Author + information
- Published online November 15, 2017.
- Gary S. Mintz, MD∗ ()
- ↵∗Address for correspondence:
Dr. Gary S. Mintz, Cardiovascular Research Foundation, 1700 Broadway, Ninth Floor, New York, New York 10019.
It is the 40th anniversary of percutaneous coronary intervention (PCI), 30 years since the first transmural intravascular images of the coronary arteries using grayscale intravascular ultrasound (IVUS), and 20 years since the first optical coherence tomographic images. Although they were initially used for research purposes, and although the contributions made to understanding atherosclerosis, mechanisms of successful (and failed) catheter-based treatments, and restenosis cannot be overestimated, IVUS and optical coherence tomography (OCT) are now used to answer questions that occur during routine PCI (1): 1) Is this left main stenosis significant? 2) Which (and what) is the culprit lesion? 3) Is this a high-risk plaque or patient? 4) What is the likelihood of distal embolization or periprocedural myocardial infarction during stent implantation? 5) How do I optimize acute stent results? 6) Why did this stent thrombose or restenose? However, as addressed in this issue of JACC: Cardiovascular Interventions by Buccheri et al. (2), the major clinical utility of intravascular imaging is to optimize stent implantation.
In their Bayesian network meta-analysis of 17,882 patients treated with either bare-metal or drug-eluting stents from 17 randomized clinical trials (RCTs) and 14 registries that used matching algorithms for statistical adjustment, Buccheri et al. show that 1) IVUS significantly reduced all-cause mortality (the primary outcome) compared with angiography alone; 2) IVUS or optical coherence tomographic guidance was associated with significant and consistent reductions of major adverse cardiac events and cardiovascular mortality compared with angiographic guidance; 3) benefits of IVUS guidance were also statistically significant for myocardial infarction, stent thrombosis, and repeat revascularization; 4) using rank probability analysis, coronary angiography was rated as the worst guidance strategy; 5) there were no efficacy differences between IVUS and OCT; and 6) results were consistent across multiple sensitivity analyses including just drug-eluting stent implantation.
Although the most statistically sophisticated, this is merely the latest in a series of similar meta-analyses beginning in the bare-metal stent era (3) and continuing with drug-eluting stent implantation (4), with 3 others just recently published, one including 8 RCTs (5), one focusing on complex lesions (6), and the third addressing left main coronary artery stenting with mortality as the primary endpoint (7). Each of these many meta-analyses differed in terms of study inclusion criteria, and each used a different combination (but not all) of the many published RCTs and registries (4); however, the principal findings—significant reductions in major adverse cardiac events and hard clinical endpoints of death, cardiac death, myocardial infarction, and stent thrombosis—were consistent with those of the present study. This is not to say that all stent implantation procedures should be intravascular imaging guided; the greatest benefits were in high-risk or complex lesions and/or patients (4).
The elephant in the room is, of course, the fact that intravascular imaging is used infrequently outside of Japan, notably in Europe, where Buccheri et al.’s study was done; and outside of Japan, use statistics are skewed by a few high-volume institutions. Given the available data, this seems inexplicable. What are the main excuses and explanations?
First, no (or insufficient) data. This is again debunked in the current meta-analysis (2).
Second, lack of a large-scale, definitive RCT. This dismisses IVUS-XPL, in which 1,400 patients with long lesions treated with a single type of stent (≥28-mm XIENCE V) were randomized to IVUS versus angiographic guidance (8).
Third, cost. Alberti et al. (9) reported that IVUS guidance was dominant and cost effective, especially in patients with comorbid conditions who were at a higher risk for cardiac events and was cost saving when the benefits continued beyond 1 year.
Fourth, safety. Although 1 study suggested that IVUS use was associated with more periprocedural MIs (10), this was not shown in the comprehensive meta-analysis by Ahn et al. (11). Recently, van der Sijde et al. (12) reported 1,142 procedures using OCT and 2,476 procedures using IVUS. Imaging-related complications were rare, did not differ between the 2 methods, and were self-limiting after retrieval of the imaging catheter or easily treatable with no major adverse events, hospital stay prolongation, or permanent harm compared with procedures in which imaging was not performed. Common sense is always important. And to the contrary, in patients with pre-existing renal insufficiency, IVUS-guided stent implantation can be done without the use of angiographic contrast (13).
Fifth, angiography is sufficient. However, angiography has known limitations in assessing vessel size and plaque burden, lesion calcium and eccentricity, and stent expansion and geographic miss and complications. In the present network meta-analysis, coronary angiographic guidance was rated as the worst of the 3 strategies (2).
Sixth, lack of a guideline Class I indication. Hopefully, recent RCTs such as IVUS-XPL (8) and meta-analyses such as the present study by Buccheri et al. (2) will convince those writing the guideline to change this.
Finally, education. A recent TCTMD poll asked finishing or recently graduated interventional fellows to detail training in IVUS or OCT as “none,” “rudimentary,” “sufficient to perform the procedures, make measurements, and use the information clinically,” or “expert.” Overall, 52% reported no or rudimentary education in IVUS, and 68% reported no or rudimentary education in OCT.
It seems that most interventional cardiologists do not know the research, do not believe or choose to ignore the data, or do not know how to use IVUS and/or OCT, and most training directors do not consider intravascular imaging to be an important part of the PCI curriculum.
It is not necessary to be expert at both IVUS or OCT; the 2 are equivalent in guiding and optimizing most stent implantation procedures. However, a modern coronary interventionalist should be able (or be trained) to properly perform, interpret, and use at least 1 of these 2 technologies. An alternative strategy that has been successful at several institutions such as New York-Presbyterian Hospital/Columbia University Medical Center, Washington Hospital Center, and Erasmus MC’s Thoraxcentre is to train dedicated intravascular imagers, often technologists, who can interpret images, make measurements, provide information to the operator, and troubleshoot as necessary. This is analogous to the function of highly trained technologists in an echocardiography laboratory.
Finally, even using intravascular imaging, we can we still do better. For example, in IVUS-XPL, imaging-guided procedures that met the optimization endpoint (minimum stent area greater than the distal reference lumen) had a 1-year major adverse cardiac event rate of 1.5%; however, only one-half met this endpoint (8). It is not clear whether images were interpreted incorrectly, careful measurements were not made or used to control the procedure (i.e., mere use vs. actual guidance), it was impossible to achieve optimal stent expansion despite high-pressure post-dilation, noncompliant lesions went unrecognized and were not prepared with 1 of the available plaque modification devices, or the necessary tools do not exist to optimize stent implantation procedures. We can only speculate the impact if intravascular imaging were used routinely beginning with pre-PCI imaging to assess plaque characteristics (i.e., calcification) and select the best stent size and length and post-PCI to control when the procedure is or is not finished (i.e., truly optimized).
The purpose of any study is not just to add lines to curricula vitae but to nudge clinical practice in the direction indicated by the evidence. Time will tell whether the meta-analysis by Buccheri et al. (2) convinces some naysayers, only supports those who already believe the existing data, or is merely added to many RCTs, registries, and meta-analyses showing the superiority of IVUS and/or OCT versus angiographic guidance but that are then ignored.
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
The Cardiovascular Research Foundation receives grant or fellowship support from Boston Scientific, Volcano, and St. Jude Medical. Dr. Mintz is a consultant to or receives honoraria from Boston Scientific, Volcano, ACIST, and Infraredx.
- 2017 American College of Cardiology Foundation
- Mintz G.S.,
- Guagliumi G.
- Buccheri S.,
- Franchina G.,
- Romano S.,
- et al.
- Mintz G.S.
- Bavishi C.,
- Sardar P.,
- Chatterjee S.,
- et al.
- Fan Z.G.,
- Gao X.F.,
- Li X.B.,
- et al.
- Ye Y.,
- Yang M.,
- Zhang S.,
- Zeng Y.
- Ahn J.M.,
- Kang S.J.,
- Yoon S.H.,
- et al.
- van der Sijde J.N.,
- Karanasos A.,
- van Ditzhuijzen N.S.,
- et al.