EFFECTS OF FILTERING ON RIGHT VENTRICULAR ELECTROGRAMS RECORDED FROM ENDOCARDIAL CATHETERS IN HUMANS

Criteria for identifying electrically abnormal myocardium based on analysis of electrogram, amplitude, duration, and morphology have recently been proposed as an aid to catheter mapping in humans. Electrogram characteristics are likely to be influenced by analog filtering and the distance between recording electrodes. These parameters can be varied and differ among electrophysiol‐ogy laboratories. To determine the magnitude of the effect of low pass and high pass filtering and intere‐lectrode distance on human bipolar right ventricular electrograms eight patients were studied. The signal from a single bipolar electrode pair was recorded simultaneously on five channels with variable filtering. As the high pass filter frequency increased above 10 Hz, the electrogram duration and amplitude decreased. The decrease in amplitude was described by a monoexponential function with a 30% decrease in amplitude at a filter setting of 30 Hz and a 69% decrease at a filter setting of 100 Hz. As the low pass filter frequency decreased from 2,500 to 250 Hz, electrogram characteristics were not significantly altered. Low pass filtering at 300 Hz decreased electrogram amplitude slightly. Neither high nor low pass filter settings or interelectrode distance affected electrogram timing. However, filtering could markedly alter electro‐gram morphology, introducing the possibility for errors in visual assessment of electrograms. Electrogram duration increased linearly with increasing distance between recording electrodes. The effects of filtering on electrograms were similar regardless of interelectrode spacing. Thus, varying the high pass filter setting over a range that is used clinically (from 1 to 100 Hz) can substantially alter electrogram duration, amplitude, and morphology. Low pass filtering at 250 Hz and above had no effect on these electrograms although it is possible that alterations would be observed in signals from areas of slow conduction that we did not study. The filter setting and interelectrode distance must be considered when applying proposed criteria for distinguishing normal from abnormal electrograms. Copyright © 1990, Wiley Blackwell. All rights reserved