- Email: [email protected]
Hemodynamics and Patient Safety During Pump-off Studies of an Axial-flow Left Ventricular Assist Device
MECHANICAL CIRCULATORY SUPPORT
Hemodynamics and Patient Safety During Pump-off Studies of an Axial-flow Left Ventricular Assist Device Timothy J. Myers, BS, CCRA, O. H. Frazier, MD, Hernando S. Mesina, BS, Branislav Radovancevic, MD, and Igor D. Gregoric, MD Background: Axial-flow left ventricular assist devices (LVADs), when inactivated, may result in regurgitant blood flow. We assessed the effects of regurgitant pump flow with the intraventricular Jarvik 2000 Heart LVAD (Jarvik Heart, Inc., New York, NY) on hemodynamics and patient safety under pump-off conditions. Methods: Thirty patients being supported by a Jarvik 2000 as a bridge to heart transplantation underwent pump-off studies. Hemodynamics, vital signs and cognitive function were monitored; Doppler echocardiographic studies were done with the pump turned off for 5 minutes if tolerated. Regurgitant flow was assessed in terms of the difference between left ventricular and right ventricular outflow tract cardiac output (LVOT CO – RVOT CO). Results: During pump-off periods, the mean regurgitant flow was 0.42 ⫾ 0.41 liter/min, and the mean arterial blood pressure was 63.1 ⫾ 11.6 mm Hg. There was no regurgitant flow when the pump was on. Three patients did not tolerate the pump being off for periods of 5 minutes; in these tests, the mean regurgitant flow rate was 0.54 ⫾ 0.50 liter/min, the mean arterial blood pressure was 52.8 ⫾ 9.8 mm Hg, and the mean pump-off time was 3.1 ⫾ 1.1 minutes. All patients remained conscious during the pump-off period, and none showed lasting adverse effects. Conclusions: Our findings suggest that patients being supported with the axial-flow Jarvik 2000 Heart LVAD can generally tolerate pump-off times of 5 minutes. J Heart Lung Transplant 2006;25:379 – 83. Copyright © 2006 by the International Society for Heart and Lung Transplantation.
Axial-flow pumps for the provision of long-term mechanical circulatory support have been under development for the last decade and are now undergoing clinical trials.1–3 These ventricular assist devices are designed to be small, easy to operate and reliable. Compared with pulsatile pumps, they are smaller and less complex; they do not require valves, a compliance chamber or external venting. They have the potential for use in smaller patients and have demonstrated the ability to provide support for long periods without mechanical failure. The Jarvik 2000 Heart (Jarvik Heart, Inc., New York, NY) is a battery-operated axial-flow left ventricular assist device (LVAD) that, under normal operating conditions, provides an uninterrupted flow of blood from the left ventricle to the aorta throughout the cardiac cycle.4 Flow through the pump is dependent on the rotation of its impeller and the differential pressure From the Texas Heart Institute, Houston, Texas. Submitted August 2, 2005; revised November 22, 2005; accepted November 27, 2005. Reprint requests: O. H. Frazier, MD, Texas Heart Institute, P.O. Box 20345, MC 2-114A, Houston, TX 77225-0345. Tel: 832-355-3000. Fax: 832-355-6798. E-mail: [email protected] Copyright © 2006 by the International Society for Heart and Lung Transplantation. 1053-2498/06/$–see front matter. doi:10.1016/ j.healun.2005.11.459
across the pump. When the impeller is not rotating, as when the pump is off, blood may flow from the aorta to the left ventricle, causing shunting and ineffectual blood flow. This scenario has not been observed clinically with the Jarvik 2000. At our institution, we test each patient before hospital discharge for the ability to withstand a temporary pump stoppage in the unlikely event this should occur. In the present study, we assessed the effects of regurgitant pump flow on hemodynamics and patient safety under pump-off conditions.
PATIENTS AND METHODS Device The Jarvik 2000 system consists of a blood pump, an external analog control unit and a DC battery power supply.4 The system and techniques for implanting it have previously been described in detail.4 – 6 The pump provides uninterrupted blood flow from the left ventricle to the ascending or descending aorta. The pump outflow graft was placed on the ascending aorta in 7 patients (23%) and on the descending aorta in 23 (77%). The pump’s impeller spins at speeds ranging from 8,000 to 12,000 rpm, which are manually adjusted in increments of 1,000 rpm. 379
380
Myers et al.
The Journal of Heart and Lung Transplantation April 2006
Patients Thirty patients who were being supported with the Jarvik 2000 pump as a bridge to heart transplantation during the period of this study were enrolled. All met the study inclusion criteria and signed an informed consent form. The study population consisted of 24 men and 6 women. The mean age was 50.6 ⫾ 12.9 years (range 19 to 70 years), and the mean body surface area was 1.92 ⫾ 0.19 m2 (range 1.6 to 2.29 m2). At the time this manuscript was prepared, 5 patients were still being supported, 22 had undergone heart transplantation, and 3 had died during support. The mean duration of support was 170 ⫾ 168 days (range 13 to 763 days). Patients were monitored during brief periods with the pump off. A 5-minute pump-off limit was chosen because, during the period of this study, Jarvik 2000 – supported patients were required to be able to change out the system’s batteries and controller in less than 5 minutes. The pump would be off for these 5 minutes until the battery power was reinstated. Vital signs were monitored and recorded during the pump-off period. All patients were receiving anti-coagulation therapy to achieve an international normalized ratio (INR) of 1.5 to 2.0, and the mean INR value for the group was 1.8. In each case, 5,000 U of heparin was administered just before the pump was turned off. Doppler echocardiographic studies were used to assess left ventricular outflow tract cardiac output (LVOT CO, liters/min), right ventricular outflow tract cardiac output (RVOT CO, liters/min), left ventricular end-systolic dimension (LVESd) and left ventricular
end-diastolic dimension (LVEDd). As previously reported,7 the LVOT stroke volume was obtained by multiplying the LVOT area (derived from the LVOT diameter8) by the velocity–time integral (VTI) obtained by placing the pulsed Doppler sample volume in the LVOT (3chamber or apical 5-chamber view). The RVOT stroke volume was obtained in a similar fashion by measuring the RVOT diameter and RVOT VTI obtained in the parasternal short-axis view by pulsed Doppler echocardiography. Cardiac output was obtained by multiplying the derived stroke volume by the heart rate.
Figure 1. Left ventricular outflow tract cardiac output (LVOT CO) and right ventricular outflow tract cardiac output (RVOT CO) as measured by Doppler echocardiography. The difference between LVOT CO and RVOT CO is equal to flow through the pump (“pump flow”). With the pump off, LVOT CO is greater than RVOT CO and represents regurgitant flow through the pump plus bronchial arterial blood flow. The mean rate of regurgitant flow for all studies was 0.42 liters/min. *p ⬍ 0.05, as compared with pump-off value.
RESULTS Studies were performed in 30 patients. No study lasted ⬎5 minutes. All patients remained alert and conscious during testing, and none showed any lasting adverse effects. Changes in RVOT CO, LVOT CO, LVOT CO ⫺ RVOT CO, diastolic blood pressure and mean arterial blood pressure were recorded and studied. With the pump off, the mean LVOT CO was 4.46 ⫾ 0.89 liters/min, the mean RVOT CO was 4.03 ⫾ 0.71 liters/min and the mean LVOT CO ⫺ RVOT CO value was 0.42 ⫾ 0.41 liters/min (Figure 1). With the pump on, the LVOT CO ⫺ RVOT CO value represented estimated flow through the pump because there is no method at present for estimating or directly measuring flow through this device. In most patients, when the pump speed reached 12,000 rpm, the cardiac output was largely represented by pump flow. The systolic blood pressure was not significantly different with the pump on or off (Figure 2). However, the diastolic and mean arterial blood pressures significantly increased (p ⬍ 0.05) with increasing pump speed. With the pump off, the mean values for LVEDd and LVESd were 7.36 ⫾ 0.7 cm and 6.46 ⫾ 0.7 cm, respectively (Table 1). With the pump
Figure 2. Systolic, diastolic and mean arterial blood pressures with pump off and at increasing pump speeds. Systolic pressure remained unchanged, but diastolic and mean arterial pressures rose significantly when the pump was on. *p ⬍ 0.05, as compared with pump-off value.
The Journal of Heart and Lung Transplantation Volume 25, Number 4
Myers et al.
381
Table 1. Left Ventricular End-diastolic Dimension (LVEDd), Left Ventricular End-systolic Dimension (LVESd), Heart Rate, and Respiratory Rate With Pump Off and at Increasing Pump Speeds Pump speed (rpm) 0 8,000 9,000 10,000 11,000 12,000
LVEDd (mm) 7.36 ⫾ 0.7 6.97 ⫾ 0.86 6.84 ⫾ 0.75 6.65 ⫾ 0.89 6.49 ⫾ 0.70 6.16 ⫾ 0.84
LVESd (mm) 6.46 ⫾ 0.8 6.20 ⫾ 0.66 6.10 ⫾ 0.70 5.90 ⫾ 0.88 5.79 ⫾ 0.76 5.55 ⫾ 0.87
on, LVEDd and LVESd progressively decreased as pump speed increased. When compared with the pump-off values, these changes became significant when the pump was operating at speeds of ⱖ10,000 rpm. The LVEDd decreased to 6.65 ⫾ 0.89 cm at 10,000 rpm and decreased further to 6.16 ⫾ 0.84 cm at 12,000 rpm (Table 1). Likewise, the LVESd was 5.90 ⫾ 0.88 cm at 10,000 rpm and 5.55 ⫾ 0.87 cm at 12,000 rpm (Table 1). The mean heart rate and respiratory rate did not change significantly at any of the experimental settings (p ⬍ 0.05).
Heart rate (bpm) 92.7 ⫾ 19.1 90.5 ⫾ 18.9 90.6 ⫾ 18.5 89.7 ⫾ 18 91.0 ⫾ 19.4 91.0 ⫾ 16.9
Respiratory rate (rpm) 15.8 ⫾ 2.1 15.5 ⫾ 2.3 14.5 ⫾ 1.8 14.4 ⫾ 2.2 14.5 ⫾ 2.0 15.6 ⫾ 1.2
Pump-off testing was terminated in 3 cases because of patient complaints, which included dyspnea, lightheadedness, diaphoresis, flushing, arm tingling and nausea. Most of these complaints were considered mild by the 3 patients, and the symptoms resolved when pump support was resumed. The mean pump-off time during these studies was 3.1 ⫾ 1.1 minutes. The most characteristic change in the 3 studies was a decrease in mean arterial blood pressure to 52.8 ⫾ 9.8 mm Hg, as compared with a mean arterial blood pressure of 63.1 ⫾
Figure 3. Two-dimensional pulsed Doppler echocardiograms obtained in the parasternal long axis view. The sample volume was placed within the outlet cannula just proximal to a descending thoracic aorta anastomosis. Arrows indicate forward systolic flow out of the cannula and into the descending aorta. Arrowheads indicate diastolic-only reverse flow into the cannula (below baseline in figure) with pump off (A) and forward flow into the aorta (above baseline in figure) with pump on and operating at low speed (B) and higher speed (C).
382
Myers et al.
11.6 mm Hg for all other studies. The mean regurgitant flow of 0.54 ⫾ 0.50 liter/min was comparable to that for the whole group (0.42 ⫾ 0.41 liter/min). The respiratory rate was lower, and the mean arterial pressure was higher. Echocardiograms of patients who did not tolerate the pump off and those who did were similar. However, the 3 patients whose pump-off studies were terminated early appeared to be more dependent on pump support at the time of testing, even though there were no clinical parameters that were predictive of which patients would not tolerate the pump-off test. Because of the small sample size, no meaningful statistical comparisons could be made. Initially, because of restrictions in the approved protocol, only 9 patients were allowed to be discharged from the hospital; however, all 30 patients were medically suitable for discharge. All of the discharged patients were able to change out their external batteries without difficulty after the pump had been off for 5 minutes. No inadvertent pump-off episodes have occurred in outpatients. DISCUSSION Axial-flow cardiac support presents a new form of circulatory physiology.9 Blood flow through axial-flow pumps depends primarily on the differential pressure across the pump when the impeller speed is constant. This differential pressure is the difference between left ventricular pressure and aortic pressure, which is always changing throughout the cardiac cycle. At the start of diastole, left ventricular pressure is at its lowest; this is also when the differential pressure is greatest. At the end of systole, left ventricular pressure and aortic pressure are nearly equal; this is when the differential pressure is lowest. As the left ventricular pressure rises during systole, pump flow increases because the differential pressure decreases. During diastole, when the left ventricular pressure decreases (thus increasing the differential pressure), the blood flow decreases but remains antegrade through the pump (Figure 3A). However, when the pump is off, blood may flow through the pump in both directions (Figure 3B and C). The present results demonstrate that regurgitant flow through an idle Jarvik 2000 Heart axial-flow LVAD is minimal and does not usually have immediate clinical consequences. Whether this holds true when the pump is turned off for longer periods will depend on the degree to which the patient’s native heart function recovers. In this study, regurgitant flow was measured directly by transthoracic Doppler echocardiography.10 There is, however, a constant error inherent in this methodology; it calculates RVOT CO and LVOT CO as equivalent.
The Journal of Heart and Lung Transplantation April 2006
The bronchial artery blood supply to the lungs returns directly through the pulmonary venous system, thereby bypassing the right ventricle. This natural shunt normally represents a 5% to 10% increase in LVOT CO over RVOT.11 However, normal echocardiographic assessment calculates the RVOT and LVOT CO as equivalent, thereby negating the effect in the calculations. In many cases of LVAD support, unloading of the left ventricle leads to improved cardiac function,12,13 which allows adequate cardiac output and blood pressure to be maintained. In the case of support with an axial-flow LVAD, our results suggest that regurgitant flow is limited in the Jarvik 2000 and, in the case of device stoppage, it should not cause acute decompensation. REFERENCES 1. Frazier OH, Delgado RM, Kar B, et al. First clinical use of the redesigned HeartMate II left ventricular assist system in the United States: a case report. Tex Heart Inst J 2004;31:157–9. 2. Frazier OH, Myers TJ, Gregoric ID, et al. Initial clinical experience with the Jarvik 2000 implantable axial-flow left ventricular assist system. Circulation 2002;105: 2855– 60. 3. Goldstein DJ, Zucker M, Arroyo L, et al. Safety and feasibility trial of the MicroMed DeBakey ventricular assist device as a bridge to transplantation. JAMA 2005;45:962–3. 4. Frazier OH, Myers TJ, Jarvik RK, et al. Research and development of an implantable, axial-flow left ventricular assist device: the Jarvik 2000 heart. Ann Thorac Surg 2001;71(suppl):S125–32. 5. Westaby S, Frazier OH, Pigott DW, Saito S, Jarvik RK. Implant technique for the Jarvik 2000 Heart. Ann Thorac Surg 2002;73:1337– 40. 6. Siegenthaler MP, Martin J, Frazier OH, Beyersdorf F. Implantation of the permanent Jarvik-2000 left-ventricular-assist-device: surgical technique. Eur J Cardiothorac Surg 2002;21:546 – 8. 7. Stainback RF, Croitoru M, Hernandez A, et al. Echocardiographic evaluation of the Jarvik 2000 axial-flow LVAD. Tex Heart Inst J 2005;32:263–70. 8. Weyman AE. Standard plane positions—standard imaging planes. In: Weyman AE, editor. Principles and practice of echocardiography, 2nd ed. Philadelphia: Lea & Febiger; 1994:98 –123. 9. Frazier OH, Myers TJ, Westaby S, Gregoric ID. Clinical experience with an implantable, intracardiac, continuous flow circulatory support device: physiologic implications and their relationship to patient selection. Ann Thorac Surg 2004;77:133– 42. 10. Kitabatake A, Inoue M, Asao M, et al. Noninvasive evaluation of the ratio of pulmonary to systemic flow in atrial septal defect by duplex Doppler echocardiography. Circulation 1984;69:73–9.
The Journal of Heart and Lung Transplantation Volume 25, Number 4
11. Olsen DB, Long JW. Simplified right–left balance for the implanted artificial heart. In: Akutsu T, Koyanagi H, editors. Heart replacement: artificial heart 3. Tokyo: Springer; 1990:235– 43. 12. Khan T, Okerberg K, Hernandez A, et al. Assessment of myocardial recovery using dobutamine stress echocardi-
Myers et al.
383
ography in LVAD patients. J Heart Lung Transplant 2001; 20:202–3. 13. Razeghi P, Myers TJ, Frazier OH, Taegtmeyer H. Reverse remodeling of the failing human heart with mechanical unloading. Emerging concepts and unanswered questions. Cardiology 2002;98:167–74.
Hemodynamics and Patient Safety During Pump-off Studies of an Axial-flow Left Ventricular Assist Device Timothy J. Myers, BS, CCRA, O. H. Frazier, MD, Hernando S. Mesina, BS, Branislav Radovancevic, MD, and Igor D. Gregoric, MD Background: Axial-flow left ventricular assist devices (LVADs), when inactivated, may result in regurgitant blood flow. We assessed the effects of regurgitant pump flow with the intraventricular Jarvik 2000 Heart LVAD (Jarvik Heart, Inc., New York, NY) on hemodynamics and patient safety under pump-off conditions. Methods: Thirty patients being supported by a Jarvik 2000 as a bridge to heart transplantation underwent pump-off studies. Hemodynamics, vital signs and cognitive function were monitored; Doppler echocardiographic studies were done with the pump turned off for 5 minutes if tolerated. Regurgitant flow was assessed in terms of the difference between left ventricular and right ventricular outflow tract cardiac output (LVOT CO – RVOT CO). Results: During pump-off periods, the mean regurgitant flow was 0.42 ⫾ 0.41 liter/min, and the mean arterial blood pressure was 63.1 ⫾ 11.6 mm Hg. There was no regurgitant flow when the pump was on. Three patients did not tolerate the pump being off for periods of 5 minutes; in these tests, the mean regurgitant flow rate was 0.54 ⫾ 0.50 liter/min, the mean arterial blood pressure was 52.8 ⫾ 9.8 mm Hg, and the mean pump-off time was 3.1 ⫾ 1.1 minutes. All patients remained conscious during the pump-off period, and none showed lasting adverse effects. Conclusions: Our findings suggest that patients being supported with the axial-flow Jarvik 2000 Heart LVAD can generally tolerate pump-off times of 5 minutes. J Heart Lung Transplant 2006;25:379 – 83. Copyright © 2006 by the International Society for Heart and Lung Transplantation.
Axial-flow pumps for the provision of long-term mechanical circulatory support have been under development for the last decade and are now undergoing clinical trials.1–3 These ventricular assist devices are designed to be small, easy to operate and reliable. Compared with pulsatile pumps, they are smaller and less complex; they do not require valves, a compliance chamber or external venting. They have the potential for use in smaller patients and have demonstrated the ability to provide support for long periods without mechanical failure. The Jarvik 2000 Heart (Jarvik Heart, Inc., New York, NY) is a battery-operated axial-flow left ventricular assist device (LVAD) that, under normal operating conditions, provides an uninterrupted flow of blood from the left ventricle to the aorta throughout the cardiac cycle.4 Flow through the pump is dependent on the rotation of its impeller and the differential pressure From the Texas Heart Institute, Houston, Texas. Submitted August 2, 2005; revised November 22, 2005; accepted November 27, 2005. Reprint requests: O. H. Frazier, MD, Texas Heart Institute, P.O. Box 20345, MC 2-114A, Houston, TX 77225-0345. Tel: 832-355-3000. Fax: 832-355-6798. E-mail: [email protected] Copyright © 2006 by the International Society for Heart and Lung Transplantation. 1053-2498/06/$–see front matter. doi:10.1016/ j.healun.2005.11.459
across the pump. When the impeller is not rotating, as when the pump is off, blood may flow from the aorta to the left ventricle, causing shunting and ineffectual blood flow. This scenario has not been observed clinically with the Jarvik 2000. At our institution, we test each patient before hospital discharge for the ability to withstand a temporary pump stoppage in the unlikely event this should occur. In the present study, we assessed the effects of regurgitant pump flow on hemodynamics and patient safety under pump-off conditions.
PATIENTS AND METHODS Device The Jarvik 2000 system consists of a blood pump, an external analog control unit and a DC battery power supply.4 The system and techniques for implanting it have previously been described in detail.4 – 6 The pump provides uninterrupted blood flow from the left ventricle to the ascending or descending aorta. The pump outflow graft was placed on the ascending aorta in 7 patients (23%) and on the descending aorta in 23 (77%). The pump’s impeller spins at speeds ranging from 8,000 to 12,000 rpm, which are manually adjusted in increments of 1,000 rpm. 379
380
Myers et al.
The Journal of Heart and Lung Transplantation April 2006
Patients Thirty patients who were being supported with the Jarvik 2000 pump as a bridge to heart transplantation during the period of this study were enrolled. All met the study inclusion criteria and signed an informed consent form. The study population consisted of 24 men and 6 women. The mean age was 50.6 ⫾ 12.9 years (range 19 to 70 years), and the mean body surface area was 1.92 ⫾ 0.19 m2 (range 1.6 to 2.29 m2). At the time this manuscript was prepared, 5 patients were still being supported, 22 had undergone heart transplantation, and 3 had died during support. The mean duration of support was 170 ⫾ 168 days (range 13 to 763 days). Patients were monitored during brief periods with the pump off. A 5-minute pump-off limit was chosen because, during the period of this study, Jarvik 2000 – supported patients were required to be able to change out the system’s batteries and controller in less than 5 minutes. The pump would be off for these 5 minutes until the battery power was reinstated. Vital signs were monitored and recorded during the pump-off period. All patients were receiving anti-coagulation therapy to achieve an international normalized ratio (INR) of 1.5 to 2.0, and the mean INR value for the group was 1.8. In each case, 5,000 U of heparin was administered just before the pump was turned off. Doppler echocardiographic studies were used to assess left ventricular outflow tract cardiac output (LVOT CO, liters/min), right ventricular outflow tract cardiac output (RVOT CO, liters/min), left ventricular end-systolic dimension (LVESd) and left ventricular
end-diastolic dimension (LVEDd). As previously reported,7 the LVOT stroke volume was obtained by multiplying the LVOT area (derived from the LVOT diameter8) by the velocity–time integral (VTI) obtained by placing the pulsed Doppler sample volume in the LVOT (3chamber or apical 5-chamber view). The RVOT stroke volume was obtained in a similar fashion by measuring the RVOT diameter and RVOT VTI obtained in the parasternal short-axis view by pulsed Doppler echocardiography. Cardiac output was obtained by multiplying the derived stroke volume by the heart rate.
Figure 1. Left ventricular outflow tract cardiac output (LVOT CO) and right ventricular outflow tract cardiac output (RVOT CO) as measured by Doppler echocardiography. The difference between LVOT CO and RVOT CO is equal to flow through the pump (“pump flow”). With the pump off, LVOT CO is greater than RVOT CO and represents regurgitant flow through the pump plus bronchial arterial blood flow. The mean rate of regurgitant flow for all studies was 0.42 liters/min. *p ⬍ 0.05, as compared with pump-off value.
RESULTS Studies were performed in 30 patients. No study lasted ⬎5 minutes. All patients remained alert and conscious during testing, and none showed any lasting adverse effects. Changes in RVOT CO, LVOT CO, LVOT CO ⫺ RVOT CO, diastolic blood pressure and mean arterial blood pressure were recorded and studied. With the pump off, the mean LVOT CO was 4.46 ⫾ 0.89 liters/min, the mean RVOT CO was 4.03 ⫾ 0.71 liters/min and the mean LVOT CO ⫺ RVOT CO value was 0.42 ⫾ 0.41 liters/min (Figure 1). With the pump on, the LVOT CO ⫺ RVOT CO value represented estimated flow through the pump because there is no method at present for estimating or directly measuring flow through this device. In most patients, when the pump speed reached 12,000 rpm, the cardiac output was largely represented by pump flow. The systolic blood pressure was not significantly different with the pump on or off (Figure 2). However, the diastolic and mean arterial blood pressures significantly increased (p ⬍ 0.05) with increasing pump speed. With the pump off, the mean values for LVEDd and LVESd were 7.36 ⫾ 0.7 cm and 6.46 ⫾ 0.7 cm, respectively (Table 1). With the pump
Figure 2. Systolic, diastolic and mean arterial blood pressures with pump off and at increasing pump speeds. Systolic pressure remained unchanged, but diastolic and mean arterial pressures rose significantly when the pump was on. *p ⬍ 0.05, as compared with pump-off value.
The Journal of Heart and Lung Transplantation Volume 25, Number 4
Myers et al.
381
Table 1. Left Ventricular End-diastolic Dimension (LVEDd), Left Ventricular End-systolic Dimension (LVESd), Heart Rate, and Respiratory Rate With Pump Off and at Increasing Pump Speeds Pump speed (rpm) 0 8,000 9,000 10,000 11,000 12,000
LVEDd (mm) 7.36 ⫾ 0.7 6.97 ⫾ 0.86 6.84 ⫾ 0.75 6.65 ⫾ 0.89 6.49 ⫾ 0.70 6.16 ⫾ 0.84
LVESd (mm) 6.46 ⫾ 0.8 6.20 ⫾ 0.66 6.10 ⫾ 0.70 5.90 ⫾ 0.88 5.79 ⫾ 0.76 5.55 ⫾ 0.87
on, LVEDd and LVESd progressively decreased as pump speed increased. When compared with the pump-off values, these changes became significant when the pump was operating at speeds of ⱖ10,000 rpm. The LVEDd decreased to 6.65 ⫾ 0.89 cm at 10,000 rpm and decreased further to 6.16 ⫾ 0.84 cm at 12,000 rpm (Table 1). Likewise, the LVESd was 5.90 ⫾ 0.88 cm at 10,000 rpm and 5.55 ⫾ 0.87 cm at 12,000 rpm (Table 1). The mean heart rate and respiratory rate did not change significantly at any of the experimental settings (p ⬍ 0.05).
Heart rate (bpm) 92.7 ⫾ 19.1 90.5 ⫾ 18.9 90.6 ⫾ 18.5 89.7 ⫾ 18 91.0 ⫾ 19.4 91.0 ⫾ 16.9
Respiratory rate (rpm) 15.8 ⫾ 2.1 15.5 ⫾ 2.3 14.5 ⫾ 1.8 14.4 ⫾ 2.2 14.5 ⫾ 2.0 15.6 ⫾ 1.2
Pump-off testing was terminated in 3 cases because of patient complaints, which included dyspnea, lightheadedness, diaphoresis, flushing, arm tingling and nausea. Most of these complaints were considered mild by the 3 patients, and the symptoms resolved when pump support was resumed. The mean pump-off time during these studies was 3.1 ⫾ 1.1 minutes. The most characteristic change in the 3 studies was a decrease in mean arterial blood pressure to 52.8 ⫾ 9.8 mm Hg, as compared with a mean arterial blood pressure of 63.1 ⫾
Figure 3. Two-dimensional pulsed Doppler echocardiograms obtained in the parasternal long axis view. The sample volume was placed within the outlet cannula just proximal to a descending thoracic aorta anastomosis. Arrows indicate forward systolic flow out of the cannula and into the descending aorta. Arrowheads indicate diastolic-only reverse flow into the cannula (below baseline in figure) with pump off (A) and forward flow into the aorta (above baseline in figure) with pump on and operating at low speed (B) and higher speed (C).
382
Myers et al.
11.6 mm Hg for all other studies. The mean regurgitant flow of 0.54 ⫾ 0.50 liter/min was comparable to that for the whole group (0.42 ⫾ 0.41 liter/min). The respiratory rate was lower, and the mean arterial pressure was higher. Echocardiograms of patients who did not tolerate the pump off and those who did were similar. However, the 3 patients whose pump-off studies were terminated early appeared to be more dependent on pump support at the time of testing, even though there were no clinical parameters that were predictive of which patients would not tolerate the pump-off test. Because of the small sample size, no meaningful statistical comparisons could be made. Initially, because of restrictions in the approved protocol, only 9 patients were allowed to be discharged from the hospital; however, all 30 patients were medically suitable for discharge. All of the discharged patients were able to change out their external batteries without difficulty after the pump had been off for 5 minutes. No inadvertent pump-off episodes have occurred in outpatients. DISCUSSION Axial-flow cardiac support presents a new form of circulatory physiology.9 Blood flow through axial-flow pumps depends primarily on the differential pressure across the pump when the impeller speed is constant. This differential pressure is the difference between left ventricular pressure and aortic pressure, which is always changing throughout the cardiac cycle. At the start of diastole, left ventricular pressure is at its lowest; this is also when the differential pressure is greatest. At the end of systole, left ventricular pressure and aortic pressure are nearly equal; this is when the differential pressure is lowest. As the left ventricular pressure rises during systole, pump flow increases because the differential pressure decreases. During diastole, when the left ventricular pressure decreases (thus increasing the differential pressure), the blood flow decreases but remains antegrade through the pump (Figure 3A). However, when the pump is off, blood may flow through the pump in both directions (Figure 3B and C). The present results demonstrate that regurgitant flow through an idle Jarvik 2000 Heart axial-flow LVAD is minimal and does not usually have immediate clinical consequences. Whether this holds true when the pump is turned off for longer periods will depend on the degree to which the patient’s native heart function recovers. In this study, regurgitant flow was measured directly by transthoracic Doppler echocardiography.10 There is, however, a constant error inherent in this methodology; it calculates RVOT CO and LVOT CO as equivalent.
The Journal of Heart and Lung Transplantation April 2006
The bronchial artery blood supply to the lungs returns directly through the pulmonary venous system, thereby bypassing the right ventricle. This natural shunt normally represents a 5% to 10% increase in LVOT CO over RVOT.11 However, normal echocardiographic assessment calculates the RVOT and LVOT CO as equivalent, thereby negating the effect in the calculations. In many cases of LVAD support, unloading of the left ventricle leads to improved cardiac function,12,13 which allows adequate cardiac output and blood pressure to be maintained. In the case of support with an axial-flow LVAD, our results suggest that regurgitant flow is limited in the Jarvik 2000 and, in the case of device stoppage, it should not cause acute decompensation. REFERENCES 1. Frazier OH, Delgado RM, Kar B, et al. First clinical use of the redesigned HeartMate II left ventricular assist system in the United States: a case report. Tex Heart Inst J 2004;31:157–9. 2. Frazier OH, Myers TJ, Gregoric ID, et al. Initial clinical experience with the Jarvik 2000 implantable axial-flow left ventricular assist system. Circulation 2002;105: 2855– 60. 3. Goldstein DJ, Zucker M, Arroyo L, et al. Safety and feasibility trial of the MicroMed DeBakey ventricular assist device as a bridge to transplantation. JAMA 2005;45:962–3. 4. Frazier OH, Myers TJ, Jarvik RK, et al. Research and development of an implantable, axial-flow left ventricular assist device: the Jarvik 2000 heart. Ann Thorac Surg 2001;71(suppl):S125–32. 5. Westaby S, Frazier OH, Pigott DW, Saito S, Jarvik RK. Implant technique for the Jarvik 2000 Heart. Ann Thorac Surg 2002;73:1337– 40. 6. Siegenthaler MP, Martin J, Frazier OH, Beyersdorf F. Implantation of the permanent Jarvik-2000 left-ventricular-assist-device: surgical technique. Eur J Cardiothorac Surg 2002;21:546 – 8. 7. Stainback RF, Croitoru M, Hernandez A, et al. Echocardiographic evaluation of the Jarvik 2000 axial-flow LVAD. Tex Heart Inst J 2005;32:263–70. 8. Weyman AE. Standard plane positions—standard imaging planes. In: Weyman AE, editor. Principles and practice of echocardiography, 2nd ed. Philadelphia: Lea & Febiger; 1994:98 –123. 9. Frazier OH, Myers TJ, Westaby S, Gregoric ID. Clinical experience with an implantable, intracardiac, continuous flow circulatory support device: physiologic implications and their relationship to patient selection. Ann Thorac Surg 2004;77:133– 42. 10. Kitabatake A, Inoue M, Asao M, et al. Noninvasive evaluation of the ratio of pulmonary to systemic flow in atrial septal defect by duplex Doppler echocardiography. Circulation 1984;69:73–9.
The Journal of Heart and Lung Transplantation Volume 25, Number 4
11. Olsen DB, Long JW. Simplified right–left balance for the implanted artificial heart. In: Akutsu T, Koyanagi H, editors. Heart replacement: artificial heart 3. Tokyo: Springer; 1990:235– 43. 12. Khan T, Okerberg K, Hernandez A, et al. Assessment of myocardial recovery using dobutamine stress echocardi-
Myers et al.
383
ography in LVAD patients. J Heart Lung Transplant 2001; 20:202–3. 13. Razeghi P, Myers TJ, Frazier OH, Taegtmeyer H. Reverse remodeling of the failing human heart with mechanical unloading. Emerging concepts and unanswered questions. Cardiology 2002;98:167–74.