Author + information
- Received December 1, 2008
- Revision received March 3, 2009
- Accepted March 10, 2009
- Published online July 21, 2009.
- Francis D. Pagani, MD, PhD⁎,⁎ (, )
- Leslie W. Miller, MD†,
- Stuart D. Russell, MD‡,
- Keith D. Aaronson, MD⁎,
- Ranjit John, MD§,
- Andrew J. Boyle, MD§,
- John V. Conte, MD‡,
- Roberta C. Bogaev, MD∥,
- Thomas E. MacGillivray, MD¶,
- Yoshifumi Naka, MD#,
- Donna Mancini, MD#,
- H. Todd Massey, MD⁎⁎,
- Leway Chen, MD⁎⁎,
- Charles T. Klodell, MD††,
- Juan M. Aranda, MD††,
- Nader Moazami, MD‡‡,
- Gregory A. Ewald, MD‡‡,
- David J. Farrar, PhD§§,
- O. Howard Frazier, MD∥,
- HeartMate II Investigators†
- ↵⁎Reprint requests and correspondence to:
Dr. Francis D. Pagani, Section of Cardiac Surgery, University of Michigan Health System, Cardiovascular Center, Room 5161, 1500 East Medical Center Drive, SPC#5864, Ann Arbor, Michigan 48109
Objectives This study sought to evaluate the use of a continuous-flow rotary left ventricular assist device (LVAD) as a bridge to heart transplantation.
Background LVAD therapy is an established treatment modality for patients with advanced heart failure. Pulsatile LVADs have limitations in design precluding their use for extended support. Continuous-flow rotary LVADs represent an innovative design with potential for small size and greater reliability by simplification of the pumping mechanism.
Methods In a prospective, multicenter study, 281 patients urgently listed (United Network of Organ Sharing status 1A or 1B) for heart transplantation underwent implantation of a continuous-flow LVAD. Survival and transplantation rates were assessed at 18 months. Patients were assessed for adverse events throughout the study and for quality of life, functional status, and organ function for 6 months.
Results Of 281 patients, 222 (79%) underwent transplantation, LVAD removal for cardiac recovery, or had ongoing LVAD support at 18-month follow-up. Actuarial survival on support was 72% (95% confidence interval: 65% to 79%) at 18 months. At 6 months, there were significant improvements in functional status and 6-min walk test (from 0% to 83% of patients in New York Heart Association functional class I or II and from 13% to 89% of patients completing a 6-min walk test) and in quality of life (mean values improved 41% with Minnesota Living With Heart Failure and 75% with Kansas City Cardiomyopathy questionnaires). Major adverse events included bleeding, stroke, right heart failure, and percutaneous lead infection. Pump thrombosis occurred in 4 patients.
Conclusions A continuous-flow LVAD provides effective hemodynamic support for at least 18 months in patients awaiting transplantation, with improved functional status and quality of life. (Thoratec HeartMate II Left Ventricular Assist System [LVAS] for Bridge to Cardiac Transplantation; NCT00121472)
Heart transplantation remains the most successful treatment option for patients with advanced heart failure refractory to medical therapy (1). As a consequence of limited donor availability (1), left ventricular assist device (LVAD) therapy has become an established treatment for patients with advanced heart failure as either a bridge to transplantation (BTT) (2) or as permanent support as an alternative to transplantation (3,4). Historically, patients have been supported by devices engineered with pulsatile design (i.e., HeartMate IP1000, VE, or XVE, Thoratec PVAD or IVAD, Thoratec Corporation, Pleasanton, California; or Novacor, World Heart Corporation, Oakland, California) (2,5–8). These devices are designed with an internal pumping chamber and inflow and outflow valves permitting cyclic filling and emptying with pump actuation elicited by either pneumatic or electrical systems (9). Previous studies have demonstrated the efficacy of these devices with regard to improvement in survival to transplantation (2) and improvement in survival compared with optimal medical management for patients with advanced heart failure, not candidates for transplantation (3).
Pulsatile devices have limitations in their design that preclude their practical use for extended mechanical circulatory support (MCS). These limitations include a large pump size, requirement for extensive surgical dissection for implant, a large body habitus of the recipient, the presence of a large-diameter percutaneous lead for venting air, and audible pump operation (2,7). A critical limitation of the majority of these devices has been the high incidence of reoperation for device exchange for device infection or malfunction (3,10–12).
The REMATCH (Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure) trial demonstrated a survival advantage for LVAD therapy over optimal medical management for patients with advanced heart failure who were not eligible for transplantation. Although demonstrating the potential of MCS in providing improved survival, this trial demonstrated the risk of mechanical failure and device-related complications inherent in the pulsatile HeartMate VE LVAD (3,12). In patients surviving up to 2 years on device support, nearly 65% underwent replacement for infection or malfunction (12).
The development of continuous-flow rotary pump technology represents an innovative design for LVADs (13–18). These devices have the advantage of a smaller pump size and potential for greater mechanical reliability by simplification of the pumping mechanism (13,14). Reports from clinical trials of these newer pump designs have demonstrated efficacy in providing hemodynamic support and favorable risk-to-benefit ratio (15–18).
The HeartMate II LVAD is a continuous-flow rotary pump that has completed a U.S. Food and Drug Administration (FDA)-approved pivotal trial in 133 BTT patients (18). Since this report, 336 additional patients have undergone implantation of the HeartMate II LVAD as of April 2008 through a continued-access protocol approved by the FDA. We report on the first 281 patients entered into this clinical evaluation who have completed study end points or at least 18-month follow-up after LVAD implantation.
Patients were enrolled in the study conducted at 33 centers in the U.S. between March 2005 and April 2008 (18). The study was supervised by the Thoratec Corporation. The clinical affairs and biostatistics departments at Thoratec designed the trial in consultation with the FDA and clinical investigators. Coordinators at each site collected study data, which was forwarded to the data analysis center of the sponsor. The academic authors vouch for the completeness and accuracy of the data and the analyses. A data and safety monitoring board, consisting of 4 independent physicians and 1 biostatistician who were not investigators in the study, met routinely to review study compliance, adverse events, quality of life, and outcomes of patients. These 5 committee members were compensated for their time, but none have any financial interest in the Thoratec Corporation or stand to gain financially from the outcome of the trial. A clinical events committee of 4 independent physicians who were not involved in the conduct of the trial reviewed, classified, and adjudicated the causes of death and all adverse events. The study was conducted in compliance with FDA regulations for Good Clinical Practices. The protocol was approved by the FDA and the institutional review board at each participating center.
Patients with heart failure who were on a waiting list for heart transplantation at each center were eligible for study enrollment. Patients were required to have symptoms of New York Heart Association (NYHA) functional class IV heart failure and to be ill enough to have high priority for transplantation (United Network for Organ Sharing status 1A or 1B). A complete list of study inclusion and exclusion criteria have been reported (18). All participating patients provided written informed consent before enrolling in the study.
Data collection baseline assessment
Baseline data were obtained upon patient consent and enrollment into the study. Information collected for baseline data have been previously reported (18).
The continuous-flow LVAD used in this study was the HeartMate II LVAS (Thoratec Corporation), which is a rotary pump with axial flow design (Fig. 1)(18). The system design and operating performance of the device have been previously described (18).
Surgical implantation of the HeartMate II LVAS was conducted according to the HeartMate II LVAS “Instructions for Use.” Post-operative treatment included initiation of an anticoagulation regimen (18).
Follow-up after device implantation
Post-operative medical care was managed according to each investigator's usual practice. Details of the protocol for patient follow-up after device implantation have been previously reported (18). Readmissions to the hospital and adverse events were recorded throughout the study as they occurred with the use of standardized definitions (18). All deaths of patients and causes of death were determined at autopsy when possible, examination of medical records, or by interviews with family members. Final adjudication was determined by the clinical events committee. Major device malfunctions were defined as those resulting in death or requiring device replacement.
The principal outcomes assessed through 18 months after enrollment were the proportions of patients who had undergone transplantation, had undergone explantation of the device because of recovery of ventricular function, or continued with ongoing MCS. The proportion of patients who died on support and those who were withdrawn from the study were also followed for 18 months. Patients who had the original continuous-flow pump replaced with another identical device remained in the study, whereas patients who had the original pump replaced with a different type of device were withdrawn from the study.
Differences between measures of continuous variables before and after implantation were analyzed by a paired ttest. McNemar's test was used for comparisons between paired categorical variables. The level of statistical significance was set at p < 0.05. All statistical comparisons were 2 sided. Biochemical and hemodynamic variables are presented as mean (±SD), and median and range were used where appropriate. Discrete variables are presented as percentages. Adverse events are presented both as the percentage of patients who had the event and as event rates per patient-year. Survival analysis for patients continuing on mechanical support was performed with the Kaplan-Meier method. Patients were censored for transplantation, recovery of the natural heart, and withdrawal from the study. The Kaplan-Meier method also was used for analysis of freedom from death due to major device malfunction or need for device replacement for all causes.
A total of 469 patients meeting study-entry criteria were enrolled into the study as of April 2008 and underwent implantation of the continuous-flow LVAD as a BTT. The initial 281 of the 469 patients that completed study end points or have at least 18 months of follow-up with ongoing device support comprise the patients in this report. These 281 patients include additional follow-up of the first 133 patients implanted as part of a primary cohort and previously reported (18). Most of the subjects were men with a median age of 54 years (Table 1).The most frequent heart failure etiology was nonischemic cardiomyopathy. All patients had symptoms of advanced heart failure despite optimal medical management with oral medications. The majority of patients in the study were receiving an intravenous inotrope, with approximately one-third requiring ≥2 inotropes. Patients not administered inotrope therapy were intolerant to it as the result of ventricular arrhythmias. Forty-five percent of patients were on concomitant support with an intra-aortic balloon pump. Previous biventricular pacing therapy failed in a significant proportion of these patients. Pre-operative hemodynamic and laboratory assessment were consistent with a group of patients with advanced heart failure requiring inotrope and/or intra-aortic balloon pump support.
Of the 281 patients, 222 (79%) either received a transplant, recovered cardiac function and underwent device explantation, or remained alive with ongoing LVAD support at 18-month follow-up (Fig. 2).At 18 months, 157 (55.8%) patients had received a heart transplant, 58 (20.6%) remained alive with ongoing LVAD support, 56 (19.9%) patients died, 7 (2.5%) patients recovered cardiac function and underwent device explantation, and 3 (1%) patients were withdrawn from the study after device explantation and exchange for another type of LVAD. The median time to transplantation was 118 days (range 10 to 545 days), median time to death was 64 days (range 0 to 797 days), and the median time to pump removal after cardiac recovery was 302 days (range 161 to 558 days). The median duration of support for all patients was 155 days (range 0 to 1,026 days), with a cumulative follow-up of 181 patient-years.
Overall survival for the patients who continued on LVAD support was 82% (95% confidence interval [CI]: 77% to 87%) at 6 months, 73% (95% CI: 66% to 80%) at 1 year, and 72% (95% CI: 65% to 79%) at 18 months (Fig. 3).Of the 157 patients that received a transplant, post-transplant survival was 96% at 30 days and 86% at 1 year. As of June 2008, 42 patients were alive with device support with a median duration of 1.6 years (longest duration 3.1 years). Of these patients, 71% remained active on the transplant list, and 29% were not listed (Table 2).Only 3 patients were not listed for irreversible medical conditions or degree of organ dysfunction.
Average LVAD estimated blood flow at 6 months of support was 5.6 ± 0.9 l/min (flow index 2.83 ± 0.45 l/min/m2) at a pump speed of 9,467 ± 499 rpm. Left ventricular end-diastolic dimension determined by echocardiography reduced from 69.7 ± 12.3 mm at baseline to 59.2 ± 15.1 mm at 1 week and 56.7 ± 14.5 mm at 6 months. Anticoagulation with warfarin resulted in an average international normalized ratio throughout support of 2.1 ± 0.8 (median 2.0). Twenty-five (8.9%) patients underwent transplantation during their initial hospital stay, and 33 (11.7%) patients died on LVAD support before discharge.
Two hundred twenty (78%) patients were discharged from the hospital with the LVAD, with a median hospital stay after surgery of 25 days (range 8 to 180 days). The median number of days out of hospital before transplantation, readmission, or death was 55.5 (range 1 to 892 days). One hundred forty nine (68%) patients required rehospitalization after discharge, with a median duration of rehospitalization of 5 days (range 0 to 209 days).
Causes of death
The primary causes of death were sepsis in 11 patients (4%), stroke in 10 (4%; ischemic: n = 5, 2%; hemorrhagic: n = 5, 2%), and right heart failure in 7 (3%). Other causes included device-related deaths in 7 patients (3%), multiorgan failure in 5 (2%), anoxic brain injury in 3 (1%), bleeding in 3 (1%), and other causes in 10 (4%) (cancer, respiratory failure, hyperthermia, air embolism, and unknown). There were a total of 7 device-related deaths, 4 that were attributed to malfunction of implanted components (pump thrombosis in 2, inflow graft that was twisted during implantation in 1, and outflow elbow disconnect in 1) and 3 that were attributed to external components (severed percutaneous lead in 1, power loss in 2). Only 13 of the deaths in the 281 (4.6%) patients (or 23% of the 56 deaths) occurred after 6 months of device support and included 5 from sepsis, 2 from power loss of the LVAD, 2 from withdrawal of support, and 1 each from cancer, respiratory failure, hemorrhage, and unknown.
Bleeding requiring transfusion and surgery were the most common adverse events (see Miller et al.  for definition) in the study and were primarily observed within the first 30 days of device implantation (Table 3).Stroke was observed in 25 (8.9%) patients, with the event rate greatest in the first 30 days. Ischemic strokes were more common than hemorrhagic strokes. Five of the 15 (33%) ischemic strokes occurred within 48 h. Almost one-half of the strokes were fatal (10 of 25; 40%). After the first 30 days, the event rates for ischemic and hemorrhagic stroke were 0.05 and 0.03 events per patient year, respectively. Six (2%) additional patients had transient ischemic attacks that completely reversed. Other nonstroke neurologic events, including seizures and confusion, were observed in 15 (5%) patients. Psychological symptoms developed in 16 (6%) patients. Localized infection not related to the device occurred in 84 (30%) patients. Infection associated with the percutaneous lead was observed in 41 (14%) patients, and there were 5 (2%) pre-peritoneal pump pocket infections. Respiratory and renal failure occurred in 72 (26%) and 30 (11%) patients, respectively. Thirty-six (13%) patients had right heart failure requiring inotrope support for more than 14 days. Of the 53 (19%) patients that developed post-operative right heart failure, survival with continued LVAD support at 18 months was 62 ± 8%. Twenty-six of these 53 (49%) patients with right heart failure received a transplant. Seventeen (6%) patients received temporary support with right ventricular assist devices (median time of support 11.5 days; range 0 to 148 days). The median duration of post-operative inotrope support was 9 days.
A total of 11 (4%) patients underwent 12 HeartMate II LVAD replacements for either device thrombosis (n = 4; 1.4%) (2 of whom subsequently died), surgical complications at the time of LVAD implantation (n = 3; 1.1%), percutaneous lead wire damage (n = 4; 1.4%), or for development of device infection (n = 1; 0.3%). One patient had 2 of the pump replacements (thrombosis and infection). Of the 12 pump replacements, 9 (75%) were replaced with another HeartMate II LVAD, and the patients continued in the study, and 3 (25%) were replaced with another type of LVAD device, and the patients were withdrawn from the study. There were no failures of the mechanical pumping mechanism. The median time to pump replacement was 106 days (range 0 to 672 days). The freedom from major device malfunction resulting in death (n = 4) or device replacement for all causes (malfunction, thrombosis, or infection; 9 without deaths) was 96% (95% CI: 95% to 99%) at 6 months, 93% (95% CI: 90% to 98%) at 1 year, and 92% (95% CI: 88% to 97%) at 18 months (Fig. 4).
Fifty-nine percent of patients (165 of 281) required 368 operations or procedures after device implantation. The majority of these occurred within 30 days of device implantation (245 of 368; 67%), and most were required for re-explorations or sternal closures for bleeding complications (177 of 245; 72%) followed by temporary RVAD insertion or removal (n = 18; 7%), tracheostomy (n = 12; 5%), implantable cardioverter defibrillator insertion or replacement (n = 9; 4%), infection (n = 7; 3%); pump replacement (n = 4; 2%), and various other cardiac (n = 15; 6%) and noncardiac (n = 3; 1%) procedures. After 30 days, the most frequent indication for reoperation was for infection complications (49 of 123; 40%), bleeding (n = 19; 9%), pump replacement (n = 8; 7%), RVAD insertion or removal (n = 3; 2%), implantable cardioverter defibrillators (n = 5; 4%), and various other noncardiac (n = 24; 20%) and cardiac procedures (n = 14; 11%).
Functional assessment, quality of life, and end-organ function
Functional assessment with 6-min walk and NYHA functional classification were performed for patients remaining on device support up to 6 months (Table 4).Of 109 patients with paired values at baseline and at 6 months, only 14 of 109 (13%) were able to perform a 6-min walk test at baseline, compared with 97 (89%) patients after 6 months of support, and there was a significant improvement in distance walked between baseline and 6 months.
Additionally, NYHA functional classification improved from 3.9 ± 0.3 at baseline (with 0% of patients in NYHA functional class I or II) to 1.8 ± 0.7 at 6 months (with 83% in functional class I or II). Quality of life assessed by both Minnesota Living With Heart Failure and the Kansas City Cardiomyopathy Questionnaire was significantly improved at 6 months compared with baseline (Table 4), with mean scores improving over 25 U, or 41% and 75%, respectively. Hepatic (total bilirubin, serum aspartate aminotransaminase, and serum alanine aminotransaminase) and renal (blood urea nitrogen) function significantly improved from baseline to 6 months, but changes in serum creatinine were not statistically significant.
Continuous-flow rotary pumps represent a novel design concept for LVADs. Findings from this study have validated the efficacy and safety profile of this design for patients awaiting transplantation. At 18 months after LVAD implantation, a majority of patients (79%) underwent transplantation, had cardiac recovery, or remained alive with ongoing LVAD support. Of the patients remaining alive with ongoing LVAD support, the majority were actively listed for transplant, obtained satisfactory renal and hepatic function and functional status, and were free from significant neurological injury. For patients that received a transplant, 1-year survival after transplant survival was similar (86% at 1 year) to that reported for transplant recipients by the Scientific Registry of Transplant Recipients (1). These data confirm the feasibility of extended LVAD support with a continuous-flow LVAD and importantly provide support for evaluating this technology for permanent MCS as an alternative to transplantation.
Major adverse events and deaths were associated with LVAD implantation (Table 3). Most deaths occurred within the first 3 months and were attributable to stroke, infection, or multiorgan failure. The skewed distribution of mortality and morbidity events early after implantation suggests that the acuity of patient illness contributes significantly to early deaths and complications. Improvements in patient selection, timing of LVAD implantation, and improvements in device technology should lead to improvements in outcomes. The refinement of risk assessment scores for patients being evaluated for LVAD therapy will assist in defining optimal patient selection and timing of LVAD intervention (19,20). Multivariate analyses have shown the importance of pre-existing liver and renal dysfunction, coagulation abnormalities, and lower serum albumin as risk factors that can identify patients who may be too ill for LVAD surgery (19).
There have been no randomized comparisons of implantable LVAD therapy with optimal medical management in patients awaiting transplantation because of the perceived benefits of LVAD therapy in patients at imminent risk of death and the potential ethical issues of withholding life-saving therapy. Further, there have been no randomized comparisons of different types of implantable LVAD designs in this population. Data pertaining to the efficacy and safety profile of a continuous-flow, rotary pump from this study can be placed in perspective when compared historically with previous FDA-approved clinical studies investigating the use of devices with pulsatile design for patients undergoing transplantation.
In these studies, which evaluated >500 patients during a 10-year period, overall survival to transplantation or survival on LVAD support at 6 months was approximately 70% with an average support duration typically of 1 to 3 months and was a consistent finding in each of the clinical studies (2,6,8,21). In a comparison with devices with pulsatile design, the continuous-flow, rotary pump demonstrated at least equal efficacy with regard to hemodynamic support, ability to improve renal and hepatic function, rates of transplantation, and overall patient survival but typically over longer durations of support compared with earlier studies with pulsatile devices. Actuarial survival on LVAD support for patients in this study was 73% at 1 year and 72% at 18 months and appeared superior to survival observed during pump support with a pulsatile device (53% at 1 year) (2).
Importantly, significantly fewer deaths were observed during late follow-up (6 to 18 months) in the present study, suggesting that the incidence of major events contributing to deaths such as stroke, infection, and device malfunction were lower with the new design. This late reduction in deaths has not been previously observed with pulsatile pumps for either BTT (2) or destination therapy (3) and suggests that refinements in device technology have improved late outcomes. Importantly, excellent late survival on LVAD support (12 to 18 months) was maintained in the absence of continuing high rates of attrition to transplantation (Fig. 2), suggesting that significant device complications requiring urgent transplantation were not prevalent.
The rates of major adverse events appear less for continuous-flow rotary pumps when compared with devices with pulsatile design when comparing rates of adverse events with similar definitions to historical trials (2). The adverse event rates (in events per patient-year) of bleeding requiring surgery (0.45 vs. 1.47), percutaneous lead infection (0.26 vs. 3.49), stroke (0.14 vs. 0.44), nonstroke neurological events (0.09 vs. 0.67), and right heart failure requiring a right ventricular assist device (0.09 vs. 0.30) were significantly less in this current study compared with those observed during clinical evaluation of a pulsatile pump, respectively (2). This comparison is historical in nature, and differences in rates of adverse events may have been influenced by differences in acuity of patient illness or improvements in patient management over time. However, these improvements are consistent, at least in part, with improvements in LVAD technology.
Durability and reliability of LVAD design are among the most significant features for extended use of MCS devices. Device replacements in this study were relatively infrequent despite a cumulative support duration of approximately 180 years. When device replacement was required, it was largely for device thrombosis and infection. No mechanical failures of the pumping mechanism were observed, although there was damage to wires in the percutaneous leads and 2 deaths from problems with grafts and connectors.
Recently, a safety advisory regarding the potential for wear and fatigue of the percutaneous lead to occur over time was released by Thoratec Corporation. The advisory provided an estimate of the probability of the need for pump replacement due to percutaneous lead damage to be 1.3% at 12 months, 6.5% at 24 months, and 11.4% at 36 months and also noted percutaneous lead care instructions for patients and caregivers as well as methods to diagnose clinically-relevant wear and fatigue (22). As patients with LVADs become more active and are supported for longer durations, design improvements are being made to improve the durability of the percutaneous lead, and modular designs are under consideration that will permit external exchange of the percutaneous lead in the event of a lead fracture without the need for pump exchange.
The absence of mechanical failures of the pumping mechanism is very encouraging in the prospect of long-term therapy for patients with advanced heart failure. Device exchange is a significant cause of morbidity and mortality (11,12). Data from the present study with a continuous-flow rotary pump demonstrate an important trend toward a low rate of device replacement required from all causes at 18 months. These data suggest that small durable devices can lead to a reduction in infectious and mechanical pump complications and support the concept and feasibility of extended pump support.
The use of continuous-flow rotary pumps has raised a number of concerns. Continuous-flow rotary pumps significantly reduce but do not eliminate the normal pressure amplitude associated with the native circulation that may influence long-term organ function. This initial concern has now been refuted by longer-term follow-up from this present study that has demonstrated preservation of organ function (23). Continuous-flow rotary pumps are currently designed to operate at fixed rotor speeds that do not automatically adjust to changes in left ventricular pre-load. Pre-load changes could limit the capacity of the pump output and patient exercise capacity or alternatively might result in excessive flow in low filling conditions associated with left ventricular collapse and ventricular arrhythmias. However, recent reports have suggested that the hydrodynamic characteristics of the continuous-flow rotary pump used in this clinical study allow flow to increase during exercise and appear to have no or insignificant limitations in patient exercise performance compared with pulsatile pumps (24).
Additionally, although the risk of pump thrombosis and thromboembolism (particularly stroke) is low, it has not been eliminated with the new pump design. Current management strategies for continuous-flow devices require long-term antithrombotic therapy with warfarin, thereby increasing the patient's risk of hemorrhagic events. The optimum level of anticoagulation to minimize both thromobembolic and hemorrhagic risks has not been determined from this present study because there was variability in the anticoagulation management among the participating centers. Although the smaller percutaneous lead in the continuous-flow device has appeared to reduce the rates of device-associated infection, infection remains a potential concern. Additional follow-up is needed to establish the durability and reliability of the pump mechanism over the longer term (>18 months).
The study was a nonrandomized comparison. Thus, direct comparisons of the current study of a continuous-flow rotary pump to medical therapy or to other studies of pulsatile- or continuous-flow pump designs cannot be performed. Second, the study was performed in a BTT population. There was a significant attrition of patients to transplantation (50%) by 1 year. Although this was the intent of the study, attrition of patients impacts the interpretation of data obtained from the patients remaining on device support as a result of the potential bias from giving priority to transplantation of patients experiencing the least benefit from the device.
A continuous-flow rotary pump LVAD with axial design provides safe, reliable, and effective hemodynamic support in patients awaiting transplantation with improved quality of life and functional capacity. Furthermore, LVAD therapy with continuous-flow rotary pumps with extended support is associated with a very low rate of device malfunction or infection requiring device exchange. Continuous-flow rotary pumps provide a superior alternative to pumps with a pulsatile design in patients awaiting transplantation.
A complete list of study investigators has been previously published (N Engl J Med 2007;357:885–96). Dr. Pagani has received research grant support from Thoratec Corporation, was site PI for the HeartMate LVAD trial, has received research grant support from Terumo Heart Corporation, and was national Co-PI for the DuraHeart LVAD trial. Dr. Miller has received research grant support and honoraria for talks at academic medical centers from Thoratec Corporation. Dr. Russell has been a consultant and received research grant support from Thoratec Corporation. Dr. Aaronson has been a consultant for Thoratec Corporation (reimbursed for travel expenses only). Dr. John has received research support from Thoratec Corporation and has served on the Scientific Advisory Board of Ventracor. Dr. Boyle has been a consultant for Thoratec Corporation. Dr. Conte has been a PI for Thoratec Corporation, ABiomed, and Heartware, and has been a trainer for Thoratec/Abiomed. Dr. Bogaev has been a consultant to Thoratec Corporation. Dr. Naka has been a speaker for Thoratec Corporation and a consultant for Cardiomems. Dr. Farrar is an employee and stock holder of Thoratec Corporation. Drs. Miller and Pagani contributed equally to this work.
- Abbreviations and Acronyms
- Food and Drug Administration
- left ventricular assist device
- mechanical circulatory support
- New York Heart Association
- Received December 1, 2008.
- Revision received March 3, 2009.
- Accepted March 10, 2009.
- American College of Cardiology Foundation
- El-Banayosy A.,
- Arusoglu L.,
- Kizner L.,
- et al.
- Dowling R.D.,
- Park S.J.,
- Pagani F.D.,
- et al.
- Goldstein D.J.,
- Zucker M.,
- Arroyo L.,
- et al.
- Esmore D.,
- Spratt P.,
- Larbalestier R.,
- et al.
- Göttel P.,
- Hetzer R.,
- Schmid C.,
- et al.
- Lietz K.,
- Long J.W.,
- Kfoury A.G.,
- et al.
- Matthews J.C.,
- Koelling T.M.,
- Pagani F.D.,
- Aaronson K.D.
- ↵United States Food and Drug Administration, MedWatch, Food and Drug Administration Safety Information and Adverse Event Reporting Program, 2008 Safety Alerts for Human Medical Products (Drugs, Biologics, Medical Devices, Special Nutrionals, and Cosmetics). Reported October 2008 http://www.fda.gov/medwAtch/safety/2008/safety08.htm#HeartMate. Accessed January 10, 2009.