Resuscitation 85 (2014) 689–693
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Experimental paper
Left ventricular thrombus development during ventricular fibrillation and resolution during resuscitation in a swine model of sudden cardiac arrest夽 Gavin R. Budhram ∗ , Timothy J. Mader, Lucienne Lutfy, David Murman, Abdullah Almulhim Tufts University School of Medicine, Department of Emergency Medicine, Baystate Medical Center, 759 Chestnut Street, Springfield, MA 01199, United States
a r t i c l e
i n f o
Article history: Received 10 August 2013 Received in revised form 16 January 2014 Accepted 22 January 2014 Keywords: Cardiac arrest Ventricular fibrillation Intracardiac thrombus Left ventricular thrombus Ultrasound Echocardiography
a b s t r a c t Background: Intracardiac thrombus is a well-known complication of low-flow cardiac states including acute myocardial infarction and atrial fibrillation. Little is known, however, about the formation of intracardiac (left ventricular [LV]) thrombus during the extreme low-flow state of cardiac arrest. Objective: Using a swine model of sudden cardiac arrest, we examined the sonographic development of LV thrombus over time after induction of ventricular fibrillation (VF) and resolution of thrombus with cardiopulmonary resuscitation (CPR). Methods: This observational study was IACUC approved. Forty-five Yorkshire swine were sedated, intubated, and instrumented under general anesthesia before VF was electrically induced. Sonographic data was collected immediately after VF induction and at 2-min intervals thereafter. Following 12 min of untreated VF, resuscitation was initiated with closed chest compressions using an oxygen-powered mechanical resuscitation device. Observations were continued during attempted resuscitation. At the end of the experiment, the animals were euthanized while still at a surgical depth of anesthesia. The data was analyzed descriptively. Results: Sonographic evidence of LV thrombus was observed in 43/45 animals (95.6% [95%CI: 85.2%, 98.8%]). Thrombus was detected within 6 min in 39/45 (86.7% [95%CI: 73.8%, 93.8%]) animals that developed thrombus. Thrombus resolved within 2 min after initiation of chest compressions in 31/43 (72.1% [95%CI: 57.3%, 83.3%]) animals. Conclusion: Similar to other low-flow cardiac states, LV thrombus develops early in the natural history of VF arrest and resolves quickly once forward flow is re-established by chest compressions. Institutional protocol number: 154600-8. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Intracardiac thrombus formation is known to complicate several low-flow cardiac states, including severe acute myocardial infarction (AMI), atrial fibrillation (AF), and severe cardiomyopathy. The initial clinical model for studying left ventricular thrombus was AMI because the onset of the pathophysiological process could be identified and all three prerequisites for thrombus formation are present (endothelial injury, hypercoagulable state, and blood stasis). Decreased left ventricular systolic performance after AMI is
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2014.01.030. ∗ Corresponding author. E-mail address: [email protected] (G.R. Budhram). http://dx.doi.org/10.1016/j.resuscitation.2014.01.030 0300-9572/© 2014 Elsevier Ireland Ltd. All rights reserved.
known to correlate strongly with the formation of LV thrombus.1–6 Intracardiac thrombus formation complicates 7–46% of AMIs, most frequently with acute anterior or apical infarctions.7–9 Thrombi tend to develop within the first week after infarction, and are preceded by severe apical wall-motion abnormalities characterized by akinesis, dyskinesis, or frank aneurysm.3,7 Left ventricular thrombus is also commonly found in AF, another low-flow state. In a prospective study of 539 patients with AF > 48 h and without chronic anticoagulation, 13.1% were found to have atrial thrombi by transesophageal echocardiography (TEE).10 Manning et al. also found a 13% incidence in a similar group of patients with AF.11 LV thrombus is also found in a significant proportion of patients with severe congestive heart failure.12–14 In a study of 45 patients with dilated cardiomyopathy in a sinus rhythm and not on anticoagulation, Bakalli et al. LV thrombus formation in 13.3% of patients.
690
G.R. Budhram et al. / Resuscitation 85 (2014) 689–693
He also identified a significant correlation between thrombus formation and reduced ejection fraction (EF).14 Although cardiac arrest may be thought of as the ultimate low-flow state, echocardiographic studies of thrombus formation during cardiac arrest are lacking. Only one observational study by Varriale et al. of 18 patients with in-hospital arrest and sonographic cardiac standstill mentions the development of a “stagnant gel-like echodensity” after 20–30 min of CPR consistent with early thrombus formation in 10 patients.15 Using a swine model of sudden cardiac arrest, we examined the sonographic development of LV thrombus with time after induction of ventricular fibrillation (VF) and resolution during resuscitation. Swine are generally considered to be an excellent model of human myocardial electrophysiology, superior to dogs for the purpose of this study (similar-ordered species) and are far superior to rodents and rabbits (next species down) whose hearts are too small to produce human-like VF and bodies too small to conduct CPR and advance cardiac life support (ACLS) interventions.
2. Methods This was a single group observational study done in conjunction with an experimental protocol evaluating two resuscitation drug schemes. The study was approved by the Institutional Animal Care and Utilization Committee (IACUC), and conducted in our USDAcertified laboratory. We used locally obtained domestic Yorkshire swine, weighing approximately 30–35 kg. Animals were delivered and acclimated in advance of use. Our standard animal preparation and resuscitation procedures have been previously described in detail.16,17 Briefly, at the time of surgery, we sedated the animal with intramuscular TKX [telazol (5 mg kg−1 ), ketamine (2.5 mg kg−1 ), and xylazine (2.5 mg kg−1 )]. We provided inhaled isoflurane to facilitate advanced airway placement with a size-appropriate cuffed endotracheal tube and established intravenous access. We established a surgical plane of anesthesia using a rapid IV injection of propofol (2 mg kg−1 ) and maintained this with a continuous infusion (80 mics kg−1 min−1 ), titrated to effect. During preparation, we ventilated the animals with room air, using a volume-cycled ventilator, and adjusted the tidal volume and ventilation rate to maintain eucapnea (EtCO2 = 38–42 torr). We secured surface electrodes configured to correspond to a standard lead II electrocardiogram (ECG). After the airway is secured and a surgical depth of anesthesia was established, we placed an arterial introducer (8.5 Fr) into the femoral artery and a venous introducer (8.5 Fr) into the femoral vein under direct visualization. We inserted micro-manometer tipped pressure catheters (Mikro-Tip, Millar Instruments, Houston, TX) through the introducers and advance them into the ascending aorta and right atrium, respectively. The ECG tracing and all pressure data were continuously monitored and digitally recorded via a commercially available software package (LabChart, v.7.2.3, AD Instruments, Colorado Springs, CO). Immediately before induction of VF, the propofol infusion was discontinued, and the ventilator was disconnected. We induced VF by delivering a 3 s, 60-Hz, 100-mA transthoracic alternating current using a PowerStat variable transformer. Continuous transthoracic echocardiography (TTE) was performed by one of three emergency physicians: a fellowship trained ultrasound program director, an emergency ultrasound fellow, or an emergency physician credentialed in ultrasound in accordance with American College of Emergency Physician (ACEP) guidelines for bedside ultrasound. Ultrasound was performed using a Sonsite M-Turbo (Bothell, WA) and a 5-1 MHz phased array transducer. A subcostal transducer position was used, though the images correspond to an apical four-chamber window on the swine anatomy.
The presence or absence of hyperechoic thrombus in the LV was recorded every 2 min from time zero. No other interventions were performed for 12 min. When it became apparent that thrombus developed rapidly and reliably after induction of VF, we attempted to estimate how much of the LV was filled by thrombus over time. This estimation was performed on the last 10 animals enrolled. The sonographer visually assessed the percentage of the LV filled by thrombus at each 2-min interval from the two-dimensional sonographic image. After 12 min of untreated VF, resuscitation began with initiation of closed chest compressions. We used an oxygen-powered mechanical resuscitation device (Life-StatTM Mechanical CPR System, Michigan Instruments, Grand Rapids, MI) that provides standardized chest compressions in the anterior–posterior direction at a rate of 100 per minute with complete chest recoil. The device was programmed to provide compressions and ventilation (Vt = 500 cm3 , Fi 02 = 100%) in a ratio of 30:2. We gradually adjusted the compression depth over the first 20 s to match the animal’s size and generate a systolic peak pressure of between 50 and 60 mm Hg, typically to a depth of 1.25–2 in. After 30 s of mechanical chest compressions (MCC), which simulated the time necessary to obtain vascular access, animals were randomized to one of two groups (control or treatment) in a 2:1 ratio for drug delivery. After 2.5 min of MCC to adequately circulate the administered drugs, the first rescue shock (RS) was delivered. All defibrillation attempts were delivered at a fixed dose of energy (120 J) with a proprietary Rectilinear BiphasicTM defibrillation waveform (E SeriesTM , Zoll Medical Corp., Chelmsford, MA) through adult-sized defibrillator pads (Adult Plus multifunction electrode pads, Philips Healthcare, Andover, MA). We classified a RS as successful (VF termination) only if there was a restoration of organized electrical activity. Conversion to pulseless electrical activity (PEA) necessitated treatment according to the appropriate ACLS algorithm. We classified rescue shocks as failed if VF was not terminated (i.e., VF was the post-shock rhythm) or if the post-shock rhythm was asystole. We defined return of spontaneous circulation (ROSC) as the combination of an organized ECG rhythm with a systolic pressure greater than 80 mm Hg, sustained for at least 60 s continuously. Short-term survival required a minimum of 20 min of sustained ROSC. If the RS failed to terminate VF and restore spontaneous circulation, MCC resumed, and additional drugs (per protocol) were given and circulated for 3 min before the next attempted defibrillation. This latter sequence was repeated as long as a shockable rhythm persists or until 20 min of failed resuscitation had been attempted. Investigators continued to obtain subcostal TTE images every 2 min and recorded the time when thrombus disappeared completely from the LV. If any RS terminated VF with ROSC, aggressive supportive care was provided. At the end of the experiment, the animals were euthanized with a rapid IV bolus injection of Fatal-Plus Solution (1 cm3 /10 lbs) while still at a surgical depth of anesthesia. The data was analyzed descriptively using a commercially available software package (Stata/SE v.10.0 for Macintosh, College Station, TX).
3. Results Forty-nine animals were studied. Complete data was collected on 45/49 pigs. Four animals were excluded from data analysis because of inadequate sonographic visualization of the heart by TTE. Development of LV thrombus was observed in 43/45 animals during the 12-min period of observation. Representative images are shown in Fig. 1A–D.
G.R. Budhram et al. / Resuscitation 85 (2014) 689–693
691
Fig. 1. Development of LV thrombus over time. (A) (upper left) shows empty left ventricle at time 0. (B) (upper right) shows thrombus after 2 min (white arrows). (C) (lower left) shows thrombus filling the LV after 8 min. (D) (lower right) shows thrombus resolution after 2 min of chest compressions.
A summary of time to thrombus development and resolution is shown in Fig. 2. Of the animals that developed LV thrombus, 39/43 (86.7%) did so in less than 6 min. No animals that developed LV thrombus took longer than 10 min for thrombus to become visible by TTE. Only 2/45 (4.4%) animals never developed LV thrombus. The thrombus was generally observed to resolve quickly once chest compressions were initiated. Fig. 2 again summarizes the time to thrombus resolution. Of the 43 animals that developed LV thrombus, 31/43 (72.0%) completely resolved the thrombus in
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Experimental paper
Left ventricular thrombus development during ventricular fibrillation and resolution during resuscitation in a swine model of sudden cardiac arrest夽 Gavin R. Budhram ∗ , Timothy J. Mader, Lucienne Lutfy, David Murman, Abdullah Almulhim Tufts University School of Medicine, Department of Emergency Medicine, Baystate Medical Center, 759 Chestnut Street, Springfield, MA 01199, United States
a r t i c l e
i n f o
Article history: Received 10 August 2013 Received in revised form 16 January 2014 Accepted 22 January 2014 Keywords: Cardiac arrest Ventricular fibrillation Intracardiac thrombus Left ventricular thrombus Ultrasound Echocardiography
a b s t r a c t Background: Intracardiac thrombus is a well-known complication of low-flow cardiac states including acute myocardial infarction and atrial fibrillation. Little is known, however, about the formation of intracardiac (left ventricular [LV]) thrombus during the extreme low-flow state of cardiac arrest. Objective: Using a swine model of sudden cardiac arrest, we examined the sonographic development of LV thrombus over time after induction of ventricular fibrillation (VF) and resolution of thrombus with cardiopulmonary resuscitation (CPR). Methods: This observational study was IACUC approved. Forty-five Yorkshire swine were sedated, intubated, and instrumented under general anesthesia before VF was electrically induced. Sonographic data was collected immediately after VF induction and at 2-min intervals thereafter. Following 12 min of untreated VF, resuscitation was initiated with closed chest compressions using an oxygen-powered mechanical resuscitation device. Observations were continued during attempted resuscitation. At the end of the experiment, the animals were euthanized while still at a surgical depth of anesthesia. The data was analyzed descriptively. Results: Sonographic evidence of LV thrombus was observed in 43/45 animals (95.6% [95%CI: 85.2%, 98.8%]). Thrombus was detected within 6 min in 39/45 (86.7% [95%CI: 73.8%, 93.8%]) animals that developed thrombus. Thrombus resolved within 2 min after initiation of chest compressions in 31/43 (72.1% [95%CI: 57.3%, 83.3%]) animals. Conclusion: Similar to other low-flow cardiac states, LV thrombus develops early in the natural history of VF arrest and resolves quickly once forward flow is re-established by chest compressions. Institutional protocol number: 154600-8. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Intracardiac thrombus formation is known to complicate several low-flow cardiac states, including severe acute myocardial infarction (AMI), atrial fibrillation (AF), and severe cardiomyopathy. The initial clinical model for studying left ventricular thrombus was AMI because the onset of the pathophysiological process could be identified and all three prerequisites for thrombus formation are present (endothelial injury, hypercoagulable state, and blood stasis). Decreased left ventricular systolic performance after AMI is
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2014.01.030. ∗ Corresponding author. E-mail address: [email protected] (G.R. Budhram). http://dx.doi.org/10.1016/j.resuscitation.2014.01.030 0300-9572/© 2014 Elsevier Ireland Ltd. All rights reserved.
known to correlate strongly with the formation of LV thrombus.1–6 Intracardiac thrombus formation complicates 7–46% of AMIs, most frequently with acute anterior or apical infarctions.7–9 Thrombi tend to develop within the first week after infarction, and are preceded by severe apical wall-motion abnormalities characterized by akinesis, dyskinesis, or frank aneurysm.3,7 Left ventricular thrombus is also commonly found in AF, another low-flow state. In a prospective study of 539 patients with AF > 48 h and without chronic anticoagulation, 13.1% were found to have atrial thrombi by transesophageal echocardiography (TEE).10 Manning et al. also found a 13% incidence in a similar group of patients with AF.11 LV thrombus is also found in a significant proportion of patients with severe congestive heart failure.12–14 In a study of 45 patients with dilated cardiomyopathy in a sinus rhythm and not on anticoagulation, Bakalli et al. LV thrombus formation in 13.3% of patients.
690
G.R. Budhram et al. / Resuscitation 85 (2014) 689–693
He also identified a significant correlation between thrombus formation and reduced ejection fraction (EF).14 Although cardiac arrest may be thought of as the ultimate low-flow state, echocardiographic studies of thrombus formation during cardiac arrest are lacking. Only one observational study by Varriale et al. of 18 patients with in-hospital arrest and sonographic cardiac standstill mentions the development of a “stagnant gel-like echodensity” after 20–30 min of CPR consistent with early thrombus formation in 10 patients.15 Using a swine model of sudden cardiac arrest, we examined the sonographic development of LV thrombus with time after induction of ventricular fibrillation (VF) and resolution during resuscitation. Swine are generally considered to be an excellent model of human myocardial electrophysiology, superior to dogs for the purpose of this study (similar-ordered species) and are far superior to rodents and rabbits (next species down) whose hearts are too small to produce human-like VF and bodies too small to conduct CPR and advance cardiac life support (ACLS) interventions.
2. Methods This was a single group observational study done in conjunction with an experimental protocol evaluating two resuscitation drug schemes. The study was approved by the Institutional Animal Care and Utilization Committee (IACUC), and conducted in our USDAcertified laboratory. We used locally obtained domestic Yorkshire swine, weighing approximately 30–35 kg. Animals were delivered and acclimated in advance of use. Our standard animal preparation and resuscitation procedures have been previously described in detail.16,17 Briefly, at the time of surgery, we sedated the animal with intramuscular TKX [telazol (5 mg kg−1 ), ketamine (2.5 mg kg−1 ), and xylazine (2.5 mg kg−1 )]. We provided inhaled isoflurane to facilitate advanced airway placement with a size-appropriate cuffed endotracheal tube and established intravenous access. We established a surgical plane of anesthesia using a rapid IV injection of propofol (2 mg kg−1 ) and maintained this with a continuous infusion (80 mics kg−1 min−1 ), titrated to effect. During preparation, we ventilated the animals with room air, using a volume-cycled ventilator, and adjusted the tidal volume and ventilation rate to maintain eucapnea (EtCO2 = 38–42 torr). We secured surface electrodes configured to correspond to a standard lead II electrocardiogram (ECG). After the airway is secured and a surgical depth of anesthesia was established, we placed an arterial introducer (8.5 Fr) into the femoral artery and a venous introducer (8.5 Fr) into the femoral vein under direct visualization. We inserted micro-manometer tipped pressure catheters (Mikro-Tip, Millar Instruments, Houston, TX) through the introducers and advance them into the ascending aorta and right atrium, respectively. The ECG tracing and all pressure data were continuously monitored and digitally recorded via a commercially available software package (LabChart, v.7.2.3, AD Instruments, Colorado Springs, CO). Immediately before induction of VF, the propofol infusion was discontinued, and the ventilator was disconnected. We induced VF by delivering a 3 s, 60-Hz, 100-mA transthoracic alternating current using a PowerStat variable transformer. Continuous transthoracic echocardiography (TTE) was performed by one of three emergency physicians: a fellowship trained ultrasound program director, an emergency ultrasound fellow, or an emergency physician credentialed in ultrasound in accordance with American College of Emergency Physician (ACEP) guidelines for bedside ultrasound. Ultrasound was performed using a Sonsite M-Turbo (Bothell, WA) and a 5-1 MHz phased array transducer. A subcostal transducer position was used, though the images correspond to an apical four-chamber window on the swine anatomy.
The presence or absence of hyperechoic thrombus in the LV was recorded every 2 min from time zero. No other interventions were performed for 12 min. When it became apparent that thrombus developed rapidly and reliably after induction of VF, we attempted to estimate how much of the LV was filled by thrombus over time. This estimation was performed on the last 10 animals enrolled. The sonographer visually assessed the percentage of the LV filled by thrombus at each 2-min interval from the two-dimensional sonographic image. After 12 min of untreated VF, resuscitation began with initiation of closed chest compressions. We used an oxygen-powered mechanical resuscitation device (Life-StatTM Mechanical CPR System, Michigan Instruments, Grand Rapids, MI) that provides standardized chest compressions in the anterior–posterior direction at a rate of 100 per minute with complete chest recoil. The device was programmed to provide compressions and ventilation (Vt = 500 cm3 , Fi 02 = 100%) in a ratio of 30:2. We gradually adjusted the compression depth over the first 20 s to match the animal’s size and generate a systolic peak pressure of between 50 and 60 mm Hg, typically to a depth of 1.25–2 in. After 30 s of mechanical chest compressions (MCC), which simulated the time necessary to obtain vascular access, animals were randomized to one of two groups (control or treatment) in a 2:1 ratio for drug delivery. After 2.5 min of MCC to adequately circulate the administered drugs, the first rescue shock (RS) was delivered. All defibrillation attempts were delivered at a fixed dose of energy (120 J) with a proprietary Rectilinear BiphasicTM defibrillation waveform (E SeriesTM , Zoll Medical Corp., Chelmsford, MA) through adult-sized defibrillator pads (Adult Plus multifunction electrode pads, Philips Healthcare, Andover, MA). We classified a RS as successful (VF termination) only if there was a restoration of organized electrical activity. Conversion to pulseless electrical activity (PEA) necessitated treatment according to the appropriate ACLS algorithm. We classified rescue shocks as failed if VF was not terminated (i.e., VF was the post-shock rhythm) or if the post-shock rhythm was asystole. We defined return of spontaneous circulation (ROSC) as the combination of an organized ECG rhythm with a systolic pressure greater than 80 mm Hg, sustained for at least 60 s continuously. Short-term survival required a minimum of 20 min of sustained ROSC. If the RS failed to terminate VF and restore spontaneous circulation, MCC resumed, and additional drugs (per protocol) were given and circulated for 3 min before the next attempted defibrillation. This latter sequence was repeated as long as a shockable rhythm persists or until 20 min of failed resuscitation had been attempted. Investigators continued to obtain subcostal TTE images every 2 min and recorded the time when thrombus disappeared completely from the LV. If any RS terminated VF with ROSC, aggressive supportive care was provided. At the end of the experiment, the animals were euthanized with a rapid IV bolus injection of Fatal-Plus Solution (1 cm3 /10 lbs) while still at a surgical depth of anesthesia. The data was analyzed descriptively using a commercially available software package (Stata/SE v.10.0 for Macintosh, College Station, TX).
3. Results Forty-nine animals were studied. Complete data was collected on 45/49 pigs. Four animals were excluded from data analysis because of inadequate sonographic visualization of the heart by TTE. Development of LV thrombus was observed in 43/45 animals during the 12-min period of observation. Representative images are shown in Fig. 1A–D.
G.R. Budhram et al. / Resuscitation 85 (2014) 689–693
691
Fig. 1. Development of LV thrombus over time. (A) (upper left) shows empty left ventricle at time 0. (B) (upper right) shows thrombus after 2 min (white arrows). (C) (lower left) shows thrombus filling the LV after 8 min. (D) (lower right) shows thrombus resolution after 2 min of chest compressions.
A summary of time to thrombus development and resolution is shown in Fig. 2. Of the animals that developed LV thrombus, 39/43 (86.7%) did so in less than 6 min. No animals that developed LV thrombus took longer than 10 min for thrombus to become visible by TTE. Only 2/45 (4.4%) animals never developed LV thrombus. The thrombus was generally observed to resolve quickly once chest compressions were initiated. Fig. 2 again summarizes the time to thrombus resolution. Of the 43 animals that developed LV thrombus, 31/43 (72.0%) completely resolved the thrombus in
