Results
Twenty-five patients underwent first-time PVI as described, between November 2016 and May 2017: 13 persistent AF, 12 PAF; 19 male (76%); mean age 57 [SD: 14] years and mean CHA2DS2-VASc score 1.3 [SD: 1.3]. Complete PVI was achieved in all without spontaneous / dormant recovery of PV conduction, following mean 16.2 [SD: 3.1] minutes of RF, without procedural complications. The respiratory motion-triggering cohort comprised 8 cases (32%); considering age, body mass index and RF duration required for case completion, there were no significant differences between cohorts with and without respiratory motion threshold triggering.
Comparing ACCURESP™ RMA “on” versus “off”, the number of auto-annotated LAPW sites and total LAPW RF duration were 82 and 98, and 1091s and 1006s, respectively. Annotated RF data according to RMA setting are shown in table 1. For each group (i.e. left or right-sided), per-site RF duration and FTI were significantly greater with RMA “on” versus “off”: i.e. left-sided mean RF duration 13.1s versus 9.9s (p=0.0003) and median FTI 156g.s versus 114g.s (p=0.0003), respectively; right-sided mean RF duration 13.5s versus 10.6s (p=0.006) and median FTI 228g.s versus 166g.s (p=0.04), respectively. Analysis combining left and right sides also demonstrated significantly greater mean ITD with RMA “on”; i.e. 6.0mm versus 4.8mm (p=0.002).
Table 2 shows comparisons of annotated RF data at site-1 according to RMA setting: At the left-side and comparing ACCURESP™ RMA “on” versus “off” the site 1-to-2 ITD (6.6mm versus 5.3mm, p=0.07), RF duration (16.0s versus 15.1s, p=0.16) and FTI (185g.s versus 163g.s, p=0.33) were greater with RMA “on”, but without statistical significance. At the right side the site 1-to-2 ITD was greater with RMA “on” (7.2mm versus 5.0mm, p=0.13), but without statistical significance. Combined data analysis demonstrated that auto-annotated site 1-to-2 ITD was significantly greater with RMA “on” versus “off” (i.e. 6.7mm versus 5.2mm, p=0.02), while the difference in site-1 RF duration (15.7s versus 15.1s, p=0.09) and FTI (240g.s versus 198g.s, p=0.38) was not statistically significant.
Analyses at sites of deliberate catheter motion: Auto-annotated site 1-to-2 transitions
Comparing auto-annotated data (RMA “on” minus “off”) at 3 site 1-to-2 transitions concurrent with loss of tissue contact (i.e. 0g CF), the maximum difference in RF duration, FTI and ITD was -0.6s, -17g.s and 2.2mm respectively, with no difference in impedance drop (table 3 and figure 1, plus supplementary figures 1-2).
Four site 1-to-2 transitions were effected with constant catheter-tissue contact and associated with UE morphology change from pure R (site-1 completion) to RS (site-2 onset). In this group, the maximum difference (RMA “on” minus “off”) in RF duration, FTI, ITD and impedance drop was 11.3s, 139g.s, 2.6mm and 3.3Ω respectively (table 3), with the first indication of catheter movement represented via RMA “off” annotation in all cases (figure 2 and supplementary figures 3-5). The greatest difference occurred when site 1-to-2 ITD with RMA “off” was 4.1mm (figure 2): at 15.2s following RF onset there was an abrupt increase in catheter position shift and SD, with a corresponding change in CF waveform indicating deliberate catheter motion “per protocol” at 15s, however while annotated site 1-to-2 transition according to RMA “off” coincided with these changes (blue vertical line), the RMA “on” timing of annotated site 1-to-2 transition occurred 11.3s later (red vertical line).
The remaining 9 deliberate site 1-to-2 transitions during constant catheter-tissue contact were associated with continuous pure R UE morphology at both site-1 end and site-2 onset. Comparison of annotated data (RMA “on” minus “off”) demonstrated a difference in site-1 RF duration >1s in 8; range -1.3 – 8.6s, mean 3.7 [SD: 4.0] s (supplementary table 1). In this group, maximal differences in site-1 annotated RF duration, FTI, ILD and impedance drop were 8.6s, 208g.s, 7.7mm and 1.4Ω respectively, with RMA “on” resulting in greater values for annotated data in 7/9 transitions. When considering multiple measures of catheter position stability, the appropriate indication of deliberate catheter motion occurred with RMA “off” in all 9 transitions; 8/9 demonstrated ≥1s difference in annotation timing (supplementary figure 13).
For all 13 deliberate site 1-to-2 transitions achieved with constant catheter-tissue contact, the relationship between differences in annotated RF data (RMA “on” minus “off”) and ITD (with RMA “off”) is shown in figure 3. There was a strong negative correlation between the difference in annotated RF duration and ITD – Pearson r -0.68 (95% confidence interval (CI) -0.91 to -0.13, p=0.02). Consequently, while the maximum difference in annotated RF duration with ≥6mm site 1-to-2 ITD was 1.1s, ≤5mm ITD was associated with maximal difference in annotated RF duration of 11.3s. There was a moderate negative correlation between the difference in annotated FTI and ITD – Pearson r -0.47 (95% CI -0.81 to 0.11, p=0.10) and a moderate negative correlation between the difference in impedance drop and ITD – Pearson r -0.53 (95% CI -0.84 to 0.05, p=0.07).
Data supplement figures 7 – 15 demonstrate remaining annotated site 1-to-2 transitions, with corresponding position shift, SD, CF and impedance data.