Does the Magnitude of the Electrocardiogram QT Interval Dispersion Predict Stroke Outcome?

OBJECTIVE: 1) Determine if admission QT interval dispersion (QTd) predicts short term clinical outcome in acute ischemic stroke (AIS) and 2) Determine if brain lesion location (insular vs non-insular cortex) predicts QTd. BACKGROUND: QTd, the maximal inter-lead difference in QT interval on a 12-lead ECG, is an indirect measure of myocardial repolarization heterogeneity. Data are conflicting on whether QTd predicts adverse outcome in acute stroke. DESIGN/METHODS: Admission ECGs from 403 consecutive patients admitted for acute stroke symptoms were reviewed. A neuroradiologist reviewed all MRIs/CTs for insular cortex involvement. Favorable clinical outcomes at discharge were defined as mRS of 0-1, discharge NIHSS score <2, and home disposition. Multiple logistic regression was performed for each outcome measure with QTd as the predictor, controlling for age, gender, race, admission NIHSS, history of cardiovascular disease, and history of stroke. A logistic regression analysis determined the association between insular infarct and QTd. Exclusion criteria were: hemorrhagic stroke (n=44), TIA (n=102), ECG unavailable/poor quality (n=75), evidence of atrial fibrillation (n=13), bundle branch block (n=18), ST elevation MI (n=1), pacemaker rhythm (n=5). RESULTS: Of 145 AIS patients in the final analysis, median age = 65 years (IQR 56,75, range 28-94); male 38%, black 68%, median QTd =78 msec (IQR 59,98, range 8-241) median admission NIHSS = 4 (IQR 2,6, range 0-29); 28% had a history of cardiovascular disease; 24% had prior stroke. QTd did not predict clinical outcome for mRS (95%CI 0.99-1.01 p=0.85) NIHSS at discharge (95%CI 0.98-1.01, p =0.30), and discharge disposition, (95%CI 0.99-1.01, p =0.81). Insular cortex involvement did not correlate with QTd magnitude (95%CI 0.99-1.02, p=0.45). CONCLUSIONS: We were unable to demonstrate that QTd is useful in predicting clinical outcome at discharge in AIS.