Resisted Inspiration: A New Technique to Aid in the Detection of CSF-Venous Fistulas

DISCUSSION

In the current study, we found that resisted inspiration can dramatically decrease SVC venous pressure and concurrently increase CSF pressure, which has the effect of increasing the CSF-to-venous pressure gradient. Increasing this gradient will likely improve detection of a CVF using myelography, a notoriously challenging diagnosis. This is the first description of resisted inspiration as an adjunctive maneuver during myelography to potentially help increase the conspicuity of a CVF.

MR myelography, CT myelography, and digital subtraction myelography have been described as different diagnostic techniques to identify a CVF.^4,5 Each of these techniques requires the passage of contrast from the CSF to the venous system, which is driven by the CSF-to-venous pressure gradient. One prior article has described the effects of respiration on CVF detection. Amrhein et al⁸ have shown the benefit of end inspiration with the visualization of CVF on a CT myelogram and digital subtraction myelography. We have built on this work by showing that resisted inspiration dramatically decreases the venous pressure and increases the CSF pressure to an even greater extent compared with normal (nonresisted) inspiration.

Deep inspiration leads to negative intrathoracic pressure, increased cardiac return, and subsequent decreased venous pressure.^9,10 If normal inspiration can generate decreased SVC venous pressure, then resisted inspiration that requires greater effort from the muscles of inspiration should further decrease venous pressure. Gutzeit et al¹¹ showed that resisted inspiration increases the SVC/inferior vena cava blood flow for opacification of the pulmonary arteries and thereby aids in the detection of pulmonary emboli on CT angiography. While this effect was seen with resisted inspiration, they only noted minimal effect with normal inspiration.

Most interesting and somewhat unexpected, inspiration (both resisted and normal) leads to an increase in CSF pressure. Downward contraction of the diaphragm increases intrabdominal pressure and leads to an influx of blood into the epidural veins and, thus, increases pressure on the thecal sac.¹²

The CSF-to-venous pressure gradient should dictate the flow of CSF and, therefore, intrathecal contrast across a fistula. A 1-way valve between the SVC and azygos vein resists reflux of high pressure from the SVC into the azygos system but would facilitate low pressure in the SVC into the azygos system.¹³ This outcome is important as new and effective treatments become available for CVF, which include transvenous endovascular embolization¹⁴ in addition to surgical ligation and percutaneous fibrin glue injection.¹⁵

There are several limitations to our study. First, we studied a small number of separate patients for venous and CSF manometry. Ideally, we would study the pressure differences in a larger number of patients who would undergo both venous and CSF manometry. However, the venous manometry changes with resisted inspiration were large and consistent across all 4 patients and markedly different from both breathing at rest and normal inspiration. These results have already caused a change in the myelography technique at our institution. Finally, we used SVC pressure as a surrogate for pressure within the azygos system. While this is the most common venous drainage pathway of CVF, there are other venous drainage patterns⁷ that may not be affected by pressure changes in the SVC to the same degree.