Smoking Cessation in RA: Q&A with Dr. Deborah Symmons

May 3, 2016

Smoking and rheumatoid arthritis are intertwined. In this Q&A, Deborah Symmons, M.D., of the University of Manchester discusses the research on smoking and rheumatoid arthritis.

Smoking and rheumatoid arthritis are intertwined. Cigarette consumption is one of the strongest environmental risk factors for the disease. And now, new research finds, smoking is also a risk factor for early death among rheumatoid arthritis patients who do smoke.

The study did not examine the mechanisms for this link, but other research suggests that there are physiological reasons that smoking might be particularly deadly for people with rheumatoid arthritis. Eventually, researchers may even be able to target smokers at genetic risk for the disease in order to prevent rheumatoid arthritis damage before it starts.

It's been long known that people with rheumatoid arthritis are more likely to be smokers than people without the disease. There is also good reason to believe that smoking may be the precipitating factor for RA in people who are genetically predisposed to the disease. A review published in April 2016 in the journal Arthritis & Rheumatology lays out the case: Smoking is the most robust risk factor for the disease, and people who both smoke and carry two HLA-CRB1 alleles that code for a shared epitope (SE) are 21 times more likely than those without those HLA-SE alleles to develop anticitrulline-positive rheumatoid arthritis. Meanwhile, there is evidence that localized autoimmunity in the mucosa of the mouth, lungs and guts may precede systemic rheumatoid arthritis. Smoking may be one trigger of this autoimmunity in the oral mucosa and in the lungs.

Smoking may also complicate the health of people with established RA; for example, current smokers are less responsive to methotrexate and TNF inhibitors, according to a 2010 study in Arthritis & Rheumatology. The reason may be that smokers metabolize methotrexate differently than non-smokers, the authors wrote, or it may be that smokers have a more severe disease course than non-smokers, regardless of treatment.

Some research supports the latter interpretation. A recent paper in the journal Mediators of Inflammation tested the blood of 543 patients with the disease for RA biomarkers. After controlling for age, disease activity, gender and body mass index, the researchers found that smokers had lower levels of IGF1 and leptin than non-smokers. Because IGF1 deficiency is associated with osteoporosis and atherosclerosis, among other effects, and low leptin is linked to poor survival in cardiovascular disease and bone loss, the decreases in IGF1 and leptin could explain a variety of effects in the progression of rheumatoid arthritis, the researchers wrote. 

"Smoking-induced low levels of IGF1 and leptin may support the aberrant T-cell formation at the preclinical stage of arthritis, in the poor IGF-1 bone remodelling and progressive joint damage in the overt arthritis, and for the early cardiovascular mortality in RA," they wrote.

Mortality is the topic of a new study in Arthritis Care & Research which used U.K. electronic medical records to follow 5,677 patients with rheumatoid arthritis, 26 percent of whom were current smokers, 24 percent of whom had previously smoked, and 40 percent of whom were never-smokers. The study found that current smokers were nearly twice as likely to die prematurely as never-smokers (HR 1.98 (1.56-2.53)). This risk stemmed from an increased rate of mortality from circulatory disease and lung cancer (HR 1.96 (1.22-2.90) and 23.2 (5.15-105), respectively). The analysis found that former smoking was to blame for 6 deaths per 1,000 person-years, and current smoking was to blame for 15 deaths per 1,000 person-years. The good news for patients is that mortality risk drops with each year after smoking cessation - in other words, it's always worth it to quit.

Rheumatology Network spoke with study researcher Deborah Symmons, M.D., a professor of rheumatology and musculoskeletal epidemiology at the University of Manchester, and the research lead for the prevention of co-morbidity at the NIHR Manchester Musculoskeletal Biomedical Research Unit, to explain the findings and how physicians should approach the issue of smoking with their patients.  [[{"type":"media","view_mode":"media_crop","fid":"48248","attributes":{"alt":"Deborah Symmons, M.D.","class":"media-image media-image-right","id":"media_crop_6064725061123","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"5749","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"font-size: 13.008px; line-height: 1.538em; float: right;","title":"Deborah Symmons, M.D.","typeof":"foaf:Image"}}]]

RN: Why is it important to understand smoking and RA? Is there reason to think that smoking might be more damaging for someone with RA than for someone without the disease? 

Symmons: It is already well recognized that smoking is a risk factor for the development of RA – particularly sero-positive RA. Therefore the prevalence of smoking amongst patients with RA is higher than in the background population. Smoking is likely to be at least as damaging for someone with RA as for someone of the same age and sex who does not have RA. It is also likely that smoking could be more damaging particularly with regards to risk of chest infections. We know that RA itself is a risk factor for infections and this can be amplified by the immunosuppressant therapy taken by many patients. If we then add smoking to the mix, then the risk of infection is likely to rise still further. Published evidence suggests that many patients continue to smoke after developing RA (i.e. smoking cessation programs don’t work well in this patient group) and our motivation in doing this study was to provide quantitative evidence to patients with RA and their carers, that smoking continues to be harmful after the development of disease.

What was new about the way you decided to tackle this question?

We wanted to examine the harms of smoking within the RA population (i.e. not compared to the general population). We used a large electronic primary care based database and a validated algorithm to identify patient with new-onset RA. This enabled us to study thousands of patients from close to the diagnosis of their disease – something which is not possible within a hospital clinic setting. It also enabled us to generalize our results to the whole RA population in England. It was unique in that we studied smoking in a time-varying manner (i.e. we did not just look at baseline smoking status but studied the effect of stopping and restarting smoking during the course of RA.) We also controlled for diabetes, CYD and DMARD and steroid treatment in a time-varying manner.  It was also unique that we included a competing risk analysis. 

Were you surprised by your results? How did they fit into previous research on this topic?

On the whole we were not surprised by the direction of the results – although we were glad that our study had sufficient power for many of the associations to be statistically significant.

We were somewhat surprised that former smokers had a greater increased risk of mortality from respiratory infection than current smokers. We were interested to see that most of the excess risks reported were confined to heavy smokers and also to note the year-on-year improvement in mortality risk after stopping smoking. We were disappointed to note that only 2 percent of current smokers gave up smoking after a single attempt during the period of follow-up – although this did confirm other published reports that rates of stopping smoking in RA are low. 

Why do you think you saw the results you did? Does having rheumatoid arthritis make smoking more risky?

On the whole the direction of the results are the same as one would expect within a sample of the general population. Smoking does not increase the risk of all types of cancer. Lung cancer is the cancer most strongly associated with smoking and that was the cancer which we found with the highest relative risk (23-fold increase in current smokers). I think the fact that, for some causes of death, former smokers had a higher risk than current smokers can be explained by the reason that smokers quit. If someone quits smoking after having a heart attack or hospital admission for a chest infection then that increased risk will persist for sometime after stopping smoking. 

We can see that, at baseline, former smokers had a higher prevalence of cardiovascular disease, diabetes and cardiovascular medications. We can’t comment on whether the relative risks of smoking are higher in this RA population than they would be in an age and sex matched sample of the English population, since we didn’t include a control sample. Also relative risk is related not only to the mortality risk amongst smokers but also the risk among non-smokers. We are not claiming that all the excess risk of mortality seen in RA cohorts is due to smoking.

What should be the take-away message for rheumatologists from this study? 

We have shown that patients with RA who continue to smoke have a higher mortality risk than those who never smoked.  More importantly, we have shown that the risk of mortality falls year-on-year in former smokers. We hope that rheumatologists can share this information with their patients who will then be motivated to stop smoking.

How do you approach smoking and smoking cessation in the clinic? What can rheumatologists or GPs do to help patients who smoke? 

I ask patients at every visit whether or not they smoke. If they do smoke, I point out the harms (even if I have done so before). I also point out that it is not a case that the damage is done and there is no point in stopping. 

Although some work has been done, we need more materials on smoking cessation that are specific to RA accentuating the harms such as those shown in this study and the harms relating to poorer treatment response, etc. In the absence of these, patients should at least be referred to and encouraged to follow smoking cessation programs developed for the general population.

Your paper does not address this directly, but do you think there is a prevention angle to the link between smoking and RA? For example, would it be worthwhile to target smokers with a family history or other biomarkers before RA develops? 

As you say, our paper doesn’t really relate to this question. There is an increasing body of opinion to suggest that we do now know enough about the genetic and environmental risk factors for RA to develop strategies which could prevent the development of the disease in those at highest risk. However, these would need to be tested in the context of a trial just like any other prevention strategies.

With regards to smoking, it does seem self-evident that people who know that they have a high background risk of RA because of their family history should be encouraged by their GPs and family members not to smoke. I am not sure it would be practical to screen all smokers to see if they have a family history of RA and then target those who do for smoking cessation. I think, at a population level, a reduction in the number of new cases of RA is not the strongest reason to advocate stopping smoking – but it might be at the individual level for those at highest risk. 

 

References:

Joseph RM, Movahedi M, Dixon WG, Symmons DP. Smoking-related mortality in patients with early rheumatoid arthritis - a retrospective cohort study using the Clinical Practice Research Datalink. Arthritis Care & Research 2016. doi:10.1002/acr.22882.

Mankia K, Emery P. Review: Preclinical Rheumatoid Arthritis: Progress Toward Prevention. Arthritis & Rheumatology 2016;68(4):779–788. doi:10.1002/art.39603.

Klareskog L, Stolt P, Lundberg K, et al. A new model for an etiology of rheumatoid arthritis: Smoking may trigger HLA–DR (shared epitope)–restricted immune reactions to autoantigens modified by citrullination. Arthritis Rheum Arthritis & Rheumatism 2005;54(1):38–46. doi:10.1002/art.21575.

Saevarsdottir S, Wedrén S, Seddighzadeh M, et al. Patients with early rheumatoid arthritis who smoke are less likely to respond to treatment with methotrexate and tumor necrosis factor inhibitors: Observations from the Epidemiological Investigation of Rheumatoid Arthritis and the Swedish Rheumatology Reg. Arthritis & Rheumatism 2010;63(1):26–36. doi:10.1002/art.27758.

Erlandsson MC, Medina RD, Silfverswärd ST, Bokarewa MI. Smoking Functions as a Negative Regulator of IGF1 and Impairs Adipokine Network in Patients with Rheumatoid Arthritis. Mediators of Inflammation 2016;2016:1–8. doi:10.1155/2016/3082820.