Abstract
OBJECTIVE: N-butyl cyanoacrylate (NBCA) must reach the venous side beyond the fistulous point (venous penetration) for the successful transarterial embolization of dural arteriovenous fistulas (DAVFs). The purpose of this study is to identify several factors associated with the venous penetration of NBCA. METHODS: One hundred and thirty-four transarterial embolization procedures using an NBCA-lipiodol mixture (NBCA-Lip) in 38 consecutive patients with DAVFs treated at Kumamoto University Hospital were reviewed. DSA images before, during, and after the injection of NBCA-Lip, as well as procedure records, were reviewed with particular interest in the venous penetration of NBCA-Lip and factors potentially associated with the venous penetration, including volume rates of NBCA-Lip, artery injected, types of feeding artery injected (proper feeder or proximal feeder), status of injection (wedged or non-wedged), position of a microcatheter in relation to the sharp bend of the feeder, and use of adjunctive techniques. The correlation between venous penetration of NBCA-Lip with these factors was statistically analyzed using XLSTAT software (Lumina Decision Systems, Campbell, CA, USA). RESULTS: Among 134 NBCA-Lip injections, venous penetration was achieved in 68 injections (50.7%) and was significantly associated with injection from the proper feeder (odds ratio [OR], 11.9; 95% confidence interval [CI], 4.6-31.3; p <0.001), wedged injection (OR, 5.6; 95% CI, 2.6-11.9; p <0.001), injection in the non-curved position (OR, 15.9; 95% CI, 6.8-36.9; p <0.001), and injection with adjunctive techniques (OR, 2.1; 95% CI, 1.0-4.4; p = 0.04). Multivariate logistic regression analysis showed that wedged injection (OR, 8.8; 95% CI, 3.0-26.0; p <0.001) and injection in the non-curved position (OR, 25.0; 95% CI, 9.1-100.0; p <0.001) independently predicted venous penetration. CONCLUSION: The venous penetration of injected NBCA-Lip is significantly affected by the position and status of the microcatheter tip.