Background During automated radiofrequency (RF) annotation-guided pulmonary vein isolation (PVI), respiratory motion adjustment (RMA) is recommended, yet lacks in vivo validation. Methods Following contact force (CF) PVI (continuous RF, 30W) using general anaesthesia and automated RF annotation-guidance (VISITAG™: force-over-time 100% minimum 1g; 2mm position stability; ACCURESP™ RMA “off”) in 25 patients, we retrospectively examined RMA settings “on” versus “off” at the left atrial posterior wall (LAPW). Results Respiratory motion detection occurred in 8, permitting offline retrospective comparison of RMA settings. Significant differences in LAPW RF auto-annotation occurred according to RMA setting, with curves displaying catheter position, CF and impedance data indicating “best-fit” for catheter motion detection using RMA “off”. Comparing RMA “on” versus “off”, respectively: Total annotated sites 82 versus 98; median RF duration per-site 13.3s versus 10.6s (p<0.0001); median force time integral 177g.s versus 130g.s (p=0.0002); mean inter-tag distance (ITD) 6.0mm versus 4.8mm (p=0.002). Considering LAPW annotated site 1-to-2 transitions resulting from deliberate catheter movement, 3 concurrent with inadvertent 0g CF demonstrated <0.6s difference in RF duration. However, 13 deliberate catheter movements during constant tissue contact (ITD range 2.1 – 7.0mm) demonstrated (mean) site-1 RF duration difference 3.7s (range: -1.3 to 11.3s): considering multiple measures of catheter position instability, the appropriate indication of deliberate catheter motion occurred with RMA “off” in all. Conclusions ACCURESP™ respiratory motion adjustment importantly delayed the identification of deliberate and clinically relevant catheter motion during LAPW RF delivery, rendering auto-annotated RF display invalid. Operators seeking greater accuracy during auto-annotated RF delivery should avoid RMA use.