A randomized prospective mechanistic cardiac magnetic resonance study correlating catheter stability, late gadolinium enhancement and 3 year clinical outcomes in robotically assisted vs. standard catheter ablation

Abstract

Aims: To prospectively compare cardiac magnetic resonance late gadolinium enhancement (LGE) findings created by standard vs. robotically assisted catheter ablation lesions and correlate these with clinical outcomes.

Methods and results: Forty paroxysmal atrial fibrillation patients (mean age 54 ± 13.8 years) undergoing first left atrial ablation were randomized to either robotic-assisted navigation (Hansen Sensei(®) X) or standard navigation. Pre-procedural, acute (24 h post-procedure) and late (beyond 3 months) scans were performed with LGE and T2W imaging sequences and percentage circumferential enhancement around the pulmonary vein (PV) antra were quantified. Baseline pre-procedural enhancements were similar in both groups. On acute imaging, mean % encirclements by LGE and T2W signal were 72% and 80% in the robotic group vs. 60% (P = 0.002) and 76%(P = 0.45) for standard ablation. On late imaging, the T2W signal resolved to baseline in both groups. Late gadolinium enhancement remained the predominant signal with 56% encirclement in the robotic group vs. 45% in the standard group (P = 0.04). At 6 months follow-up, arrhythmia-free patients had an almost similar mean LGE encirclement (robotic 64%, standard 60%, P = 0.45) but in recurrences, LGE was higher in the robotic group (43% vs. 30%, P = 0.001). At mean 3 years follow-up, 1.3 procedures were performed in the robotic group compared with 1.9 (P < 0.001) in the standard to achieve a success rate of 80% vs. 75%.

Conclusion: Robotically assisted ablation results in greater LGE around the PV antrum. Effective lesions created through improved catheter stability and contact force during initial treatment may have a role in reducing subsequent re-do procedures.

Keywords: Atrial fibrillation; Cardiac magnetic resonance imaging; Catheter ablation; Long-term follow-up; Pulmonary vein isolation; Robotic-assisted.

© The Author 2015. Published by Oxford University Press on behalf of the European Society of Cardiology.

Figures

Figure 1

Flow chart outlining number of times procedure performed, outcome at each stage and the use of anti-arrhythmic drugs. Both groups had 20 patients each undergoing first ablation procedure. In the robotic group, seven patients underwent a second procedure while three patients did not undergo repeat procedures and one patient required a pacemaker. In the standard group, 13 patients underwent a second procedure, four patients a third procedure, and one patient undergoing a fourth procedure. Three patients declined further procedures and one patient underwent a pacemaker implantation.

Figure 2

(A) Cumulative number of arrhythmia free patients over a 3 years mean follow-up period. (B) Anti-arrhythmic use pre- and post-ablation in both groups at follow-up.

Figure 3

Serial T2W (A) and LGE (B) CMR scans performed on a single patient following robotic and standard navigated catheter ablation at 3 time points. In the T2W series, both groups demonstrate an increase in signal intensity and atrial wall thickness following ablation which resolves on the late follow-up scans. In the LGE series (B), baseline images in the first column show no significant DE signal (tissue injury/necrosis) compared with acute post-ablation images in the second column. The late scans in the third column shows that areas of LGE signal become less diffuse and more defined with sharper borders in comparison to the acute scans.

Figure 4

Example of a series of 3-D LA reconstructed shells (at 3 time points) in two patients to compare robotic (A) vs. standard catheter ablation (B). Blue represents areas of T2 signal while red represents LGE. More ‘islands' of red are seen in the robotic assisted LA shell implying greater tissue injury achieved. The first row represents the raw data transferred onto the atrial shell, each for T2 and LGE (Row A). The second row represents the semi-automated quantification of T2 and LGE signals (Row B). The third row demonstrates the combination of T2 and LGE from the second row onto a single atrial shell (Row C).

Figure 5

(A) Scatter-boxplot shows a comparison of pre, acute and late T2, LGE and combined T2&LGE for both robotic (green) and standard (brown) ablation. Each individual scatter plot represents the raw data for that specific group. The dots within each group have been dispersed horizontally to optimize visualization and clarity. The boxplots on the other hand represent median (red line), 95% confidence intervals (pink box) and 1 standard deviation (green/brown box). An overall higher enhancement is seen post-procedure in the robotic group compared with the standard group. The higher percentage encirclements by LGE assessment in the robotic group continued to remain significantly more in the follow-up scans. The overall lower standard deviation in the robotic group suggests better consistency in creating tissue injury. This is most likely a function of catheter stability resulting in better tissue contact. (B) Overall, a higher percentage encirclement is noted in the robotic group with statistical significance achieved in the recurrences cohort. This may have potential consequences for re-do ablation procedures.

Figure 6

Scatter plot describes the relationship between energy delivered and scar created as quantified on LGE–CMR by percentage PV encirclement. Both robotic and standard ablation datasets have been categorized into no recurrences vs. recurrences. The blue circles representing robotic procedures appear less dispersed and have a minimum LGE value of around 40%. The red circles representing standard catheter ablation are widely dispersed and have a minimum LGE value of 15% to 30% despite high total energy delivery. This suggests better catheter–tissue contact conferred by the robotic system yielding higher percentages of PV antrum scar.