顔面肥厚性瘢痕リスクの予測因子：損傷タイプと炎症サイクルに基づく後向き研究

Facial hypertrophic scar is a common clinical problem after trauma and surgery. Different injury types and healing processes may affect its risk, but a systematic analysis of relevant predictive factors is lacking.

To analyze the relationship between different types of facial injuries and inflammation-healing-related factors and the occurrence of hypertrophic scar, and to construct a predictive model.

Sixty patients who developed facial scars between June 2020 and June 2025 were retrospectively included. The primary outcome was the occurrence of hypertrophic scar. Firth-corrected univariate and multivariate logistic regression analyses were used to screen predictive factors, and the model performance was evaluated using ROC curves and bootstrap internal validation.

Hypertrophic scar occurred in 29 of the 60 patients, with an incidence rate of 48.3%. The highest incidence of hypertrophic scar was observed in burn patients (87.5%). Univariate analysis showed that burns, infection, and prolonged epithelialization time were significantly associated with hypertrophic scarring. In multivariate analysis, delayed epithelialization, infection, and burns remained independent risk factors. The apparent AUC of the predictive model was 0.815, and the bootstrap-corrected AUC was 0.792, indicating good calibration performance.

Burn injury type, infection, and delayed epithelialization are important predictors of hypertrophic scarring in the head and face. The model constructed based on clinical indicators has good predictive ability and can provide a reference for early risk assessment.