OP 0011 : Repair of Erosions Occurs Almost Exclusively in Damaged Joints without Swelling: Post Hoc Analysis of Radiographic Data from Year 1 of the TEMPO Study
Authors: van der Heijde, Lukas, Fatanejad, Landewe
Background:
Recent conflicting data have raised questions about the progressive nature of RA joint damage. On one hand, the observation of repair of joint erosions in clinical trials of TNF inhibitors has raised the possibility that radiographic damage in RA may be reversible. On the other hand, the observation of radiographic progression in joints without apparent symptoms has suggested that bony erosion in RA may occur via mechanisms unrelated to adjacent synovitis. To further explore this issue, van der Heijde et al compare radiographic healing and clinically observed swelling from subjects enrolled in the TEMPO trial.
Methods:
The design of the TEMPO trial has been previously described on this site (link to http://www.hopkins-arthritis.org/news-archive/2004/TEMPO.html). For this post-hoc analysis, two X-ray readers, blinded to treatment allocation and sequence, twice evaluated all radiographs of the hands and wrists from the first year of TEMPO. Mean erosion change scores were calculated as the mean of the first reading from each reader. Joints exhibiting repair (i.e. regression in radiographic score over time) were identified and compared to changes in swelling from the clinical examination.
Results:
11,159 individual joints were evaluated. Among these, 553 (5%) exhibited repair over time and had available information on change in swelling. Among joints exhibiting repair, only 1 joint of 553 (0.2%) showed an increase in swelling over the time of repair compared to 130 joints of the 10,497 (1.3%) that did not exhibit repair (p=0.0337). Similarly, 318 of the 553 repaired joints (57.5%) demonstrated a reduction in swelling over the time of repair compared to 4,224 of the 10, 497 joints without repair (40.3%; p<0.0001).
Subjects with baseline radiographic damage and no change or improvement in swelling had significantly lower mean change in erosion scores compared to subjects with swelling that worsened (see table). However, there was no difference in mean change in erosion scores according to change in swelling in subjects without baseline radiographic damage (see table).
Change in Erosion Scores According to Change in Swelling, by Presence of Baseline Damage
|
Baseline Damage Present |
Baseline Damage Absent |
||
|
Mean Change in Erosion Score |
95% CI |
Mean Change in Erosion Score |
95% CI |
No Change or Improvement in Swelling |
-0.09 |
-0.11, -0.06 |
0.01 |
0.00, 0.01 |
Swelling Worsened |
0.06 |
-0.02, 0.14 |
0.03 |
0.01, 0.04 |
Conclusions:
Repair of RA erosions in response to therapy is greatest in joints with improvement in swelling, with the effect the most pronounced in patients with baseline radiographic damage.
Editorial Comment:
This is an elegant approach to answering a complex question. While the findings may seem intuitive (that joints with persistent synovitis are less likely to show regression of erosions with treatment), this has not previously been demonstrated. Even so, whether radiographic healing actually occurs or is a phenomenon related to insensitivity of sequential radiographic interpretation remains controversial. Demonstrating a clear link between improvement in clinical joint swelling and repair, as evidenced by this abstract, would appear to support the notion that repair of erosions does occur. It is important to note that even with aggressive therapy, possible repair of radiographic erosions was only seen in 5% of hand/wrist joints and that ultimately the goal of RA treatment remains prevention, rather than repair, of damage.