Algorithm for Endovascular Treatment of Patients With Spinal Arteriovenous Malformations

   Objective: To compare the endovascular treatment results in patients with spinal arteriovenous malformations (AVM) based on the proposed algorithm.

   Materials and methods: We retrospectively analyzed the endovascular treatment results in 72 patients with various types of spinal AVMs for 2014-2021. We formed 2 main groups of patients based on the developed indications for neurophysiological monitoring and provocative tests (NFM and PT): group 1 (n = 63) was treated according to the algorithm, and group 2 (n = 9) was treated before the algorithm was implemented. Group 1 was divided into subgroup 1.1 (n = 42) including patients with no indications for NFM and PT and subgroup 1.2 (n = 21) with patients indicated for NFM and PT. Subgroup 1.2 was further divided into subsubgroup 1.2A (n = 2) with patients indicated for NFM and PT yet to be informative due to severe neurological deficit and subsubgroup 1.2B (n = 19) with patients that had indications for and successfully underwent NFM and PT. We compared patients between groups 1 and 2, subsubgroup 1.2B and group 2 to evaluate the effectiveness of the algorithm (radical nature of the treatment, functional status assessment, complications).

   Results: Radical nature of spinal AVM treatment in group 1 was 79 % compared with 44 % in group 2 (P = 0.043). There was a significant improvement in motor function in group 1 compared with group 2 in each follow-up period (Р ≤ 0.007). Comparison of subsubgroup 1.2B and group 2 showed no significant differences (P = .05). The treatment led to complications in 5 patients (7 % of the total number of patients with spinal AVMs): 4 patients in group 2 and 1 patient in subsubgroup 1.2B. The effectiveness of the developed criteria was indirectly confirmed by difference in complications number between subsubgroup 1.2B and group 2 (P = 0.001).

   Conclusions: Group 1 showed better treatment results, significant clinical improvement, high radical nature of treatment, and a low percentage of complications compared with group 2. The proposed algorithm proved effective for main tasks of endovascular treatment of spinal AVMs.

1. Thron A. Spinal dural arteriovenous fistulas. Radiologe. 2001;41(11):955–960. (In German). PMID: 11765536. doi: 10.1007/s001170170031

2. Lad SP, Santarelli JG, Patil CG, Steinberg GK, Boakye M. National trends in spinal arteriovenous malformations. Neurosurg Focus. 2009;26(1):1–5. PMID: 19228104. doi: 10.3171/FOC.2009.26.1.E10

3. Yu JX, Hong T, Krings T, et al. Natural history of spinal cord arteriovenous shunts: an observational study. Brain. 2019;142(8):2265–2275. Published correction appears in Brain. 2019;142(11):e62. PMID: 31211368. doi: 10.1093/brain/awz153

4. Tissen TP. Endovascular treatment of arterial-venous malformations of the spinal cord. Russian Journal of Neurosurgery. 2007;(3):35–42. (In Russ.).

5. Gross BA, Du R. Spinal juvenile (type III) extradural-intradural arteriovenous malformations. J Neurosurg Spine. 2014;20(4):452–458. PMID: 24527826. doi: 10.3171/2014.1.SPINE13498

6. Grin’ AA, Sinkin MV, Aleynikova IB, Kordonskiy АYu. Intraoperative neurophysiological monitoring during thoracoscopic removal of the paravertebral spinal tumor (from practice). Russian Journal of Neurosurgery. 2018;20(4):75–79. (In Russ.). doi: 10.17650/1683-3295-2018-20-4-75-79

7. Niimi Y, Sala F, Deletis V, Setton A, de Camargo AB, Berenstein A. Neurophysiologic monitoring and pharmacologic provocative testing for embolization of spinal cord arteriovenous malformations. AJNR Am J Neuroradiol. 2004;25(7):1131–1138. PMID: 15313696. PMCID: PMC7976537.

8. Sala F, Beltramello A, Gerosa M. Neuroprotective role of neurophysiological monitoring during endovascular procedures in the brain and spinal cord. Neurophysiol Clin. 2007;37(6):415–421. PMID: 18083497. doi: 10.1016/j.neucli.2007.10.004

9. Jahangiri FR, Sheryar M, Al Okaili R. Neurophysiological monitoring of the spinal sensory and motor pathways during embolization of spinal arteriovenous malformations--propofol: a safe alternative. Neurodiagn J. 2014;54(2):125–137. PMID: 25080772.

10. Li X, Zhang HQ, Ling F, He C, Ren J. Differences in the electrophysiological monitoring results of spinal cord arteriovenous and intramedullary spinal cord malformations. World Neurosurg. 2019;122:e315–e324. PMID: 30339909. doi: 10.1016/j.wneu.2018.10.032

11. Perfiliyev AM, Kiselev VS, Chishchina NV, Rzayev DA. Endovascular embolization of spinal intramedullary arteriovenous malformation at the cervical level: case report and literature review. Voprosy neirokhirurgii imeni N. N. Burdenko. 2020;84(3):82–87. (In Russ.). doi: 10.17116/neiro20208403182

12. Perfilyev AM, Chishchina NV, Kiselev VS, Rzaev JA. Possibilities of endovascular embolization of spinal cord arteriovenous malformations using neurophysiological monitoring and provocative pharmacological tests. Russian Journal of Neurosurgery. 2020;22(2):14–24. (In Russ.). doi: 10.17650/1683-3295-2020-22-2-14-24

13. Takai K. Spinal arteriovenous shunts: angioarchitecture and historical changes in classification. Neurol Med Chir (Tokyo). 2017;57(7):356–365. PMID: 28515372. PMCID: PMC5566708. doi: 10.2176/nmc.ra.2016-0316

14. Aminoff MJ, Logue V. The prognosis of patients with spinal vascular malformations. Brain. 1974;97(1):211–218. PMID: 4434169. doi: 10.1093/brain/97.1.211

15. Deletis V. Intraoperative neurophysiology and methodologies used to monitor the functional integrity of the motor system. In: Deletis V, Shils J. Neurophysiology in Neurosurgery: A Modern Intraoperative Approach. Academic Press; 2002:25–51.

16. Feliciano CE, de León-Berra R, Hernández-Gaitán MS, Torres HM, Creagh O, Rodríguez-Mercado R. Provocative test with propofol: experience in patients with cerebral arteriovenous malformations who underwent neuroendovascular procedures. AJNR Am J Neuroradiol. 2010;31(3):470–475. PMID: 19892816. PMCID: PMC7963975. doi: 10.3174/ajnr.A1855

17. Li X, Zhang HQ, Ling F, et al. Intraoperative neurophysiological monitoring during the surgery of spinal arteriovenous malformation: sensitivity, specificity, and warning criteria. Clin Neurol Neurosurg. 2018;165:29–37. PMID: 29289918. doi: 10.1016/j.clineuro.2017.12.016

18. The R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; 2022.