[Clinical outcomes following microsurgery and endovascular embolization in the management of spinal dural arteriovenous fistula: A meta-analysis study].

[OBJECTIVE] To compare the clinical effect of microsurgery and endovascular embolization in the treatment of spinal dural arteriovenous fistula (SDAVF) by meta-analysis.

[METHODS] A systematic review was performed to retrieve all relevant literature about surgical treatment or endovascular embolization of SDAVF up to December 2019 through PubMed, Embase, Web of Science, Cochrane Central Register of Controlled Trials Results, CNKI, Wanfang Data, and SinoMed. The Chinese and English key words included: "SDAVF", "spinal dural arteriovenous fistula", "spinal AVM", "spinal vascular malformation and treatment". The included studies were evaluated using the Newcastle-Ottawa scale. The early failure rate, long-term recurrence, neurological recovery, and complications were evaluated and the clinical effects of the two methods in the treatment of SDAVF were compared by using RevMan 5.3 software. And a further subgroup analysis of the therapeutic effect of endovascular embolization with different embolic agents was conducted.

[RESULTS] A total of 46 studies involving 1 958 cases of SDAVF were included, in which 935 cases were treated by microsurgery and 1 023 cases were treated by endovascular embolization. The funnel plot demonstrated that there was no publication bias. The results of meta-analysis showed that the incidence of early surgical failure was lower than that of endovascular embolization (=0.20, 95%: 0.13-0.30, < 0.05), and the long-term recurrence was also lower than that of endovascular embolization (=0.36, 95%: 0.22-0.58, < 0.05). The improvement of neurological function in the surgical patients is significantly higher than that in the patients treated with endovascular embolization (=2.86, 95%: 1.36-5.99, < 0.05). There was no significant difference in the occurrence of complications in these two groups (=1.52, 95%: 0.88-2.64, =0.14). In the cases of endovascular embolization, the risk of treatment failure or recurrence was higher with Onyx glue than with n-butyl 2-cyanoacrylate (NBCA), and the difference was statistically significant (=4.70, 95%: 1.55-14.28, < 0.05).

[CONCLUSION] Although the treatment of dural arteriovenous fistulas by intravascular embolization has been widely used, the clinical effect of microsurgery is still better than that of endovascular embolization. Large scale and high-quality randomized controlled trials are required to validate the efficacy and safety of endovascular treatment in SDAVF patients.