Usefulness of CHA2DS2-VASc scoring system for predicting risk of embolism in patients with cardiac tumours: a single-centre study

Abstract

Abstract Background The CHA2DS2-VASc score is the most used in predicting thromboembolic risk in patients with atrial fibrillation (AF). However, a higher CHA2DS2-VASc score predicts cardiovascular events even in patients without AF. Up to date, there is limited evidence about the association between CHA2DS2-VASc score and the embolic risk in patient with cardiac tumours (CT). Purpose To evaluate the role of the CHA2DS2-VASc score in predicting embolism and to identify other markers of embolization in patients with CT. Methods We included all patients with a CT who had a complete echocardiographic evaluation and a histologic diagnosis. We excluded patients with cardiac thrombi because of their higher embolic risk. Results Our sample consisted of 221 patients (mean age 60.9±15.6, 7% female). 132 (59.7%) patients had benign CT, usually left-sided (84.8%). In the malignant tumours, 28 (31.5%) were primary and 61 (68.5%) were metastasis. Compared to benign tumours, malignant ones were predominantly right-sided (47.2%) or with pericardial (20.2%) or great-vessels involvement (16.9%). A total of 59 patients (26.6%) experienced an embolic event, of which 33 patients had peripheral embolism, while 28 had pulmonary embolism (PE), 2 patients had both. Patients with embolism were older (p=0.013), had a higher prevalence of diabetes or previous stroke (p=0.019 and p<0.001), had left-sided and not-infiltrating CT (p=0.001 and p=0.04). We did not find any differences in AF prevalence, atrium volumes or other CT characteristics. Moreover, patients who developed an embolic event had a higher CHA2DS2-VASc score than those without embolism (p<0.001). In particular, a CHA2DS2-VASc score >3 had an AUC=0.835 in predicting a peripheral embolic event in the overall population (p<0.001). At multivariate analysis, only CHA2DS2-VASc and left-side localization were significantly associated with embolization (p<0.001 and p=0.009). Also in patients with right-sided CT, CHA2DS2-VASc score (p=0.017), together with some tumours characteristics (mobility, pericardial effusion and absence of myocardial infiltration (p<0.04 for all) was associated with PE. Embolization did not impact survival when considering the overall population. Otherwise, when focusing on patients with malignant CT, those who had an embolic event had a worse prognosis (p=0.02), as well as those with PE (p=0.037). Conclusions CHA2DS2-VASc and left-side localization are the best markers of embolism. In particular, CHA2DS2-VASc seems to predict embolization in CT, regardless of histology or localization. Many patients with CT and embolism may not be offered surgical treatment given their assumed high-risk profile. By contrast, our analysis showed that survival is not related to the embolic event per se, but by histology. In patients with CT and high CHA2DS2-VASc score, further studies are needed to evaluate the best therapeutic strategy to minimize the embolic risk. Funding Acknowledgement Type of funding sources: None.