The beneficial effect on all-cause mortality of coronary artery bypass grafting surgery (CABG) observed at 4 and 5 years in women with complex coronary disease seen in the SYNTAX trial is gone at 10 years.
If anything, the results suggest a mortality benefit for CABG over percutaneous coronary intervention (PCI) mainly for men (adjusted hazard ratio [HR], 0.76; 95% CI, 0.56 – 1.02) and not for women (adjusted HR, 0.90; 95% CI, 0.54 – 1.51) in the SYNTAX Extended Survival (SYNTAXES) study.
The sex–treatment interaction for all-cause mortality was significant at 5 years (P = .025) but not at 10 years (P = .952).
“I’m becoming very humble with trials because I’m not expecting the convergence of the curve. I was expecting like a surge, a further divergence,” senior author Patrick Serruys, MD, PhD, National University of Ireland, Galway, said in an interview. “You could say, at the end of the day, everybody dies. And that’s the life expectancy factor.”
Although female patients had slightly lower anatomic SYNTAX scores at randomization (27.0 vs 29.2), they were on average 4 years older than men (mean age, 68 years) and had higher prevalence rates of diabetes, hypertension, and chronic kidney disease, he noted. “The other explanation is that we know that the bypass graft, the saphenous bypass graft, became vulnerable around 7 years; that’s probably the half-life.”
Overall, mortality in both men and women tended to be lower after CABG than after PCI, although the differences were not statistically significant, the authors reported August 17 in the Journal of the American College of Cardiology.
The 1800-patient SYNTAX trial showed no difference in all-cause mortality at 5 years between CABG and PCI, although CABG was associated with fewer major adverse cardiac and cerebrovascular events (MACCE) and more favorable results among those with complex, three-vessel disease.
The findings were confirmed in 10-year follow-up reported last year from SYNTAXES, which analyzed only all-cause mortality.
Female sex, however, was an independent predictor of mortality with PCI at 4-years follow-up (HR, 2.87) in SYNTAX and led to sex being incorporated into the SYNTAX score II to help guide revascularization decisions. Notably, this interaction for all-cause mortality has not been seen in other studies.
Treatment Effect by Sex
In the new prespecified subgroup analysis, women had a higher crude rate of all-cause mortality at 10 years than men (32.8% vs 24.7%; log-rank P = .002).
This held true whether women were in the PCI group (33.0% vs 27.0%; log-rank P = .053) or the CABG group (32.5% vs 22.5%; log-rank P = .017).
In women, the mortality rate was significantly higher with PCI than with CABG at 5 years, but was no longer different at 10 years (33.0% vs 32.5%; log-rank P = .601). This was largely due to an uptick in deaths between 5 and 10 years in those treated with CABG, compared with PCI.
In men, the mortality rate was similar between PCI and CABG at 5 years, but tended to be higher with PCI at 10 years (27.0% vs 22.5%; log-rank P = .082).
Asked about the possible late benefit for CABG in men, Serruys replied: “Of course, everyone had made a hypothesis — ‘Let’s look at the use of internal mammary arteries in these patients, etcetera’ — but I must be honest, we don’t have an explanation so far.”
Commenting on the study to theheart.org | Medscape Cardiology, Roxana Mehran, MD, Mount Sinai School of Medicine, New York City, said with just 402 women and using a no-longer-available, first-generation (Taxus) stent, the findings are, unfortunately, not informative.
“For me, it would be important for these investigators to share their data for women so we can do a patient-based analysis to better figure out the differential between first-generation stents and how well we’re doing,” Mehran said.
“What’s really important is to have a study where you actually collect female-specific risk factors that are never, ever looked at, [such as] age at menopause or having had pregnancy-related complications, that predispose these women to more of an atherosclerotic risk. And, even so, to better understand their anatomy and what suits them better,” she said. “I just don’t think we know enough or have put enough effort into understanding the biology that is sex-specific and different for men and women.
Revising SYNTAX Score II
Given the lack of a sex–treatment interaction in the analysis, Serruys and colleagues suggest that the SYNTAX score II “should be re-evaluated for the prediction of all-cause mortality at 10 years.”
Lending further support to this is the fact that SYNTAX score II was similar between women who died at 5 to 10 years and those who died in the first 5 years after CABG (31.8 vs 31.6).
“The authors rightfully ask whether the SYNTAX II score should be revised to remove female sex, and given the current study result this appears warranted,” Arnold H. Seto, MD, MPA, Long Beach Veterans Administration Hospital, California, said in a related editorial.
He points out that women in SYNTAXES treated with CABG tended to have a survival time 0.51 years longer than women treated with PCI (P = .07). Nonetheless, the lack of confirmation for a sex-specific treatment interaction in any other study — EXCEL, FREEDOM, BEST, PRECOMBAT, BARI, or MASS — strongly suggests that the interaction seen in SYNTAX is likely a “type 1 error.”
Rather than focusing on early mortality, which may represent relatively rare events that are susceptible to chance, Seto suggests “other end points such years of life saved, quality adjusted life-years, and MACE may better capture the benefits of different revascularization decisions, even if they have a higher risk for bias.”
A new risk model, SYNTAX score 2020, has been developed and will be published imminently, Serruys told theheart.org | Medscape Cardiology.
The SYNTAX Extended Survival study was supported by the German Foundation of Heart Research. The SYNTAX trial, during 0- to 5-year follow-up, was funded by Boston Scientific. Both sponsors had no role in study design or data collection, analyses, and interpretation, nor were they involved in the decision to publish the final manuscript. Serruys has received personal fees from Biosensors, Micel Technologies, Sinomedical Sciences Technology, Philips/Volcano, Xeltis, and HeartFlow, outside the submitted work. Seto reported research grants from Philips and Acist, and honoraria from Terumo, Getinge, Boston Scientific, General Electric, and Janssen.