Neurology International (Dec 2021)

Diagnostic Yield of TEE in Patients with Cryptogenic Stroke and TIA with Normal TTE: A Systematic Review and Meta-Analysis

  • Shamik Shah,
  • Preeti Malik,
  • Urvish Patel,
  • Yunxia Wang,
  • Gary S. Gronseth

DOI
https://doi.org/10.3390/neurolint13040063
Journal volume & issue
Vol. 13, no. 4
pp. 659 – 670

Abstract

Read online

Introduction: The role of transesophageal echocardiography (TEE) in cryptogenic stroke and transient ischemic attack (TIA) with normal transthoracic echocardiography (TTE) remains controversial in the absence of definite guidelines. We aimed to perform a systematic review and meta-analysis to estimate an additional diagnostic yield and clinical impact of TEE in patients with cryptogenic stroke and TIA with normal TTE. Methods: We performed a systematic review of cohort studies on PubMed using the keywords ‘cryptogenic stroke’, cryptogenic TIA’, ‘TEE’, and ‘TTE’ with matching MeSH terms. We included studies with patients who had cryptogenic stroke or TIA and had normal TTE findings, where the study intended to obtain TEE on all patients and reported all TEE abnormalities. The studies containing patients with atrial fibrillation were excluded. All studies were evaluated for internal and external validity. Inverse variance random effects models were used to calculate the effect size, the number needed to diagnose, and the 95% confidence interval. Results: We included 15 studies with 2054 patients and found LA/LAA/aortic thrombus, valvular vegetation, PFO-ASA, valvular abnormalities, and complex aortic plaques on TEE. Of these, 37.5% (29.7%–45.1%) of patients had additional cardiac findings on TEE. Management of 13.6% (8.1%–19.1%) of patients had changed after TEE evaluation. Based on current guidelines, it should change management in 4.1% (2.1%–6.2%) of patients and could potentially change management in 30.4% (21.9%–38.9%) of patients. Sensitivity analysis was also performed with only class II studies to increase internal validity, which showed additional cardiac findings in 38.4% (28.5%–48.3%), changed management in 20.2% (8.7%–31.8%), should change management in 4.7% (1.5%–7.9%), and could potentially change management in 30.4% (17.8%–43.0%) of patients. Conclusions: The diagnostic yield of TEE to find any additional cardiac findings in patients with cryptogenic stroke or TIA is not only high, but it can also change management for certain cardiac abnormalities. TTE in cryptogenic stroke or TIA may mitigate future risks by tailoring the management of these patients.

Keywords