Background/Purpose: The DAS28 is a measure of disease activity in rheumatoid arthritis (RA). DAS28 stands for ‘disease activity score’ and the number 28 refers to the 28 joints that are examined in this assessment [1, 2]. DAS28 is a composite outcome measure that assesses how many joints in the hands, wrists, elbows, shoulders, and knees are swollen (SJ) and/or tender (TJ), the erythrocyte sedimentation rate (ESR) or C reactive protein (CRP) in the blood to measure the degree of inflammation, the patient’s general health (GH) on a Visual Analogue Score (a simple scale) to assess how they are feeling on that day. These results are then fed into a complex mathematical formula to produce the overall disease activity score. A DAS28 of greater than 5.1 implies active disease. In the UK the NICE has approved the use of TNFα<span”>inhibitors to treat patients with rheumatoid arthritis in accordance with the British Society for Rheumatology guidelines. Having a DAS28 score equal or greater than 5.1 on two occasions is one of the criteria required. Failure of anti-TNF treatment is defined as having a difference between baseline and 6 months post-treatment DAS28 score of < 0.6. Although DAS28 score is a validated tool for the identification of patients who are likely to benefit form anti-TNF treatment, it takes into account a number of subjective parameters which are likely to influence the overall result and reach the NICE threshold for the initiation of anti-TNF treatment [3]. The purpose of this study is to explore how an increase in the reported number of tender joints and/or GH would affect DAS28 score and to identify patients who although qualify for, might nor benefit from anti-TNF treatment.

Methods:   We retrospectively investigated the GH, TJ, SJ, ESR, baseline DAS28 score and the DAS28 score 6 months after treatment of patients who qualified for anti-TNF treatment. The patients were divided in responders and non-responders to treatment. T-test was used to compare continuous variables between groups for parametric data and Mann-Whitney U test for non-parametric data. Pearson’s correlation was used to investigate statistical dependence between variables. Multiple regression analysis was performed to assess the ability of a number of factors to predict the degree of reduction in DAS28 score after treatment.

Results:   We reviewed the data of patients who received treatment with biologic factors for the last 2 years across East Kent Hospitals. One hundred and ten patients qualifying for and having received anti-TNF treatment for 6 months were included in the study. All baseline parameters where similar between the two groups. Eighty six patients had reduction in the DAS28 by 0.6 or more (responders) and 24 had reduction of less than 0.6 (non-responders). There was a statistically significant difference in the swollen-to-tender joints ratio between the responders and non responders (median difference = -0.55, 99.9%CI 0.46 – 0.64, p<0.001). Baseline GH and the number of tender joints were most strongly positively correlated with higher DAS28 score post treatment among all baseline factors (Pearson’s r = 0.87 and 0.84 respectively, p<0.001). Multiple regression analysis revealed that a model containing the GH and the swollen-to-tender joints ratio was able to explain 70% of the variance in post-treatment DAS28 score reduction (F (2, 107)= 129, p<0.001). GH and the swollen-to-tender joints ratio each made a strong contribution to explaining the DAS28 score reduction (beta= 0.48 and 0.38 respectively, p<0.001 and p<0.05 respectively). Patients with high swollen-to-tender joints ratio and low GH were more likely to have a bigger reduction in DAS28 score after treatment.

Conclusion:   We have identified a number of patients who fail anti-TNF treatment and in whom the DAS28 score is mainly the product of an increased score in
the subjective components of it (GH and TJ). DAS28 score is the main criterion for the administration of anti-TNF treatment in the UK. We challenge the current practice of using DAS28 score per se for the identification of patients suitable for anti-TNF treatment and we recommend that more focus should be placed upon the individual components of it, especially on those which are objective and reproducible, and finally rely less on subjective criteria such as the GH and the number of tender joints, especially when a discrepancy between the number of swollen and tender joints exists (as in our population). A large study, in the UK population, involving thousands of patients is needed, to adjust the weights of the individual components and increase the reliability of the DAS28 score for the better identification of patients who will really benefit from anti-TNF treatment. This will save the NHS approximately £10,000 per patient and also avoid serious side effects associated with anti-TNF drugs, in this sub-group of patients who, based on our results, is unlikely to benefit from this class of medications. We suggest that the swollen-to-tender joints ratio merits further investigation as a possible predictor of treatment success or failure as other investigators have also suggested.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-das28-score-may-misread-disease-activity-in-rheumatoid-arthritis-patients/