Introduction: ED observation units offer an alternative to hospitalization for the rapid diagnosis and etiologic evaluation of clinically suspected TIA. While cardiac embolism is thought to account for at least 25% of ischemic cerebrovascular events, the diagnostic yield and clinical impact of TTE in an ED observation unit has yet to be determined.
Methods: We conducted a retrospective review of 676 patients with suspected TIA within prospectively collected registry data of an ED observation unit between 2013-2017. Patients were considered eligible regardless of clinical risk score and routinely underwent DW-MRI/MRA and TTE. New onset atrial fibrillation was considered an exclusion criterion. A multivariable logistic regression was used to identify clinical predictors of abnormal TTE findings of possible cardioembolic etiology. Fisher’s exact test was performed to compare 30-day outcomes of patients who did or did not receive a TTE.
Results: Among 676 patients with clinically suspected TIA who underwent ED observation, a TTE was performed on 69% (n=465). After completion of the observation period, a final diagnosis of ischemic stroke, TIA, or mimic accounted for 14%, 44%, and 42% respectively. Of those diagnosed with stroke or TIA, abnormal TTE findings suggestive of a low or high-risk source occurred in 12% (n=35) and 2% (n=6) respectively. Age, gender, ECG abnormalities, heart murmur, history of diabetes, hypertension, hyperlipidemia, atrial fibrillation, anticoagulant use, cardiac surgery, coronary disease or stenting did not predict abnormal TTE findings (chi2 p=.78 for low-risk, p=.31 for high-risk TTE findings). These findings led to initiation of anticoagulation in 2 patients (0.7%). There was no difference in 30-day clinical outcomes (recurrent TIA/stroke, ED recidivism or hospital admission) in patients that did or did not have a TTE (p=0.92).
Conclusions: In our cohort of ED patients undergoing ED observation for clinically suspected TIA, a TTE informed of a possible cardioembolic source in 14% of patients with a final diagnosis of TIA or ischemic stroke. These abnormalities were not predicted by clinical variables alone and tended to not lead to a change in clinical management.