How good is the agreement between clinical diagnoses and classification criteria fulfilment in axial spondyloarthritis? Results from the SPARTAKUS cohort

Figures & data

Figure 1. Concordance between clinical International Classification of Diseases, 10th Revision (ICD-10) diagnoses and fulfilment of classification criteria. *M45.9; **M46.1, M46.8, M46.9 (no patient with M46.0 was identified); †Assessment of SpondyloArthritis international Society (ASAS) axial spondyloarthritis (axSpA) and/or modified New York (mNY) criteria for ankylosing spondylitis (AS). u-axSpA, undifferentiated axial spondyloarthritis; r-axSpA, radiographic axial spondyloarthritis; nr-axSpA, non-radiographic axial spondyloarthritis.

Table 1. Characteristics of patients fulfilling Assessment of SpondyloArthritis international Society (ASAS) axial spondyloarthritis (axSpA) and/or modified New York criteria versus patients not fulfilling any such criteria.

Variable	axSpA (n = 198)	No axSpA (n = 41)
Male sex	106 (54)	11 (27)*
Age (years)	51 ± 14	49 ± 11
Family history of SpA	77 (39)	11 (27)
Symptom duration (years)	26 ± 14	21 ± 13
HLA-B27 positive	163 (84)	4 (10)**
Back pain ≥ 3 months, with onset < 45 years	195 (99)	37 (93)*
Inflammatory back pain (ASAS definition)	162 (82)	23 (56)**
Sagittal lumbar flexion (modified Schober’s test) (cm)	4.0 ± 1.4	4.4 ± 1.0
Lateral lumbar flexion (cm)†	13.7 ± 5.7	14.8 ± 3.8
Chest expansion (cm)	4.8 ± 1.9	5.2 ± 1.7
Vital capacity (L)	3.8 ± 1.1	3.7 ± 0.9
Sacroiliitis on plain X-ray	141 (71)	2 (5)**
SI-joint bone marrow oedema on MRI‡	52 (49)	2 (6)**
Peripheral arthritis	100 (51)	17 (42)
Dactylitis	22 (11)	3 (7)
Heel enthesitis	79 (40)	17 (41)
History of uveitis	77 (39)	2 (5)**
Skin and/or nail psoriasis	17 (9)	0 (0)*
IBD§	20 (10)	6 (15)
Good response of back pain to NSAID	144 (73)	23 (56)*
Elevated CRP in the presence of back pain	129 (65)	22 (54)
MASES (0–15)	4.6 ± 3.9	7.1 ± 3.6**
VAS global (mm)	34 ± 25	39 ± 28
VAS pain (mm)	34 ± 27	39 ± 27
VAS fatigue (mm)	37 ± 27	44 ± 32
Evaluator’s VAS global (mm)	17 ± 14	20 ± 15
BASDAI	3.1 ± 2.3	3.9 ± 2.4
BASFI	2.3 ± 2.3	2.8 ± 2.2
BASMI	3.1 ± 1.7	2.6 ± 1.0
ASDAS-CRP	1.9 ± 1.0	2.1 ± 1.0
CRP (mg/L)	4.2 ± 8.2	3.3 ± 3.8
EQ-5D utility||	0.71 ± 0.25	0.63 ± 0.28*
Ongoing csDMARDs	51 (26)¶	11 (27)
Azathioprine	4 (2)	1 (2)
Methotrexate	31 (16)	7 (17)
Sulfasalazine	15 (8)	3 (7)
Ongoing bDMARDs	93 (47)	19 (46)
Adalimumab	19 (10)	5 (12)
Certolizumab pegol	16 (8)	2 (5)
Etanercept	29 (15)	7 (17)
Golimumab	6 (3)	2 (5)
Infliximab	19 (10)	2 (5)
Secukinumab	4 (2)	1 (2)

Data are shown as n (%) or mean ± sd.

SpA, spondyloarthritis; HLA, human leucocyte antigen; SI, sacroiliac; MRI, magnetic resonance imaging; IBD, inflammatory bowel disease; NSAID, non-steroidal anti-inflammatory drug; CRP, C-reactive protein; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; VAS, visual analogue scale; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; ASDAS-CRP, Ankylosing Spondylitis Disease Activity Score using C-reactive protein; EQ-5D, EuroQol 5-Dimensions; csDMARD, conventional synthetic disease-modifying anti-rheumatic drug; bDMARD, biological disease-modifying anti-rheumatic drug.

†Mean of right and left lateral lumbar flexion.

‡Previous or current SI-joint bone marrow oedema according to the ASAS definition.

§Of the 26 cases with known IBD, nine had ulcerative colitis and 17 Crohn’s disease.

||For EQ-5D, the standard British time trade-off-based preference set was used to assign utility values.

¶One AS patient was treated with concomitant methotrexate and sulfasalazine.

Missing data, n (%): HLA-B27, 4 (2); symptom duration, 1 (0.4); back pain < 45 years, 1 (0.4); back pain improved by exercise but not by rest, 2 (2); SI-joint MRI, 91 (46) (MRI was only actively performed within the protocol if required for classification purposes); chest expansion, 20 (8); vital capacity, 21 (9); MASES, 20 (8); VAS global/pain/fatigue, 9 (4); evaluator’s VAS global, 13 (10); BASDAI, 18 (8); BASFI, 20 (8); BASMI, 10 (4); ASDAS-CRP, 33 (14); CRP, 2 (1); EQ-5D, 20 (8).

*p < 0.05, **p < 0.001 for comparison by the chi-squared test/Fisher’s exact test, Mann–Whitney U-test, or Student’s t-test, as appropriate.

Table 2. Clinical diagnoses and predictive values of clinical diagnoses of ankylosing spondylitis (AS) (M459) and undifferentiated axial spondyloarthritis (u-axSpA) (M461, M468, M469) for classification criteria fulfilment.

Ankylosing spondylitis
	Total	Fulfilling mNY criteria for AS		Fulfilling ASAS criteria for r-axSpA			Not fulfilling mNY criteria for AS				Not fulfilling ASAS criteria for r-axSpA
Clinical diagnosis	n	Yes (n)	PPV	Yes (n)		PPV	n		NPV		n	NPV
AS (M459)	119	84	71	82		69	35		n/a		37	n/a
Not having AS	120	48	n/a	53		n/a	72		60		67	56
Undifferentiated axial spondyloarthritis
	Total	Fulfilling ASAS criteria fornr-axSpA						Not fulfilling ASAS criteria for nr-axSpA
Clinical diagnosis	n	Yes (n)			PPV			n		NPV
u-axSpA (M461, M468, M469)	120	36			30			84		n/a
Not having u-axSpA	119	21			n/a			98		82

mNY criteria, modified New York criteria for ankylosing spondylitis; ASAS, Assessment of SpondyloArthritis international Society; r-axSpA, radiographic axial spondyloarthritis; nr-axSpA, non-radiographic axial spondyloarthritis; PPV, positive predictive value; NPV, negative predictive value; n/a, not applicable.

Table 3. Characteristics of the radiographic axial spondyloarthritis/ankolysing spondylitis (r-axSpA/AS) and non-radiographic axial spondyloarthritis (nr-axSpA) patients.

Variable	r-axSpA/AS (n = 141)	nr-axSpA (n = 57)
Male sex	88 (62)**	18 (32)
Age (years)	54 ± 14**	45 ± 12
Family history of SpA	59 (42)	18 (32)
Symptom duration (years)	29 ± 15**	19 ± 11
HLA-B27 positive	115 (84)	48 (84)
Back pain ≥ 3 months with onset < 45 years	133 (94)	57 (100)
Inflammatory back pain (ASAS definition)	117 (83)	45 (79)
Sagittal lumbar flexion (modified Schober’s test) (cm)	3.9 ± 1.5*	4.4 ± 1.2
Lateral lumbar flexion (cm)†	12.8 ± 5.8**	15.9 ± 4.7
Chest expansion (cm)	4.7 ± 1.9	5.1 ± 1.9
Vital capacity (L)	3.7 ± 1.1	3.8 ± 1.0
Sacroiliitis on plain X-ray	141 (100)**	0 (0)
SI-joint MRI available	67 (48)*	40 (70)
SI-joint bone marrow oedema on MRI‡	32 (48)	20 (50)
Peripheral arthritis	68 (49)	31 (55)
Dactylitis	10 (7)*	12 (21)
Heel enthesitis	54 (39)	25 (44)
History of uveitis	62 (44)*	15 (27)
Skin and/or nail psoriasis	12 (9)	5 (9)
IBD§	18 (13)	2 (4)
Good response of back pain to NSAID	106 (76)	38 (68)
Elevated CRP in the presence of back pain	92 (65)	37 (65)
MASES (0–15)	4.1 ± 3.7*	5.7 ± 4.1
VAS global (mm)	32 ± 25	38 ± 24
VAS pain (mm)	32 ± 27	37 ± 25
VAS fatigue (mm)	35 ± 27	41 ± 28
Evaluator’s VAS global (mm)	16 ± 14	19 ± 15
BASDAI	3.1 ± 2.3	3.2 ± 2.1
BASFI	2.3 ± 2.4	2.1 ± 2.1
BASMI	3.4 ± 1.8**	2.3 ± 1.1
ASDAS-CRP	1.9 ± 1.0	1.9 ± 0.9
CRP (mg/L)	4.6 ± 9.2	2.6 ± 2.6
EQ-5D utility	0.72 ± 0.25	0.68 ± 0.26
Ongoing csDMARDs	36 (26)||	13 (23)
Azathioprine	3 (2)	1 (2)
Methotrexate	24 (17)	7 (12)
Sulfasalazine	10 (7)	5 (8)
Ongoing bDMARDs	66 (47)	27 (47)
Adalimumab	16 (11)	3 (5)
Certolizumab pegol	8 (6)	8 (14)
Etanercept	22 (16)	7 (12)
Golimumab	4 (3)	2 (3.5)
Infliximab	14 (10)	5 (9)
Secukinumab	2 (1)	2 (3.5)

Data are shown as n (%) or mean ± sd.

SpA, spondyloarthritis; HLA, human leucocyte antigen; ASAS, Assessment of SpondyloArthritis international Society; SI, sacroiliac; MRI, magnetic resonance imaging; IBD, inflammatory bowel disease; NSAID, non-steroidal anti-inflammatory drug; CRP, C-reactive protein; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; VAS, visual analogue scale; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; ASDAS-CRP, Ankylosing Spondylitis Disease Activity Score using C-reactive protein; EQ-5D, EuroQol 5-Dimensions; csDMARD, conventional synthetic disease-modifying anti-rheumatic drug; bDMARD, biological disease-modifying anti-rheumatic drug.

†Mean of right and left lateral lumbar flexion.

‡Previous or current SI-joint bone marrow oedema according to the ASAS definition.

§Of the 20 cases with known IBD, eight had ulcerative colitis and 12 Crohn’s disease.

||One AS patient was treated with concomitant methotrexate and sulfasalazine.

Missing data, n (%): HLA-B27, 4 (2); symptom duration, 1 (0.4); back pain < 45 years, 1 (0;.4); back pain improved by exercise but not by rest, 2 (2); chest expansion, 20 (8); vital capacity, 21 (9); SI-joint MRI, 91 (46) (MRI was only actively performed within the protocol if required for classification purposes); VAS global/pain/fatigue, 9 (4); evaluator’s VAS global, 13 (10); BASDAI, 18 (8); BASFI, 20 (8); BASMI, 10 (4); ASDAS-CRP, 33 (14); CRP, 2 (1); MASES, 20 (8).

*p<0.05, *p>0.001 for comparison with the nr-axSpA group by     the chi-squared test/Fisher’s exact test, Mann–Whitney U-test, or Student’s t-test, as appropriate.