**2. Pathogenesis**

Many factors have been involved in the pathogenesis of DAVFs; however, the exact pathophysiology remains unclear.

Some patients with DAVFs have been found to have venous sinus occlusion, which was proposed to be the direct cause for the development of the fistula. While others were found to have venous sinus thrombosis which resulted in the formation of small dural arteriovenous shunts by the invasion of angiogenic factors, released from the organizing thrombus, to a group of small dural arteries. However, it has been noted that DAVF can precede either venous sinus occlusion or thrombosis, because of its association with venous hypertension [11, 12].

Venous hypertension may lead to chronic focal decrease in both regional cerebral blood flow and cerebral perfusion pressure. The decrease in the cerebral blood flow will cause cerebral ischemia, which increases the expression of vascular endothelial growth factor (VEGF) via hypoxia-inducible factor-1α (HIF-1α) upregulation, leading eventually to the formation of DAVF. Also, some theories suggest that if venous hypertension was present, a small arteriovenous anastomosis might open up and generates DAVF when the shunting enlarges [13].

Besides occurring incidentally, the development of DAVFs has been caused by a wide range of events with head trauma, with or without skull fractures, being the most common one. Other preceding events include surgical operations like craniotomy, hormonal alterations like what happens in cases of pregnancy and menopause, infections like cases of otitis and sinusitis, and tumors, particularly meningiomas [14].

## **3. Classification**

The Borden-Shucart and Cognard systems are both known to be the most commonly used systems to classify DAVFs. However, there have been many classification schemes submitted for DAVF [10]. The Borden classification system (**Table 1**) classified DAVFs into three main types according to the presence or absence of cortical venous drainage (CVD) and the location of venous drainage [15]. In type I lesion, the drainage goes to the dural sinus or meningeal vein in an anterograde fashion. In type II lesion, the drainage to dural sinus goes in an anterograde fashion, however, the high pressure within the DAVF causes blood to flow in a retrograde way to subarachnoid veins. In type III lesions, the drainage goes directly and completely to subarachnoid veins [11].

The Cognard grading scale (**Table 2**) classifies DAVFs into five main types based on the direction of the flow whether anterograde or retrograde, presence or absence of a CVD, presence of venous ectasia whether ectatic or nonectatic cortical vein and the location of the fistula [14].

The presence of a CVD (Borden type II and III, Cognard types IIb–V) or absence (Borden type I, Cognard types I, IIa) and the pattern of the venous drainage all are considered major factors by the previous classification systems and were used to assess the risk of hemorrhage or any neurologic deficits [16]. The absence of a CVD


**Table 1.** *Borden classification of DAVF.*

*Arteriovenous Fistulas: The Pathological Bridge DOI: http://dx.doi.org/10.5772/intechopen.89724*


**Table 2.** *Cognard classification of DAVF.*

#### **Figure 1.** *Magnetic resonance angiography demonstrating the arteriovenous fistula.*

#### **Figure 2.**

*Digital subtraction angiography showed dural arterio-venous fistula with pseudo aneurysm at the abnormal anastomoses between the right temporal artery and the superior sagittal sinus and the facial vein.*

has been recognized as a favorable factor and the patients are usually asymptomatic or have symptoms like pulsatile tinnitus or exophthalmos [15]. On the other hand, presence of a CVD is an unfavorable factor that puts the dural fistula in a higher

risk group. DAVF has a higher chance of being symptomatic if the type of either classification systems was higher. Also, symptomatic lesions have a higher risk of hemorrhage than asymptomatic lesions [16] (**Figures 1** and **2**).
