Circulating periostin levels in patients with AS: association with clinical and radiographic variables, inflammatory markers and molecules involved in bone formation.

Abstract

The aim of this study was to evaluate serum periostin levels in patients with AS in comparison with healthy controls as well as their association with clinical, inflammatory and radiographic parameters and molecules involved in bone formation. Serum samples for periostin, total Dickkopf-1 (Dkk-1), sclerostin, VEGF and inflammatory markers were obtained from 65 TNF inhibitor-naive patients with AS. The BASDAI, BASFI, modified Stoke AS Spine Score and BASRI for the spine (BASRI-s) were assessed for each patient. Serum periostin levels were also measured in 36 sex-, age- and BMI-matched controls. Serum periostin levels were significantly lower in AS patients compared with controls [234.4 pg/ml (s.e.m. 7.5) vs 291.4 (s.e.m. 8.3), respectively; P < 0.001]. Periostin levels were higher in AS patients with elevated CRP (P = 0.005), high BASDAI (P = 0.014) and low BASRI-s (P = 0.033) and were correlated with BMI (r = -0.304, P = 0.014), ESR (r = 0.395, P = 0.001), CRP (r = 0.413, P = 0.001), BASRI-s (r = -0.242, P = 0.047) and sclerostin (r = -0.280, P = 0.024). In multiple regression analysis, periostin levels were an independent variable of CRP (β = 0.160, P = 0.009) and sclerostin levels (β = -0.311, P = 0.012). Our data suggest that periostin levels are low in patients with AS. Among AS patients, periostin levels are higher in those with higher disease activity, higher systemic inflammation and less extensive radiographic damage. Periostin is independently associated with CRP and sclerostin levels.