Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose: CBP is often the starting point for a suspicion of axSpA. In the ASAS-criteria for axial SpA either MRI of the SI-joints or HLA-B27-testing are dominant. But, CBP is an extremely common presenting symptom and not all patients can be followed up by MRI and/or HLAB27 testing. This analysis was undertaken to investigate which characteristics of back-pain forecast a positive HLA-B27.
Objectives: To evaluate which inflammatory characteristics of CBP are associated with the presence of HLA-B27 in patients with a suspicion of axSpA.
Methods: Baseline dataset from the EsPeranza cohort (<45 years old, symptoms duration 3-24 months and with inflammatory back pain -IBP- or asymmetrical arthritis or spinal/joint pain plus ≥1 SpA features) was used. For this study, only data from all patients with axial symptoms and HLA-B27 assessed were analysed. Univariable and multivariable logistic regression analyses were employed to estimate odds ratio for the association between IBP characteristics (morning stiffness, improve with exercise and not with rest, alternating buttock pain, insidious onset, awakening at 2nd half of night and good response to NSAID) and their different combinations with a positive HLA-B27 (local lab testing). Furthermore, diagnostic utility measures were also calculated.
Results: Data from 653 patients (54.2% male, mean (SD) age 33.0 (7.1) years and mean (SD) symptoms duration 11.0 (6.6) months. A total of 270 (41%) patients were HLA-B27 positive) were included in this analysis. Table shows the association between each separate characteristic (1A) and each possible IBP definition (1B) with a positive HLA-B27. Awakening at second half of night (OR=1.53;p<0.05) and good response to NSAID (OR=1.46;p<0.05) were significantly and positively associated with a positive HLA-B27. Among the existing criteria to define IBP, the ASAS criteria had the highest specificity (81%), but insufficient sensitivity. The addition of these two characteristics to the Calin-definition of IBP (88%) as well as the addition of NSAID response to the ASAS-definition of IBP (86%) just increased the specificity slightly.
Conclusion: Awakening at second half of night and good response to NSAIDs are distinguishing IBP characteristics associated with the presence of HLA-B27 in patients with suspected axSpA. However, the addition of these characteristics to the existing IBP definitions in the decision to test HLA-B27 does not significantly improve diagnostic efficiency in patients with suspected axSpA. Further, the most specific IBP definition for a positive HLA-B27 is the ASAS-definition.
Acknowledgements: The EsPeranza Program has been supported by an unrestricted grant from Pfizer
Disclosure of Interest: None declared
Table: Association between each of the CBP characteristics and each of the possible IBP definitions with a positive HLA-B27.
|
|
HLA-B27 + N=270 |
HLA-B27 – N=383 |
Univariable analysis |
Multivariable analysis |
Diagnostic utility measures |
|||||
|
|
N (%) |
N (%) |
OR |
OR |
Sen |
Spe |
PPV |
NPV |
LR+ |
LR- |
|
Table 1A: Individual Characteristic of IBP |
||||||||||
|
Morn. Stiff > 30 min |
171 (63.3) |
216 (56.4) |
1.34* |
1.10 |
63.3 |
43.6 |
44.2 |
62.8 |
1.12 |
0.84 |
|
Imp. exercise, not rest |
91 (33.7) |
114 (29.8) |
1.20 |
|
33.7 |
70.2 |
44.4 |
60.0 |
1.13 |
0.94 |
|
Alter. buttock pain |
86 (31.9) |
110 (28.7) |
1.16 |
|
31.9 |
71.3 |
43.9 |
59.7 |
1.11 |
0.96 |
|
Insidious onset |
184 (68.1) |
241 (62.9) |
1.26 |
|
68.1 |
37.1 |
43.3 |
62.3 |
1.08 |
0.86 |
|
Awake 2nd half night |
149 (55.2) |
163 (42.6) |
1.66*** |
1.53** |
55.2 |
57.4 |
47.8 |
64.5 |
1.30
|
0.78 |
|
Response to NSAIDs |
182 (67.4) |
217 (56.7) |
1.58*** |
1.46** |
67.4 |
43.3 |
45.6 |
65.4 |
1.19
|
0.75 |
|
Table 1B: Table 1B: IBP Definition |
||||||||||
|
Calin criteria |
98 (36.3) |
99 (25.8) |
1.63** |
– |
36.3 |
74.2 |
49.7 |
62.3 |
1.41 |
0.86 |
|
Berlin criteria |
173 (64.1) |
194 (50.7) |
1.74** |
– |
64.1 |
49.3 |
47.1 |
66.1 |
1.26 |
0.73 |
|
ASAS criteria |
85 (31.5) |
74 (19.3) |
1.92*** |
– |
31.5 |
80.7 |
53.5 |
62.3 |
1.63
|
0.85 |
|
Night + NSAID response (2/2) |
108 (40.0) |
109 (28.5) |
1.68** |
– |
40.0 |
71.5 |
49.8 |
62.8 |
1.40 |
0.84 |
|
Calin + Night + NSAID response (6/7) |
58 (21.5) |
45 (11.7) |
2.06** |
– |
21.5 |
88.3 |
56.3 |
61.5 |
1.84 |
0.89 |
|
Berlin + NSAID response (3/5) |
154 (57.0) |
153 (39.9) |
2.00*** |
– |
57.0 |
60.1 |
50.2 |
66.5 |
1.43 |
0.72 |
|
ASAS + NSAID response (5/6) |
66 (24.4) |
51 (13.3) |
2.11*** |
– |
24.4 |
86.7 |
56.4 |
61.9 |
1.83
|
0.87 |
|
ASAS + Buttock + NSAIDs (6/7)
|
29 (10.7) |
22 (5.7) |
1.98** |
|
10.7 |
94.3 |
56.9 |
60.0 |
1.88
|
0.95 |
*p<0.1;**p<0.05;***p<0.01
Disclosure:
V. Navarro-Compán,
None;
J. J. Aznar,
None;
L. F. Linares,
None;
E. Collantes-Estévez,
None;
R. B. M. Landewé,
None;
D. van der Heijde,
None;
P. Zarco,
None.
« Back to 2014 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/which-characteristics-of-inflammatory-back-pain-cbp-forecast-the-presence-of-hla-b27-results-from-the-esperanza-cohort/