV vs. 27.9 ± 10.5 %LV, difference: 7.7 ± 5.4; p < 0.001) with limits of agreement of ±10.6 %LV.
Comparison of AAR assessed by T2-edema versus ESA
AAR by ESA showed only a moderate correlation with the AAR determined by T2-weighted imaging (r = 0.56; p < 0.001). Furthermore, a significant difference between these 2 approaches (35.7 ± 10.9 %LV vs. 20.0 ± 11.7 %LV, difference 15.7 ± 10.7; p < 0.001) with limits of agreement of 21.4 %LV was evident (Fig. 4B).
Comparison of AAR assessed by ESA versus APPROACH-score
There was only moderate correlation between ESA and angiographic scoring (r = 0.44; p < 0.001). Consequently, the AAR was significantly different between these 2 approaches (20.0 ± 11.7 %LV vs. 27.9 ± 10.5 %LV, difference: 8.0 ± 11.8; p < 0.001) with higher values of the AAR using the APPROACH-score. The limits of agreement were 23.1 %LV in our Bland-Altman analysis (Fig. 4C).
Influence of MSI on AAR measurement
The extent of the AAR assessed by T2-weighted CMR and the APPROACH-score was independent from the amount of MSI (mean 49.3 ± 27.3). Thus, no significant correlation between T2-weighted imaging as well as the APPROACH-score with the MSI could be observed (r = −0.21, p = 0.77; r = 0.08, p = 0.26, respectively). In contrast, the ESA showed a significant negative correlation to the MSI (r = −0.61; p < 0.001). When categorized by quartiles of MSI, patients with a large MSI had a significantly smaller AAR as assessed by the ESA method, in comparison to T2-weighted imaging and the APPROACH-score (Fig. 5).
Correlation of infarct size on AAR measurement
Only a weak correlation was seen between infarct size and T2-weighted imaging (r = −0.48; p < 0.001) or APPROACH-score (r = −0.34; p < 0.001), whereas there was a strong correlation of ESA to infarct size (r = −0.81; p < 0.001). This was also observed when the infarct size was categorized in quartiles (Fig. 6).
This is the largest study to date to compare and validate the assessment of myocardium at risk with 2 different CMR methods in a large cohort of consecutive STEMI patients. We could show that AAR assessment with T2-weighted CMR correlates well with angiographic scoring by the APPROACH-score, whereas AAR by ESA seems to be dependent on MSI and may not reflect the true AAR.
AAR assessment in AMI
The reliable assessment of the AAR in relation to infarct size enables determination of myocardial salvage and, consequently, of efficacy of reperfusion therapy in patients with STEMI. This is important for the evaluation of new therapeutic approaches in both interventional and pharmacological strategies for improvement of reperfusion success and reduction of reperfusion injury.
In the current study, we were able to show that there is excellent correlation between the AAR measured by T2-weighted imaging and the angiographic APPROACH-score, which is an anatomically and prognostically validated measure of the extent of myocardial jeopardy. This is in line with a recently published study of 50 AMI patients, which also found that AAR estimated by T2-weighted CMR was a predictor of the APPROACH-score (17). In contrast, AAR determination by ESA in our study showed only a moderate correlation to both T2-weighted and the angiographic AAR. The main reason for these differences might be that the ESA method significantly underestimates the AAR in patients with major myocardial salvage. Ubachs et al. (16) could similarly demonstrate in a small study cohort that the ESA is not suitable for AAR assessment in patients with early reperfusion as well as aborted infarction.
The characteristic CMR pattern of an aborted infarction is the detection of myocardium at risk in the absence of LGE (25). In these patients, the ESA allows no reliable assessment of the AAR, as the ESA method would generate an AAR of 0%, whereas the true salvaged AAR is 100%. Eleven patients in the present study showed no signs of infarction by LGE but did, however, show evidence of ischemic myocardium by T2-weighted imaging. Consequently, LGE or ESA imaging does not allow for determination of AAR in aborted infarction, whereas T2-weighted imaging does.
A recent study by Wright et al. (10) observed a better correlation between ESA and T2-weighted imaging than our study did (r = 0.77). However, in our study, a significant portion of patients was treated within 2 h following symptom onset with consequently a high degree of myocardial salvage (28 patients), whereas in the study by Wright et al. (10) the shortest pain-to-balloon time was 120 min. Recent publications found the highest MSI in AMI reperfused the first 2 h after onset of symptoms (15,26).
We and others found only moderate correlation between angiographic risk scores and infarct ESA, in contrast to the results reported by Ortiz-Perez et al. (8). This might be due in part to different study populations, because Ortiz-Perez et al. (8) only studied patients with TIMI (Thrombolysis In Myocardial Infarction) flow grade 0 before intervention and patients with a pain-to-balloon times >60 min, indicating a lack of patients with high MSI.
Similar to a previous study (17), we found a systematic overestimation of T2-weighted AAR in relation to angiographic AAR by the APPROACH-score. This might be explained by a significant increase of myocardial thickness in the infarcted region, which has been shown in experimental (27,28), patient (29), and imaging studies (30), as compared to noninfarcted areas. The APPROACH-score (3), however, is based on older autopsy studies (19,20,31) and was created to obtain prognostic information for overall coronary heart disease and is not exclusively dedicated to AMI patients. Thus, the APPROACH-score does not account for such an ischemia-related increased wall thickness. Notably, early postmortem analysis with microspheres in an animal model (6) and comparison to SPECT in humans (13) showed no significant differences in comparison to the T2-weighted CMR–detected AAR.
Although it is possible to image the AAR with SPECT, determination of the AAR and myocardial salvage in routine clinical practice by SPECT is impractical and limited to specialized centers. Major limitations of SPECT include the need of the perfusion tracer injection before reperfusion, imaging with a gamma camera within a few hours, the need of 2 SPECT scans with radiation exposure. In contrast, CMR allows AAR delineation as well as estimation of the salvaged myocardium retrospectively in 1 scan after the acute event. Notably, the AAR by T2-weighted imaging does not change during the first week after infarction, enabling myocardial salvage assessment after stabilization of the AMI patient (13). Additionally, CMR provides important information about ventricular anatomy, function, and microvascular status (32).
Some patients had to be excluded because of missing angiographic scoring for side branch lesions or codominant coronary anatomy. In addition, we did not assess several clinical variables such as the collateral blood flow or preconditioning, which might affect myocardial salvage. Several limitations of T2-weighted edema imaging have to be mentioned such as signal intensity variability caused by phased-array coils (although we used a body coil in the data acquisition of the present analysis), high signal from slow moving ventricular chamber blood that can mimic and mask elevated T2 in subendocardial myocardium, motion artifacts, and also the partly subjective nature of T2-weighted image interpretation. Technical developments such as T2-mapping may provide an even more robust and qualitative approach to edema imaging (33). In our study, we also performed quantitative analysis by manual drawing of the edema border contours, which might be inferior to a semiautomatic analysis. However, we have previously shown an excellent intra- and interobserver variability as well as reproducibility of T2-weigted edema imaging and LGE measurements with manual planimetry (14,22), minimizing an influence on the data.
In conclusion, CMR can provide information on myocardium at risk and salvaged myocardium using T2-weighted edema imaging, with a good correlation to the angiographic APPROACH-score. AAR determination by ESA seems to depend on myocardial salvage and underestimates the AAR, especially in patients with high myocardial salvage and, therefore, cannot be recommended for AAR assessment.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- area at risk
- acute myocardial infarction
- cardiac magnetic resonance
- endocardial surface area
- interquartile range
- late gadolinium enhancement
- left ventricle
- myocardial salvage index
- single-photon emission computed tomography
- ST-segment elevation myocardial infarction
- Received October 11, 2010.
- Revision received February 15, 2011.
- Accepted February 24, 2011.
- American College of Cardiology Foundation
- Jang I.K.,
- Weissman N.J.,
- Picard M.H.,
- et al.,
- EVOLVE Investigators
- Ndrepepa G.,
- Mehilli J.,
- Schwaiger M.,
- et al.
- Aletras A.H.,
- Tilak G.S.,
- Natanzon A.,
- et al.
- Friedrich M.G.,
- Abdel-Aty H.,
- Taylor A.,
- Schulz-Menger J.,
- Messroghli D.,
- Dietz R.
- Ortiz-Perez J.T.,
- Meyers S.N.,
- Lee D.C.,
- et al.
- Wright J.,
- Adriaenssens T.,
- Dymarkowski S.,
- Desmet W.,
- Bogaert J.
- Abdel-Aty H.,
- Cocker M.,
- Meek C.,
- Tyberg J.V.,
- Friedrich M.G.
- Carlsson M.,
- Ubachs J.F.,
- Hedstrom E.,
- Heiberg E.,
- Jovinge S.,
- Arheden H.
- Eitel I.,
- Desch S.,
- Fuernau G.,
- et al.
- Ubachs J.F.,
- Engblom H.,
- Erlinge D.,
- et al.
- Berry C.,
- Kellman P.,
- Mancini C.,
- et al.
- Thiele H.,
- Hildebrand L.,
- Schirdewahn C.,
- et al.
- Lee J.T.,
- Ideker R.E.,
- Reimer K.A.
- Thiele H.,
- Kappl M.J.,
- Conradi S.,
- Niebauer J.,
- Hambrecht R.,
- Schuler G.
- Thiele H.,
- Schindler K.,
- Friedenberger J.,
- et al.
- Ganame J.,
- Messalli G.,
- Dymarkowski S.,
- et al.
- Francone M.,
- Bucciarelli-Ducci C.,
- Carbone I.,
- et al.
- de Waha S.,
- Desch S.,
- Eitel I.,
- et al.