Uncomplicated TLR After PCI Linked With Worse Mortality

Treating nonemergent, uncomplicated restenosis after PCI is associated with an increased risk of mortality, suggesting that target lesion revascularization may not be an “entirely benign” procedure, according to the results of a new meta-analysis.  

In a pooled analysis of 21 randomized trials that included 32,882 patients, TLR was an independent predictor of all-cause mortality after adjusting for potential confounding variables (HR 1.23; 95% CI 1.04-1.45).

“Even so-called uncomplicated TLR procedures, those that were not associated with an acute coronary syndrome and those in which the procedures themselves are not complicated, are still associated with an increased risk of long-term mortality,” senior investigator Gregg Stone, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), told TCTMD. “So, the implications would be that we need to add TLR to a list of complications which can occur after coronary stenting, and that can contribute to worse long-term prognosis to patients.”

In an editorial, Harold Dauerman, MD (University of Vermont Medical Center, Burlington), writes that the “underlying premise of this pooled analysis is that TLR is not a nuisance but a real risk for increased mortality.” Moreover, given the improved understanding from clinical, pathological, and imaging studies, which suggest that stent restenosis might share high-risk features of de novo lesions, it’s plausible that TLR even in uncomplicated patients is not a harmless procedure. 

We need to add TLR to a list of complications which can occur after coronary stenting, and that can contribute to worse long-term prognosis to patients. Gregg Stone

Davide Capodanno, MD, PhD (University of Catania, Italy), who was not involved in the study, said the prognostic value of TLR has been debated in the past, particularly when it has been combined with cardiac death and target-vessel MI in the broadly used endpoint of target lesion failure. Some have argued that the TLF composite can be unfair to PCI in head-to-head comparisons versus CABG surgery because hard endpoints (death, MI) are mixed with the “softer” TLR.

“I think this study reinforces the value of TLF as a device-oriented endpoint of stent performance, because all its components are clinically meaningful,” Capodanno told TCTMD. “In clinical practice, we do our best to optimize PCI and avoid restenosis, which is not a ‘piece of cake’ anyway for even the uncomplicated patient who has to come back to the hospital and reassume a [dual antiplatelet therapy (DAPT)] regimen.”

Approximate 25% Higher Risk of Death

In the meta-analysis, which was published April 18, 2018, in JACC: Cardiovascular Interventions, 7.2% of patients underwent a nonemergent, uncomplicated TLR procedure at a median of 271 days after PCI. Comparatively, 2.5% of patients underwent nonemergent, uncomplicated non-TLR revascularization of the target vessel at a median of 537 days after PCI. Patients undergoing TLR were younger, more likely to be female, and more likely to have diabetes and other comorbidities compared with patients without target vessel revascularization.

During follow-up, 1,739 patients without target vessel revascularization died (incidence 1.87 per 100 patient-years). Comparatively, 144 who underwent nonemergent, uncomplicated TLR and 49 who underwent non-TLR target vessel revascularization died (incidence 2.45 and 2.67 per 100 patient-years, respectively).

After adjusting for confounders, nonemergent TLR was associated with an increased risk of death whereas non-TLR target vessel revascularization was not. Among patients with TLR, those who had an MI after the revascularization procedure had a significantly increased risk of death compared with those without MI. Overall, MI occurring during follow-up after TLR was associated with a nearly fourfold higher risk of death (HR 3.82; 95% CI 2.44-5.99).

The researchers state that most of the included studies did not mandate routine angiographic follow-up and, as a result, could not determine the impact of restenosis alone on the risk of death. Additionally, the studies could not determine whether there was a correlation between restenosis severity and mortality.

To TCTMD, Stone said the risks associated with TLR after PCI might not be fully appreciated by all physicians. While it’s on their radar, it’s a remote, distant concern. He noted that their analysis excluded TLR presenting as MI, as well as TLR complicated by an MI after the procedure, so the observed mortality risk is likely underestimated.

How Does TLR Confer Risk?

As to why uncomplicated TLR poses a long-term risk of mortality, Stone said the mechanisms are uncertain. Patients with TLR might be more likely to develop MI in the future, which might be due to the TLR procedure, or it might be the result of disrupting other plaques within the target vessel. Additionally, it may be that patients undergoing TLR are more prone to developing aggressive atherosclerosis and progressive disease.

Capodanno said TLR might increase the risk of death due to the small but inherent risks of repeat PCI and restarting DAPT, or the risk arising from complications such as MI and stent thrombosis. “On the other hand, having TLR may be a marker of being a more complex patient, clinically and/or anatomically, which is something difficult to adjust for, particularly in the presence of potentially unidentified confounders,” he said.  

I’ve always had the feeling that a low ejection fraction predisposes to restenosis. Robert Byrne

Robert Byrne, MBBCh, PhD (Deutsches Herzzentrum München, Germany), who was not involved in the study, conducted a similar analysis in 2014 along with lead investigator Salvatore Cassese, MD (Deutsches Herzzentrum München). In 10,004 patients who underwent successful PCI at two Munich centers, the presence of restenosis on coronary angiography 6 to 8 months after the procedure was associated with a similarly increased risk of mortality at 4 years.

The newest meta-analysis excluded patients who had a periprocedural MI as well as patients who died within 24 hours of TLR, which was done to include only patients stable at presentation. But Byrne said downstream factors, such as changes in kidney function or the use of DAPT that resulted in a bleeding event, might have contributed to the observed mortality risk.

“Or it could be that the reason they have restenosis is due to other factors that aren’t captured,” he said. “For example, we don’t know what the ejection fraction was in the patients with and without restenosis. It might be that the patients with restenosis had worse ejection fraction and it wasn’t accounted for in the multivariable adjustment. I’ve always had the feeling that a low ejection fraction predisposes to restenosis.”

Such limitations are inherent to analyses such as this one, as well as their own, said Byrne.

Minimizing the Risk of TLR 

Stone suggested that best-practice procedures and techniques, as well as contemporary devices, used to further lower the rate of angiographic restenosis and the need for TLR procedures could improve the long-term prognosis of patients. For example, he suggested the use of image-guided PCI to ensure the best stent implantation possible, appropriately pretreating certain lesions (such as calcified lesions that are associated with an increased risk of TLR), and lesion preparation with cutting balloons or atherectomy when appropriate.

Additionally, he recommends the use of contemporary DES. In the meta-analysis, which includes trials using BMS as well as first- and second-generation DES, use of current DES, particularly everolimus-eluting stents, was associated with lower rates of TLR. “I do think interventionalists can do a lot by using the techniques and technologies to minimize the risk of TLR,” he said.

In the editorial, Dauerman asks whether TLR should be a target for quality improvement. It’s an attractive target, he writes, given that it’s modifiable with technology, such as better DES. He points out, however, that before efforts are put into reducing TLR with newer stent technology, it’s possible that the incidence may be too low to study in clinical trials. The pooled analysis suggested TLR occurs in 7.2% of patients, which is consequential, but the trials included spanned decades, with nearly 20% of patients receiving a bare-metal stent.

“Thus, the pursuit of any superior stent product with respect to TLR (whether metallic or bioresorbable) will require creative study designs or enriched patient groups with enhanced TLR rates in order to avoid the ambiguity of noninferiority trials and magical endpoints,” writes Dauerman.

Note: Stone and several co-authors of the study are faculty members of the Cardiovascular Research Foundation, the publisher of TCTMD.