Article Text

Download PDFPDF
47 A comparison of two algorithms in the prediction of accessory pathway localization in children with wolff-parkinson-white
  1. D Moran1,
  2. W Dewals2,
  3. H Couti3,
  4. C DeAsmundis3,
  5. A Benatar2
  1. 1Vrije Universitaire Brussel, Brussels, Belgium
  2. 2Paediatric Cardiology Department, Vrije Universitair Brusse, Brussels, Belgium
  3. 3Heart Rhythm Management Centre, Vrije Universitair Brusse, Brussels, Belgium


Introduction Several algorithms have been developed to predict the localization of the accessory pathways (AP) in patients with Wolff-Parkinson-White (WPW). The majority of these have focused on accessory pathways in the adult population. Boersma et al. developed an algorithm specifically designed for the paediatric population.

Aim We aim to compare the accuracy of predicting the localization of the AP using the adult algorithm according to d’Avila et al. and the paediatric algorithm by Boersma. Both algorithms are based on QRS polarity.

Method We present our single centre experience of children with WPW. The data was collected retrospectively from April 2007 to November 2015. Patients aged <18 years with documentation of pre-excitation on a 12 lead resting ECG and with eminent evidence of an accessory pathway on electrophysiological study (EPS) were included. Patients with concealed accessory pathways (i.e., retrograde conduction) were excluded. The ECG was analysed blinded from EPS results. The primary outcome concerns an exact match between the predicted localization and site of ablation, but because both algorithms describe different categorization for AP localization, an agreement was made on which localizations could be accepted as a match in a secondary analysis. Since Boersma’s algorithm points out multiple possible localizations, a match was accepted if any one of these sites corresponded with localization on EPS.

Results Inclusion criteria were met for 36 patients. Sixty-nine percent of the patients were boys. Median age was 13 years (4–18 years). The algorithm by Boersma provided an exact match in 29 patients (80%) while the one by d’Avila only provided an exact match in 24 patients (66%). If we expand our match as previously described, the accuracy of the algorithm by d’Avila augments to 75%.

Conclusion The algorithm by Boersma provided a more correct, though less detailed localization of the AP in 80% of this paediatric population. Using the more detailed adult algorithm by d’Avila this result was almost equaled on condition that we accepted some previously agreed sites as a match.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.