Is early remission associated with improved survival or is arthritis persistency associated with increased mortality in early arthritis?- comparisons with the general population

Jessica AB van Nies, MD,

Other Contributors:

June 28, 2013

Dear editor,

With interest we read the article of Sciré et al.,(1) evaluating the association between remission and mortality in patients with inflammatory polyarthritis (IP). This study shows that patients achieving remission early in the disease course, have an improved survival rate compared to patients never achieving remission. Based on this observation the authors conclude that achieving remission early in the disease influences the long -term outcome. The question arises which processes underlie their observation. Rheumatoid Arthritis (RA) has been associated with an increased mortality risk, which has been attributed to the effects of long -term inflammation.(2) We hypothesized that the observation in which patients with remission have an improved survival rate, might actually be driven by the association between prolonged inflammation and increased mortality risk. We addressed this hypothesis by taking advantage of long-term outcomes of 556 patients with early undifferentiated arthritis (UA) or RA that were included in the Leiden EAC between 1993-1998.(3) Treatment in this era was delayed and DMARDs were mild. We previously observed that patients included in this era had an increased mortality rate compared to the general Dutch population(4); an effect which was not present in patients that were diagnosed more recently and treated more aggressively.(4)
Presently, we evaluated the remission status in relation to mortality. In our study remission was defined as the persistent absence of synovitis for at least one year after cessation of contingent DMARD-therapy. This is the most stringent definition of remission and differs of the definitions used by Sciré et al., which were based on the number of tender and swollen joints at a point in time. Presence of DMARD-free sustained remission was assessed within 3 years after inclusion (similar to Sciré et al.) and also at any point during follow-up (median 15 years, IQR 13.2-17.1). Mortality data on patients were tracked nationally using the civic registries (Gemeentelijke Basis Administratie). First, mortality rates were compared between patients with versus patients without remission, similar as performed by Sciré et al. Secondly, we compared the mortality rates of both groups with the mortality rates of the general Dutch population. Kaplan Meier-curves were constructed. Cox proportional hazard regression models for multivariable comparisons were made; all analyses were adjusted for age and sex. Subsequent adjustments were made also for rheumatoid factor and smoking (ever versus never). Patients were censored at date of death or 1st May 2012 whichever came first. Standardized mortality ratios (SMR) were calculated using the general Dutch population matched on age, sex and inclusion year (Statistics Netherlands). Out of 556 patients, 162 (29%) patients died during the total follow-up of 7,682 person years. The mean age was 52.6?17.2, 60% was female. DMARD-free sustained remission was achieved in 124 (22%) patients within 3 years after inclusion and in 182 (33%) patients during the total follow-up.
Patients that achieved remission within 3 years had a reduced mortality compared to patients without remission (HR 0.49 95%CI 0.32-0.75), also after additional adjustments for rheumatoid factor and smoking (HR 0.60 95%CI 0.39-0.93). Likewise, patient that achieved remission at any point during follow-up had a lower mortality risk (HR 0.41 95%CI 0.28-0.60 and HR 0.50 95%CI 0.33-0.74, respectively)
Next we questioned whether patients with remission had a decreased mortality rate compared to the general Dutch population and/or whether patients without remission had an increased mortality rate.

We therefore compared the mortality rates of both patients groups to the mortality rates of matched controls from the general Dutch population. Patients who did not achieve remission during three years of disease or during the total follow-up duration had an significantly increased mortality rate (SMR 1.49 95% CI 1.26-1.76 and 1.66 95%CI 1.40-1.98 respectively). Patients that achieved remission within the first three years or the total follow-up duration did not have an increased mortality (SMR 0.70 95% CI 0.47-1.02 and SMR 0.66 0.47-0.92 respectively).

In conclusion, we studied the relation between DMARD-free sustained remission and mortality. Similar as Sciré et al. we observed that patients that achieve remission have an improved survival compared to patients without remission. Furthermore, when using the general population as a reference, we observed that patients with persistent arthritis had 49-66% increased mortality risk. This is in line with previous observations , in these patients and in other studies, which show that rheumatoid arthritis is associated with a decreased survival.(1;4) This increased mortality risk was no longer present in patients that achieved DMARD-free sustained remission, indicating that the survival probability returns to normal in patients that are cured from RA. In our view the slightly decreased SMR in these patients can be explained by a healthy inclusion bias or by good medical care while being under specialist's control.(4)


(1) Sciré CA, Lunt M, Marshall T, et al. Early remission is associated with improved survival in patients with inflammatory polyarthritis: results from the Norfolk Arthritis Register. Ann Rheum Dis 2013 Jun 7;Epub ahead of print doi:10.1136/annrheumdis-2013-203339 .

(2) Sokka T, Abelson B, Pincus T. Mortality in rheumatoid arthritis: 2008 update. Clin Exp Rheumatol 2008;26(5 Suppl 51):S35-S61.

(3) de Rooy DP, van der Linden MP, Knevel R, et al. Predicting arthritis outcomes--what can be learned from the Leiden Early Arthritis Clinic? Rheumatology (Oxford) 2011;50(1):93-100.

(4) van Nies JA, de Jong Z., van der Helm-van Mil AH, et al. Improved treatment strategies reduce the increased mortality risk in early RA patients. Rheumatology (Oxford) 2010;49(11):2210-6.

Conflict of Interest:

None declared

Conflict of Interest

None declared