- Email: [email protected]
Characterisation of Atrial Activation Patterns in Human Persistent Atrial Fibrillation: Multiple Wavelets or Focal Drivers?
at CFAE sites were characterised into six types reflecting possible underlying mechanisms: (1) focal activations with radial spread, (2) rotor, (3) passive-WF activation, (4) collision of ≥2 WFs, (5) WF conduction block and (6) disorganised activity. Type 1 and 2 activation patterns are consistent with an active process while types 3–6 represent passive patterns. Results: The % of CFAE seen was PLAW 16.2 ± 13.1%, LAA 10.5 ± 9.6%, RAA 17.6 ± 13.8%, RSPV-LA 13.9 ± 5.8%; p = NS. Wavefront patterns at sites of CFAE were highly heterogeneous with multiple (2.9 ± 1.4) activation subtypes seen per map. Activation at CFAE sites were passive in 56.6 ± 32.4%, disorganised in 28.7 ± 33.9%, wavefront collision in 6 ± 9% and focal in 4.5 ± 7.9%. Focal activations were always intermittent and never sustained. A rotor was only seen in three maps; one sustained rotor (rotational frequency ∼5 Hz) and two intermittent rotors. It was only in the sustained rotor that CFAE sites correlated with the centre of the rotors rotational core. Conclusion: Our study shows that the majority of CFAE in persistent AF are activated passively by wavefronts or disorganised activity. http://dx.doi.org/10.1016/j.hlc.2012.05.291 282 Characterisation of Activation Patterns at Sites of High Dominant Frequency During Human Persistent AF G. Lee 1,∗ , S. Kumar 1 , A. Teh 1 , A. Madry 1 , S. Spence 1 , J. Morton 1 , P. Sparks 1 , P. Kistler 1,2 , J. Kalman 1 1 The 2 The
Royal Melbourne Hospital, Australia Baker IDI and The Alfred Hospital, Australia
Introduction: Experimental animal models suggest that persistent AF is maintained by high frequency drivers that are located within the LA. Dominant frequency (DF) has been used as a tool to identify these drivers; however, activation at sites of high DF has not been characterised. Methods: High-density epicardial mapping of the posterior LA wall, LAA, RAA and RSPV-LA-jxn was performed in 18 pts with persistent AF undergoing openheart surgery. A High DF (HDF) site was identified if local DF was >20% of adjacent sites. A new mapping technique was developed which allowed continuous animation and analysis of activation wavefronts during sustained AF. The activation at HDF sites was characterised into four types: (1) focal activation with radial spread, (2) rotor, (3) passive activation and (4) disorganised activity. Types 1 and 2 represent activation patterns that are consistent with an AF driver. Results: A mean of 3.9 ± 4.0 HDF sites were identified per map with a mean DF of 9.0 ± 2.5 Hz. The majority of activations at HDF sites were classified as passive (60.8 ± 29.7%) or represented disorganised activity (23.2 ± 24.7%). Focal activations accounted for 5.1 ± 6.3%. Three maps showed rotor activity; one showed sustained rotor activity (rotational frequency ∼5 Hz) and the other two showed intermittent activity. It was only in the sustained AF rotor that HDF sites correlated with the centre of the rotors rotational core.
CSANZ 2012 Abstracts
S117
Conclusion: The majority of activations at sites of HDF are passive or represent disorganised activity. They do not identify putative AF drivers in the form of rotors or focal sources. http://dx.doi.org/10.1016/j.hlc.2012.05.292 283 Characterisation of Activation Patterns at Sites of Short Atrial Fibrillation Cycle Length (AFCL) During Human Persistent AF: Do They Identify Drivers? G. Lee 1,∗ , S. Kumar 1 , A. Teh 1 , A. Madry 1 , S. Spence 1 , J. Morton 1 , P. Sparks 1 , P. Kistler 2 , J. Kalman 1 1 The 2 The
Royal Melbourne Hospital, Australia Baker IDI and The Alfred Hospital, Australia
Introduction: AFCL has been used as a surrogate marker to identify sites that act as putative AF drivers; however, activation at these sites has not been characterised. Methods: High-density epicardial mapping of the posterior LA wall, LAA, RAA and RSPV-LA-jxn was performed in 18 pts with persistent AF undergoing openheart surgery. Beat-to-beat AFCL was measured from each electrode during 10 s of AF. Short AFCL driver sites (AFCLD) were defined as sites where mean AFCL was less that the 25th percentile calculated for the plaque. A new mapping technique was developed which allowed dynamic analysis of activation wavefront patterns during sustained AF. Activations at AFCL-D sites were characterised into four types: (1) focal activation with radial spread, (2) rotor, (3) passive activation from adjacent sites and (4) disorganized activity. Types 1 and 2 represent activation patterns consistent with an AF driver. Results: The mean and median plaque AFCL was 162.8 ± 39 ms and 158.3 ± 47.2 ms. 7.5 ± 12.9 electrode sites per plaque were identified as an AFCL-D. The cycle length of AFCL-D sites was significantly shorter than non-AFCLD sites, 111.9 ± 39 ms vs 176.8 ± 38.4 ms, respectively; p < 0.05. Activation patterns at AFCL-D sites were classified as passive (60.8 ± 29.7%) or disorganised activity (23.2 ± 24.7%). Focal activations at AFCL-D were seen in 5.1 ± 6.3% of maps. Only one sustained rotor (5 Hz) was seen, AFCL-D sites correlated with the centre of the rotor’s rotational core. Conclusion: AFCL-D sites are activated passively or represent disorganised activity. They fail to reliably identify LA drivers in the form of rotors or focal activations. http://dx.doi.org/10.1016/j.hlc.2012.05.293 284 Characterisation of Atrial Activation Patterns in Human Persistent Atrial Fibrillation: Multiple Wavelets or Focal Drivers? G. Lee 1,∗ , A. Teh 1 , S. Kumar 1 , A. Madry 1 , S. Spence 1 , J. Morton 1 , P. Sparks 1 , P. Kistler 1,2 , J. Kalman 1 1 The 2 The
Royal Melbourne Hospital, Australia Baker IDI and The Alfred Hospital, Australia
Introduction: The aim of this study was to characterise activation patterns in human persistent AF.
ABSTRACTS
Heart, Lung and Circulation 2012;21:S1–S142
S118
Heart, Lung and Circulation 2012;21:S1–S142
CSANZ 2012 Abstracts
ABSTRACTS
Methods: High-density epicardial mapping of the posterior LA wall, LA and RA appendages (LAA, RAA) and RSPV-LA-jxn was performed in 18 pts with persistent AF. A new mapping technique was developed which allowed dynamic analysis of activation wavefront (WF) patterns during sustained AF. Based on the work by Konings, activations were characterised into seven subtypes: (1) single broad-WF, (2) 1-narrow WF, (3) 2-narrow WFs, (4) ≥3-narrow WFs, (5) rotors, (6) focal activation or (7) disorganized activity. Results: A total of 36 sites were mapped. There was marked heterogeneity in activation patterns between patients and within the same patient. The majority of maps were highly dynamic with multiple unstable patterns seen within the same recording (mean of 3.8 ± 1.6 subtypes/map). Only 3/33 maps (9%) displayed a consistent organised pattern of activation. The most common activation patterns seen were wavefronts (56.6 ± 32.4%) and disorganised activity (28.9 ± 33.7%). Focal activations accounted for 11.8 ± 17.3% of activations and were shortlived (1–2 beats) at each site of origin. Three maps showed rotor activity; one showed sustained rotor activity (rotational frequency ∼5 Hz) and two intermittent activity (1–7 revolutions with a rotational frequency 6–10 Hz). Conclusion: Our data shows that human persistent AF is characterised by highly heterogeneous patterns of wavefront activation and disorganised activity. Organised drivers in the form of rotors and focal activations are rare. http://dx.doi.org/10.1016/j.hlc.2012.05.294 285 Comparison of Pre-acquired CT Versus Real Time Assessment of Oesophageal Position in AF Ablation A. Gavin ∗ , C. Singleton, J. Bowyer, A. McGavigan Cardiology Department, Flinders Medical Centre, Adelaide, Australia Introduction: Atrio-oesophageal fistula is a rare but often fatal complication of catheter ablation for atrial fibrillation (AF). Various strategies are employed to evaluate oesophageal position in relation to the posterior left atrium (LA). These include segmentation of the oesophagus from a pre-acquired CT scan and direct, real time assessment of oesophageal position using contrast at the time of the procedure. Methods: 114 patients with drug-refractory AF underwent CT scanning prior to AF ablation. The LA and oesophagus were segmented from this scan. The oesophagus was deemed ostial if it crossed directly behind any of the pulmonary vein (PV) os, antral if it passed within 5 mm of a PV os or midline. Under general anaesthetic at the time of ablation the same patients were administered contrast via an oro-gastric tube to outline the oesophagus. Catheters were placed at the PV os and the oesophageal position in relation to the PVs was established radiographically using a postero-anterior view. Results: The real-time assessment of oesophageal position correlated with the CT scan in 59% of patients. In 34%
the oesophagus was more right sided on direct visualisation whilst in 7% it was more left sided. Conclusions: Segmentation of the oesophagus from the CT scan did not correlate the real time oesophageal position at the time of the procedure in over 40% of patients under general anaesthesia. Reliance on determination of oesophageal position by previously acquired CT may be misleading at best and provide a false sense of security when ablating in the posterior LA. http://dx.doi.org/10.1016/j.hlc.2012.05.295 286 Concealed and Manifest Premature His Bundle Depolarisations Simulating Heart Block and Tachycardia H. Lim 1,2,∗ , R. Mahajan 1 , M. Alasady 1 , A. Brooks 1 , J. Kim 1 , A. Ganesan 1 , H. Abed 1 , S. Nayyar 1 , K. RobertsThomson 1 , P. Sanders 1 1 Centre for Heart Rhythm Disorders, Royal Adelaide Hospital and The University of Adelaide, Adelaide, Australia 2 Austin and Northern Health, Melbourne, Australia
Introduction: Concealed para-hisian extrasystoles are a rare but potential underlying cause of heart block on surface ECG and have been purported to simulate tachycardia. Methods: N/A. Results: A 58 year-old woman with a history of migraines and depression presented with palpitations and pre-syncopal sympoms. There was no background or family history of cardiac diseases and echocardiography was normal. Holter monitoring showed various degrees of heart block (figure, strip A). There were occasional runs of narrow complex tachycardia, shown in strip B. During electrophysiological study, episodic narrow complex premature beats and bursts of tachycardia were noted with the earliest recordable endocardial activity being the His recording (strip C). Interestingly, there was no retrograde conduction to the atria which remained with sinus activation. In addition, His ectopy was observed to induce heart block (strip D). Concealed and manifest premature His depolarisations should be suspected in cases of unexplained PR prolongation, heart block, blocked atrial beats, retrogradely conducted atrial beats and premature junctional beats. Conclusions: Concealed and manifest premature depolarisations from the His bundle can mimic heart block and other forms of tachycardia. This diagnosis should be suspected when premature junctional beats are observed in conjunction with intermittent first and second degree heart block and non-conducted atrial premature beats.
Royal Melbourne Hospital, Australia Baker IDI and The Alfred Hospital, Australia
Introduction: Experimental animal models suggest that persistent AF is maintained by high frequency drivers that are located within the LA. Dominant frequency (DF) has been used as a tool to identify these drivers; however, activation at sites of high DF has not been characterised. Methods: High-density epicardial mapping of the posterior LA wall, LAA, RAA and RSPV-LA-jxn was performed in 18 pts with persistent AF undergoing openheart surgery. A High DF (HDF) site was identified if local DF was >20% of adjacent sites. A new mapping technique was developed which allowed continuous animation and analysis of activation wavefronts during sustained AF. The activation at HDF sites was characterised into four types: (1) focal activation with radial spread, (2) rotor, (3) passive activation and (4) disorganised activity. Types 1 and 2 represent activation patterns that are consistent with an AF driver. Results: A mean of 3.9 ± 4.0 HDF sites were identified per map with a mean DF of 9.0 ± 2.5 Hz. The majority of activations at HDF sites were classified as passive (60.8 ± 29.7%) or represented disorganised activity (23.2 ± 24.7%). Focal activations accounted for 5.1 ± 6.3%. Three maps showed rotor activity; one showed sustained rotor activity (rotational frequency ∼5 Hz) and the other two showed intermittent activity. It was only in the sustained AF rotor that HDF sites correlated with the centre of the rotors rotational core.
CSANZ 2012 Abstracts
S117
Conclusion: The majority of activations at sites of HDF are passive or represent disorganised activity. They do not identify putative AF drivers in the form of rotors or focal sources. http://dx.doi.org/10.1016/j.hlc.2012.05.292 283 Characterisation of Activation Patterns at Sites of Short Atrial Fibrillation Cycle Length (AFCL) During Human Persistent AF: Do They Identify Drivers? G. Lee 1,∗ , S. Kumar 1 , A. Teh 1 , A. Madry 1 , S. Spence 1 , J. Morton 1 , P. Sparks 1 , P. Kistler 2 , J. Kalman 1 1 The 2 The
Royal Melbourne Hospital, Australia Baker IDI and The Alfred Hospital, Australia
Introduction: AFCL has been used as a surrogate marker to identify sites that act as putative AF drivers; however, activation at these sites has not been characterised. Methods: High-density epicardial mapping of the posterior LA wall, LAA, RAA and RSPV-LA-jxn was performed in 18 pts with persistent AF undergoing openheart surgery. Beat-to-beat AFCL was measured from each electrode during 10 s of AF. Short AFCL driver sites (AFCLD) were defined as sites where mean AFCL was less that the 25th percentile calculated for the plaque. A new mapping technique was developed which allowed dynamic analysis of activation wavefront patterns during sustained AF. Activations at AFCL-D sites were characterised into four types: (1) focal activation with radial spread, (2) rotor, (3) passive activation from adjacent sites and (4) disorganized activity. Types 1 and 2 represent activation patterns consistent with an AF driver. Results: The mean and median plaque AFCL was 162.8 ± 39 ms and 158.3 ± 47.2 ms. 7.5 ± 12.9 electrode sites per plaque were identified as an AFCL-D. The cycle length of AFCL-D sites was significantly shorter than non-AFCLD sites, 111.9 ± 39 ms vs 176.8 ± 38.4 ms, respectively; p < 0.05. Activation patterns at AFCL-D sites were classified as passive (60.8 ± 29.7%) or disorganised activity (23.2 ± 24.7%). Focal activations at AFCL-D were seen in 5.1 ± 6.3% of maps. Only one sustained rotor (5 Hz) was seen, AFCL-D sites correlated with the centre of the rotor’s rotational core. Conclusion: AFCL-D sites are activated passively or represent disorganised activity. They fail to reliably identify LA drivers in the form of rotors or focal activations. http://dx.doi.org/10.1016/j.hlc.2012.05.293 284 Characterisation of Atrial Activation Patterns in Human Persistent Atrial Fibrillation: Multiple Wavelets or Focal Drivers? G. Lee 1,∗ , A. Teh 1 , S. Kumar 1 , A. Madry 1 , S. Spence 1 , J. Morton 1 , P. Sparks 1 , P. Kistler 1,2 , J. Kalman 1 1 The 2 The
Royal Melbourne Hospital, Australia Baker IDI and The Alfred Hospital, Australia
Introduction: The aim of this study was to characterise activation patterns in human persistent AF.
ABSTRACTS
Heart, Lung and Circulation 2012;21:S1–S142
S118
Heart, Lung and Circulation 2012;21:S1–S142
CSANZ 2012 Abstracts
ABSTRACTS
Methods: High-density epicardial mapping of the posterior LA wall, LA and RA appendages (LAA, RAA) and RSPV-LA-jxn was performed in 18 pts with persistent AF. A new mapping technique was developed which allowed dynamic analysis of activation wavefront (WF) patterns during sustained AF. Based on the work by Konings, activations were characterised into seven subtypes: (1) single broad-WF, (2) 1-narrow WF, (3) 2-narrow WFs, (4) ≥3-narrow WFs, (5) rotors, (6) focal activation or (7) disorganized activity. Results: A total of 36 sites were mapped. There was marked heterogeneity in activation patterns between patients and within the same patient. The majority of maps were highly dynamic with multiple unstable patterns seen within the same recording (mean of 3.8 ± 1.6 subtypes/map). Only 3/33 maps (9%) displayed a consistent organised pattern of activation. The most common activation patterns seen were wavefronts (56.6 ± 32.4%) and disorganised activity (28.9 ± 33.7%). Focal activations accounted for 11.8 ± 17.3% of activations and were shortlived (1–2 beats) at each site of origin. Three maps showed rotor activity; one showed sustained rotor activity (rotational frequency ∼5 Hz) and two intermittent activity (1–7 revolutions with a rotational frequency 6–10 Hz). Conclusion: Our data shows that human persistent AF is characterised by highly heterogeneous patterns of wavefront activation and disorganised activity. Organised drivers in the form of rotors and focal activations are rare. http://dx.doi.org/10.1016/j.hlc.2012.05.294 285 Comparison of Pre-acquired CT Versus Real Time Assessment of Oesophageal Position in AF Ablation A. Gavin ∗ , C. Singleton, J. Bowyer, A. McGavigan Cardiology Department, Flinders Medical Centre, Adelaide, Australia Introduction: Atrio-oesophageal fistula is a rare but often fatal complication of catheter ablation for atrial fibrillation (AF). Various strategies are employed to evaluate oesophageal position in relation to the posterior left atrium (LA). These include segmentation of the oesophagus from a pre-acquired CT scan and direct, real time assessment of oesophageal position using contrast at the time of the procedure. Methods: 114 patients with drug-refractory AF underwent CT scanning prior to AF ablation. The LA and oesophagus were segmented from this scan. The oesophagus was deemed ostial if it crossed directly behind any of the pulmonary vein (PV) os, antral if it passed within 5 mm of a PV os or midline. Under general anaesthetic at the time of ablation the same patients were administered contrast via an oro-gastric tube to outline the oesophagus. Catheters were placed at the PV os and the oesophageal position in relation to the PVs was established radiographically using a postero-anterior view. Results: The real-time assessment of oesophageal position correlated with the CT scan in 59% of patients. In 34%
the oesophagus was more right sided on direct visualisation whilst in 7% it was more left sided. Conclusions: Segmentation of the oesophagus from the CT scan did not correlate the real time oesophageal position at the time of the procedure in over 40% of patients under general anaesthesia. Reliance on determination of oesophageal position by previously acquired CT may be misleading at best and provide a false sense of security when ablating in the posterior LA. http://dx.doi.org/10.1016/j.hlc.2012.05.295 286 Concealed and Manifest Premature His Bundle Depolarisations Simulating Heart Block and Tachycardia H. Lim 1,2,∗ , R. Mahajan 1 , M. Alasady 1 , A. Brooks 1 , J. Kim 1 , A. Ganesan 1 , H. Abed 1 , S. Nayyar 1 , K. RobertsThomson 1 , P. Sanders 1 1 Centre for Heart Rhythm Disorders, Royal Adelaide Hospital and The University of Adelaide, Adelaide, Australia 2 Austin and Northern Health, Melbourne, Australia
Introduction: Concealed para-hisian extrasystoles are a rare but potential underlying cause of heart block on surface ECG and have been purported to simulate tachycardia. Methods: N/A. Results: A 58 year-old woman with a history of migraines and depression presented with palpitations and pre-syncopal sympoms. There was no background or family history of cardiac diseases and echocardiography was normal. Holter monitoring showed various degrees of heart block (figure, strip A). There were occasional runs of narrow complex tachycardia, shown in strip B. During electrophysiological study, episodic narrow complex premature beats and bursts of tachycardia were noted with the earliest recordable endocardial activity being the His recording (strip C). Interestingly, there was no retrograde conduction to the atria which remained with sinus activation. In addition, His ectopy was observed to induce heart block (strip D). Concealed and manifest premature His depolarisations should be suspected in cases of unexplained PR prolongation, heart block, blocked atrial beats, retrogradely conducted atrial beats and premature junctional beats. Conclusions: Concealed and manifest premature depolarisations from the His bundle can mimic heart block and other forms of tachycardia. This diagnosis should be suspected when premature junctional beats are observed in conjunction with intermittent first and second degree heart block and non-conducted atrial premature beats.