Endovascular intervention for carotid blowout syndrome and predictors of recurrence: A retrospective and multicenter cohort study

Abstract

Introduction

Carotid blowout syndrome (CBS) is a potentially life-threatening complication of head and neck cancer and associated treatment. In this study, we assess the safety and efficacy of deconstructive and reconstructive procedures with a focus on CBS recurrence.

Methods

We conducted a multicenter retrospective analysis of a prospectively maintained database and identified 80 consecutive neurointerventions for CBS from 2016 to 2020. Patients were divided into 2 groups: deconstructive embolization (68 patients) and reconstructive stenting (12 patients). A comparative analysis was performed between the two groups.

Results

The CBS recurrence rate was 23.8 % with 84.2 % of recurrences occurring within 90 days of the primary event. The median time to rebleeding was 8.0 days (IQR: 2.0 – 28.5) with a mortality rate of 26.3 %. There was no significant difference in rates of peri-operative ischemic stroke (1.5 % vs. 0 %, p=0.672) or peri-operative mortality (1.5 % vs. 0 %, p=0.670). CBS recurrence was significantly higher in the reconstructive group (58.3 % vs. 17.6 %, p=0.002). On multivariate analysis, reconstructive stenting independently predicted rebleeding (adjusted hazard ratio 8.31, 95 % CI: 2.34–29.59, p=0.001). There was no significant association between CBS recurrence and pre-operative (p=0.600) or post-operative (p=0.275) anticoagulant/antiplatelet use.

Conclusion

CBS remains a challenging and potentially catastrophic complication of head and neck cancers. Reconstructive procedures, including stenting, predicted CBS recurrence independent of bleeding site or tumor invasion. Postoperative surveillance based on time intervals to CBS recurrence and engineering advancements including improved vessel reconstruction devices have the potential to reduce rehemorrhage rates and improve patient outcomes. Further clinical investigations amongst larger cohorts are needed.