Robotic catheter ablation required less fluoroscopy time and appeared to reduce operator fatigue compared with manual procedures in a randomized trial. But a comparable success rate and technical problems may make uptake a rough slog.
Waqas Ullah, MBBS, of the cardiovascular biomedical research group at St. Bartholomew’s Hospital in London, and colleagues initially designed a superiority trial to compare robotic catheter ablation (Sensei X, Hansen Medical) with manual catheter ablation as a single-procedure treatment for atrial fibrillation. They terminated the study early when an interim analysis showed no difference between the groups.
At that point, they had randomized 166 patients between 2008 and 2012. Of those, physicians manually ablated 78 patients and robotically ablated 79 patients. The ablation protocol was the same for both groups.
Eleven of the robotic ablation cases crossed over to manual, with seven due to system failures and two to usage difficulties. The single-procedure success rate at 12 months, which was the primary endpoint, was not significantly different, at 33 percent in the robotic group and 24 percent in the manual group. They found no difference in the combined rate for major and minor complications.
One patient in each group died, which they wrote was “surprising and not representative of our experiences of ablation outside the trial.”
Procedure time was longer in the robotic group compared with the manual group (289 vs. 273 minutes) but fluoroscopy time was shorter (46 vs. 50 minutes). The radiation dose was lower in the robotic group, at 3,354 vs. 4,136 Gy*cm2. Procedure and fluoroscopy times in each operator’s first 10 robotic procedures were longer, reflecting a likely learning curve.
Operator fatigue, measured on a five-point scale with five being very fatigued as reported by physicians, was three for manual for two for robotic ablations. Operators also reported a lower number of catheter displacements with the robotic approach.
“Of note, the death that occurred in the RRN [robotic] arm was primarily due to atrial perforation (with concomitant retroperitoneal bleed) in a patient within the first 10 patients of an operator’s learning curve,” Ullah et al wrote. They emphasized that their work and other studies have not shown a relationship between a learning curve and complication rates but “increased vigilance at the start of one’s RRN experience seems prudent.”
Their study was underpowered to demonstrate either superiority or noninferiority, they wrote, and an ongoing noninferiority trial that avoids the learning curve issue and uses contemporary equipment and methods may overcome the limitations in their study.
“Although there appear certain advantages to the RRN system, the crossover rate and lack of outcomes benefits are important considerations for the uptake of this technology,” they cautioned.