Editor's Corner: Practice-Changing Clinical Trial Results at TCT 2016? Not Exactly (All Save One) | Peter C. Block

CardioSource WorldNews Interventions | This year’s annual Transcatheter Therapeutics Conference (TCT) featured a host of late breaking clinical trials that promised to change clinical practice. Conference-goers looked with anticipation to answers for a lot of clinical questions, of which I have chosen these (arguably) important ones. The answer to one will indeed change clinical practice.:

  1. Will the EXCEL and NOBLE Trials tell us whether surgery or stenting should be chosen for treatment of left main disease?
  2. Will bioresorbable scaffolds live up to their hype and promise and soon replace current non-resorbable stents?
  3. Is Impella, rather than intra-aortic balloon support, the proper choice for interventionalists treating patients with cardiogenic shock?
  4. Can an “old” trial (RESPECT) finally give us an answer to whether PFO closure is clinically beneficial?
  5. Will the EXCEL and NOBLE Trials tell us whether surgery or stenting should be chosen for treatment of left main disease?

The EXCEL and NOBLE Trials were the most talked about at TCT. Both randomized trials compared surgery to stenting for unprotected left main disease and left both cardiac surgeons and interventionalists happy because the results were conflicting (or so it appeared). EXCEL randomized patients with low- or intermediate –SYNTAX scores to treatment with a Xience stent or CABG. Patients were followed for 3 years. Death, stroke, and myocardial infarction (MI) were comparable in the two groups. Interestingly there were more periprocedural MI’s and 30 day ST-elevation MI’s in the surgery group. The NOBLE Trial followed patients to 5 years (perhaps an important difference). In the NOBLE Trial patients received either a Biomatrix biolimus eluting stent or CABG surgery. All-cause mortality was similar in the two groups at 5 years, but MACE (specifically non-procedural MI) and (strangely) strokes were significantly higher in the stented group. Interpretation of these apparently contradictory outcomes for the two trials then became the focus of many conversations. EXCEL (~1900 patients) clearly showed that PCI is an acceptable alternative to surgery, and that the risk of MI appears lower in the PCI group (at least out to 3 years). NOBLE (~1200 patients) did not tally periprocedural MI, but the longer follow-up in NOBLE may indicate that late MI and more repeat coronary revascularization in the PCI group indeed may make surgery a more favorable alternative. Once the trials were dissected it became clear that neither was without its flaws. EXCEL did not have the same long term follow up as NOBLE, and if periprocedural MI’s were taken out of the surgical group, later follow up would be similar to the NOBLE outcomes. Some NOBLE patients (~10%) received a first generation stent before the trial switched over to a modern stent, and stent thrombosis was more prevalent in NOBLE patients compared to EXCEL patients. In addition, in NOBLE the late stroke rate in the PCI group was unexpected. Many simply passed that off as a play of chance.

The two trials offer lots of controversy but no clear answer as to which strategy is superior. However, we do have more information to share with our patients faced with this decision. PCI is certainly a valid alternative to surgery for low or intermediate risk patients with left main disease. Early results appear to favor PCI. However, if outcomes longer than 3 years are part of the conversation, it appears that surgery still plays an important role in therapy. The bottom line is really not a lot different from what many of us are currently advocating. If the left main disease is focal and easily treated with stenting, PCI may be a good choice. If the left main disease is complex and at a bifurcation, and other more distal disease is present, surgery, with its good long term results, may be a better choice.

Will bioresorbable scaffolds live up to their hype and promise and soon replace current non-resorbable stents?

Two studies presented at TCT reported on longer term outcomes of bioresorbable scaffolds compared to the Xience V everolimus-eluting stent – ABSORB II and ABSORB CHINA. ABSORB II reported on 3-year outcomes which demonstrated greater late lumen loss compared with Xience and surprisingly also showed worse vasomotion in the bioresorbable scaffold patients compared to the supposedly more “rigid” metal stent. In addition, vasomotor reactivity was also worse in the bioresorbable scaffold, (0.047 mm for Absorb vs 0.056 mm for Xience, p = 0.49 for superiority). Interestingly, one of the primary endpoints of the trial was vasomotion which was expected to be absent in the Xience arm, and explanations for the unanticipated outcome are still lacking. More importantly - angiographic late lumen loss was greater in the bioresorbable arm, which led to the trail failing non-inferiority (0.371 mm vs 0.250 mm, P=0.78 for non-inferiority, P = 0.0003 for difference). Most troublesome was the fact that there was a low but real incidence of late bioresorbable scaffold thrombosis. Long term outcomes reported a total of 8/335 scaffold thromboses at very long term follow up. None of the Xience patients (n=166) had late occlusions. The device-oriented composite endpoint (cardiac death, target vessel MI, and target lesion revascularization) was more common in the bioresorbable scaffold group (10.4% vs 4.9%, p = 0.043). ABSORB CHINA reported more favorable though different outcomes: composite of all-cause death, MI, and revascularization (10.1% vs 11.4%, p = 0.66); device-related cardiac death, target vessel MI, or target lesion revascularization (4.2% vs 4.6%, p = 0.82) and major adverse cardiac events (5.1% vs 5.1%, p =1.00; target vessel failure (5.5% vs 6.8%, p = 0.57). However, ABSORB CHINA also reported a low incidence of late scaffold thrombosis. The results of these trials were a disappointment, but it is unlikely that bioresorbable scaffolds will disappear. Smart developers are already pursuing smaller, thinner, and more user-friendly scaffolds, and more trials are in the offing. On the other hand, the remarkably positive outcomes demonstrated by current drug eluting stents in these trials will give bioresorbable scaffolds a high bar to cross.

Is Impella, rather than intra-aortic balloon support, the proper choice for interventionalists treating patients with cardiogenic shock?

TCT offered a randomized trial of Impella versus intra-aortic balloon pump (IABP) support that might have answered the question that has lingered so long. Instead – another disappointment. The Impella clearly provides hemodynamic support compared to IABP, and thus Impella has been used preferentially by interventionalists in the past years –despite the fact that no trials have provided evidence of clinical benefit of Impella over IABP. The IMPELLA Trial had only 48 patients with cardiogenic shock randomized to Impella or IABP, and thus was underpowered. There was no difference in mortality at 30 days (IABP-50%; Impella 46%). At 6 months, mortality was identical at 50% in both groups. In a separate meta-analysis which included the IMPELLA data, a total of 95 patients were reviewed with the same results – no difference between IABP and Impella at 30 days or 6 months. It seems that the better hemodynamic support that Impella provides is simply not translated to better clinical outcomes – a sobering conclusion. Despite the negative data, surely further studies will be undertaken to try and identify subgroups of patients with shock that benefit, or if earlier institution of support might be clinically helpful. There is more to come.

Can an “old” trial (RESPECT) finally give us an answer to whether PFO closure is clinically beneficial?

With all of the above disappointments, one “old” trial –RESPECT- finally had its day to shine. The RESPECT Trial was begun about a decade ago, and tested whether closure of a patent foramen ovale (PFO) in patients with cryptogenic stroke was superior to medical therapy. Despite many years of follow up, differences in “closure” versus medical therapy groups were not seen. The original 980-person trial showed superiority of PFO closure over medical management in as-treated and (pre-specified) per-protocol analyses. However, early on in follow up, RESPECT failed to show superiority in its primary end point of significant risk reduction of recurrent stroke in the “intention to treat” analysis. But after 5 years the closure group emerged as having fewer recurrent cerebrovascular events. Just at TCT was beginning, the US Food and Drug Administration approved the Amplatzer PFO Occluder device for recurrent stroke prevention in patients with a PFO and a history of cryptogenic stroke. That indeed will change clinical practice going forward. Many of us initially closed PFO’s when closure devices became available (though not FDA-approved for PFO closure). However, when randomized trial data did not appear to show benefit, PFO closure became a rare event in the cath lab. Note that current FDA approval comes with careful wording that neurologists and cardiologists should be collaborative in the decision making process – much like the “Heart Team” approach to transcatheter valve replacement. Such collaboration should help to more carefully select those patients who might best benefit from closure and also should include a careful evaluation of alternate sources of thromboemboli.


Peter C. Block, MD, is a professor of medicine and cardiology at Emory University Hospital and School of Medicine in Atlanta.

Read the full November/December issue of CardioSource WorldNews Interventions at ACC.org/CSWNI

Clinical Topics: Invasive Cardiovascular Angiography and Intervention

Keywords: CardioSource WorldNews Interventions, Transcatheter Cardiovascular Therapeutics, Drug-Eluting Stents, Follow-Up Studies, Myocardial Infarction, Risk, Stents, Stroke, Thrombosis


< Back to Listings