Up to one-fourth of ischemic strokes remain cryptogenic after standard evaluation. The ESUS concept for embolic strokes of undetermined source was defined in 2014, the patients being supposed to benefit from anticoagulation, but the studies are negative as of this date.

Transthoracic and transesophageal (TEE) echocardiography are the main diagnostic tools for cardiac and aortic sources of embolism, the second being superior. However, its therapeutic impact is not fully determined, since it is not recommended in ESUS.

We aimed to assess prospectively, in consecutive ESUS patients admitted to our stoke unit, the rate of treatment change induced by transesophageal echocardiography (TEE), and to identify predictive factors for such modifications.

Acute ESUS patients were prospectively included to undergo TEE within seven days. Demographic, laboratory, MRI and echocardiographic data were recorded. The primary endpoint was the rate of TEE induced therapeutic modification.

Between October 2016 and May 2018, 154 ESUS patients entered the analyses. TEE changed the treatment in 22 patients (14.3%): anticoagulation in 7, cardiac surgery in 2, dual antiplatelet therapy for complex aortic arch atheroma in 10, antibiotics in 2 and atrial septal defect closure in 1. Patients with TEE induced treatment modification were significantly older (68.9 vs. 59.0 years, P<0.001), and had more vascular risk factors: diabetes mellitus (36.4% vs. 15.9%, P=0.036), hypertension (68.2% vs. 43.9%, P=0.035), hypercholesterolemia (72.7% vs. 28.8%, P<0.001) and more CRP (63.6% vs. 33.3%, P=0.007), hs Troponin I (50.0% vs. 22.7%, P=0.007) and NT-pro-BNP (31.8% vs. 9.8%, P=0.011) elevation. There was a trend to more multiple infarcts on DWI MRI in treatment change group (50.0% vs. 31.8%, P=0.097).

TEE lead to a treatment modification in 14% of cases. Selected ESUS patients may benefit from TEE.