|
|
||||||||
Center for Cardiac Support and the Heart Failure Clinic, Texas Heart Institute at St. Luke's Episcopal Hospital, Houston, Texas
Accepted for publication August 4, 2009.
* Address correspondence to Dr Gregoric, Texas Heart Institute, MC 3-147, PO Box 20345, Houston, TX 77225-0345 (Email: epongratz{at}heart.thi.tmc.edu).
| Abstract |
|---|
|
|
|---|
Methods: We reviewed the records of 10 patients (6 men; 4 women), aged 62 ± 12 years, who presented with cardiac arrest or SRCS. Eight patients, 5 undergoing cardiopulmonary resuscitation (CPR) and 3 with SRCS, received TandemHeart support in the catheterization laboratory and had AVR after undergoing hemodynamic stabilization. The other 2 patients went directly to the operating room while undergoing CPR, for emergency AVR and received the device for postcardiotomy cardiogenic shock. All 10 patients were intubated, on maximal vasopressor support, and 7 had an intraaortic balloon pump. The preoperative Society of Thoracic Surgeons mortality risk was 74.9% ± 24.5%.
Results: The 8 patients who received the TandemHeart in the catheterization laboratory were supported for 6.4 ± 3.8 days and had significantly improved renal function before AVR. One patient died of sepsis 34 days after AVR, The other 7 were discharged home (ejection fraction, 0.42 ± 0.14) and were alive 2 to 43 months later. The 2 patients who received the device in the operating room after AVR died on days 8 and 21, respectively.
Conclusions: Prompt placement of the TandemHeart in these critically ill patients yields the shortest "emergency department door to left ventricular unloading time," improves end-organ function, and allows AVR to be performed electively.
| Introduction |
|---|
|
|
|---|
The TandemHeart Percutaneous Ventricular Assist Device (pVAD; Cardiac Assist Inc, Pittsburgh, PA) is a promising new option for temporary circulatory support. It can be quickly inserted through the femoral route in the cardiac catheterization laboratory or can be placed surgically during a cardiac procedure [9, 10]. Unlike the IABP, the pVAD achieves significant ventricular unloading, producing flow rates of up to 4 L/min at 7500 rpm [9].
Thiele and coauthors [11] used the TandemHeart pVAD to treat patients in cardiac shock and reported a 30-day mortality rate of 44%. The same device has been successfully used in a limited number of patients undergoing percutaneous coronary interventions [9, 12–16], bridging to transplantation [17], and temporary support for SRCS [18, 19]. We reported the use of the TandemHeart pVAD in a patient with critical aortic valve stenosis in 2006 [20]; since then, the literature has contained a few other reports of the device being used in this patient population [21, 22]. In the present study, we retrospectively examined our use of the TandemHeart pVAD as a rescue therapy for 10 patients with cardiac arrest or SRCS before (n = 8) or after (n = 2) AVR.
| Patients and Methods |
|---|
|
|
|---|
Eight of the 10 patients, 5 with cardiac arrest and 3 with SRCS, underwent emergency pVAD insertion in the catheterization laboratory with the goal of unloading the LV, reversing SRCS, and stabilizing the patient's condition sufficiently to permit elective AVR. In 1 patient, the atrial septum was thickened and scarred from previous cardiac operations, and attempts to cross it for left atrial access were unsuccessful. Because this patient had been undergoing cardiopulmonary resuscitation (CPR) for more than an hour, we placed the oxygenator on the pVADs outflow-limb tubing, creating temporary extracorporeal membrane oxygenation (ECMO) perfusion. The patient's neurologic function was confirmed to be normal 6 hours later, and AVR was performed.
The other 2 patients, who were undergoing CPR, were rushed directly to the operating room for emergency AVR. They could not be weaned from CPB because of postcardiotomy cardiogenic shock and so the pVAD was placed intraoperatively for LV support.
For this retrospective analysis, we analyzed the LV ejection fraction, cardiac index, central venous pressure, systolic blood pressure, pulmonary capillary wedge pressure, mean arterial pressure, urine output, and inotropic therapy requirements, as well as laboratory values for blood urea nitrogen (BUN), creatinine, total bilirubin, serum glutamic pyruvic transaminase (SGPT), serum glutamic oxaloacetic transaminase (SGOT), lactate dehydrogenase (LDH), sodium, chloride, and potassium. We also analyzed the duration of pVAD support, the 30-day mortality, and the long-term follow-up results.
The TandemHeart pVAD implantation procedure has been described elsewhere [13]. The insertion procedure for the 8 patients who underwent pVAD placement in the cardiac catheterization laboratory lasted an average of 15 minutes. The pVAD was placed into 2 other patients by a surgeon in the operating room. Postoperative management of the device in these patients did not differ substantially from device management in the pVAD recipients with cardiogenic shock related to other etiologies.
In the catheterization laboratory, after pumping was initiated, the flow rate was slowly increased to 3 to 5 L/min. Anticoagulation was achieved with heparin as previously described [9, 10]. After AVR, the effects of heparin were reversed with protamine. In the intensive care unit, after chest-tube blood drainage decreased to less than 100 mL/h, heparin therapy was initiated at a low dosage (200 U/h). When bleeding stopped, heparinization was continued according to the TandemHeart protocol [11, 13]. Once the patient had recovered sufficiently that the device was no longer needed, it was removed in the operating room as described previously, and the femoral access site was surgically repaired [10, 11].
Paired t tests were used to compare hemodynamic and laboratory variables recorded before and after pVAD placement. All analyses were performed with SAS 9.1 software (SAS Institute, Cary, NC). Data are expressed as the mean ± standard deviation or as the median and range.
| Results |
|---|
|
|
|---|
The last available ejection fraction, obtained within the month before cardiac arrest or SRCS developed, was 0.183 ± 0.075 (range, 0.10 to 0.35); the peak aortic valve gradient was 36.5 ± 7.4 mm Hg, corresponding to the low ejection fraction. Just before pVAD implantation, all patients were receiving maximal vasopressor support; 6 patients underwent CPR for more than 20 minutes, and 1 patient had CPR for less than 5 minutes. The other 3 patients had an ejection fraction of 0.167 ± 0.058 (range, 0.10 to 0.20). Seven patients received an IABP before undergoing pVAD support. At device implantation, all 10 patients were intubated and receiving mechanical ventilation. Six patients were in acute renal failure and had anuria. The Society of Thoracic Surgeons preoperative mortality risk score for AVR [23] in these patients was 74.9% ± 24.5%. One patient underwent coronary stent placement immediately after pVAD implantation. Another patient had concomitant CABG and AVR.
In the 8 patients who received the pVAD in the catheterization laboratory, the average duration of support before AVR was 6.4 ± 3.8 days (range, 1 to 7 days), and the average flow was 3.1 ± 0.8 L/min. Placement of the device resulted in significant improvements in the levels of creatinine in these 8 patients (2.0 ± 0.9 vs 1.1 ± 0.3 mg/dL; p = 0.002), BUN (46.5 ± 26.5 vs 28.9 ± 13.3 mg/dL; p = 0.023), and LDH (919 ± 464 vs 339 ± 90 U/L; p = 0.047; Table 1). Before pVAD placement, 6 of the 8 patients were anuric; during TandemHeart support, their urine output improved and reached 155 ± 78 mL/h.
|
The 8 patients who received the pVAD in the catheterization laboratory preoperatively had an uneventful AVR procedure. During the AVR operation, the pVAD cannulas were left in place. After AVR, 7 of these 8 patients continued to be supported by the pVAD for 4.1 ± 1.7 days (range, 2 to 7 days); the patient who had preoperative ECMO did not require postoperative support. Seven patients were discharged home after 28 ± 13 days and remained alive 2 to 43 months after AVR; the 1 remaining patient died of sepsis after 34 days. After undergoing an uneventful AVR, the 2 patients who received the pVAD in the operating room for postcardiotomy cardiogenic shock continued to be supported by the device until they died of multiorgan failure 8 and 21 days later, respectively.
The total support period for all 10 patients averaged 10.9 ± 5.6 days. There were no perioperative complications related to device implantation, except that 1 patient required antegrade arterial reperfusion with a 12F cannula, placed in the superficial femoral artery below the pVAD arterial cannula, due to lower-extremity ischemia. The other 9 patients underwent pVAD implantation and support without complications. Intraoperative transesophageal echocardiography showed no thrombi in the left or right atrium or around the transseptal cannula before its removal. No patients required closure of the iatrogenic atrial septal defect that remained after transseptal cannula removal.
| Comment |
|---|
|
|
|---|
We used the TandemHeart pVAD to resuscitate 8 critically ill patients before AVR; 7 of these patients are alive and doing well, and 1 patient died of sepsis 34 days after AVR. We also used this device to resuscitate 2 other patients who were in cardiac arrest preoperatively and were rushed to the operating room while undergoing CPR. Both of these patients received the pVAD for postcardiotomy cardiogenic shock after AVR, and both died in the postoperative period.
Other pulsatile or nonpulsatile temporary mechanical circulatory support devices have been used for postcardiotomy cardiogenic shock [24, 25]. The TandemHeart provides similar flows as those devices, so we do not believe that its performance was insufficient in our 2 patients with postcardiotomy cardiogenic shock. On the contrary, we have used the TandemHeart successfully in other patients who had postcardiotomy cardiogenic shock after CABG, and other investigators have reported similar usage in this setting [26]. The most important factor in treating cardiogenic shock is immediate restoration of end-organ perfusion. Therefore, prompt institution of percutaneous mechanical circulatory support in the catheterization laboratory may contribute to better outcomes than rushing a patient with SRCS or with CPR in progress to the operating room for definitive CPB-assisted cardiac operations. Furthermore, use of the TandemHeart allows the patient's condition to stabilize so that the cardiac procedure may be performed electively.
Very limited data are available regarding comparable patient populations. Our patients were in end-stage circulatory collapse: 7 were undergoing CPR, 1 other patient had SRCS despite IABP support, and all required mechanical ventilation. Early studies yielded a surgical mortality rate of 50% for emergency aortic valve operations [27]. Patients with critical aortic stenosis and an ejection fraction of less than 0.35 without cardiogenic shock who have undergone AVR have had surgical mortality rates of 18% [28] to 21% [29] (in the latter group, only 1 of 52 patients analyzed had cardiogenic shock). In 2001 Monin and coworkers [30] reported 50% surgical mortality in patients with a low ejection fraction, critical aortic stenosis, and no contractile reserve. Use of percutaneous aortic balloon valvuloplasty as an alternative treatment in these patients has also resulted in extremely high mortality rates, ranging from 43% [31] to 71% [3]. Recently, Christ and coworkers [32] described 5 patients with critical aortic stenosis, cardiogenic shock, and multiple organ failure who underwent emergency AVR and survived. However, only 2 of these patients required mechanical ventilation and had been resuscitated preoperatively. The multiple organ failure score (Goris) [33] for the entire group was 6.8 ± 0.5. In our cohort of patients, the same score was 8.7 ± 1.4, which indicates a higher risk of death.
SRCS due to critical aortic valve stenosis necessitates immediate therapy to reverse end-organ failure and prevent death [34]. Until recently, the most commonly used treatment was IABP insertion and maximal inotropic support, followed by immediate AVR in patients with SRCS or undergoing CPR, but the outcomes of this approach were not encouraging.
Some authors have reported the use of long-term left ventricular assist devices (LVADs) to salvage patients in cardiogenic shock, but the results have not been satisfactory [35, 36]. With the use of extracorporeal membrane oxygenation (ECMO), outcomes have improved, but morbidity and mortality are still high [4, 7, 37]. Pagani and coworkers [7] used ECMO, and John and associates [25] used the CentriMag Ventricular Assist System (Levitronix, Waltham, MA) as a bridge to a long-term LVAD in patients with cardiogenic shock to stabilize their cardiac condition. We also have reported a series of patients with cardiogenic shock who underwent successful TandemHeart bridging to a long-term LVAD and had 100% survival [19]. We agree with John and coworkers [25] that rapid stabilization of the patients' condition was of primary importance in achieving better survival.
Because our current patients had SRCS or CPR in progress, they were not immediately good candidates for a long-term LVAD. To have resuscitated these patients with a surgically placed LVAD would have necessitated an emergency redo sternotomy in 6 patients, which by itself, would have posed a very high surgical risk. Furthermore, use of the Impella Recover microaxial-flow LVAD (Impella CardioSystems AG, Aachen, Germany) as a percutaneous device is not feasible in patients with critical aortic valve stenosis because the Impella must be placed through the aortic valve [38, 39]. This series shows the feasibility of using a TandemHeart preoperatively and during postoperative recovery in a cohort of patients with critical aortic valve stenosis and cardiac arrest or SRCS.
The primary goal in the treatment of SRCS is to reverse the hypoperfusion associated with end-organ damage and multisystem organ failure. Thiele and associates [40] analyzed the hemodynamic and metabolic effects of the TandemHeart vs the IABP in patients in cardiogenic shock after a myocardial infarction. Use of the pVAD improved end-organ perfusion, increased the cardiac index and output, decreased the pulmonary capillary wedge pressure and pulmonary artery pressure, and decreased the serum lactate level. It is not known how long a patient can be in cardiogenic shock before it becomes irreversible. We believe that the "emergency department door to LV unloading" time is crucial in that regard. Prompt availability of the pVAD in the catheterization laboratory allows immediate placement and unloading vs delayed unloading with surgically placed circulatory support devices. By providing enough flow (3 to 5 L) to reverse cardiac arrest or SRCS immediately, the TandemHeart offers a new treatment choice for these high-risk patients. Because end-organ function is improved before definitive surgical therapy is undertaken, emergency operations are avoided. In addition, the TandemHeart unloads the LV and allows successful resuscitation and defibrillation in patients undergoing CPR.
In our cohort, 8 patients underwent pVAD placement before AVR, and 7 were long-term survivors; in contrast, the 2 patients who received the pVAD for postcardiotomy cardiogenic shock after AVR both died. This limited experience suggests that optimizing end-organ function with the pVAD in severely ill patients before AVR may have an advantage for better survival. Given that the entire group had a Society of Thoracic Surgeons preoperative AVR mortality risk [23] of 74% ± 24%, the overall outcome was satisfactory, as 7 patients with preoperative pVAD support are alive and doing well after a mean follow-up of 13 months.
For patients with a transseptal cannula in their left atrium, a slightly different air removal technique is needed at the end of the AVR operation. This technique has been described elsewhere and is easily learned [10].
In conclusion, use of the TandemHeart pVAD may improve survival in high-risk cardiac surgical candidates with cardiac arrest or SRCS due to critical aortic valve stenosis. Indeed, using the percutaneous VAD to achieve end-organ recovery and assess the patient's neurologic status after CPR but before AVR should optimize outcomes in this patient group, which otherwise has a prohibitively high surgical risk.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
S. Kennon and A. Jain Circulatory support in severe aortic stenosis Heart, May 15, 2011; 97(10): 783 - 784. [Full Text] [PDF] |
||||
![]() |
L. Samuels Invited Commentary Ann. Thorac. Surg., December 1, 2009; 88(6): 1826 - 1827. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |