Background/Purpose:

BASDAI has been extensively used to assess disease activity in ankylosing spondylitis (AS) BASDAI score ≥4 represent high disease activity and has been suggested as an eligibility criterion for initiation of anti-TNF therapy.ASDAS is a new composite clinical tool to assess disease activity in AS. ASDAS values of 2.1 and 3.5 have been selected as cut-off values to define high and very high disease activity, respectively and ASDAS ≥ 2.1 has been proposed as a suitable cut-off for eligibility for anti-TNF therapy The aim of this study was to estimate the corresponding BASDAI and ASDAS cut-off values for initiation of anti-TNF therapy and to assess the agreement between disease activity states defined by BASDAI and ASDAS values

Methods:

Patients with complete baseline data for ASDAS based on CRP and BASDAI were included. Mean ASDAS and BASDAI values were compared by Spearman correlation. Receiver operating characteristic (ROC) curves were constructed to determine the cut-off value of ASDAS that correspond to BASDAI score of 4 and to asses the cut-off value of BASDAI that correspond to ASDAS values of 2.1 and 3.5 Each cut-off point was calculated on the basis of the best trade-off values between sensitivity and specificity. The Kappa statistic was used to test the agreement between the disease activity states according to ASDAS and BASDAI, as well as the agreement between the clinical improvement as assessed by the two instruments (for patients with available data)

Results:

396 patients (291 M; 44 ±12.0) were identified with complete data at baseline Mean disease duration was 9.4 ± 8.2.Mean BASDAI, and ASDAS scores were 3.6 ± 2.3 and 2.9 ± 1.1 There was good correlation between ASDAS and BASDAI when tested by Spearman (p<0.001; r=0.9). The best trade-off ASDAS value corresponding to BASDAI score of 4 was 2.9 (80% sensitivity and 77% specificity; AUC: 0.88) whereas the best trade-off BASDAI values corresponding to ASDAS≥2.1 and ≥3.5 were 2.4 (84% sensitivity and 83% specificity; AUC: 0.90) and 3.7 (89% sensitivity and 69% specificity; AUC: 0.87), respectively. Similar cut-off values were obtained for different gender and age groups(≤40 / >40 years).Overall percent agreement of BASDAI≥4with ASDAS≥2.1 and ASDAS≥ 3.5were 67 and 76%, with κ valuesof 0.39 and 0.48, respectively. There was a moderate agreement between both the major clinical response measured by BASDAI50 and that by ΔASDAS≥2 (κ=0.441) and the minimal clinical improvement as measured by ΔBASDAI≥ 2and  ΔASDAS≥1.1 (κ=0.545)

Conclusion:

The ASDAS value that corresponds to the recommended BASDAI cut-off ≥4 for initiation of anti-TNF therapy is higher than the recommended ASDAS threshold of ≥ 2.1. Agreement between high disease activity states as defined by BASDAI and that by ASDAS is only fair, whereas agreement between clinical response as measured by BASDAI and that by ASDAS seems to be better.

Table1: Agreement between BASDAI≥4and high and very high disease activity states according to ASDAS

		ASDAS
		<2.1	≥2.1	<3.5	≥3.5	Total
BASDAI	<4	99	129	211	17	228
	≥4	1	167	78	90	168
	Total	100	296	289	107	396

---

« Back to ACR/ARHP Annual Meeting 2013

ACR Meeting Abstracts - https://acrabstracts.org/abstract/agreement-between-disease-activity-states-and-improvement-scores-as-defined-by-bath-ankylosing-spondylitis-disease-activity-index-and-ankylosing-spondylitis-disease-activity-score-cut-off-va/