Abstract
OBJECTIVES: This study aimed to examine associations between disease activity/severity markers across multiple PsA domains and structural damage development in the spine in a large longitudinal cohort. METHODS: We analysed patients from a prospective longitudinal PsA cohort and evaluated structural damage in the spine by assessing the development of new syndesmophytes (a binary outcome) on radiographs of the cervical and lumbar spine. We used Cox regression to examine the association between syndesmophyte development and disease activity and severity variables, including nail involvement, PASI, swollen joint count, SPARCC enthesitis score, dactylitis, cDAPSA, modified Steinbrocker score, radiographic sacroiliitis score (range 0-8), BASDAI, ESR and CRP. Separate multivariable models were conducted and adjusted for age, sex, PsA duration, BMI, smoking status, decade of observation, HLA-B*27, HLA-A*29, NSAID use and, where applicable, radiographic sacroiliitis score and b/tsDMARD use. RESULTS: Among 1584 PsA patients (median age 45), 218 (13.8%) had syndesmophytes at baseline (clinic entry). Median follow-up was 8.5 years [IQR: 4.4-14.3]. Among 922 patients without baseline syndesmophytes who underwent repeat radiographs, 166 (18.0%) developed new syndesmophytes, with a median time to development of 5.8 years [IQR: 2.4-9.9]. The incidence of new syndesmophytes appeared to decline in more recent calendar years. In multivariable models, higher ESR (HR 1.02, 95% CI 1.01-1.03) and higher radiographic sacroiliitis score (HR 1.19, 95% CI 1.07-1.32) were associated with syndesmophyte development. CONCLUSION: Syndesmophytes developed in approximately one-fifth of patients with PsA during long-term follow-up. Systemic inflammation and radiographic sacroiliitis are major, independent predictors of spinal structural damage.