Clinical Application of Prophylactic Percutaneous Left Ventricular
Assist Device (Tandem Heart‚Ñ¢) in High-Risk Percutaneous
Cor

Introduction of novel devices for cardiac support during complex percutaneous coronary intervention (PCI) has revolutionized the treatment of coronary artery disease (CAD) and has led to an increasing number of high-risk PCI procedures. Traditionally, the intra-aortic balloon pump (IABP) has been used to provide hemodynamic stability during coronary interventions in high-risk patients. However, the mortality rate continues to be high (30–75%) despite the use of IABPs in the setting of acute myocardial infarction complicated by cardiogenic shock; insertion of a percutaneous transseptal ventricular assist device (PTVA) has shown to improve outcomes.¹ The TandemHeart^™ (TH) (Cardiac Assist, Inc., Pittsburg, Pennsylvania) is a PTVA system which can be used to support the failing heart in high-risk PCI.^2–6 Ventricular assist has been found to be particularly useful when IABP fails to provide desired hemodynamic support during PCI.⁷ This patient subset includes those with arrhythmias, very poor left ventricular ejection fraction (LVEF), and patients with left main coronary artery (LMCA) involvement who have been refused cardiac surgery on account of comorbid conditions or who were at prohibitive risk of death or major complications during surgery. The versatile nature of the TH, the ease of device implantation in the cardiac catheterization laboratory and the excellent short-term hemodynamic support provided by this device have overcome many of the limitations of the IABP.^7–9 We report our experience in the first 20 consecutive patients in whom the TH was utilized as a short-term hemodynamic support option for high-risk PCI. To our knowledge, this is one of the largest series involving the use of this device in high-risk patients. Also not reported previously for carrying out an interventional coronary procedure is our technique of preclosure of the arterial access site using 2 Perclose^™ devices (Abbott Vascular, Redwood City, California) for minimization of vascular complications.

Methods

Twenty consecutive high-risk elderly patients in whom this recently-introduced PTVA was used at the Mount Sinai Medical Center, New York City from April 2004 to November 2005 during PCI, were retrospectively analyzed from a large interventional database. The Institutional Review Board of the Mount Sinai Medical Center approved this study. The patients were considered high-risk because of their complex coronary anatomy, low LVEF (12/20, 60%), comorbid conditions and/or refusal for coronary artery bypass surgery (CABG) by the patient or the operating surgeon due to prohibitive surgical risk.
Of the 8 patients who had a LVEF > 50%, 6 had LMCA involvement. Two others had complex bifurcation lesions with significant angina and dominant left and right coronary systems, respectively, and hence were considered high risk. Considering that the mean age of the patients in our series was about 70 years, and that the majority of the patients underwent rotational atherectomy as a plaque-modifying strategy, a more durable hemodynamic support was thought necessary in anticipation of prolonged procedure times. The utility of the TH in cardiogenic shock has been previously documented,^7,8 and hence those patients with this condition were excluded from this analysis.
Informed consent was obtained from all patients. Vascular access was obtained using the standard Seldinger technique, and 2 suture-mediated Perclose devices were deployed (without knotting the sutures down to the surface of the arteries) in the femoral arteries. Implantation of the TH was done prior to starting the PCI. Implantation time was defined as the time from the start of the TH implantation to the start of PCI. All PCI procedures were performed using conventional techniques.

Patients were followed for inhospital major adverse cardiac events (MACE) defined as death, myocardial infarction or stroke. One-year clinical follow up was obtained by telephone calls to the patients. Angiographic follow up was obtained if clinically necessary or when mandated by the institution’s study protocol, as is the case for LMCA stenting.

PTVA Implantation Procedure
Femoral artery preclosure technique using the Perclose device. All patients were subjected to an ipsilateral oblique projection femoral artery angiogram using a 4 Fr arterial sheath dilator before upsizing to a larger French arterial sheath. The purpose of this angiogram was to delineate the angiographic anatomy of the ilio-femoral system and to rule out peripheral vascular disease. If the vessel anatomy was found conducive (stick in the common femoral artery, artery size > 5 mm), the 4 Fr sheath dilator was exchanged for a 6 Fr sheath. If the stick was found to be in the superficial femoral artery or profunda femoris artery, the 4 Fr sheath dilator was removed, manual compression applied and hemostasis achieved. Using the angiographic anatomy obtained previously, a higher stick was then attempted under fluoroscopy and a 6 Fr sheath was introduced into the common femoral artery. The 6 Fr introducer sheath was removed, leaving the J-tip wire in the artery. The 2 Perclose devices were then sequentially deployed at 90 degree angles to each other (2 o’clock and 11 o’clock positions), yielding two sets of suture limbs spaced at 90 degree intervals. These were then harvested and secured using artery forceps. Care was taken not to advance the suture knots down to the arterial wall at this stage. The 6 Fr sheath was then reinserted over the J-wire. This access site was used for subsequent placement of the arterial canula of the TH. Two patients had unfavorable femoral artery angiographic anatomy for the preclosure technique, thus their groins were manually held postprocedure.
TandemHeart implantation technique. The right femoral vein was accessed for transseptal puncture which was performed using the Brockenbrough needle and a Mullins sheath as is done during balloon mitral valvuloplasty. After confirming the position of the Mullins sheath in the left atrium, unfractionated heparin was given to achieve a target activated clotting time > 300 seconds. The Mullins sheath was exchanged for the 14/21 Fr two-stage dilator over a 0.038 inch J-tip 260 cm Amplatz Super Stiff^™ guidewire (Boston Scientific Corp., Natick, Massachusetts). Next, the 21 Fr TH transseptal canula was advanced along with the 14 Fr obturator over the Amplatz Super Stiff guidewire and placed in the left atrium. The position of the canula in the left atrium was confirmed by injecting dye and also by drawing blood and assessing the saturation level. The obturator and the wire were then removed and clamps were applied for temporary hemostasis. Care was taken to place all the sideholes of the TH canula into the left atrium to avoid possible right-to-left shunting during device operation. The peripheral end of the canula was sutured to the skin of the patient’s thigh and clamped. The left femoral artery 6 Fr sheath was then exchanged for the 15 Fr arterial perfusion cannula of the TH device with the distal end of the canula lying above the aortic bifurcation. The peripheral end of the canula was similarly sutured to the patient’s thigh and clamped. The femoral venous canula was connected to the inflow conduit and the femoral arterial canula to the outflow conduit of the TH pump after performing de-airing according to the specified protocol. A heparinized saline infusate was started according to the product specification, which provided hydrodynamic bearing, anticoagulation and local cooling for the motor of the pump. The pump was then connected to the TH controller and its speed adjusted to provide a support of 2.5 to 3.0 L/minute. In all cases, the TH device was removed after the procedure before the patient left the catheterization laboratory.
Perclose suture application technique after removal of the TH cannulae. After completion of the procedure, the venous canula was removed from the left atrium and placed in the inferior vena cava, after which the pump was stopped. The 15 Fr arterial cannula was removed from the left femoral artery over a 0.038 inch J-tip wire and temporary manual compression was applied. The first set of suture limbs and its knot were then advanced down to the arterial wall using the knot advancer. The knots were deployed in the same sequence as their initial preclosure and tied. Additional manual compression was applied when needed.

Results

Clinical and procedure-related characteristics of the study patients are summarized in Table 1. Briefly, the mean age of the 20 study patients was 70 ± 11 years (6 females). The mean LVEF was 38 ± 18% (range 15–60%). The patient group was considered to represent a high-risk subset on account of associated high-risk features (10 had a LVEF £ 30%, 8 had LMCA involvement, 3 underwent prior CABG).Preclosure sutures were placed successfully in 18/20 patients using the technique described. Implantation of the TH was successful in 100% of patients. During the procedure, rotational atherectomy was performed in 17/20 (85%) cases as a plaque-modifying strategy. The implantation time, procedure time and duration of TH support were 32 ± 10, 92 ± 12 and 72 ± 41 minutes, respectively. The average hospital stay of the patients was 2 ± 1 days.
Follow up. No patient experienced any MACE during the index hospitalization. Clinical follow up at 1 year was available in 19/20 (95%) patients. Two patients died, but their cause of death could not be ascertained. Angiographic follow up was available for 7 patients, 2 of whom had in-stent re-stenosis of their LMCA after a mean interval of 5.5 months from the date of the index procedure and were referred for CABG. Both of these patients had a LVEF < 40%, which improved to > 40% on follow up. Of the 17 patients who were alive, 15 were free of MACE at 1 year.
Vascular complications. Of the 20 patients, there was 1 minor vascular complication in the form of a moderate-sized hematoma which was treated conservatively. This patient was on chronic anticoagulation therapy with coumadin for atrial fibrillation and had Perclose sutures predeployed. However, no ischemic complications were noted.

Discussion
The main findings of our study include the feasibility and ease of TH implantation in patients as a prophylactic device for hemodynamic support during high-risk PCI and the success of the suture-mediated preclosure technique in achieving hemostasis after removal of large-bore arterial cannulae.
With the advent of newer percutaneous modalities to support a poorly functioning ventricle, comparisons with conventional hemodynamic support systems like the IABP seeminevitable. Two existing randomized, controlled trials^7,8 have compared the efficacy of the IABP to that of the TH PTVA in patients presenting with cardiogenic shock. These studies concluded that percutaneous LV assist provides better and more durable hemodynamic support during high-risk PCI. The critically ill patients were more responsive, and those in cardiogenic shock had early reversal of their shock after TH implantation when compared to IABP use. However, none of the studies showed a mortality benefit, even when evaluated for short-term outcomes, and there was a trend toward higher vascular complications with TH implantation. Despite this, it was recommended that a LV assist device should be considered for additional hemodynamic support in instances where an IABP does not provide adequate hemodynamic support, or where it fails.
The patient subset in our series was considered high-risk given that the majority of the patients were elderly with significant atherosclerotic burden, most of them (85%) required rotational atherectomy for plaque modification and stent delivery, 40% had LMCA involvement, and 50% in the present series had a LVEF £ 30%. We anticipated long procedure times (mean procedure time > 100 minutes), and it was felt that longer-lasting LV assistance was needed, hence we chose the TH as the support device. We think that the TH would be a better alternative than the IABP in following circumstances:
1. Atrial fibrillation/arrhythmia;
2. Patients with a low LVEF (< 20%);
3. Attempted PCI on the LMCA irrespective of the EF;
4. Complex coronary anatomy subset where rotational atherectomy/other plaque modification strategies are employed that may prolong the procedure time;
5. Where procedure time is expected to exceed 60 minutes and hemodynamic support is considered prudent.
In an early clinical experience with elective PTVA implantation by Vranckx et al² before high-risk PCI, encouraging short-term results were obtained. None of the 3 patients in their study were in cardiogenic shock and they remained free of MACE up to 7 months postprocedure. In our series too, none of the patients were in cardiogenic shock; however, the clinical or angiographic substrate was considered to be extremely high-risk for PCI, and given the superior hemodynamic profile offered by the TH, it was considered prudent to percutaneously implant the TH electively before the procedure as a temporary hemodynamic support option. Eight patients who underwent LMCA intervention either opted for PCI or were refused by surgeons on account of comorbidities (2 had a low EF with CHF, 1 had cancer, 1 had an aortic aneurysm with CHF and 4 patients refused CABG). As per the institutional protocol, an expert surgical opinion was first sought, and only after refusal from the patient or from the operating surgeon was LMCA PCI attempted. Six out of 8 patients with LMCA involvement had calcified vessels and underwent rotational atherectomy as a plaque-modification strategy. Moreover, 6/20 (30%) high-risk patients in our series had atrial fibrillation, and by default, the TH was considered the device of choice in these patients given the superior hemodynamic profile on the TH in patients with arrhythmia.⁸
In a study by Kar et al, the TH was implanted in 18 patients prior to PCI, 11 of whom were in cardiogenic shock.¹⁰ Despite encouraging hemodynamic profile, shortterm mortality (30 days) in the cardiogenic shock group was high; 6/11 (54%) patients failed to survive at 30 days. This was in contrast to 6/7 (86%) patients not in cardiogenic shock who managed to survive 30 days, suggesting that earlier institution of temporary hemodynamic support before cardiogenic shock sets in could lead to better short-term outcomes. Kar et al further confirmed this in another study where the TH was implanted percutaneously prior to high-risk PCI (unprotected LMCA or left main equivalent) in 5 patients who were not eligible for CABG, with 100% procedural success.¹¹
Therefore, implantation of the TH PTVA early in the course of a planned high-risk PCI may provide vital hemodynamic support, which in turn translates into better periprocedural outcomes. The success rate of TH implantation in their series was 100%, with an average implantation time of 45–60 minutes. We had similar implantation success rates with shorter implantation times (mean 31 ± 11) minutes.
While the TH provides good hemodynamic support for carrying out high-risk procedures, long-term outcomes and prognosis remain unaltered. Clinical follow up was available for 19 patients and angiographic follow up in 8 patients. Two patients died in our series (both with LMCA involvement), 1 at day 4, and the other at 10 months, the cause of which could not be ascertained. Acute stent thrombosis could possibly explain death in 1 of these patients, and progressive heart failure and/or sudden cardiac death might have been responsible in the other patient (LVEF 15%). Eight patients underwent repeat procedures due to recurrent angina, 2 patients with LMCA restenosis underwent CABG, while 6 patients underwent revascularization procedure for native vessel disease progression and/or in-stent restenosis not involving the LMCA. Ten patients remained symptom-free at 1 year following the index procedure. The follow-up outcomes reflect the advanced atherosclerotic burden and high risk underlying this anatomical subset and probably remain unaffected by the type of periprocedural hemodynamic support.
Vascular complications have always been a concern after deployment of large-sized arterial cannulae. Bleeding and loss of distal pulse have been reported as complications of implantation of large-bore arterial cannulae.^6–9 These have been largely related to the size of the arterial puncture and the presence of peripheral vascular disease, respectively. Use of the Perclose device as a method of achieving hemostasis postprocedure for large arteriotomies (10–14 Fr) has been shown to result in earlier ambulation and discharge from the hospital.^12–14 However, these studies have been conducted with the use of either single or large-sized Perclose devices (8 or 10 Fr). Moreover, the size of the arteriotomy for which manual compression may be considered relatively safe and effective is not well defined. A hypothesis that 12 Fr is the maximum size ofthe arteriotomy that can safely be closed with manual compression was proposed by Quin and Kim.¹⁵ Anthony Lee et al recently reported the use of two 6 Fr Perclose Proglide devices for large arteriotomies (12–24 Fr femoral artery sheaths) performed for endovascular repair of aortic aneurysms.¹⁶ Two Proglide devices were used for preclosing 279 femoral arteries, with a reported technical success rate of 94.3%. We used a similar methodology for preclosure of the femoral arteries that were used for insertion of the arterial canula of the TH device (15 Fr sheath). As a standard protocol, we used femoral artery angiography in an ipsilateral oblique projection to assess the suitability of the vascular access site and to rule out peripheral vascular disease before TH implantation. This may have lead to the low vascular complication rate observed in our study. The femoral artery angiographic anatomy of 2/20 patients was found unfavorable (focal common femoral artery stenosis in one and peripheral vascular disease in the other) for the preclosure technique, thus these groins were manually held for prolonged periods without any complication.

Conclusion
The TH as a left ventricular assist device provides durable hemodynamic support for high-risk PCIs with short implantation times without affecting the long-term follow-up outcomes. When combined with the technique of Perclose preclosure, vascular complications can be minimized.

References

1. Chen JM, DeRose JJ, Slater JP, et al. Improved survival rates support left ventricular assist device implantation early after myocardial infarction. J Am Coll Cardiol 1999;33:1903–1908.
2. Vranckx P, Foley DP, de Feijter PJ, et al. Clinical introduction of the TandemHeart, a percutaneous left ventricular assist device, for circulatory support during highrisk percutaneous coronary intervention. Int J Cardiovasc Intervent 2003;5:35–39.
3. Kar B, Forrester M, Gemmato C, et al. Use of the TandemHeart percutaneous ventricular assist device to support patients undergoing high-risk percutaneous coronary intervention. J Invasive Cardiol 2006;18:93–96.
4. Kar B, Butkevich A, Civitello AB, et al. Hemodynamic support with a percutaneous left ventricular assist device during stenting of an unprotected left main coronary artery. Tex Heart Inst J 2004;31:84–86.
5. Kar B, Forrester M, Gemmato C, et al. Use of the TandemHeart percutaneous ventricular assist device to support patients undergoing high-risk percutaneous coronary intervention. J Invasive Cardiol 2006;18:A6.
6. Aragon J, Lee MS, Kar S, Makkar RR. Percutaneous left ventricular assist device: “TandemHeart” for high-risk coronary intervention. Catheter Cardiovasc Interv 2005;65:346–352.
7. Burkhoff D, Cohen H, Brunckhorst C, O'Neill WW. A randomized multicenter clinical study to evaluate the safety and efficacy of the TandemHeart percutaneous ventricular assist device versus conventional therapy with intra-aortic balloon pumping for treatment of cardiogenic shock. Am Heart J 2006;152:469, e1–e8.
8. 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 2005;26:1276–1283.
9. Thiele H, Lauer B, Hambrecht R, et al. Reversal of cardiogenic shock by percutaneous left atrial-to-femoral arterial bypass assistance. Circulation 2001;104:2917–2922.
10. Kar B, Adkins LE, Civitello AB, et al. Clinical experience with the TandemHeart percutaneous ventricular assist device. Tex Heart Inst J 2006;33:111–115.
11. Kar B, Forrester M, Gemmato C, et al. Use of the TandemHeart percutaneous ventricular assist device to support patients undergoing high-risk percutaneous coronary intervention. J Invasive Cardiol 2006;18:A6.
12. Mylonas I, Sakata Y, Salinger M, et al. The use of percutaneous suture-mediated closure for the management of 14 French femoral venous access. J Invasive Cardiol 2006;18:299–302.
13. Tron C, Koning R, Eltchaninoff H, et al. A randomized comparison of a percutaneous suture device versus manual compression for femoral artery hemostasis after PTCA. J Interv Cardiol 2003;16:217–221.
14. Michaels AD, Ports TA. Use of a percutaneous arterial suture device (Perclose) in patients undergoing percutaneous balloon aortic valvuloplasty. Catheter Cardiovasc Interv 2001;53:445–447.
15. Quinn SF, Kim J. Percutaneous femoral closure following stent-graft placement: Use of the Perclose device. Cardiovasc Intervent Radiol 2004;27:231–236.
16. Lee WA, Brown MP, Nelson PR, et al. Percutaneous access for endovascular aortic aneurysm repair (“Preclose” technique). J Vasc Surg 2007;45:1095–1101.