Published Online
on October 19, 2009

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Google Scholar Articles by Elvan, A. Articles by deBakker, J. M.T. PubMed PubMed Citation Articles by Elvan, A. Articles by deBakker, J. M.T. Related Collections Arrhythmias, clinical electrophysiology, drugs

Original Article

Dominant Frequency of Atrial Fibrillation Correlates Poorly with Atrial Fibrillation Cycle Length

Arif Elvan^1,3; Andre C. Linnenbank²; Marnix W. vanBemmel¹; Anand R. Ramdat Misier¹; Peter Paul H.M. Delnoy¹; Willem P. Beukema¹ and Jacques M.T. deBakker²

¹ Isala Klinieken, Zwolle, The Netherlands;
² Academic Medical Centre, Amsterdam & The Interuniv Cardio Inst, Utrecht, The Netherlands

^* Corresponding author; email: a.elvan{at}isala.nl

Background—Localized sites of high frequency during atrial fibrillation (AF) are used as target sites to eliminate AF. Spectral analysis is used experimentally to determine these sites. Purpose of this study was to compare dominant frequencies (DF) with AF cycle length (AFCL) of unipolar and bipolar recordings.

Methods and Results—Left (LA) and right atrial (RA) endocardial electrograms were recorded during AF in 40 patients with lone AF, using two 20 polar catheters. Mean age was 53±9.9 years. Unipolar and bipolar electrograms were recorded simultaneously during 16s at 2 RA and 4 LA sites. AFCLs and DFs were determined. QRS subtraction was performed in unipolar signals. DFs were compared with mean, median and mode of AFCLs. 4800 unipolar and 2400 bipolar electrograms were analysed. Intraclass correlation was poor for all spectral analysis protocols. Best correlation was accomplished with DFs from unipolar electrograms compared to median AFCL (Intraclass correlation coefficient, ICC = 0.67). A gradient in median AFCL of >25% was detected in 16/40 patients. In 13 out of 16 patients (81%) with a frequency gradient of >25%, the site with highest frequency was located in the LA (posterior LA in 8 patients). The site with shortest median AFCL and highest DF corresponded in 25% if unipolar, and in 31% if bipolar electrograms were analysed.

Conclusions—DFs from unipolar and bipolar electrograms recorded during AF correlated poorly with mean, median and mode AFCL. If a frequency gradient >25% existed, the site with highest DF corresponded to the site of shortest median AFCL in only 25% of patients. Since spectral analysis is being used to identify ablation sites, these data may have important clinical implications.

Key Words: atrial fibrillation • dominant frequency • spectral analysis