Discussion
The main findings of these investigations into RMA use during automated
RF annotation at the LAPW are:
At sites of deliberate catheter motion between first and second
auto-annotated sites effected without loss of catheter-tissue contact,
RMA “off” demonstrated suitable accuracy towards catheter motion
detection in all cases.
Conversely, RMA “on” resulted in delayed identification of true
catheter tip motion, resulting in delayed auto-annotation of site
1-to-2 transition and important error in per-site RF parameters. Put
simply, RMA “on” effectively renders automated RF annotation
“blind” to the immediate occurrence of clinically important catheter
displacement events.
These results may be better understood when considering the VISITAG™
Module annotation “system logic” and how this is modified by RMA use.
Briefly, the ablation catheter tip position is measured in a “rolling
window” of 60 sites per second (i.e. intervals of 16/17ms), from which
is calculated the standard deviation (SD). With RMA (i.e. ACCURESP™)
“off”, RF auto-annotation occurs when both every 1s of position data
is within twice (2x) the user-defined position SD, and a total
consecutive minimum of 3s is within (1x) the SD. However, with RMA
“on”, the position stability filter operates over a minimum of two
respiratory cycles, using position data “adjusted” to end-expiration.
In this situation, automated RF annotation occurs when these position
stability targets are met following data adjustment . Importantly,
annotation only occurs when CF filter preferences are also satisfied: in
contrast to position stability filtering, the CF filter is applied
continuously regardless of RMA setting (i.e. independent of the
respiratory cycle). Therefore, with the force-over-time 100% minimum 1g
CF filter used during this present report, RF annotation “end” logic
is fulfilled by any 0g CF event at any stage in the respiratory
cycle – i.e. regardless of RMA setting.