Smoking Habits Following Cancer Diagnosis and Heart Failure
Menée à partir de données de l'Assurance maladie coréenne portant sur 548 320 patients atteints d'un cancer (âge médian : 59 ans ; durée médiane de suivi : 7,3 ans), cette étude de cohorte analyse l'association entre des pratiques tabagiques et le risque d'insuffisance cardiaque en lien avec les traitements anticancéreux
Résumé en anglais
Introduction: A previous study reported that changes in smoking habits after cancer diagnosis are associated with the risk of atherosclerotic cardiovascular disease.1 Heart failure (HF), given its different pathophysiologic characteristics and association with certain chemotherapeutics (eg, anthracyclines or antihuman epidermal growth factor receptor 2 [ERBB2; formerly HER2] agents), represents a distinct cardiovascular event among individuals with cancer.
Methods: In this cohort study, we identified 548 320 individuals with cancer from the Korean National Health Insurance database2 meeting the following criteria: (1) diagnosed with cancer (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] codes C00-C97, and rare and intractable disease code V193)3 at age 20 years or older between calendar years 2005 and 2014, (2) survived for 3 or more years, (3) underwent health examinations within 3 years before and after cancer diagnosis, and (4) did not have HF or atherosclerotic cardiovascular disease before the 3-year survival date (index date). This study, from January 15 to October 31, 2023, followed the STROBE reporting guideline and was approved by the institutional review board of Severance Hospital. Informed consent was waived because this study was based on deidentified administrative data. Participants’ self-reported current smoking status was assessed in both prediagnosis and postdiagnosis health examinations. Smoking habit in patients who did not smoke prediagnosis was classified as either sustained nonsmoking or initiated/relapsed smoking based on postdiagnosis smoking status. Smoking habit in patients who smoked prediagnosis was classified as either quit or continued smoking.1 Participants were followed up from the index date until the earliest incident of HF (first-ever hospitalization with ICD-10 code I50),4 death, or December 31, 2021. We calculated hazard ratios (HRs) and 95% CIs for HF using a cause-specific Cox proportional hazards model, with participants censored at competing death events.5
Results: The median age was 59 (IQR, 51-69) years; 52.0% were female and 48.0% were male. Of 449 834 patients who did not smoke prediagnosis, 10 043 individuals (2.2%) initiated or relapsed into smoking postdiagnosis. Of 98 486 patients who smoked prediagnosis, 50 967 (51.8%) quit smoking postdiagnosis. Over a median follow-up of 7.3 (IQR, 5.4-9.9) years, 27 053 incident HF events occurred. Compared with sustained nonsmoking, the multivariable-adjusted risk of HF was 36% higher among individuals who initiated/relapsed into smoking (HR, 1.36; 95% CI, 1.26-1.47). The risk was also higher among those who quit (HR, 1.17; 95% CI, 1.12-1.22) or continued smoking (HR, 1.54; 95% CI, 1.48-1.61) (Figure, A). Compared with continued smoking, the risk was 24% lower among patients who quit (HR, 0.76; 95% CI, 0.72-0.80). When stratified by cancer type, the findings were consistent across upper gastrointestinal; lower gastrointestinal; hepatobiliary, pancreatic, or other gastrointestinal; lung; breast or female genital organs; male genital organs; and urinary tract cancers (Table).vA total of 37 866 participants (6.9%) had been exposed to anthracyclines or anti-ERBB2 agents before the index date. Regardless of the exposure, smoking initiation/relapse was associated with a higher risk of HF than sustained nonsmoking, while smoking cessation was associated with a lower risk of HF than continued smoking. Participants exposed to anthracyclines or anti-ERBB2 agents who continued smoking exhibited the highest risk of HF (HR, 2.43; 95% CI, 2.03-2.91 vs nonexposed sustained nonsmoking) (Figure, B).
Discussion: The risk of HF in patients with cancer can vary by cancer type, use of cardiotoxic anticancer therapies, and presence of shared risk factors.6 In our study, patients who had received anthracyclines or anti-ERBB2 agents and continued smoking showed the highest risk of HF, representing a high-priority group for HF surveillance and smoking cessation counseling. Nevertheless, smoking discouragement remains a widely beneficial approach, as evidenced by consistent results across 7 major cancer types/groups and exposure status to cardiotoxic anticancer therapies.
Study limitations include (1) potential heterogeneity in smoking parameters within each group, (2) inability to differentiate between HF with reduced vs preserved ejection fraction, and (3) lack of information on pathologic subtype and stage of cancer. Despite these limitations, our study highlights the critical need for smoking prevention and cessation among individuals with cancer, particularly those treated with cardiotoxic anticancer therapies.
