The ‘Smoker’s Paradox’ Takes a New Hit From Long-term Analysis

There’s more evidence to snuff out the idea that tobacco smokers fare better after PCI, the so-called “smoker’s paradox,” according to new 10-year data from five randomized trials that confirms worse outcomes compared with patients who did not smoke at baseline.

The new DECADE cooperation data are “robust” enough to put the long-held theory to rest and compel cardiologists to put greater emphasis on smoking cessation efforts, senior author Salvatore Cassese, MD, PhD (Deutsches Herzzentrum München, Germany), told TCTMD.

The findings, published online last week in JACC: Cardiovascular Interventions with first author Antonia Presch, MD (Deutsches Herzzentrum München, Germany), show that smoking at baseline increases the 10-year risk of all-cause death, MI, and definite stent thrombosis in almost 10,000 patients undergoing PCI with DES. There was, notably, a lower risk of repeat revascularization among those who smoked at baseline.

“Our results are not as much surprising, but I think they are much stronger than the results from other studies,” said Cassese, highlighting the randomized data and rigorous 10-year follow-up protocols. He called the repeat revascularization finding “very strange,” and said further examination is needed to rule out the play of chance.

A similar patient-level analysis of 10 randomized studies published in 2020 also demonstrated poorer outcomes in smokers following a STEMI, but it followed patients only out to 1 year. Even with all the evidence showing the negative effects of tobacco, as well as the cessation efforts in place, plenty of cardiovascular disease patients continue to smoke.

In an accompanying editorial, Michael A. Kutcher, MD (Wake Forest University School of Medicine, Winston-Salem, NC), argues that the new data place “another and perhaps final ‘nail in the coffin’ of the outdated ‘smoker’s paradox.’” However, he adds, “the investigators may have uncovered a new 10-year ‘smoker’s repeat revascularization paradox’ that could engender controversary and discussion. Whether this is a clinically meaningful entity or just a statistical anomaly should be assessed in other studies.”

‘Applicable to Real Practice’

For the study, the researchers included 9,527 patients receiving PCI with DES, including just under one-quarter who reported smoking at baseline. The five trials were ISAR-TEST 4, ISAR-TEST 5, SORT OUT III, SIRTAX, and EXAMINATION.

Overall, 23.5% and 31.9% of patients in the smoking and nonsmoking groups, respectively, died within 10 years, including 11.7% and 16.8% dying from cardiovascular causes. MI was reported in 9.9% and 7.3%, respectively

On multivariate analysis, smoking was linked with higher 10-year risks of all-cause death (HR 1.45; 95% CI 1.33-1.59), CV death (HR 1.59; 95% CI 1.41-1.80), and definite stent thrombosis (HR 2.09; 95% CI 1.34-3.26). MI risk was found to be comparable regardless of smoking status in the first 30 days as well as from 30 days through 1 year following PCI, but it increased significantly from 1 to 10 years in patients who smoked at baseline (HR 1.60; 95% CI 1.36-1.90).

Risks of target lesion revascularization (HR 0.91; 95% CI 0.84-0.99) and target vessel revascularization (HR 0.91; 95% CI 0.85-0.97) were lower in smokers over 10 years, but there was a similar risk of non-target vessel revascularization in smokers and nonsmokers.

Cassese believes these data should mark the end of the smoker’s paradox, which was largely observed in studies conducted during the “interventional prehistoric era” when thrombolysis was heavily used and DES weren’t as common. “We have a relatively contemporary cohort, because these studies have been done 15 years ago, but a lot of [these] stents are still used,” he said. “It’s something that is much more applicable to real practice.”

He emphasized the harms of all tobacco products beyond cigarettes, which should be curtailed, especially among patients with CVD. “Everyone knows that tobacco smoking is something which is dangerous and is risky for your health status, regardless if you have coronary artery disease or not,” Cassese said. “But I think the new tobacco products that have been marketed . . . should be the new focus.”

Limitations, Future Directions

Geir Frivold, MD, MPH (Loma Linda University School of Medicine, CA), who wasn’t involved in the study, urged caution in interpreting the results due to several limitations, including the fact that patient smoking status post-PCI was not tracked.

“This significantly limits our ability to assess the long-term impact of smoking on stent patency,” he told TCTMD via email. “Given that a major cardiac event often prompts smoking cessation, it is likely that many baseline smokers quit at some point postprocedure. However, the true smoking status of these patients years later remains unknown.”

Additionally, he said, without autopsy data, “it is impossible to determine whether cardiac deaths in smokers potentially were due to stent occlusion or other causes.”

Tor Ole Klemsdal, MD, PhD (Medisinsk Klinikk, Oslo Univeristetssykehus, Norway), who also commented on the findings, told TCTMD that the study results reinforce the work that has already been done to help people quit smoking but underscore the struggle still ahead.

“Although the proportion of daily smokers has fallen steadily in most western countries including the US and Norway, the proportion is still very high among patients admitted with coronary heart disease,” he said in an email. “Active smoking cessation that includes pharmacotherapy help should be offered to all patients with known or suspected atherosclerotic disease (as in general) and must not be forgotten.”