Long-Term Embolic Outcomes After Detection of Left Ventricular Thrombus by Late Gadolinium Enhancement Cardiovascular Magnetic Resonance Imaging: A Matched Cohort Study

Abstract

Background: Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging is more sensitive than echocardiography for the detection of intracardiac thrombus because of its unique ability to identify thrombus based on tissue characteristics related to avascularity. The long-term prognostic significance of left ventricular (LV) thrombus detected by LGE CMR is unknown.

Methods: We performed a matched cohort study of consecutive adult patients with LV thrombus detected by LGE CMR who were matched on the date of CMR, age, and LV ejection fraction to up to 3 patients without LV thrombus. We investigated the long-term incidence of a composite of embolic events: stroke, transient ischemic attack, or extracranial systemic arterial embolism. We also compared outcomes among patients with LV thrombus detected by LGE CMR stratified by whether the LV thrombus was also detected by echocardiography or not.

Results: Of 157 LV thrombus patients, 155 were matched to 400 non-LV thrombus patients. During a median follow-up of 3.3 years, the cumulative incidence of embolism was significantly higher in LV thrombus patients compared with the matched non-LV thrombus patients (P<0.001), with annualized rates of 3.7% and 0.8% for LV thrombus and matched non-LV thrombus patients, respectively. LV thrombus was the only independent predictor of the composite embolic end point (hazard ratio, 3.99 [95% CI, 1.54-10.35]; P=0.004). The cumulative incidence of embolism was not different in patients with LV thrombus that was also detected by echocardiography versus patients with LV thrombus not detected by echocardiography (P=0.25).

Conclusions: Despite contemporary antithrombotic treatment, LV thrombus detected by LGE CMR is associated with a 4-fold higher long-term incidence of embolism compared with matched non-LV thrombus patients. LV thrombus detected by LGE CMR but not by echocardiography is associated with a similar risk of embolism as that detected by both LGE CMR and echocardiography.

Keywords: adult; echocardiography; embolism; humans; magnetic resonance imaging; prognosis; thrombosis.

Figures

Figure 1.. Examples of study patients.

Patient A had LV thrombus detected on LGE CMR and echocardiography and had an embolic outcome. Patient B had LV thrombus detected on LGE CMR but not echocardiography, and also had an embolic outcome. Patient C had no LV thrombus detected on LGE CMR and had no embolic outcome.

Figure 2.. Incidence of embolism in LV thrombus patients compared with matched non-LV thrombus patients.

Kaplan-Meier curves demonstrate the cumulative incidence of the composite embolic endpoint in the LV thrombus (in red) and in the matched non-LV thrombus (in blue) groups. Note the significant difference in the cumulative incidence of embolic events between the 2 groups.

Figure 3.. Incidence of embolism in patients with LV thrombus detected by echocardiography compared with patients with LV thrombus not detected by echocardiography.

Kaplan-Meier curves demonstrate the cumulative incidence of the composite embolic endpoint in patients with LV thrombus detected by echocardiography (in red) and in patients with LV thrombus not detected by echocardiography (in blue). Note the lack of a significant difference in the cumulative incidence of embolic events between the 2 groups.