| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1998;66:1662-1669
© 1998 The Society of Thoracic Surgeons
a Cardiothoracic Surgery and Artificial Heart Program, University of Arizona Health Sciences Center, Tucson, Arizona, USA
b Loyola University Medical Center, Maywood, Illinois, USA
c LDS Hospital, Salt Lake City, Utah, USA
d University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
Accepted for publication May 20, 1998.
Address reprint requests to Dr Copeland, Cardiovascular and Thoracic Surgery, The University of Arizona Health Sciences Center, 1501 N. Campbell Avenue, Room 4402, Tucson, AZ 85724
e-mail: (jgc{at}u.arizona.edu)
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and methodsResultsCommentReferences |
|---|
Methods. The CardioWest trial started in 1993 in six United States institutions under an investigational device exemption from the Food and Drug Administration. Four centers contributed 27 implant and 18 matched retrospective control patients.
Results. Of the implant patients, 25 (93%) received a transplant, 24 (89% of the total, 96% of those transplanted) were discharged and are currently surviving. In the control group, 10 patients died awaiting transplantation, 8 received a transplant, and 7 were discharged with 6 surviving (p = 0.00001). All adverse events were documented with respect to time. Thirteen serious adverse events occurred, 11 of which occurred in the 2 patients that died during implant.
Conclusions. In a selected group of patients with end-stage heart disease, use of the CardioWest total artificial heart is lifesaving. When compared with the series of matched retrospective controls, a significant improvement in survival was found in the CardioWest implant group.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
The hypothesis of this study was that the CardioWest TAH could provide successful bridge to transplantation in a selected population of decompensating potential cardiac recipients and result in improved survival when compared with a control group. All implant and control patients met standard cardiac transplant selection criteria, had sufficiently large body and heart sizes, had clinical evidence of both left and right heart failure, and were hemodynamically decompensating despite inotropic and other forms of support considered to be beyond the usual.
Our TAH bridge-to-transplant trial that had begun in 1985 [1] and spanned 6 years [2–4] supported the hypothesis. In 1992, the Symbion (Tempe, AZ) registry reported the results of all 198 patients who received the TAH [5]. One hundred forty-three patients (72%) were transplanted and 85 (43%) discharged. Infections occurred in only 16 patients (9.4%). There were nine strokes (5.3%) and seven transient ischemic attacks (4.1%) during 170 months of the study. One center reported no neurologic events in 60 consecutive implants [6]. On the basis of this information, experience with sizing the device, and markedly improved implant hemostasis with the use of aprotinin and improved technique [7], we believed that TAH implantation could salvage critically ill transplant candidates. Further, we believed it offered these candidates excellent cardiac output and a chance for transplantation that was not available with either left ventricular assist devices or paracorporeal biventricular assist devices [8, 9].
With this in mind, we applied to the Food and Drug Administration for a new investigational device exemption to further study this device, formerly the Symbion TAH [10] renamed the CardioWest TAH. Two small device alterations had been made. The "skin button" on the transcutaneous drivelines was replaced with a Dacron (C.R. Bard, Inc, Billerica, MA) velour wrap, and a drop of silicone oil was added to the housing side of the diaphragm to prevent sticking.
Preliminary information on the international experience from 1993 to 1995 with the CardioWest TAH indicating outcome result parity with all other bridge-to-transplant devices has been published previously [11]. Presently, there have been 91 CardioWest TAH implants in the United States, Canada, and France. Sixty (69%) of those patients have received a transplant, 55 (92% of those transplanted and 63% of those implanted) have been discharged from the hospital after transplantation. The mean age of these patients was 45 years and the mean implant time was 30 days (range, 1–186 days).
We believe that the data in this study support the contention that the CardioWest TAH is not only efficacious when compared with matched controls, but also is the device of choice in a selected group of potential cardiac transplant recipients with hemodynamic decompensation, right heart failure, or early-onset low output syndrome.
|
Material and methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
Comparisons of control and implant data were made using Student’s t test. Kaplan-Meier curves were calculated for both groups and for all adverse events and tested for equality of survival distributions using log-rank, Breslow, and Tarone-Ware test statistics.
Adverse events are defined as follows. Bleeding is defined as excessive perioperative bleeding, any reoperation to control bleeding or treat tamponade, or any severe drop in hematocrit requiring transfusion as defined below. Any surgical procedure (implant, transplant, or other) requiring perioperative transfusion of eight or more units of packed red blood cells (PRBCs) or producing thoracic drainage of more than 200 mL per hour, 4 hours postoperatively, will be considered excessive bleeding. The perioperative period is defined to start at the initiation of a surgical procedure and end 48 hours after its completion. After 48 hours, transfusion of three units of PRBCs administered within a 24-hour period is also defined as bleeding.
Device malfunction is defined as any malfunction of the implantable device, including but not limited to valve failure or rupture of the diaphragm, driveline, or connections, which results in failure to pump blood or directs injury or potential injury to the patient. Also, any malfunction of the console, including but not limited to failure to respond to controls, failure to alarm appropriately during an emergency, or other component failure which results in failure to pump blood or directs injury or potential injury to the patient. This category includes all reoperation events to repair or replace the device. Injury is defined as the occurrence of an adverse event resulting from the malfunction.
Fit complication is defined as inadequate mediastinal space to accommodate the TAH as evidenced by compression of thoracic structures or impaired TAH performance. Compression of the thoracic organs or surrounding structures by the TAH is indicated by necrotic tissue, inability to close sternum, ascites, atelectasis, or low cardiac output resulting from kinked conduits or cuffs obstruction of venous return.
Hemodynamic insufficiency is defined as cardiac index less than 2.0 L/m2 per minute or systolic arterial blood pressure of less than 90 mm Hg for more than 4 hours.
Hemolysis is defined as plasma-free hemoglobin levels greater than 30 mg/dL persisting for 72 hours after bypass or blood transfusion. The patient must have two consecutive levels greater than 30 mg/dL sampled 24 hours apart.
Hepatic dysfunction is defined as elevated total bilirubin higher than 5.0 mg/dL. A second event occurs only when the total bilirubin has been less than 5.0 mg/dL for 7 days.
Infection is defined as positive culture or clinical sign of sepsis with negative culture. One infection is defined as multiple organisms at one site, repeated positive cultures of one organism at one site, or multiple sites culturing one organism. Clinical signs may include elevated white blood cell count, hypotension, or fever. In this study, infections have been subdivided into serious and nonserious. Serious is defined as infections causing death, contributing to death, or delaying transplant. Nonserious infections are all others.
A neurologic event is defined as any new episode of focal or global neurologic dysfunction such as transient ischemic attack, cerebral vascular accident, or seizure activity. The nature of the neurological event is considered to be thromboembolic, hemorrhagic, ischemic, metabolic, or medication related.
Peripheral thromboembolism is defined as a noncerebral thromboembolic event resulting in motor, sensory, or ischemic impairment.
Renal dysfunction is defined as serum creatinine level higher than 5.0 mg/dL or any postimplant dialysis treatment. A second event is documented only when the serum creatinine remains below 5.0 mg/dL for 7 days (in an undialyzed patient). Renal dysfunction requiring dialysis is documented as a single event for the duration of the dialysis treatment.
Reoperation is defined as corrective or exploratory procedure of any kind after the implant surgical procedure. Causes for reoperation include bleeding, repositioning of the device, and events unrelated to the device but having a direct bearing on the patient’s condition.
Respiratory dysfunction is defined as postoperative ventilatory support required for 10 days or more, or reintubation for respiratory dysfunction.
Adverse events not classified by the definitions listed above and which in the investigator’s judgment are of comparable importance were categorized as miscellaneous.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Of the 25 patients who received transplants, 24 survived (96%) and were discharged. One death occurred on the 15th posttransplant day from hemodynamic insufficiency secondary to multivessel coronary artery disease in the donor heart. Thus, the total survival was 24 of 27 (89%). The mean follow-up time has been 20 months (range, 3–48 months) with no other deaths.
During TAH support, all 27 implanted patients had at least one adverse event (median, 5 events per patient; range, 1–22). There were 175 adverse events (Table 2, Figs 1 and 2). In Table 2, the number of adverse events, number of patients, and the percentage of patients having specific events and device-related and nonrelated adverse events are shown. In Figure 1, the relationship of total specific adverse events to device-related adverse events are shown. Most adverse events were not device related. Only in the device malfunction (5 events in 3 patients), hemodynamic insufficiency (5 events in 3 patients), neurologic (17 events in 9 patients), and splenic emboli (2 events in 1 patient) groups were most events device related (21 of 29 device related). Of the two deaths during TAH support, the one secondary to diaphragm tear was device related. That patient also accounted for device-related adverse events in other categories, including 1 device malfunction, 3 hemodynamic insufficiency (3 of the 5 events), 1 hepatic dysfunction, 1 neurologic, 1 reoperation, general, and 1 respiratory dysfunction. His total number of adverse events was 23, including 8 device-related events. The other patient that died on TAH support died from infection and multiple organ failure. He had 8 adverse events including 5 infections (4 serious), 1 renal dysfunction, and 1 respiratory dysfunction. These 2 patients, in addition to representing the two deaths, accounted for 11 of the 13 complications that we called serious (causing death, contributing to death, or delaying transplantation). Another serious complication was an infected decubitus ulcer that was not sufficiently healed to allow transplantation for 75 days. The last serious infection was a driveline infection that prevented transplantation for 29 days (Table 3).
|
|
|
|
In Figure 2, the percentage and number of patients with an adverse event is shown with respect to time. This summary figure shows that nearly all of the adverse events occurred in the first weeks after implantation. Time-related, event-specific analysis is shown in Table 4.
|
All adverse events were reviewed and documented in detail. A summary of event-specific information follows. All events that met the bleeding definition were at 11 days or less after implantation. Five were associated with reoperation. None of the bleeding events influenced clinical outcome.
In all but one case, device malfunction refers to momentary changes in state of consciousness that did not affect outcome due to driveline kinks (2 patients), air tank change (1 patient), resetting the vacuum (1 patient), and to the one fatal diaphragm tear that led to death during the 17th week. In the first 4 weeks, the distribution over time appeared to be random (Table 3). Three of these events, weeks 11, 13, and 14 involved the first patient in the series.
Splenic emboli were documented by computed tomographic scan on two occasions in the same patient (weeks 7 and 16) when left upper quadrant pain was noted. Neither of these episodes affected clinical outcome.
The one fit complication required reoperation on the second day after implant. Lateral repositioning successfully relieved an inferior vena caval stenosis.
Hemodynamic insufficiency was attributed to medication effect while the patient was sleeping in one case and to hypovolemia in another case and was of no consequence in either situation. The other three occurrences of this event were in the same patient, who had the diaphragm tear with cardiac indices of 2.0 L/m2 per minute or less on days 115, 116, and 118 (all in the 16th week) before his death in the 17th week.
The patient with the fatal diaphragm tear had hepatic dysfunction as a consequence of his device malfunction during the 17th week after implant just before his death. The other 11 cases of hepatic dysfunction were in the first 3 weeks after implant. Ten events were residual from before implant, which resolved with device support. The episode in the third week was attributed to a septic episode. Neither the sepsis nor the hepatic dysfunction affected the outcome.
None of the 57 nonserious infection events in 24 patients affected clinical outcome. These events included 21 positive sputum cultures, 10 genitourinary infections, 6 positive blood cultures, 6 positive driveline cultures, 4 decubitus ulcers, 2 stool pathogens, and 1 each of the following: oral herpes, oral candida, great toe infection, and bronchial secretions.
The serious infection events were separated from other infections because they caused death, contributed to death, or delayed transplantation. As seen in Table 4, all occurred during the first month of implantation. Delay of transplant for 75 days was caused by a coccygeal decubitus ulcer. In another patient, delay of 29 days was caused by a medial driveline infection. Both patients underwent successful transplantations and are alive. The remaining three infections occurred in the patient who died of mediastinitis and sepsis during the third implant week. The infections were Pseudomonas from a chest tube site (day 9), Enterobacter from sputum (day 1), and gram-negative rods and Serratia from mediastinal drainage (day 2).
There were 17 neurologic events in 9 patients, most occurred in the first 2 months (Table 4). There were 9 transient ischemic attacks characterized by neurologic symptoms that resolved within 24 hours. In addition, there were 3 seizures, 2 impaired states of consciousness secondary to anoxic or metabolic causes, 1 retinal hemorrhage, 1 retinal embolus, and 1 stroke. The stroke, a hemiplegia, presumed due to a left middle cerebral artery embolus, occurred on the day of transplantation. All residual effects resolved within 3 months after transplantation.
All episodes of renal dysfunction occurred in the first and second weeks after implant and were attributed to the patients’ preoperative conditions. One event occurred in the patient who died of mediastinitis and sepsis during the third week after implant. No other renal events affected clinical outcome.
Eight chest reoperations in 7 patients were done for bleeding (n = 4), tamponade (n = 2), device repositioning (n = 1), and sternal debridement (n = 1). None of these events affected clinical outcome.
There were 31 general reoperations in 13 patients, with 24 procedures done in the first through third weeks after implant. The procedures included bronchoscopy (n = 12), nasal cautery (n = 9), endoscopic gastroduodenoscopy, thoracentesis (n = 3), arm fasciotomy (n = 1), laparoscopic cholecystectomy (n = 1), and exploratory laparotomy (n = 1). Clinical outcome was not affected by any of these events.
There were 11 respiratory dysfunction events, all reintubations, in 7 patients including 2 events in the 2 patients who died. Ten events occurred in the first and second weeks after implant. Nine did not affect clinical outcome [13].
Eighteen control patients met entry criteria for the study and were similar to the implant group in demographics, risk factors, hemodynamics (the mean cardiac index for the implanted group was significantly lower than that of the controls), and laboratory values (Table 1). Ten of these 18 patients died before transplantation, after entrance criteria were met, 1 to 43 days (mean 7.64, median 3 days). All had hemodynamic insufficiency. Eight patients were transplanted after a mean waiting time of 23.1 days (range, 6–98 days). There was one immediate death because of donor heart failure and one late death during the first year after transplant. All 6 patients surviving at 1 year were also alive 2 years after transplant. Thus, 8 of 18 patients in the control group (44%) were transplanted, 7 of 8 (39% of the total) were discharged home, and 6 (33%) of 18 survived 1 to 2 years after transplantation.
Actuarial survival curves for these two groups are shown in Figure 3. They document a significant (p < 0.00001) survival benefit in the use of the CardioWest TAH.
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
In this study, of the 27 patients implanted with the CardioWest TAH, 25 (93%) were transplanted and 24 (89%) were discharged home and are surviving to the present. The international experience with this device between 1993 and 1995 in 40 implants was a 65% discharge rate that compared favorably to those reported for other bridge-to-transplant devices, including the Novacor (Oakland, CA) and the TCI-Heartmate (Woburn, MA) [11]. That percentage, as noted in the introduction, has increased to 67% internationally. The difference in the CardioWest experience in this study and the international experience appears to be in patient selection. The careful selection criteria requiring that, at the time of device implantation, patients are acceptable for transplantation immediately excludes some patients with progressive hemodynamic deterioration. In addition, many patients are excluded because of age, size, and type of preimplant support. Thus, if one looks at the percentage of patients surviving transplantation, for this study 96% and for the international experience 92% to 93%, the difference in total outcome results from a greater attrition rate on device support in the international group. The selection criteria for device implantation possibly are not as tight in the international group, and the selection criteria for this study are not designed for what may be the eventual use of the device, but instead for the purpose of scientific comparison of the implant group with a matched control group.
The other striking finding in this study is the wide gap in survival rates between the implant and control groups. This finding confirms that, by using selection criteria in end-stage heart disease, we can with some certainty predict the imminent demise of a patient. Such a statement may sound elementary to the seasoned clinician, but is fundamental to obtaining FDA approval for the commercialization of any type of artificial heart. The wide gap in survival between patients who received implants and controls also confirms the efficacy of the CardioWest TAH. This efficacy is seen in a selected group of patients in imminent danger of dying and with evidence of right and left heart failure (mean central venas pressure 19.7 mm Hg and pulmonary vascular resistance 233 dynes · sec/cm5).
Among the 175 adverse events in this study were 2 deaths and 13 serious adverse events that were the cause of death, contributed to death, or delayed transplantation. Eleven of these serious adverse events were in the 2 patients who died. The two other events included 1 decubitus ulcer and one driveline infection that delayed transplantation. The other 160 events were judged not to affect clinical outcome. Among those were 9 transient ischemic attacks, 1 retinal embolus, 1 stroke, and 2 splenic emboli. Most of these events occurred within the first or second week after implantation. The stroke occurred on the day of transplantation and might have been influenced by that procedure, but it has been counted as a device-related event. The linearized rate for stroke was 0.97% per patient-year; the linear rate for all emboli was 12.7%. None of the embolic events affected clinical outcome. One might infer that in a study with a fixed anticoagulant protocol, a lower rate might have been obtained. In addition, in a less rigorous study, some of these embolic events might have been missed.
One hundred four (59%) of the adverse events occurred within the first 2 weeks after implantation. Many of these were residual problems relating to the poor condition of the patient before implantation. All ten early cases of hepatic dysfunction and seven of the eight cases of renal dysfunction were reversed with device support. Several other early events were related to nonserious infections and to minor reoperations such as bronchoscopy and nasal cautery. Nearly all of these early problems were overcome, as were all but the 11 nonfatal events (6%) that contributed to the deaths of 2 patients.
The patient who died of device malfunction developed a tear in the first of four layers of the left ventricular diaphragm. This tear was eventually diagnosed by cardiac catheterization and transesophageal echocardiogram, but not before the patient had severe multiple organ failure. Reoperation to replace the ventricle was considered, but rejected because of his moribund state. This represents the only diaphragm tear in the history of this device dating back to the initial Jarvik-100 hearts and encompassing over 16 patient-years. Examination of the diaphragm by routine and scanning electron microscopic techniques failed to identify a cause. It is therefore assumed to be a random and rare event.
We believe that 89% survival to discharge of those implanted (25 of 27 patients) is remarkable. The reliability of this device as a biventricular blood pump provides a physiologic hemodynamic state that allows time for healing, nutrition, and improved general status so that a patient becomes a better risk for transplantation. In this study, 24 of the 25 transplanted patients are long-term survivors. This survival rate is considerably better than that in nonbridged patients of 80% at 1 year, which has been repeatedly published in the registry of the International Society for Heart and Lung Transplantation [14]. In our experience with 23 consecutive bridge-to-transplant patients with the CardioWest TAH, the posttransplant survival has been 100%. We believe that the CardioWest TAH is a valuable device for bridge to transplantation in patients who qualify as cardiac transplant candidates, have biventricular failure, large hearts, adequate body size (>1.7 m2), and are rapidly decompensating.
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. G. Copeland, R. G. Smith, R. K. Bose, P. H. Tsau, P. E. Nolan, and M. J. Slepian Risk factor analysis for bridge to transplantation with the CardioWest total artificial heart. Ann. Thorac. Surg., May 1, 2008; 85(5): 1639 - 1644. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Aggarwal, F. Cheema, M. C. Oz, and Y. Naka Long-Term Mechanical Circulatory Support Card. Surg. Adult, January 1, 2008; 3(2008): 1609 - 1628. [Full Text]
|
|
|
|
 |
|
 J. G. Copeland, R. G. Smith, F. A. Arabia, P. E. Nolan, G. K. Sethi, P. H. Tsau, D. McClellan, M. J. Slepian, and the CardioWest Total Artificial Heart Investigator Cardiac Replacement with a Total Artificial Heart as a Bridge to Transplantation N. Engl. J. Med., August 26, 2004; 351(9): 859 - 867. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D R Wheeldon Mechanical circulatory support: state of the art and future perspectives Perfusion, July 1, 2003; 18(4): 233 - 243. [Abstract] [PDF]
|
|
|
|
 |
|
 J. G. Copeland Multicenter bridge to transplantation with the HeartMate assist device: Evaluation from another perspective J. Thorac. Cardiovasc. Surg., February 1, 2003; 125(2): 228 - 230. [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Simmons Currents in Contemporary Ethics J. Law Med. Ethics, September 1, 2001; 29(3-4): 401 - 406. [PDF]
|
|
|
|
|
|
J. G. Copeland, F. A. Arabia, R. G. Smith, G. K. Sethi, P. E. Nolan, and M. E. Banchy Arizona experience with CardioWest total artificial heart bridge to transplantation Ann. Thorac. Surg., August 1, 1999; 68(2): 756 - 760. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Nose, K.-i. Nakata, M. Yoshikawa, G. V. Letsou, A. Fujisawa, E. Wolner, and H. Schima Development of a totally implantable biventricular bypass centrifugal blood pump system Ann. Thorac. Surg., August 1, 1999; 68(2): 775 - 779. [Abstract] [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 |
