Localization of accessory pathways in the Wolff-Parkinson-White pattern--physician versus computer interpretation of the same algorithm

Abstract

Background: There are several published algorithms for the prediction of accessory pathway (AP) location in the Wolff-Parkinson-White syndrome from the 12-lead electrocardiogram (ECG). Most depend on stepwise criteria, and minor disagreements between observers over QRS transition point or delta wave axis may lead to different classification of pathway location. We compared the utility of a computerized program in identifying pathway location from the ECG using the algorithm published by Fitzpatrick and coworkers(3) against physician assessment with the same algorithm.

Methods: Thirty-one 12-lead ECGs with an overt preexcitation pattern were examined by three physicians and AP localized to one of eight anatomical sites using the Fitzpatrick algorithm, with disagreements resolved by consensus. Similarly, pathway location was determined by the Glasgow ECG program with the Fitzpatrick algorithm incorporated into its logic.

Results: The agreement between each physician and their consensus was 28/31, 29/31, and 29/31. Similarly, assessment by the Glasgow program produced agreement with the physician consensus in 29/31 cases. Of the 24 patients who underwent radiofrequency ablation, the program localized the pathway to the true or adjacent annular region in 20, compared to 20/24 by physician assessment of the algorithm, producing a similar predictive accuracy to published data.

Conclusion: This study has shown that incorporation of the Fitzpatrick algorithm for AP location into a widely used computer program results in the same level of performance as that of experienced physicians and may be useful in clinical practice as an aid to referral for electrophysiological study and ablation.