All trials of ad hoc PCI discussed below used registry data; randomized trials of ad hoc PCI have not been conducted.
Ad Hoc PCI in the Prestent Era
Ten studies[5-14] have reported the results of ad hoc PTCA. Of these, seven compared ad hoc with delayed PTCA, and all found similar overall angiographic success and acute complication rates for ad hoc and delayed PTCA (Table 1). Kimmel et al. reported that ad hoc angioplasty was associated with an increased risk of acute complications in patients with unstable angina, multivessel coronary artery disease (CAD), advanced age, and multilesion angioplasty. Overall, these studies provided evidence that ad hoc angioplasty, compared to delayed angioplasty, was safe in selected patients.
Table 1. Studies of Ad Hoc Versus Delayed Percutaneous Coronary Intervention
|O'Keefe et al. (1985–1986)||120/404||89||91||0||1.2||0.8||1.4||1.6||3.4|
|O'Keefe et al. (1984–1988)||73/5,351||95||95||0.5||0||0.9||0.5||2.3||0.5|
|Lund et al. (1991–1992)||124/?||92.1||88.4||NA||NA||NA||NA||NA||NA|
|Rozenman et al. (1989–1992)||1,719/2,069||93.9||92.9||0.8||1.3||1.0||1.3||0.5||0.3|
|Kimmel et al. (1992–1995)a||6,152/29,548||NA||NA||0.29||0.16||0.73||0.15||1.3||1.09|
|Le Feuvre et al. (1990–2000)||1,809/631||92c||88c||0.9||0.4||2.2||2.3||0.6||0.9|
|Panchamukhi and Flaker (1995–1996)||244/113||92||91||0||0||NA||NA||0.8||0|
|Stent era (>50% of PCI utilized stents)|
|Shubrooks et al. (1997)a||1,748/2,388||93.7b||93.6b||0.6||0.5||2.0||2.6||0.9||0.8|
|Goldstein et al. (1995–1998)a||38,411/23,462||NA||NA||0.46||0.56||NA||NA||NA||NA|
|Krone et al. (2001–2003)a||41,524/27,004||91.7c||92.5c||0.13||0.16||NA||NA||0.59c||0.34c|
|Feldman et al. (2001–2002)a||28,904/18,116||NA||NA||0.4||0.4||NA||NA||0.2||0.3|
|Hannan et al. (2003–2005)a||38,431/8,134||NA||NA||0.25c||0.45c||0.85||0.95||NA||NA|
|Good et al. (2004)a||557/23||97.7||100||0.7||0||3.8c||8.7c||0.4||0|
Ad Hoc PCI in the Stent Era
Shubrooks et al. reported the outcome of 4,136 PCIs performed in seven New England centers in 1997. Ad hoc PCI was performed in 42% of PCIs with similar clinical success and ischemic complication rates compared to delayed PCI. Vascular complication rates were lower in patients undergoing ad hoc PCI (0.6% vs. 1.6%, P = 0.006).
Goldstein et al. reported outcomes of 62,873 PCIs performed in 33 centers from 1995 to 1998, using data from the New York State Department of Health PCI database. Ad hoc PCI was performed in 62% of PCIs with similar mortality as delayed PCI overall but with increased risk of mortality in “high-risk” subgroups [i.e., those with congestive heart failure (odds ratio = 1.6; P = 0.04) or class IV angina (odds ratio = 1.6; P = 0.04)].
Feldman et al. reported the outcome of 47,020 patients undergoing PCI from 2000 to 2001 using data from the same New York State PCI Registry analyzed by Goldstein et al. years earlier. Ad hoc PCI was performed in 61%. Mortality, major adverse cardiac events (death, emergency CABG, or stroke), and incidence of renal failure/dialysis during hospitalization were similar for ad hoc and delayed PCI. Patients with high-risk features (age >80 years, class IV angina, congestive heart failure on admission, renal failure, and multivessel CAD) had similar in-hospital clinical outcomes after either treatment approach.
Krone et al. reported the outcomes of 68,528 PCIs with stable angina from 2001 to 2003 using data from the American College of Cardiology National Cardiovascular Data Registry. Ad hoc PCI was performed in 61%. While ad hoc PCI was associated with lower success rates, and slightly more frequent unplanned CABG and emergency repeat PCIs, the differences between ad hoc and delayed PCI became nonsignificant in a multivariate analysis. Procedural mortality, cerebrovascular events, and renal failure were similar between groups.
Hannan et al. reported outcomes of 46,565 PCIs between 2003 and 2005 using data from the New York PCI Reporting System. Ad hoc PCI was performed in 83% of PCIs. Adjusted in-hospital mortality rates were similar for ad hoc and delayed PCI. Ad hoc PCI was associated with lower rates of renal failure (0.07% vs. 0.14%) and myocardial infarction (MI) (0.85% vs. 0.95%) compared with delayed PCI, although P values were not reported. After 36 months of follow-up, ad hoc PCI was associated with lower mortality (risk-adjusted hazard ratio 0.76, P < 0.0001). The mortality reduction associated with ad hoc PCI was present in “high-risk” groups (women, age ≥ 75 years, multivessel disease, congestive heart failure, and class IV angina). The data did not explain why delayed PCI was associated with higher mortality overall and in high-risk subgroups; the authors suggested it could be due to increased morbidity associated with a second PCI procedure or unidentified biases in their data.
Good et al. reported the outcomes of 580 PCIs in 2004 from a single center. Ad hoc PCI was performed in 96% of PCIs. Delayed PCI patients were older with a higher frequency of prior MI, congestive heart failure, chronic kidney disease, left ventricular systolic dysfunction, and prior CABG. Outcomes were similar for both groups except for a higher incidence of periprocedural MI in the delayed PCI group (8.7% vs. 3.8%, P = 0.023).
Ad Hoc PCI in the Current Era
The prevalence of ad hoc PCI has increased over the past decade.[15-20] This increase is in part due to the proven efficacy of PCI (usually performed ad hoc) for acute coronary syndromes (ACS) and to studies suggesting that ad hoc PCI is safe and effective compared to delayed PCI.[15, 17, 19, 20]
The appropriateness of ad hoc PCI has been challenged recently, particularly for patients at either end of the spectrum of CAD—those with mild CAD in whom medical therapy might be sufficient and those with extensive and complex CAD for whom the relative benefit of PCI versus CABG has been questioned.[22, 23] For example, it has been suggested that, for patients with stable ischemic heart disease (SIHD), delay or deferral to discuss treatment options and to intensify medical therapy may be appropriate.[22, 24, 25] For patients with extensive (i.e., complex multivessel, or unprotected left main) CAD, it has been suggested that a “heart team” approach allowing input from both an interventional cardiologist and a cardiac surgeon may be preferable.[2, 23]