OCT Reveals Potential Causes of Very Late Stent Thrombosis

A new OCT study is shedding light on the potential roots of very late stent thrombosis after implantation of early- and new-generation DES, finding that one-third of cases likely arose from strut malapposition and nearly 3 in 10 from neoatherosclerosis. Uncovered stent struts—long assumed to be an important contributor to late thrombotic risk—were less common than the other factors in this series.

Take Home: OCT Reveals Potential Causes of Very Late Stent Thrombosis

“An important message from the paper is that nowadays, in the cath lab, we can unravel the reasons or the causes underlying very late stent thrombosis, or stent thrombosis in general, by using optical coherence tomography,” senior author Lorenz Räber, MD, PhD, of Bern University Hospital in Switzerland, told TCTMD.

The paper was published early online this week in Circulation and according to Räber is the largest study to date using OCT to investigate very late stent thrombosis.

Four Factors Identified

For their study, Räber and colleagues used OCT to examine 64 patients who developed very late stent thrombosis a mean of 5 years after their index procedures. Four European centers participated in the study. OCT pullbacks were performed after restoration of flow and analyzed at 0.4 mm intervals. In all, imaging results for 38 early- and 20 newer-generation DES were available for analysis.

A “putative cause” for the events was identified using OCT for 98% of cases, including strut malapposition (34.5%), neoatherosclerosis (27.6%), uncovered struts (12.1%), and stent underexpansion (6.9%).

These findings open the door for interventionalists to better understand how altering the initial procedure might reduce long-term risk, Räber suggested. For instance, both malapposition and underexpansion may be addressed by post-dilatation.

The idea that PCI optimization can have such long-lasting impact has not been fully appreciated, he said. “It was more [thought] that probably this is a course that you cannot impact upon [but instead is] a natural consequence. It seems at least that by perfecting results, you can potentially attenuate the risk of very late stent thrombosis.”

Another way these OCT results might alter care, he said, is that they highlight the role of neoatherosclerosis. Earlier research done by Räber and colleagues has shown that, “if a patient has progressive coronary artery disease in previously untreated lesions, he is also more prone to develop [in-stent] neoatherosclerosis. And therefore probably optimal medical treatment with high-dose statin therapy, etc, may attenuate the risk of neoatherosclerosis and thereby [reduce] the risk of very late stent thrombosis,” he proposed.

But while it is important to understand the underlying mechanisms of very late stent thrombosis, the current study does not speak to whether OCT should be performed in individual patients and how it should affect their management, Räber noted.

How Much Does Malapposition Matter?

The relevance of malapposition and whether or not it should be corrected immediately after PCI has been debated in the past, Räber said. Data are conflicting, and there are no sizeable RCTs addressing the question. Indeed, uncovered struts have typically been viewed as a more important contributor to very late stent thrombosis than incomplete stent apposition—an assumption not borne out in the current study.

Commenting on the study for TCTMD via email, Gary S. Mintz, MD, of the Cardiovascular Research Foundation (New York, NY), pointed out that not all of the malapposition identified in Räber’s analysis represents an opportunity for technical improvements during the initial procedure.

“There is a distinction between acute malapposition, which can be corrected at the time of PCI, and late stent malapposition that can either persist from the time of implantation (ie, acute malapposition that is detectable by OCT in half of the DES post-PCI, but that resolves in 70% of cases) or malapposition that develops during the follow-up period,” Mintz said. Studies such as the one by Räber only assess late malapposition without making the critical distinction between persistent malapposition and late acquired malapposition, he added.

Mintz points out, however, that there are more than a half-dozen large studies, each with more than 500 patients, showing no relationship between acute malapposition and very late stent thrombosis.

To better understand characteristics of malapposition that might affect decision making, Räber et al compared thrombosed with non-thrombosed regions. Uncovered and malapposed struts were more common in the thrombosed regions (P < .001), a pattern that was consistent when looking separately at both early- and newer-generation DES. While the maximal length of malapposed or uncovered struts also was greater in thrombosed compared with non-thrombosed regions (3.40 mm vs 1.29 mm; P < .001), the maximal or average axial malapposition distance did not differ.

Based on confidence intervals for the maximum length of malapposition and uncovered stents, Räber et al showed that malapposed segments greater than .99 mm and uncovered stent segments greater than 1.63 mm were associated with very late thrombus formation, the paper notes.

In particular, the 1-mm cutoff for malapposition might prove useful when operators are deciding whether to further optimize a given PCI. “If it’s below, you might not correct it,” Räber said, stressing that this implication is “purely hypothetical” and that many factors must come together for thrombosis to occur.

“I’m by no means saying that all the regions that are longer than 1 mm will subsequently lead to thrombosis. I would never say that, because certainly many, many patients live their entire life with malapposition,” he emphasized. “But I am saying that, at least in this cohort, we observed that if you have a thrombus, the malapposition in this thrombus is almost always longer than 1 mm.”

Taniwaki M, Radu MD, Zaugg S, et al. Mechanisms of very late drug-eluting stent thrombosis assessed by optical coherence tomography. Circulation. 2016;Epub ahead of print.

Related Stories:


  • Räber reports receiving speaker fees and research support from St. Jude Medical.
  • Mintz reports that CRF receives research or fellowship grant support from Boston Scientific, InfraReDx, and St Jude and that he serves a consultant or member of the speaker’s bureau for ACIST, Boston Scientific, InfraReDx, St Jude, and Volcano.

We Recommend