Last call on nMARQ™ safety Free

This editorial refers to ‘Safety and efficacy of the nMarq catheter for paroxysmal and persistent atrial fibrillation’ by M. Vurma et al., on pages 1164–1169.

Pulmonary vein isolation (PVI) has evolved into an accepted treatment option for symptomatic paroxysmal and persistent atrial fibrillation (AF) patients. As conventional point-by-point PVI is time consuming and requires considerable levels of manual skills, new types of specific ablation catheters have been developed to facilitate the procedure, such as the cryoballoon and the duty-cycled phased RF ablation catheter (PVAC). In this new, competitive field of ‘single shot devices’, Biosense Webster introduced the nMARQ™, a new circular, irrigated multi-3 mm-electrode catheter. In the very first clinical study on this new ablation system, a worrisome high degree of thermal oesophageal injury (33%, as assessed by endoscopy) was reported when applying 25 and 15 W maximum in unipolar and bipolar mode, respectively.¹ The authors already noted that lower power settings were associated with less oesophageal damage. This correlation between power settings and collateral damage was further substantiated by another European study, showing that thermal oesophageal damage occurred in three out of six patients treated with 20 W unipolar and 10 W bipolar ablation for maximum 60 s, compared with 2 out of 15 patients treated with 15 W unipolar and 10 W bipolar ablation for maximum 30 s.^2,3 After two fatalities, one oesophago-pericardial fistula and one sepsis, in the only multi-centre study in 374 AF patients, the manufacturer adjusted the recommended upper unipolar power limit to 20 W.⁴ An even more prudent approach by Burri et al.⁵ treating 50 AF patients with 15 W was associated with a very high AF-recurrence rate of 54% after 15 months as well as with a still relatively high degree of procedural complications (one case of phrenic nerve palsy and two cases pericardial effusion).

In the case of nMARQ™, the window between the desired therapeutic effect of PVI and the well-known procedural risks seems very small or even non-existent in comparison to other techniques.⁶ This notion should now be considered a fact after the paper by Vurma et al.⁷ published in this issue of EP-Europace. The authors are to be applauded for resolving this safety issue and, more importantly, preventing further potentially fatal treatment of AF patients, as the catheter now is recalled. In this prospective registry of 327 patients treated by a single, experienced operator using conservative, recommended power settings were included; 37 patients were treated with the second-generation nMARQ™ catheter introduced in February 2015. Despite proton-pump inhibitors and short duration, low power ablation at the posterior wall, two patients developed atrio-oesophageal fistulas and died. As these patients were treated with the second-generation catheter, this can be expressed as a 5.4% incidence of fistula in this subset of patients. The study came to an immediate halt after these two fatalities.

The atrio-oesophageal fistula is the most feared complication of PVI due to a nearly 100% fatality rate. Fortunately, its incidence generally is low, with five fistulas reported among 45 115 procedures included in a large-scale survey from 546 centres worldwide.⁸ Comparing these figures to the published percentage of fistulas in patients treated with nMARQ™, the manufacturer rightfully recalled this catheter.

Looking back at all studies on the nMARQ™ catheter, the obvious question is whether fatalities could have been prevented. Procedural faults are to be excluded as, like in this study by Vurma et al.,⁷ recommended power setting were used, operator experience was high, and other preventive measures such as proton-pump inhibitors were taken. The true issue here relates to the device itself and the way it has been launched in routine care for AF patients. The nMARQ™ catheter essentially is a completely new invasive technology, which was introduced in many electrophysiology labs several years ago. Worrisome results from several, small single-centre studies were published shortly after its introduction, but only after the fatal fistulas in the sole multi-centre trial by Mahida et al.,⁴ adjustments to the technology were made.

It is appropriate to reflect on the optimal way to introduce new and potentially harmful, invasive technology for several reasons. Preventing unnecessary fatalities due to immature technology as well as maintaining our credibility in the eyes of society and regulatory authorities, in particular, should come to mind after looking back at the body of evidence supporting safe introduction of the nMARQ™ catheter. Fragmented and uncoordinated introduction of new and potentially harmful technology can not only lead to avoidable risks, it might also lead to overlooking advantages of this technology due to, for example, repetitive learning curves in many centres, or poor collection of substantial data, as reflected in a multiplicity of smaller studies.

Recommendations on the design of trials on treatment of AF patients published in the 2012 HRS/EHRA/ECAS expert consensus report have importantly contributed to better structuring AF research.⁹ The accumulation of cautioning results of several, separate studies, peaking in the present study by Vurma et al.,⁷ should bring the international cardiological community to consider including recommendations on introducing new, potentially harmful technology in routine care in the next consensus report, especially if a solid scientific basis with appropriate clinical trials is lacking. A coordinated approach by means of shared protocols and data collection, open to a limited number of selected centres for the initial introduction, will help us in rightfully maintaining the innovative nature of cardiology.

Conflict of interest: none declared.

References

Author notes