Secondary Logo

Journal Logo

Adult Circulatory Support

Clinical Experience and Patient Outcomes Associated With the TandemHeart Percutaneous Transseptal Assist Device Among a Heterogeneous Patient Population

Tempelhof, Michael W.*; Klein, Liviu*; Cotts, William G.*; Benzuly, Keith H.*; Davidson, Charles J.*; Meyers, Sheridan N.*; McCarthy, Patrick M.; Malaisrie, Chris S.; McGee, Edwin C.; Beohar, Nirat*

Author Information
doi: 10.1097/MAT.0b013e318218679e
  • Free


Advances in medical therapy for acute myocardial infarction, coronary artery disease (CAD), hypertension, and diabetes, in conjunction with the aging population, have led to an increased prevalence of heart failure (HF) in the United States.1–3 Approximately 6 million Americans have HF, of which an estimated 100,000 have advanced, stage D disease.4 Despite improvements in therapy, the 1-year mortality rate for patients hospitalized with an acute HF episode is as high as 30%.4,5 Patients with advanced HF have an estimated 6-month mortality >60%.6 Although medical and resynchronization therapies may improve HF symptoms, the prognosis and survival of patients with acute cardiogenic shock (CS) or refractory advanced HF is dismal. As such, these patients often require mechanical circulatory support (MCS) or cardiac transplantation.

Patients with severe CAD may present with CS or persevered low systolic ejection fraction. Such high-risk patients have minimal reserve with which to tolerate complex percutaneous coronary interventions (PCI) and often have significant comorbid conditions that preclude cardiac surgery. Periprocedural hemodynamic support through the use of a percutaneous ventricular assist device (PVAD) may allow for the treatment of complex CAD patients for whom there are no alternative therapies.

The technology and practicality of MCS devices for advanced HF patients have improved dramatically since their introduction 25 years ago. Before the development of extracorporeal temporary assist devices, temporary support was through the use of extracorporeal membrane oxygenation (ECMO). Use of ECMO remains a valid therapeutic, bridge-to-bridge option; however, it often requires invasive surgical intervention and is associated with a high mortality risk. Traditionally, intra-aortic balloon pumps (IABPs) were used for acutely decompensated or refractory HF patients. However, IABPs provide a limited amount of hemodynamic support, have a finite duration of use, and limit the mobility of patients. Despite these limitations, there is evidence to suggest the use of IABPs as a short-term bridge to more permanent hemodynamic support among high-surgical-risk patients with CS.7

Long-term, implantable ventricular assist devices (VADs) provide much more robust hemodynamic support for longer durations in comparison with IABPs. Their use as bridge-to-recovery, destination therapy, or as bridge-to-heart transplantation for patients with advanced HF refractory to conventional therapy has been well documented.8,9 Over the past 5 years, percutaneous VADs, such as TandemHeart (Cardiac Assist Inc., Pittsburgh, PA) or Impella (ABIOMED, Inc., Danvers, MA), have been used in hospitalized patients with advanced HF as bridge-to-recovery or to a more permanent implantable VAD (bridge-to-bridge). These devices also provide excellent hemodynamic support (up to 7 L/min) but limit patient mobility.

Impella is a catheter-based, impeller-driven, axial flow pump that crosses the aortic valve and aspirates blood directly from the left ventricle and ejects it directly into the ascending aorta. The Impella LP 2.5 device is implanted through a femoral percutaneous approach and can deliver a limited output of up to 2.5 L/min. The larger Impella LP 5.0 device provides higher cardiac support (5.0 L/min), but use of the 5.0 system is limited as implantation requires femoral surgery and is associated with higher rates of bleeding and access site complications compared with the 2.5-L system.10

TandemHeart is an extracorporeal continuous-flow centrifugal assist device that is connected to an inlet catheter that crosses the atrial septum in the left atrium where oxygenated blood is withdrawn and returned to an outlet cannula into the femoral artery. The device can be placed percutaneously through the femoral vein and provide immediate hemodynamic support with cardiac outputs of up to 5 L/min. TandemHeart has been demonstrated to provide temporary hemodynamic support across a wide range of clinical indications, including during high-risk PCI or during bridge-to-recovery, bridge-to-bridge, or bridge-to-transplantation for acute and advanced HF patients.11 Despite its commercial availability since 2004, there is a paucity of literature describing the clinical outcomes and safety associated with the use of TandemHeart system. Previous reports on the use of TandemHeart are mostly case reports, and the few trials are limited by analyzing only particular subsets of patients. We present the clinical outcomes and safety associated with the use of TandemHeart in a consecutive series of patients requiring PVAD support, representing the largest published database with the use of this device among patients with various indications for TandemHeart implantation.


From March 2007 to March 2010, 25 consecutive patients with hemodynamic compromise secondary to acutely decompensated HF, refractory-advanced HF, following postpericardiotomy surgery (coronary artery bypass graft surgery, aortic root replacement, or aortic root replacement with aortic valve replacement [AVR]), or high-risk percutaneous intervention underwent successful implantation of TandemHeart at our institution. Patient data were retrospectively analyzed by medical chart abstraction. Patients with CS despite inotropic therapy and/or IABP support met the criteria for TandemHeart implantation if their cardiac index was <2.5 L/min/m2 and there was an evidence of hepatic and/or renal failure. Pertinent data during the index hospitalization and follow-up visits were reviewed and entered into the registry for each patient. This study was approved by our Institutional Review Board committee, and because of the retrospective collection of data, informed consent was not obtained from the subjects.

Clinical characteristics, hemodynamic variables, laboratory and echocardiographic data, use of inotropic agents, and/or mechanical support before and after TandemHeart implantation were collected and analyzed (Table 1). All postimplantation hemodynamic and laboratory data were measured 24–36 hours after device implantation. All postimplantation echocardiograms were completed 24–48 hours after device implantation while the PVAD was temporarily on hold or immediately after device explantation. Complications incurred up to 30 days after device implantation were recorded. Patient outcomes (survival, death, transplantation status, and implanted device status) at 30-day, 90-day, and long-term follow-up (>90 days) were recorded.

Table 1
Table 1:
Baseline Characteristics Before TandemHeart Support

In accordance with the 2007 International Registry for Mechanically Assisted Circulatory Support (INTERMACS) report,12 the INTERMACS risk profile was calculated for each patient before device implantation. Previously defined and assessed INTERMACS scale uses a seven-level classification scheme that classifies advanced HF patients according to their hemodynamic status before VAD implantation to predict outcomes for patients receiving MCS. The University of Michigan right ventricular failure risk score (MRVFRS) was also calculated for each patient before device implantation.14 The MRVFRS predicts the risk for developing right ventricular (RV) failure after VAD implantation for patients requiring MCS. The key elements of the risk score include the need for vasopressor medications (4 points), presence of liver function abnormalities (2 points for aspartate aminotransferase >80; 2.5 points for total bilirubin >2), and renal dysfunction (three points for serum creatinine >2.3 mg/dl).14

Implantation and management of the TandemHeart system have been described earlier.11 TandemHeart insertion was performed percutaneously in the cardiac catheterization laboratory. Under fluoroscopic guidance, the inflow cannula was inserted into the femoral vein, and using the standard trans-septal technique, a 21-French cannula was inserted into the left atrium over a graduated dilator and heavy-duty guidewire. The outflow cannula then returns oxygenated blood to the femoral artery. Intraoperative placement of the TandemHeart has also been well described.15 In a similar fashion (to the percutaneous insertion) and before sternotomy, the TandemHeart's left atrial transseptal cannula is placed into the femoral vein, advanced to the cavoatrial junction. Sternotomy, cannulation for double-venous return, and cardiopulmonary bypass are then performed (before advancing into the left atrium). Average device speeds for all patients were set between 5,000 and 7,000 rpm.

Patients were weaned from TandemHeart based on real-time assessment of their hemodynamic and end-organ functional status. Weaning protocols did not vary for ST-segment elevation myocardial infarction (STEMI) patients. Device flow rates were continuously adjusted to maintain mixed venous oxygen saturation (SVO2) >60 and mean arterial pressure (MAP) >60 mm Hg. Patients with adequate hemodynamics and end-organ function at PVAD flow rates of <2 L/d for approximately 2 days were gradually weaned from TandemHeart. Device explantation occurred surgically during VAD implantation, open heart transplant, or when clinically indicated at bedside with manual compression.

Data are expressed as mean and SD. Paired Student's t-tests were used to compare hemodynamic parameters and laboratory values pre- and postdevice implantation. All analyses were performed with the use of STATA SE 11 (StataCorp, College Station, TX).


Baseline Assessment

The registry consisted of 25 patients with a mean age of 52 (±12) years and 17 (68%) were men. The indications for TandemHeart support were advanced HF with CS in 14 (56%) patients, STEMI with CS in 5 patients (20%), postpericardiotomy with CS in 4 (16%) patients (coronary artery bypass graft [1], aortic root replacement [1], or aortic root replacement with AVR [2]), and high-risk support (PCI and ventricular tachycardia [VT] ablation) in 2 (8%) patients with end-stage ischemic cardiomyopathy. Sixteen (64%) patients required mechanical ventilation before TandemHeart implantation. The INTERMACS score before device implantation ranged between 1 and 4, with most patients (88%) having a score of 1 (36%) and 2 (52%).15–17 The mean preimplantation MRVFRS was 5.0 ± 3.26.

Metabolic Parameters

Mean preimplantation bilirubin, aspartate aminotransferase, and alanine transaminase levels were 5.72 (±2.94) mg/dl, 1,026 (±2,645) U/L, and 2,645 (±1,500) U/L, respectively. The mean Cockcroft-Gault estimated glomerular filtration rate was 39 (±19) ml/min/m2 preimplantation and 42 (±20) ml/min/m2 after implantation (>24 hours) (p = 0.5). The mixed SVO2 increased from 55.14 ± 13.34 to 66.43 ± 7.43 (p = 0.008) after implantation.

Cardiac Support

Before TandemHeart implantation, intravenous inotropic agents were used in 17 (68%) patients. Among these, five (30%) patients were treated with a single agent, five (30%) required two agents, and seven (40%) patients required a combination of three or more inotropic agents. Fifteen patients (60%) required IABP support and one patient received an Impella 2.5 L device before TandemHeart implantation. Inotropic support was still necessitated in most patients (14) immediately (<24 hours) after TandemHeart insertion before the reassurance of adequate perfusion as indicated by improvements in mixed SVO2 and/or urine output. Patients not requiring inotropic support before device implantation were chronic HF patients presenting with CS (3), postpericardiotomy with CS (2), high-risk PCI (1), and high-risk VT ablation (1) patients. Of the patients who did not require inotropic support immediately (<24 hours) after device implantation, the majority (5/7) did not require inotropic support at any point after device implantation. New, postimplantation inotropic dependence was noted in two patients, both were postpericardiotomy patients. Postimplantation inotrope dependence was no longer necessary in one postpericardiotomy patient and two patients with chronic HF in CS.

Hemodynamic Parameters

The mean systemic arterial pressures, right atrial pressures, pulmonary artery systolic pressures, mean pulmonary artery pressures, pulmonary capillary wedge pressures, and mean cardiac indexes at baseline and after device implantation are listed in Table 2. There was a significant improvement in all hemodynamic values. Pre- and postimplantation left ventricular (LV) ejection fractions (LVEF) were 21.5% (±15%) and 24.5% (±10.5%), respectively (p = 0.06). The mean left ventricular end-diastolic diameter (LVEDD) preimplantation was 54 (±13) mm, which improved to 47 (±13) mm postimplantation (p = 0.008). Severe RV dysfunction was present in seven patients preimplantation and in 10 patients postimplantation. The mean preimplantation MRVFRS for patients with postimplantation RV failure was 6.4 ± 1.3.

Table 2
Table 2:
Hemodynamic Values Before and Immediately After Assist Device Implantation

Survival and Clinical Outcomes

The duration of support ranged from 1 to 8 days (mean 4.8 ± 2 days). Overall, rates of 30-day, 90-day, and long-term survival were 56%, 52%, and 36%, respectively. At the time of study follow-up, six patients (20%) were successfully bridged to recovery, seven (28%) patients were successfully bridged to implantable VAD as bridge-to-transplant, and three (12%) patients were successfully bridged to transplantation (Table 3). Subset analysis revealed a high rate of mortality among STEMI patients, as two patients (40%) died while on TandemHeart support, and only two (40%) survived to the 30-day and long-term follow-up landmarks. Early, high mortality rates were similarly noted in patients requiring TandemHeart support postpericardiotomy as two (50%) died while on PVAD support (both after aortic root replacement with AVR). The remaining two postpericardiotomy patients were successfully discharged and did not require hemodynamic support at long-term follow-up. High-risk percutaneous intervention patients demonstrated a similar rate of mortality as one (50%) patient (post-VT ablation) died of sepsis at the 90-day follow-up period but was alive and well at the 30-day landmark. The other high-risk patient (PCI) was alive at the long-term follow-up. Among HF patients with CS (14), four (28%) died while on TandemHeart support. In addition, 2 (14%), 0 (0%), and 3 (21%) patients died at the 30-day, 90-day, and long-term follow-up landmarks, respectively. Overall, 14 patients (56%) survived to hospital discharge.

Table 3
Table 3:
Outcomes and Complications According to Indication for Device Implantation

Complications and Safety

No patients experienced a procedural complication during TandemHeart insertion. There were no cases of cardiac tamponade, thromboembolic events, arrhythmias, or TandemHeart system failure noted. A total of 13 device-related in-hospital complications were observed (Table 4). Ten (77%) of these complications were vascular in etiology, three of which were minor bleeding from the groin site. One patient did experience significant groin site bleeding requiring multiple blood transfusions and subsequently died without evidence for an underlying hematological process. The venous cannula migrated into the right atrium in two patients, resulting in hypoxemia and ultimately the death of one patient from hypoxic respiratory failure. Two patients experienced a groin site dehiscence and subsequent infection, requiring surgical debridement. In addition, two patients experienced groin dehiscence requiring surgical arterial repair but did not develop infection or significant bleeding. Finally, one patient required surgical fasciotomy for right leg compartment syndrome and subsequently died 2 days after device implantation from multiple system organ failure. All five STEMI patients developed a device-related complication (Table 3). Complications of local, groin site infections occurred among STEMI patients. Bleeding complications occurred among chronic HF patients.

Table 4
Table 4:
Complications and Safety Associated With TandemHeart Implantation

Device Removal and Complications

Device explantation occurred surgically in 18 patients, at the bedside with FemoStop device assistance in three patients, and four patients died with the TandemHeart device still in place. Patients with bedside device removal were noted to have only one complication (33%); minor bleeding. Patients with surgical device explantation had seven (38%) vascular complications including pseudoaneurysm, groin dehiscence, femoral artery tear, and compartment syndrome.


Effective therapies for patients with acutely decompensated HF and/or advanced HF with CS are limited. Such patients often decompensate rapidly, necessitating immediate hemodynamic stabilization. A minority of patients are able to achieve hemodynamic stability with inotropic and/or IABP therapy. In addition, initial responders to inotropic agents and/or IABP therapy may decompensate over time and will require advanced MCS. Therefore, the majority of HF patients will require advanced mechanical support. Appreciating that the potential for heart transplant is limited as donor availability remains exceedingly low, stabilized, decompensated HF patients must receive MCS as a bridge-to-bridge (VAD) or bridge-to-transplant. Several devices for temporary mechanical support have been used to appropriately bridge patients in refractory CS. However, patient outcomes with these different temporary circulatory support devices have varied and have been underreported.18,19

In this series, most patients requiring TandemHeart support for advanced HF had concurrent CS (14). Intravenous, inotropic support was required in 70% of patients before device implantation. Of these patients not requiring inotropic support (7), three were in CS from advanced HF, two were post-STEMI patients, and two were postpericardiotomy patients. Less surprisingly, both the patients with high-risk percutaneous intervention did not require inotropic support before or after device implantation. Two HF patients in CS were able to discontinue inotropic support immediately after TandemHeart implantation. The low dependence of inotropic support pre- and postimplantation among chronic HF patients in CS was an unanticipated finding and one that warrants further investigation. Evidence for hemodynamic improvement after device implantation was demonstrated with moderate improvements in the LVEF, cardiac output, and LVEDD (3.0%, 1.2 L/min, and 7 mm, respectively) and could correlate with the aforementioned inotropic dependence of the patient subgroups.

Before the development of extracorporeal temporary assist devices, temporary, hemodynamic support for patients in CS was conferred by ECMO. Although still a therapeutic option for acutely ill patients, ECMO therapy often requires invasive surgical procedures and has a high associated mortality risk among patients who require cardiopulmonary bypass. In addition, ECMO provides incomplete unloading of the left ventricle, may increase LV afterload, and thus many ECMO patients require simultaneous decompression of the left ventricle through continued inotropic support. The first significant series reporting on the clinical experience with ECMO for HF patients was published by Pagani et al.19 Of the 25 patients who were placed on ECMO, only 8 survived to bridge-to-bridge therapy and 3 were successfully bridged to transplantation. Of the 30 patients who underwent LVAD implantation without ECMO support, 23 had a successful bridge to transplantation. In a more contemporary series, Hoefer et al.20 reported similar findings among a patient cohort of 131 HF patients receiving ECMO therapy. There was a similarly high overall mortality of 42%. Twenty-one percent of patients were successfully bridge to bridge (BTB), and 12% were successfully BTT. The high mortality and poor BTB and BTT rates associated with ECMO requires physicians to consider the implantation of extracorporeal temporary assist devices for hemodynamic support, as use of these devices are associated with improved outcomes. In fact, a recent case report concluded that the TandemHeart system can provide superior ECMO more safely than surgical ECMO implementation among patients with cardiopulmonary respiratory failure.21

Previous MCS has included the use of IABP and the surgical insertion of extracorporeal assist devices. Since their introduction into practice >30 years ago, IABPs have been used to support patients with CS in a variety of clinical scenarios. The major advantage of IABPs is their ease of implementation. IABPs can provide a limited hemodynamic support and require a certain residual level of LV function.22 IABPs reliance on sinus rhythm for maximal hemodynamic benefit further limits their use among high-risk HF patients and postcardiothoracic surgery patients who often have underlying arrhythmias.

A multicenter randomized trial has compared the safety and efficacy of the TandemHeart with IABP in patients with CS. Compared with the IABP, the TandemHeart significantly reduced LV preload, increased cardiac output, and improved end-organ dysfunction.23 However, complications of limb ischemia and severe bleeding are more common with the TandemHeart than IABP.23,24 Despite better hemodynamic support provided by TandemHeart compared with IABP support, no study has demonstrated an overall survival benefit with TandemHeart therapy versus IABP therapy. It therefore seems reasonable that before TandemHeart implantation, physicians assess the patients' risk for developing complications, as the hemodynamic benefit conferred by this device over IABP may not bestow a benefit in survival.

The Impella Recover PVAD system is available in two models: LP 2.5 and LP 5.0. Implantation of the Impella Recover systems requires femoral arterial but not venous access. The Impella Recover LP 2.5 is suitable for percutaneous implantation, whereas the larger Impella Recover LP 5.0 catheter requires surgical cut down of the femoral artery for device insertion. The Impella Recover LP 2.5 system and the LP 5.0 system are capable of providing 2.5 and 5.0 L/min of cardiac support, respectively. The advantages of the Impella system include short implantation times (average 50 minutes) and moderate hemodynamic support with relatively low rates of associated complications.25 The hemodynamic effect of the Impella Recover LP 2.5 device has been reported.26 Among their small patient cohort, the investigators reported no significant unloading of the left ventricle, and there were no statistically significant changes in stroke volume, cardiac output, and ejection fraction. These findings are consistent with the outcomes we experienced with one patient who received the Impella Recover LP 2.5. With inadequate improvements in cardiac output and ejection fraction, the Impella device was exchanged for a TandemHeart device, after which time the patient's hemodynamic status improved significantly. Currently, there are no studies which compare the Impella devices directly with IABP or the TandemHeart.

This study supports the assertion that TandemHeart provides significant hemodynamic support among patients with CS regardless of the etiology. The hemodynamic improvements provided by the device unloaded the left ventricle, and there were significant improvements in metabolic parameters, which may prevent and/or reverse organ dysfunction among patients with refractory, end-stage HF.24

These aforementioned, hemodynamic changes have been associated with improvements in myocardial strain and tissue perfusion at both the coronary and peripheral tissue level.27 Our results, demonstrating the favorable hemodynamic effects of TandemHeart are congruent with previous reports. Thiele et al.27 demonstrated that among STEMI patients receiving TandemHeart therapy, there were dramatic improvements in vital organ function, which the authors associate with changes in the hemodynamic parameters after device implantation. Notably, the authors found improvements in cardiac output (before support 3.5 ± 0.8 L/min versus with support 4.8 ± 1.1 L/min, p < 0.001), pulmonary capillary wedge pressure (before support 21 ± 4 mm Hg versus with support 14 ± 4 mm Hg, p < 0.001), mean pulmonary arterial pressure (before support 31 ± 8 mm Hg versus with support 23 ± 6 mm Hg, p < 0.001), and central venous pressure (before support 13 ± 4 mm Hg versus with support 9 ± 3 mm Hg, p < 0.001). With its favorable improvements in hemodynamics and broad applicability among HF patients, high-risk patients with percutaneous interventions,28 acute myocardial infarction complicated by CS, and decompensated HF; TandemHeart use for patients in CS is a reasonable therapeutic option for improving systemic circulation, tissue perfusion, and organ function.

Patient risk stratification is a crucial element for improving outcomes in patients requiring temporary MCS. The rapid diagnosis and treatment of CS are essential but are often based on limited information. Hemodynamic and metabolic data, including urine output, serum creatinine, mixed SVO2, pulmonary artery pressure, central venous pressure, and mean arterial pressures, may help guide the overall patient selection, the management, and potential weaning decisions for mechanically supported patients. The INTERMACS score has been previously validated as a useful classification scheme to risk-stratify HF patients and predict outcomes after assist device implantation.29,30

In the current series, eight patients died while on TandemHeart support, and all eight patients were in profound CS (INTERMACS mean score of 1.38) at the time of treatment initiation. Most of the deaths occurred soon after device implantation and were related to multiple organ system failure. These outcomes are similar to those reported by other groups and highlight the importance of early device implantation and patient risk stratification.27 Our study, with a mean INTERMACS score of 1.8, demonstrated 30-day, 90-day, and long-term survival rates of 56%, 48%, and 36%, respectively, with similar rates of mortality regardless of the indication for device implantation. In similar patient cohorts, INTERMACS scores between 1.5 and 3.0 have demonstrated the overall predicted survival after implantation between 30% and 50%, congruent with our findings.19 Our data indicates that INTERMACS scores between 1 and 2 are associated with more profound decompensated HF, higher mortality rate, and worse overall survival, whereas INTERMACS scores of 3 and 4 were associated with better outcomes. This data and data from other study groups suggest that patients with an INTERMACS score close to or <2 are reasonable candidates for TandemHeart implantation.

RV dysfunction was present in seven HF patients before device implantation, and three (42%) of these patients died before the 30-day landmark. Patients with RV dysfunction at baseline should be considered a higher mortality risk cohort than similar patients without severe RV dysfunction where the 30-day mortality rate was noted to be lower (34%). MCS should therefore be considered early among HF patients with concomitant severe RV dysfunction.22 Although never validated and developed among a VAD patient cohort, the application of the MRVFRS in this study suggests that it may significantly predict patients who are at high risk for developing RV failure after PVAD implantation. Three patients developed postimplantation RV failure and had a mean MRVFRS of 6.4, which was 1.4 points higher than the cohort, mean MRVFRS of 5.0. According to the study authors, the MRVFRS would have predicted an odds ratio of 7.6 for patients with RV failure and 2.8 for those patients who did not develop postimplantation RV failure.14

The in-hospital device-related complication rate was relatively high, with the caveat that our study was performed among a high-risk patient cohort. The most common complication was access site-related bleeding. The rate of bleeding was not greater than what has previously been reported or expected, as direct cannulation of the femoral vessels and aggressive anticoagulation therapy are necessary with the use of TandemHeart. Implementation of an ultrasound-guided puncture of the femoral artery with iliac angiography has the potential to minimize access site-related complications and was not routinely performed on these patients.27 A prudent, preinsertion risk assessment should therefore include a angiographic runoff study to asses vessel diameter where clinically feasible and was not completed for all patients in this cohort.

Similar rates of local complications were noted regardless of the method of device explanation. However, the severity of complications associated with the surgical explantation of the TandemHeart device was more severe and frequent compared with bedside explantation. Patients with bedside device explantation were likely more clinically stable and therefore less likely to develop complications compared with patients requiring/undergoing surgical device explantation. Notably, all STEMI patients developed a device-related complication, likely reflective of the urgent environment in which vascular access was attained for these patients. The urgency of device implantation may also explain why complications of infection only occurred among STEMI patients.

Careful attention must be directed to avoid accidental dislodgement of the arterial transseptal cannula, which occurred in two of our patients. When the transseptal cannula falls into the right atrium, significant right-to-left shunting results in acute decompensation and could be prevented through the meticulous securement of the cannula at the skin site.27 Both these events occurred early in our experience with the TandemHeart device, suggesting the existence of an institutional learning curve associated with device implantation. We noted two cases of infections during the peri-implantation period, which is congruent with previous reports. Importantly, there were no mechanical device failures reported and no device-associated thromboembolic events. Overall, postimplantation complication rates were similar to previously published rates. Bleeding and vascular complications were the major issues associated with implantation of this device.


Applicability of this single-center experience to other centers is a limitation of this study. Although we used the TandemHeart under current, reasonable practices, there are currently no uniformly accepted guidelines of how best to implement TandemHeart therapy.

Despite this being the largest single-center reported experience with TandemHeart, sample size is a study limitation. Our rates of complications and death are high; however, the study was performed in high-risk patients.


To the best of our knowledge, this study represents one of the largest published registry analyzing the clinical outcomes and safety associated with the use of the TandemHeart mechanical support system for patients with acute decompensated HF and/or refractory, advanced HF, postpericardiotomy, and high-risk percutaneous interventions. Our assessment suggests that the TandemHeart device is a safe and effective therapeutic option as bridge-to-recovery or bridge-to-bridge for patients with CS regardless of the etiology. This device allows for proper long-term decision making regarding patient management while minimizing preoperative risk factors in patients who are in refractory CS if implantation of a long-term device is indicated. This study demonstrates the use of TandemHeart to provide hemodynamic support for patients with acutely decompensated HF, advanced HF with CS, postpericardiotomy, and during high-risk cardiac interventions in patients not deemed to be appropriate surgical candidates. The initiation and termination of TandemHeart support has been demonstrated to occur safely in both elective and emergent situations in a diverse and complex subset of patients.


1. Fang J, Mensah GA, Croft JB, Keenan NL: Heart failure-related hospitalization in the U.S., 1979 to 2004. J Am Coll Cardiol 52: 428–434, 2008.
2. Jessup M, Abraham W, Casey D, et al: 2009 focused update: ACCF/AHA Guidelines for the Diagnosis and Management of Heart Failure in Adults: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines: Developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation 119: 1977–2016, 2009.
3. Fonarow GC, Heywood JT, Heidenreich PA, et al: Temporal trends in clinical characteristics, treatments, and outcomes for heart failure hospitalizations, 2002 to 2004: Findings from Acute Decompensated Heart Failure National Registry (ADHERE) Am Heart J 153: 1021–1028, 2007.
4. Lloyd-Jones D, Adams RJ, Brown TM, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee: Heart disease and stroke statistics—2010 update: A report from the American Heart Association. Circulation 121: e46–e215, 2010.
5. Allen LA, Rogers JG, Warnica JW, et al; Escape Investigators and Coordinators: High mortality without ESCAPE: The registry of heart failure patients receiving pulmonary artery catheters without randomization. J Card Fail 14: 661–669, 2008.
6. Fonarow GC: Epidemiology and risk stratification in acute heart failure. Am Heart J 155: 200–207, 2008.
7. Hausmann H, Potapov EV, Koster A, et al: Prognosis after the implantation of an intra-aortic balloon pump in cardiac surgery calculated with a new score. Circulation 106(suppl I): I203–I206, 2002.
8. McCarthy PM, Smedira NO, Vargo RL, et al: One hundred patients with the HeartMate left ventricular assist device: Evolving concepts and technology. J Thorac Cardiovasc Surg 115: 904–912, 1998.
9. Pagani FD, Lynch W, Swaniker F, et al: Extracorporeal life support to left ventricular assist device bridge to heart transplant: A strategy to optimize survival and resource utilization. Circulation 100(suppl 19): II206–II210, 1999.
10. Granfeldt H, Hellgren L, Dellgren G, et al: Experience with the Impella recovery axial-flow system for acute heart failure at three cardiothoracic centers in Sweden. Scand Cardiovasc J 43: 233–239, 2009.
11. Kar B, Adkins LE, Civitello AB, et al: Clinical experience with the TandemHeart percutaneous ventricular assist device. Tex Heart Inst J 33: 111–115, 2006.
12. International Registry for Mechanically Assisted Circulatory Support (INTERMACS). Available at: Accessed March 12, 2010.
13. Gregoric ID, Bruckner BA, Jacob L, et al: Techniques and complications of TandemHeart ventricular assist device insertion during cardiac procedures. ASAIO J 55: 251–254, 2009.
14. Matthews JC, Koelling TM, Pagani FD, Aaronson KD: The right ventricular failure risk score a pre-operative tool for assessing the risk of right ventricular failure in left ventricular assist device candidates. J Am Coll Cardiol 51: 2163–2172, 2008.
15. Kirklin JK, Naftel DC, Stevenson LW, et al: INTERMACS database for durable devices for circulatory support: First annual report. J Heart Lung Transplant 27: 1065–1072, 2008.
16. Stevenson LW, Couper G: On the fledgling field of mechanical circulatory support. J Am Coll Cardiol 50: 748–751, 2007.
17. Wilson S, Mudge G Jr, Stewart GC, Givertz M: Evaluation for a ventricular assist device: Selecting the appropriate candidate. Circulation 119: 2225–2232, 2009.
18. Morgan JA, Stewart AS, Lee BJ, et al: Role of the Abiomed BVS 5000 device for short-term support and bridge to transplantation. ASAIO J 50: 360–363, 2004.
19. Pagani FD, Aaronson KD, Swaniker F, Bartlett RH: The use of extracorporeal life support in adult patients with primary cardiac failure as a bridge to implantable left ventricular assist device. Ann Thorac Surg 71(suppl 3): S77–S81; discussion S82–S85, 2001.
20. Hoefer D, Ruttmann E, Poelzl G, et al: Outcome evaluation of the bridge-to-bridge concept in patients with cardiogenic shock. Ann Thorac Surg 82: 28–33, 2006.
21. Herlihy JP, Loyalka P, Jayaraman G: Extracorporeal membrane oxygenation using the TandemHeart system's catheters. Tex Heart Inst J 36: 337–341, 2009.
22. Ferguson JJ III, Cohen M, Freedman RJ Jr, et al: The current practice of intra-aortic balloon counterpulsation: Results from the Benchmark Registry. J Am Coll Cardiol 38: 1456–1462, 2001.
23. Burkhoff D, Cohen H, Brunckhorst C, O'Neill WW; TandemHeart Investigators Group: A randomized multicenter clinical study to evaluate the safety and efficacy of the TandemHeart percutaneous ventricular assist device versus conventional therapy with intraaortic balloon pumping for treatment of cardiogenic shock. Am Heart J 152: 469.e1–469.e8, 2006.
24. Thiele H, Sick P, Boudriot E, et al: Randomized comparison of intra-aortic balloon support with a percutaneous left ventricular assist device in patients with revascularized acute myocardial infarction complicated by cardiogenic shock. Eur Heart J 26: 1276–1283, 2005.
25. Henriques JP, Remmelink M, Baan J Jr, et al: Safety and feasibility of elective high-risk percutaneous coronary intervention procedures with left ventricular support of the Impella Recover LP 2.5. Am J Cardiol 97: 990–992, 2006.
26. Valgimigli M, Steendijk P, Serruys PW, et al: Use of Impella Recover(R) LP 2.5 left ventricular assist device during high-risk percutaneous coronary interventions: Clinical, haemodynamic and biochemical findings. EuroIntervention 2: 91–100, 2006.
27. Thiele H, Lauer B, Hambrecht R, et al: Reversal of cardiogenic shock by percutaneous left atrial-to femoral arterial bypass assistance. Circulation 104: 2917–2922, 2001.
28. Vranckx P, Foley P, De Feijter PJ, et al: Clinical introduction of the Tandemheart, a percutaneous left ventricular assist device, for circulatory support during high-risk percutaneous coronary intervention. Int J Cardiovasc Intervent 5: 35–39, 2003.
29. Stevenson LW, Pagani FD, Young JB, et al: INTERMACS profiles of advanced heart failure: The current picture. J Heart Lung Transplant 28: 535–541, 2009.
30. Alba AC, Rao V, Ivanov J, et al: Usefulness of the INTERMACS scale to predict outcomes after mechanical assist device implantation. J Heart Lung Transplant 28: 827–833, 2009.
Copyright © 2011 by the American Society for Artificial Internal Organs