Intravascular imaging: a glass half empty or half full?

Graphical Abstract

More evidence supports the use of intravascular imaging in percutaneous coronary intervention. The glass is half full but there is more evidence to come. Abbreviation: PCI, percutaneous coronary intervention.

Open in new tabDownload slide

This editorial refers to ‘Intravascular imaging during percutaneous coronary intervention: temporal trends and clinical outcomes in the USA’, by R. Fazel et al., https://doi.org/10.1093/eurheartj/ehad430.

Without angiography, there is no percutaneous coronary intervention (PCI). However, angiography has several well-recognized limitations. For PCI strategy, angiography is imprecise in determining plaque morphology, vascular remodelling, and atherosclerosis burden. Likewise, for PCI optimization, angiography is limited in resolution to detect suboptimal stent expansion, malapposition, residual dissection, and plaque protrusion. These limitations may be overcome by intravascular imaging (IVI), which provides high-resolution cross-sectional tomographic imaging of the vessel wall. So why is a tool, with a one-off cost, that can overcome the limitations of angiography so infrequently used?

Perhaps the most common reason for lack of adoption of IVI is the notion that individual practitioner’s outcomes are already excellent with angiography-guided PCI. Indeed, while contemporary head-to-head randomized controlled trials comparing drug-eluting stents (DESs) report 1-year target vessel failure rates of ∼5%–6%,¹ longer term data suggest a persistent 1%–2% accrual of events year by year, with 5 year rates reaching ∼15%.² Another common misconception that impacts adoption of IVI is the belief that there is lack of clinical data to support IVI use. On the contrary, large observational cohort studies, randomized trials, and meta-analyses have shown that IVI-guided DES implantation reduces major adverse cardiovascular events (MACE), including target lesion revascularization, stent thrombosis, and cardiac mortality, compared with angiography guidance.³ Furthermore, registries⁴ and the recent RENOVATE-COMPLEX-PCI randomized controlled trial⁵ refute the idea that IVI is only beneficial in simple lesions, and on the contrary shows greater benefit in complex patients and lesions. So, while AHA/ACC and ESC PCI guidelines do not recommend IVI at the Ia level of evidence, there is support in the guidelines at level IIa, higher than the use of plaque modification for severely calcified lesions at IIb for example. Of course another major impediment to widespread adoption of IVI has been cost. Unfortunately there is a paucity of data on this topic, partly due to the large geographic variations in cost, but also due to lack of randomized datasets available to allow comparison. Nevertheless, recent data support the concept that IVI can be cost-effective, if not cost-saving, with even greater cost-benefit in complex lesions.⁶ With longer follow-up data showing a continued divergence of events favouring IVI over time,⁷ the health economics could be even more favourable than previously reported.

Undoubtedly the most precious resource we have in the catheterization laboratory is time. Correspondingly, the additional time required to activate, perform, and interpret IVI during PCI serves as a major impediment to routine clinical use. However, IVI studies with protocol-mandated pre-, intra-, and post-PCI imaging report procedure length to be extended by ∼15 min,⁸ which can be shortened to ∼6 min when integrated systems are used.⁹ One of the greatest obstacles to use of IVI in routine clinical practice has been difficulty with image interpretation and lack of a standardized workflow. While image interpretation on 20 MHz intravascular ultrasound (IVUS) may be simpler, it lacks the detail to detect post-PCI complications that have been associated with outcomes.⁸ On the contrary, high-definition IVUS and optical coherence tomography (OCT) have been criticized for providing ‘too much’ information, with clinicians struggling to determine what is relevant. Fortunately, in recent years, there have been considerable advances in teaching image interpretation at local and international conferences, as well as through periodicals.³ Moreover, standardized workflows such as the MLD MAX (Pre-PCI strategize Morphology, Length, Diameter, Post-PCI optimize Medial Dissection, Apposition, eXpansion) algorithm and the European Consensus Document have simplified which information in IVI is most clinically impactful along with how to practically apply it.¹⁰ These documents are also aimed to be universally applicable to all IVI, highlighting the benefits of IVUS and OCT in specific clinical scenarios, including complex PCI. Finally, safety, with concerns of ischaemia and dissection from catheter manipulation, has been raised as another barrier to widespread adoption of IVI. However, multiple studies have identified the risk of IVI to be very low. One study of >3500 consecutive studies adjudicated by an independent safety committee reported intra-procedural complications to be ∼0.5% with no MACE as a consequence of the procedure.¹¹

In this issue of the European Heart Journal, Fazel et al.¹² report an analysis of temporal trends and clinical outcomes of IVI in the USA. In a retrospective cohort study of Medicare beneficiaries treated with either inpatient or outpatient PCI, they analysed 1-year mortality and MACE [death, myocardial infarction (MI), repeat PCI, or coronary artery bypass grafting]. As these studies are often heavily influenced by selection bias, they used falsification endpoints of hospitalized pneumonia and hip fracture to assess for unmeasured confounding. Of the 1.2 million patients undergoing PCI during a 7-year period, 10.5% underwent IVI-guided PCI. Between 2013 and 2019, IVI use increased from 9.5% to 15.4%. IVI use was associated with lower mortality, MI, repeat PCI, and MACE, with no association of the falsification endpoints.

So is the glass half empty or half full? An absolute increase of ∼6%, and relative increase of 62% in 7 years in the absence of class Ia societal guidelines is impressive growth. With real-world large dataset analyses such as those reported by Fazel et al., coupled with the upcoming ILUMIEN IV (NCT# 3507777), OCTOBER (NCT #3171311), OCTIVUS (NCT #3394079), IMPROVE (NCT# 4221815), and IVUS-CHIP (NCT #4854070) randomized trials, the glass certainly seems half full. Sadly, interventional cardiology behaviour suggests otherwise. While we all set out with evidence-based patient-centred intentions aimed at reducing MACE, as soon as we gown lead, our decision-making often changes to reducing procedural risk. So, as the clock ticks 5 pm with five patients waiting one ponders: I can probably obtain an angiographically suitable result; there is no real consequence to not using IVI; the device and software are complex to set up and practically utilize; and the information made available is not so easy to interpret. This behaviour is reflected in the current report. The median IVI use was a disappointing 3.9%, operator variation was high with an odds ratio of 3.4, and ∼25% of operators never used IVI over a 7-year period. So in the presence of overwhelming data to support IVI use, how do we move the needle? The answer is by recognizing that when technology is adjunctive, adoption is driven by two predominant factors—fear and finance.

So how does fear influence change? As a corollary, the major inflection point in the adoption of intracoronary physiology above and beyond randomized controlled evidence or integration into the guidelines was the institution of the appropriate use criteria by the Centers for Medicare & Medicaid Services (CMS). Intracoronary physiology use increased 3.8-fold in 5 years, reaching ∼40% of indicated lesions in 2014.¹³ Fear of litigation, removal of hospital privileges, and withholding of payment by CMS and private payers were driving reasons for increased adoption of intracoronary physiology. While performance and interpretation of intracoronary physiology is relatively simple, in contrast, implementation of IVI can be challenging. Fear of embarrassment from being unfamiliar with the technology, unable to interpret images, or apply action based on findings is a major impediment to IVI adoption. In a recent experiment, we evaluated interventional cardiology fellows’ self-perceived training sufficiency vs. core competency-based preparedness for independent practice for IVUS. Nearly 80% of fellows reported themselves as expert or sufficiently trained in IVUS, whereas an abysmal 15% had independence in core competencies when they were tested.¹⁴

They say money makes the world go round. This is indeed true with regards to IVI. Stark differences in adoption exist across geographies. In a recent societal survey, IVI was used in >95% of procedures in Japan, but only in 10.4% in Europe.¹⁵ In Japan, procedural reimbursement is tied to IVI use, which is a quality metric. Marked variations in the cost of catheters is another major issue, for example in Asia-Pacific, where third-party distribution can create a substantial cost upcharge. Finally, in fee-for-service systems such as the USA, the minimal physician reimbursement for IVI use is another impediment.

It is of course important to understand the limitations of the report by Fazel et al. Critically, the selected falsification endpoints do not take into account procedural and lesion characteristics, leaving the possibility that unmeasured confounders impact the results. Likewise, it is not possible to determine how the IVI was used and in what vessel. Further, there is no comparison of IVUS vs. OCT or data on pharmacological therapy. Finally, the analysis is limited to Medicare patients.

It appears that the burden of evidence in favour of IVI-guided PCI is reaching critical mass, especially with multiple major randomized controlled trials planned for presentation this year. In this regard, the study by Fazel et al., highlighting the impact of IVI in the real world, makes the glass at least half full.

Declarations

Disclosure of Interest

Z.A.A. reports institutional research grants to St Francis Hospital from Abbott, Philips, Boston Scientific, Abiomed, Acist Medical, Medtronic, Cardiovascular Systems Inc., Shockwave, Chiesi, Teleflex, and Opsens; being a consultant to AstraZeneca, Abiomed, Boston Scientific, and Cathworks Philips; and having equity in Shockwave Medical. D.S. reports nothing to disclose.

References

Author notes