Significance of Late Gadolinium Enhancement in Cardiovascular Magnetic Resonance Imaging (CMR

129Herz 32 · 2007 · Nr. 2 © Urban & Vogel 129Herz 32 · 2007 · Nr. 2 © Urban & Vogel

© Urban & Vogel 2007Herz

Significance of Late Gadolinium Enhancement in Cardiovascular Magnetic Resonance Imaging (CMR)Matthias Vöhringer, Heiko Mahrholdt, Ali Yilmaz, Udo Sechtem1

AbstractCardiovascular magnetic resonance imaging (CMR) permits optimal differentiation between normal and diseased myocardium with the use of gadolinium-based contrast agents and special magnetic reso-nance pulse sequences. Imaging is performed 10–20 min after contrast agent application to produce so-called late gadolinium enhancement (LGE) images which depict diseased myocardium with excellent re-producibility. Areas showing LGE correspond to zones of myocyte necrosis or myocardial fibrosis as shown by comparison with histopathology. Typical patterns of hyperenhancement exist in ischemic heart disease

but also in dilated cardiomyopathy, hypertrophic car-diomyopathy and other inflammatory or infiltrative myocardial disease and are described in this article. LGE-CMR is helpful to distinguish advanced ischemic heart disease from nonischemic dilated cardiomyop-athy. In ischemic heart disease LGE can also predict the functional recovery after revascularization proce-dures by directly showing the remaining viable myo-cardium. LGE may also become useful to predict ma-lignant arrhythmias in patients with ischemic heart disease or nonischemic cardiomyopathy. This may lead in future to an increased role of LGE-CMR as a prognostic tool.

Die Aussagekraft des Late Gadolinium Enhancement in der kardiovaskulären Magnetresonanztomographie (CMR)

ZusammenfassungMit speziellen Magnetresonanzsequenzen (s. Abbil-dung 2) und Gadolinium-(Gd-)basierten Kontrast-mitteln kann das sog. Late Gadolinium Enhance-ment (LGE) dargestellt werden (s. Abbildung 1). Eine relative Anreicherung von Gd und damit ein LGE ent-steht, wenn im Rahmen einer akuten Nekrose myo-kardiale Zellmembranen rupturiert sind und damit das Verteilungsvolumen von Gd zunimmt. Zugrunde liegen kann aber auch bei chonischen Prozessen ein im Rahmen des fibrotischen Umbaus vergrößerter extrazellulärer Raum im Myokard (s. Abbildung 3). Die verschiedenen myokardialen Erkrankungen füh-ren zu unterschiedlicher, typischer Ausprägung von LGE (s. Abbildungen 5 bis 10). Aufgrund dieser krank-heitstypischen Bilder kann LGE-CMR bei der Diffe-rentialdiagnose bei Patienten mit unklaren kardi-alen Krankheitsbildern, Herzinsuffizienz, Kardio-myopathien, Speichererkrankungen oder Myokarditis

sehr nützlich sein. Zuverlässig kann z.B. eine fortge-schrittene ischämische Herzerkrankung von dilata-tiven Kardiomyopathien nichtischämischer Genese unterschieden werden. Die LGE-CMR bietet auch zu-nehmend Möglichkeiten zur individuellen Progno-seabschätzung. Anhand des transmuralen Aus-maßes von Infarkten lässt sich der von einer revas-kularisierenden Maßnahme zu erwartende Gewinn abschätzen. Zunehmendes Interesse gilt der wei-teren Charakterisierung des mit Ausmaß, Verteilung und Homogenität von LGE-Arealen (und damit von myokardialen Narben) verknüpften Risikos für das Auftreten maligner Rhythmusstörungen. Entspre-chende Studien gibt es sowohl für ischämische als auch für dilatative und hypertrophe Kardiomyopa-thien. Eingang in die gültigen Leitlinien für entspre-chende therapeutische Maßnahmen wie Implanta-tion automatischer Kardioverter-Defibrillatoren ha-ben diese Daten bisher noch nicht gefunden.

Schlüsselwörter: Magnetresonanz-tomographie · Late Gadolinium Enhan-cement · Ischämische Herzkrankheit · Dilatative Kardiomyopathie · Hypertrophe Kardiomyo-pathie · Inflammato-rische Kardio myopathie · Infiltrative Kardiomyo-pathie

Key Words: Cardiac magnetic resonance · Late gadolinium enhance ment · Ischemic heart disease · Dil ated cardiomyopathy · Hypertrophic cardiomyopathy · Inflammatory heart disease · Infiltrative heart disease

1 Department of Cardiology and Pulmonology, Robert Bosch Hospital, Stutt-gart, Germany.

Herz 2007;32:129–37

DOI 10.1007/s00059-007-2972-5

Technical AspectsThe primary effect of most cardiovascular magnet-ic resonance imaging (CMR) contrast agents cur-rently approved for use in humans is shortening of the longitudinal relaxation time (T1). Consequent-ly, the goal of most CMR pulse sequences for eval-

uation of contrast enhancement is to make image intensities a strong function of T1 (T1-weighted images).

Early approaches to acquiring T1-weighted im-ages of the heart often employed ECG-gated spin echo techniques in which one k-space line was ac-

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

130 Herz 32 · 2007 · Nr. 2 © Urban & Vogel

quired in each cardiac cycle, resulting in image acqui-sition over several minutes during free breathing. Consequently, image quality was adversely affected by artifacts due to respiratory motion, partial volume from motional averaging over the respiratory cycle, and modest T1-weighting due to limited choices of repetition time.

Since the early use of ECG-gated spin echo im-aging a number of improvements have been made.

One of the most important is the use of k-space seg-mentation [13], which means that multiple k-space lines are acquired during each cardiac cycle. This re-sults in a reduction of imaging time to the point, where the entire image can be acquired during a sin-gle breath hold.

In addition, the preparation of magnetization prior to image acquisition by using an inversion pulse does not only increase the degree of T1-weighting, but also nulls most signal from normal myocardium. This will result in an improvement in signal intensity ratio between enhanced and normal myocardium of up to 500% compared to most spin echo techniques (Figure 1) [48].

Figure 2 shows this optimized segmented in-version recovery gradient echo (IR GRE) sequence in more detail. Following the R-wave of the ECG a delay period (“trigger delay”) is used to ensure that acquisition of the image occurs in diastole to minimize cardiac motion. The magnetization of the heart is then prepared by a nonselective 180° inver-sion pulse to increase T1-weighting. The inversion delay time (TI) is defined as the time between this 180° pulse and the center of acquisition of the seg-mented k-space lines (lines 1–23 in Figure 2). For correct implementation, the TI must be selected manually to null signal from normal myocardium. This TI varies from patient to patient as a function of dose and time due to gadolinium contrast kinet-ics [29].

If applied correctly, late gadolinium enhance-ment (LGE) using a segmented inversion recovery gradient echo pulse sequence is highly reproducible [29, 54] and has been extensively validated in animal models and a variety of patient cohorts. This article will review the clinical significance of this technique.

Mechanisms of Late Gadolinium Enhancement (LGE)

The likely mechanism of myocardial LGE is demon-strated in Figure 3. The mechanism is based on two simple facts. First, gadolinium chelates are extracel-lular contrast agents that are inert and cannot cross cell membranes [40, 55]. Second, in normal myocar-dium, myocytes are densely packed and thus myocyte intracellular space forms the majority (~85%) of the volume [19].

Conceptually, it then follows that the volume of distribution of gadolinium in a hypothetical voxel of normal myocardium is small (Figure 3a, K indicates high potassium concentration which is typical for the intracellular milieu in a myocyte), and the overall number of gadolinium molecules (Gd) is low. In the setting of acute myocardial damage, there is myocyte membrane rupture, which allows additional gadolini-

Same imaging session

T1 SE Segmented IR GRE

Figures 1a and 1b. LGE imaging of a subendocardial myocardial infarct (arrows). Note the distinct improvement in contrast and resolution of the segmented inver-sion recovery gradient echo technique (IR GRE, b) compared to T1 spin echo imag-ing (T1 SE, a; adapted by permission, [48]).Abbildungen 1a und 1b. LGE eines subendokardialen Myokardinfarkts (Pfeile). Be-achtenswert ist die deutliche Verbesserung von Kontrast und Auflösung durch die segmentierte Inversion-Recovery-Gradientenechotechnik (IR GRE, b) im Vergleich zur T1-Spinechotechnik (T1 SE, a; Nachdruck mit Erlaubnis, [48]).

a b

ECG

Trigger

Nonselective180° inversion

Triggerdelay

250!350 msTI

1 2 12 23

!1!2 !12 !23

Nonselective180° inversionMz infarct

Mz normal

R R R

Figure 2. Timing diagram for the segmented inversion recovery gradient echo se-quence (IR GRE) with TI set to null normal myocardium after contrast agent ad-ministration (adapted by permission, [48]).Abbildung 2. Diagramm zur Veranschaulichung der segmentierten Inversion-Re-covery-Gradientenechotechnik (IR GRE). TI wird so gewählt, dass normales Myo-kard nach Kontrastmittelgabe genullt wird (Nachdruck mit Erlaubnis, [48]).

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

131Herz 32 · 2007 · Nr. 2 © Urban & Vogel

um to diffuse into what was previously intracellular space (indicated by the high content in sodium [Na]). This in turn results in increased gadolinium concen-tration and therefore LGE (Figure 3b).

In the setting of chronic myocardial damage, myocytes have been replaced with collagenous scar (symbolized by black ribbons in Figure 3c). Thus, the interstitial space is also expanded [40], which again leads to increased gadolinium concentration and therefore LGE.

These mechanisms likely apply to many forms of acute and chronic myocardial damage, independent of the underlying cause (ischemia, inflammation, etc.), and may help to understand the different pat-terns of LGE found in different myocardial disorders [31].

However, for correct image interpretation it is important to be aware of some special situations. One is the “no-reflow phenomenon” that can be found early after acute myocardial infarction. The “no-re-flow phenomenon” may be explained by intracapil-lary red blood cell stasis in the central necrotic region of a larger infarct that is caused by capillary plugging resulting in tissue hypoperfusion [14, 18]. Conse-quently, no-reflow zones will appear dark as com-pared to the surrounding regions of LGE due to de-layed contrast penetration [30] (Figure 4). The size of the no-reflow region depends on the time between contrast injection and imaging and is larger when im-aging is started early.

Another special situation is diffuse plexiform fibrosis, which does not lead to detectable LGE for two reasons. First, the new optimized segmented in-version recovery sequence is sensitive to regional differences in gadolinium accumulation rather than to an overall increase of gadolinium concentration, because the technique depends on the ability to “null” signal from “remote” (presumably normal) myocardium (see Technical Aspects). Therefore, cardiac disorders that lead to focal regions of scar-ring will cause enhancement, whereas disorders that lead to global changes such as diffuse interstitial fi-brosis will not. Second, it should be noted that the voxel resolution of LGE-CMR is approximately 1.8 mm ! 1.2 mm ! 6 mm. Hence, only complete scar tissue comprising several voxels will be visible on CMR images as a bright area. Regional formation of smaller scars dispersed as islands within normal myocardium which may, for instance, occur in myo-carditis may result in grayish areas on LGE-CMR images. These areas may be more difficult to distin-guish from normal myocardium. In summary, LGE-CMR depicts areas of necrosis or scarring in vivo that previously could only be detected at au-topsy, but it is not suitable to delineate diffuse re-ticular interstitial fibrosis.

Significance of LGE in Ischemic Heart Disease

Animal studies invariably showed the presence of LGE in both acute and chronic myocardial infarction [14, 40]. In these studies LGE was closely correlated to histopathologic findings. LGE reflects the acute ischemic injury and only occurs in areas of irrevers-ibly injured myocardium. The resulting scar is re-stricted to the supply area of the affected coronary artery. As predicted by the wavefront theory scar for-mation always includes the subendocardium and spreads to a variable extent from there to the epicar-dium [31, 41] (Figures 5 and 6). The reproducibility of

Normal myocardium

Intact cell membrane Ruptured cell membrane Collagen matrix

Acute damage Scar

Figures 3a to 3c. Mechanism for LGE in acute and chronic myocardial damage. See text for details (adapted by permission, [30]).Abbildungen 3a bis 3c. Mechanismus des LGE bei akuter und chronischer myokar-dialer Schädigung. Details s. Text (Nachdruck mit Erlaubnis, [30]).

a b c

Figures 4a to 4d. The “no-reflow” phenomenon demonstrated by LGE in repeated images, acquired at the same location over time (from left to right). The bottom labels refer to the time after contrast administration. The “no-reflow” zone often originates from the subendocardium, initially appears black surrounded by larger regions of LGE (black arrow, a), and slowly takes up contrast over time (white ar-rows, b to d). This can be explained by microvascular damage originating from the subendocardium (wavefront phenomenon), which impedes penetration of the contrast into this area of the infarct (adapted by permission, [30]).Abbildungen 4a bis 4d. „No-reflow“-Phänomen, dargestellt im zeitlichen Verlauf (Bezeichnungen beziehen sich auf die Zeit nach Kontrastmittelgabe). Die initial schwarz erscheinende „no-reflow“-Zone beginnt meist subendokardial und ist von größeren LGE-Regionen umgeben (schwarzer Pfeil, a). Die Kontrastierung er-folgt mit zeitlicher Verzögerung (weiße Pfeile, b bis d). Dies kann durch eine mikro-vaskuläre Schädigung erklärt werden, die die Penetration von Kontrastmittel in das Zentrum des Myokardinfarkts behindert (Nachdruck mit Erlaubnis, [30]).

a b c d

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

132 Herz 32 · 2007 · Nr. 2 © Urban & Vogel

LGE measurements of infarct size has proven to be excellent [29, 51]. Infarct size by LGE also correlates well with clinical findings in acute infarcts [17]. For large infarcts LGE has the same high sensitivity for infarct detection as the current gold standard sin-gle-photon emission computed tomography (SPECT)

imaging [22]. However, due to its higher spatial reso-lution LGE-CMR is clearly superior to SPECT in the detection of small subendocardial infarctions (sensi-tivity of 92% for LGE-CMR vs. 28% for SPECT) [53]. Occasionally, it may be challenging to distin-guish such small subendocardial hyperenhancements from the bright blood in the left ventricular cavity. The additional use of a short inversion time can help in this case and further improve the diagnostic accu-racy of LGE-CMR [15].

LGE is also useful for detecting very small in-farcts which may occur during interventional proce-dures [42]. Although such infarcts may not be of im-mediate clinical or functional relevance [4], their presence and detection may be relevant for the pa-tients’ long-term prognosis. This was recently dem-onstrated in patients with small clinically unrecog-nized infarcts which were identified by LGE-CMR. During a median follow-up of 16 months, 31 of 195 patients (18%) experienced major adverse cardiac events (MACE), including 17 deaths. LGE demon-strated the strongest unadjusted associations with MACE and cardiac mortality. Patients in the lowest tertile of LGE-involved myocardium (mean left ven-tricular mass, 1.4%) still experienced a more than sevenfold increased risk for MACE. By multivariable analyses, LGE was the strongest predictor of MACE and cardiac mortality [25].

The high spatial resolution of CMR also permits detection of infarcts of the right ventricle. LGE-CMR detects right ventricular infarction more frequently than current standard diagnostic techniques [24].

LGE alone cannot distinguish between acute and chronic infarcts. This limitation may be overcome by additionally assessing myocardial edema with T2-weighted sequences or using different contrast agents [2, 45].

LGE-CMR is not only able to show the presence of irreversible myocardial damage, but it is unique in its ability of showing its transmural extent and the re-maining viable myocardium. This is of high relevance to predict the prognosis after revascularization in acute myocardial infarcts [5] and to estimate the po-tential benefit of revascularization procedures. The transmural extent of LGE is negatively correlated to the functional outcome after revascularization [10, 20]. CMR has proven to be at least of equal accuracy as nuclear imaging in predicting the benefit of revas-cularization [23]. However, it needs to be pointed out that there seems to be no clear threshold of transmu-rality that excludes functional recovery after revascu-larization [32]. Therefore, outcome remains difficult to predict even by CMR when scar involves between 25% and 75% of the myocardium. Low-dobutamine stress CMR may be more reliable than LGE-CMR for prediction of recovery following revascularization

a b c

Figures 5a to 5c. LGE in a patient with subendocardial myocardial infarction (ar-rows). Note the remaining viable myocardium in the epicardium. a) Typical pat-tern. b) Short axis. c) Long axis.Abbildungen 5a bis 5c. LGE bei einem Patienten mit einem subendokardialen In-farkt (Pfeile). Beachtenswert ist das epikardial verbliebene vitale Myokard. a) Gra-phik mit typischem LGE-Muster. b) Schnitt in der kurzen Achse. c) Schnitt in der langen Achse.

a b c

Figures 6a to 6c. LGE in a patient with transmural myocardial infarction (black ar-rows). There is pericardial thickening overlying the infarct zone (small white ar-rows). a) Typical pattern. b) Short axis. c) Long axis.Abbildungen 6a bis 6c. LGE bei einem Patienten mit einem transmuralen Infarkt (schwarze Pfeile). Im Infarktgebiet besteht eine perikardiale Verdickung (kleine weiße Pfeile). a) Graphik mit typischem LGE-Muster. b) Schnitt in der kurzen Ach-se. c) Schnitt in der langen Achse.

a b c

Figures 7a to 7c. LGE in a patient with dilated cardiomyopathy and a streaky mid-wall LGE in the septum, the so-called midwall sign (arrows). a) Typical pattern. b) Short axis. c) Long axis.Abbildungen 7a bis 7c. LGE bei einem Patienten mit dilatativer Kardiomyopathie und streifigem LGE im Septum, einem „midwall sign“ (Pfeile). a) Graphik mit ty-pischem LGE-Muster. b) Schnitt in der kurzen Achse. c) Schnitt in der langen Achse.

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

133Herz 32 · 2007 · Nr. 2 © Urban & Vogel

cause different therapeutic options exist. The typical subendocardial or transmural LGE in a coronary supply area is indicative of ischemic heart disease [33, 50, 58]. In the presence of heart failure and obstruc-tive coronary heart disease this pattern of LGE oc-curs in virtually all patients. However, ischemic pat-

[56] or it may improve the diagnostic accuracy of LGE when performed additionally [9].

LGE-CMR is able to detect areas within larger infarcts exhibiting the “no-reflow phenomenon” (see above) which represents a more severe form of tissue injury and is associated with a worse prognosis [59].

More recently, CMR was employed to predict susceptibility for malignant arrhythmias. Infarct sur-face area and mass, as measured by CMR, were bet-ter identifiers of patients who had a substrate for in-ducible monomorphic ventricular tachycardia than left ventricular ejection fraction [6]. When core and periinfarct regions as depicted on LGE-CMR images are separated using a computer-assisted, semiauto-matic algorithm based on signal-intensity thresholds (core > 3 SDs [standard deviations] and periphery 2–3 SDs above remote normal myocardium), patients with an above-median % LGE periphery are at a sig-nificantly higher risk for death compared with those with a below-median % LGE periphery [60]. This in-dicates that the extent of the periinfarct zone as char-acterized by CMR may provide incremental prognos-tic value beyond left ventricular systolic volume index or ejection fraction. It is hypothesized that the so es-timated “patchiness” of infarcts may be the substrate for electric reentry mechanisms. Infarct characteris-tics as assessed by CMR may thus prove to become unique and valuable noninvasive predictors of post-myocardial infarction mortality.

Another potential application of LGE-CMR is to plan and predict the outcome of cardiac resynchro-nization therapy (CRT) by the distribution and the amount of scar. CRT does not reduce left ventricular dyssynchrony in patients with transmural scar tissue in the posterolateral left ventricular segments diag-nosed by LGE-CMR, resulting in clinical and echo-cardiographic nonresponse to CRT [7]. Moreover, percent total scar is significantly higher in the nonre-sponse versus response group and predicts a lack of response by receiver-operating characteristic analy-sis [57].

Significance of LGE in Nonischemic Cardiomyopathies

LGE may also occur in nonischemic cardiac diseases such as dilated cardiomyopathy or hypertrophic car-diomyopathy.

Dilated cardiomyopathy (DCM) is defined as di-lation and impairment of function of one or both ven-tricles. DCM may be genetically determined or may be caused by chronic myocarditis, toxic agents, mus-cle dystrophies or neuromuscular diseases, metabolic and storage diseases, or systemic inflammatory dis-eases. It is important to differentiate between ad-vanced diffuse ischemic heart disease and DCM be-

a b c

Figures 8a to 8c. LGE in a patient with hypertrophic cardiomyopathy. Note the patchy LGE at the junction of the right ventricle and the interventricular septum (arrows). a) Typical pattern. b) Short axis. c) Long axis.Abbildungen 8a bis 8c. LGE bei einem Patienten mit hypertropher Kardiomyopa-thie. Beachtenswert ist das LGE an den Insertionsstellen des rechten Ventrikels (Pfeile). a) Graphik mit typischem LGE-Muster. b) Schnitt in der kurzen Achse. c) Schnitt in der langen Achse.

a b c

Figures 9a to 9c. LGE in a patient with acute myocarditis proven by endomyocar-dial biopsy. Note the epicardial LGE in the inferolateral wall (arrows). a) Typical pattern. b) Short axis. c) Long axis.Abbildungen 9a bis 9c. LGE bei einem Patienten mit akuter Myokarditis, die durch Myokardbiopsie bestätigt wurde. Beachtenswert ist das epikardiale LGE (Pfeile). a) Graphik mit typischem LGE-Muster. b) Schnitt in der kurzen Achse. c) Schnitt in der langen Achse.

a b c

Figures 10a to 10c. LGE obtained early (5 min) after contrast application in a pa-tient with cardiac amyloidosis proven by endomyocardial biopsy (arrows). a) Typi-cal pattern. b) Short axis. c) Long axis.Abbildungen 10a bis 10c. LGE bei einem Patienten mit histologisch nachgewie-sener kardialer Amyloidose, das sich früh nach Kontrastmittelgabe (5 min) dar-stellt (Pfeile). a) Graphik mit typischem LGE-Muster. b) Schnitt in der kurzen Achse. c) Schnitt in der langen Achse.

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

134 Herz 32 · 2007 · Nr. 2 © Urban & Vogel

terns of scar may also be observed in 10–15% of pa-tients with diffuse global reduction of left ventricular function felt to have DCM as they do not have ob-structive coronary disease but only diffuse plaque formation within the coronary arteries [33, 43, 50]. This ischemic pattern of scar without corresponding obstruction in the supplying coronary artery might be caused by transient occlusion by a thrombus or em-bolus or coronary vasospasm occurring in addition to the underlying myocardial abnormality. In these pa-tients, the extent of subendocardial scar does not ex-plain the extent of wall motion abnormalities.

Two other patterns of LGE are more frequent in patients with DCM: no LGE and a streaky or patchy enhancement in the midwall of the left ven-tricle [33] (Figure 7). Midwall LGE has an adverse prognosis as compared to DCM without LGE [3]. The histological basis for LGE is replacement fibro-sis which is found at necropsy in approximately half of the patients dying of the disease [43]. Recently, the midwall LGE pattern was found to be associated with active or borderline myocarditis by Dallas cri-teria in patients with a clinical presentation of chronic myocarditis and depressed left ventricular function or repetitive ventricular arrhythmias. This indicates that LGE-CMR may be able to identify the cause of ventricular dysfunction in a subset of patients with DCM [12].

As LGE is predictive of an adverse prognosis in patients with DCM, a potential clinical application is to provide such patients with an antiarrhythmic de-vice. Such a strategy in DCM is supported by the find-ing that DCM patients with predominance of scar distribution involving 26–75% of wall thickness are more prone to have inducible ventricular tachycar-dia. LGE-CMR may hence identify high-risk patients with nonischemic cardiomyopathy currently missed by ejection fraction criteria [38].

In patients with hypertrophic cardiomyopathy (HCM), LGE is a frequent finding [11, 34]. Hyperen-hancement is the result of a number of pathologic processes that result in different forms of fibrosis (re-placement scar or myocyte dropout) or in relation to myocardial disarray and subsequent local interstitial expansion. The different patterns of hyperenhance-ment are likely to be linked to the different patho-logic processes occurring in different patients, and the different stages that such processes have reached at the time of scanning. In a histopatho logic compara-tive study in a patient who underwent heart trans-plantation shortly after CMR imaging, there was ex-cellent correlation of LGE and fibrosis [35]. LGE appears most frequently at the junctions of interven-tricular septum and the right ventricle (Figure 8). Other typical locations are the most hypertrophied areas which are usually located in the septum. LGE

occurs in a multifocal, patchy pattern. In advanced stages of the disease there is progressive scarring which may finally lead to wall thinning [34]. The total amount of scar correlates with the clinical risk factors currently used to estimate the patient’s need for an antiarrhythmic device [34]. However, it has yet to be shown convincingly that LGE-CMR can provide ad-ditional or better information for risk assessment in HCM.

Significance of LGE in Inflammatory Heart Disease

The most common cause of inflammatory heart dis-ease in Europe is acute viral myocarditis. The histo-logical substrate for LGE in acute myocarditis is myocyte necrosis and not replacement fibrosis as in chronic myocarditis. LGE therefore represents the severity and focality of inflammation, which is de-termined by the patient’s disposition and the infec-tious viral agent. Differences in patient population, clinical or biopsy inclusion criteria are the likely ex-planation of the varying incidence of LGE observed in acute myocarditis ranging from 44% to 88% [1, 27]. The distribution of LGE is typically patchy in the epicardium of the inferolateral wall or more band-like in the midwall of the septum (Figure 9). The patterns of LGE may be related to the type of virus or viruses causing the inflammation [28]. The specificity of LGE in acute myocarditis is very high up to 100% [49]. In selected patients, the sensitivity of endomyocardial biopsy is improved if biopsies can be obtained from the region of LGE as com-pared to nonenhancing regions [27]. Epicardial LGE in acute myocarditis decreases with healing of myo-carditis [31] which may be explained by the patchy nature of the inflammation. With shrinkage of the small scattered scars which are still surrounded by normal myocytes, expansion of the extracellular space per voxel may become so small that enhance-ment becomes invisible. Another possibility is that there is some expansion of the interstitium by ede-ma which may be prominent in the acute stage but disappears with healing [16].

Inflammation in the myocardium is also part of the clinical spectrum in patients with Chagas’ disease. LGE is frequently found in these patients and some-times even before clinical manifestation of the dis-ease [44]. The pattern of LGE is similar to that in acute viral myocarditis [8]. It also affects predomi-nantly the epi- and midventricular layer of the infero-lateral wall. The amount of LGE seems to be corre-lated with the severity of ventricular dysfunction and severe arrhythmias [44].

In sarcoidosis cardiac involvement is difficult to diagnose clinically but present in 20–30% of patients

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

135Herz 32 · 2007 · Nr. 2 © Urban & Vogel

in autopsy studies [47]. LGE typically occurs in car-diac sarcoidosis as areas of focal, patchy hyperen-hancement usually located subepicardial or in the midwall [21, 52]. In a study with examination of con-secutive patients with biopsy-proven pulmonary sar-coidosis, LGE has emerged as an excellent tool to detect cardiac involvement [46].

Significance of LGE in Infiltrative Heart Disease

Cardiac amyloidosis is characterized by deposition of amyloid in the interstitium causing expansion of the interstitial space. This allows gadolinium-based MR contrast agents to diffuse into the interstitium and leads to LGE despite the absence of larger amounts of necrosis or fibrosis. The hyperenhancement is dif-fuse throughout the ventricle but pronounced in the subendocardium (Figure 10). It is best demonstrated early after contrast injection. Therefore, imaging should be started already 5 min after contrast injec-tion. If this is respected, LGE has an excellent diag-nostic accuracy of 97% [26]. It can be expected that LGE should be able to demonstrate regression of amyloid with treatment, but this has not yet been re-ported.

Another storage disease with cardiac involve-ment is Anderson-Fabry disease. It leads to diffuse cardiac hypertrophy due to intramyocardial accumu-lation of sphingolipids. Although this is a diffuse pro-cess throughout the myocardium, patients with more severe disease may show focal inferolateral midwall LGE [36]. LGE is caused by fibrosis as shown by cor-relation with autopsy findings [37], but it remains un-known why fibrosis appears focal and predominantly in the inferolateral wall.

ConclusionLGE is usually not performed on its own but is part in a CMR examination protocol providing comprehen-sive information in patients with heart disease [39]. In such a protocol LGE is a powerful tool to determine the presence and extent of myocardial diseases and can help the clinician to identify the etiology of heart-related symptoms and heart failure. However, the specificity of LGE for differentiating between the various forms of nonischemic myocardial disease ap-pears to be rather low indicating that endomyocardial biopsy cannot be replaced by CMR.

There is increasing evidence that the total amount and the pattern of distribution of LGE are of prognostic value especially with respect to its poten-tial ability to predict malignant arrhythmias. This may lead to an expanding prognostic role of LGE-CMR.

Conflict of interest: None. The authors declare that they had no financial or personal relations to other parties whose interests could have affected the content of this article in any way, either positively or negatively.

References1. Abdel-Aty H, Boye P, Zagrosek A, et al. Diagnostic perfor-

mance of cardiovascular magnetic resonance in patients with suspected acute myocarditis: comparison of different approaches. J Am Coll Cardiol 2005;45:1815–22.

2. Abdel-Aty H, Zagrosek A, Schulz-Menger J, et al. Delayed enhancement and T2-weighted cardiovascular magnetic resonance imaging differentiate acute from chronic myo-cardial infarction. Circulation 2004;109:2411–6.

3. Assomull RG, Prasad SK, Lyne J, et al. Cardiovascular mag-netic resonance, fibrosis, and prognosis in dilated cardio-myopathy. J Am Coll Cardiol 2006;48:1977–85.

4. Barbier CE, Bjerner T, Johansson L, et al. Myocardial scars more frequent than expected: magnetic resonance imag-ing detects potential risk group. J Am Coll Cardiol 2006;48:765–71.

5. Barclay JL, Egred M, Kruszewski K, et al. The relationship be-tween transmural extent of infarction on contrast en-hanced magnetic resonance imaging and recovery of con-tractile function in patients with first myocardial infarction treated with thrombolysis. Cardiology 2006;108:217–22.

6. Bello D, Fieno DS, Kim RJ, et al. Infarct morphology identi-fies patients with substrate for sustained ventricular tachy-cardia. J Am Coll Cardiol 2005;45:1104–8.

7. Bleeker GB, Kaandorp TA, Lamb HJ, et al. Effect of postero-lateral scar tissue on clinical and echocardiographic im-provement after cardiac resynchronization therapy. Circu-lation 2006;113:969–76.

8. Bocchi EA, Kalil R, Bacal F, et al. Magnetic resonance imag-ing in chronic Chagas’ disease: correlation with endomyo-cardial biopsy findings and gallium-67 cardiac uptake. Echocardiography 1998;15:279–88.

9. Bodi V, Sanchis J, Lopez-Lereu MP, et al. Usefulness of a com-prehensive cardiovascular magnetic resonance imaging assessment for predicting recovery of left ventricular wall motion in the setting of myocardial stunning. J Am Coll Cardiol 2005;46:1747–52.

10. Choi KM, Kim RJ, Gubernikoff G, et al. Transmural extent of acute myocardial infarction predicts long-term improve-ment in contractile function. Circulation 2001;104:1101–7.

11. Choudhury L, Mahrholdt H, Wagner A, et al. Myocardial scarring in asymptomatic or mildly symptomatic patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 2002;40:2156–64.

12. De Cobelli F, Pieroni M, Esposito A, et al. Delayed gadolini-um-enhanced cardiac magnetic resonance in patients with chronic myocarditis presenting with heart failure or recur-rent arrhythmias. J Am Coll Cardiol 2006;47:1649–54.

13. Edelman RR, Wallner B, Singer A, et al. Segmented turbo-FLASH: method for breath-hold MR imaging of the liver with flexible contrast. Radiology 1990;177:515–21.

14. Fieno DS, Kim RJ, Chen EL, et al. Contrast-enhanced mag-netic resonance imaging of myocardium at risk: distinction between reversible and irreversible injury throughout in-farct healing. J Am Coll Cardiol 2000;36:1985–91.

15. Foo TK, Wolff SD, Gupta SN, et al. Enhanced viability imag-ing: improved contrast in myocardial delayed enhance-ment using dual inversion time subtraction. Magn Reson Med 2005;53:1484–9.

16. Hiramitsu S, Morimoto S, Kato S, et al. Transient ventricular wall thickening in acute myocarditis: a serial echocardio-graphic and histopathologic study. Jpn Circ J 2001;65:863–6.

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

136 Herz 32 · 2007 · Nr. 2 © Urban & Vogel

17. Ingkanisorn WP, Rhoads KL, Aletras AH, et al. Gadolinium delayed enhancement cardiovascular magnetic resonance correlates with clinical measures of myocardial infarction. J Am Coll Cardiol 2004;43:2253–9.

18. Kim RJ, Fieno DS, Parrish TB, et al. Relationship of MRI de-layed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation 1999;100:1992–2002.

19. Kim RJ, Judd RM, Chen EL, et al. Relationship of elevated 23Na magnetic resonance image intensity to infarct size af-ter acute reperfused myocardial infarction. Circulation 1999;100:185–92.

20. Kim RJ, Wu E, Rafael A, et al. The use of contrast-enhanced magnetic resonance imaging to identify reversible myo-cardial dysfunction. N Engl J Med 2000;343:1445–53.

21. Kiuchi S, Teraoka K, Koizumi K, et al. Usefulness of late gad-olinium enhancement combined with MRI and 67-Ga scin-tigraphy in the diagnosis of cardiac sarcoidosis and disease activity evaluation. Int J Cardiovasc Imaging 2007:in press.

22. Klein C, Nekolla SG, Bengel FM, et al. Assessment of myo-cardial viability with contrast-enhanced magnetic reso-nance imaging: comparison with positron emission to-mography. Circulation 2002;105:162–7.

23. Kuhl HP, Lipke CS, Krombach GA, et al. Assessment of revers-ible myocardial dysfunction in chronic ischaemic heart dis-ease: comparison of contrast-enhanced cardiovascular magnetic resonance and a combined positron emission tomography-single photon emission computed tomogra-phy imaging protocol. Eur Heart J 2006;27:846–53.

24. Kumar A, Abdel-Aty H, Kriedemann I, et al. Contrast-en-hanced cardiovascular magnetic resonance imaging of right ventricular infarction. J Am Coll Cardiol 2006;48:1969–76.

25. Kwong RY, Chan AK, Brown KA, et al. Impact of unrecog-nized myocardial scar detected by cardiac magnetic reso-nance imaging on event-free survival in patients present-ing with signs or symptoms of coronary artery disease. Circulation 2006;113:2733–43.

26. Maceira AM, Joshi J, Prasad SK, et al. Cardiovascular mag-netic resonance in cardiac amyloidosis. Circulation 2005;111:186–93.

27. Mahrholdt H, Goedecke C, Wagner A, et al. Cardiovascular magnetic resonance assessment of human myocarditis: a comparison to histology and molecular pathology. Circula-tion 2004;109:1250–8.

28. Mahrholdt H, Wagner A, Deluigi CC, et al. Presentation, pat-terns of myocardial damage, and clinical course of viral myocarditis. Circulation 2006;114:1581–90.

29. Mahrholdt H, Wagner A, Holly TA, et al. Reproducibility of chronic infarct size measurement by contrast-en-hanced magnetic resonance imaging. Circulation 2002;106:2322–7.

30. Mahrholdt H, Wagner A, Judd RM, et al. Assessment of myo-cardial viability by cardiovascular magnetic resonance im-aging. Eur Heart J 2002;23:602–19.

31. Mahrholdt H, Wagner A, Judd RM, et al. Delayed enhance-ment cardiovascular magnetic resonance assessment of non-ischaemic cardiomyopathies. Eur Heart J 2005;26:1461–74.

32. Mahrholdt H, Wagner A, Parker M, et al. Relationship of con-tractile function to transmural extent of infarction in pa-tients with chronic coronary artery disease. J Am Coll Car-diol 2003;42:505–12.

33. McCrohon JA, Moon JC, Prasad SK, et al. Differentiation of heart failure related to dilated cardiomyopathy and coro-nary artery disease using gadolinium-enhanced cardiovas-cular magnetic resonance. Circulation 2003;108:54–9.

34. Moon JC, McKenna WJ, McCrohon JA, et al. Toward clinical risk assessment in hypertrophic cardiomyopathy with gad-

olinium cardiovascular magnetic resonance. J Am Coll Car-diol 2003;41:1561–7.

35. Moon JC, Reed E, Sheppard MN, et al. The histologic basis of late gadolinium enhancement cardiovascular magnetic resonance in hypertrophic cardiomyopathy. J Am Coll Car-diol 2004;43:2260–4.

36. Moon JC, Sachdev B, Elkington AG, et al. Gadolinium en-hanced cardiovascular magnetic resonance in Anderson-Fabry disease. Evidence for a disease specific abnormality of the myocardial interstitium. Eur Heart J 2003;24:2151–5.

37. Moon JC, Sheppard M, Reed E, et al. The histological basis of late gadolinium enhancement cardiovascular magnetic resonance in a patient with Anderson-Fabry disease. J Car-diovasc Magn Reson 2006;8:479–82.

38. Nazarian S, Bluemke DA, Lardo AC, et al. Magnetic reso-nance assessment of the substrate for inducible ventricu-lar tachycardia in nonischemic cardiomyopathy. Circulation 2005;112:2821–5.

39. Poon M, Fuster V, Fayad Z. Cardiac magnetic resonance im-aging: a “one-stop-shop” evaluation of myocardial dys-function. Curr Opin Cardiol 2002;17:663–70.

40. Rehwald WG, Fieno DS, Chen EL, et al. Myocardial magnetic resonance imaging contrast agent concentrations after re-versible and irreversible ischemic injury. Circulation 2002;105:224–9.

41. Reimer KA, Lowe JE, Rasmussen MM, et al. The wavefront phenomenon of ischemic cell death. 1. Myocardial infarct size vs duration of coronary occlusion in dogs. Circulation 1977;56:786–94.

42. Ricciardi MJ, Wu E, Davidson CJ, et al. Visualization of dis-crete microinfarction after percutaneous coronary inter-vention associated with mild creatine kinase-MB elevation. Circulation 2001;103:2780–3.

43. Roberts WC, Siegel RJ, McManus BM. Idiopathic dilated car-diomyopathy: analysis of 152 necropsy patients. Am J Car-diol 1987;60:1340–55.

44. Rochitte CE, Oliveira PF, Andrade JM, et al. Myocardial de-layed enhancement by magnetic resonance imaging in patients with Chagas’ disease: a marker of disease severity. J Am Coll Cardiol 2005;46:1553–8.

45. Saeed M, Weber O, Lee R, et al. Discrimination of myocardial acute and chronic (scar) infarctions on delayed contrast en-hanced magnetic resonance imaging with intravascular magnetic resonance contrast media. J Am Coll Cardiol 2006;48:1961–8.

46. Serra JJ, Monte GU, Mello ES, et al. Images in cardiovascular medicine. Cardiac sarcoidosis evaluated by delayed-en-hanced magnetic resonance imaging. Circulation 2003;107:e188–9.

47. Silverman KJ, Hutchins GM, Bulkley BH. Cardiac sarcoid: a clinicopathologic study of 84 unselected patients with sys-temic sarcoidosis. Circulation 1978;58:1204–11.

48. Simonetti OP, Kim RJ, Fieno DS, et al. An improved MR imag-ing technique for the visualization of myocardial infarc-tion. Radiology 2001;218:215–23.

49. Skouri HN, Dec GW, Friedrich MG, et al. Noninvasive imaging in myocarditis. J Am Coll Cardiol 2006;48:2085–93.

50. Soriano CJ, Ridocci F, Estornell J, et al. Noninvasive diagnosis of coronary artery disease in patients with heart failure and systolic dysfunction of uncertain etiology, using late gadolinium-enhanced cardiovascular magnetic resonance. J Am Coll Cardiol 2005;45:743–8.

51. Thiele H, Kappl MJ, Conradi S, et al. Reproducibility of chron-ic and acute infarct size measurement by delayed enhance-ment-magnetic resonance imaging. J Am Coll Cardiol 2006;47:1641–5.

52. Vignaux O. Cardiac sarcoidosis: spectrum of MRI features. AJR Am J Roentgenol 2005;184:249–54.

Vöhringer M, et al. Late Gadolinium Enhancement in CMR

137Herz 32 · 2007 · Nr. 2 © Urban & Vogel

53. Wagner A, Mahrholdt H, Holly TA, et al. Contrast-enhanced MRI and routine single photon emission computed tomog-raphy (SPECT) perfusion imaging for detection of subendo-cardial myocardial infarcts: an imaging study. Lancet 2003;361:374–9.

54. Wagner A, Mahrholdt H, Thomson L, et al. Effects of time, dose, and inversion time for acute myocardial infarct size measurements based on magnetic resonance imaging-de-layed contrast enhancement. J Am Coll Cardiol 2006;47:2027–33.

55. Weinmann HJ, Brasch RC, Press WR, et al. Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. AJR Am J Roentgenol 1984;142:619–24.

56. Wellnhofer E, Olariu A, Klein C, et al. Magnetic resonance low-dose dobutamine test is superior to SCAR quantifica-tion for the prediction of functional recovery. Circulation 2004;109:2172–4.

57. White JA, Yee R, Yuan X, et al. Delayed enhancement mag-netic resonance imaging predicts response to cardiac re-synchronization therapy in patients with intraventricular dyssynchrony. J Am Coll Cardiol 2006;48:1953–60.

58. Wu E, Judd RM, Vargas JD, et al. Visualisation of presence, location, and transmural extent of healed Q-wave and non-Q-wave myocardial infarction. Lancet 2001;357:21–8.

59. Wu KC, Zerhouni EA, Judd RM, et al. Prognostic significance of microvascular obstruction by magnetic resonance imag-ing in patients with acute myocardial infarction. Circula-tion 1998;97:765–72.

60. Yan AT, Shayne AJ, Brown KA, et al. Characterization of the peri-infarct zone by contrast-enhanced cardiac magnetic resonance imaging is a powerful predictor of post-myocar-dial infarction mortality. Circulation 2006;114:32–9.

Address for CorrespondenceUdo Sechtem, MDAbteilung für Kardiologie und PulmologieRobert-Bosch-KrankenhausAuerbachstraße 11070376 StuttgartGermanyPhone/Fax (+49/711) 8101-3456e-mail: [email protected]