| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;81:1365-1371
© 2006 The Society of Thoracic Surgeons
Department of Surgery, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania
Accepted for publication November 22, 2005.
* Address correspondence to Dr Acker, Division of Cardiothoracic Surgery, 6th Floor Silverstein, Hospital of the University of Pennsylvania, 34th and Spruce Street, Philadelphia, PA 19104-4283 (Email: michael.acker{at}uphs.upenn.edu).
Presented at the Fifty-second Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 10–12, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
METHODS: A retrospective review of the ventricular assist device (VAD) database at the Hospital of the University of Pennsylvania was instituted.
RESULTS: From April 1995 to February 2005, 49 patients received LVAD support for CS-AMI (group I). The majority of these patients suffered anterior wall myocardial infarctions. This group of patients was compared with a separate cohort of 61 patients with chronic ischemic cardiomyopathy who received LVAD support (group II). The VAD support successfully bridged 38 (74%) group I patients and 37 (61%) group II patients to heart transplantation. Of the 38 patients transplanted in group I, 33 (87%) were discharged from the hospital. In group II, 36 of the 37 patients transplanted (97%) survived to hospital discharge. The overall in-hospital mortality rates for the series were 33% for group I patients, and 41% for group II patients.
CONCLUSIONS: Left ventricular assist device support in patients with CS-AMI is a safe and effective therapy which should be incorporated into the standard treatment paradigm for appropriate patients presenting with this lethal disease.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Experience with the use of left ventricular assist devices (LVADs) for support of the circulation in patients with CS-AMI is limited. Most reports in the literature are anecdotal or consist of small subsets of patients drawn from larger LVAD series. The largest existing series comes from a multicenter trial, which reported a 24% mortality rate with the use of LVADs in 17 patients with CS-AMI [4]. More recently, Park and colleagues [5] reported an 85% rate of successful bridge to transplantation and a 29% mortality rate with the use of LVAD support in 7 patients with CS-AMI. Although the data are limited, the significantly lower mortality rates observed with LVAD support of the circulation compared with IABP and revascularization alone necessitate further investigation into the potential incorporation of LVADs into the standard treatment paradigm for CS-AMI.
One of the surgical dilemmas when implanting an LVAD into a patient with an acute anterior wall myocardial infarction is the safety of apical cannulation in the presence of acutely infarcted apical myocardium, which is typically necrotic and friable. Ventricular disruption and bleeding from the cannulation site are major concerns with lethal consequences. Although left atrial cannulation is an option, it is suboptimal as it affords inadequate left ventricular decompression and limits LVAD inflow. Furthermore, left atrial cannulation and CS-AMI have both been shown to be independent risk factors for the development of left ventricular thrombus and stroke [6].
The purpose of this report is twofold. First, we present our experience using LVADs in the treatment of patients with CS-AMI. Second, in order to prove the safety and efficacy of left apical cannulation into acutely infarcted myocardium, we compared outcomes and complications of this group of patients with a separate cohort of patients with chronic ischemic cardiomyopathy who received LVAD support during the same time period at our institution.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Patient demographics are listed in Table 1. The mean time from AMI to LVAD implantation for group I was 6.4 ± 7.3 days. The length of hospitalization prior to LVAD implantation was significantly less in group I patients (3 ± 4 days vs 12 ± 21 days, p < 0.001). At the time of implantation, 88% of these patients were already receiving mechanical circulatory support with an IABP. The two groups were closely matched in age, gender, and body habitus (as assessed by body surface area). The major risk factor profiles for coronary artery disease (diabetes mellitus, hypertension, smoking) were also comparable in both groups. Twenty-one (43%) patients in group I and 31 (51%) patients in group II had undergone prior CABG. Both groups had evidence of mild renal dysfunction preoperatively (mean creatinine 1.6 ± 0.8 in group I vs 1.8 ± 1.0 in group II, p = 0.27).
|
Right ventricular (RV) failure was defined as LVAD output less than 1.8 L · min · m2 with elevated central venous pressure (> 20 mm Hg) that did not respond to inotropic agents or pulmonary vasodilators (nitric oxide or inhaled prostacyclin). Right ventricular assist devices (RVADs) were implanted for RV failure or RV arrhythmias refractory to medical therapy, and were inserted through similar techniques as described above for the LVAD. Inflow cannulas were placed into either the right atrium or right ventricle, and outflow cannulas were anastomosed to the pulmonary artery.
Patients in this series were supported by a variety of short-term (BioMedicus [Medtronics, Inc, Eden Prairie, MN]; Abiomed BVS 5000 [Abiomed Cardiovascular, Danvers, MA]), and long-term (Thoratec, HeartMate IP [implantable pneumatic], HeartMate VE [vented electric; Thoratec Laboratories Corp, Pleasanton, CA]) devices. Although we never primarily implanted Biomedicus pumps, patients on Biomedicus support were transferred from outside hospitals to our institution for transplant evaluation. The VAD selection was based on both patient characteristics and device availability. Patients with a body surface area (BSA) less than 1.8 m2 usually received an Abiomed BVS (biventricular support) 5000 (only one patient in group I) or a Thoratec LVAD. From 1995 to 1998, patients with a BSA greater than 1.8 m2 received a HeartMate IP LVAD because of its higher flow capacity. When the HeartMate VE LVAD became available in 1998, we began implanting this device for all patients with a BSA greater than 1.8 m2. This device is a portable system that allows for outpatient bridge to transplant, and was the most common type of LVAD implanted in our series in either group. Patients with biventricular failure at the time of LVAD insertion received the Thoratec biventricular assist device system (BiVAD). Those patients who developed refractory RV failure after HeartMate LVAD implantation were supported by a hybrid combination of devices, most commonly a HeartMate VE LVAD and a Thoratec RVAD (see Table 2).
|
Statistical Analysis
All data are reported as means ± standard deviation. Continuous data were compared between the two groups by an independent sample Student t test. Nominal data were compared between the two groups by a 
2 or Fisher's exact test for a 2 x 2 table. Statistical significance was defined as a p value less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Both cohorts had similar cardiopulmonary bypass and cross-clamp times (see Table 3). The rate of reoperation was approximately 50% in both groups, and most commonly due to bleeding. Other reasons for return to the operating room included tamponade, cannula exchange, sternal closure, and VAD exchange.
|
|
Of the 38 patients transplanted in group I, 33 (87%) were discharged from the hospital. In group II, 36 of the 37 patients transplanted (97%) survived to hospital discharge. All posttransplant deaths occurred within 30 days in both groups. Once transplanted, patients in group I tended to have a shorter length of stay than group II patients (23 ± 15, group I vs 35 ± 33 days, group II, p = 0.105), but there was no difference in the overall length of stay between the two groups. The most common cause of death in either group was multisystem organ failure. There was a single case of successful VAD explantation without transplantation in a group II patient. The overall in-hospital mortality rates for the series were 33% for group I patients and 41% for group II patients (see Table 5).
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Currently, the standard treatment of patients with CS-AMI involves mechanical support of the circulation with an IABP [7]. The use of IABP has had a significant impact on survival, but in-hospital mortality rates remain over 50% [8]. The efficacy of LVAD circulatory support as a bridge to transplant in the setting of postcardiotomy cardiogenic shock has been well-established [4, 9]. In recent years, reports documenting the use of LVAD support in the setting of post-MI cardiogenic shock have produced in-hospital mortality rates of 24% to 44% [4, 5, 10, 11]. Although these series are small the reduction in mortality compared with conventional therapy is significant, which raises the possibility for a definitive role of LVADs in the treatment of CS-AMI.
The current study describes a decade of experience at the Hospital of the University of Pennsylvania with two separate groups of patients suffering from ischemic heart disease. In order to prove the safety and efficacy of LVAD use in patients with CS-AMI (group I), we compared outcomes with a separate cohort of patients with chronic ischemic cardiomyopathy. The chronic patients (group II) represent a population of patients with end-stage heart failure for which LVAD insertion has proven to be highly effective as a bridge to transplant [4, 9, 12]. Although the two groups represent different presentations of heart failure (acute versus chronic), their demographic profiles are similar. The comparable rates of hypertension, diabetes mellitus, smoking, and prior CABG in each group highlight the common ischemic etiology of their heart disease. Besides a slight difference in age, the only significant difference in preoperative risk factors was the higher percentage of group I patients who had an IABP placed prior to LVAD insertion (88% vs 56%, p < 0.001).
A major concern when implanting LVADs into patients with acute anterior wall myocardial infarctions is the safety of LV cannulation. It is widely believed that a significant risk of left ventricular disruption and bleeding exists when cannulating the acutely infarcted, friable apex. Although left atrial cannulation is an alternative option, left ventricular cannulation is preferable because it achieves superior decompression of the ventricle, provides greater LVAD inflow, and reduces the risk of LV thrombus formation and stroke [6, 13]. If the ventricle is not maximally decompressed, blood pools in the ventricle and becomes stagnant in a segment of infarcted myocardium. This becomes a nidus for thrombus formation and subsequent embolic phenomena, resulting in cerebrovascular and other end organ injuries.
Effective LV decompression also achieves a reduction in LV pressure, which will lower the risk of ventricular tearing and bleeding at the cannulation site [10, 14]. Previous authors have maintained that LVADs can be safely implanted into acutely infarcted, friable myocardium by modifying their surgical technique. This involves placing cannulation sutures through the full-thickness of the infarcted ventricular myocardium and reinforcing their suture line with pericardium or Teflon felt [5, 10, 14]. Our cannulation technique in CS-AMI (group I) patients, the majority of whom had anterior wall infarcts, consisted of securing the cannula with interrupted, pledgeted, horizontal mattress sutures through the full-thickness of the infarcted myocardium. If significant bleeding was observed, additional sutures and/or BioGlue (BioGlue Surgical Adhesive; CryoLife, Inc, Kennesaw, GA) were applied to the cannulation site. This strategy appears effective as there were no cases of ventricular disruption and no reoperations for bleeding from the apical cannulation site.
Postoperative complication rates after VAD placement were compared between the groups. Both groups had comparable rates of reoperation, RV failure, gastrointestinal bleeding, and stroke. Although there were no differences between the two groups in preoperative creatinine levels, patients in group I had a significantly higher rate of requiring hemodialysis. This is most likely due to severe renal hypoperfusion attendant with cardiogenic shock prior to circulatory stabilization with the LVAD. The VAD-related infections were a common problem in both groups (51% in group I vs 33% in group II, p = 0.87). The incidence of driveline infections and endocarditis were equivalent in both groups; however, sepsis tended to occur more commonly in group I patients (37% vs 20% in group II, p = 0.06). This may be due to a relative state of immunosuppression in these patients secondary to the profound systemic inflammatory response syndrome associated with post-MI cardiogenic shock [15, 16].
Included in this report is a subset of patients who received bridge to bridge therapy prior to cardiac transplantation. In a previous report from our group we described the use of ECMO as a temporary bridge to VAD support for patients with intractable cardiogenic shock [17]. This bridge to bridge method of circulatory support, utilizing both ECMO and short-term VAD support as an initial bridge to a long-term VAD, was effective in both acute (group I) and chronic (group II) heart failure patients. Of the 26 patients who received bridge to bridge therapy, 14 (54%) patients were transplanted and discharged from the hospital.
The high use of BiVADs in both groups (39% in group I vs 36% in group II, p = 1.00) in this series reflects our belief in the importance of RV support in the overall outcomes of these critically ill patients (Table 2). At the time of LVAD insertion in both acute and chronic heart failure patients, right ventricular function is always evaluated. In addition to RV failure and intractable arrhythmias, we believe that the presence of shock with multisystem organ failure is an indication for biventricular support. We feel that in these patients BiVADs provide better flows with reduced ionotropic requirements. Furthermore, in our experience the reduction in central venous pressures afforded by BiVAD support (vs LVAD alone) improves overall resuscitation in multiorgan system failure.
With the recent development of percutaneous ventricular assist devices (pVADs), we envision an expanding role for VAD support, which may ultimately replace IABP in the treatment of CS-AMI. The pVAD can provide circulatory support similar to an IABP and can decompress the left ventricle, thus limiting infarct size and subsequent ventricular remodeling. Ideally this would enable patients in CS-AMI who are treated with a pVAD to be bridged to recovery. Initial reports of the successful implementation of pVADs in patients with CS-AMI are beginning to emerge [18, 19]. These short-term devices will likely result in morbidity and mortality comparable with conventional LVADs, and add flexibility to the treatment algorithm as they will be used as a bridge to recovery, a bridge to bridge, or as a bridge to transplant. In summary, this report represents a decade of experience at a single institution using LVAD circulatory support to treat patients in CS-AMI. It is the largest reported series to date and our in-hospital mortality rate of 33% is significantly lower than the 60% to 70% mortality rate achieved by conventional treatment algorithms [3, 20, 21]. By comparing outcomes and complications with a cohort of patients for which the benefits of LVAD support have been firmly established, we have validated the safety of LVAD inflow cannulation into the acutely infarcted, friable apex and demonstrated that, with proper technique, there is no additional risk of bleeding or ventricular disruption. These data advocate the incorporation of LVADs into the standard treatment paradigm for patients in CS-AMI.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
I have two questions. The first is a general question regarding timing. The frequent scenario is that these patients present with MI and shock, have an intra-aortic balloon pump placed and then undergo an attempted revascularization, either PCI or CABG. After revascularization there is usually some delay to assess whether revascularization affected improvement in myocardial recovery. By the time an LVAD is considered, the patient may be in advanced shock. Were there any patients in the series in whom LVAD was used primarily, even before consideration of revascularization? Conversely, in patients who have progressed into shock, were there patients in the series who were supported with temporary less invasive mechanical measures such as percutaneous ECMO or percutaneous Tandem Heart LVAD (to make the patients a better candidate for a larger LVAD procedure)?
My second question relates to the incidence of RVAD support. Thirty or 40% of the patients in both groups required an RVAD. For patients who present with an acute myocardial infarction due to an LAD occlusion, usually the right ventricle is spared. The right coronary and the blood supply to the right ventricle may be completely normal. Why was the incidence of RVAD requirement so high? Was the RVAD support temporary or permanent?
Thank you, and again, this was an excellent paper on an important topic.
DR LESHNOWER: Thank you. Those are excellent questions and I will try to answer them in reverse order here.
We made the decision on the use of RVADs usually at the time of operation, and as I stated, we were pretty liberal about using RVADs, which is shown with our high incidence. If these patients required significant inotropes for RV failure, if their CVPs were greater than 18, and also an intraop decision was made, we went for an RVAD placement pretty aggressively. That just reflects our belief with these patients who have other systemic signs of advanced multisystem organ failure.
To go back to the other questions, we have not used any of the percutaneous LVADs yet in these patients, and none of these patients were referred to us prior to any revascularization attempts. Generally, the way it works is that when patients at Penn come in in cardiogenic shock, they are evaluated by the heart failure service. The cardiologists and a surgeon are both involved and are aware early, and decisions are made keeping both parties up to date simultaneously.
DR HOOSHANG BOLOOKI (Miami, FL): Let me first congratulate you and your colleagues for a good study. It has involved a great deal of work in a group of very sick patients. When we initially started this type of therapy in acute phase of myocardial infarction, many years ago, two problems emerged: first, the left ventricle was small, and infarcted area was large and necrotic making insertion of the LVAD cannula into the apex, through an infarcted myocardium unsustainable. Second, we just did not have enough left ventricular space and volume or sufficient preload to maintain LVAD output. My question is did you use specific selection criteria such as, number of days after infarction or a minimum LV diastolic volume before accepting the post infarction patient with shock for LVAD insertion? I noticed that not many of your patients were operated within the first few hours or days after infarction and shock. Some of the patients were up to 30 days post infarction, which means they had healed infarcts that makes it easier to work with than the patients we were dealing with.
Congratulations again, and I hope this work will be a milestone in dealing with patients in cardiogenic shock.
DR LESHNOWER: Thank you for the question. I think the majority of the acute MI group, the LVADs were placed probably within the first 72 hours, although we do have some that go out further. And as far as evaluating which patients were and were not candidates, all of these patients had to be eligible for transplant or they were immediately excluded from LVAD candidacy.
DR MICHAEL MACK (Dallas, TX): As a follow-up to those two questions, do you have the Tandem Heart available?
DR LESHNOWER: We do.
DR MACK: Do you find now that you have a percutaneous LVAD available that you tend to use that more as a bridge to a bridge and let the infarct mature some and the ventricle dilate some before you go ahead and place the LVAD, or has it not changed your treatment paradigm?
DR LESHNOWER: I would say it has not changed the treatment paradigm yet. I know we have placed one or two, but we have not placed it for this specific scenario.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. M. Cheng, C. A. den Uil, S. E. Hoeks, M. van der Ent, L. S.D. Jewbali, R. T. van Domburg, and P. W. Serruys Percutaneous left ventricular assist devices vs. intra-aortic balloon pump counterpulsation for treatment of cardiogenic shock: a meta-analysis of controlled trials Eur. Heart J., September 1, 2009; 30(17): 2102 - 2108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Shamloo, J. L. Taylor, T. Yusufali, and N. D' Attellis Can a broken heart be fixed? J Intensive Care Med, September 1, 2009; 24(5): 338 - 343. [PDF]
|
|
|
|
 |
|
 C. W. Hamm, H. Möllmann, J.-P. Bassand, and F. van de Werf CHAPTER 16 Acute Coronary Syndromes ESC Textbook of Cardiovascular Medicine, January 1, 2009; 2(1): med-9780199566990-chapter - med-9780199566990-chapter. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Authors/Task Force Members, F. Van de Werf, J. Bax, A. Betriu, C. Blomstrom-Lundqvist, F. Crea, V. Falk, G. Filippatos, K. Fox, K. Huber, et al. Management of acute myocardial infarction in patients presenting with persistent ST-segment elevation: The Task Force on the management of ST-segment elevation acute myocardial infarction of the European Society of Cardiology: Eur. Heart J., December 1, 2008; 29(23): 2909 - 2945. [Full Text] [PDF]
|
|
|
|
 |
|
 H. R. Reynolds and J. S. Hochman Cardiogenic Shock: Current Concepts and Improving Outcomes Circulation, February 5, 2008; 117(5): 686 - 697. [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]
|
|
|
|
 |
|
 L. E. Samuels, E. C. Holmes, K. Hagan, R. S. Boova, S. Janzer, and D. Kocovic Cardiogenic Shock: Collaboration Between Cardiac Surgery and Cardiology Subspecialties to Bridge to Recovery Ann. Thorac. Surg., May 1, 2007; 83(5): 1863 - 1864. [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 |
