Spinal Cord Vascular Shunts: Spinal Cord Vascular Malformations and Dural Arteriovenous Fistulas

Leodante da Costa, MD; Amir R. Dehdashti, MD; Karel G. terBrugg E, MD


Neurosurg Focus. 2009;26(1):E6 

In This Article

Abstract and Introduction

Object: Spinal cord vascular malformations are rare, fascinating lesions. In this paper, the authors' goal was to demonstrate how these lesions, more specifically spinal cord arteriovenous malformations and dural arteriovenous fistulas, are assessed, classified, and managed at their institution. They also highlight some aspects of classification and management that may be different from the views of others.
Methods: The authors reviewed the 20-year experience at the senior author's institution regarding the management of spinal cord vascular malformations. They discuss the management, surgical and endovascular treatment results, and the classification that resulted from the combined experience of 3 major reference centers.
Results: The accumulated knowledge on embryological and pathophysiological aspects in such a rare disease resulted in a more global, patient-oriented (and not radiologically oriented) approach to spinal cord shunts.
Conclusions: The multiple classifications proposed for spinal cord vascular malformations reflect the continuous advancement of the authors' understanding. They adopt a classification based on new physiological and genetic data that treats these lesions as expressions of more complex disease processes and not simply a morphological target, with direct impact on therapeutic options.

The first clinical description of a spinal vascular malformation was in 1890 by Berenbruch,[10] who did not recognize the lesion as a vascular entity at surgery but only later at autopsy. Two decades later, Krause[32] recognized a spinal vascular malformation at surgery. Credit for first recognizing that spinal vascular malformations can cause SAH must be given to Heboldt in 1885[28] and Gaupp in 1888,[23] who referred to these lesions as "hemorrhoids of the pia mater." Since then, multiple eponyms and classifications have been used for the various types of vascular malformations of the spine and spinal cord. Although reflecting the continuous evolution in our understanding of the subject, the many classifications and variable nomenclature attest to our difficulty in understanding and comparing them. In 1943, Wyburn-Mason[57] recognized 2 main types of spinal vascular lesions: the venous type, consisting of an abnormal mass of sinuous, turgid, blue, pial veins, usually located posteriorly and below the midthoracic region of the cord; and the arteriovenous type, consisting of a fistulous capillary bed, preferentially located in the cervical enlargement anteriorly or at the lumbar enlargement, indicating for the first time that true medullary AVMs are different from what we know today as spinal dural fistulas.

The introduction of selective spinal angiography in the late 1960s and early 1970s and the studies of large series of patients by Houdart,[30] Di Chiro,[15] Baker,[6] Djindjian,[16,17,18] and their colleagues further improved our anatomical and hemodynamic understanding of those rare lesions. Aminoff and colleagues[3,4,5] later extensively reviewed the pathophysiology, clinical presentation, and prognosis of vascular malformations of the spinal cord. Confusion was propagated by attempts to classify spinal AVMs based on morphological aspects.[31,42,51] From an anatomical point of view, we prefer to distinguish paraspinal (including vertebral/vertebral) and epidural locations from dural and intradural locations of the AVM or AVF, realizing that the neurological symptoms will nearly always be related to the venous drainage pattern of the lesion no matter where the nidus is situated.

Vascular malformations of the spine and spinal cord are considered uncommon lesions, representing 3-16% of all spinal space-occupying lesions.[39,41,54,57] Dural AVFs are more common than spinal cord AVMs, the latter of which have an incidence of ~ 10% that of brain AVMs.[14,47] In this article, we aim to demonstrate the way spinal AVMs, specifically spinal cord AVMs and spinal DAVFs, are assessed, classified, and managed at our institution, highlighting some aspects of classification and management that may be different from the views of others.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.