Diagnosis
A detailed assessment of LVT is crucial for patient management and prognosis. Cardiac magnetic resonance (CMR) is now considered the diagnostic standard for detection of LVT with cine-CMR and contrast-enhanced CMR (CE-CMR) being the most useful modalities (Fig.1). This is particularly true in patients with mural or small LVT.12 The only study that included surgical or pathological evidence of LVT for confirmation of LVT showed that CE-CMR has a diagnostic sensitivity of 88% and specificity of 99%.13 No other study has subsequently shown that an alternative modality is superior to CE-CMR imaging in LVT detection.14
Cardiac magnetic resonance imaging has not only the strength to provide better spatial resolution for morphological definition15 but can characterize and differentiate the avascular LVT from neighboring structures after contrast administration. Moreover, CE-CMR can distinguish between acute versus older thrombus (Fig.1). Newer CMR sequences (such as T2* to identify ferrous products of hemoglobin breakdown and the use of long inversion time imaging to selectively null the normal myocardium) may provide further diagnostic advantages above and beyond CE-CMR imaging.16, 17
Contrast-enhanced CMR proved that an imaging delay after acute MI of more than 5 days was associated with significantly higher LVT detection rates compared to imaging performed within 5 days. Of note, CE-CMR performed between 9–14 days post-MI provided the highest detection rate.18, 19
Despite its diagnostic superiority, CMR remains a time consuming and expensive test, not available in all centers. Indeed, it is impractical to perform and repeat CMR in all patients with high-risk MI. A more practical approach may be to perform a transthoracic echocardiography (TTE) as the first-line imaging modality to screen for LVT in all patients with recent MI. Accordingly, current European Society of Cardiology guidelines recommend routine TTE during hospital stay in all patients to exclude LVT after MI (Class I, Level B).20Furthermore, considering that LVT could develop at various times after MI, performing repeated TTE rather than a single CE-CMR could have an even greater clinical impact in patients with satisfactory ultrasound quality.
The diagnosis of LVT by TTE should be defined as a mass in the LV cavity located adjacent to an area of LV wall dyssynergia and seen from at least two views (usually apical and short axis, Fig.2). Care must be taken to exclude the most common causes for an erroneous diagnosis of a thrombus (false tendons, trabeculae, technical artifacts and tangentially-cut LV wall).1, 12 Usually, LVT have a homogeneous texture with a softer echo density than myocardium, which suggests that the thrombus may be relatively recent and still “in a growing phase”, whereas an older thrombus tends to have a smoother surface and is typically more static.21
On the basis of available data it is possible to extrapolate that, in comparison with CE-CMR, non-contrast TTE has a sensitivity of 24%-33%, a specificity of 94%-95%, an accuracy of 82%, a positive predictive value of 57% and a negative predictive value of 85%.14 A low sensitivity may be of concern because TTE is the examination performed regularly in daily practice to search for LVT. However, it is difficult to generalize these data to the current “real life” since previous studies evaluating TTE used multiple gold standards, or none at all, and were conditioned by subjective image quality and the use of the off-axis projections.14 Indeed, TTE is more operator-dependent than CE-CMR. Varying gain setting and depth of field, as well as using transducers with different frequencies in multiple positions and orientations, are helpful approaches to minimize the false-positive studies. This important notion is highlighted by the finding that TTE performance varies highly according to the exam indication: if LVT search is prespecified, sensitivity is multiplied by 2 (60% vs. 26%) and positive predictive value by 3 (75% vs. 21%) as compared with unfocused routine TTE.22 When the search for LVT is prespecified in high-risk patients with recent anterior MI and low ejection fraction, the accuracy of TTE as compared with DE-CMR is even better (sensitivity and specificity 94.7% and 98.5%, respectively).19 Thus, we believe that the attempt to uncritically combine results from the literature to provide a summary estimate of the diagnostic accuracy of TTE might have led to inaccurate or misleading results, at least when considering Echo labs with high standards of quality in TTE. Indeed, considering the importance of the echogenicity, TTE ideally should be scored for diagnostic quality using previously validated quantitative tools.12
In short, we must emphasize that TTE accuracy can be excellent if performed specifically for LVT search with a standardized protocol and that non-visualized LVT are usually mural and small (Table).