Article Text
PDF
Statistics from Altmetric.com
In the previous article in this series, we discussed the cardiovascular magnetic resonance (CMR) techniques used in the assessment of ischaemic heart disease and their clinical application in patients with acute myocardial infarction. In this article we address the use of CMR in the chronic phase of infarction, focusing on the assessment of function and viability in patients with ischaemic cardiomyopathy or (suspected) prior infarction.
General considerations
A basic imaging protocol in patients with ischaemic cardiomyopathy or (suspected) old myocardial infarction includes cine imaging for the assessment of ventricular and valvular function, and delayed contrast enhanced (DE) imaging for the assessment of regional scar and viability. Total (maximum) examination time for the assessment of ventricular function and viability will be 30 min, which can be reduced to around 20 min if contrast is given before the examination. If the presence and distribution of scar is the only clinical concern, DE imaging may provide the answer within 5 min. Low dose dobutamine stress cine is a good alternative method for assessing viability in patients with (relative) contraindications to gadolinium based contrast agents—for example, in advanced renal failure. Depending on the clinical situation, adenosine first-pass imaging or high dose dobutamine stress cine can be added to detect ischaemia related perfusion defects or wall motion abnormalities, respectively. An extensive description of these techniques is beyond the scope of this article.
Function
Global ventricular function
Global left ventricular function strongly influences prognosis and management and its assessment is an essential part of the work-up in a patient with prior myocardial infarction. As in acute infarction, the selection of candidates for implantable cardioverter defibrillator (ICD) or cardiac resynchronisation devices relies heavily on left ventricular ejection fraction (LVEF). Cine imaging ensures the quantification of global function with the highest reproducibility and accuracy possible, thus avoiding both unnecessary implantations and deaths.
Right ventricular dysfunction is a well known risk factor of adverse outcome in patients with ischaemic heart disease. Quantified with cine CMR and defined as ejection fraction <40%, it was a significant predictor of mortality in a group of 147 patients with prior (>30 days) myocardial infarction, and remained so after adjustment for age, infarct size and LVEF.1 Only 16% of the patients with right ventricular dysfunction had right ventricular hyperenhancement, suggesting that mechanisms other than ischaemia play a role in its pathogenesis.1
Regional function
The high resolution SSFP (steady state free precision) cine sequence allows the detailed evaluation of regional wall thickness and thickening. Chronic scarring leads to regional thinning which may be extreme in remodelled ventricles (<4 mm)(figure 1). Decreased end-diastolic wall thickness is one of the parameters used in the evaluation of myocardial viability (see below).
End-diastolic (A) and end-systolic (B) still frames showing a severely deformed left ventricle with a massive aneurysm after inferolateral myocardial infarction. The delayed contrast enhanced image (C) shows the exact definition of the border between scar and healthy myocardium, as well as two small thrombi (arrows).
Infarct related complications
The unrestricted access to the heart and the chest allows the full assessment of severely deformed and remodelled ventricles or (pseudo-) aneurysms before reconstructive surgery (figure 1).2 Cine imaging is generally sufficient to show the deformation in detail; it also provides geometric information on mitral annular size and papillary muscles which may play a role in the preoperative evaluation of functional mitral regurgitation.2 DE imaging may help to diagnose and delineate (pseudo-) aneurysms by highlighting the borders of the scarred region, and is highly sensitive for the detection of left ventricular thrombi (figure 1C).
Regional low signal intensity may be seen within chronically infarcted regions on SSFP cines (figure 2). These correspond to lipomatous metaplasia, which refers to the transformation of fibrotic scars into fat. Although fat normally has high signal intensity on SSFP images, the intravoxel presence of myocytes and fat may cause phase cancellation and signal loss. T1 weighted spin echo imaging with and without fat suppression, or inversion recovery imaging (DE) before contrast injection (with adjusted inversion time), can be used to confirm this diagnosis (figure 2).3 The clinical significance of this finding that may also be detected with CT is unknown, although an association with remodelling and aneurysm formation has been suggested.4
Short axis view in a patient with old anteroseptal infarction. Lipomatous metaplasia (markers) leads to phase cancellation and signal loss in SSFP cine imaging (A), regional high signal intensity in T1 weighted spin echo imaging (B), and very low signal in T2 weighted spin echo imaging with fat suppression (C). See also subcutaneous fat (arrows).
Infarct visualisation
DE CMR can be considered the current gold standard for in vivo infarct visualisation at every stage of myocardial infarction. Serial DE imaging has shown that infarct size decreases by 19–31% in the first months after the acute event (figure 3).5 6 Infarct regression is the result of the healing process and the replacement of inflammation, haemorrhage, oedema and necrotic myocytes by collagenous scar. It is more pronounced in patients with larger infarct size and microvascular obstruction, and in patients with remodelling at follow-up.7
Three chamber delayed contrast enhanced view in a patient with anteroseptal infarction acquired 4 days (A) and 4 months (B) after reperfusion, showing infarct regression and disappearance of microvascular obstruction.
Chronic myocardial infarction is depicted as regional areas of hyperenhancement of varying extent, ranging from small subendocardial with preserved wall thickness to completely transmural with advanced wall thinning (figure 1C, figure 4). Often, several degrees of transmural extent can be found within one infarction, and the enhancement pattern may be irregular and patchy in patients with chronic coronary artery disease and a history of multiple infarctions or revascularisations (figure 4). Not surprisingly, ECG markers of transmurality (Q wave, non-Q wave) are poorly related to the actual extent of hyperenhancement.8 Q waves have low sensitivity for the detection of chronic infarction, especially of the (infero-) lateral wall, and their presence reflects the total size of the infarct rather than infarct transmurality.9 When Q waves are present, their location correlates well with the distribution according to DE.10
Relatively smooth subendocardial hyperenhancement in a patient with 6-day-old reperfused inferolateral infarction (A) and irregularly shaped and patchy enhancement in a patient with a history of coronary artery bypass grafting but no history of infarction (B).
The high sensitivity of DE for the detection of chronic scar facilitates the non-invasive diagnosis of unsuspected or previously non-diagnosed coronary artery disease.11 Although regional hyperenhancement per se is not specific for ischaemic heart disease, location and extent generally allow for the reliable differentiation from non-ischaemic myocardial disease.12 Ischaemic scar has a transmural or subendocardial distribution, whereas non-ischaemic disease can be found in the subepicardial or mid part of the left ventricular wall, outside the usual coronary artery territories.
Several studies have now shown that regional scar is strongly related to outcome and has prognostic value beyond global volumes and ejection fraction assessed with cine CMR, as well as other commonly used clinical markers.13 The size of the scarred region might be less relevant in terms of prognosis than its presence. One study evaluated 195 patients without known prior myocardial infarction but with known or suspected coronary artery disease, and found that the presence of any hyperenhancement was the strongest multivariable predictor of major adverse cardiovascular end points during a 16 month follow-up period.13
In addition to the presence, transmural extent and total size of the hyperenhanced region, recent studies have focused on the heterogeneity of signal in the infarct and its direct border zone.14 15 DE can visualise and quantify this ‘peri-infarct’ or ‘grey’ zone, which is thought to represent a mixture of viable myocytes and collagen fibres and a potential substrate of ventricular arrhythmia (figure 5). The extent of tissue heterogeneity has been related to mortality and increased inducibility of ventricular arrhythmia in patients with previous myocardial infarction.14 15 Also, papillary muscles with heterogeneous contrast uptake have been suggested as a possible origin of ventricular arrhythmia in patients with old myocardial infarction.16 However, a uniform approach to the assessment of signal heterogeneity on DE images is lacking and further study is needed to define its significance.
The ‘grey’ zone. Midventricular short axis view in a patient with old anteroseptal infarction showing an intensely bright subendocardial zone (marker) surrounded by a more patchy mid wall and subendocardial enhancement (arrows).
Viability
There is general agreement that patients with ischaemic cardiomyopathy (defined as ejection fraction <40% with evidence of prior infarction or hibernation, and ≥2 vessel coronary artery disease) should undergo revascularisation if there is significant potential for recovery of function.17 CMR has several ways to assess the presence of viability in dysfunctional myocardium: morphological evaluation of end-diastolic wall thickness (EDWT) using cine imaging; functional evaluation of inotropic reserve using low dose dobutamine stress cine imaging; evaluation of myocyte integrity and regional scar by DE imaging. Although first pass perfusion imaging may be used to demonstrate impaired resting blood flow in hibernating myocardium, it is not used for the prediction of functional outcome and is therefore not routinely included in the CMR imaging protocol of myocardial viability.18
End-diastolic wall thickness
Both echocardiographic and CMR studies have demonstrated a low likelihood of functional recovery after successful revascularisation in segments with chronically scarred, thinned myocardium.19 20 Using cine CMR imaging, it was shown that segmental EDWT <5.5 mm (based on a mean −2.5 SD in normal individuals) had a 92% sensitivity and a 56% specificity for the prediction of functional improvement.19 Echocardiographic studies have used similar cut-offs with comparable results, confirming the high negative predictive value of EDWT, with <6 mm virtually excluding functional recovery.20 However, current SSFP cine sequences allow a considerably better delineation of the blood–endocardial and epicardial border; we have seen on several occasions that wall thinning, even when extreme, did not preclude functional recovery, as long as there was no or minimal regional scarring at DE imaging (figure 6; see also supplementary video). Although extensive literature data are lacking, others have reported similar findings.21 This is why we currently consider severely thinned (<4 mm) segments viable if there is no or minimal hyperenhancement. However, further work is needed to define appropriate cut-offs for SSFP cine imaging.
Prediction of functional recovery in a patient presenting with heart failure and proximally occluded left anterior descending artery 1 year after percutaneous angioplasty and stenting. End-diastolic still frame (A) and end-systolic still frame (B) in two chamber view show pronounced wall thinning and lack of wall thickening (markers). Two chamber (C) and short axis (D) delayed contrast enhanced (DE) view showed no hyperenhancement. Follow-up cine 6 months after re-intervention showed normal wall thickness and full recovery of wall thickening (E, end-systolic, F, end-diastolic) with an increase in ejection fraction from 39% to 61% (see also supplementary videos 1 and 2).
Low dose dobutamine stress cine imaging
Low dose dobutamine stress imaging (LDDS) (5–10 μg/kg/min) can be used to assess the contractile reserve of dysfunctional but viable myocardium, and its clinical use has been demonstrated in a large number of studies using both echocardiography and cine CMR.22 23 A positive response (detectable increase in wall motion or thickening) requires both a significant amount of intact myofibrillar units as well as sufficient perfusion reserve, which is why LDDS techniques typically have lower sensitivity but higher specificity when compared to other (nuclear) techniques.24 Sensitivity can be notably reduced in segments with advanced wall motion abnormality, which potentially limits its use in patients that could benefit most from revascularisation—that is, the ones with the most severe left ventricular dysfunction.25
In a direct comparison, LDDS cine CMR was as effective as echocardiography in predicting functional recovery after revascularisation, with an overall diagnostic accuracy of 88% and 83%, respectively.22 The addition of high dose stress has been shown to improve diagnostic accuracy in echocardiography protocols. High dose dobutamine (up to 40 μg/kg/min with optional atropine to achieve a predefined target heart rate) induces ischaemia and may cause deterioration of wall motion in segments that showed improvement at low dose. This so-called ‘biphasic response’ is considered the best predictor of functional recovery after revascularisation.26 So far, the use of high dose dobutamine stress in cine CMR viability protocols has not yet been reported.
DE imaging
The use of DE imaging in viability assessment has several advantages compared to other techniques. DE offers superior spatial resolution (1–2 mm in-plane, slice thickness 4–6 mm) and high contrast between hyperenhanced scarred regions and non-enhanced viable regions. This allows for the transmural evaluation of scar even in segments with pronounced wall thinning. DE CMR is quick, reproducible and robust, and safer and less dependent on technician and observer skills than dobutamine stress imaging. Finally, it is the only technique that provides side-by-side visualisation of viable and non-viable parts. Thus, it allows for the assessment of viability as a gradual rather than a binary (present–non-present) phenomenon.
Functional improvement after revascularisation
Kim et al were the first to show that the transmural extent of scar strongly predicted functional outcome after revascularisation.27 The likelihood of segmental functional recovery was inversely related to the segmental extent of hyperenhancement (SEH): 78% of dysfunctional segments without hyperenhancement improved, whereas only one segment with >75% hyperenhancement improved. Using a cut-off value of ≤25% SEH to define viability, positive and negative predictive values were, respectively, 71% and 79% for segments with any baseline dysfunction, and 88% and 89% for segments with akinesia or dyskinesia. Furthermore, the total number of viable segments was strongly related to the degree of improvement of LVEF. Other studies have since confirmed the potential of DE in viability assessment, although the total amount of evidence is still limited compared to other techniques.
Comparison to other techniques
Several studies have shown that DE has good correlation with 18F fluorodeoxyglucose positron emission tomography (FDG PET), which has long been the reference imaging standard of viability.28 In 26 patients with chronic ischaemic cardiomyopathy, Kuhl et al showed that SEH was 9±14%, 33±25%, and 80±23% in severely dysfunctional segments with normal FDG uptake/perfusion, FDG/perfusion mismatch, and matched defects, respectively.28 A cut-off value of 37% SEH predicted PET viability with 96% sensitivity and 84% specificity. Two studies so far have compared FDG PET and DE to the clinical reference standard functional outcome after revascularisation.29 30 Both found comparable overall diagnostic accuracy, with one suggesting a higher negative predictive accuracy for FDG PET and the other for DE at follow-up periods of 17 days and 6 months, respectively. Knuesel et al postulated that functional improvement would only occur if a residual viable rim of sufficient wall thickness was present.31 Both DE and FDG PET were acquired in 10 patients with ischaemic cardiomyopathy. After 11 months, 85% of dysfunctional segments with both a thick viable rim (defined as >4.5 mm) and preserved FDG uptake improved, whereas all other segments (including thin segments with preserved FDG uptake and thick metabolically non-viable segments) had low improvement rates. However, the number of patients in this study was low and, as stated above, wall thickness may be less relevant if there is no or little hyperenhancement (figure 6; supplementary videos).
Several studies have compared DE imaging to LDDS cine imaging.32–34 The two techniques correlate well at the extremes of the enhancement spectrum: most segments with no or limited subendocardial scar (<25% SEH) demonstrate a contractile reserve whereas largely transmurally involved segments (>75% SEH) do not. In segments with intermediate ranges of segmental extent of enhancement (25–75%), the use of LDDS cine has been suggested to improve the detection of viable segments that may improve after revascularisation.33 34 However, the studies published so far were small and showed only small differences between both techniques, which leaves this matter as yet unresolved.
Lack of improvement despite viability
The use of DE imaging may help to explain why left ventricular function does not always improve after revascularisation despite the presence of viability. First, its high spatial resolution allows the identification of subendocardial scar, which may preclude functional improvement despite subepicardial viability. Second, DE imaging may help to identify new, clinically undetected areas of necrosis between the baseline study and the follow-up study. Procedure related myocardial necrosis may be substantial and has been shown to be an important negative predictor of functional outcome.35 Finally, the timing of the follow-up study to assess recovery of function may play an important role. Bondarenko et al recently reported on the relation between long term functional outcome after revascularisation and baseline segmental extent of hyperenhancement.36 Thirty-five patients with ischaemic cardiomyopathy (ejection fraction 39±11%) underwent cine and DE imaging 1 month before and 3, 6 and 24±12 months after revascularisation. Throughout the entire study period, the likelihood of improvement was strongly and inversely related to SEH (figure 7A). Interestingly, segments from all SEH groups continued to improve throughout the whole study and, at the end of the study period, the improvement rate was higher than previously reported in all SEH groups except in the 76–100% SEH. For example, at 3 months, 56% of segments without hyperenhancement had improved, which is relatively low compared to Kim et al.27 At long term follow-up, however, practically all (93%) had improved, and segments with 1–25, 26–50, 51–75, and 76–100% SEH were 2, 5, 11, and 86 times less likely to improve (multilevel analysis, p<0.001). The time course of improvement was considerably more delayed in segments with more extensive hyperenhancement at baseline (multilevel analysis, p<0.001) (figure 7B). These results demonstrate that DE imaging strongly predicts functional outcome even at long term follow-up, that the time course of improvement may be considerably delayed, and that both likelihood and time course are related to the baseline amount of scar. The delayed time course suggests that long follow-up periods may be required to assess the full potential of recovery of dysfunctional but viable myocardium.
Hibernation and the time course of functional improvement after revascularisation. (A) Likelihood of functional improvement after revascularisation in relation to baseline segmental extent of hyperenhancement (SEH), expressed as a percentage of total number of dysfunctional segments, at 3 months (blue bars), 6 months (red bars), and 2 years (white bars) follow-up. All dysfunctional segments are included (n=258). (B) Time course of regional functional improvement in relation to baseline SEH, shown as the relative percentage of improvement at 3 months (blue bars), 6 months (red bars), and 2 years (white bars) follow-up. Only segments with functional improvement are included (n=159). Reprinted with permission from Bondarenko O, Beek AM, Twisk JW, et al. Time course of functional recovery after revascularisation of hibernating myocardium: a contrast-enhanced cardiovascular magnetic resonance study. Eur Heart J 2008;29:2000–5.36
Device therapy
CMR cine imaging is the current gold standard for the quantification of ventricular function, which is the critical step in the selection of patients for implantable cardioverter-defibrillator (ICD) or cardiac resynchronisation therapy (CRT). The timing of ICD implantation may be difficult in patients with significant viability who have undergone revascularisation, since functional recovery may be considerably delayed.
Significant mechanical dyssynchronous contraction between the septal and the lateral wall is usually easily recognised on a regular four chamber or mid ventricular short axis SSFP cine. MRI tagging allows for the detailed 3D assessment of intramural deformation including strain, velocity and torsion. Although myocardial tagging is the ideal technique to assess the process of mechanical dyssynchrony, its current use is largely limited to research because postprocessing requires considerable know-how and is still relatively time consuming. Several quantitative measures of dyssynchrony have recently been proposed for both (plain) SSFP cine imaging as well as for myocardial tagging, all awaiting further clinical validation.37 38
DE imaging may be of value in the selection of CRT candidates by identifying potential non-responders by demonstrating regional transmural scar in the posterolateral segments or a large amount of total scar.39 40
Further study is also needed to determine whether the presence of peri-infarct DE signal heterogeneity may help to identify patients at even higher risk of ventricular arrhythmia (see ‘Infarct visualisation’ above).
Conclusions
The management of patients with ischaemic cardiomyopathy and a history of myocardial infarction critically depends on the assessment of left ventricular function and viability. For these indications, cardiovascular MRI has outgrown its status of a promising research tool and has established itself as a reliable alternative to echocardiography and nuclear techniques.
Use of CMR in patients with ischaemic cardiomyopathy and chronic myocardial infarction: key points
In a patient with ischaemic cardiomyopathy or (suspected) old myocardial infarction, cardiovascular magnetic resonance (CMR) should be considered:
To assess viability and the likelihood of functional improvement after revascularisation, preferably using DE imaging, with low dose dobutamine stress cine as a good alternative—for example, in patients with advanced renal failure.
For the quantification of left ventricular volumes and ejection fraction in the work-up of candidates for implantable cardioverter defibrillator (ICD) or cardiac resynchronisation therapy (CRT) devices.
To assess severely deformed and remodelled ventricles before reconstructive surgery.
To detect left (or right) ventricular thrombi.
In a patient with ischaemic cardiomyopathy or (suspected) old myocardial infarction, CMR may be used to:
Detect previously undiagnosed underlying coronary artery disease.
Identify potential CRT non-responders by showing regional transmural scar in the posterolateral wall or a large amount of global scar.
Identify patients at increased risk by showing signal heterogeneity in the peri-infarct region using delayed contrast enhanced imaging.
You can get CPD/CME credits for Education in Heart
Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to: http://heart.bmj.com/misc/education.dtl
RCP credits: Log your activity in your CPD diary online (http://www.rcplondon.ac.uk/members/CPDdiary/index.asp)—pass mark is 80%.
EBAC credits: Print out and retain the BMJ Learning certificate once you have completed the MCQs—pass mark is 60%. EBAC/ EACCME Credits can now be converted to AMA PRA Category 1 CME Credits and are recognised by all National Accreditation Authorities in Europe (http://www.ebac-cme.org/newsite/?hit=men02).
Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal's username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.
References
- ↵
- ↵
- ↵
- ↵
- ↵
Landmark paper that set the stage for delayed contrast enhanced CMR.
- ↵
- ↵
- ↵
This paper clarifies the relation between the Q wave/non-Q wave ECG classification of myocardial infarction and infarct extent according to DE imaging.
- ↵
- ↵
- ↵
This paper shows the feasibility of DE imaging to differentiate between ischaemic and non-ischaemic causes of cardiomyopathy.
- ↵
Excellent review on the DE contrast patterns seen in ischaemic and non-ischaemic cardiomyopathies.
- ↵
This paper demonstrates that the finding of regional myocardial scar in patients with suspected ischaemic heart disease has prognostic relevance.
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
The first study to demonstrate the potential of DE imaging to predict functional recovery of hibernating myocardium after revascularisation.
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
This study demonstrated that the time course of functional improvement of hibernating myocardium after revascularisation can be considerably protracted, and that both likelihood and time course are predicted by the segmental extent of hyperenhancement.
- ↵
- ↵
- ↵
- ↵
Footnotes
Competing interests In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The authors have no competing interests.
Provenance and peer review Commissioned; not externally peer reviewed.