Development of a predictive model for the relationship between serum pan-immunoinflammatory index levels and scar formation in facial burn patients.

OBJECTIVES: This study aims to develop a predictive model for scar risk in patients with facial burns using the Pan-Immune Inflammation Value (PIV) and other serological markers. METHODS: A retrospective cohort study was conducted on 367 patients with facial burns treated at a single institution between June 2021 and June 2023. Patients were categorized based on the presence of the scar 7 days post-treatment. Serum markers, including PIV, TNF-α, IL-10, EPO, TGF-β1, and ICAM-1, were measured. Multivariate logistic regression was employed to identify independent predictors of scar formation. A predictive model was developed and validated using a test set of 144 patients. RESULTS: Scar formation was associated with elevated levels of TNF-α and ICAM-1, and reduced levels of IL-10, EPO, and TGF-β1, indicating a pro-inflammatory profile. Patients with scars showed higher symptom severity, emotional distress, and functional impairment. The predictive model, incorporating these markers, achieved an AUC of 0.815 in the training set and 0.845 in the test set, demonstrating good predictive performance. CONCLUSION: Elevated pro-inflammatory markers and altered PIV levels were significant predictors of scar formation in patients with facial burns.