Putting Ad Hoc PCI on Pause

Abstract

AD HOC IS A LATIN PHRASE, LITERALLY MEANING “FOR this.” It typically implies a solution designed for a particular problem or task—that is, something not broadly generalizable. When the term was introduced more than 2 decades ago to cardiologists in the context of percutaneous coronary intervention (PCI), it suggested the unique circumstances that justified combining coronary angiography and PCI into the same setting. This approach was uncommon in the early years of PCI when most procedures were performed days to weeks after the initial diagnostic test. Initially, the hesitation to perform ad hoc PCI resulted from technical challenges. Older imaging systems limited the ability of cardiologists to rapidly and accurately review angiograms. In addition, operating rooms and cardiac surgeons had to be on stand-by whenever PCI was performed because of the threat of acute closure of the coronary artery and other life-threatening complications. Over the years, these issues have largely been resolved, however, by advances in imaging systems and procedural techniques, such as the introduction of coronary stents. As a result, ad hoc PCI is now routinely performed, and its use has increased considerably in recent years. In New York state, for example, more than 80% of PCIs are currently performed ad hoc and, in many cases, this makes perfect sense. During acute coronary syndromes, including STelevation myocardial infarction, the ability to perform PCI immediately after coronary angiography saves lives and improves outcomes. In elective cases, ad hoc PCI avoids the cost and inconvenience associated with a second procedure and arterial puncture. Yet now that it is clear that ad hoc PCI can be performed, a more fundamental question arises: Should it? Recent clinical trials, such as COURAGE and SYNTAX, have fueled concerns about the widespread use of PCI. For many patients with stable coronary disease, PCI does not improve survival when compared with optimal medical therapy. For patients with more severe disease for whom coronary artery bypass grafting is considered, PCI does as well as surgery for 1-year survival, but leads to more repeat procedures and the need for prolonged antiplatelet medications. Although findings from these trials are largely undisputed, the use of ad hoc PCI raises challenges for incorporating their lessons into routine practice. Specifically, ad hoc PCI provides less opportunity for patients and their physicians to thoughtfully consider a range of clinically equivalent treatment options after the coronary anatomy is known. By making the diagnostic test an automatic gateway to the therapeutic procedure, pressure is placed on the cardiologist and patient to make immediate decisions that may favor PCI even in elective settings. In one study, for instance, patients with stronger indications for coronary artery bypass grafting were more likely to be recommended surgery after coronary angiography when it was performed at hospitals without the ability to perform ad hoc PCI. At the other extreme are ongoing concerns about how frequently PCI is performed when medical therapy appears suitable. Part of this results from the well-described “oculo-stenotic” reflex, ie, the tendency to treat blockages, even when clinically silent, based on benefits attributed to PCI that are not supported by the literature. There are other issues to consider as well. In their seminal article, Myler and colleagues highlighted 2 specific challenges with ad hoc PCI. First, they noted that obtaining adequate informed consent for ad hoc PCI may be difficult. Patients typically receive conscious sedation during coronary angiography, prompting most centers to adopt an approach in which risks of PCI are simultaneously discussed before the coronary anatomy is known. However, risks of PCI differ substantially from the diagnostic test and may vary from patient to patient based on lesion characteristics and location. A generic consent obtained prior to coronary angiography fails to capture these subtleties. In addition, Myler et al pointed out possible concerns about the ethics of self-referral. At its core, this issue highlights the potential hazards of merging diagnostic tests (ie, coronary angiography) and therapeutic procedures (ie, PCI) into the same setting. If the cardiologist who performs and interprets the coronary angiogram will also perform the PCI,