This series from Mount Sinai Hospital extends the reported experience with prophylactic deployment of a percutaneous transseptal cardiopulmonary bypass system during high-risk elective coronary angioplasty. These highly experienced operators employed a dual-“preclose” technique in the majority of patients, implanting arterial sutures before pump deployment that permitted immediate postprocedure arterial closure. The short-term outcome was excellent, reported complications minimal, and intermediate-term outcomes reasonable for the high-risk patient cohort described.
How should this report influence clinical practice? Temporary cardiopulmonary support, ranging from intra-aortic balloon counterpulsation to full oxygenator pumps for cardiopulmonary bypass, has long been part of the interventional toolbox.¹ Different bypass systems are commercially available, including a transseptal left atrial inflow system (TandemHeart^®, CardiacAssist, Inc., Pittsburgh, Pennsylvania) described here, including a retrograde transaortic pump (Impella^®, Abiomed, Inc., Danvers, Massachusetts) that is investigational in the United States,² and including “conventional” right atrial and distal aortic percutaneous bypass (cardiopulmonary bypass, CPS) with or without left ventricular decompression.³ Each has advantages and disadvantages. The TandemHeart does not require an oxygenator, but requires operators to be skilled in transseptal access and is perhaps most difficult to deploy in emergencies. The Impella requires only arterial access, but requires transaortic access to the left ventricle. Peripheral CPS requires arterial and venous access, but is straightforward to deploy during cardiopulmonary resuscitation.⁴ Prophylactic application of these aggressive support systems has waned at many clinical centers as coronary interventional tools have matured, especially in the form of highly-deliverable stents.
Some patients need cardiopulmonary support because they cannot tolerate the incremental myocardial ischemia and dysfunction imposed by the interventional procedure. In some, the ischemia is transient and reverses quickly. In others, the ischemia may be more sustained, for example, during intentional (ablative) or unintentional athero- or thromboembolism. In still others, an unexpected catastrophe requires temporary support to avert death pending surgical or percutaneous rescue. However, intolerable ischemia and cardiopulmonary collapse remain relatively infrequent, even during elective high-risk percutaneous coronary intervention.
Support can have other “soft” benefits, specifically reducing operator stress that might otherwise detract from optimal conduct of complex procedures. Transient iatrogenic myocardial dysfunction causing pain, pulmonary edema, ventricular arrhythmia and hypoperfusion syndromes might lead to procedure compromises, such as lesscomplete revascularization that might detract from long-term outcome.
Support can have important disadvantages such as complications of large-caliber devices introduced into diseased peripheral arteries, of left atrial access, if attempted, of pump induction of a systemic inflammatory state via activation of complement and other injurious humeral factors, of high-dose antithrombins required by the bypass pump, of postprocedure hemostasis, and of the high cost and complexity of the disposable pump equipment.
Retrospective comparisons of prophylactic and rescue support using an intra-aortic balloon pump (IABP) show a worse outcome for patients requiring bailout IABP during elective percutaneous coronary intervention.^5,6 These observations do not imply causality.
In this report, Dr. Rajdev and colleagues reassure us that in talented hands and selected patients, the risks of more aggressive prophylactic support appear acceptable in the short term using a transseptal bypass system. This represents a remarkable evolution from the early registry experience with cardiopulmonary support when complications were frequent.³ However, reports like this are always difficult to interpret. In such talented hands, how many of these patients really needed hemodynamic support?
Are there better tools available to identify in advance patients who need support? Which is the “best” support device for elective or emergency procedures? Is direct left atrial access really necessary, or would “ordinary” cardiopulmonary support be satisfactory? Is the risk of transseptal puncture and implantation of a large transseptal catheter justified in an elective setting? Is the incremental bleeding risk acceptable in light of the higher requisite antithrombin treatment? Can such tools be applied with comparable success and safety by less-experienced operators? Is there a “ standby” mode that can be employed, such as prophylacticvascular or even transseptal access? Is standby a satisfactory alternative, or does it waste precious support time that might otherwise have been lifesaving? How many patients need to be treated in order for one life-threatening complication to be averted?
Unfortunately, we are left with more questions than answers. Will clinicians ever have sufficient equanimity to answer these questions in a clinical experiment?