Three-Dimensional Mapping in Interventional Electrophysiology: Techniques and Technology

Douglas L. Packer, M.D.


J Cardiovasc Electrophysiol. 2005;16(10):1110-1116. 

In This Article

General Requirements of Cardiac Mapping Systems

At a minimum, a computer-based cardiac mapping system should be able to (1) accurately replicate the cardiac anatomy underlying an arrhythmia; (2) provide a plausible representation of activation of that chamber, as linked to the specific anatomic site of data acquisition; (3) readily capture and intelligibly display other details of physiology; and (4) catalogue the site of interventions.

The first requirement provides the context for the arrhythmia. A mapping system should faithfully replicate the anatomy of a chamber under examination and those structures both entering and exiting that chamber. In short, the geometry needs to look like the chamber under study. The extent to which this is accomplished is a matter of both "man vs machine." The resolution of anatomy is a function of the number of points taken in the process of creating the surrogate geometry. The more points, the closer the image rendering comes to replicating the chamber under evaluation. This is an operator issue. The problem from the mapping system side comes from the algorithms used to graphically connect three or more points sampled by a roving catheter to create a surface segment and subsequent volume.

While interpolation between points along an uncomplicated surface readily produces a clear image of that surface, the process is strained at the junction between chambers and sites of entering or exiting "veins and valves," or at areas of complex structures. The system must easily preserve the complexities of anatomy at those points, along with intervening acute and oblique angles between structures, without losing requisite detail. Some systems are more prone to "interpolation obliteration" of those junctions with smoothing over the angles defining the underlying structures. One way to minimize this is to treat multiple veins or neighboring structures as separate volumes or maps, which some systems readily allow.

Another challenge of mapping systems is created by the inherent difficulty in displaying three-dimensional (3D) structures on a monitor screen in two-dimensional views. Here, the ability to show multiple views simultaneously is of paramount importance to give the 3D perspective. The addition of virtual endoscopic views from within a chamber is enormously helpful. Transparency features incorporated into the system can be useful or create confusion. Other optimal features are as outlined in Table 1 . These challenges of anatomic rendering and display are not a trivial matter, since, under the best of circumstances these features allow the user to clearly understand and navigate the underlying geometry. Under the worst of circumstances, bad geometry may be dangerous. It should be noted that the accuracy of the surrogate geometries would become increasingly important, as interventions are guided by those very images. Remarkably, very few validation studies addressing these issues in any mapping system are available. Additional investigations will therefore be required to insure that the surrogate mapping geometries accurately depict actual anatomy.

A second function of a mapping system is to catalogue local physiology as linked to the anatomic site of data acquisition. The system should readily "arrange" sequential sampling site activation times and voltages within the context of the entire surface geometry to provide the global indication of activation sequence. Furthermore, resulting depictions of chamber activation must be consistent with the first principles of cardiac electrophysiology. A map of sinus rhythm or any arrhythmia must be "plausible," whether the activation sequence is displayed in terms of progressive activation times, voltage transients, or any other physiologic parameter. While mapping systems should make a case easier, the user still has the obligation of knowing when it is serving up "jewels" or "junk" without relegating this responsibility solely to an industry representative.

Again, success in this process is part operator and part machine. While large circuits can be dissected with relatively few mapping points, progressively more points at a sufficient density are required to resolve arrhythmia circuits of decreasing size or increasing complexity. A reasonable goal for the ideal system is to ultimately provide adequate resolution and mechanistic disclosure on the same order as found with optical mapping systems. The use of advanced computational capabilities should simplify this process by cataloging all sites of data acquisition with the accompanying electrograms without confusing interpolation across lines of block or other boundaries of the circuits. This should be extended to allow mapping of rapid tachycardias, nonsustained arrhythmias, or tachycardias with complex activation pathways involving several different chambers.

Third, the system must lend itself readily to capturing and intelligibly displaying other details or the "physiologic quirks" contributing to an arrhythmia. This implies system versatility in extracting relevant features from electrograms or other sensors and providing real-time parametric displays. Most electrophysiologists are familiar with unipolar or bipolar voltage mapping to reflect underlying tissue integrity and pathophysiology in patients with a prior myocardial infarction. "Scar mapping" has been used, for example, to identify the site of possible circuits in patients with unstable ventricular tachycardias (VT) that defy activation sequence mapping.[1,2] An example of this approach is shown in Figure 1. Some investigators have used a 1.0 mV cut off to reflect scar and a 0.5 mV cutoff to reflect dense scar. It should be noted, however, that the presence of dense scar does not exclude the possibility of pathways within that scar that are incapable of generating a 0.5–1.0 mV signal. In atrial mapping, for example, scar cutoffs of <0.5 mV may better detect the underlying tissue pathophysiology and reflect relevant active circuit components.[1] Any mapping system should also chronicle the voltage changes of repolarization, even if this would require alternative signal amplification and filtering.

Electroanatomic voltage maps, generated during the evaluation of a patient with rapid unstable ventricular tachycardia. Panel A shows the voltage map with manually adjusted voltage settings (1.0 and 0.5 mV) to more specifically identify the interface between normal tissue (purple) and low-voltage scar (red). The yellow through blue isovoltage lines occurred at the assumed boundary between normal and abnormal tissue. Panel B shows the creation of an ablative line (maroon dots) along the subendocardial border zone of the infarction, which was successful in eliminating the patient's VT. Also shown is the site of 12 of 12 matches between the underlying ventricular tachycardia and the QRS morphology during pacing (black arrow). MV (white arrow) indicates the position of the mitral annulus while LVA is the left ventricular apex.

Mapping systems should identify and catalogue the presence and location of double potentials, fractionated electrograms, or the switching sequences of activation around a line of fixed or physiologic block, and archive and display these "quirks" in real-time, 3D space. Ideally, it should also be possible to exploit any characteristic of the electrogram at each site to better explain an arrhythmia. For example, it would be highly useful if signal amplitude, width, fractionation, or evidence of specific patterns of temporal activation, such as repetitive firing could be translated into specific maps through straightforward signal processing.

In theory, the ideal advanced mapping system should also display mechanical events, such as motion, wall stress or tension, or any other fourth- or fifth-dimensional contraction parameter and allow easy visualization, understanding, and metrics to assess those processes. Activation mapping of several chambers may thereby give a clear-cut indication of inter- or intraventricular dysynchrony, while voltage or other mapping over the course of a single cardiac cycle, may disclose the possibility of intramural dysynchrony.

Finally, the ability to catalogue the site of an intervention (either performed or planned), such as a specific ablation, or marking the site of cellular or other factor injections is highly useful. While this annotation might be done with a marking pen on an acetate film taped over a fluoroscopic monitor, such an approach is eclipsed by the capability of CPUs with high computational speed. The characteristics of available systems are chronicled in Table 1 .


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: