Good afternoon ladies and gentlemen. I'm very grateful to the organisers for inviting me to present in this session on mapping atrial fibrillation in 2020. My talk is on non-contact mapping and my disclosure at the bottom. My name is Andrew Grace and I'm from the University of Cambridge and I also work at the Royal Papworth Hospital in my clinical role. I'm first going to consider the general issues attached to non-contact mapping and the case of need to develop this as a system as we are at this point in time to improve the outcomes in those undergoing atrial fibrillation ablation. I'll then go on to the concepts that underpin this in terms of the physics and seeking down the sources. I'll then give some evidence of efficacy and utility of this mode of mapping in terms of eradication of source and then I will conclude. In terms of non-contact mapping it's got a long and storied history. An argument one could easily make is that the first non-contact electrical recordings relative relevant to our field resulted in the electrocardiogram and the electrocardiogram I think it is safe to say underpins the whole of cardiology. The work that was done in the 19th century by some outstanding physicists and medics interested in physics culminated in the work of Gabriel Lippmann 1872 is a capillary electrometer that could resolve bioelectrical signals through the interaction of mercury and sulfuric acid. This allowed Augustus Waller to record the electrocardiogram and Eindhoven to further develop that through the string galvanometer. These are non-contact electrical recordings obtained from the heart. In general cardiologists do not immediately grasp the idea that non-contact intercardiac electrograms are self-evident because of course much of our field of cardiac electrophysiology has been based upon contact recordings but again going back almost 35 years now Brunette Cardi working out of Turin initially a person with a great interest in electrocardiograms in general took it upon himself with his group to measure intracardiac electrocardiograms. If you like this is in a open chest canine on the left an olive-shaped probe placed within the chamber the agenda was to really provide capability for mapping ventricular tachycardia. So in this left ventricular preparations these electrodes are placed through the wall from the endo-epicardium rather to the endocardium with epicardial, mid-myocardial and endocardial locations for recording and then electrograms were recorded from this probe placed within the chamber. These the sickle early spatial negativity was recorded the relatively low amplitude inter-cavitory potentials and resolutions restricted to about 1.5 centimeters so of course this would not be suitable for mapping complex arrhythmias such as atrial or ventricular fibrillation. The trigger point I think to thinking about moving non-contact mapping into the modern space came from the requirement for seeking out non-pulmonary vein triggers. The system the insight system based on initially to Cardi's work that had been employed had all sorts of issues attached to it and has never really picked off picked up in in the clinical domain in a way that one would have would have thought it might have been developed certainly provided many insights into ventricular tachycardia and similar and this is the wonderful work of Richard Schilling. In terms of the case of need for non-contact mapping came from work from Sanjeev I think mainly as a trigger using a contact mapping system that a pair of system ultimately the basket placed in the left and the right atria problem is the practicalities these reasonably plausible maps but the problem is sequential acquisition is required and one sees the gaps between the electrodes placed in this complex three-dimensional chambers required interpolation in order to fill in these spaces also picking up much in where far-field noise and other artifact and overall the physical basis is such a proposition of contact mapping is not ideally placed within a more global domain of electromagnetic field theory. So Cutis Medical have set out on this scheme going back something like 10 years and I've been involved with them more or less from the outset to develop a non-contact mapping tool for macroscopic source detection to get over the problems that arose from the contact systems and to get a global view of these complex arrhythmias moving in multiple directions that are very unstable so that mapping could be achieved at a high level of resolution and the sort of levels of resolution now I think supported by our data is down to one or two millimeters so this is pretty spectacular. This is the cathode many of you will be familiar with this now it's about two and a half centimeters across it's placed in the chamber of interest it has 48 ultrasound transducers and 48 unipolar electrodes records large amounts of data and of most interest to us 150,000 endocardiac unipolar voltage samples are taken every second. The pole physical basis of the system the thing that differentiates I think from anything else out there is the calculation of charge density. The charges are a fundamental property of matter along with things like mass so physicists are very familiar with these as I'm sure you would have been when you were in college and similar but something that we've moved away from in general in terms of cardiologists but charge density calculation has now enabled reliable and effective non-contact mapping and the idea is that one wants to get down to the continuous charge layer that constitutes the activation wave moving through the myocardium that's the positive and negative charge across the cell membranes and the wave of activation moving it along as this as this charged layer in the myocardium and the point is that this is much more compact this layer than the voltage capacitive polarization field of the voltage field if you like that emerges from that whether that's measured within the chamber whether it's measured by the electrocardiogram or whether it's measured through a contact mapping approach a contact mapping approach with a bipolar electrode for example will simply record aspects of the voltage field now in order to calculate this the charge layer one needs excellent accurate distance measurements between the probe and the endocardial shell if you like and that's achieved by ultrasound the ultrasound localized as a surface it also has practical value of minimizing motion errors providing this very wonderful set of images that map precisely onto CT images and MRI images that have been obtained from patients beforehand and this provides a platform also for ablation but the most important aspect in terms of the way the system works is to get this accurate distance for the calculation of the charge density and that's fully explained in our paper that appeared later in terms of the calculations again this is not some newfangled set of mathematics if you like it's all based on Poisson you know going back to the early part of the 19th century that relates potential to charge so one essentially samples the voltage and then one can resolve the local source and exclude the sum of the distant sources and that results if you like in this sort of burnt orange under here a much sharper signal greater sensitivity and specificity in terms of the electrical outputs if you like the charge sources rather than the elect the voltage outputs and much more resolved so the processes of acquiring the electrograms hundred and fifty thousand every second solving for charge density at fixed times across the endocardial layer and then generate an activation and this is a sort of typical sort of image this shows a rotational activation across the posterior wall a number of patterns have been observed using this technology is something like 1,200 patients I think now have undergone investigation there were a number of trials dramatic SVT was to set up the basic functionality and safety of the system we presented that at the Art Association in 2016 but this a series of patterns have been observed focal patterns localized rotational activation and localized irregular activation again these are now being sub-segmented in terms of subgroups as more observations have been made but the important thing I think to say is that all this fits into the concept conventional conceptual models of atrial fibrillation that go back you know hundred years something of this order and you stand a tell beautifully summarized those in nature in 2002 the single focus idea which would encompass Hasiga's work coming from the palmy vein or from the right atrium initially to these the circus movement activity that Sanji has observed and also multiple re-entry as observed by such excellent workers as Moritz Alessi and Jimmy Cox and so this is not against some newfangled but I think it's reasonable to say that I feel that we've obtained the first full chamber non perturbed maps of atrial fibrillation that have so far been obtained so are these maps any good any use if you like for to move forward the field and obviously fix our patients I mean the mechanistic insights I think are very important from my own personal perspective I think those are possibly of the most relevance but in terms the immediacy of required to improve ablation approaches studies are ongoing this is a published study now this is the uncover AF trial led by Stefan Williams published in circulation circulation arrhythmia EP last year 127 patients were published typical sort of AF but these are proper full-on persistent AF cases the first onset being something like three years beforehand and being persistent for a couple of years and although the duration and the total time in AF was less than 12 months they'd had it a good deal of time they tended to have good left ventricular function this was an adaptive therapy protocol the process of placing the catheter within the chambers was pretty standard the first of all build up the ultrasound anatomy of the chamber then a map of atrial fibrillation was obtained in all patients and recorded and then PVI was performed as per in the individual institutions usual protocol after PVI then remapping occurred and new sites were observed about four maps per case were completed the great thing about this system again you've got the anatomical structure but actually the electrical map can be remapped extremely rapidly so you can immediately you know go back and see what it is that you've achieved with a particular set of deliveries of ablation lesions and the concept is either to sort of fragment and if you like pick out components of a central core but then linking that to some to for example anchoring to the PVI site that's in the nearest locality the Outcomes were pretty good. This is comparing the Uncover AF data to the STAR-AF outcomes and Atul Verma was a co-author also of this paper and the freedom from greater 30 seconds of AF after multiple procedure in Uncover was 93% and obviously somewhat superior than what's reported in STAR-AF. Similarly, arrhythmia burden is massively down. The percentage of patients with AF greater than 30 seconds was trivial after single procedures and reduced further with more procedures. One of the things I think of most interest in terms of the idea of the importance of non-PVI triggers is that the more targets were ablated in addition to PVI and the more pattern types then you see there's an odds ratio of impact and success 9.39 if greater than three targets were ablated in addition to PVI when compared to other groups. So this is I think supports the concept of the observation and the accurate delineation of these alternative sites and that then their resolution through ablation can provide added value over and above simple PVI. So in conclusion in terms of contemporary non-contact mapping the calculation of charge density is basically allows this to occur in the absence of that charge density calculation then the voltage can give you a non-contact map but doesn't give you the resolution that can be obtained through charge density which is a four-fold increment. The reason that there's such value is that this charge layer is the true source of the cardiac electrical field but the actual maps that are obtained are plausible, they're actionable and they fit into the whole literature of atrial fibrillation as it's emerged over this long period of time. From a more general perspective looking forward I think this contemporary non-contact mapping is a form of deep phenotype and it allows us to map on to genetic and genomic information. I mean my particular route of attack is going to be through single cell biology in my own group and thank you for your attention.

non-contact mapping

atrial fibrillation

electrocardiogram

CathoMAP

charge density