- Email: [email protected]
Recording and interpreting unipolar electrograms to guide catheter ablation
HANDS ON
Recording and interpreting unipolar electrograms to guide catheter ablation Usha B. Tedrow, MD, MSc, William G. Stevenson, MD, FHRS From Brigham and Women’s Hospital, Boston, Massachusetts.
Introduction Electrophysiology laboratories commonly use closely spaced bipolar recordings for mapping. However, unipolar recordings have some useful features that can provide additional complimentary information, provided the limitations of these recordings and the particular recording techniques are recognized.
Unipolar recordings A cardiac electrogram is the result of the voltage difference between two recording electrodes. For clinical systems, the exploring distal electrode is connected to the anodal (positive) input of the recording amplifier. For unipolar recordings, the cathodal (negative) input of the amplifier is connected to a remote electrode (referred to as an indifferent or reference electrode) that is distant from the heart. In most systems, selection of the unipolar recording mode connects the indifferent input to the Wilson central terminal, which can have substantial electrical noise. Noise may be reduced by using an indifferent electrode in the inferior vena cava. The amplifier is left in the bipolar configuration, but the negative input is connected to the intravascular electrode, which in our laboratory is located 15 to 20 cm proximal to the tip electrode of a custom hexapolar catheter positioned at the His bundle.1 The genesis of a unipolar recording is illustrated in Figure 1. A depolarization wavefront propagating toward the exploring electrode generates a positive deflection. As the wavefront reaches the electrode and propagates away, the deflection sweeps steeply negative. Thus, an RS complex is generated. In a sheet of uniformly conducting tissue, depolarization of tissue beneath the electrode coincides with the maximum negative slope (– dV/dt) of the signal. The morphology of the recording indicates the direction of wavefront propagation. When the exploring electrode is located at the site of initial activation (e.g., the left-hand side of the tissue in Figure 1A), depolarization produces a wavefront KEYWORDS Ablation; Conduction block; Mapping ABBREVIATION VT ⫽ ventricular tachycardia (Heart Rhythm 2011;8: 791–796) Dr. Tedrow has received consulting fees from St. Jude Medical and Boston Scientific. Address reprint requests and correspondence: Dr. Usha B. Tedrow, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115. E-mail address: [email protected].
that spreads away from the electrode, generating a monophasic QS complex.1,2 At sites remote from the tachycardia origin, an initial R wave is recorded. For focal arrhythmias, a QS complex is typically recorded at the successful ablation site (Figure 2A). A QS complex may be recorded over an area larger than the successful ablation site and should not be the only mapping finding used to guide ablation. Successful ablation is unusual, however, at sites with an initial R wave in the unipolar recordings (Figure 2B). It is also important to realize that a QS complex can be recorded when the exploring electrode is not in contact with the myocardium. With poor or absent electrode contact, the initial S wave is slurred, suggesting that the electrogram is a far-field signal. The major disadvantage of unipolar recordings is the contribution of substantial far-field signal generated by depolarization of tissue remote from the recording electrode. In normal tissue, the maximum negative slope is a good indication of local depolarization. In abnormal regions, such as infarct scars, the tissue beneath the recording electrode may be small relative to the surrounding myocardium outside the scar. A large farfield signal can obscure low-amplitude local potentials of interest (Figure 5A). With clinical recording systems, unipolar signals are of limited value for mapping in areas of scar unless they are filtered to remove far-field signal (discussed later).
Bipolar recordings Bipolar recordings are obtained by connecting two electrodes in the area of interest to the recording amplifier (Figure 1B).2 Much of the far-field signal is similar at each electrode and is subtracted out. In homogeneous tissue, the initial peak of the bipolar signal coincides with depolarization beneath the recording electrode and is typically selected for activation time. Bipolar recording facilitates the identification of local depolarization in abnormal areas of infarction or scar (Figure 5). Multiple sharp deflections that represent asynchronous activation of multiple myocyte bundles are often recorded in scars and may be obscured in unipolar recordings. In bipolar recordings, however, a potential of interest may originate beneath either or both recording electrodes (Figure 5B). Ablation energy is applied only to the distal electrode, so ablation may fail when the focus is beneath the proximal electrode.
1547-5271/$ -see front matter © 2011 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2010.12.038
792
Heart Rhythm, Vol 8, No 5, May 2011
Figure 1 Genesis of unipolar and bipolar electrograms. A: Schematic diagram showing the unipolar recording electrode on a uniform sheet of tissue. Excitation of this tissue proceeds from left to right; the lighter gray areas indicate depolarized tissue. Theoretical recordings from three different sites along the tissue are shown in the bottom panel. Recordings from the initial site of depolarization generate a QS complex. Recordings from the last site to be depolarized generate a monophasic R wave (see text for discussion). B: Simulated unipolar and bipolar electrograms. The unipolar signal seen at the positive input to the recording amplifier (Uni-1) is an R-S. The unipolar signal from the second electrode (Uni-2) is in the negative input to the amplifier and therefore is inverted. The summation of these two signals generates the bipolar signal (Bi 1-2). Much of the far-field signal is the same at the two electrodes and therefore is subtracted out. C: Setup for unipolar recording, using either the Wilson central terminal or an indifferent electrode in the inferior vena cava (IVC). RA indicates the right arm; LA, the left arm; and LL the left leg ECG electrodes. (Reproduced with permission from Stevenson WG, Soejima K. Recording techniques for clinical electrophysiology. J Cardiovasc Electrophysiol 2005;16:1017–1022.)
In contrast to the amplitude of unipolar recordings, the amplitude of a bipolar electrogram is influenced by the direction of wavefront propagation.2 Signal amplitude is largest for a wavefront propagating parallel to the axis of the recording electrodes and is reduced when propagation is perpendicular to the electrodes. The bipolar electrogram is also influenced by catheter orientation relative to the tissue. If the proximal electrode is off the tissue, a “semi-unipolar” recording results that has less far-field signal than the unipolar recording from the electrode that is in contact with the tissue. These factors can potentially influence voltage maps as well as activation maps.
High-pass filtering
Signal processing
Filtering unipolar recordings
Electrograms can be considered the sum of waveforms of different frequencies. The distance between the signal source and the recording electrodes also influences the signal. The amplitude of the signal diminishes as a function of signal frequency. The faster the frequency, the greater it will decrease in amplitude with respect to distance from the source. Thus, we recognize that “sharp potentials” are often due to local depolarization and rounded potentials are often “far field” from depolarization of distant tissue. Techniques of analog-to-digital conversion and filtering also influence the frequency content of electrograms.
Unipolar signals are commonly high-pass filtered at low corner frequencies of 0.05 to 0.5 Hz, which removes baseline drift without substantial alteration of the electrogram appearance. With these filter settings, a QS complex is observed at the site of origin of focal arrhythmias. The far-field component of unipolar signals can be reduced by high-pass filtering at corner frequencies of 30 Hz and above. This can facilitate identification of low-amplitude local potentials that would be otherwise obscured, (Figure 5B, with low-amplitude potential seen under electrode 1), but alters the morphology such that QS and RS
High-pass filters attenuate low-frequency signals, but filters are imperfect and do not completely suppress these frequencies. The selected corner frequency provides an indication of the frequencies that will be attenuated, but the slope relating amplitude to frequency varies with different filters. In general, high-pass filtering can be viewed as differentiating the signal, such that the amplitude is proportional not just to the potential amplitude but also to the rate of change of the signal. The effect of differentiating signals, which can introduce additional peaks and complexity, is shown in Figure 1B.
Tedrow and Stevenson
Unipolar Electrograms for Guiding Catheter Ablation
793
Figure 2 Recordings of focal ventricular ectopy from the lateral papillary muscle. Shown from the top to bottom are ECG leads and recordings from the mapping catheter: bipolar (ABL1-2, ABL2-3, ABL3-4) and unipolar high-pass (30 Hz) filtered (U1 hp, U2 hp) and unipolar high-pass filtered at 0.5 Hz (U1) electrograms. Vertical dashed line from a reference point on the premature beat is shown. Each panel shows a sinus beat followed by a premature beat. A: The ablation catheter is on a successful ablation site. The premature beat has a sharp QS complex in U1, and the onset of the QS, the bipolar electrogram peak, and high-pass filtered unipolar electrogram coincide. The sharp initial component of in the bipolar signal (arrow) is present in the U1 hp signal, indicating that it is beneath the ablating electrode. B: Remote from the focus, the U1 signal on the premature beat has an initial r wave. A limitation of the unipolar recording is also shown: a slurred QS complex is present on the sinus beats due to poor electrode contact, which is common when mapping along papillary muscles.
morphologies are not reliable indications of proximity to a focal source.
Simultaneous unipolar and bipolar recordings We commonly record bipolar signals from the distal (electrodes 1–2), mid (electrodes 2–3), and proximal (electrodes 3– 4) pairs and unipolar recordings from electrodes 1 and 2 (Figure 5). Unipolar recordings for electrode 1 are often recorded on two channels, with one high-pass filtered at 30 Hz to facilitate identification of low-amplitude local potentials, and the second filtered at 0.5 Hz for assessment of QS complexes and far-field signal. Comparison of the electrograms and activation times among the simultaneous recordings provides an indication of wavefront direction.1 If activation is earlier at the proximal electrodes than at the distal electrodes, the distal electrode is moved in the direction toward the proximal electrode site in search of earlier activation. In addition, at sites remote from the focus, onset of the initial R wave in the minimally filtered unipolar recording precedes the first peak of the bipolar electrogram (Figure 2A). In contrast, at the earliest site of activation, this unipolar signal has a QS configuration with the rapid S-wave downstroke coinciding with the initial peak of the bipolar signal (Figure 2B).
Mapping specific arrhythmias Focal ventricular and atrial tachycardias For focal tachycardias, we seek a QS complex with a sharp downstroke and good electrode contact. We then fine tune the catheter position from the unipolar and
bipolar signals to achieve the earliest activation time (Figures 2 and 3). At the focus, the onset of the qS unipolar signal coincides with the first peak of the bipolar signal and is earlier or simultaneous with unipolar 1 compared to unipolar 2 recordings (Figures 2, 4, and 5). A qS signal alone may be seen over a small area. and the earliest site of activation within that area should be sought.
Atrial fibrillation Failure to achieve permanent, transmural lesions is a major cause of failed catheter ablation for atrial fibrillation. Otomo et al evaluated unipolar recordings to identify RF ablation lesion transmurality in porcine atria.3,4 Prior to ablation, unipolar electrograms had an RS morphology. After ablation, transmural atrial lesions showed loss of the S wave, consistent with wavefront propagation toward the recording site from all sides. In contrast, the S waves remained when the lesion was not transmural. For bipolar recordings, ablation diminished electrogram amplitude, but the changes were more variable. Whether analysis of unipolar electrograms will be a useful endpoint for ablation at various sites in human atria remains to be determined.
Accessory pathways Unipolar electrograms filtered at 0.5 Hz are useful for mapping accessory pathways.5,6 During anterograde (A-V) conduction, a P-qS is seen at the ventricular insertion region (Figure 4). A QS complex alone is not sufficient to guide ablation, as it may be seen over a small region or if catheter contact is poor.
794
Heart Rhythm, Vol 8, No 5, May 2011
Figure 3 Recordings from a patient with focal right atrial tachycardia (AT). Shown from top to bottom are ECG leads, recordings from the mapping catheter: bipolar (ABL1-2, ABL2-3, ABL3-4) and unipolar high-pass (30 Hz) filtered (U1 hp, U2 hp) and unipolar high-pass filtered at 0.5 Hz (U1) electrograms, followed by intracardiac recordings from the lateral right atrium (RA). Dashed vertical line marks p-wave onset. A: Recordings from a site adjacent to the successful ablation site. In U1, the atrial signal has is an rs complex (arrow) whose onset precedes the bipolar 1-2 signal peak. B: Recordings from the successful ablation site. The onset of the qs complex in U-1 is simultaneous with the first peak of the bipolar signal in ABL1-2.
However, ablation usually fails at sites where an initial r wave is inscribed. These features can be helpful for mapping anterograde activation during atrial fibrillation.5 During VA conduction, the atrial electrogram at the atrial insertion site is expected to have an initial negative deflection, but this is often difficult to define.
Scar-related ventricular tachycardia In scar-related arrhythmias, fibrosis creates asynchronous activation of myocyte bundles giving rise to fractionated bipolar electrograms, with multiple rapid components. Resulting complex electrograms have variable contributions from tissue beneath the recording electrodes and remote tissue (Figure 5). Low-amplitude local potentials may be obscured by far-field signals from the larger mass of myocardium outside the scar (Figure 5B). High-pass filtering can be helpful but increases noise with some systems, such that low-amplitude isolated potentials may still be obscured. At the reentrant circuit exit in the infarct border (Figure 5A), fractionated presystolic electrograms are seen in the bipolar and filtered unipolar recordings. The minimally filtered unipolar recording may have a dominant S wave as the wavefront propagates away from the exit region across the ventricle, creating a QS complex. The reliability of this finding for selecting exits for ablation has not been determined. The unipolar electrogram morphology is variable at sites in the scar.
Substrate mapping Definition of scar regions in electroanatomic maps is an important tool for guiding ablation7–10 In the ventricles, a bipolar electrogram amplitude less than 1.5 to 1.55 mV is specific for
scar.7,9 Electrogram amplitude does not separate dense fibrosis in electrically unexcitable scar, where pacing fails to capture, from small channels of excitable myocardium that produce very-low-amplitude signals.8 Unipolar electrograms have also been evaluated for substrate mapping.9,11–14 The amplitude of unipolar electrograms high-pass filtered at 1 Hz is threefold to fivefold greater than that of bipolar electrograms high-pass filtered at 30 Hz. In comparisons with infarct scars detected by magnetic resonance imaging, Codreanu et al9 found that the best electrogram amplitude threshold for characterizing scar was 6.52 mV for unipolar electrograms and 1.54 mV for bipolar electrograms, and bipolar electrogram amplitude was a better discriminator of scar. Wijnmaalen et al14 found that greater than 50% scar transmurality produced unipolar electrogram amplitudes less than 8 mV, but with substantial variability. In an ovine infarct model, scar was better correlated with bipolar filtered as compared to unipolar minimally filtered electrogram amplitude.12 At present, bipolar electrograms are preferred for defining endocardial scar. Scars causing ventricular tachycardia (VT) can be endocardial, intramural or epicardial in location. Identification of intramural or epicardial scar from endocardial mapping is desirable, particularly in consideration of percutaneous epicardial mapping. Marchlinski et al hypothesized that the contribution of far-field signal to unipolar electrograms might allow detection of epicardial scar. In patients with right ventricular scar-related VT, they found that an endocardial unipolar voltage less than 5.5 mV in regions with normal bipolar electrograms predicted overlying epicardial right ventricular scar.15
Tedrow and Stevenson
Unipolar Electrograms for Guiding Catheter Ablation
795
Figure 4 A: Recordings from an anterogradely conducting right posterior septal accessory pathway during sinus rhythm. Shown are recordings from the mapping catheter: bipolar (ABL1-2, ABL2-3) and unipolar high-pass (30 Hz) filtered (U1 hp, U2 hp) and unipolar high-pass filtered at 0.5 Hz (U1) electrograms, followed by recordings from the right atrium (RA), His-bundle region (His), coronary sinus proximal to distal (CSp, CSd), and right ventricular apex (RVA). Preexcitation is present on the first beat and is absent on the second beat at this successful ablation site, allowing clear recognition of atrial and ventricular signals. On the preexcited beat, the large ventricular signal has a qS configuration (arrow). A small initial positive deflection is likely due to the overlapping atrial signal. In the absence of preexcitation, the unipolar signal has a large initial r wave. Ventricular activation precedes the delta wave by approximately 30 ms. Note that the V in the bipolar signal is on time with the onset of the unipolar ventricular signal. B, C: Recordings during atrial pacing with anterograde conduction over a right-sided accessory pathway. At the tricuspid annulus remote from the pathway, the unipolar electrogram filtered at 0.5 Hz (U1) has an rS configuration. At the successful ablation site (C), the U1 recording has a qS configuration. Tracings are labeled as in panel A.
Changes in electrogram amplitude can be used to assess creation of ablation lesions in the ventricles, but postablation pacing threshold has been shown to better correlate with lesion size.16
Conclusion Although bipolar recordings provide sufficient information for most catheter mapping purposes, unipolar recordings and manipulation of filtering can provide additional helpful information and are easily implemented with most recording systems. Unipolar electrograms have important potential future applications in the catheter ablation of atrial fibrillation and characterization of substrate for ventricular arrhythmias.
References 1. Delacretaz E, Soejima K, Gottipaty VK, et al. Single catheter determination of local electrogram prematurity using simultaneous unipolar and bipolar recordings to replace the surface ECG as a timing reference. Pacing Clin Electrophysiol 2001;24(4 Pt I):441– 449. 2. de Bakker JMT, Hauer RNW, Simmers TA. Activation mapping: unipolar versus bipolar recording. In Zipes DP, Jalife J, editors : Cardiac Electrophysiology : From Cell to Bedside. Second Edition. Philadelphia: WB Saunders, 1995:1068. 3. Michaud GF, Cutro R. Information at our catheter tips: unipolar electrogram morphology makes another comeback! Heart Rhythm 2010;7:1301–1302. 4. Otomo K, Uno K, Fujiwara H, et al. Local unipolar and bipolar electrogram criteria for evaluating the transmurality of atrial ablation lesions at different catheter orientations relative to the endocardial surface. Heart Rhythm 2010;7: 1291–1300. 5. Hindricks G, Kottkamp H, Chen X, et al. Successful radiofrequency catheter ablation of right sided accessory pathways during sustained atrial fibrillation. Eur Heart J 1995;16:967–970.
796
Heart Rhythm, Vol 8, No 5, May 2011
Figure 5 Recordings from a patient with ventricular tachycardia (VT) due to an anterior wall myocardial infarction. Shown from top in each panel are ECG leads, bipolar recordings from the mapping catheter distal (ABL1-2) and middle (ABL 2-3) electrodes, unipolar 1 high-pass filtered at 30 Hz (U1 hp, U2 hp), and unipolar 1 signal with minimal high-pass filtering (U1). A, B: Sustained monomorphic VT with a cycle length of 510 ms is present. Beneath the tracings is a schematic of the left ventricle viewed from the right anterior oblique projection. Dark gray indicates normal ventricular myocardium. Lighter gray region is the apical infarct, containing branching channels (white). Arrows indicate excitation wavefronts. In A, the recording site is in the reentrant circuit exit. Electrograms precede QRS onset. The U1 signal has a small r wave and large S wave, consistent with spread of the excitation wavefront away from the exit, across the ventricle. Activation of this large mass of tissue obscures some local signals in the infarct border. In B, the recording site is from an isthmus in the center of the infarct region. A local low-amplitude isolated potential is present, on the distal ablation bipolar and unipolar recordings (U1 hp, U1). The U1 signal has less far-field component than at the exit region (A) because it is further from the border of the infarct. The U1 far-field signal has a complex configuration that does not reliably indicate the direction of propagation. C: Recordings during sinus rhythm from an isthmus region in the center of the anterior wall infarct. A late potential is present, visible on the distal bipolar and unipolar recordings and faintly visible on the U2 hp recording.
6. Haissaguerre M, Dartigues JF, Warin JF, et al. Electrogram patterns predictive of successful catheter ablation of accessory pathways. Value of unipolar recording mode. Circulation 1991;84:188 –202. 7. Marchlinski FE, Callans DJ, Gottlieb CD, et al. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation 2000;101:1288 –1296. 8. Soejima K, Suzuki M, Maisel WH, et al. Catheter ablation in patients with multiple and unstable ventricular tachycardias after myocardial infarction: short ablation lines guided by reentry circuit isthmuses and sinus rhythm mapping. Circulation 2001;104:664 – 669. 9. Codreanu A, Odille F, Aliot E, et al. Electroanatomic characterization of post-infarct scars comparison with 3-dimensional myocardial scar reconstruction based on magnetic resonance imaging. J Am Coll Cardiol 2008; 52:839 – 842. 10. Deneke T, Muller KM, Lemke B, et al. Human histopathology of electroanatomic mapping after cooled-tip radiofrequency ablation to treat ventricular tachycardia in remote myocardial infarction. J Cardiovasc Electrophysiol 2005;16: 1246 –1251. 11. Brunckhorst CB, Delacretaz E, Soejima K, et al. Impact of changing activation sequence on bipolar electrogram amplitude for voltage mapping of left ventric-
12.
13.
14.
15.
16.
ular infarcts causing ventricular tachycardia. J Interv Card Electrophysiol 2005;12:137–141. Sivagangabalan G, Pouliopoulos J, Huang K, et al. Comparison of electroanatomic contact and noncontact mapping of ventricular scar in a postinfarct ovine model with intramural needle electrode recording and histological validation. Circ Arrhythm Electrophysiol 2008;1:363–369. Koch KC, vom Dahl J, Wenderdel M, et al. Myocardial viability assessment by endocardial electroanatomic mapping: comparison with metabolic imaging and functional recovery after coronary revascularization. J Am Coll Cardiol 2001; 38:91–98. Wijnmaalen AP, van der Geest RJ, van Huls van Taxis CF, et al. Head-to-head comparison of contrast-enhanced magnetic resonance imaging and electroanatomical voltage mapping to assess post-infarct scar characteristics in patients with ventricular tachycardias: real-time image integration and reversed registration. Eur Heart J 2010;32:104 –114. Polin GM, Haqqani H, Tzou W, et al. Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular dysplasia/cardiomyopathy. Heart Rhythm 2011;8:76 – 83. Sapp JL, Soejima K, Cooper JM, et al. Ablation lesion size correlates with pacing threshold: a physiological basis for use of pacing to assess ablation lesions. Pacing Clin Electrophysiol 2004;27:933–937.
Recording and interpreting unipolar electrograms to guide catheter ablation Usha B. Tedrow, MD, MSc, William G. Stevenson, MD, FHRS From Brigham and Women’s Hospital, Boston, Massachusetts.
Introduction Electrophysiology laboratories commonly use closely spaced bipolar recordings for mapping. However, unipolar recordings have some useful features that can provide additional complimentary information, provided the limitations of these recordings and the particular recording techniques are recognized.
Unipolar recordings A cardiac electrogram is the result of the voltage difference between two recording electrodes. For clinical systems, the exploring distal electrode is connected to the anodal (positive) input of the recording amplifier. For unipolar recordings, the cathodal (negative) input of the amplifier is connected to a remote electrode (referred to as an indifferent or reference electrode) that is distant from the heart. In most systems, selection of the unipolar recording mode connects the indifferent input to the Wilson central terminal, which can have substantial electrical noise. Noise may be reduced by using an indifferent electrode in the inferior vena cava. The amplifier is left in the bipolar configuration, but the negative input is connected to the intravascular electrode, which in our laboratory is located 15 to 20 cm proximal to the tip electrode of a custom hexapolar catheter positioned at the His bundle.1 The genesis of a unipolar recording is illustrated in Figure 1. A depolarization wavefront propagating toward the exploring electrode generates a positive deflection. As the wavefront reaches the electrode and propagates away, the deflection sweeps steeply negative. Thus, an RS complex is generated. In a sheet of uniformly conducting tissue, depolarization of tissue beneath the electrode coincides with the maximum negative slope (– dV/dt) of the signal. The morphology of the recording indicates the direction of wavefront propagation. When the exploring electrode is located at the site of initial activation (e.g., the left-hand side of the tissue in Figure 1A), depolarization produces a wavefront KEYWORDS Ablation; Conduction block; Mapping ABBREVIATION VT ⫽ ventricular tachycardia (Heart Rhythm 2011;8: 791–796) Dr. Tedrow has received consulting fees from St. Jude Medical and Boston Scientific. Address reprint requests and correspondence: Dr. Usha B. Tedrow, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115. E-mail address: [email protected].
that spreads away from the electrode, generating a monophasic QS complex.1,2 At sites remote from the tachycardia origin, an initial R wave is recorded. For focal arrhythmias, a QS complex is typically recorded at the successful ablation site (Figure 2A). A QS complex may be recorded over an area larger than the successful ablation site and should not be the only mapping finding used to guide ablation. Successful ablation is unusual, however, at sites with an initial R wave in the unipolar recordings (Figure 2B). It is also important to realize that a QS complex can be recorded when the exploring electrode is not in contact with the myocardium. With poor or absent electrode contact, the initial S wave is slurred, suggesting that the electrogram is a far-field signal. The major disadvantage of unipolar recordings is the contribution of substantial far-field signal generated by depolarization of tissue remote from the recording electrode. In normal tissue, the maximum negative slope is a good indication of local depolarization. In abnormal regions, such as infarct scars, the tissue beneath the recording electrode may be small relative to the surrounding myocardium outside the scar. A large farfield signal can obscure low-amplitude local potentials of interest (Figure 5A). With clinical recording systems, unipolar signals are of limited value for mapping in areas of scar unless they are filtered to remove far-field signal (discussed later).
Bipolar recordings Bipolar recordings are obtained by connecting two electrodes in the area of interest to the recording amplifier (Figure 1B).2 Much of the far-field signal is similar at each electrode and is subtracted out. In homogeneous tissue, the initial peak of the bipolar signal coincides with depolarization beneath the recording electrode and is typically selected for activation time. Bipolar recording facilitates the identification of local depolarization in abnormal areas of infarction or scar (Figure 5). Multiple sharp deflections that represent asynchronous activation of multiple myocyte bundles are often recorded in scars and may be obscured in unipolar recordings. In bipolar recordings, however, a potential of interest may originate beneath either or both recording electrodes (Figure 5B). Ablation energy is applied only to the distal electrode, so ablation may fail when the focus is beneath the proximal electrode.
1547-5271/$ -see front matter © 2011 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2010.12.038
792
Heart Rhythm, Vol 8, No 5, May 2011
Figure 1 Genesis of unipolar and bipolar electrograms. A: Schematic diagram showing the unipolar recording electrode on a uniform sheet of tissue. Excitation of this tissue proceeds from left to right; the lighter gray areas indicate depolarized tissue. Theoretical recordings from three different sites along the tissue are shown in the bottom panel. Recordings from the initial site of depolarization generate a QS complex. Recordings from the last site to be depolarized generate a monophasic R wave (see text for discussion). B: Simulated unipolar and bipolar electrograms. The unipolar signal seen at the positive input to the recording amplifier (Uni-1) is an R-S. The unipolar signal from the second electrode (Uni-2) is in the negative input to the amplifier and therefore is inverted. The summation of these two signals generates the bipolar signal (Bi 1-2). Much of the far-field signal is the same at the two electrodes and therefore is subtracted out. C: Setup for unipolar recording, using either the Wilson central terminal or an indifferent electrode in the inferior vena cava (IVC). RA indicates the right arm; LA, the left arm; and LL the left leg ECG electrodes. (Reproduced with permission from Stevenson WG, Soejima K. Recording techniques for clinical electrophysiology. J Cardiovasc Electrophysiol 2005;16:1017–1022.)
In contrast to the amplitude of unipolar recordings, the amplitude of a bipolar electrogram is influenced by the direction of wavefront propagation.2 Signal amplitude is largest for a wavefront propagating parallel to the axis of the recording electrodes and is reduced when propagation is perpendicular to the electrodes. The bipolar electrogram is also influenced by catheter orientation relative to the tissue. If the proximal electrode is off the tissue, a “semi-unipolar” recording results that has less far-field signal than the unipolar recording from the electrode that is in contact with the tissue. These factors can potentially influence voltage maps as well as activation maps.
High-pass filtering
Signal processing
Filtering unipolar recordings
Electrograms can be considered the sum of waveforms of different frequencies. The distance between the signal source and the recording electrodes also influences the signal. The amplitude of the signal diminishes as a function of signal frequency. The faster the frequency, the greater it will decrease in amplitude with respect to distance from the source. Thus, we recognize that “sharp potentials” are often due to local depolarization and rounded potentials are often “far field” from depolarization of distant tissue. Techniques of analog-to-digital conversion and filtering also influence the frequency content of electrograms.
Unipolar signals are commonly high-pass filtered at low corner frequencies of 0.05 to 0.5 Hz, which removes baseline drift without substantial alteration of the electrogram appearance. With these filter settings, a QS complex is observed at the site of origin of focal arrhythmias. The far-field component of unipolar signals can be reduced by high-pass filtering at corner frequencies of 30 Hz and above. This can facilitate identification of low-amplitude local potentials that would be otherwise obscured, (Figure 5B, with low-amplitude potential seen under electrode 1), but alters the morphology such that QS and RS
High-pass filters attenuate low-frequency signals, but filters are imperfect and do not completely suppress these frequencies. The selected corner frequency provides an indication of the frequencies that will be attenuated, but the slope relating amplitude to frequency varies with different filters. In general, high-pass filtering can be viewed as differentiating the signal, such that the amplitude is proportional not just to the potential amplitude but also to the rate of change of the signal. The effect of differentiating signals, which can introduce additional peaks and complexity, is shown in Figure 1B.
Tedrow and Stevenson
Unipolar Electrograms for Guiding Catheter Ablation
793
Figure 2 Recordings of focal ventricular ectopy from the lateral papillary muscle. Shown from the top to bottom are ECG leads and recordings from the mapping catheter: bipolar (ABL1-2, ABL2-3, ABL3-4) and unipolar high-pass (30 Hz) filtered (U1 hp, U2 hp) and unipolar high-pass filtered at 0.5 Hz (U1) electrograms. Vertical dashed line from a reference point on the premature beat is shown. Each panel shows a sinus beat followed by a premature beat. A: The ablation catheter is on a successful ablation site. The premature beat has a sharp QS complex in U1, and the onset of the QS, the bipolar electrogram peak, and high-pass filtered unipolar electrogram coincide. The sharp initial component of in the bipolar signal (arrow) is present in the U1 hp signal, indicating that it is beneath the ablating electrode. B: Remote from the focus, the U1 signal on the premature beat has an initial r wave. A limitation of the unipolar recording is also shown: a slurred QS complex is present on the sinus beats due to poor electrode contact, which is common when mapping along papillary muscles.
morphologies are not reliable indications of proximity to a focal source.
Simultaneous unipolar and bipolar recordings We commonly record bipolar signals from the distal (electrodes 1–2), mid (electrodes 2–3), and proximal (electrodes 3– 4) pairs and unipolar recordings from electrodes 1 and 2 (Figure 5). Unipolar recordings for electrode 1 are often recorded on two channels, with one high-pass filtered at 30 Hz to facilitate identification of low-amplitude local potentials, and the second filtered at 0.5 Hz for assessment of QS complexes and far-field signal. Comparison of the electrograms and activation times among the simultaneous recordings provides an indication of wavefront direction.1 If activation is earlier at the proximal electrodes than at the distal electrodes, the distal electrode is moved in the direction toward the proximal electrode site in search of earlier activation. In addition, at sites remote from the focus, onset of the initial R wave in the minimally filtered unipolar recording precedes the first peak of the bipolar electrogram (Figure 2A). In contrast, at the earliest site of activation, this unipolar signal has a QS configuration with the rapid S-wave downstroke coinciding with the initial peak of the bipolar signal (Figure 2B).
Mapping specific arrhythmias Focal ventricular and atrial tachycardias For focal tachycardias, we seek a QS complex with a sharp downstroke and good electrode contact. We then fine tune the catheter position from the unipolar and
bipolar signals to achieve the earliest activation time (Figures 2 and 3). At the focus, the onset of the qS unipolar signal coincides with the first peak of the bipolar signal and is earlier or simultaneous with unipolar 1 compared to unipolar 2 recordings (Figures 2, 4, and 5). A qS signal alone may be seen over a small area. and the earliest site of activation within that area should be sought.
Atrial fibrillation Failure to achieve permanent, transmural lesions is a major cause of failed catheter ablation for atrial fibrillation. Otomo et al evaluated unipolar recordings to identify RF ablation lesion transmurality in porcine atria.3,4 Prior to ablation, unipolar electrograms had an RS morphology. After ablation, transmural atrial lesions showed loss of the S wave, consistent with wavefront propagation toward the recording site from all sides. In contrast, the S waves remained when the lesion was not transmural. For bipolar recordings, ablation diminished electrogram amplitude, but the changes were more variable. Whether analysis of unipolar electrograms will be a useful endpoint for ablation at various sites in human atria remains to be determined.
Accessory pathways Unipolar electrograms filtered at 0.5 Hz are useful for mapping accessory pathways.5,6 During anterograde (A-V) conduction, a P-qS is seen at the ventricular insertion region (Figure 4). A QS complex alone is not sufficient to guide ablation, as it may be seen over a small region or if catheter contact is poor.
794
Heart Rhythm, Vol 8, No 5, May 2011
Figure 3 Recordings from a patient with focal right atrial tachycardia (AT). Shown from top to bottom are ECG leads, recordings from the mapping catheter: bipolar (ABL1-2, ABL2-3, ABL3-4) and unipolar high-pass (30 Hz) filtered (U1 hp, U2 hp) and unipolar high-pass filtered at 0.5 Hz (U1) electrograms, followed by intracardiac recordings from the lateral right atrium (RA). Dashed vertical line marks p-wave onset. A: Recordings from a site adjacent to the successful ablation site. In U1, the atrial signal has is an rs complex (arrow) whose onset precedes the bipolar 1-2 signal peak. B: Recordings from the successful ablation site. The onset of the qs complex in U-1 is simultaneous with the first peak of the bipolar signal in ABL1-2.
However, ablation usually fails at sites where an initial r wave is inscribed. These features can be helpful for mapping anterograde activation during atrial fibrillation.5 During VA conduction, the atrial electrogram at the atrial insertion site is expected to have an initial negative deflection, but this is often difficult to define.
Scar-related ventricular tachycardia In scar-related arrhythmias, fibrosis creates asynchronous activation of myocyte bundles giving rise to fractionated bipolar electrograms, with multiple rapid components. Resulting complex electrograms have variable contributions from tissue beneath the recording electrodes and remote tissue (Figure 5). Low-amplitude local potentials may be obscured by far-field signals from the larger mass of myocardium outside the scar (Figure 5B). High-pass filtering can be helpful but increases noise with some systems, such that low-amplitude isolated potentials may still be obscured. At the reentrant circuit exit in the infarct border (Figure 5A), fractionated presystolic electrograms are seen in the bipolar and filtered unipolar recordings. The minimally filtered unipolar recording may have a dominant S wave as the wavefront propagates away from the exit region across the ventricle, creating a QS complex. The reliability of this finding for selecting exits for ablation has not been determined. The unipolar electrogram morphology is variable at sites in the scar.
Substrate mapping Definition of scar regions in electroanatomic maps is an important tool for guiding ablation7–10 In the ventricles, a bipolar electrogram amplitude less than 1.5 to 1.55 mV is specific for
scar.7,9 Electrogram amplitude does not separate dense fibrosis in electrically unexcitable scar, where pacing fails to capture, from small channels of excitable myocardium that produce very-low-amplitude signals.8 Unipolar electrograms have also been evaluated for substrate mapping.9,11–14 The amplitude of unipolar electrograms high-pass filtered at 1 Hz is threefold to fivefold greater than that of bipolar electrograms high-pass filtered at 30 Hz. In comparisons with infarct scars detected by magnetic resonance imaging, Codreanu et al9 found that the best electrogram amplitude threshold for characterizing scar was 6.52 mV for unipolar electrograms and 1.54 mV for bipolar electrograms, and bipolar electrogram amplitude was a better discriminator of scar. Wijnmaalen et al14 found that greater than 50% scar transmurality produced unipolar electrogram amplitudes less than 8 mV, but with substantial variability. In an ovine infarct model, scar was better correlated with bipolar filtered as compared to unipolar minimally filtered electrogram amplitude.12 At present, bipolar electrograms are preferred for defining endocardial scar. Scars causing ventricular tachycardia (VT) can be endocardial, intramural or epicardial in location. Identification of intramural or epicardial scar from endocardial mapping is desirable, particularly in consideration of percutaneous epicardial mapping. Marchlinski et al hypothesized that the contribution of far-field signal to unipolar electrograms might allow detection of epicardial scar. In patients with right ventricular scar-related VT, they found that an endocardial unipolar voltage less than 5.5 mV in regions with normal bipolar electrograms predicted overlying epicardial right ventricular scar.15
Tedrow and Stevenson
Unipolar Electrograms for Guiding Catheter Ablation
795
Figure 4 A: Recordings from an anterogradely conducting right posterior septal accessory pathway during sinus rhythm. Shown are recordings from the mapping catheter: bipolar (ABL1-2, ABL2-3) and unipolar high-pass (30 Hz) filtered (U1 hp, U2 hp) and unipolar high-pass filtered at 0.5 Hz (U1) electrograms, followed by recordings from the right atrium (RA), His-bundle region (His), coronary sinus proximal to distal (CSp, CSd), and right ventricular apex (RVA). Preexcitation is present on the first beat and is absent on the second beat at this successful ablation site, allowing clear recognition of atrial and ventricular signals. On the preexcited beat, the large ventricular signal has a qS configuration (arrow). A small initial positive deflection is likely due to the overlapping atrial signal. In the absence of preexcitation, the unipolar signal has a large initial r wave. Ventricular activation precedes the delta wave by approximately 30 ms. Note that the V in the bipolar signal is on time with the onset of the unipolar ventricular signal. B, C: Recordings during atrial pacing with anterograde conduction over a right-sided accessory pathway. At the tricuspid annulus remote from the pathway, the unipolar electrogram filtered at 0.5 Hz (U1) has an rS configuration. At the successful ablation site (C), the U1 recording has a qS configuration. Tracings are labeled as in panel A.
Changes in electrogram amplitude can be used to assess creation of ablation lesions in the ventricles, but postablation pacing threshold has been shown to better correlate with lesion size.16
Conclusion Although bipolar recordings provide sufficient information for most catheter mapping purposes, unipolar recordings and manipulation of filtering can provide additional helpful information and are easily implemented with most recording systems. Unipolar electrograms have important potential future applications in the catheter ablation of atrial fibrillation and characterization of substrate for ventricular arrhythmias.
References 1. Delacretaz E, Soejima K, Gottipaty VK, et al. Single catheter determination of local electrogram prematurity using simultaneous unipolar and bipolar recordings to replace the surface ECG as a timing reference. Pacing Clin Electrophysiol 2001;24(4 Pt I):441– 449. 2. de Bakker JMT, Hauer RNW, Simmers TA. Activation mapping: unipolar versus bipolar recording. In Zipes DP, Jalife J, editors : Cardiac Electrophysiology : From Cell to Bedside. Second Edition. Philadelphia: WB Saunders, 1995:1068. 3. Michaud GF, Cutro R. Information at our catheter tips: unipolar electrogram morphology makes another comeback! Heart Rhythm 2010;7:1301–1302. 4. Otomo K, Uno K, Fujiwara H, et al. Local unipolar and bipolar electrogram criteria for evaluating the transmurality of atrial ablation lesions at different catheter orientations relative to the endocardial surface. Heart Rhythm 2010;7: 1291–1300. 5. Hindricks G, Kottkamp H, Chen X, et al. Successful radiofrequency catheter ablation of right sided accessory pathways during sustained atrial fibrillation. Eur Heart J 1995;16:967–970.
796
Heart Rhythm, Vol 8, No 5, May 2011
Figure 5 Recordings from a patient with ventricular tachycardia (VT) due to an anterior wall myocardial infarction. Shown from top in each panel are ECG leads, bipolar recordings from the mapping catheter distal (ABL1-2) and middle (ABL 2-3) electrodes, unipolar 1 high-pass filtered at 30 Hz (U1 hp, U2 hp), and unipolar 1 signal with minimal high-pass filtering (U1). A, B: Sustained monomorphic VT with a cycle length of 510 ms is present. Beneath the tracings is a schematic of the left ventricle viewed from the right anterior oblique projection. Dark gray indicates normal ventricular myocardium. Lighter gray region is the apical infarct, containing branching channels (white). Arrows indicate excitation wavefronts. In A, the recording site is in the reentrant circuit exit. Electrograms precede QRS onset. The U1 signal has a small r wave and large S wave, consistent with spread of the excitation wavefront away from the exit, across the ventricle. Activation of this large mass of tissue obscures some local signals in the infarct border. In B, the recording site is from an isthmus in the center of the infarct region. A local low-amplitude isolated potential is present, on the distal ablation bipolar and unipolar recordings (U1 hp, U1). The U1 signal has less far-field component than at the exit region (A) because it is further from the border of the infarct. The U1 far-field signal has a complex configuration that does not reliably indicate the direction of propagation. C: Recordings during sinus rhythm from an isthmus region in the center of the anterior wall infarct. A late potential is present, visible on the distal bipolar and unipolar recordings and faintly visible on the U2 hp recording.
6. Haissaguerre M, Dartigues JF, Warin JF, et al. Electrogram patterns predictive of successful catheter ablation of accessory pathways. Value of unipolar recording mode. Circulation 1991;84:188 –202. 7. Marchlinski FE, Callans DJ, Gottlieb CD, et al. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation 2000;101:1288 –1296. 8. Soejima K, Suzuki M, Maisel WH, et al. Catheter ablation in patients with multiple and unstable ventricular tachycardias after myocardial infarction: short ablation lines guided by reentry circuit isthmuses and sinus rhythm mapping. Circulation 2001;104:664 – 669. 9. Codreanu A, Odille F, Aliot E, et al. Electroanatomic characterization of post-infarct scars comparison with 3-dimensional myocardial scar reconstruction based on magnetic resonance imaging. J Am Coll Cardiol 2008; 52:839 – 842. 10. Deneke T, Muller KM, Lemke B, et al. Human histopathology of electroanatomic mapping after cooled-tip radiofrequency ablation to treat ventricular tachycardia in remote myocardial infarction. J Cardiovasc Electrophysiol 2005;16: 1246 –1251. 11. Brunckhorst CB, Delacretaz E, Soejima K, et al. Impact of changing activation sequence on bipolar electrogram amplitude for voltage mapping of left ventric-
12.
13.
14.
15.
16.
ular infarcts causing ventricular tachycardia. J Interv Card Electrophysiol 2005;12:137–141. Sivagangabalan G, Pouliopoulos J, Huang K, et al. Comparison of electroanatomic contact and noncontact mapping of ventricular scar in a postinfarct ovine model with intramural needle electrode recording and histological validation. Circ Arrhythm Electrophysiol 2008;1:363–369. Koch KC, vom Dahl J, Wenderdel M, et al. Myocardial viability assessment by endocardial electroanatomic mapping: comparison with metabolic imaging and functional recovery after coronary revascularization. J Am Coll Cardiol 2001; 38:91–98. Wijnmaalen AP, van der Geest RJ, van Huls van Taxis CF, et al. Head-to-head comparison of contrast-enhanced magnetic resonance imaging and electroanatomical voltage mapping to assess post-infarct scar characteristics in patients with ventricular tachycardias: real-time image integration and reversed registration. Eur Heart J 2010;32:104 –114. Polin GM, Haqqani H, Tzou W, et al. Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular dysplasia/cardiomyopathy. Heart Rhythm 2011;8:76 – 83. Sapp JL, Soejima K, Cooper JM, et al. Ablation lesion size correlates with pacing threshold: a physiological basis for use of pacing to assess ablation lesions. Pacing Clin Electrophysiol 2004;27:933–937.