The value of joint ultrasonography in predicting arthritis in seropositive patients with arthralgia: a prospective cohort study

Table 3 Added value of ultrasound over clinical parameters according to a clinical prediction rule

Ultrasound abnormalities	Arthritis, yes	Arthritis, no	OR (95% CI)^d	p value
Clinical prediction rule risk-groups^a	n = 51	n = 112	OR (95% CI)^d	p value
Synovial thickening^b (16 joints)			1.5 (0.7–3.4)	0.3
Clinical prediction rule			3.5 (2.2–5.4)	< 0.01
Low risk	2/8 (25%)	19/67 (28%)
Intermediate risk	5/13 (38%)	5/25 (20%)
High risk	12/30 (40%)	6/20 (30%)
Synovial thickening (10 joints, no MTP)			6.1 (CI 1.6–23.2)	< 0.01
Clinical prediction rule			3.5 (CI 2.2–5.5)	< 0.01
Low risk	0/8 (0%)	3/67 (4%)
Intermediate risk	4/13 (31%)	1/25 (4%)
High risk	6/30 (20%)	0/20 (0%)
Power Doppler^b (16 joints)^c			1.7 (0.3–10.2)	0.5
Clinical prediction rule			3.6 (2.3–5.6)	< 0.01
Low risk	0/8 (0%)	4/67 (6%)
Intermediate risk	2/13 (15%)	1/25 (4%)
High risk	0/30 (0%)	0/20 (0%)

ACPA anti-citrullinated protein antibody, CI confidence interval, MTP metatarsophalangeal, OR odds ratio, RF rheumatoid factor
^aRisk groups based on the clinical prediction rule described in reference number 4
^bResults are presented for synovial thickening and power Doppler in at least one joint (present, %)
^cSame results when excluding MTP joints
^dLogistic regression analysis (note that the prediction rule risk groups were combined)

ISSN: 1478-6362