SAT0271 Predictors of physical function and mobility in nonradiographic axial spondyloarthritis and ankylosing spondylitis

Background The definition of axial spondyloarthritis is evolving with a new ASAS classification. Maintenance of physical function and patient mobility are primary goals in the treatment of ankylosing spondylitis (AS) and nonradiographic axial SpA (nr-axSpA). Bath AS Functional Index (BASFI) and Bath AS Mobility Index (BASMI) were developed to measure physical function and mobility in AS patients. HAQ-S is a measure of physical function for Spondyloarthropathies. Predictors for these outcomes may be different for nr-axSpA. Objectives To identify physical function and mobility predictors in nr-axSpA and AS patients. Methods Baseline data from ABILITY-1 (N=185) and ATLAS (N=315) clinical trials were analyzed. Separate stepwise regression analyses were conducted for AS and nr-axSpA groups. For physical function, the dependent variables were BASFI for (AS) and HAQ-S for (nr-axSpA). Model selection included age, sex, disease duration, Bath AS Disease Activity Index (BASDAI), CRP level, SF-36 Mental Component Summary (MCS), nocturnal pain, total back pain, Patient’s Global Assessment (PGA), Physician’s Global Assessment (PhGA), and BASMI. For mobility the dependent variable was BASMI for both diseases. Model selection included the same variables plus other measures of physical function; BASFI, SF-36 Physical Component Summary (PCS), and SF-36 Physical Function (PF). Results Physical Function: Based on the stepwise regression, disease activity, age and mobility were predictors in both AS and nr-axSpA. For nr-axSpA, these were the only predictors for physical function. For AS patients, PGA of disease activity explained most of the variance in the model. CRP levels were also significant predictors of physical function in AS (Table 1). Mobility: Based on the stepwise regression, age and physical function were predictors of mobility in AS (BASFI) and nr-axSpA patients (HAQ-S) (Table 2). For AS patients, disease activity, PhGA of disease activity, and disease duration were also predictors of BASMI. Table 1. Predictors of BASFI in AS (ATLAS) and HAQ-S in nr-axSpA Patients (ABILITY-1) ATLASABILITY-1 VariablePartial R-SquareP-ValueVariablePartial R-SquareP-Value PGA0.40