In this issue of the Journal, Biswajit Kar, et al. report their initial experience using the TandemHeart Percutaneous Ventricular Assist Device (CardiacAssist, Inc., Pittsburgh, Pennsylvania) in patients who were undergoing a high-risk percutaneous coronary intervention (PCI). The authors chose patients who were at risk for hemodynamic collapse during the PCI. These patients had an unprotected left main stenosis or the equivalent, and were not considered to be candidates for cardiac surgery because of their multiple medical comorbidities.
       The TandemHeart is a left ventricular assist device that can be implanted percutaneously. The device decompresses the left ventricle and maintains adequate perfusion of major organs. A transseptal puncture is performed initially, then a 21 Fr cannula is advanced from a femoral vein to the left atrium. The inflow into the centrifugal pump, therefore, is oxygenated blood from the left atrium. The pump can deliver up to 3.5 liters through a 15 Fr or 17 Fr femoral arterial cannula.
       Existing options for hemodynamic support of high-risk PCI patients have included an intra-aortic balloon pump¹ and percutaneous femoral-femoral bypass using an extracorporeal membrane oxygenator and a centrifugal pump.² The TandemHeart has advantages over these alternatives. The device clearly provides more reliable circulatory support than intra-aortic balloon counterpulsation.³ Percutaneous femoral-femoral bypass can provide similar circulatory support, but the oxygenator of the extracorporeal system limits the duration of use to 6 hours or less. Beyond 6 hours, there are predictable major complications that occur with ongoing use of the oxygenator. While the TandemHeart is recommended for 6 hours, it was used for up to 48 hours in this study, and even longer periods of support have been reported.
       The disadvantages of the TandemHeart include procedural complexity, vascular complications and high cost. An initial transseptal puncture is required. Interventional cardiologists who have not participated in radiofrequency ablations or septal closures may have limited experience performing transseptal punctures. Many operators will prefer intracardiac ultrasound guidance to advance a large 21 Fr catheter safely across the interatrial septum. Only a few interventional cardiologists are likely to be comfortable advancing a Brockenbrough needle under fluoroscopy alone. The 21 Fr venous cannula may cause a venous thrombus while implanted, or an iatrogenic atrial septal defect when removed. Similarly, the large arterial cannula may be associated with bleeding, traumatic arterial injury or limb ischemia. Finally, it should be recognized that there is a significant expense associated with this technology. The disposables for the TandemHeart now cost $15,000 per case. There is also an initial capital investment for the console, although that cost may be included in the disposables for institutions that perform a higher volume of cases.
       Despite these limitations, the TandemHeart may become a valuable resource in the cardiac catheterization laboratory. The authors are to be commended for their successful use of the TandemHeart in 5 very high-risk PCI patients. They emphasized that patients should be selected very cautiously, and suggested that the TandemHeart may reduce morbidity and mortality in these patients.
       Patient selection for the prophylactic use of the TandemHeart should be done cautiously, but it will remain difficult. Study patients had a left main equivalent and required a more “complex” or a prolonged PCI. They also had multiple medical comorbidities and other cardiac disorders that made their surgical mortality unacceptable. While the study patients did in fact require hemodynamic support, many similar unprotected left main patients are stable throughout their PCI.⁴ One might argue that the use of the TandemHeart in such patients is more difficult and costly than the unprotected left main stenting procedure itself. While experience with the device will make implantation more routine, the large arterial and venous cannulas will cause complications. Perhaps high-risk PCI patients should receive a provisional TandemHeart. A smaller 6 Fr catheter could be placed in the left atrium and a 5 Fr sheath inserted into the contralateral femoral artery. An iliac angiogram would be performed to insure that it was feasible to insert a 15 Fr arterial sheath. The TandemHeart itself would only be implanted in those who did not tolerate transient ischemia during the PCI.
       The authors acknowledged that it is too early to claim that the TandemHeart reduces mortality and morbidity for complex unprotected left main stenting cases. The study mortality rate was 20%. However, it was their impression that an intra-aortic balloon would not have provided adequate hemodynamic support. The TandemHeart allowed the PCI procedure to be completed in all study patients. Since the cause of mortality in this high-risk diverse patient population is often multifactorial, prevention of or resuscitation from shock may be the most important endpoints to evaluate. Further investigation of the TandemHeart in high-risk PCI patients, as well as those presenting with cardiogenic shock, is clearly warranted. Given the aging population and the favorable clinical results of drug-eluting stents for multivessel disease,5 PCI in patients with left main obstructive disease is likely to become increasingly common. The appropriate use of alternate technologies such as the TandemHeart and the percutaneous impella pump needs to be carefully evaluated. Large databases such as the American College of Cardiology National Cardiovascular Data Repository (ACC-NDR) will be critical for studying how these devices are used and for evaluating in-hospital outcomes. If usage becomes more common, appropriately powered randomized trials with both clinical and economic outcomes will be needed to demonstrate efficacy and insure that scarce resources are being well spent.

1. Aguirre FV, Kern MJ, Bach R, et al. Intraaortic balloon pump support during high risk coronary angioplasty. Cardiology 1994;84:175–186.
2. Ott RA, Mills TC, Tobis JM, et al. ECMO assisted angioplasty for cardiomyopathy patients with unstable angina. ASAIO Trans 1990;36:M483–M485.
3. 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. European Society of Cardiology, 2005.
4. Chieffo A, Stankovic G, Bonizzoni E, et al. Early and mid-term results of drug-eluting stent implantation in unprotected left main. Circulation 2005;111:791–795.
5. Serruys P, for the Arts ll Trial. American College of Cardiology 2005 Scientific Sessions. March 6–9, 2005,Orlando, Florida.