Blood biomarker-based risk factors and predictive model for middle ear cholesterol granuloma

Middle ear cholesterol granuloma (CG) is an uncommon inflammatory lesion often arising in patients with persistent otitis media with effusion (OME). Although local pathological mechanisms have been well described, the contribution of systemic factors has not been clearly defined.

To investigate blood-derived risk indicators for middle ear CG and to establish a predictive model capable of early risk assessment in OME patients.

This retrospective case-control study included 31 patients with surgically and histologically verified CG and 60 patients with uncomplicated OME. Demographic characteristics, comorbidities, and laboratory parameters were compared. Logistic regression analyses were applied to identify independent predictors. A multivariable prediction model was generated, and its diagnostic performance was assessed using receiver operating characteristic (ROC) analysis.

Hypertension (OR = 22.77, 95% CI: 2.82-183.85; P = 0.003), higher total cholesterol levels (OR = 1.87, 95% CI: 1.15-3.03; P = 0.011), and increased serum albumin (OR = 1.25, 95% CI: 1.03-1.50; P = 0.022) were identified as independent risk factors for CG. Eosinophil count demonstrated a negative association (OR = 0.90, 95% CI: 0.83-0.99; P = 0.025). While single blood indicators showed modest performance (AUCs < 0.65), the combined model yielded stronger discrimination with an AUC of 0.801 (95% CI: 0.699-0.904; P < 0.001).

This study highlights systemic metabolic and inflammatory contributors-including hypertension and dyslipidemia-to the development of middle ear CG. The composite model offers a practical tool for identifying high-risk OME patients and may support earlier surveillance and clinical intervention.