J Interv Card Electrophysiol DOI 10.1007/s10840-015-0008-2
MULTIMEDIA REPORT
Detection of left atrial thrombus by intracardiac echocardiography in patients undergoing ablation of atrial fibrillation Chenni S. Sriram 1 & Javier E. Banchs 1 & Talal Moukabary 1 & Raman Moradkhan 1 & Mario D. Gonzalez 1,2
Received: 11 February 2015 / Accepted: 6 April 2015 # Springer Science+Business Media New York 2015
Abstract Background The role of intracardiac echocardiography (ICE) to detect thrombus within left atrium (LA) before atrial fibrillation (AF) ablation despite a recent transesophageal echocardiogram (TEE) is not well defined. We examined the prevalence of LA/left atrial appendage (LAA) thrombus using ICE immediately prior to AF ablation in patients in whom anticoagulation was not withheld. Methods We analyzed 122 consecutive patients (62.6 ± 10.8 years, 90 males, CHA2DS2-VASc score 2.4±1.5, persistent AF 29.5 %) who underwent an ICE-guided AF ablation 1 day after a negative (n=120) or inconclusive (n=2) TEE for LA thrombus. LA was imaged with ICE from the right atrium, coronary sinus, and right ventricular inflow tract (RVIT). ICE and TEE images were compared for LAA area, thrombus, and spontaneous echo contrast (SEC). Results LAA was adequately visualized in 99 and 100 % of patients with TEE and ICE, respectively. RVIT was the best ICE view for LAA visualization. The LAA 2-D-area measured by TEE was 4.9± 0.5 vs. 5 ± 0.5 cm2 by ICE (P=NS). ICE identified a thrombus in seven patients with a previous negative TEE, leading to cancellation of ablaElectronic supplementary material The online version of this article (doi:10.1007/s10840-015-0008-2) contains supplementary material, which is available to authorized users. * Mario D. Gonzalez [email protected] 1
Heart & Vascular Institute, Penn State University, Hershey, PA, USA
2
Penn State Heart & Vascular Institute, Milton S. Hershey Medical Center, Penn State University, 500 University Drive, Hershey, PA 17033, USA
tion. It ruled out a thrombus in two patients with an inconclusive TEE. Thrombi were found in the LAA (n=4), atrial septum (n = 2), and left superior pulmonary vein (n= 1). SEC during TEE was more frequent in patients with thrombus on ICE than those without (85.7 vs. 17.4 %; p=0.03; positive predictive value 23.1 %, negative predictive value 98.9 %). Conclusions The results of our staged imaging approach suggest that ICE has a complimentary value in re-screening the LA/LAA for thrombus after a recent negative or equivocal TEE. The presence of SEC during TEE increases the probability of finding a thrombus with ICE, which could potentially be dislodged during catheter manipulation.
Keywords Atrial fibrillation . Left atrium . Left atrial appendage . Intracardiac echocardiography . Spontaneous echo contrast . Thrombus . Transesophageal echocardiography
Abbreviations AF Atrial fibrillation BMI Body mass index CI Confidence interval ICE Intracardiac echocardiography INR International normalized ratio LA Left atrium LAA Left atrial appendage LVEF Left ventricular ejection fraction RVIT Right ventricular inflow tract SEC Spontaneous echo contrast TEE Transesophageal echocardiography
J Interv Card Electrophysiol
1 Introduction The role of intracardiac echocardiography (ICE) to guide ablation of atrial fibrillation (AF) and other complex arrhythmias is well recognized [1–14]. However, the additional role of ICE in identifying an atrial thrombus immediately before ablation of AF is not well studied. Transesophageal echocardiography (TEE) is routinely used prior to ablation of AF and remains the gold standard to exclude thrombus within the left atrial appendage (LAA) [15]. Despite a negative TEE result, catheter ablation of AF can result in manifest or silent embolic events [16]. Several mechanisms have been implicated, including de novo thrombus formation, endothelial damage, air embolism, and inadequate anticoagulation [17]. However, catheter dislodgement of an overlooked thrombus within the left atrium (LA) is a potential mechanism not previously recognized. Therefore, in the present study, we analyzed the prevalence and location of LA/LAA thrombus detected with ICE immediately before ablation of AF following a TEE performed the day before the procedure.
2.2 Electrophysiology study and intracardiac echocardiography All procedures were performed during therapeutic anticoagulation under sedation with intravenous propofol. To monitor arterial blood pressure, a small cannula was placed in the left radial artery. Venous access was obtained using vascular ultrasound guidance (Site Rite, Bard) to prevent inadvertent arterial puncture. A bolus of 5000 units of heparin was administered immediately after vascular access. Four multipolar electrode catheters were percutaneously advanced into the coronary sinus, His bundle region, right atrial appendage, and right ventricle. A 10-Fr ICE catheter (Sound Star, 64-element, 5.5–10.0 MHz, Biosense Webster, Inc.) was advanced percutaneously through the left femoral vein into the right atrium. Ultrasound images of the LA, pulmonary veins, mitral annulus, left atrial appendage, and esophagus were acquired and processed by the Carto-Sound system (Carto-3, Biosense Webster Inc.). Even though all patients underwent TEE the day before the procedure, the left atrium and left atrial appendage were re-screened for potential atrial thrombus.
2 Materials and methods 2.1 Patients
2.3 Transesophageal echocardiography and intra-cardiac echocardiographic imaging
We analyzed 122 consecutive patients at our institution who underwent ICE-guided AF ablation preceded by a negative (n = 120) or inconclusive (n = 2) pre-procedural TEE. Written informed consent was obtained, and the study was approved by the Institutional Review Board. In each patient, we recorded the following demographic information: age, sex, body mass index (BMI), left ventricular ejection fraction (LVEF), type of AF [15] (paroxysmal, persistent, long- standing persistent), CHA2DS2VASc score [18], history of any previous ablation for AF, type of oral anti-coagulant used, and presenting cardiac rhythm. In patients anticoagulated with warfarin, international normalized ratio (INR) was checked once a week on the month before the ablation to ascertain that they were indeed in the therapeutic range. An INR obtained on the day of the procedure also confirmed that they were therapeutic. The presence or absence of a pericardial effusion observed with ICE at the start and end of the procedure was documented [3, 19], and a limited transthoracic echocardiogram to rule out pericardial effusion was obtained in all patients approximately 4 h after the procedure. Any major peri-procedural complication that was documented included: cardiac tamponade, significant pericardial effusion, transient ischemic attack/stroke, femoral arterial–venous fistula, pseudo-aneurysm, and hematoma requiring blood transfusion. The cardiac rhythm at discharge was also noted.
Online analysis of the recorded clips of transesophageal echocardiography and intra-cardiac echocardiography using the Synapse™ ProSolv® cardiovascular-Version-4.0.4 imaging software (FUJIFILM Medical Systems USA, Inc.) was reviewed by two independent operators. SEC was defined as dynamic smoke-like non-homogenous amorphous echos with the LA/LAA, with a characteristic slow swirling motion that was distinct from white noise artifact. During ICE imaging, we routinely screened for SEC at 5.5–10 MHz (mostly at 7.5 MHz) and distinguished this from background noise and speckle by manipulation of the gain settings. SEC was subjectively graded as mild, moderate, or severe. LA/LAA thrombus was diagnosed if there was a clearly defined circumscribed echo-dense intracavitary mass that was acoustically distinct from the endocardium and pectinate muscles of the LAA. Presence of thrombus was corroborated in two or more views and the largest measurement of thrombus size (mm2) was taken [20–22]. LA sludge was defined subjectively as an echo dense lesion more viscid than SEC, but less dense than thrombus [23]. The TEE was performed according to standard guidelines [24] to exclude any LA/LAA thrombus 1 day prior to the scheduled ablation. Echo contrast was not routinely used during the TEE exam. The LAA was best visualized with the transducer probe in the mid-esophagus, and panoramic views of LA/LAA were obtained by gradually varying the degree of rotation of the probe and the angle of interrogation. Detailed
J Interv Card Electrophysiol
views of the pectinate muscles and lobes of the LAA were thus obtained. The largest 2-D cross-sectional area (cm2) of the LAA was calculated from the best visualized view. The presence or absence of SEC or thrombus within the LA/LAA was noted after optimization of the various parameters such as frame rates, harmonics, and dynamic range. This cohort of patients with a negative or inconclusive TEE result for LA/LAA thrombus was subsequently scheduled to undergo radiofrequency ablation using a CARTO 3 system (Biosense Webster, CA, USA). ICE imaging was obtained during ablation of atrial fibrillation to identify LA/LAA thrombus and to reconstruct the anatomy of the LA, LAA, and pulmonary veins [3]. The CARTO-Sound image integration system was obtained using a 10-Fr (3.3 mm tip) 64-element phased-array (5.5 to 10 MHz) ICE catheter (SoundStar, Biosense Webster, CA, USA) percutaneously inserted through an 11-Fr sheath via the left femoral vein. The ICE catheter was initially advanced into the right atrium under fluoroscopic guidance. The ICE catheter manipulation included rotation, advancement and withdrawal, and the use of two jog dials (antero-posterior and right to left deflections). Imaging of the LA and LAA was obtained from the right atrium, proximal coronary sinus, and right ventricular inflow tract (RVIT). The RVIT, due to its unique location in relation to the aorta, mitral, annulus and LAA [25], offers an excellent location to adequately visualize these structures. To obtain an adequate view of the LAA, the tip of the ICE catheter was advanced just beyond the tricuspid valve and the transducer, imaging the leaflets of the aortic valve, the LAA, and the left superior pulmonary vein (Fig. 1a, b). Multiple planes were obtained for threedimensional reconstructions of the LA, LAA, and pulmonary veins. Most ICE imaging was performed with the transducer set at 7.5 MHz, and fine-tuning was done to optimize the image quality, taking frame rates, harmonics, and dynamic range into consideration [3]. Different panoramic views of the inter-atrial septum, LA, LAA, and individual pulmonary veins were obtained from the right atrium by gradual clockwise rotation of the catheter. The ICE catheter was then deflected across the tricuspid valve using the jog dial and gently advanced into the right ventricular inflow tract. By gradually rotating the ICE catheter in a clockwise direction within the RVIT, the LA, the aortic valve leaflets in short axis, and the LAA (visualized from its mouth upto the apex) were observed. The ICE catheter was then withdrawn across the tricuspid annulus and positioned at/just within the proximal coronary sinus using the pre-existent CS electrophysiological catheter as a landmark. Sonographic images of the LA roof, LAA, right superior, and left superior pulmonary veins were thus obtained. In each patient, the best visualized images of the LAA obtained by TEE and ICE were compared both qualitatively and quantitatively by measuring the largest 2-D cross-
a
ICE Catheter Tip in RVIT RAO
LAO
ICE ICE CS os
b
CS os
LAA Visualized from RVIT
LAAA
Fig. 1 a ICE catheter tip in RVIT. Fluoroscopy images showing the tip of the intracardiac echo catheter positioned in the right ventricular inflow tract as seen in the right anterior oblique (left image) and left anterior oblique (right image) views. Coronary sinus ostium (CS os) is also labelled. b LAA visualized from RVIT. ICE image obtained from the RVIT position shows adequate visualization of the LAA
sectional area. The LA/LAA was re-screened for thrombus using ICE in all patients despite a negative result on the previous TEE. We analyzed the dataset for any association between SEC within the LA/LAA during TEE to subsequently detect LA/LAA thrombus during ICE. We documented the presence or absence of SEC within the LA/LAA with ICE imaging. Those patients in whom a thrombus was found with ICE, we requested an independent review of the previously performed TEE. 2.4 Statistical analysis Data are expressed as mean±SD along with 95 % confidence interval (CI) as well as percentage when appropriate. A twotailed Student’s paired t test was used to compare the LAA cross-sectional area (cm2) obtained with each imaging modality. Other continuous and non-continuous variables were compared by the use of Wilcoxon rank sum test and Fisher’s exact test, respectively. Pearson’s correlation coefficient (r) was calculated for the presence or absence of SEC between TEE and
J Interv Card Electrophysiol
ICE. Binary logistic regression analysis was performed to analyze the effect of multiple discrete independent variables on the presence or absence of any LA/LAA thrombus detected with ICE. A P value
MULTIMEDIA REPORT
Detection of left atrial thrombus by intracardiac echocardiography in patients undergoing ablation of atrial fibrillation Chenni S. Sriram 1 & Javier E. Banchs 1 & Talal Moukabary 1 & Raman Moradkhan 1 & Mario D. Gonzalez 1,2
Received: 11 February 2015 / Accepted: 6 April 2015 # Springer Science+Business Media New York 2015
Abstract Background The role of intracardiac echocardiography (ICE) to detect thrombus within left atrium (LA) before atrial fibrillation (AF) ablation despite a recent transesophageal echocardiogram (TEE) is not well defined. We examined the prevalence of LA/left atrial appendage (LAA) thrombus using ICE immediately prior to AF ablation in patients in whom anticoagulation was not withheld. Methods We analyzed 122 consecutive patients (62.6 ± 10.8 years, 90 males, CHA2DS2-VASc score 2.4±1.5, persistent AF 29.5 %) who underwent an ICE-guided AF ablation 1 day after a negative (n=120) or inconclusive (n=2) TEE for LA thrombus. LA was imaged with ICE from the right atrium, coronary sinus, and right ventricular inflow tract (RVIT). ICE and TEE images were compared for LAA area, thrombus, and spontaneous echo contrast (SEC). Results LAA was adequately visualized in 99 and 100 % of patients with TEE and ICE, respectively. RVIT was the best ICE view for LAA visualization. The LAA 2-D-area measured by TEE was 4.9± 0.5 vs. 5 ± 0.5 cm2 by ICE (P=NS). ICE identified a thrombus in seven patients with a previous negative TEE, leading to cancellation of ablaElectronic supplementary material The online version of this article (doi:10.1007/s10840-015-0008-2) contains supplementary material, which is available to authorized users. * Mario D. Gonzalez [email protected] 1
Heart & Vascular Institute, Penn State University, Hershey, PA, USA
2
Penn State Heart & Vascular Institute, Milton S. Hershey Medical Center, Penn State University, 500 University Drive, Hershey, PA 17033, USA
tion. It ruled out a thrombus in two patients with an inconclusive TEE. Thrombi were found in the LAA (n=4), atrial septum (n = 2), and left superior pulmonary vein (n= 1). SEC during TEE was more frequent in patients with thrombus on ICE than those without (85.7 vs. 17.4 %; p=0.03; positive predictive value 23.1 %, negative predictive value 98.9 %). Conclusions The results of our staged imaging approach suggest that ICE has a complimentary value in re-screening the LA/LAA for thrombus after a recent negative or equivocal TEE. The presence of SEC during TEE increases the probability of finding a thrombus with ICE, which could potentially be dislodged during catheter manipulation.
Keywords Atrial fibrillation . Left atrium . Left atrial appendage . Intracardiac echocardiography . Spontaneous echo contrast . Thrombus . Transesophageal echocardiography
Abbreviations AF Atrial fibrillation BMI Body mass index CI Confidence interval ICE Intracardiac echocardiography INR International normalized ratio LA Left atrium LAA Left atrial appendage LVEF Left ventricular ejection fraction RVIT Right ventricular inflow tract SEC Spontaneous echo contrast TEE Transesophageal echocardiography
J Interv Card Electrophysiol
1 Introduction The role of intracardiac echocardiography (ICE) to guide ablation of atrial fibrillation (AF) and other complex arrhythmias is well recognized [1–14]. However, the additional role of ICE in identifying an atrial thrombus immediately before ablation of AF is not well studied. Transesophageal echocardiography (TEE) is routinely used prior to ablation of AF and remains the gold standard to exclude thrombus within the left atrial appendage (LAA) [15]. Despite a negative TEE result, catheter ablation of AF can result in manifest or silent embolic events [16]. Several mechanisms have been implicated, including de novo thrombus formation, endothelial damage, air embolism, and inadequate anticoagulation [17]. However, catheter dislodgement of an overlooked thrombus within the left atrium (LA) is a potential mechanism not previously recognized. Therefore, in the present study, we analyzed the prevalence and location of LA/LAA thrombus detected with ICE immediately before ablation of AF following a TEE performed the day before the procedure.
2.2 Electrophysiology study and intracardiac echocardiography All procedures were performed during therapeutic anticoagulation under sedation with intravenous propofol. To monitor arterial blood pressure, a small cannula was placed in the left radial artery. Venous access was obtained using vascular ultrasound guidance (Site Rite, Bard) to prevent inadvertent arterial puncture. A bolus of 5000 units of heparin was administered immediately after vascular access. Four multipolar electrode catheters were percutaneously advanced into the coronary sinus, His bundle region, right atrial appendage, and right ventricle. A 10-Fr ICE catheter (Sound Star, 64-element, 5.5–10.0 MHz, Biosense Webster, Inc.) was advanced percutaneously through the left femoral vein into the right atrium. Ultrasound images of the LA, pulmonary veins, mitral annulus, left atrial appendage, and esophagus were acquired and processed by the Carto-Sound system (Carto-3, Biosense Webster Inc.). Even though all patients underwent TEE the day before the procedure, the left atrium and left atrial appendage were re-screened for potential atrial thrombus.
2 Materials and methods 2.1 Patients
2.3 Transesophageal echocardiography and intra-cardiac echocardiographic imaging
We analyzed 122 consecutive patients at our institution who underwent ICE-guided AF ablation preceded by a negative (n = 120) or inconclusive (n = 2) pre-procedural TEE. Written informed consent was obtained, and the study was approved by the Institutional Review Board. In each patient, we recorded the following demographic information: age, sex, body mass index (BMI), left ventricular ejection fraction (LVEF), type of AF [15] (paroxysmal, persistent, long- standing persistent), CHA2DS2VASc score [18], history of any previous ablation for AF, type of oral anti-coagulant used, and presenting cardiac rhythm. In patients anticoagulated with warfarin, international normalized ratio (INR) was checked once a week on the month before the ablation to ascertain that they were indeed in the therapeutic range. An INR obtained on the day of the procedure also confirmed that they were therapeutic. The presence or absence of a pericardial effusion observed with ICE at the start and end of the procedure was documented [3, 19], and a limited transthoracic echocardiogram to rule out pericardial effusion was obtained in all patients approximately 4 h after the procedure. Any major peri-procedural complication that was documented included: cardiac tamponade, significant pericardial effusion, transient ischemic attack/stroke, femoral arterial–venous fistula, pseudo-aneurysm, and hematoma requiring blood transfusion. The cardiac rhythm at discharge was also noted.
Online analysis of the recorded clips of transesophageal echocardiography and intra-cardiac echocardiography using the Synapse™ ProSolv® cardiovascular-Version-4.0.4 imaging software (FUJIFILM Medical Systems USA, Inc.) was reviewed by two independent operators. SEC was defined as dynamic smoke-like non-homogenous amorphous echos with the LA/LAA, with a characteristic slow swirling motion that was distinct from white noise artifact. During ICE imaging, we routinely screened for SEC at 5.5–10 MHz (mostly at 7.5 MHz) and distinguished this from background noise and speckle by manipulation of the gain settings. SEC was subjectively graded as mild, moderate, or severe. LA/LAA thrombus was diagnosed if there was a clearly defined circumscribed echo-dense intracavitary mass that was acoustically distinct from the endocardium and pectinate muscles of the LAA. Presence of thrombus was corroborated in two or more views and the largest measurement of thrombus size (mm2) was taken [20–22]. LA sludge was defined subjectively as an echo dense lesion more viscid than SEC, but less dense than thrombus [23]. The TEE was performed according to standard guidelines [24] to exclude any LA/LAA thrombus 1 day prior to the scheduled ablation. Echo contrast was not routinely used during the TEE exam. The LAA was best visualized with the transducer probe in the mid-esophagus, and panoramic views of LA/LAA were obtained by gradually varying the degree of rotation of the probe and the angle of interrogation. Detailed
J Interv Card Electrophysiol
views of the pectinate muscles and lobes of the LAA were thus obtained. The largest 2-D cross-sectional area (cm2) of the LAA was calculated from the best visualized view. The presence or absence of SEC or thrombus within the LA/LAA was noted after optimization of the various parameters such as frame rates, harmonics, and dynamic range. This cohort of patients with a negative or inconclusive TEE result for LA/LAA thrombus was subsequently scheduled to undergo radiofrequency ablation using a CARTO 3 system (Biosense Webster, CA, USA). ICE imaging was obtained during ablation of atrial fibrillation to identify LA/LAA thrombus and to reconstruct the anatomy of the LA, LAA, and pulmonary veins [3]. The CARTO-Sound image integration system was obtained using a 10-Fr (3.3 mm tip) 64-element phased-array (5.5 to 10 MHz) ICE catheter (SoundStar, Biosense Webster, CA, USA) percutaneously inserted through an 11-Fr sheath via the left femoral vein. The ICE catheter was initially advanced into the right atrium under fluoroscopic guidance. The ICE catheter manipulation included rotation, advancement and withdrawal, and the use of two jog dials (antero-posterior and right to left deflections). Imaging of the LA and LAA was obtained from the right atrium, proximal coronary sinus, and right ventricular inflow tract (RVIT). The RVIT, due to its unique location in relation to the aorta, mitral, annulus and LAA [25], offers an excellent location to adequately visualize these structures. To obtain an adequate view of the LAA, the tip of the ICE catheter was advanced just beyond the tricuspid valve and the transducer, imaging the leaflets of the aortic valve, the LAA, and the left superior pulmonary vein (Fig. 1a, b). Multiple planes were obtained for threedimensional reconstructions of the LA, LAA, and pulmonary veins. Most ICE imaging was performed with the transducer set at 7.5 MHz, and fine-tuning was done to optimize the image quality, taking frame rates, harmonics, and dynamic range into consideration [3]. Different panoramic views of the inter-atrial septum, LA, LAA, and individual pulmonary veins were obtained from the right atrium by gradual clockwise rotation of the catheter. The ICE catheter was then deflected across the tricuspid valve using the jog dial and gently advanced into the right ventricular inflow tract. By gradually rotating the ICE catheter in a clockwise direction within the RVIT, the LA, the aortic valve leaflets in short axis, and the LAA (visualized from its mouth upto the apex) were observed. The ICE catheter was then withdrawn across the tricuspid annulus and positioned at/just within the proximal coronary sinus using the pre-existent CS electrophysiological catheter as a landmark. Sonographic images of the LA roof, LAA, right superior, and left superior pulmonary veins were thus obtained. In each patient, the best visualized images of the LAA obtained by TEE and ICE were compared both qualitatively and quantitatively by measuring the largest 2-D cross-
a
ICE Catheter Tip in RVIT RAO
LAO
ICE ICE CS os
b
CS os
LAA Visualized from RVIT
LAAA
Fig. 1 a ICE catheter tip in RVIT. Fluoroscopy images showing the tip of the intracardiac echo catheter positioned in the right ventricular inflow tract as seen in the right anterior oblique (left image) and left anterior oblique (right image) views. Coronary sinus ostium (CS os) is also labelled. b LAA visualized from RVIT. ICE image obtained from the RVIT position shows adequate visualization of the LAA
sectional area. The LA/LAA was re-screened for thrombus using ICE in all patients despite a negative result on the previous TEE. We analyzed the dataset for any association between SEC within the LA/LAA during TEE to subsequently detect LA/LAA thrombus during ICE. We documented the presence or absence of SEC within the LA/LAA with ICE imaging. Those patients in whom a thrombus was found with ICE, we requested an independent review of the previously performed TEE. 2.4 Statistical analysis Data are expressed as mean±SD along with 95 % confidence interval (CI) as well as percentage when appropriate. A twotailed Student’s paired t test was used to compare the LAA cross-sectional area (cm2) obtained with each imaging modality. Other continuous and non-continuous variables were compared by the use of Wilcoxon rank sum test and Fisher’s exact test, respectively. Pearson’s correlation coefficient (r) was calculated for the presence or absence of SEC between TEE and
J Interv Card Electrophysiol
ICE. Binary logistic regression analysis was performed to analyze the effect of multiple discrete independent variables on the presence or absence of any LA/LAA thrombus detected with ICE. A P value
