There's now even more evidence that smoking can be deadly for rheumatoid arthritis (RA) patients, British researchers reported.
In a study with mean follow-up of about 5 years, death rates were more than doubled compared with those who have never smoked, according to Deborah P.M. Symmons, MD, and colleagues from the Manchester Musculoskeletal Biomedical Research Unit.
Their study, published in Arthritis Care & Research, found that smokers were at significantly increased risk of death from any cause, as well as from circulatory disease and lung cancer, and that the mortality risk fell significantly each year after quitting.
"These results emphasize that smoking cessation programs should be a focus for patients newly diagnosed with RA," said Jeffrey A. Sparks, MD, of Harvard Medical School in Boston, who was not involved in the study.
"It is currently unclear whether a new diagnosis of RA encourages smoking cessation -- similar to how a new diagnosis of cardiovascular disease, diabetes, or hypertension finally makes patients quit smoking," Sparks told MedPage Today.
Patients with RA have a mortality risk about 1.5-fold higher over specified time periods than the general population, and they also have higher rates of smoking and other cardiovascular risk factors. However, it has not been clear whether smoking contributes to mortality risk or what effects of cessation might be.
To investigate this, Symmons and colleagues analyzed data for 5,677 patients with RA from the Clinical Practice Research Datalink (CPRD), a U.K. database of primary care electronic medical records. The database holds the records of almost 13 million patients, with active patients representing about 7% of the U.K. population.
In the study cohort, 67.8% were female, and the median age at RA diagnosis was 61.4 years. At baseline, 40.3% were never smokers, 34.1% were former smokers, and 25.6% were current smokers.
The researchers treated smoking status in a time-varying fashion. "By allowing smoking status to vary throughout follow-up, our study reflects more accurately the patients' exposure," the authors wrote.
During a mean follow-up of 4.7 years, 16% changed smoking status, including 348 current smokers who stopped smoking after a single attempt.
After linking CPRD data with mortality data, researchers found that during 26,679 person-years of follow-up, 574 patients died. This provided a crude mortality rate of 21.5 per 1,000 person-years (95% CI 19.8-23.3).
After adjustment for age and sex, the mortality rates for never, former, and current smokers were 16.2 (95% CI 13.7-18.6), 22.4 (95% CI 19.7-25.2), and 31.6 (95% CI 25.3-37.9) per 1,000 person-years, respectively. This translates into about six deaths per 1,000 person-years attributable to former smoking, and 15 deaths per 1,000 person-years attributable to current smoking.
In the time-varying smoking status analysis, current smokers had a significantly higher risk of death than never smokers (HR 2.18, 95% CI 1.73-2.76) or former smokers (HR 1.77, 95% CI 1.43-2.20) after adjustments for age and sex. This higher risk persisted after full adjustments that included socioeconomic status, body mass index, cardiovascular disease, and diabetes, among other factors.
In former smokers, the mortality risk fell significantly each year after quitting. This was particularly true for those who had been heavy smokers, whose fully adjusted hazard ratio fell to 0.85 (95% CI 0.77-0.94) for each year since cessation.
The major cause of death was circulatory disease, with a mortality rate of 8.3 per 1,000 person-years. In general, the age and sex adjusted mortality rates were lowest in those who had never smoked and highest in those who currently smoked.
However, for respiratory-related mortality, former smokers had a nonsignificant increased risk of death. "This could be because patients with pre-existing respiratory disorders are more likely to stop smoking," the authors wrote.
It's also possible, they suggested, that patients who continued to smoke were in some ways healthier because the risk or presence of smoking-related disorders such as cardiovascular disease may have motivated other less healthy smokers to quit.
Current smokers also had significantly increased risk of dying from lung cancer, but there was no significant association between current smoking and deaths due to all-site cancer. Risk of mortality due to lung cancer was also increased in former smokers compared to never smokers.
A limitation of the study was the reliance on an algorithm to identify cases of RA, but the authors pointed out that the algorithm they used has a sensitivity and specificity of over 80% and that the included patients were considered to be "highly probable cases."
Other possible weaknesses of the study were the limited follow-up time, and misclassification of smoking status as this information can be influenced by how often patients visit their primary care physician and how accurately patients report their smoking behavior.
"Future research is needed to quantify whether a diagnosis of RA compels patients to quit smoking compared with the general population, and to institute RA-specific smoking cessation programs," Sparks said.
Dixon and Movahedi were supported by an MRC Clinician Scientist Fellowship. The work was further supported by the Arthritis Research U.K. Center for Epidemiology.
Sparks had no financial disclosures.
Aucun commentaire:
Enregistrer un commentaire