Poor reliability of P-wave terminal force V 1 in ischemic stroke.

INTRODUCTION: Several ECG markers are postulated to represent underlying atrial remodelling and have been associated with ischemic stroke. P-wave terminal force in lead V1 (PTFV1 ) is one such marker. We examined the factors that contribute to the reliability of PTFV1 and its association with ischemic stroke.

MATERIAL AND METHODS: Four hundred and thirty-five patients that presented with an ischemic stroke or transient ischemic attack (TIA) were identified through a prospectively maintained multi-site institutional stroke database. Control group consisted of age matched patients without prior history of an ischemic stroke or TIA. All patients underwent a 12-lead ECG and 24-hour Holter monitoring during the study period to exclude atrial fibrillation.

RESULTS: Morphology consistent with PTFV1 occurred commonly in both the stroke/TIA and control groups. There was no significant difference in the median PTFV1 value between the stroke 3.96 mV ms [Interquartile range (IQR) 2.78-5.58] and control 4.23 mV ms [IQR 2.91-5.57] groups. Measurements of PTFV1 demonstrated excellent intra-observer reliability on assessment of the same P-wave (Intra class correlation (ICC) 0.91, p < 0.001) with narrow limits of agreement 2.21 to -2.95 mV ms. A change in the P wave assessed led to a significant reduction in reliability (ICC 0.79, p < 0.001). Inter-observer, inter P-wave assessment demonstrated further reduction in reliability (ICC 0.68, p < 0.002) with wide limits of agreement 6.17 to -5.78 mV ms, indicating significant under and overestimation of PTFV1 .

CONCLUSION: The utility of PTFV1 as a clinical marker for ischemic stroke is limited by the reduction in reliability associated with inter-observer and inter P-wave measurements.