Clinical Heterogeneity and Risk Stratification of Carotid Blowout Syndrome After Definitive Radiation Therapy for Nasopharyngeal Carcinoma.

Abstract

Carotid blowout syndrome (CBS) is a rare but devastating complication following radiation therapy for nasopharyngeal carcinoma (NPC). This study aimed to identify clinically actionable risk factors for CBS and to explore heterogeneity among anatomically high-risk patients. Patients with NPC treated with a single course of definitive radiation therapy were retrospectively analyzed. Propensity score matching was applied to balance baseline characteristics between patients with and without CBS. Clinical, anatomic, and dosimetric variables were evaluated using logistic regression, with variable selection by least absolute shrinkage and selection operator. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUC). Heterogeneity analyses were performed among patients with macroscopic internal carotid artery (ICA) encasement. After matching, 31 patients with CBS and 62 matched controls were analyzed. Macroscopic ICA encasement and hypertension were strongly associated with CBS occurrence. In multivariable analysis, ICA encasement, hypertension, and ICA D0.5cc remained independent predictors of CBS, yielding good discrimination (AUC = 0.83). Among patients with macroscopic ICA encasement, those who developed CBS more frequently had hypertension and tended to receive higher irradiation to the pericarotid soft tissues, although dose differences were not statistically significant. Notably, progression to nasopharyngeal soft tissue necrosis with ICA exposure was observed exclusively in patients who developed CBS, whereas no patient who remained CBS-free demonstrated overt soft tissue necrosis during follow-up. CBS risk after definitive radiation therapy for NPC is primarily driven by anatomic vulnerability and systemic vascular factors. Pericarotid soft tissue injury and impaired post-radiation therapy healing appear to modulate progression from anatomic risk to clinical events. These findings support a risk-adapted surveillance and management strategy for patients with ICA encasement.