Electroanatomic mapping
Theoretical considerations
During conventional electrophysiology (EP) procedures, catheters are manually navigated with the use of single or bi-plane fluoroscopy. An inherent limitation of fluoroscopic navigation is that orientation of the catheter relative to the cardiac anatomy can only be appreciated in two dimensions. Thus it may make complex procedures challenging and is associated with radiation exposure for both patient and physician. In 1996 Ben-Haim and Josephson published a report of a new technology referred to as nonfluoroscopic in vivo navigation and mapping ( Fig. 7.1 ). Navigation was achieved by a magnetic sensor inserted in an EP recording catheter and magnetic field radiators placed below the operating bed. This discovery led to an era of innovation with several three-dimensional (3D) electroanatomic mapping (EAM) systems being developed for clinical use in the EP laboratory.
A fundamental principle of 3D EAM is the ability to collect anatomic (location) and electrical (electrograms) data simultaneously. This means that as the catheter connected to the 3D EAM system is moved in a cardiac chamber, serial recording of location data allows replication of the catheter tip’s location in real time on a computer monitor. When location data are acquired at specific anatomic locations, a meshlike geometry can be created that replicates the chamber’s anatomy. As location data are being collected, the catheters also record electrograms. Acquired electrograms can be displayed on the 3D chamber anatomy at their collected location coordinates. Each electrogram can be annotated for local activation time (LAT) or amplitude (voltage) and the annotation displayed, following a color code, on the map. Therefore, during this mapping process, the operator creates a 3D electroanatomic map of the chamber of interest ( Fig. 7.2 A).
Impedance-based location measurements (or a combination with magnetic field–based technology) are also used by some 3D EAM systems. This is performed by emitting a small current from the catheter electrodes and sensor patches placed on the patient. The resulting current ratio recorded at each sensor, per electrode, allows calculation of location data and consequent catheter visualization and mapping.
EAM has rendered mapping and ablation of complex arrhythmias, such as postoperative right atrial flutters and ventricular tachycardia (VT), feasible when previously this was a challenge. , Randomized ablation trials using EAM in patients with supraventricular tachycardia have demonstrated similar acute procedural success rates and reduced fluoroscopy time. , EAM is also useful for activation and voltage mapping and for tagging locations of specific electrograms of interest that allow substrate-modification ablation approaches (see Chapter 16 ).
The three systems used in clinical practice for EAM are the CARTO mapping platform (currently in its CARTO 3 version 7), the EnSite Precision platform, and the Rhythmia HDx system, which has the capability to record and map from 64 electrodes simultaneously.
Activation, voltage, and propagation mapping
Electrograms are acquired at specific anatomic locations, as the catheter moves across the chamber of interest, and is assigned to a local activation timing. The local activation time (LAT) of each electrogram is then compared with an automatically selected but often user-modified stable fiducial point. This is a surface electrocardiography (ECG) lead, or an intracardiac electrogram recorded by a diagnostic catheter in a stable position within the heart. The comparison occurs within in a fixed period, the window of interest (WOI), encompassing only one reference signal. If, for example, sinus rhythm is being mapped in the right atrium, the reference could be a clear surface P wave (V1 is usually convenient), and the WOI would have to span a period before the P wave and a period after it. With present systems, regions of red color indicate sites of “early activation” and activation becomes progressively later proceeding through the colors of the rainbow to yellow-green and finally the blue and purple hues that define the sites of late activation relative to the reference point. These colors are displayed as a time bar adjacent to the 3D map. Thus, if the annotated LAT of an electrogram is “early” in the window compared with the reference signal (i.e., occurs before it), it will be assigned a red color to be displayed as on the 3D map. As signals get later compared with the reference, the color eventually will change to purple. If colors are missing, then, a portion of the window has not been mapped with EGMs, either because the window has been set up incorrectly, mapping has been insufficient, or activation encompasses more time than the window covers. A common example of this is a perimitral flutter with epicardial conduction over the coronary sinus (CS).
Two other concepts pertain to activation mapping: isochrones and propagation. Isochronal mapping is a type of LAT mapping whereby color coding is done based on groups of EGMs with the same LAT (isochrones) as opposed to depicting and annotating each EGM distinctly. The maps appear similar, but crucially the area of a color (i.e., how much tissue is activated simultaneously) is purported to correlate to conduction speed ( Fig. 7.2 C and D).
Voltage maps of the recorded electrograms can be created with colors representing the maximal bipolar/unipolar voltage amplitude. These maps provide a method by which to quantify cardiac scarring and have significant value in modern cardiac EP procedures ( Fig. 7.2 B).
Mapping during arrhythmia also allows the creation of an actual propagation map of the tachycardia. This is a “moving” version of LAT that displays spread of activation wave front throughout a cardiac cycle as a live wave front traversing the cardiac chamber. It allows visual appreciation of the mapped arrhythmia mechanism and a visual estimation of relative conduction velocity through various sites.
Practical points of 3D mapping
Several problems can influence accuracy and should be eliminated as much as possible during the mapping process. They include the inherent noise of the location system, the reproducibility of the fiducial point on the ECG, the reproducibility of cardiac mechanics on a beat-to-beat basis, and gating of image acquisition to the cardiac cycle and respiratory phase. Maps created using EAM systems are also subject to additional variability depending on accurate annotation of electrogram qualities, consistent catheter contact with tissue, distributed sampling of the entire structure of interest, density of location “points” in the map, type of rhythm being mapped, direction of activation wave front propagation, and the size and spacing of the electrodes used to acquire the data.
Patient movement and respiratory stability
A basic requirement of accurate 3D EAM maps is that the patient remains still throughout the entire procedure. CARTO has arguably developed the most accurate navigation technology based on magnetic localization, which, as discussed, depends on creating a matrix of location data based on the reach of the intracardiac catheter in the magnetic field. However, this approach means it is also most sensitive to patient movements relative to the magnetic field, despite it also using impedance-based localization technology via sensor patches on the patient. EnSite Precision and Rhythmia HDx rely less on magnetic localization, using impedance primarily, and therefore are less sensitive to patient movements. However, localization by impedance is heavily limited by even minor changes in tissue characteristics (e.g., tidal volume or fluid status) that occur continuously during EP procedures, as well as background current noise. , If a patient moves during mapping, the entire collected map will “shift” with respect to where the heart and the catheters truly lie. This is a challenge to correct, by moving either patches, equipment, or the patient, and further mapping of the same map impossible. Many operators employ general anesthesia to avoid problems related to patient movement
Movement of the catheter caused by inspiration and expiration can be problematic when precise ablation is needed. Although patient tidal volumes can be controlled during general anesthesia procedures, it is preferable to employ technology available within the 3D EAM systems. All major system use thoracic impedance measurements to calculate respiratory motion and allow location data or EGM only during a predefined phase of respiration (typically end expiration).
Another challenge to accurate localization of catheters and anatomy by 3D EAM is the nonuniformity of cardiac anatomy in patients. For example, an extremely dilated left ventricle (LV) in a patient with scar-related VT may mean that the LV apex is “out of range” of the mapping range of any system.
Reference electrogram
During activation mapping, the electrical fiducial point needs to be stable and have a clearly reproducible annotation point. With atrial tachycardias, a stable intracardiac atrial electrogram such as a CS catheter electrogram is usually chosen because the P wave can be indistinct on the body surface ECG. With VTs, a large, reproducibly identifiable component of a QRS complex is typically chosen. Advances in 3D EAM systems can also take advantage of specific algorithms to identify the best reference signal from the body surface or stable intracardiac signals. Once selected, 3D EAM systems will annotate the reference signal automatically for the duration of the case. It is critically important, before mapping starts, to assess the automatic annotation of this signal because it cannot be changed after acquisition of the first point, to avoid map distortion. For voltage mapping, the reference can be any reliable and stable signal; errors in timing annotation are less important because the variable mapped is amplitude. However, because voltage and activation are almost always collected simultaneously, a stable reference selection should mark the start of any 3D mapping procedure.
Once the fiducial point is selected, it will provide time 0 ms within the WOI of any activation map. Fig. 7.3 shows how changing the reference while mapping the same rhythm (in this case an atrial tachycardia), changes the appearance of the LAT map. These differences are critical because the LAT map appearance guides ablation.
Window of interest
The WOI is defined by specifying an interval around the reference point, with symmetric or asymmetric boundaries, within which the mapping system records electrical signals for annotation. The WOI usually covers most of the cardiac cycle length in duration but strictly no more than one cardiac cycle in time—that is, not more than two reference signals.
Determination of the WOI is based on the rhythm being mapped when the diagnosis is known or clinically suspected. In focal tachycardias and ectopic beats, the true early site of activation is rarely earlier than 50 ms from the start of the surface ECG (P wave or QRS), and the window of interest is easily chosen to include the P wave/QRS with the left boundary beginning 40 to 50 ms before the reference point. In reentrant tachycardias the WOI is usually chosen to be 10 or 20 ms less than the tachycardia cycle length, and different methods to position it around the reference point have been proposed.
For atrial macroreentry, the most used calculation is the DePonti equation, which is calculated to center the WOI around activation occurring in the mid-diastolic pathway supporting reentry. However, a practical approach is choosing a timing interval slightly less than the tachycardia cycle and distributing this evenly around the reference, which is 0 ms. Another common practical approach is to set the left boundary at 40 ms before P-wave onset and extend forward for the remainder of the cycle. For ventricular substrate mapping during sinus or atrially paced rhythms, the principle is to include the entire ventricular component of activation and restoration in the WOI so as to correctly annotate peak voltage and identify any abnormally late activation. Therefore the WOI should exclude atrial activation and can be symmetric or asymmetric, depending on the reference chosen. During ventricular pacing, care must be taken to exclude the stimulation artifact from the WOI (even if chosen as the reference) to avoid its annotation as true voltage.
It is important to consider certain common errors regarding the WOI as illustrated in Fig. 7.4 . If the WOI is too short, for any rhythm, a certain section of activation will be “missed.” This may not be a significant problem when mapping a focal tachycardia, where the most important information is the pattern of activation after breakout; however, in the case of reentry or even late potential mapping, this can change the map dramatically because the entire circuit or channel may not be visualized. Conversely, if the WOI is too large, areas of adjacent early activation and late activation may be labeled as the same. Another common pitfall of mapping of a reentrant tachycardia with a 3D mapping system is the perception that the site of slow conduction that facilitates tachycardia perpetuation is the zone where the “earliest” activation sites meet the “latest” ones (early meets late area). Although in some cases this might be correct, many times it is not. The activation timing of a site depends on the positioning of the WOI around the reference point; thus the “early meets late” zone is also dependent on this factor.
