European Journal of Radiology 83 (2014) 778–782
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Delayed-enhancement magnetic resonance imaging at 3.0 T using 0.15 mmol/kg of contrast agent for the assessment of chronic myocardial infarction Jun Yang a,∗,1 , Heng Ma a,1 , Jing Liu a,1 , Chunxiao Wang a , Yinghong Shi a , Haizhu Xie a , Futao Huo a , Fengli Liu a , Kai Lin b,∗∗ a b
Yuhuangding Hospital, Yantai, Shandong Province, China Department of Radiology, Northwestern University, Chicago, IL, USA
a r t i c l e
i n f o
Article history: Received 12 October 2013 Received in revised form 27 November 2013 Accepted 10 January 2014 Keywords: Imaging Magnetic resonance imaging Myocardial infarction
a b s t r a c t Objective: A recent international, multicenter, double-blinded, randomized trial shows delayed-enhanced magnetic resonance imaging (DE-MRI) using contrast doses of ≥0.2 mmol/kg is effective in the detection and assessment of myocardial infarction (MI), and 0.1 mmol/kg is not enough; intermediate doses between 0.1 and 0.2 mmol/kg have not been tested. The aim of this study was to prospectively test the performance of DE-MRI using 0.15 mmol/kg of contrast agent for the detection of MI. Materials and methods: A total of 31 consecutive patients with chronic MI underwent DE-MRI at 3.0 T using both 0.15 mmol/kg and 0.2 mmol/kg of contrast agent in random order and on separate days. Infarction segment and infarction size were compared on DE-MRI images using a 17-segment model. Bland–Altman analysis was used to analyze correlation and agreement between global infarct sizes. Results: DE-MRI showed enhanced myocardium in all the 31 patients with chronic MI. There was no significant difference between the 0.15 mmol/kg and 0.2 mmol/kg images in all 31 patients based on the infarction segment (7.87 ± 2.72 vs. 7.81 ± 2.64, respectively; p = 0.33). There was no significant difference between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition (16.3 ± 7.8% vs. 16.4 ± 7.9%, respectively; p = 0.87). A strong correlation between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition was indicated through Bland–Altman analysis. Conclusion: DE-MRI at 3.0 T using 0.15 mmol/kg of contrast agent is effective for the assessment of MI. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In patients with coronary artery disease (CAD) and left ventricular dysfunction, the distinction between viable and nonviable myocardium with delayed-enhanced magnetic resonance imaging (DE-MRI) allows the prediction of functional recovery after surgical or interventional revascularization [1]. Guidelines on myocardial
∗ Corresponding author at: Yuhuangding Hospital, No. 20, Yuhuangding East Road, Yantai, Shandong Province 264000, China. Tel.: +86 535 6691999; fax: +86 535 6240341. ∗∗ Corresponding author at: Department of Radiology, Northwestern University, Suite 1600, 737 N. Michigan Avenue, Chicago, IL 60611, USA. Tel.: +1 312 363 8469; fax: +1 312 926 5991. E-mail addresses: [email protected] (J. Yang), [email protected] (H. Ma), [email protected] (J. Liu), [email protected] (C. Wang), [email protected] (Y. Shi), [email protected] (H. Xie), [email protected] (F. Huo), [email protected] (F. Liu), [email protected] (K. Lin). 1 These authors have contributed equally to this work. 0720-048X/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2014.01.012
infarction (MI) diagnostics and viability assessment published by the European Society of Cardiology (ESC) recognize cardiac MR as a Class-I method [2]. Infarcted myocardium exhibits delayed hyperenhancement after administration of gadolinium-based contrast material and can be imaged by using inversion-recovery MR techniques. The mechanism of DE-MRI is generally owing to an increased signal intensity of infarcted nonviable myocardium compared with that of healthy myocardium. These effects result from the delayed washout kinetics of contrast material in infarcted myocardium due to leakage into the interstitial space caused by microvascular damage. The alternative methods in the assessment of MI have significant limitations. The biomarkers such as troponin I and T and creatine kinase-MB (CK-MB) are insensitive for old MI, do not provide information on infarct location. The ECG is insensitive for chronic non-Q-wave infarcts. Single-photon emission computed tomography (SPECT) is useful tool for the diagnosis of MI, but has relatively poor spatial resolution. DE-MRI has superior spatial resolution, and it can be used to differentiate transmural and nontransmural MI.
J. Yang et al. / European Journal of Radiology 83 (2014) 778–782
Studies in animal experiments show a close correlation with the histopathologically determined area of MI [3,4]. Studies in humans show that DE-MRI is effective in determining the presence, location, and extent of acute and chronic MI [5–7]. Most of previous DE-MRI studies, however, are limited in the optimal dose of contrast for the diagnosis of MI. A recent international, multicenter, double-blinded, randomized trial shows DE-MRI using contrast doses of ≥0.2 mmol/kg is effective in the detection and assessment of both acute and chronic MI, and 0.1 mmol/kg is not enough [8]. In their study, however, intermediate doses between 0.1 and 0.2 mmol/kg have not been tested. The aim of the present study was to prospectively evaluate whether 0.15 mmol/kg of contrast agent is effective for the assessment of MI.
779
Table 1 Characteristics of the study population. Characteristics
No. of patients
Age (years) Range Sex (male/female) Risk factors for CAD Hypertension Hypercholesterolemia Smoking Diabetes Family history CAD classification Single-vessel Multi-vessel
55 ± 19 41–69 18/13 12 (39%) 17 (55%) 13 (42%) 5 (16%) 16 (32%) 6 (19%) 25 (81%)
Values are n (%) or mean ± standard deviation. CAD, coronary artery disease as defined by coronary artery stenosis ≥50%.
2. Materials and methods 2.1. Patients The study protocol was approved by the institutional review board. Written informed consent was obtained from each patient. A total of 31 consecutive patients (18 men, 13 women; age range, 41–69 years; mean age, 55 years ± 19) with chronic MI were prospectively enrolled between January 2012 and May 2013. All patients included in this study were confirmed to be positive to have single-vessel or multi-vessel disease by invasive X-ray angiography, and all were identified as chronic MI on the basis of occurring more than 1 year since the initial presentation of acute MI. Patients with contraindications for MRI (metallic implants such as pacemakers, defibrillators, cerebral aneurysm clips, ocular metallic deposits, severe claustrophobia), non-sinus rhythm (atrial fibrillation/flutter), dyspnea, renal insufficiency (estimated glomerular filtration rate assessed by creatinine clearance < 60 ml/min/1.73 m2 ), or a history of hemolytic anemia or other hemoglobinopathy and those who were pregnant or breastfeeding also were excluded. 2.2. DE-MRI imaging All patients were imaged in the supine position with a 3.0 T clinical scanner (Signa EXCITE, GE Healthcare, Waukesha, WI, USA) with a 4-element phased-array surface coil. Each patient underwent two DE-MRI scans with different contrast doses. One was 0.15 mmol/kg and the other was 0.2 mmol/kg. The two scans were performed in random order and on separate days. After localization of the heart, 8–10 short-axis slices were obtained to cover the entire left ventricle (LV) from base to apex. Two- and 4-chamber long-axis views also were obtained. DE-MRI was performed at 10 min after contrast material administration (Magnevist, Schering, Berlin, Germany). An inversion recovery fast gradient-echo pulse sequence was used to assess myocardial viability. Imaging parameters were: TR 6 ms, TE 3 ms, 36 cm FOV, 8-mm slice thickness, matrix 192 × 160, TI 180–270 ms (adjusted to null the signal of normal myocardium), 1 NEX, flip angle 20◦ .
percentage of LV myocardium was calculated by summing the segments with delayed enhancement (each weighted by the midpoint of the range of enhancement for the given segmental score; i.e., 1 = 13%, 2 = 38%, 3 = 63%, 4 = 88%) and dividing by 17 [8].
2.4. Statistics A statistical software program, SPSS 13.0 (SPSS Inc., Chicago, IL, USA), was used for statistical analysis. Continuous variables were expressed as mean ± standard deviation. Categorical variables were expressed as absolute number (percentage). Continuous data were compared using t test, which was paired where appropriate. The Wilcoxon signed rank test was performed to evaluate the difference in global infarct size percentage between the two dosages. Bland–Altman analysis was used to analyze correlation and agreement between global infarct size. The inter-observer agreement was determined with the kappa test. Statistical tests were 2-tailed and a p-value < 0.05 was considered to be significant.
3. Results All acquisitions were successful and all images acquired were of sufficient quality for the final analysis. The characteristics of the study population are summarized in Table 1.
3.1. Infarction segment DE-MRI showed enhanced myocardium in all the 31 patients with chronic MI. A total of 527 segments were analyzed, of which 242 segments with scar were found in both 0.15 mmol/kg and 0.2 mmol/kg images. Only 2 segments with scar in 1 patient were found in the 0.15 mmol/kg images, which were not detected in the 0.2 mmol/kg images. There was no significant difference between the 0.15 mmol/kg and 0.2 mmol/kg images in all 31 patients based on the infarction segment (7.87 ± 2.72 vs. 7.81 ± 2.64, respectively; p = 0.33) (Fig. 1).
2.3. DE-MRI image analysis The images were independently assessed by 2 experienced readers who were unaware of patient identity, contrast dose, or any associated clinical information. Any disagreement in the diagnosis between the 2 readers was settled by a consensus reading. DE-MRIs were graded for the presence, location, and extent of enhanced myocardium using a 17-segment model and a 5-point scale for each segment (0 = no delayed enhancement, 1 = 1–25%, 2 = 26–50%, 3 = 51–75%, 4 = 76–100%) [8]. Global infarct size as a
3.2. Infarction size There was no significant difference between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition (16.3% ± 7.8% vs. 16.4% ± 7.9%, respectively; p = 0.87). A strong correlation between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition was indicated through Bland–Altman analysis (Fig. 2).
780
J. Yang et al. / European Journal of Radiology 83 (2014) 778–782
Fig. 1. A 55-year-old man with chronic myocardial infarction. DE-MRI images using 0.15 mmol/kg of contrast agent (A) appears virtually identical to those using 0.2 mmol/kg (B).
3.3. Inter-observer agreement Inter-observer agreement of the segments for presence or absence of scar was evaluated by using the kappa test. The kappa score of 0.2 mmol/kg DE-MRI was 0.89, and that of 0.15 mmol/kg DE-MRI was 0.91 (p < 0.05 for both). 4. Discussion In the present study, we have prospectively compared the diagnostic value of DE-MRI at 3.0 T using 0.15 mmol/kg and 0.2 mmol/kg of contrast agent for the assessment of chronic MI. Our results demonstrate that DE-MRI using 0.15 mmol/kg enables effective
assessment of infarction segment and infarction size with high agreement compared with DE-MRI using 0.2 mmol/kg. Regarding the assessment of MI using DE-MRI in CAD, this is to the best of our knowledge the first study to compare measurements of infarction segment and infarction size in 0.15 mmol/kg and 0.2 mmol/kg of contrast agent acquisitions in a cohort of patients with chronic MI. We found strong and significant correlations between 0.15 mmol/kg and 0.2 mmol/kg of contrast agent acquisitions. These results confirm and extend the findings of published single-center and multi-center studies evaluating the accuracy of DE-MRI for detecting infarction in patients with acute or chronic MI [5–9]. Unlike these previous studies that involved different groups with different doses of contrast, in the current study each
Fig. 2. Bland–Altman analysis indicates strong correlation in the infarction size between DE-MRI using 0.15 mmol/kg and that using 0.2 mmol/kg.
J. Yang et al. / European Journal of Radiology 83 (2014) 778–782
patient underwent two DE-MRI scans with different contrast doses. Therefore, the results from the current study should reflect the comparison value of DE-MRI more accurately. We believe these findings indicate that DE-MRI using 0.15 mmol/kg can effectively assess MI. The current consensus criteria for the diagnosis of MI, which are fundamentally based on cardiac enzymes, the 12-lead ECG, and echocardiography, have important limitations [10]. Although cardiac enzymes are highly sensitive for acute MI, they are insensitive for chronic MI, and do not provide information on infarction segment and infarction size [10–13]. The presence of Q waves in the ECG has been used as a marker of prior MI. However, the ECG is insensitive for chronic non-Q-wave infarcts [14–18]. Although echocardiography is subsidiary to the ECG in stable patients presenting with ST-segment elevation MI, its role may be more important in patients with a non-diagnostic ECG [19]. Echocardiography is generally very accurate, but the sensitivity to detect acute MI varies widely [20]. Using echocardiography, chronic MI was reported to remain undetected more frequently, because of the disappearance of segmental wall motion abnormalities after several weeks [19,5]. DE-MRI may overcome most of these limitations. Studies in animal models and humans show an excellent match of infarction segment by DE-MRI to histopathology [3–7]. Recent studies reported silent myocardial infarctions (MIs) are prevalent among diabetic patients and inflict significant morbidity and mortality. DE-MRI can detect unrecognized MI in diabetic patients [21,22]. Moreover, because of high spatial resolution (more than 40-fold greater than SPECT), DE-MRI can even detect small infarcts in the absence of Q waves [6,16,23,24]. These findings demonstrate that DE-MRI is the best available technique for the detection and assessment of MI. Previous multi-center trial shows DE-MRI using contrast doses of ≥0.2 mmol/kg is effective in the detection and assessment of both acute and chronic MI, and 0.1 mmol/kg is not enough [8]. Our results demonstrate that DE-MRI using 0.15 mmol/kg of contrast agent is enough for the assessment of MI. Lower doses of contrast mean a lower cost, a darker blood pool to detect subendocardial infarcts [25], and decreased side effects such as nephrogenic systemic fibrosis in those with severe renal impairment. Several important limitations exist in the present study. First, our study was a consecutive analysis of patients in a single center; thus, it would be important to validate these results in multiple centers. Second, the population was relatively small, and the statistical power was relatively low. Third, the performance of both contrast doses was studied at 3.0 T. Additional 1.5 T studies are therefore needed to confirm our findings. Fourth, we have not tested the performance of DE-MRI at 3.0 T using contrast agent doses of
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Delayed-enhancement magnetic resonance imaging at 3.0 T using 0.15 mmol/kg of contrast agent for the assessment of chronic myocardial infarction Jun Yang a,∗,1 , Heng Ma a,1 , Jing Liu a,1 , Chunxiao Wang a , Yinghong Shi a , Haizhu Xie a , Futao Huo a , Fengli Liu a , Kai Lin b,∗∗ a b
Yuhuangding Hospital, Yantai, Shandong Province, China Department of Radiology, Northwestern University, Chicago, IL, USA
a r t i c l e
i n f o
Article history: Received 12 October 2013 Received in revised form 27 November 2013 Accepted 10 January 2014 Keywords: Imaging Magnetic resonance imaging Myocardial infarction
a b s t r a c t Objective: A recent international, multicenter, double-blinded, randomized trial shows delayed-enhanced magnetic resonance imaging (DE-MRI) using contrast doses of ≥0.2 mmol/kg is effective in the detection and assessment of myocardial infarction (MI), and 0.1 mmol/kg is not enough; intermediate doses between 0.1 and 0.2 mmol/kg have not been tested. The aim of this study was to prospectively test the performance of DE-MRI using 0.15 mmol/kg of contrast agent for the detection of MI. Materials and methods: A total of 31 consecutive patients with chronic MI underwent DE-MRI at 3.0 T using both 0.15 mmol/kg and 0.2 mmol/kg of contrast agent in random order and on separate days. Infarction segment and infarction size were compared on DE-MRI images using a 17-segment model. Bland–Altman analysis was used to analyze correlation and agreement between global infarct sizes. Results: DE-MRI showed enhanced myocardium in all the 31 patients with chronic MI. There was no significant difference between the 0.15 mmol/kg and 0.2 mmol/kg images in all 31 patients based on the infarction segment (7.87 ± 2.72 vs. 7.81 ± 2.64, respectively; p = 0.33). There was no significant difference between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition (16.3 ± 7.8% vs. 16.4 ± 7.9%, respectively; p = 0.87). A strong correlation between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition was indicated through Bland–Altman analysis. Conclusion: DE-MRI at 3.0 T using 0.15 mmol/kg of contrast agent is effective for the assessment of MI. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In patients with coronary artery disease (CAD) and left ventricular dysfunction, the distinction between viable and nonviable myocardium with delayed-enhanced magnetic resonance imaging (DE-MRI) allows the prediction of functional recovery after surgical or interventional revascularization [1]. Guidelines on myocardial
∗ Corresponding author at: Yuhuangding Hospital, No. 20, Yuhuangding East Road, Yantai, Shandong Province 264000, China. Tel.: +86 535 6691999; fax: +86 535 6240341. ∗∗ Corresponding author at: Department of Radiology, Northwestern University, Suite 1600, 737 N. Michigan Avenue, Chicago, IL 60611, USA. Tel.: +1 312 363 8469; fax: +1 312 926 5991. E-mail addresses: [email protected] (J. Yang), [email protected] (H. Ma), [email protected] (J. Liu), [email protected] (C. Wang), [email protected] (Y. Shi), [email protected] (H. Xie), [email protected] (F. Huo), [email protected] (F. Liu), [email protected] (K. Lin). 1 These authors have contributed equally to this work. 0720-048X/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2014.01.012
infarction (MI) diagnostics and viability assessment published by the European Society of Cardiology (ESC) recognize cardiac MR as a Class-I method [2]. Infarcted myocardium exhibits delayed hyperenhancement after administration of gadolinium-based contrast material and can be imaged by using inversion-recovery MR techniques. The mechanism of DE-MRI is generally owing to an increased signal intensity of infarcted nonviable myocardium compared with that of healthy myocardium. These effects result from the delayed washout kinetics of contrast material in infarcted myocardium due to leakage into the interstitial space caused by microvascular damage. The alternative methods in the assessment of MI have significant limitations. The biomarkers such as troponin I and T and creatine kinase-MB (CK-MB) are insensitive for old MI, do not provide information on infarct location. The ECG is insensitive for chronic non-Q-wave infarcts. Single-photon emission computed tomography (SPECT) is useful tool for the diagnosis of MI, but has relatively poor spatial resolution. DE-MRI has superior spatial resolution, and it can be used to differentiate transmural and nontransmural MI.
J. Yang et al. / European Journal of Radiology 83 (2014) 778–782
Studies in animal experiments show a close correlation with the histopathologically determined area of MI [3,4]. Studies in humans show that DE-MRI is effective in determining the presence, location, and extent of acute and chronic MI [5–7]. Most of previous DE-MRI studies, however, are limited in the optimal dose of contrast for the diagnosis of MI. A recent international, multicenter, double-blinded, randomized trial shows DE-MRI using contrast doses of ≥0.2 mmol/kg is effective in the detection and assessment of both acute and chronic MI, and 0.1 mmol/kg is not enough [8]. In their study, however, intermediate doses between 0.1 and 0.2 mmol/kg have not been tested. The aim of the present study was to prospectively evaluate whether 0.15 mmol/kg of contrast agent is effective for the assessment of MI.
779
Table 1 Characteristics of the study population. Characteristics
No. of patients
Age (years) Range Sex (male/female) Risk factors for CAD Hypertension Hypercholesterolemia Smoking Diabetes Family history CAD classification Single-vessel Multi-vessel
55 ± 19 41–69 18/13 12 (39%) 17 (55%) 13 (42%) 5 (16%) 16 (32%) 6 (19%) 25 (81%)
Values are n (%) or mean ± standard deviation. CAD, coronary artery disease as defined by coronary artery stenosis ≥50%.
2. Materials and methods 2.1. Patients The study protocol was approved by the institutional review board. Written informed consent was obtained from each patient. A total of 31 consecutive patients (18 men, 13 women; age range, 41–69 years; mean age, 55 years ± 19) with chronic MI were prospectively enrolled between January 2012 and May 2013. All patients included in this study were confirmed to be positive to have single-vessel or multi-vessel disease by invasive X-ray angiography, and all were identified as chronic MI on the basis of occurring more than 1 year since the initial presentation of acute MI. Patients with contraindications for MRI (metallic implants such as pacemakers, defibrillators, cerebral aneurysm clips, ocular metallic deposits, severe claustrophobia), non-sinus rhythm (atrial fibrillation/flutter), dyspnea, renal insufficiency (estimated glomerular filtration rate assessed by creatinine clearance < 60 ml/min/1.73 m2 ), or a history of hemolytic anemia or other hemoglobinopathy and those who were pregnant or breastfeeding also were excluded. 2.2. DE-MRI imaging All patients were imaged in the supine position with a 3.0 T clinical scanner (Signa EXCITE, GE Healthcare, Waukesha, WI, USA) with a 4-element phased-array surface coil. Each patient underwent two DE-MRI scans with different contrast doses. One was 0.15 mmol/kg and the other was 0.2 mmol/kg. The two scans were performed in random order and on separate days. After localization of the heart, 8–10 short-axis slices were obtained to cover the entire left ventricle (LV) from base to apex. Two- and 4-chamber long-axis views also were obtained. DE-MRI was performed at 10 min after contrast material administration (Magnevist, Schering, Berlin, Germany). An inversion recovery fast gradient-echo pulse sequence was used to assess myocardial viability. Imaging parameters were: TR 6 ms, TE 3 ms, 36 cm FOV, 8-mm slice thickness, matrix 192 × 160, TI 180–270 ms (adjusted to null the signal of normal myocardium), 1 NEX, flip angle 20◦ .
percentage of LV myocardium was calculated by summing the segments with delayed enhancement (each weighted by the midpoint of the range of enhancement for the given segmental score; i.e., 1 = 13%, 2 = 38%, 3 = 63%, 4 = 88%) and dividing by 17 [8].
2.4. Statistics A statistical software program, SPSS 13.0 (SPSS Inc., Chicago, IL, USA), was used for statistical analysis. Continuous variables were expressed as mean ± standard deviation. Categorical variables were expressed as absolute number (percentage). Continuous data were compared using t test, which was paired where appropriate. The Wilcoxon signed rank test was performed to evaluate the difference in global infarct size percentage between the two dosages. Bland–Altman analysis was used to analyze correlation and agreement between global infarct size. The inter-observer agreement was determined with the kappa test. Statistical tests were 2-tailed and a p-value < 0.05 was considered to be significant.
3. Results All acquisitions were successful and all images acquired were of sufficient quality for the final analysis. The characteristics of the study population are summarized in Table 1.
3.1. Infarction segment DE-MRI showed enhanced myocardium in all the 31 patients with chronic MI. A total of 527 segments were analyzed, of which 242 segments with scar were found in both 0.15 mmol/kg and 0.2 mmol/kg images. Only 2 segments with scar in 1 patient were found in the 0.15 mmol/kg images, which were not detected in the 0.2 mmol/kg images. There was no significant difference between the 0.15 mmol/kg and 0.2 mmol/kg images in all 31 patients based on the infarction segment (7.87 ± 2.72 vs. 7.81 ± 2.64, respectively; p = 0.33) (Fig. 1).
2.3. DE-MRI image analysis The images were independently assessed by 2 experienced readers who were unaware of patient identity, contrast dose, or any associated clinical information. Any disagreement in the diagnosis between the 2 readers was settled by a consensus reading. DE-MRIs were graded for the presence, location, and extent of enhanced myocardium using a 17-segment model and a 5-point scale for each segment (0 = no delayed enhancement, 1 = 1–25%, 2 = 26–50%, 3 = 51–75%, 4 = 76–100%) [8]. Global infarct size as a
3.2. Infarction size There was no significant difference between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition (16.3% ± 7.8% vs. 16.4% ± 7.9%, respectively; p = 0.87). A strong correlation between the infarction size obtained from 0.15 mmol/kg acquisition and that from 0.2 mmol/kg acquisition was indicated through Bland–Altman analysis (Fig. 2).
780
J. Yang et al. / European Journal of Radiology 83 (2014) 778–782
Fig. 1. A 55-year-old man with chronic myocardial infarction. DE-MRI images using 0.15 mmol/kg of contrast agent (A) appears virtually identical to those using 0.2 mmol/kg (B).
3.3. Inter-observer agreement Inter-observer agreement of the segments for presence or absence of scar was evaluated by using the kappa test. The kappa score of 0.2 mmol/kg DE-MRI was 0.89, and that of 0.15 mmol/kg DE-MRI was 0.91 (p < 0.05 for both). 4. Discussion In the present study, we have prospectively compared the diagnostic value of DE-MRI at 3.0 T using 0.15 mmol/kg and 0.2 mmol/kg of contrast agent for the assessment of chronic MI. Our results demonstrate that DE-MRI using 0.15 mmol/kg enables effective
assessment of infarction segment and infarction size with high agreement compared with DE-MRI using 0.2 mmol/kg. Regarding the assessment of MI using DE-MRI in CAD, this is to the best of our knowledge the first study to compare measurements of infarction segment and infarction size in 0.15 mmol/kg and 0.2 mmol/kg of contrast agent acquisitions in a cohort of patients with chronic MI. We found strong and significant correlations between 0.15 mmol/kg and 0.2 mmol/kg of contrast agent acquisitions. These results confirm and extend the findings of published single-center and multi-center studies evaluating the accuracy of DE-MRI for detecting infarction in patients with acute or chronic MI [5–9]. Unlike these previous studies that involved different groups with different doses of contrast, in the current study each
Fig. 2. Bland–Altman analysis indicates strong correlation in the infarction size between DE-MRI using 0.15 mmol/kg and that using 0.2 mmol/kg.
J. Yang et al. / European Journal of Radiology 83 (2014) 778–782
patient underwent two DE-MRI scans with different contrast doses. Therefore, the results from the current study should reflect the comparison value of DE-MRI more accurately. We believe these findings indicate that DE-MRI using 0.15 mmol/kg can effectively assess MI. The current consensus criteria for the diagnosis of MI, which are fundamentally based on cardiac enzymes, the 12-lead ECG, and echocardiography, have important limitations [10]. Although cardiac enzymes are highly sensitive for acute MI, they are insensitive for chronic MI, and do not provide information on infarction segment and infarction size [10–13]. The presence of Q waves in the ECG has been used as a marker of prior MI. However, the ECG is insensitive for chronic non-Q-wave infarcts [14–18]. Although echocardiography is subsidiary to the ECG in stable patients presenting with ST-segment elevation MI, its role may be more important in patients with a non-diagnostic ECG [19]. Echocardiography is generally very accurate, but the sensitivity to detect acute MI varies widely [20]. Using echocardiography, chronic MI was reported to remain undetected more frequently, because of the disappearance of segmental wall motion abnormalities after several weeks [19,5]. DE-MRI may overcome most of these limitations. Studies in animal models and humans show an excellent match of infarction segment by DE-MRI to histopathology [3–7]. Recent studies reported silent myocardial infarctions (MIs) are prevalent among diabetic patients and inflict significant morbidity and mortality. DE-MRI can detect unrecognized MI in diabetic patients [21,22]. Moreover, because of high spatial resolution (more than 40-fold greater than SPECT), DE-MRI can even detect small infarcts in the absence of Q waves [6,16,23,24]. These findings demonstrate that DE-MRI is the best available technique for the detection and assessment of MI. Previous multi-center trial shows DE-MRI using contrast doses of ≥0.2 mmol/kg is effective in the detection and assessment of both acute and chronic MI, and 0.1 mmol/kg is not enough [8]. Our results demonstrate that DE-MRI using 0.15 mmol/kg of contrast agent is enough for the assessment of MI. Lower doses of contrast mean a lower cost, a darker blood pool to detect subendocardial infarcts [25], and decreased side effects such as nephrogenic systemic fibrosis in those with severe renal impairment. Several important limitations exist in the present study. First, our study was a consecutive analysis of patients in a single center; thus, it would be important to validate these results in multiple centers. Second, the population was relatively small, and the statistical power was relatively low. Third, the performance of both contrast doses was studied at 3.0 T. Additional 1.5 T studies are therefore needed to confirm our findings. Fourth, we have not tested the performance of DE-MRI at 3.0 T using contrast agent doses of
