| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Click here to read this article as a CME activity
a Section of Cardiac Surgery, University of Michigan, Ann Arbor, Michigan
b Division of Cardiovascular Medicine, University of Michigan, Ann Arbor, Michigan
Accepted for publication April 1, 2009.
* Address correspondence to Dr Haft, Cardiovascular Center 5144, 1500 E Medical Center Dr, Ann Arbor, MI 48109 (Email: haft{at}umich.edu).
Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
| ADULT CARDIAC SURGERY:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: We performed a single-institution retrospective review of 59 consecutive patients transferred on temporary, extracorporeal MCS from 1997 to 2008. Demographics, medical history, laboratory data, and clinical status were obtained, with survival determined from the medical record and the Social Security Death Index. Univariable and multivariable analysis were performed and survival estimates were determined using the Kaplan-Meier method.
Results: Median age was 49.6 years (range, 14 to 77 years). Forty-five patients (76%) were supported for postcardiotomy failure, and 34 (58%) required biventricular support. Twenty-five (42%) survived to hospital discharge, 11 after cardiac recovery (44%), 9 with long-term implantable MCS devices (39%), and 5 after heart transplantation (22%). Eight patients discharged with implantable MCS devices underwent heart transplantation and 1 remains alive on long-term implantable MCS support. Survival was 42% ± 6% at 1 year and 38% ± 6% at 5 years. Age and renal function were independent predictors of death.
Conclusions: Nearly half of all patients transferred on temporary extracorporeal MCS survive to discharge. Most of the long-term survivors received a heart transplant. Age and renal function were independent predictors of death, suggesting that survival is maximized by considering eligibility for cardiac transplantation.
| Dr Pagani discloses that he has a financial relationship with Thoratec Corp and Terumo Corp; Dr Dyke with Thoratec Corp.
|
Cardiogenic shock can be associated with conditions such as acute myocardial infarction, viral myocarditis, postpartum cardiomyopathy, and postcardiotomy heart failure. Mortality is extremely high, particularly when refractory to intraaortic balloon counterpulsation and inotropic support [1, 2]. Mechanical circulatory support using blood pumps to support or replace cardiac function have been increasingly used as a bridge to heart transplantation or as destination therapy for selected patients unsuitable for transplant [3]. However, implantable ventricular assist devices (VADs) require lengthy operative procedures for insertion and use expensive technology currently available in only specialized higher-volume centers. Implantation of these devices to the geographically broad and poorly selected population of patients in cardiogenic shock is impractical.
Temporary extracorporeal assist devices, with lower costs and easier implantation procedures, are frequently used for patients in cardiogenic shock, particularly in hospitals that do not routinely insert implantable VADs or perform cardiac transplantation [4]. Many of these patients are transferred to regional transplant centers. We sought to determine the outcomes of these patients and to specifically identify prognostic variables for both short-term and long-term survival.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Transportation vehicles included ground ambulance, helicopter, and fixed-wing aircraft. University of Michigan flight nurses and a cardiopulmonary perfusionist cared for the patients during transport. Management protocols of patients on temporary extracorporeal support have evolved during the time of this study and are too variable for presentation.
Patients were considered for an implantable VAD if they were felt to have unrecoverable heart failure and were a suitable candidate for heart transplantation. Patients with signs of ventricular recovery were evaluated for explantation of the temporary support device. Flows were typically reduced to 1 L/min during echocardiographic imaging and invasive hemodynamic monitoring with a pulmonary artery catheter. If cardiac output, intracardiac filling pressures, and ventricular function appeared sustainable on minimal inotropic infusions, patients were explanted. Some patients were explanted without weaning trials because of bleeding or neurologic complications precluding further anticoagulation and ongoing extracorporeal support.
Data on demographics, medical history, etiology of heart failure, hemodynamic status, and laboratory values were obtained from the medical record. Survival status was obtained from the medical record or from the Social Security Death Index [5] accessed on January 13, 2009. Waiver of informed consent was granted by the Institutional Review Board, who approved collection, analysis. and reporting of the data.
Data analysis was performed using SAS 9.1 software (SAS Institute, Cary, NC). Categoric data were compared with Fisher exact tests for 2 x 2 tables or otherwise with the Pearson 
2 test. Continuous data were evaluated for normality and between-group comparisons were performed using the t test for normal data or the Mann-Whitney test for nonnormal data. Data are expressed as mean ± standard error of the mean or median (25th, 75th percentile) as appropriate.
For the purposes of further comparison, continuous data were dichotomized at the sample median. Odds ratios with 95% confidence intervals for univariable predictors (p < 0.05) of in-hospital death were generated from logistic regression analysis. Multivariable logistic regression was then used to identify independent (p < 0.05) predictors of in-hospital death. All variables with a value of p 
0.15 on univariable analysis were entered into multivariable regression in a stepwise fashion.
The likelihood ratio test (nested models) and the Akaike's information criterion (for nonnested models) were used during variable selection and model development to determine optimal fit. Kaplan-Meier survival curves were then created to evaluate in-hospital and long-term survival. Long-term survival was defined as free from death, regardless of intervening implantable mechanical support or cardiac transplant. Because of high early postoperative mortality, Wilcoxon p values are provided along with the log-rank test for survival curve comparison.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
|
|
|
|
0.15 for death were entered into multivariable logistic regression analysis. Independent predictors of in-hospital death included advanced age, affording an odds ratio for death of 7.9 (1.9, 33), and renal dysfunction, with an odds ratio for death of 6.2 (2.0, 20) per mg/dL of creatinine.
Long-Term Survival After Extracorporeal Mechanical Support
Figure 4A shows the Kaplan-Meier survival curve for the entire cohort. Relatively few late deaths are noted, and most patients discharged alive survive for at least 5 years. When patients are dichotomized by age and renal function, important and significant differences are noted in survival (Figs 4B and C), with the greatest effect on survival stratification occurring within 1 month.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
When cardiogenic shock occurs in patients in medical centers without implantable VAD programs, temporary extracorporeal devices with their lower costs and easier implant procedures can be initiated. Once hemodynamic variables are stabilized, these patients can be transferred to regional centers with expanded resources. Transportation with complex extracorporeal support devices across long distances can be hazardous to both the patient and caregivers and carries extensive costs. Thus, identification of variables that are highly predictive of death may reduce unnecessary transfers of high-risk individuals, improving resource utilization. Patients identified at low risk for death should be expeditiously transferred to a transplant facility, potentially improving outcomes.
Several regional transplant centers have previously described their experience with this patient population. Morris and colleagues [8] reported 28 patients transferred to the University of Pennsylvania on extracorporeal assist devices. Compared with our cohort, more of their patients required support after elective cardiac operations, and survival was only 32%. This may reflect a slightly older population (54 years), with a higher prevalence of diabetes (50%) and low preoperative ejection fractions. All survivors underwent transplantation, either because of center philosophy or a highly selected referral of unrecoverable hearts. Statistical analysis of predictors was not performed, and follow-up was limited to 1 year.
Gonzalez-Stawinski and colleagues [9] reported the Cleveland Clinic experience of 39 patients. Similar to our cohort, more than half of the survivors required implantable long-term assist devices or transplantation. Variables associated with death included age, female gender, history of coronary disease and myocardial infarctions, diabetes, and complex surgical procedures other than coronary artery bypass grafting. Multivariable analysis could not be performed because of inadequate sample size, and long-term survival was not described.
Hernandez and colleagues [10] reported the survival of patients requiring VAD implantation after additional cardiac operations, using the Society of Thoracic Surgeons National Cardiac Database. Their survival was 54%, which was significantly better than our experience; however, their cohort included important differences. These patients were not necessarily transferred to tertiary care centers after implant procedures, both intracorporeal and extracorporeal devices were included, and patients were more likely to receive VADs after an elective operation. In addition, patients who received heart transplants on the same hospital admission were excluded, selecting out patients with irreversible heart failure. The paucity of patients in cardiogenic shock preoperatively likely accounts for the observed difference in survival compared with our report.
In this single-center analysis of 59 individuals, more than 50 demographic, clinical, and laboratory variables were evaluated to identify independent predictors of in-hospital death in patients transferred from referring centers with extracorporeal mechanical support. Although several univariable predictors were identified, only age and renal function remained independently predictive of death on multivariable analysis. At 1 month, survival for patients of advanced age or with poor renal function, or both, was less than half that of patients aged younger than 50 years or with a serum creatinine level below 1.5 mg/dL. These risk factors would be expected, suggesting worse outcomes with either patients with more advanced acquired cardiovascular disease or those with severe organ damage from late initiation of mechanical support.
We also describe excellent long-term survival, with few late deaths in patients discharged alive. Although our approach has always been to accept any referral for cardiogenic shock, we may defer transfer of patients whose age and renal function predict in-hospital death, particularly when distance or weather influence safety and cost of transportation. We also demonstrated that most patients could not be weaned from temporary extracorporeal support and required an implantable VAD or transplant for survival. Younger patients with better renal function had improved survival, which suggests that their candidacy for heart transplantation may have influenced the decision process and contributed to our findings. Certainly, some patients died as a result of unrecoverable heart failure but were felt to be unsuitable for transplant or implantable circulatory support. Therefore, patients requiring temporary extracorporeal mechanical circulatory support devices who might be candidates for heart transplantation should ideally be referred to the regional center with the experience and resources to offer the multitude of options.
Within the last several years, newer technology aimed at mechanically supporting cardiogenic shock has become available [11, 12]. Several of these devices, including TandemHeart (CardiacAssist, Pittsburgh, PA) and Impella (ABIOMED Inc) are designed for percutaneous support and are implemented outside of the operating room, broadening the application of mechanical support for cardiogenic shock. Because most of our survivors required transplant or implantable assist devices, it is important to remember that temporary mechanical support is a bridge, either to ventricular recovery or long-term cardiac replacement. When ventricular recovery is less certain, input from regional transplant centers is essential in properly selecting patients for this therapy.
Several important limitations of this study are noteworthy, including its retrospective nature and the biases that are inherent in this type of report. In addition, this is a single-institution experience and may not translate across centers with different populations, resources, and experience. Although we were able to perform multivariable analysis, the small sample size likely underestimates the importance of additional prognostic variables.
Importantly, this review is limited to patients transferred to our facility from our referral base. Additional extracorporeal assist devices are undoubtedly implanted from these hospitals, but for a variety of reasons, the patients are not transferred. Most likely, variables from these patients would populate the extremes in our series, in which transfer was less likely to positively influence the outcome. Many of these patients are either less ill, resulting in early ventricular recovery leading to successful discharge, or severely moribund, prompting referring physicians to defer transfer of an apparently unsalvageable situation. Without knowledge of these patients, our study represents only a portion of the temporary circulatory support population, potentially biasing our findings. The outcomes of this analysis are similar to other reports, however, and the independent predictors are clinically reasonable, suggesting a valid sample of patients was used for risk factor analysis.
In conclusion, approximately 40% of patients transferred to a regional transplant center on temporary extracorporeal assist devices are discharged from the hospital alive, with excellent long-term survival. Older age and renal dysfunction are independent predictors of death. More than half of surviving patients do not have adequate ventricular recovery, necessitating implantation of long-term cardiac assist devices or heart transplantation. To improve outcomes, hospitals implanting temporary extracorporeal assist devices should communicate with regional transplant centers about the appropriateness of interhospital transfer once hemodynamic stability is achieved.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
DR HAFT: Thank you for your comments, Dr Verrier. This population of patients is quite different from the majority of our traditional LVAD [left ventricular assist device] patients. Most of our HeartMate implants are in patients with longstanding heart failure and have been medically optimized prior to surgery. The patients I presented today are in cardiogenic shock, most of whom are postcardiotomy after acute myocardial infarctions. A strategy of implanting long term VADs in this population would be impractical, particularly given the number of patients that die of multisystem organ failure or neurologic complications. Since many of these patients represent failure to wean from cardiopulmonary bypass at referring hospitals, an alternative strategy of using balloon pumps and high-dose inotropes and vasoconstrictors in order to allow transport to our facility would likely result in worse outcomes.
Certainly in the patients that are not postcardiotomy, it would be ideal if those patients could be transferred to the tertiary or quaternary care center earlier rather than later, but if you happen to be in a non-VAD or transplant hospital several hundred miles away from the tertiary care center, restoring hemodynamic stability and end-organ perfusion would be preferable prior to transport in the sickest of patients.
As far as a health policy strategy, because of the high mortality in this cohort, temporary circulatory support must be cost effective and be easy to insert by referring hospitals with relatively infrequent experience in VAD implantation. Patients can then be safely transferred to transplant or VAD centers of high volume.
Y. JOSEPH WOO (Philadelphia, PA): I have a general technical question about some of the devices. I noticed you excluded extracorporeal membrane oxygenation [ECMO], but there is certainly a group of patients who are in cardiogenic shock without respiratory issues who are transferred on ECMO. Were the majority of your patients on an ABIOMED system and were any on centrifugal pumps? Is there a temporal change in the type of devices?
DR HAFT: There has been a gradual transition in the type of devices. In our experience, the majority represented the Abiomed BVS system, some were on Biomedicus pumps [Medtronic Inc, Minneapolis, MN], a few more recently on the AB5000 system [ABIOMED Inc, Danvers, MA], and on the CentriMag [Thoratec, Pleasanton, CA] pump. So there has been an evolution in the technology.
As far as ECMO is concerned, there certainly were an additional 10 or 15 patients that we accepted in transfer on ECMO support; however, that included a population of circulatory collapse with septic shock and respiratory failure that really was distinct from this population. Although our cohort is a mixed bag with ischemics and nonischemics, it is a relatively homogeneous group with cardiogenic shock. We felt it would complicate the data to include those on ECMO.
DR LAWRENCE PATZELT (Grand Rapids, MI): Some of the patients in this presentation today are my patients that we have sent to the University of Michigan, and I want to say thank you for the support that you have given our program over the years. By definition, we are a community program, but we are not very small.
I have one comment that what is beneficial to us is the dialogue we have with the University of Michigan. I know Dr Pagani the best of the people over there who we are dealing with. And when we are in trouble, if we can make a phone call and talk things over with him, that is beneficial from both sides, try and get the patient more stable, if that is what is indicated, and maintain a dialogue about that specific patient before we transfer it to the University of Michigan. That is one point.
And then the other point is, in your own program at the University of Michigan when you have a failure to wean from the heart-lung machine, on inotropes and a balloon pump, is there an age cutoff that you would not use additional support? So if you have an 80-year-old or 85-year-old woman with status post aortic valve and you can't wean her, would you say the device is too expensive and at this age we are just not going to be able to use it in this individual, or how do you make those decisions?
DR HAFT: Thank you for your comments, Dr Patzelt. As far as addressing your first point, open lines of communication I think are absolutely essential, and certainly we have established an approach that leaves as few barriers between the referring surgeons and us as possible. This strategy may allow institution of circulatory support and transfer to our facility as expeditiously as possible.
You asked who is a candidate, and who should be excluded from mechanical circulatory support. I try and remember that temporary circulatory support for shock is a bridge, either to cardiac recovery or to implantable VAD or transplant. If you think there is a reasonable chance of cardiac recovery, then age or comorbidities should not necessarily be contraindications. I think the best example is after an acute myocardial infarction [MI] with shock. Although the infarct may be regional, the inflammatory response of that MI may cause severe global dysfunction. Mechanically supporting those patients for several days may allow significant myocardial recovery and weaning from circulatory support.
However, there are many scenarios in which the heart is unlikely to recover, such as widespread myocardial infarction from left main thrombosis, diffuse multivessel coronary occlusions, or advanced long standing heart failure with decompensation and shock. So in that scenario, the patient is more likely a bridge to an implantable VAD. If there are obvious exclusion criteria for an implantable VAD or transplant, such as advanced cirrhosis of the liver, emphysema with oxygen dependence, chronic dialysis, or significant psychiatric illness, than the patient should also be excluded from temporary circulatory support.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
