Serum macrophage migration inhibitory factor as a predictor of radiographic progression in ankylosing spondylitis: a retrospective cohort study

Abstract

OBJECTIVE: Macrophage migration inhibitory factor (MIF) is a pleiotropic pro-inflammatory cytokine implicated in the pathogenesis of ankylosing spondylitis (AS). This study aimed to evaluate whether serum MIF levels measured at baseline are associated with radiographic progression in AS patients assessed retrospectively over time. METHODS: This retrospective cohort study with concurrent healthy controls included 50 AS patients meeting the modified New York criteria and 60 age- and sex-matched healthy controls. Serum MIF was measured by ELISA at baseline. Clinical assessment included Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP). Radiographic progression was assessed retrospectively using the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) on radiographs obtained at baseline and at least 12 months later (median interval 24 months, range 12-36). Progressors were defined as patients with annualized mSASSS increase ≥ 1 unit/year. Two independent readers scored all radiographs while blinded to clinical data and time sequence. RESULTS: Serum MIF levels were significantly elevated in AS patients compared to controls (median 43.0 ng/mL vs. 11.2 ng/mL, p < 0.0001). Among AS patients, 15 (30%) were classified as progressors. After adjusting for age, sex, disease duration, baseline mSASSS, smoking, and ESR in multivariable logistic regression, elevated MIF remained independently associated with progression (OR 1.85 per 10 ng/mL increase, 95% CI 1.08-3.18, p = 0.028). MIF levels correlated with disease duration (ρ = 0.299, p = 0.035), ESR (ρ = 0.669, p < 0.0001), and mSASSS score (ρ = 0.324, p = 0.021). ROC analysis demonstrated that MIF > 51 ng/mL discriminated progressors from non-progressors with AUC 0.746 (95% CI 0.599-0.893), sensitivity 53.3%, specificity 94.3%, positive predictive value 80.0%, and negative predictive value 82.5%. CONCLUSION: Baseline serum MIF is independently associated with subsequent radiographic progression in AS patients after adjustment for key confounders. While sensitivity is modest (53%), the high specificity (94%) suggests that elevated MIF (> 51 ng/mL) may help identify patients at higher risk for progressive spinal damage. Key Points • Baseline serum MIF levels are independently associated with subsequent radiographic progression in AS after adjusting for age, sex, disease duration, baseline damage, smoking, and ESR. • MIF >51 ng/mL predicts progression with high specificity (94%) but modest sensitivity (53%), suggesting utility for identifying high-risk patients rather than screening all patients. • MIF correlates moderately with ESR (ρ=0.669) but shows independent prognostic value in multivariable models, consistent with MIF's dual role in driving inflammation and bone formation. • Rigorous radiographic methodology (blinded dual readers, ICC 0.94) and detailed ELISA protocols strengthen confidence in findings, though prospective validation in larger cohorts is needed.