[Diagnosis and treatment of carotid-cavernous fistula: analysis of 28 patients].

OBJECTIVE

To evaluate the feasibility and efficacy of endovascular treatment for different types of carotid cavernous fistula (CCF) via the approach of internal carotid artery (ICA) or inferior petrosal sinus (IPS).

METHODS

From April 2005 to June 2010, 28 CCF patients underwent endovascular treatment at our institution. There were 13 males and 15 females with a mean age of 39 years (range: 21 - 71). According to the Barrow's classification, they were classified into type A (n = 21), type B (n = 2) and type D (n = 5). Patients of type A underwent detachable balloon embolization of ipsilateral cavernous sinus or stent-graft placement via the ICA approach. Patients of types B and D received detachable coil plus n-BCA (n-butyl-2-cyanoacrylate) embolization of ipsilateral cavernous sinus via the IPS approach. The technical results, complications and therapeutic outcomes were reviewed.

RESULTS

Detachable balloons (number: 1 - 4) were used in 16 patients of type A. Angiography at immediate postembolization showed a complete occlusion of fistula in 15 patients and a small residual fistula (< 20%) in 1 patient. Five patients of type A received stent-graft placement. One stent was placed in 4 patients and 2 stents in 1 patient. Complete fistula closures with preserved ICA were documented on immediate angiogram in 3 patients whereas a large residual flow (> 50%) persisted in 1. The fistula was completely occluded after 3 detachable balloons were deployed in affected cavernous sinus through a gap between stent and vascular wall. Both fistula and ICA were occluded in 1 patient after stenting. No cerebral infarction was observed due to the adequate collateral blood flow from contralateral ICA. Complete closures of affected cavernous sinus were achieved in 6 patients of types B and D while residual flow (< 50%) persisted in 1. The number of detachable coils for each embolization ranged from 3 to 8 (mean: 6.0). The volume of n-BCA mixture varied from 1.0 to 2.1 ml (mean: 1.3). The mean duration of n-BCA injection was 65 s (range: 45 - 90). Clinical symptoms were completely relieved in 26 patients. During the mean follow-up period of 30 months (range: 12 - 60), no recurrence of clinical symptoms was observed. No thrombosis or stenosis was found in the lumina of stents.

CONCLUSION

Detachable balloon embolization is the preferential treatment for direct CCF. Detachable coil plus n-BCA embolization of cavernous sinus via the IPS approach is an efficient and safe treatment for indirect CCF.