The value of transesophageal echocardiography for embolic strokes of undetermined source.

OBJECTIVE: Our aim was to evaluate the diagnostic yield of transesophageal echocardiography (TEE) in consecutive patients with ischemic stroke (IS) fulfilling the diagnostic criteria of embolic strokes of undetermined source (ESUS).

METHODS: We prospectively evaluated consecutive patients with acute IS satisfying ESUS criteria who underwent in-hospital TEE examination in 3 tertiary care stroke centers during a 12-month period. We also performed a systematic review and meta-analysis estimating the cumulative effect of TEE findings on therapeutic management for secondary stroke prevention among different IS subgroups.

RESULTS: We identified 61 patients with ESUS who underwent investigation with TEE (mean age 44 ± 12 years, 49% men, median NIH Stroke Scale score = 5 points [interquartile range: 3-8]). TEE revealed additional findings in 52% (95% confidence interval [CI]: 40%-65%) of the study population. TEE findings changed management (initiation of anticoagulation therapy, administration of IV antibiotic therapy, and patent foramen ovale closure) in 10 (16% [95% CI: 9%-28%]) patients. The pooled rate of reported anticoagulation therapy attributed to abnormal TEE findings among 3,562 acute IS patients included in the meta-analysis (12 studies) was 8.7% (95% CI: 7.3%-10.4%). In subgroup analysis, the rates of initiation of anticoagulation therapy on the basis of TEE investigation did not differ (p = 0.315) among patients with cryptogenic stroke (6.9% [95% CI: 4.9%-9.6%]), ESUS (8.1% [95% CI: 3.4%-18.1%]), and IS (9.4% [95% CI: 7.5%-11.8%]).

CONCLUSIONS: Abnormal TEE findings may decisively affect the selection of appropriate therapeutic strategy in approximately 1 of 7 patients with ESUS.