We confirmed several established and identified some novel predictors of radiographic progression. There was a reduced risk of JSN progression over 5 years among overweight or obese patients. In female patients, this effect was seen for older (> 51 years) but not younger women. Truncal fat was associated with reduced risk of JSN progression in women but not in men. Smoking was an independent predictor of erosion progression.
In the literature, RA associated autoantibodies are consistently associated with worse radiographical outcomes [6]. Similar to our results, in two studies of RA patients with longer disease duration, RF and ACPA were associated with more joint damage, in particular erosions [30, 31].
Systemic inflammation (measured by standard laboratory markers) has been shown to predict joint damage [6], but its specific effects on erosion and JSN development have been less studied. Similar to our results, Nawata et al show baseline CRP levels to be related with both erosion and JSN progression [32].
Although some previous studies have shown COMP to be associated with levels of initial radiographic joint damage [33–35], results of studies evaluating radiographic damage over time are mixed [7–9, 33, 34, 36, 37]. Disease duration and timing of radiographic follow up in available studies on COMP and radiographic damage are likely of importance, possible explaining the above mixed results. Similar to our results, a large study on patients with early disease showed somewhat unexpectedly that COMP may be more closely related to the development of erosions than to JSN [10].
The development of erosions and JSN are considered partly separate but intertwined pathophysiological disease processes in RA. The timing of their development and whether each is a risk factor for progression of the other has been evaluated in a few studies. Consistent with our results, all available studies have shown that early JSN is predictive of future JSN progression [38–40]. Furthermore, most studies have shown early erosions to be associated with further erosion progression [38, 39]. Regarding the predictive value of early erosions for JSN and vice versa, results from previous studies are divergent [38–40]. In the present study, erosion score predicted JSN progression, while JSN was not a significant predictor of erosion progression – suggesting that erosions usually precede JSN in early RA.
Previous studies on smoking and radiographic progression have shown mixed results [2, 12, 13, 41–49]. In our previous study of prediction of total SHS progression [11] and especially in the present study, data suggest effects of smoking on radiographic progression and erosion progression in particular, that are independent of RF, disease activity, inflammation and BMI. To our knowledge, the present study is the first to show independent associations of smoking with erosion progression but not JSN progression.
In a previous study, we have shown overweight or obesity to be associated with less rapid radiographic progression [11]. Similarly, several studies have consistently shown links between higher BMI and less radiographic joint damage, summarized in a systematic review by Vidal et al [50]. In the present study, overweight or obesity was associated with less JSN progression, but not erosion progression. This is in line with a study by Alarcon et al, where higher BMI was associated more clearly with erosion-dominant than JSN-dominant pattern of joint damage [30], while BMI was negatively associated with both types joint damage in an earlier study by Hashimoto et al [39]. We found that higher ratio of truncal fat was predictive of less JSN progression in women, while no effect was observed in men. To evaluate if age influences the effects of body constitution on JSN progression in women, we used the reported median age of natural menopause of 51 years [28] and stratified the analyses according to this cut off. The negative association between overweight or obesity and JSN progression was significant in the older (mainly post-menopausal) group, while no such effect was seen in the younger. To our knowledge, this is the first study providing indications of potential effects of interaction between menopausal age and BMI on the rate of JSN progression in RA. Whether physiological age-related changes in body constitution alter levels of adipokines important in the pathogenesis of destructive arthritis or if the effects are mediated by changes in sex hormones, other hormones or chemokines, is unclear and should be further investigated.
Limitations in this study include the relatively small sample size, which affects statistical power for the multivariate analyses. As data on smoking, BMI and COMP were only available at baseline, longitudinal evaluation of the impact of these factors was not possible. Data on body composition by DXA was only available for a subset of the full cohort.
Strengths of our study include the structured longitudinal follow-up of an inception cohort from a defined catchment area. Therefore, selection bias is not a major issue in this study, and the results could be generalized to patients with RA seen in clinical practice.