Mr Miguel Sousa-Uva
This was an excellent session that proved to be even more exciting than it promised to be. The session started with Salim Yusuf’s (McMaster University, Canada) very provocative “View from the Non Interventional Cardiologist”. Yusuf reminded us that before Syntax, the information available showed that there was no evidence that PCI did better than medical treatment in patients with stable coronary artery disease. He commented on the observed significant differences in medical treatment between the two arms of the randomized trial. Indeed only 84% of patients who underwent CABG had aspirin against 91% (p=0.001) with PCI and the differences were also significant for the rates of thienopyridines (97% vs 19%), statins (87% vs 75%) and ACEI (94% vs 55%). For Yusuf, the rate of the primary end point (major adverse cardiac or cardiovascular events) of 12.4% in the CABG group vs 17.8% in the PCI group (p=0.02) from Syntax were to be expected. He predicts that these differences at 5 years will be even greater in favour of CABG with stroke rates becoming similar in both groups. He also advised not to dig into subgroup analysis concluding that the clinical implications were that CABG (with optimal medical therapy) remains the standard of care for patients with 3 vessel or left main disease.
Friedrich Mohr (Syntax co-primary investigator, Leipzig Heart Institute) started by questioning the truth of the often used statement that repeat revascularisation is “not a big deal”. He went through the Syntax main results, reminding us that antithrombotic treatment in the CABG group had been suboptimal (role of double anti platelet therapy?) and stressed that 50% of the strokes in the surgical group had no neurological damage at 1 year. He also reminded us that by study protocol, sub-analysis of cohorts were not allowed, but were used abusively to draw general conclusions. Also of note is the fact that Syntax score tertiles indicated that in 66% of patients, the best treatment was surgery. He then drew on the results of the FAME study (which showed that an FFR-guided strategy reduced MACE rates) to make the hypothesis that the higher rate of MACCE in the randomized trial compared to the registry could be due to the more complex anatomy and higher degrees of significant stenosis in the latter. He ended by stressing the need for a heart team approach and to wait for a longer follow up, foreseeing that the MACCE curves will continue to diverge with time.
“Conclusions from subgroups” was the title of Pieter Kappetein’s (Erasmus, Rotterdam) excellent talk focusing on the statistical aspects that must be kept in mind. In summary, overall comparisons between PCI and CABG should be limited to the primary end point and pre specified subgroup hypothesis. When performing subgroup analyses in a non inferiority trial, p values are invalid and the probability of falsely concluding that there is a significant difference when in fact there is none (false positive) grows exponentially with the number of subgroups under analysis. For example, in left main disease, the p value of 0.4 for MACCE at 12 months does not allow the conclusion that PCI is not inferior to CABG. Interaction is the statistical test to perform in these circumstances and sub group analysis should only be exploratory hypothesis for future studies.
Finally, Manuel Antunes (Coimbra University Hospital, Portugal) replaced David Taggart who could not be present and made the case for “The patient has the right to know” stressing the need for a real informed consent before a decision is made. He reminded the audience of the rationale behind the better results observed with CABG in the Syntax patient population: PCI treats the culprit lesion whereas CABG bypasses the lesion therefore treating not only the culprit but also future lesions. The much heralded zero restenosis rate of DES and other rosy prophecies did not materialise. Let’s wait and see what the future holds, (see also Wednesday morning, Kappetein’s presentation at Clinical trial Update III, Barcelona room), but in the mean time, the conclusions of the New England Journal of Medicine paper published last January remain valid one year on: CABG remains the standard of care for patients with triple vessel or left main disease.
The SYNTAX trial: one year on
Our mission: To reduce the burden of cardiovascular disease
© 2017 European Society of Cardiology. All rights reserved